Waters Application Notes - Glycans
Summary
Significance of the topic
Glycan analysis has become a cornerstone of biopharmaceutical characterization. Variations in glycosylation patterns critically influence drug efficacy, stability, immunogenicity and half-life. Regulatory authorities now scrutinize glycan attributes as critical quality attributes (CQAs) in submissions, and product comparability after scale-up or process changes hinges on robust glyco-profiling.
Objectives and overview of the article
This article reviews the core requirements for glycan analysis in a biopharmaceutical setting. It surveys common sample preparation and detection techniques, outlines best practices for ensuring reproducibility, and highlights the need for harmonization across laboratories to meet regulatory and quality-by-design (QbD) expectations.
Methodology and instrumentation
This review categorizes glycan analysis workflows into released glycans, glycopeptides, intact glycoproteins and monosaccharide profiling. Key sample preparation methods include:
- PNGase F and PNGase A enzymatic release of N-linked glycans
- Proteolysis for glycopeptide mapping
- Alkaline β-elimination or hydrazinolysis for O-linked glycans
- Permethylation to stabilize and facilitate MS analysis
- Amine or glycosylamine labeling for fluorescence detection
- HPAEC–PAD for underivatized monosaccharides and sialic acids
- Fluorescence detection of labeled glycans and glycopeptides
- ESI–MS and MALDI–TOF MS for mass mapping and structural assignment
Used instrumentation
- High-performance and ultra-performance liquid chromatography (HPLC, UPLC)
- Hydrophilic interaction chromatography (HILIC) columns
- High-pH anion-exchange (HPAEC) systems
- Capillary electrophoresis (HPCE/CZE)
- ESI–MS and MALDI–TOF MS platforms
- Automated sample preparation reagents (e.g., RapiFluor-MS labeling)
Main results and discussion
A case study of Myozyme® scale-up illustrated how undetected shifts in mannose-6-phosphate content triggered regulatory rejection. Agencies now demand detailed glycan profiles to control heterogeneity arising from cell line, culture conditions or downstream processing. Internally, interlaboratory studies using standard glycoproteins revealed significant variability in sialylation and antennary pattern measurements, underscoring the need for harmonized protocols.
To address this, best practices include:
- Defining CQAs such as antennary distribution, sialylation linkage, fucosylation and high-mannose content
- Employing orthogonal methods to offset biases of individual techniques
- Using well-characterized reference standards and material with known clinical history
- Establishing acceptance ranges for each glycoform and ensuring lot-to-lot consistency
Benefits and practical applications
Implementing standardized glycan analysis workflows enhances product quality control by:
- Detecting critical glycoform shifts early in process development
- Streamlining comparability studies after scale-up or process modification
- Reducing batch failure due to unexpected glycosylation changes
- Accelerating regulatory approval with comprehensive, reproducible data
Future trends and possibilities of use
Emerging directions in glycan analysis include:
- Automated, high-throughput sample preparation and labeling
- Next-generation MS and fragmentation techniques for site-specific glycosylation
- Standardized performance criteria for reproducibility, accuracy and sensitivity
- Integrated software solutions for glycan identification, quantification and reporting
- Global reference materials and interlaboratory proficiency testing
Conclusion
As glycosylation remains a pivotal determinant of biotherapeutic performance, rigorous, harmonized glycan analysis is essential. By combining complementary sample preparation protocols, orthogonal chromatographic and detection methods, and robust reference standards, biopharma organizations can satisfy regulatory demands, ensure product consistency, and expedite development.
References
- Walsh G. Nature Biotech. 28(9):917–924 (2010).
- Mack G. Nature Biotech. 26(6):592 (2008).
- ICH Q5E: Specifications for Biotechnological Products (2003).
- ICH Q6B: Specifications for Biotechnological Products (1999).
- FDA Guidance PAT—A Framework for Innovative Pharmaceutical Development (2004).
- Jefferis R. Biotechnol. Prog. 21:11–16 (2005).
- Brooks SA. Mol. Biotechnol. 28(3):241–255 (2004).
- EMA Guideline on Monoclonal Antibodies EMEA/CHMP/BWP/157653/2007 (2007).
- FDA Q8 Pharmaceutical Development (2006).
- FDA Q9 Quality Risk Management (2006).
- FDA Q10 Quality Systems (2008).
- Rathore AS. Trends Biotechnol. 27(9):546–553 (2009).
- Thobhani S et al. Glycobiology. 19(3):201–211 (2009).
- Kibe T et al. J. Clin. Biochem. Nutr. 21(1):57–63 (1996).
- Okazaki A et al. J. Mol. Biol. 336(5):1239–1249 (2004).
- Lauber MA et al. Anal. Chem. 87(10):5401–5409 (2015).
Content was automatically generated from an orignal PDF document using AI and may contain inaccuracies.
Waters Application Notes
Glycans
There are a variety of complementary techniques practiced to get the
complete story about a glycoprotein. Each technique varies in complexity
and provides a different layer of information.
This application notebook highlights a body of work that has been developed
to detail applications of innovative chemistry solutions that provide
complimentary techniques used to answer the what and why questions.
RELEASED
GLYCAN
ANALYSIS
MONOSACCHARIDE
SIALIC ACID
COMPOSITION
M
et
ho
d c
om
pl
ex
it
y
Increasing information
MIDDLE UP/DOWN -
SUBUNIT ANALYSIS
GLYCOPEPTIDE
MAPPING
INTACT PROTEIN
ANALYSIS
Stephan M. Koza, Ph.D.
Stephan joined Waters Corporation in
2011. His research, development, and
applications group has a primary focus on
the use of UPLC, HPLC, LC-MS, and sample
preparation technologies for the analysis
of biomolecules. He received his B.A. in
Chemistry from The Massachusetts College of Liberal Arts
and his Ph.D. in Analytical Chemistry from The University of
New Hampshire Prior to joining Waters he had nearly twenty
years of experience with Genetics Institute, Wyeth, and Pfizer
where he was primarily involved with biopharmaceutical
characterization and analytical method development.
Matthew Lauber, Ph.D.
Matthew is a Principal Applications
Chemist within the Consumables Business
Unit at Waters Corporation. For the last
4 years, he has applied his expertise in
protein chemistry and LC-MS based protein
characterization methods toward the
development and application of state-of-the-art reagents and
separation chemistries. Starting with his graduate studies at
Indiana University, Matthew has shown a penchant to pursue
new methodologies for puzzling analytical challenges. Most
recently, Matthew has helped lead investigations into the
rapid and sensitive analyses of released N-glycans as well as
the development of new strategies for characterizing protein
glycosylation that are based on the use of a wide-pore HILIC
column technology.
Kenneth J. Fountain III, M.S.
Kenneth received his B.S. in Biotechnology
from the Worcester Polytechnic Institute and
his M.S. in Chemistry from Tufts University.
He began his career with Waters Corporation
in 2001 as a Bioseparations Chemist.
In this role, he developed LC and LC-MS
methods for oligonucleotide, peptide, and intact protein
analysis. In 2004, he joined the Analytical R&D group at
Genzyme Corporation, where he implemented the use of
UPLC Technology for the analysis of small molecule drugs
and biomaterials. Ken returned to Waters Corporation in
2007, where he spent 7 years in the Consumables Business
Unit managing the Chemistry Applied Technology Group and
more recently, the Technology Advancement Group. Both
of these groups focus on the evaluation, development, and
application of chromatography columns, reagents, standards,
and sample preparation devices for the pharmaceutical,
biopharmaceutical, and food and environmental market
areas. In January 2015, Ken accepted a position as the
Director of the Biopharmaceutical Business Development
Group in the worldwide marketing organization at Waters.
Erin E. Chambers, Ph.D.
As a principal scientist, Erin has been
working almost exclusively on peptide and
protein bioanalysis for the last seven years,
while managing small and large molecule
bioanalysis and clinical research applications
for Waters’ Consumables Business Unit. A conversation
with her college dean helped put her on the path of scientific
study. Ultimately, Erin graduated from Yale University with
a degree in chemistry and earned her doctorate from Kings
College London. Erin loves the dynamic nature of her job,
helping to develop new methods and analytical systems that
have a profound influence on illnesses like diabetes and
Alzheimer’s disease.
Scott McCall, M.S.
Scott is a Research Chemist within the
Consumables Business Unit at Waters
Corporation. During his 8 year career
at Waters he has played a role in the
development of numerous types of column
chemistries. Most recently, his focus has
been toward investigating biomolecule separations including
peptides, proteins, and glycans. Scott has created methods
for profiling released high mannose and complex N-glycan
structures from monoclonal antibodies, and continues to work
to answer complex questions regarding glycan analysis.
Rajiv Bharadwaj, Ph.D.
Functioning in the role of Application
Specialist at Waters India Applications
Laboratory, Rajiv is instrumental in the
biopharmaceuticals and proteomics operations.
His area of focus includes high resolution
mass spectrometry based solutions, technical
support, and application development. As a day-dreamer, he
is influenced by fast technological turnover. Motivated by his
profound experience with R&D’s of Novozymes and Biocon, he
aspires to create benchmarks in analytical sciences. He has been
actively participating in scientific conferences/publications at
organizations like Proteomics Society of India and IIT Bombay,
to name a few. Trained at Indian Institute of Science (IISc),
Bangalore, he also holds a university rank for post graduation in
biotechnology. Being a gourmet food-enthusiast, Rajiv enjoys
bringing new flavors to his dishes.
Jennifer Fournier, M.S.
Jennifer joined Waters Corporation in
2004. She has worked in many different
parts of the organization. She started in Life
Science Research and Development, then
moved into the manufacturing group for the
MassPREP line of standards, and now is a
Product Manager for the same products she used to develop
and manufacture. She received her B.S. in Biotechnology and
M.S. in Biology from Worcester Polytechnic Institute. Prior to
joining Waters she taught high school biology and also worked
in manufacturing for a small pharmaceutical company that
manufactures Albuterol for inhalers. In her free time, Jennifer
enjoys spending time with her children and coaching.
Our Scientists
[ 1 ]
Michael F. Morris M.S.
As a Senior Scientist with Waters Synthesis
Group, Michael has worked on particle
synthesis and chromatographic support
development for the last 10 years.
Mike has brought 30 years experience
developing novel separations reagents
for pharmaceutical, biopharma, and diagnostic markets to
Waters’ Consumables Business Unit. Prior to joining the
synthesis group in 2006 he worked with VICAM, now a
division of Waters Corp, developing diagnostic products for
the food, agriculture, and dairy markets. Having grown up
in Pennsylvania farm country, he was initially interested in
studying Forestry but was convinced as an undergrad that
Chemistry offered a better fit to his curiosity about biological
systems. He went on to earn an M.S. in Biochemistry from
Purdue University. Mike has always been an avid outdoors
man and still spends most of his weekends in the woods and
on the water.
Darryl Brousmiche, Ph.D.
Darryl is a Principal Scientist and Manager
of the Synthesis Group in the Consumables
Business Unit at Waters Corporation.
Since joining Waters in 2003, he has
been engaged in developing new sample
prep and stationary phase materials for
chromatographic separations, as well as novel tagging
reagents. Darryl began his career studying Organic
Chemistry at the University of Ottawa before completing his
Ph.D. in Organic Photochemistry at the University of Victoria.
He then spent two years as a postdoc at the University of
California, Berkeley investigating light harvesting materials.
In his free time, Darryl enjoys camping, hiking, and skiing.
Pamela Iraneta, B.S. Chemistry, B.S.E.E.
Pamela has worked for Waters Corporation
for over 25 years. Her interest in all
forms of chromatography started while
characterizing silica sols using Sedimentation
Field Flow Fraction (SF3) – a technique
based on some of the same principles used
in chromatographic separations. She has experience in the
following chromatographic techniques: mix-mode ion exchange,
reversed-phase, HILIC, inverse-size exclusion, size exclusion
chromatography (SEC), gel permeation chromatography (GPC),
and supercritical fluid chromatography for both small molecule
and large biomolecule separations. Throughout her career at
Waters, she has made significant contributions to the Oasis
family of solid phase extraction (SPE) products, as well as the
Ostro protein precipitation and phospholipid removal plates,
and PoraPak Rxn cartridges. For column chemistries, she has
contributed to the success of ACQUITY BEH, XBridge, XSelect,
UPC2, BEH, to CORTECS, among others. Pam is a co-author on
26 peer-reviewed publications and inventor on 14 patents.
The patents are diverse in their application and have included
antimicrobial chromatographic systems, chromatographic
materials using charged surface technology, hybrid monoliths,
96-well SPE extraction devices, dried biological matrix
carriers and extraction devices, and chromatographic
materials for the separation of glyco-proteins and -peptides,
among others.
Sean M. McCarthy, Ph.D.
Sean joined Waters Corporation in 2008,
and currently serves as a Principal
Scientist and Senior Manager within the
Biopharmaceutical Business Organization.
During his tenure at Waters, he has led
applications development teams for a
variety of bioseparations application areas including protein
and peptide chromatography, oligonucleotide analysis, and
analytical method development/transfer. Sean received his
Ph.D. in Chemistry from the University of Vermont in 2005,
and prior to joining Waters completed postdoctoral training
within the Department of Pathology at the same institution,
where he focused on environmental oxidative stress
related diseases using a variety of biochemical and mass
spectrometric techniques.
Robert E. Birdsall, Ph.D.
As a Senior Scientist, Robert has an active
interest in the application of instrumentation
in the characterization of biopharmaceuticals.
Robert has spearheaded several key
initiatives since joining the Pharmaceutical
Business Operations at Waters. He
attributes his success, in the field of bioseparations, to his
academic advisors who fueled his passion and curiosity in the
sciences, the culmination of which resulted in a Doctorate in
Analytical Chemistry from Purdue University. One of Robert’s
favorite aspects of his job is developing novel applications
using cutting-edge instrumentation that can assist the
biopharmaceutical industry in bringing safe and efficacious
therapeutic products to market.
Ying Qing Yu, Ph.D.
Ying Qing is a Senior Science Manager
in the Pharmaceutical Group at Waters
Corporation. She joined Waters Corporation
in 2001, shortly after she received her
Ph.D. in Analytical Chemistry from Purdue
University. Her group’s focus is on protein
biotherapeutics characterization using the UPLC/QTof MS
platform. The projects she is currently working on range from
glycan profiling, peptide mapping, and lately, the hydrogen/
deuterium exchange mass spectrometry for higher order
structural protein analysis. She has extensive experience
in mass spectrometry, gas-phase ion chemistry, and LC
separation techniques.
Our Scientists (Continued)
[ 2 ]
Table of Contents
[ INTRODUCTION ]
A Review of Glycan Analysis Requirements (Published in BioPharm International) ................5
[ INTACT PROTEINS ]
Developing High Resolution HILIC Separations of Intact Glycosylated Proteins
Using a Wide-Pore Amide-Bonded Stationary Phase .............................................................. 11
Measuring the Glycan Occupancy of Intact mAbs Using HILIC and
Detection by Intrinsic Fluorescence ........................................................................................... 23
Intact Protein Characterization: Waters Biopharmaceutical System Solution with UNIFI ....... 26
Streamlining Compliant and Non-Compliant Intact Mass Analysis of
Biotherapeutic mAbs with the Biopharmaceutical Platform Solution with UNIFI ................... 28
[ GLYCOPEPTIDE MAPPING AND SUBUNIT ANALYSIS ]
HILIC Glycopeptide Mapping with a Wide-Pore Amide Stationary Phase ................................ 37
Mapping IgG Subunit Glycoforms Using HILIC and a Wide-Pore Amide Stationary Phase ..... 46
Structural Comparison of Infliximab and a Biosimilar via Subunit Analysis
Using the Waters Biopharmaceutical Platform with UNIFI ....................................................... 56
[ RELEASED GLYCANS ]
Rapid Preparation of Released N-Glycans for HILIC Analysis Using a
Novel Fluorescence and MS-Active Labeling Reagent .............................................................. 65
Robustness of RapiFluor-MS N-Glycan Sample Preparations and
Glycan BEH Amide HILIC Chromatographic Separations .......................................................... 76
GlycoWorks HILIC SPE Robust Glycan Sample Preparation ...................................................... 88
Quality Control and Automation Friendly GlycoWorks RapiFluor-MS
N-Glycan Sample Preparation .................................................................................................... 91
Profiling Released High Mannose and Complex N-Glycan Structures from
Monoclonal Antibodies Using RapiFluor-MS Labeling and
Optimized Hydrophilic Interaction Chromatography ................................................................ 97
[ 3 ]
Table of Contents (Continued)
Enhancing the Peak Capacity of High Molecular Weight N-Glycan
HILIC Separations with a Wide-Pore Amide Bonded Stationary Phase ................................. 106
Exploiting RapiFluor-MS Labeling to Monitor Diverse N-Glycan Structures via
Fluorescence and Mass Detection ........................................................................................... 109
New Capabilities for Monitoring Released N-Glycans through the Combined Use of
RapiFluor-MS Labeling, ACQUITY UPLC H-Class Bio System, and
Serial Fluorescence/ACQUITY QDa Mass Detection ............................................................... 113
Applying a Novel Glycan Tagging Reagent, RapiFluor-MS, and an
Integrated UPLC-FLR/QTof MS System for Low Abundant N-Glycan Analysis .................... 123
Released N-linked Glycan Analysis Using the Glycan Application Solution with UNIFI ..... 128
A Holistic Workflow for Acquisition, Processing, and Reporting Fluorescent-Labeled
Glycans Using the Biopharmaceutical Platform Solution with UNIFI ................................... 131
N-Linked Glycan Characterization and Profiling: Combining the Power of Accurate Mass,
Reference Glucose Units, and UNIFI Software for Confident Glycan Assignments .............. 139
Transferring RapiFluor-MS Labeled N-Glycans between UPLC and HPLC ............................ 149
[ RELEASED N- AND O-LINKED GLYCANS ]
Characterization of EPO N-Glycans Using RapiFluor-MS and HILIC Profiling ...................... 156
Comprehensive Characterization of the N and O-Linked Glycosylation of a
Recombinant Human EPO ....................................................................................................... 159
[ MONOSACCHARIDES ]
Future Proofing the Biopharmaceutical QC Laboratory: Chromatographic Scaling
of HPLC Monosaccharide Analyses Using the ACQUITY UPLC H-Class Bio System ........... 169
[ 4 ]
32
BioPharm International www.biopharminternational.com October 2015
C
a
l
l
is
t
a
im
a
g
e
s
/g
e
t
t
y i
m
a
g
e
s
M
ore than t wo -thirds of
recombinant biopharma-
ceutical products on the
market are glycoproteins,
and every stage of their manufacture is
carefully monitored and tested to ensure
consistency in quality, safety, and effec-
tiveness (1). Of the various aspects of
biopharmaceutical production (such as
yield, protein folding, and post-trans-
lational modifications), the host cell’s
biosynthesis of attached oligosaccharides
(glycans) is often the most difficult to
control. Selected expression systems and
even slight changes in process conditions
can alter the synthesis of glycans and
as a consequence, the physicochemical
properties (e.g., serum half-life), safety,
efficacy, and immunogenicity of the
end product. Regulatory agencies world-
wide require state-of-the-art glycan anal-
yses and the demands placed on these
methods have steadily increased as bet-
ter technologies have been developed.
Ultimately, robust, information-rich, and
reproducible methods for glycan analysis
must be included in regulatory filings
for glycoprotein-based biotherapeutics to
ensure accuracy and consistency. Method
simplification and standardization will
provide additional assurance that the gly-
can-analysis methods used are transferra-
ble between testing sites both within and
outside (e.g., contract research organiza-
tions) of the organization, ensuring better
quality and efficiency in manufacturing.
Glycans face new scrutiny
By 2008, the biotechnology company
Genzyme had developed and marketed
a review of Glycan
analysis requirements
Jennifer Fournier
The author
explores
the basic
rationale and
requirements for
standardized
glycan analysis.
Jennifer Fournier is product marketing
manager, consumables business
unit-asr, at waters corporation.
Glycosylation
ES684993_BP1015_032.pgs 10.05.2015 23:09 ADV
black
yellow
magenta
cyan
October 2015
www.biopharminternational.com BioPharm International 33
the drug Myozyme (alglucosidase
alfa) for the treatment of Pompe
disease, a rare and progressively
debilitating disorder characterized
by deficiency of lysosomal enzyme
alpha-glucosidase (GAA). The com-
pany was preparing to expand the
targeted treatment population from
primarily children to adults. Its
160-L production facility was work-
ing at capacity, so $53 million was
invested to build a 2000-L facility
for Myozyme in Allston, MA (2).
The company was ready to launch,
but FDA rejected Genzyme’s appli-
cation to sell the drug from the
2000-L plant. According to regu-
lators, the version made in the
2000 -L tank was no longer the
same drug as the one produced
in the 160 -L tank. FDA argued
that the differences in glycosyl-
ation—specifically in this case, the
composition of mannose-6-phos-
phate—meant that the drug was
no longer the biological equivalent
of the original material produced
in the 160-L bioreactor, and may
in fact introduce unknown clinical
variables. Genzyme argued that it
had already conducted a clinical
trial on the larger batch material,
demonstrating safety and effective-
ness. Ultimately, Genzyme had to
market the product from the larger
bioreactor under a different name.
The incident was a watershed
moment in the biopharmaceutical
industry, marking the emergence
of new challenges (1). First, regula-
tory authorities were beginning to
scrutinize the glycan structures of
biopharmaceutical products more
carefully based on established tech-
nical guidelines (e.g., ICH Q5E, ICH
Q6B, and FDA’s Guidance for Industry,
PAT—A Framework for Innovative
P h a r m a c e u t i c a l D e v e l o p m e n t ,
M a n u f a c t u r i n g , a n d Q u a l i t y
Assurance), yet there remained
inconsistencies in how FDA, the
European Medicines Agency (EMA),
and Japanese regulators determined
what is “biosimilar”. Second, prod-
ucts with complex glycosylation pat-
terns have the potential to easily fall
out of specification with changes
in biomanufacturing processes and
scale-up, so to meet the new regula-
tory demands, manufacturers had to
start carefully characterizing prod-
uct glycosylation and its relation to
the biological and clinical activity of
a medication, and begin monitoring
relevant glycan characteristics dur-
ing production (3, 4, 5).
In the years following FDA’s deci-
sion on Myozyme, the attention
given to glycan structure in bio-
pharmaceuticals has only increased,
reflecting improvement in analyti-
cal technology and a greater under-
standing of the role these structures
play in the physical characteristics,
stability, biological activity, and
the clinical safety and effective-
ness of a drug (6, 7). The technical
guidelines for characterizing and
monitoring glycans have changed
little since 20 08; manufact ur-
ers refer mainly to International
Conference on Har monization
(ICH) documents Q5E and Q6B
(3, 4). These documents list the
following recommendations on
characterizing glycans:
“For glycoproteins, the carbo-
hydrate content (neutral sugars,
amino sugars, and sialic acids) is
determined. In addition, the struc-
ture of the carbohydrate chains,
the oligosaccharide pattern (anten-
nary profile), and the glycosylation
site(s) of the polypeptide chain is
analyzed, to the extent possible.”
O t her g u idel i nes e x ist, set-
t i n g e x p e c t at ion s for g lyc a n
analysis, such as FDA’s Guidance
f o r I n d u s t r y, I m m u n o g e n i c i t y
Assessment for Therapeutic Protein
Products, and EMA’s 2007 mono-
graph on the characterization of
Glycosylation
Table I:
Common sample preparation methods in glycan analysis.
Sample preparation
Description
Application
PNGase F
Peptide-N-glycosidase F
Release of N-glycan chain except those
with (∝1,3)-linked core fucose
Release of complex, hybrid, and
oligomannose N-glycans
PNGase A
Peptide-N-glycosidase A
Release of N-glycan chain containing
(∝1,3)-linked core fucose
Release of complex, hybrid, and
oligomannose N-glycans
Proteolysis
The use of a protease to generate peptides
(including glycopeptides) from a glycoprotein
The peptides are often analyzed to investigate
glycosylation sites and occupancy
Alkaline beta
elimination/
hydrazinolysis
Chemical cleavage of O-linked glycans from
polypeptide chains
Primarily used in the analysis of O-linked
glycans
Permethylation
The methylation of oligosaccharide hydroxyl
groups to make glycans more amenable to
mass spectrometric (MS) analysis
MS-based characterization of glycans
including linkage analysis
Amine/
glycosylamine
labeling
Modification of glycans to facilitate
fluorescence detection
Detection of glycans and glycopeptides when
native detection is not available. Increases
options for chromatographic separation
methods. May also enhance MS analysis
ES684985_BP1015_033.pgs 10.05.2015 23:08 ADV
black
yellow
magenta
cyan
[ INTRODUCTION ]
Written by Jennifer Fournier, Product Marketing Manager at Waters, and published in BioPharm International.
[ 5 ]
34
BioPharm International www.biopharminternational.com October 2015
monoclonal antibodies (8). The
monog raph says the follow ing
on glycans:
“Glycan structures should be
characterized, and particular atten-
tion should be paid to their degree
of mannosylation, galactosylation,
fucosylation, and sialylation. The
distribution of the main glycan
structures present (often G0, G1,
and G2) should be determined.”
T hese doc u ments, however,
present few details on how to set
specification limits on glycans,
or recommend technologies and
procedures for consistent analyti-
cal results. The consequences for
this long-standing ambiguity are
that manufacturers and regulators
sometimes end up with different
ideas as to what constitutes a nec-
essary specification for a glycan
structure. Furthermore, compa-
nies submit reports to regulatory
authorities with widely different
analytical approaches. Procedures
may vary even within the same
organization, potentially leading
to inconsistent results, analytical
testing failures, and ultimately,
regulatory delays.
Quality by DesiGn
vs. Quality in practice
In 2002, in response to an increas-
ing burden on FDA of regulating
product manufacturing, and a
perception among companies that
regulatory requirements were lim-
iting flexibility in process optimi-
zation, FDA implemented changes
through its Pharmaceutical cGMP
21st Century Initiative and the
release of FDA’s process analyti-
cal technology guidance (PAT) (5).
The new approach placed greater
responsibility on the manufac-
turers to monitor quality control
through timely measurements and
corrections during processing.
A round the same time, ICH
published t wo g uidance doc u-
ments: ICH Q8 Phar maceutical
Development (7), ICH Q9 Quality
Risk Management (8), and ICH
Q10 Quality Systems Approach to
Pharmaceutical cGMP Regulations (9,
10, 11). These documents helped
to further define current scientific
and risk-based approaches to phar-
maceutical quality control.
The concept of quality by design
(QbD) was incorporated into FDA
review in 2004, which together
with the aforementioned guidelines,
emphasized a greater understanding
of the product and its manufactur-
ing process, and designing quality
control into the process, rather than
testing it after the fact (12). This
approach is particularly well-suited
to glycan analysis, which is typi-
cally associated with a complex set
of critical quality attributes (CQAs)
(such as sialylation, antennary
structure, or glycan structure het-
erogeneity) that are important to
the biological or clinical activity of
the drug. The CQAs must be identi-
fied, measured during process devel-
opment, and maintained within
required parameters (i.e., the design
space) during production.
In the case of glycans, the mea-
surement itself may introduce
uncertainty and risk, due to a high
variability of outcomes when char-
acterizing oligosaccharide chains.
An interlaboratory study present-
ing 11 industrial, regulatory, and
academic labs with the same set
of four released N-glycans demon-
strated that results were not consis-
tent between the laboratories when
Glycosylation
Table II:
Common separation methods in glycan analysis.
Separation method
Description
Application
HPAEC
High-pH anion-exchange
chromatography
Liquid chromatographic separation of
negatively charged (acidic) molecules
carried out at high pH
Separation, identification, and quantification
of glycans and glycopeptides. Analysis of
monosaccharide and/or sialic acid composition.
Often coupled with pulsed amperometric detection
(PAD) for detection of underivatized molecules
HPCE/CZE
High-performance capillary
electrophoresis/capillary-
zone electrophoresis
Separation of molecules by charge using an
electric field in a narrow capillary channel
Separation, identification, and quantification of
charged glycans. Analysis and quantification of
sialylation
HILIC
Hydrophilic-interaction,
high-performance liquid
chromatography
A variation of high-performance liquid
chromatography (HPLC) that separates
molecules using a hydrophilic stationary
phase and an organic-yet-water-miscible
liquid phase
Separation, identification, and quantification of
glycans and glycopeptides
WAX–HPLC
Weak-anion exchange–
high-performance liquid
chromatography
Separates anionic molecules based on
their degree of charge
Separation, identification, and quantification of
glycans and glycopeptides
RP–HPLC
Reverse-phase–high-
performance liquid
chromatography
Separates molecules on the basis of
differences in the strength of their interaction
with a hydrophobic stationary phase
Separation, identification, and quantification of
glycans and glycopeptides
ES684990_BP1015_034.pgs 10.05.2015 23:09 ADV
black
yellow
magenta
cyan
October 2015
www.biopharminternational.com BioPharm International 33
the drug Myozyme (alglucosidase
alfa) for the treatment of Pompe
disease, a rare and progressively
debilitating disorder characterized
by deficiency of lysosomal enzyme
alpha-glucosidase (GAA). The com-
pany was preparing to expand the
targeted treatment population from
primarily children to adults. Its
160-L production facility was work-
ing at capacity, so $53 million was
invested to build a 2000-L facility
for Myozyme in Allston, MA (2).
The company was ready to launch,
but FDA rejected Genzyme’s appli-
cation to sell the drug from the
2000-L plant. According to regu-
lators, the version made in the
2000 -L tank was no longer the
same drug as the one produced
in the 160 -L tank. FDA argued
that the differences in glycosyl-
ation—specifically in this case, the
composition of mannose-6-phos-
phate—meant that the drug was
no longer the biological equivalent
of the original material produced
in the 160-L bioreactor, and may
in fact introduce unknown clinical
variables. Genzyme argued that it
had already conducted a clinical
trial on the larger batch material,
demonstrating safety and effective-
ness. Ultimately, Genzyme had to
market the product from the larger
bioreactor under a different name.
The incident was a watershed
moment in the biopharmaceutical
industry, marking the emergence
of new challenges (1). First, regula-
tory authorities were beginning to
scrutinize the glycan structures of
biopharmaceutical products more
carefully based on established tech-
nical guidelines (e.g., ICH Q5E, ICH
Q6B, and FDA’s Guidance for Industry,
PAT—A Framework for Innovative
P h a r m a c e u t i c a l D e v e l o p m e n t ,
M a n u f a c t u r i n g , a n d Q u a l i t y
Assurance), yet there remained
inconsistencies in how FDA, the
European Medicines Agency (EMA),
and Japanese regulators determined
what is “biosimilar”. Second, prod-
ucts with complex glycosylation pat-
terns have the potential to easily fall
out of specification with changes
in biomanufacturing processes and
scale-up, so to meet the new regula-
tory demands, manufacturers had to
start carefully characterizing prod-
uct glycosylation and its relation to
the biological and clinical activity of
a medication, and begin monitoring
relevant glycan characteristics dur-
ing production (3, 4, 5).
In the years following FDA’s deci-
sion on Myozyme, the attention
given to glycan structure in bio-
pharmaceuticals has only increased,
reflecting improvement in analyti-
cal technology and a greater under-
standing of the role these structures
play in the physical characteristics,
stability, biological activity, and
the clinical safety and effective-
ness of a drug (6, 7). The technical
guidelines for characterizing and
monitoring glycans have changed
little since 20 08; manufact ur-
ers refer mainly to International
Conference on Har monization
(ICH) documents Q5E and Q6B
(3, 4). These documents list the
following recommendations on
characterizing glycans:
“For glycoproteins, the carbo-
hydrate content (neutral sugars,
amino sugars, and sialic acids) is
determined. In addition, the struc-
ture of the carbohydrate chains,
the oligosaccharide pattern (anten-
nary profile), and the glycosylation
site(s) of the polypeptide chain is
analyzed, to the extent possible.”
O t her g u idel i nes e x ist, set-
t i n g e x p e c t at ion s for g lyc a n
analysis, such as FDA’s Guidance
f o r I n d u s t r y, I m m u n o g e n i c i t y
Assessment for Therapeutic Protein
Products, and EMA’s 2007 mono-
graph on the characterization of
Glycosylation
Table I:
Common sample preparation methods in glycan analysis.
Sample preparation
Description
Application
PNGase F
Peptide-N-glycosidase F
Release of N-glycan chain except those
with (∝1,3)-linked core fucose
Release of complex, hybrid, and
oligomannose N-glycans
PNGase A
Peptide-N-glycosidase A
Release of N-glycan chain containing
(∝1,3)-linked core fucose
Release of complex, hybrid, and
oligomannose N-glycans
Proteolysis
The use of a protease to generate peptides
(including glycopeptides) from a glycoprotein
The peptides are often analyzed to investigate
glycosylation sites and occupancy
Alkaline beta
elimination/
hydrazinolysis
Chemical cleavage of O-linked glycans from
polypeptide chains
Primarily used in the analysis of O-linked
glycans
Permethylation
The methylation of oligosaccharide hydroxyl
groups to make glycans more amenable to
mass spectrometric (MS) analysis
MS-based characterization of glycans
including linkage analysis
Amine/
glycosylamine
labeling
Modification of glycans to facilitate
fluorescence detection
Detection of glycans and glycopeptides when
native detection is not available. Increases
options for chromatographic separation
methods. May also enhance MS analysis
ES684985_BP1015_033.pgs 10.05.2015 23:08 ADV
black
yellow
magenta
cyan
34
BioPharm International www.biopharminternational.com October 2015
monoclonal antibodies (8). The
monog raph says the follow ing
on glycans:
“Glycan structures should be
characterized, and particular atten-
tion should be paid to their degree
of mannosylation, galactosylation,
fucosylation, and sialylation. The
distribution of the main glycan
structures present (often G0, G1,
and G2) should be determined.”
T hese doc u ments, however,
present few details on how to set
specification limits on glycans,
or recommend technologies and
procedures for consistent analyti-
cal results. The consequences for
this long-standing ambiguity are
that manufacturers and regulators
sometimes end up with different
ideas as to what constitutes a nec-
essary specification for a glycan
structure. Furthermore, compa-
nies submit reports to regulatory
authorities with widely different
analytical approaches. Procedures
may vary even within the same
organization, potentially leading
to inconsistent results, analytical
testing failures, and ultimately,
regulatory delays.
Quality by DesiGn
vs. Quality in practice
In 2002, in response to an increas-
ing burden on FDA of regulating
product manufacturing, and a
perception among companies that
regulatory requirements were lim-
iting flexibility in process optimi-
zation, FDA implemented changes
through its Pharmaceutical cGMP
21st Century Initiative and the
release of FDA’s process analyti-
cal technology guidance (PAT) (5).
The new approach placed greater
responsibility on the manufac-
turers to monitor quality control
through timely measurements and
corrections during processing.
A round the same time, ICH
published t wo g uidance doc u-
ments: ICH Q8 Phar maceutical
Development (7), ICH Q9 Quality
Risk Management (8), and ICH
Q10 Quality Systems Approach to
Pharmaceutical cGMP Regulations (9,
10, 11). These documents helped
to further define current scientific
and risk-based approaches to phar-
maceutical quality control.
The concept of quality by design
(QbD) was incorporated into FDA
review in 2004, which together
with the aforementioned guidelines,
emphasized a greater understanding
of the product and its manufactur-
ing process, and designing quality
control into the process, rather than
testing it after the fact (12). This
approach is particularly well-suited
to glycan analysis, which is typi-
cally associated with a complex set
of critical quality attributes (CQAs)
(such as sialylation, antennary
structure, or glycan structure het-
erogeneity) that are important to
the biological or clinical activity of
the drug. The CQAs must be identi-
fied, measured during process devel-
opment, and maintained within
required parameters (i.e., the design
space) during production.
In the case of glycans, the mea-
surement itself may introduce
uncertainty and risk, due to a high
variability of outcomes when char-
acterizing oligosaccharide chains.
An interlaboratory study present-
ing 11 industrial, regulatory, and
academic labs with the same set
of four released N-glycans demon-
strated that results were not consis-
tent between the laboratories when
Glycosylation
Table II:
Common separation methods in glycan analysis.
Separation method
Description
Application
HPAEC
High-pH anion-exchange
chromatography
Liquid chromatographic separation of
negatively charged (acidic) molecules
carried out at high pH
Separation, identification, and quantification
of glycans and glycopeptides. Analysis of
monosaccharide and/or sialic acid composition.
Often coupled with pulsed amperometric detection
(PAD) for detection of underivatized molecules
HPCE/CZE
High-performance capillary
electrophoresis/capillary-
zone electrophoresis
Separation of molecules by charge using an
electric field in a narrow capillary channel
Separation, identification, and quantification of
charged glycans. Analysis and quantification of
sialylation
HILIC
Hydrophilic-interaction,
high-performance liquid
chromatography
A variation of high-performance liquid
chromatography (HPLC) that separates
molecules using a hydrophilic stationary
phase and an organic-yet-water-miscible
liquid phase
Separation, identification, and quantification of
glycans and glycopeptides
WAX–HPLC
Weak-anion exchange–
high-performance liquid
chromatography
Separates anionic molecules based on
their degree of charge
Separation, identification, and quantification of
glycans and glycopeptides
RP–HPLC
Reverse-phase–high-
performance liquid
chromatography
Separates molecules on the basis of
differences in the strength of their interaction
with a hydrophobic stationary phase
Separation, identification, and quantification of
glycans and glycopeptides
ES684990_BP1015_034.pgs 10.05.2015 23:09 ADV
black
yellow
magenta
cyan
[ INTRODUCTION ]
[ 6 ]
October 2015
www.biopharminternational.com BioPharm International 33
the drug Myozyme (alglucosidase
alfa) for the treatment of Pompe
disease, a rare and progressively
debilitating disorder characterized
by deficiency of lysosomal enzyme
alpha-glucosidase (GAA). The com-
pany was preparing to expand the
targeted treatment population from
primarily children to adults. Its
160-L production facility was work-
ing at capacity, so $53 million was
invested to build a 2000-L facility
for Myozyme in Allston, MA (2).
The company was ready to launch,
but FDA rejected Genzyme’s appli-
cation to sell the drug from the
2000-L plant. According to regu-
lators, the version made in the
2000 -L tank was no longer the
same drug as the one produced
in the 160 -L tank. FDA argued
that the differences in glycosyl-
ation—specifically in this case, the
composition of mannose-6-phos-
phate—meant that the drug was
no longer the biological equivalent
of the original material produced
in the 160-L bioreactor, and may
in fact introduce unknown clinical
variables. Genzyme argued that it
had already conducted a clinical
trial on the larger batch material,
demonstrating safety and effective-
ness. Ultimately, Genzyme had to
market the product from the larger
bioreactor under a different name.
The incident was a watershed
moment in the biopharmaceutical
industry, marking the emergence
of new challenges (1). First, regula-
tory authorities were beginning to
scrutinize the glycan structures of
biopharmaceutical products more
carefully based on established tech-
nical guidelines (e.g., ICH Q5E, ICH
Q6B, and FDA’s Guidance for Industry,
PAT—A Framework for Innovative
P h a r m a c e u t i c a l D e v e l o p m e n t ,
M a n u f a c t u r i n g , a n d Q u a l i t y
Assurance), yet there remained
inconsistencies in how FDA, the
European Medicines Agency (EMA),
and Japanese regulators determined
what is “biosimilar”. Second, prod-
ucts with complex glycosylation pat-
terns have the potential to easily fall
out of specification with changes
in biomanufacturing processes and
scale-up, so to meet the new regula-
tory demands, manufacturers had to
start carefully characterizing prod-
uct glycosylation and its relation to
the biological and clinical activity of
a medication, and begin monitoring
relevant glycan characteristics dur-
ing production (3, 4, 5).
In the years following FDA’s deci-
sion on Myozyme, the attention
given to glycan structure in bio-
pharmaceuticals has only increased,
reflecting improvement in analyti-
cal technology and a greater under-
standing of the role these structures
play in the physical characteristics,
stability, biological activity, and
the clinical safety and effective-
ness of a drug (6, 7). The technical
guidelines for characterizing and
monitoring glycans have changed
little since 20 08; manufact ur-
ers refer mainly to International
Conference on Har monization
(ICH) documents Q5E and Q6B
(3, 4). These documents list the
following recommendations on
characterizing glycans:
“For glycoproteins, the carbo-
hydrate content (neutral sugars,
amino sugars, and sialic acids) is
determined. In addition, the struc-
ture of the carbohydrate chains,
the oligosaccharide pattern (anten-
nary profile), and the glycosylation
site(s) of the polypeptide chain is
analyzed, to the extent possible.”
O t her g u idel i nes e x ist, set-
t i n g e x p e c t at ion s for g lyc a n
analysis, such as FDA’s Guidance
f o r I n d u s t r y, I m m u n o g e n i c i t y
Assessment for Therapeutic Protein
Products, and EMA’s 2007 mono-
graph on the characterization of
Glycosylation
Table I:
Common sample preparation methods in glycan analysis.
Sample preparation
Description
Application
PNGase F
Peptide-N-glycosidase F
Release of N-glycan chain except those
with (∝1,3)-linked core fucose
Release of complex, hybrid, and
oligomannose N-glycans
PNGase A
Peptide-N-glycosidase A
Release of N-glycan chain containing
(∝1,3)-linked core fucose
Release of complex, hybrid, and
oligomannose N-glycans
Proteolysis
The use of a protease to generate peptides
(including glycopeptides) from a glycoprotein
The peptides are often analyzed to investigate
glycosylation sites and occupancy
Alkaline beta
elimination/
hydrazinolysis
Chemical cleavage of O-linked glycans from
polypeptide chains
Primarily used in the analysis of O-linked
glycans
Permethylation
The methylation of oligosaccharide hydroxyl
groups to make glycans more amenable to
mass spectrometric (MS) analysis
MS-based characterization of glycans
including linkage analysis
Amine/
glycosylamine
labeling
Modification of glycans to facilitate
fluorescence detection
Detection of glycans and glycopeptides when
native detection is not available. Increases
options for chromatographic separation
methods. May also enhance MS analysis
ES684985_BP1015_033.pgs 10.05.2015 23:08 ADV
black
yellow
magenta
cyan
34
BioPharm International www.biopharminternational.com October 2015
monoclonal antibodies (8). The
monog raph says the follow ing
on glycans:
“Glycan structures should be
characterized, and particular atten-
tion should be paid to their degree
of mannosylation, galactosylation,
fucosylation, and sialylation. The
distribution of the main glycan
structures present (often G0, G1,
and G2) should be determined.”
T hese doc u ments, however,
present few details on how to set
specification limits on glycans,
or recommend technologies and
procedures for consistent analyti-
cal results. The consequences for
this long-standing ambiguity are
that manufacturers and regulators
sometimes end up with different
ideas as to what constitutes a nec-
essary specification for a glycan
structure. Furthermore, compa-
nies submit reports to regulatory
authorities with widely different
analytical approaches. Procedures
may vary even within the same
organization, potentially leading
to inconsistent results, analytical
testing failures, and ultimately,
regulatory delays.
Quality by DesiGn
vs. Quality in practice
In 2002, in response to an increas-
ing burden on FDA of regulating
product manufacturing, and a
perception among companies that
regulatory requirements were lim-
iting flexibility in process optimi-
zation, FDA implemented changes
through its Pharmaceutical cGMP
21st Century Initiative and the
release of FDA’s process analyti-
cal technology guidance (PAT) (5).
The new approach placed greater
responsibility on the manufac-
turers to monitor quality control
through timely measurements and
corrections during processing.
A round the same time, ICH
published t wo g uidance doc u-
ments: ICH Q8 Phar maceutical
Development (7), ICH Q9 Quality
Risk Management (8), and ICH
Q10 Quality Systems Approach to
Pharmaceutical cGMP Regulations (9,
10, 11). These documents helped
to further define current scientific
and risk-based approaches to phar-
maceutical quality control.
The concept of quality by design
(QbD) was incorporated into FDA
review in 2004, which together
with the aforementioned guidelines,
emphasized a greater understanding
of the product and its manufactur-
ing process, and designing quality
control into the process, rather than
testing it after the fact (12). This
approach is particularly well-suited
to glycan analysis, which is typi-
cally associated with a complex set
of critical quality attributes (CQAs)
(such as sialylation, antennary
structure, or glycan structure het-
erogeneity) that are important to
the biological or clinical activity of
the drug. The CQAs must be identi-
fied, measured during process devel-
opment, and maintained within
required parameters (i.e., the design
space) during production.
In the case of glycans, the mea-
surement itself may introduce
uncertainty and risk, due to a high
variability of outcomes when char-
acterizing oligosaccharide chains.
An interlaboratory study present-
ing 11 industrial, regulatory, and
academic labs with the same set
of four released N-glycans demon-
strated that results were not consis-
tent between the laboratories when
Glycosylation
Table II:
Common separation methods in glycan analysis.
Separation method
Description
Application
HPAEC
High-pH anion-exchange
chromatography
Liquid chromatographic separation of
negatively charged (acidic) molecules
carried out at high pH
Separation, identification, and quantification
of glycans and glycopeptides. Analysis of
monosaccharide and/or sialic acid composition.
Often coupled with pulsed amperometric detection
(PAD) for detection of underivatized molecules
HPCE/CZE
High-performance capillary
electrophoresis/capillary-
zone electrophoresis
Separation of molecules by charge using an
electric field in a narrow capillary channel
Separation, identification, and quantification of
charged glycans. Analysis and quantification of
sialylation
HILIC
Hydrophilic-interaction,
high-performance liquid
chromatography
A variation of high-performance liquid
chromatography (HPLC) that separates
molecules using a hydrophilic stationary
phase and an organic-yet-water-miscible
liquid phase
Separation, identification, and quantification of
glycans and glycopeptides
WAX–HPLC
Weak-anion exchange–
high-performance liquid
chromatography
Separates anionic molecules based on
their degree of charge
Separation, identification, and quantification of
glycans and glycopeptides
RP–HPLC
Reverse-phase–high-
performance liquid
chromatography
Separates molecules on the basis of
differences in the strength of their interaction
with a hydrophobic stationary phase
Separation, identification, and quantification of
glycans and glycopeptides
ES684990_BP1015_034.pgs 10.05.2015 23:09 ADV
black
yellow
magenta
cyan
[ 7 ]
36
BioPharm International www.biopharminternational.com October 2015
Glycosylation
comparing analyses of sialylation
and antennary structure (13). This
particular study did not address
the potentially added variability
caused by sample preparation. The
variability in outcomes may be
due in part to the availability of
numerous analytical approaches
and differences between labs as
to the selection of approach and
limitations of available equipment.
Inconsistent levels of training and
expertise in glycan analysis may
also have had an impact.
basic reQuirements
for stanDarDizeD protocols
of Glycan analysis
The establishment of a robust pro-
tocol for glycan analysis can help
extract the maximum benefit from
QbD practice; give manufacturers
greater control over product quality
and comparability between batches
and process modifications; and
ensure consistency and quality in reg-
ulatory submissions. Such a protocol
should have the following features.
Well-characterized
reference standards
A selection of known glycopro-
teins, glycopeptides, released gly-
cans, and monosaccharides will
help calibrate and validate any sys-
tem of glycoprofiling used in the
initial characterization of the prod-
uct or monitoring of the manufac-
turing process.
Well-characterized
sample standards
Isolated product with a known
clinical safety and efficacy profile
provides a reference point for com-
paring glycan structure of batch
products under different process
conditions and times.
Comprehensive identification
of critical glycan attributes
Structural features of glycans have
been linked to circulating half-life
of the glycoprotein in the blood
(sialylation); placental transport
(galactosylation); antibody-depen-
dent cell-mediated cytotoxicity
(core fucosylation); and a wide
range of effector functions, bio-
availability, and safety character-
istics (14, 15). Critical attributes
may include:
• Antennary profile
• Sia lylat ion st ate, i nc lud i ng
degree and linkage type (∝2-3
vs. ∝2-6)
• Site-specific glycosylation pro-
files and occupancy
• Fucosylation
• Galactosylation
• N-acetyl-lactosamine repeats
• High mannose residues composi-
tion
• Absence of immunogenic elements
such as N-glycolylneuraminic
acid ( Neu5Gc), deacet ylated
N-acetylneuraminic acid (Neu5Ac),
and Gal∝(1-3)Gal.
Variations in these CQAs intro-
duced by manufacturing can orig-
inate from selection of cell line,
bioreactor conditions such as nutri-
ent levels, pH or oxygen content, as
well as inadvertent modifications
during downstream purification.
Table III:
Common detection methods in glycan analysis.
Detection method
Description
Application
PAD
Pulsed amperometric
detection
Permits detection without fouling
electrodes
Detection of non-derivatized glycans
(most sugars do not absorb UV).
Frequently linked to high-pH anion-
exchange chromatography (HPAEC)
FD
Fluoresence detection
Selective fluorescent labeling and
detection, which may use derivatizing
agents
For the analysis of derivatized glycans
and glycopeptides when native
detection does not offer sufficient
sensitivity. Increases options for
chromatographic separation methods.
May also enhance mass spectrometric
(MS) analysis
ESI–MS
Electrospray ionization–
mass spectrometry
Mass measurement of gas-phase
ionized molecular species, where
ions are generated by applying a high
voltage to a liquid to create an aerosol,
with little fragmentation of molecules.
Can be directly integrated with liquid
chromatography
Mass mapping of glycans and
glycopeptides (including non-
derivatized) for identification of
sequence, antennary pattern,
modifications, and heterogeneity, etc.
MALDI–TOF MS
Matrix-assisted laser
desorption ionization–
time-of-flight mass
spectrometry
Mass measurement of gas-phase
ionized molecular species, where ions
are generated by embedding molecules
in a solid matrix, and releasing them as
ions via laser ablation
Mass mapping of glycans and
glycopeptides (including non-
derivatized) for identification of
sequence, antennary pattern,
modifications, and heterogeneity, etc.
ES684992_BP1015_036.pgs 10.05.2015 23:09 ADV
black
yellow
magenta
cyan
[ INTRODUCTION ]
[ 8 ]
October 2015
www.biopharminternational.com BioPharm International 37
Glycosylation
Establishment of ranges
of acceptable variation in
complex glycosylation patterns
M a ny g lycoprote i n s, p a r t ic u -
larly those with multiple glyco-
sylation sites, do not exist as a
single species, but as a mixture of
glycoforms. The natural complex-
ity and heterogeneity of glycan
str uct ures can have impor tant
functional relevance for a pro-
tein, and even minor, low-abun-
dance glycoform species can be
c r uc ia l. For clinica l pu r poses,
each product may have a differ-
ent tolerance or requirement for
glycoform distribution. In par-
ticular, clarity on the extent to
which low-abundance glycoforms
should be identified and moni-
tored is essential.
Adherence to best practices
in sample preparation
Selecting the most appropriate
method from the wide range of
published and commercial sample
preparation methods can be daunt-
ing. For example, purification of
glycans after release from protein
may be performed by solvent pre-
cipitation, solid-phase extraction,
or size-exclusion, hydrophobic-
interaction, or hydrophilic-inter-
ac t ion ch romatog raphy. Some
methods may lead to a non-stoi-
chiometric recovery of oligosaccha-
rides, skewing the results of glycan
profiling. Recent developments in
sample preparation have allowed
for a reduction in preparation
times and improved quantitative
yields of both high- and low-abun-
dance glycoforms (16).
Selection of glycoanalysis
technologies, methods, and strategy
There is a wide array of technolo-
gies that can be applied to glycan
analysis (see Tables I–III). A series
of detailed optimal work f lows
and best practices would help
to harmonize analytical proce-
dures between and within orga-
nizations that submit regulatory
reports. Workflows would cover
initial characterization through
to routine monitoring and quality
control. Considerations should be
made with respect to the simplic-
ity and time of analysis, as long
as the required levels of accuracy
and reproducibility are not com-
promised.
T he u s e of or t hogon a l a nd
c o m p l e m e n t a r y m e t h o d s o f
analysis help compensate for sys-
tematic errors in measurement.
These methods typically isolate
molecules and their fragments
based on different physical prop-
er t ies (e.g., h igh-per for ma nce
capillary electrophoresis [HPCE]
vs. hydrophilic interaction liq-
uid chromatography [HILIC]) or
analytical treatment (e.g., electro-
spray ionization–mass spectrom-
etry [ESI–MS] vs. matrix-assisted
l a s e r d e s o r p t i o n / i o n i z a t i o n –
time-of-flight mass spectrometry
[MALDI–TOF–MS]), and are com-
pared to compensate for potential
bias introduced by each analytical
method.
conclusion
Pharmaceutical regulatory agen-
cies worldwide have laid out the
general principles of quality con-
trol and risk management in bio-
pharmaceutical manufacturing.
Of the many CQAs that require
consideration, the variation of the
N-linked and O-linked glycosyl-
ation profiles of biotherapeutic
glycoproteins is one of the most
complex to assess. Currently, there
are numerous methods used to
elucidate these structures with
varying degrees of accuracy and
precision. In addition, the use of
these somewhat disparate meth-
odologies makes it not always pos-
sible to directly compare results
bet ween laborator ies. To meet
regulatory requirements for con-
sistent process and quality con-
trol, it would be benef icial to
establish more specific and stan-
dardized g uidelines for glycan
analysis performance with respect
to reproducibility, accuracy, and
sensitivity for the characterization
and routine monitoring of critical
glycoforms, including those of low
abundance. While such guidelines
are within purview of national
reg ulator y bodies and interna-
tional consensus organizations
(such as ICH), no such guidelines
have been released to date. The
requirements for glycan analy-
sis described in this article could
address many of the issues related
to process and quality control in
glycoprotein manufacturing.
references
1. G. Walsh, Nature Biotech. 28 (9), pp.
917–924 (2010).
2. G. Mack, Nature Biotech. 26 (6), pp.
592 (2008).
3. ICH, Q5E Specifications: Test
Procedures and Acceptance Criteria for
Biotechnological/Biological Products
,
EMA Document CPMP/ICH/5721/03
(Geneva, 2003).
4. ICH Q6B Specifications: Test Procedures
and Acceptance Criteria for
Biotechnological/Biological Products
,
EMA Document CPMP/ICH/365/96
(Geneva, 1999).
5. FDA, Guidance for Industry, PAT—A
Framework for Innovative
Pharmaceutical Development,
Manufacturing, and Quality Assurance
(Rockville, MD, Sept. 2004).
6. R. Jefferis, Biotechnol. Prog. 21, pp.
11–16 (2005).
7. S.A. Brooks, Mol. Biotechnol. 28 (3), pp.
241–255 (2004).
8. EMA, Guideline on Development,
Production, Characterization and
Specifications for Monoclonal Antibodies
and Related Products
, EMEA/CHMP/
BWP/157653/2007 (London, 2007).
9. FDA, Guidance for Industry, Q8
Pharmaceutical Development
(Rockville,
MD, May 2006).
10. FDA, Guidance for Industry, Q9 Quality
Risk Management
, (Rockville, MD, June
2006).
11. A.S. Rathore, A. Sharma, and D. Chillin,
BioPharm Int.
19, pp. 48–57 (2006).
12. A.S. Rathore, Trends Biotechnol. 27 (9),
pp. 546–553 (2009).
13. S. Thobhani et al., Glycobiology, 19 (3),
pp. 201–211 (2009).
14. T. Kibe et al., J. Clin. Biochem. Nutr. 21
(1), pp. 57–63 (1996).
15. A. Okazaki et al., J. Mol. Biol. 336 (5),
pp. 1239–1249 (2004).
16. M.A. Lauber et al., Anal. Chem. 87 (10),
pp. 5401–5409 (2015). ◆
ES684991_BP1015_037.pgs 10.05.2015 23:09 ADV
black
yellow
magenta
cyan
[ 9 ]
INTACT PROTEINS
[ 10 ]
WAT E R S S O LU T IO NS
ACQUITY UPLC® Glycoprotein BEH Amide,
300Å Column
Glycoprotein Performance Test Standard
ACQUITY UPLC H-Class Bio System
Xevo® G2 QTof Mass Spectrometer
K E Y W O R D S
ACQUITY UPLC H-Class Bio System,
BEH Amide 300Å, glycans,
glycosylated proteins, glycosylation, HILIC
A P P L I C AT IO N B E N E F I T S
■
■
Improved HILIC separations of intact
protein glycoforms through optimization
of stationary phase (bonded phase and pore
size), ion pairing, column pressurization,
and injection approaches.
■
■
MS-compatible HILIC to enable detailed
investigations of sample constituents.
■
■
Orthogonal selectivity to conventional
reversed-phase (RP) separations
for enhanced characterization of
glycoprotein samples.
■
■
Glycoprotein BEH Amide, 300Å,
1.7 µm stationary phase is QC tested via
a glycoprotein separation to ensure
consistent batch to batch reproducibility.
I N T RO DU C T IO N
Hydrophilic interaction chromatography (HILIC) has been widely adopted as a
tool for separating highly polar compounds. In fact, it has become a relatively
widespread technique for small molecule separations. By comparison, the
application of HILIC to large biomolecules has been comparatively limited
even though there are instances in which the separation selectivity of HILIC
would be highly valuable, for example during the characterization of protein
glycosylation. A standard approach to the analysis of glycans involves their
enzymatic or chemical release from their counterpart protein followed by their
chromatographic separation using HILIC. UPLC®-based separations founded
upon an optimized, sub-2-µm amide-bonded stationary phase has transformed
HILIC separations of released glycans by facilitating faster, higher resolution
separations.1-2 Although released glycan analysis is a gold-standard approach, it
has historically required lengthy and at times cumbersome sample preparation
techniques. And while the recent introduction of the GlycoWorks™ RapiFluor-
MS™ N-Glycan Kit alleviates many of these shortcomings,3 alternative means
of characterizing protein glycosylation must sometimes be investigated,4-6 for
instance when it is of interest to elucidate sites of modification.7
To enable the complementary analysis of glycans as they are still attached to
their counterpart proteins, we present an optimized HILIC stationary phase and
corresponding methods for resolving the glycoforms of intact and digested
glycoproteins. A wide-pore (300Å) amide-bonded, organosilica (ethylene
bridged hybrid; BEH)8 stationary phase is employed along with rigorously
developed methods to achieve unprecedented separations of the glycoforms of
intact proteins ranging in mass from 10 to 150 kDa.
Developing High Resolution HILIC Separations of Intact Glycosylated Proteins
Using a Wide-Pore Amide-Bonded Stationary Phase
Matthew A. Lauber, Scott A. McCall, Bonnie A. Alden, Pamela C. Iraneta, and Stephan M. Koza
Waters Corporation, Milford, MA, USA
[ 11 ]
E X P E R IM E N TA L
LC conditions
LC system:
ACQUITY UPLC H-Class Bio System
Sample temp.:
5 °C
Analytical
column temp.:
30 °C
(unless noted otherwise in the caption)
UV detection:
214/280 nm, 2 Hz
Fluorescence detection: Ex 280/Em 320 nm, 10 Hz
Flow rate:
0.2 mL/min
Injection volume:
≤1 µL (aqueous diluents). Note: It might
be necessary to avoid high organic
diluents for some samples due to the
propensity for proteins to precipitate
under ambient conditions. A 2.1 mm I.D.
column can accommodate up to a 1.2 µL
aqueous injection before chromatographic
performance is negatively affected.
Columns:
ACQUITY UPLC Glycoprotein BEH
Amide, 300Å, 1.7 µm, 2.1 x 150 mm
(p/n 176003702, with Glycoprotein
Performance Test Standard);
ACQUITY UPLC Glycoprotein BEH
Amide 300Å, 1.7 µm, 2.1 x 100 mm
(p/n 176003701, with Glycoprotein
Performance Test Standard);
ACQUITY UPLC BEH HILIC, 130Å, 1.7 µm,
2.1 x 150 mm (p/n 186003462)
XBridge BEH HILIC, 130Å, 5 µm,
2.1 x 150 mm (p/n 186004446)
ACQUITY UPLC Glycan BEH Amide, 130Å,
1.7 µm, 2.1 x 150 mm (p/n 186004742);
ACQUITY UPLC Glycan BEH Amide, 130Å,
1.7 µm, 2.1 x 100 mm (p/n 186004741);
Sample description
Glycoprotein Performance Test Standard (a formulation of bovine
RNase A and RNase B, p/n 186008010) and RNase B (Sigma R7884)
were reconstituted in 18.2 MΩ water to a concentration
of 2 mg/mL. Trastuzumab was diluted with water from its formulated
concentration of 21 mg/mL to a concentration of 2 mg/mL.
For column conditioning, the components of a vial of Glycoprotein
Performance Test Standard (100 µg) were dissolved in 25 µL of
0.1% trifluoroacetic acid (TFA), 80% acetonitrile (ACN) to create
a 4 mg/mL protein solution.
To investigate the resolution of glycan occupancy isoforms, Intact
mAb Mass Check Standard (p/n 186006552) was deglycosylated
using the following techniques. The glycoprotein (15 µg) was
reconstituted to a concentration of 0.52 mg/mL into a 28.2 μL
solution of 1% (w/v) RapiGest™ SF Surfactant and 50 mM HEPES
(pH 7.9). This solution was heated to 90 °C over 3 minutes,
allowed to cool to 50 °C, and mixed with 1.2 μL of GlycoWorks
Rapid PNGase F solution. Deglycosylation was completed by
incubating the samples at 50 °C for 5 minutes. To produce
partial deglycosylation, Intact mAb Mass Check Standard was
deglycosylated using only a 5 minute, 50 °C incubation with
PNGase F without a heat-assisted pre-denaturation.
Method conditions
(unless otherwise noted)
Column conditioning
New (previously unused) ACQUITY UPLC Glycoprotein BEH
Amide, 300Å, 1.7 µm Columns should be conditioned, before
actual test sample analyses, via two sequential injections and
separations of 40 µg Glycoprotein Performance Test Standard
(10 µL injections of 4 mg/mL in 0.1% TFA, 80% ACN) or with
equivalent loads of a test sample for which the column has been
acquired. The separation outlined in Figure 2 can be employed for
conditioning with the Glycoprotein Performance Test Standard.
[ 12 ] Developing High Resolution HILIC Separations of Intact Glycosylated Proteins Using a Wide-Pore Amide-Bonded Stationary Phase
Competitor columns: PolyHYDROXYETHYL A™, 300Å,
3 µm, 2.1 x 100 mm;
Glycoplex® A, 3 µm, 2.1 x 100 mm;
ZORBAX® RRHD 300-HILIC, 300Å,
1.8 µm, 2.1 x 100 mm;
Halo® PentaHILIC, 90Å,
2.7 µm, 2.1 x 100 mm;
SeQuant® ZIC-HILIC, 200Å,
3.5 µm, 2.1 x 100 mm;
Accucore™ Amide, 150Å,
2.6 µm, 2.1 x 100 mm;
TSKgel® Amide-80, 80Å,
3 µm, 2.0 x 100 mm
Column connector
(for coupling
150 mm columns):
0.005 x 1.75 mm UPLC SEC Connection
Tubing (p/n 186006613)
Vials:
Polypropylene 12 x 32 mm , 300 μL
Screw Neck Vial, (p/n 186002640)
Gradient used to demonstrate the progression of HILIC separation
technologies (Figure 1):
Column dimension:
2.1 x 150 mm
Mobile phase A:
0.1% (v/v) TFA, water
Mobile phase B:
0.1% (v/v) TFA, ACN
Time
%A
%B
Curve
0 20.0 80.0 6
20 80.0 20.0 6
21 20.0 80.0 6
30 20.0 80.0 6
Focused gradient for RNase B HILIC separations (Figures 2 and 5)
Column dimension:
2.1 x 150 mm
Mobile phase A:
0.1% (v/v) TFA, water
Mobile phase B:
0.1% (v/v) TFA, ACN
Time %A
%B Curve
0 20.0 80.0 6
1 34.0 66.0 6
21 41.0 59.0 6
22 100.0 0.0 6
24 100.0 0.0 6
25 20.0 80.0 6
35 20.0 80.0 6
Gradient for benchmarking/evaluations (Figure 3)
Column dimension:
2.1 x 100 mm
Mobile phase A:
0.1% (v/v) TFA, water
Mobile phase B:
0.1% (v/v) TFA, ACN
Time
%A
%B
Curve
0.0 20.0 80.0 6
0.7 30.0 70.0
6
29.3 45.0 55.0
6
30.0 80.0 20.0
6
31.3 80.0 20.0
6
32.0 20.0 80.0
6
40.0 20.0 80.0
6
Gradient employed to select a mobile phase additive (Figure 4):
Column dimension:
2.1 x 150 mm
Mobile phase A:
0.1% (v/v) TFA, water or 50 mM
ammonium formate, pH 4.4 or
0.5% (w/v) formic acid, water
Mobile phase B:
ACN
Time
%A
%B
Curve
0 20.0 80.0 6
20 80.0 20.0 6
21 20.0 80.0 6
30 20.0 80.0 6
Focused gradient for reversed phase of RNase B (Figure 6):
Column dimension:
2.1 x 150 mm
Mobile phase A:
0.1% (v/v) TFA, water
Mobile phase B:
0.1% (v/v) TFA, ACN
Time
%A
%B
Curve
0 95.0 5.0 6
1 74.5 25.5 6
21 67.5 32.5 6
22 10.0 90.0 6
24 10.0 90.0 6
25 95.0 5.0 6
35 95.0 5.0 6
[ 13 ]
Developing High Resolution HILIC Separations of Intact Glycosylated Proteins Using a Wide-Pore Amide-Bonded Stationary Phase
Focused gradient for intact trastuzumab (Figures 7 and 8)
Column dimension:
2.1 x 150 mm, with varying lengths
25 µm I.D. PEEK post-column tubing
Or two coupled 2.1 x 150 mm columns
Mobile phase A:
0.1% (v/v) TFA, water
Mobile phase B:
0.1% (v/v) TFA, ACN
Time
%A
%B
Curve
0 20.0 80.0 6
1 30.0 70.0 6
21 37.0 63.0 6
22 70.0 30.0 6
24 70.0 30.0 6
25 20.0 80.0 6
45 20.0 80.0 6
Conditions for resolving glycan occupancy isoforms
of an IgG (Figure 9):
Column dimension:
Two coupled 2.1 x 150 mm or a single
2.1 x 150 mm
Column temp.:
80 °C
Mobile phase A:
0.1% TFA, 0.3% HFIP in water
Mobile phase B:
0.1% TFA, 0.3% HFIP in ACN
Time
%A
%B
Curve
0.0 20 80 6
10.0 50 50
6
11.0 100 0
6
14.0 100 0
6
15.0 20 80
6
25.0 20 80
6
MS conditions
MS system:
Xevo G2 QTof
Ionization mode:
ESI+
Analyzer mode:
Resolution (~20 K)
Capillary voltage:
3.0 kV
Cone voltage:
45 V
Source temp.:
150 °C
Desolvation temp.:
350 °C
Desolvation gas flow: 800 L/Hr
Calibration:
NaI, 2 µg/µL from 100–2000 m/z
Acquisition:
500–4000 m/z, 0.5 sec scan rate
Data management:
MassLynx® Software (v4.1)
R E SU LT S A N D D IS C U S S IO N
Progression of HILIC technology for
glycoprotein separations
HILIC originated in the early 1990s as a separation technique
to resolve highly polar molecules using mobile phases adapted
from reversed phase chromatography.9 The HILIC separation
mechanism is largely believed to be dependent on a polar
stationary phase that adopts an immobilized water layer.9
Hydrophilic analytes partition into this immobilized water layer
and undergo interaction with the phase via a combination of
hydrogen bonding, dipole-dipole, and ionic interactions. In this
way, hydrophilic analytes will be retained on the HILIC phase
under apolar initial mobile phase conditions and later eluted
by increasing polar mobile phase concentration via use of
an LC gradient.9
Numerous HILIC or HILIC-like stationary phases have been
developed in the last two decades. Many based solely on
unbonded silica particles are widely available, so too are
HILIC phases based on polyalcohol bondings or charge bearing
surfaces, such as those with zwitterionic bondings. For the
enhanced retention and selectivity of glycans, amide bonded
phases have become increasingly popular. The ACQUITY UPLC
Glycan BEH Amide stationary phase found in Waters Glycan
Column has, for instance, found wide-spread use for high
resolution released glycan separations.
As mentioned before, HILIC has, however, not seen wide-spread
use in intact large molecule applications. Concerns that high
organic solvent concentrations can result in protein precipitation
have most likely discouraged many from attempting to develop
HILIC-based, protein separation methods. Endeavoring beyond
these perceptions, we have developed a new amide-bonded
stationary phase based on a wide-pore, organosilica (ethylene
bridged hybrid; BEH) particle that was specifically designed to
facilitate large molecule separations. It exhibits a porous network
accessible to most proteins and an average pore diameter that
does not impart significant peak broadening due to restricted
diffusion, which can occur when protein analytes are too close
in size to the average pore diameter of a stationary phase
(e.g. within a factor of 3).
[ 14 ] Developing High Resolution HILIC Separations of Intact Glycosylated Proteins Using a Wide-Pore Amide-Bonded Stationary Phase
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
5
6
7
8
9
10
11
12
13
14
15
A
21
4
Time (min)
Unbonded BEH
1.7 m
130Å
Amide-Bonded BEH
1.7 m
130Å
Amide-Bonded BEH
1.7 m
300Å
Improved
Resolution
Unbonded BEH
5 m
130Å
Glycoprotein BEH Amide,
300Å, 1.7 m
Figure 1. Progression of HILIC stationary-phase technologies for intact glycoprotein separations.
Separation of 1 µg of RNase B using 2.1 x 150 mm columns packed with stationary phases
ranging from HPLC-size unbonded organosilica (XBridge BEH HILIC, 130Å, 5 µm) to
sub-2-µm amide-bonded organosilica 300Å, 1.7 µm particles (ACQUITY UPLC Glycoprotein
BEH Amide 300Å, 1.7 µm).
Figure 2. Separations of the Glycoprotein Performance Test Standard (RNase A + RNase B
glycoforms) using a Glycoprotein BEH Amide 300Å, 1.7 µm Column versus a BEH Amide, 130Å,
1.7 µm Column. The reported resolution values were calculated using the half-height peak widths of
species 1 and 2 (RNase A and RNase B Man5 glycoforms, respectively). Fluorescence detection at Ex
280 nm and Em 320 nm and a column temperature of 45 °C were employed in this example.
EU
0.00
20.00
40.00
60.00
80.00
100.00
120.00
140.00
160.00
5.00
10.00
15.00
20.00
EU
0.00
20.00
40.00
60.00
80.00
100.00
120.00
140.00
160.00
Minutes
5.00
10.00
15.00
20.00
Rs,(1/2)
21.2
Glycoprotein BEH
Amide
300Å, 1.7 m
Rs,(1/2)
17.1
ACQUITY UPLC
Glycan BEH
Amide
130Å, 1.7 m
1
2
3
4
5
6
Glycoprotein Performance
Test Standard
300Å
130Å
Peak
Species
Rs
Rs
1
RNase A
–
–
2
RNase B (+Man 5)
21.2
17.1
3
RNase B (+Man 6)
3.5
2.8
4
RNase B (+Man 7)
2.7
2.3
5
RNase B (+Man 8)
2.6
2.3
6
RNase B (+Man 9)
3.1
2.4
The progression of HILIC technology culminating
in this new stationary phase is remarkable. The
emerging technology of large molecule HILIC can be
captured by separations of bovine ribonuclease B
(RNase B), a 13 kDa protein comprised of several high
mannose (Man5 to Man9) glycoforms. Figure 1
shows RNase B separated by several different
stationary phases. From bottom to top, increasingly
better separations of RNase B were achieved as
increasingly newer chromatographic technologies
were adopted, from 5 µm to 1.7 µm particles,
from unbonded to amide bonded particles, and
from standard pore diameter (130Å) to wide-pore
diameter (300Å) particles. It is with BEH Amide,
300Å, 1.7 µm particles that RNase B glycoforms are
best separated. The use of a wide-pore stationary
phase plays a significant role in achieving optimal
resolution. This is highlighted in Figure 2 wherein
benchmarking results are presented from the use of
a newly developed test mixture, called Glycoprotein
Performance Test Standard, which contains bovine
RNase B, its corresponding glycoforms and its
aglycosylated isoform (RNase A). Example
separations are provided for this standard wherein
a focused gradient has been used with the wide-pore
(300Å) BEH Amide as well as the standard pore size
(130Å) BEH Amide stationary phase. Notice that the
widepore amide column affords a measurable (24%)
increase in the resolution between the aglycosylated
RNase A isoform and the Man5 glycoform of
RNase B, in addition to sizeable increases in
resolution throughout the separation.
[ 15 ]
Developing High Resolution HILIC Separations of Intact Glycosylated Proteins Using a Wide-Pore Amide-Bonded Stationary Phase
Time
8.00
10.00
12.00
14.00
16.00
18.00
20.00
22.00
5.0e-2
1.0e-1
8.00
10.00
12.00
14.00
16.00
18.00
20.00
22.00
5.0e-2
1.0e-1
BEH Amide
1.7 m
130Å
BEH Amide
1.7 m
300Å
A
B
0.00
10.00
20.00
30.00
0.0
2.0e-1
4.0e-1
0.00
10.00
20.00
30.00
0.0
2.0e-1
4.0e-1
0.00
10.00
20.00
30.00
0.0
2.0e-1
4.0e-1
0.00
10.00
20.00
30.00
0.0
2.0e-1
4.0e-1
0.00
10.00
20.00
30.00
0.0
2.0e-1
4.0e-1
Poor
Retentivity
Poor Recovery
ZIC-HILIC
3.5 m
200Å
Halo
PentaHILIC
2.7 m
90Å
PolyGLYCOPLEX A
3 m
PolyHYDROXYETHYL A
3 m
300 Å
Zorbax 300 HILIC
1.8 m
300 Å
0.00
10.00
20.00
30.00
0.0
2.0e-1
4.0e-1
0.00
10.00
20.00
30.00
0.0
2.0e-1
4.0e-1
Amide-80
3 m
80Å
AccucoreAmide
2.6 m
150Å
BEH Amide
1.7 m
130Å
BEH Amide
1.7 m
300Å
Desirable Retentivity
Poor
Resolution
0.00
10.00
20.00
30.00
0.0
2.0e-1
4.0e-1
0.00
10.00
20.00
30.00
0.0
2.0e-1
4.0e-1
Figure 3. Evaluation of commercially available HILIC columns for intact glycoprotein separations. (A) UV chromatograms obtained for RNase B using 10 different
stationary phases. (B) Zoomed HILIC UV chromatograms for the highest resolution separations.
The significance of these recent developments becomes more apparent when benchmarked against other
commercially available HILIC phases. RNase B separations resulting from an evaluation of 10 different
HILIC stationary phases are shown in Figure 3. It can be seen that 6 out of the 10 evaluated materials showed
undesirable characteristics, including poor recovery and poor retention. It was only with the amide bonded
stationary phases and particle technologies based on 100Å or greater pore diameters that reasonable
separations of RNase B glycoforms could be achieved.
Mobile phase optimization, MS compatibility, and orthogonality to reversed phase
High resolution HILIC separations of protein glycoforms require that mobile phase selection be given
significant consideration. Most HILIC separations have been developed so as to rely on ammonium
salts (formate or acetate) to mitigate significant ionic interactions and to control mobile phase pH.
The suitability of such mobile phase systems to glycoproteins was evaluated using RNase B.
[ 16 ] Developing High Resolution HILIC Separations of Intact Glycosylated Proteins Using a Wide-Pore Amide-Bonded Stationary Phase
Figure 4. Optimization of mobile phase conditions for separations of intact and digested glycoproteins. (A) UV chromatograms obtained for RNase B when using
various mobile phases and a Glycoprotein BEH Amide, 300Å, 1.7 µm, 2.1 x 150 mm Column. (B) Schematic portraying the utility of ion pairing for glycoprotein
HILIC separations. Reduced hydrophilicity imparted via ion pairing with a hydrophobic, strong acid is displayed with shading. [PDB:1RBB]
Figure 4 shows the corresponding RNase B chromatogram obtained when 0.1% TFA is used as the mobile phase
modifier instead of 50 mM ammonium formate or 0.5% formic acid, two mobile phase compositions more
frequently used for HILIC separations.2,7 It is with 0.1% TFA that glycoforms are best resolved. Along with
enhancing glycoform resolution, the TFA-modified mobile phase reduced the retention of RNase B. Together
these observations highlight the significance of acidic, ion pairing mobile phases to being able to achieve high
resolution glycoprotein separations using HILIC. It is proposed that the acidic condition imparted by the TFA
ensures that acidic residues of the protein are fully protonated and thus present in their more hydrophobic state.
In addition, the ion pairing of the TFA counter ion to basic residues, ensures that cationic residues will also be
separated in a more hydrophobic form. In this way, retention of a glycoprotein onto a HILIC phase is
driven primarily by the glycans and a separation more selective to resolving differences in the glycan
modification is achieved.
0.005
0.005
0.015
0.025
8
9
10
11
12
13
14
15
16
17
18
A
280
Time (min)
50 mM Ammonium Formate
pH 4.4
0.5% FA
0.1% TFA
HOOC-
-COO
H
-NH3
+
-N
H
3 +
HO
OC-
+H
3
N-
+
H3N-
F3C-COO-
-
OOC-CF3
F3C-COO-
-
OOC-CF3
-NH3
+
-
OO
O C-CF
H
-N
3 +
+
-
OOC-CF3
+H
3
N-
F3C--COO
CO -
-COO
H
HO
OC-
HOOC-
+
H3N-
F3C-COO-
TFA
Ion Pairing
A
B
Fortunately, TFA-modified mobile phases can be readily coupled to ESI-MS, due to their volatility. This
aspect of the developed HILIC methods enables on-line characterization of the resolved glycoforms and
presents a new option for profiling a sample containing glycosylated protein. To this end, the peaks resolved
from RNase B using a BEH Amide, 300Å, 1.7 µm column were subjected to interrogation by ESI-MS.
[ 17 ]
Developing High Resolution HILIC Separations of Intact Glycosylated Proteins Using a Wide-Pore Amide-Bonded Stationary Phase
Figure 5. HILIC-MS of RNase B. (A) UV (bottom) and TIC (top) chromatograms obtained for RNase B when using a focused gradient and a Glycoprotein BEH Amide,
300Å, 1.7 µm, 2.1 x 150 mm Column. (B) Deconvoluted mass spectra obtained for each labeled peak along with corresponding glycoform identifications.
Figure 5 shows both a UV chromatogram and a corresponding total ion chromatogram (TIC) obtained
when separating RNase B. By summing and deconvoluting (MaxEnt™ 1) the mass spectra obtained for the
six labeled peaks, it was confirmed that RNase B glycoforms were being detected. In fact, the observed
deconvoluted masses support identifications of aglycosylated RNase B (RNase A) along with RNase B
modified by Man5 through Man9.
0
20000
40000
60000
80000
100000
120000
0
0.02
0.04
0.06
0.08
0.1
0.12
5
10
15
20
TIC Intensit
y
A
214
Time (min)
Blank
0E+0
5E+3
1E+4
2E+4
2E+4
3E+4
3E+4
13500
14000
14500
15000
15500
16000
Intensit
y
m/z
1
RNase B
aglycosylated
2
3
4
5
6
1)
2)
+Man5
3)
+Man6
4)
+Man7
5)
+Man8
6)
+Man9
=162 Da
=1217 Da
= 1217 Da, 2HexNAc/5Hex
= 162 Da, 1Hex
MWavg, theoretical
MWavg, observed
RNase B (4 S-S)
13682.3
13682.5
+ Man5
14899.4
14899.7
+ Man6
15061.6
15061.8
+ Man7
15223.7
15223.9
+ Man8
15385.8
15385.9
+ Man9
15548.0
15548.3
A
B
[ 18 ] Developing High Resolution HILIC Separations of Intact Glycosylated Proteins Using a Wide-Pore Amide-Bonded Stationary Phase
Figure 6. Orthogonality of reversed phase with BEH C4, 300Å,
1.7 µm and HILIC with BEH Amide, 300Å, 1.7 µm Columns. (A)
Separation of RNase B (1 µg) using an ACQUITY UPLC Protein
BEH C4, 300Å, 1.7 µm, 2.1 x 150 mm Column. (B) Separation of
RNase B (1 µg) using an ACQUITY UPLC Glycoprotein BEH Amide,
300Å, 1.7 µm, 2.1 x 150 mm Column.
Finally, it should be pointed out that the newly developed stationary phase and the demonstrated
methodologies provide new separation selectivity, one that is orthogonal and complementary to
conventional reversed phase separations. Figure 6A shows that RNase B can, for instance, be separated by
reversed-phase chromatography using a BEH C4, 300Å, 1.7 µm column so as to produce a high resolution
separation of aglyocosylated RNAse B (RNase B) from its glycosylated isoforms. By reversed phase,
however, none of high mannose glycoforms of RNase B can be resolved from one another. In contrast, a BEH
Amide, 300Å, 1.7 µm column yields baseline resolution of each major glycoform (Figure 6B).
0
0.09
0.18
0.27
A
214
0
0.02
0.04
0.06
6
8
10
12
14
16
18
20
A
214
Time (min)
+Man5
+Man6
+Man7
+Man8
+Man9
RNase B
aglycosylated
RNase B
aglycosylated
+Man5
+Man6
+Man7
+Man8
+Man9
BEH Amide
300Å 1.7 m
BEH C4
300Å 1.7 m
RNase A
RNase A
Separation of the heterogeneous glycoforms of an intact mAb
To explore the limits of this new technology, we have investigated the capabilities of resolving the
glycoforms of intact mAbs. Specifically, separations of trastuzumab have been explored. These experiments
required special considerations regarding sample injection, primarily because trastuzumab and numerous
other glycoproteins are not readily soluble in high organic concentrations. In fact, 70–80% ACN is
generally a solution condition that initiates the precipitation of proteins, such as an IgG. Accordingly,
conditions for the optimal injection of aqueous diluents were developed. It has been found that a 2.1 mm I.D.
column can accommodate an injection of aqueous sample up to 1 µL. From a 2 mg/mL aqueous sample of
trastuzumab, appropriate sample mass loads could thus be injected and HILIC separations of the IgG could
be performed. It should be mentioned that high ACN diluents can be used in intact protein HILIC, but care
must be taken to enhance the solubility of the protein sample through either the use of TFA ion pairing at
concentrations between 0.2–1.0%, the combined application of TFA and hexafluoroisopropanol (HFIP), or by
use of co-solvents, such as dimethylsulfoxide (DMSO) (data not shown).
[ 19 ]
Developing High Resolution HILIC Separations of Intact Glycosylated Proteins Using a Wide-Pore Amide-Bonded Stationary Phase
As shown in Figure 7, trastuzumab can indeed be separated into multiple chromatographic peaks using a
BEH Amide, 300Å, 1.7 µm column and an injection from a simple 100% aqueous diluent. However, at the
backpressures produced from just a 150 mm length column, a noticeably tailing profile was observed. MS
analysis indicated that the first set of peaks could be accurately assigned as the G0F/G0F, G0F/G1F, G1F/G1F,
and G1F/G2F glycoforms of intact trastuzumab. An intact IgG is a dimeric structure, with a minimum of two
N-glycan sites on two heavy chains, explaining the observation of combinatorially formed glycoforms. This
is consistent with observations by intact mass analysis of IgGs.10 The tailing component of the chromatographic
profile was in contrast found by MS to correspond to multiple, co-eluting trastuzumab glycoforms. With this
result, we proposed that on-column aggregation was occurring and that increased column pressure could be a
solution to HILIC of intact immunoglobulins, specifically since it had previously been reported that ultrahigh
pressures can be beneficial to limiting carryover and ghosting during reversed phase of intact proteins.11 The
effects of introducing additional column pressure was investigated by means of introducing varying lengths of
narrow I.D., post-column PEEK tubing. Figure 7 (darker traces) displays the effects of introducing increasingly
higher column pressure. By doubling the column pressure so that trastuzumab would elute under conditions
of approximately 7,500 psi, the putative, aggregate peaks in the chromatographic profile were eliminated. It
is encouraging that under these conditions the resulting chromatographic profile is represented by 5 major
glycoforms, which again is consistent with ESI-MS of intact trastuzumab.10 It is interesting to additionally
note that retention decreases as column pressure increases. This is a phenomenon that has been described
previously for HILIC separations of monosaccharides.12 It has been proposed that increasingly higher pressures
result in less coordination of water to the analyte and in turn reduced retention, an opposite effect to that
observed during reversed phase chromatography.12
0
0.1
0.2
10
15
20
A
21
4
Time (min)
Increasing
Pressure
3200 psi
4500 psi
7300 psi
*Pressure at retention time of the mAb
Increasing Column Pressure
Minimizes On-Column Aggregation
N-Linked Glycans
Fab
Fc
2 N-Glycans
Figure 7. Effect of column pressure on the HILIC separation of an IgG. Trastuzumab (1 µg) was separated on Glycoprotein BEH Amide, 300Å, 1.7 µm, 2.1 x 150 mm
Column with and without flow restriction. [PDB:1IGT]
[ 20 ] Developing High Resolution HILIC Separations of Intact Glycosylated Proteins Using a Wide-Pore Amide-Bonded Stationary Phase
Given that intact IgGs benefit from separations at
ultrahigh pressures, we pursued separations based on
the use of two BEH amide 300Å, 1.7 µm,
2.1 x 150 mm columns coupled with a low volume,
high pressure column connector. The separation for
intact trastuzumab obtained with these coupled
columns is displayed in Figure 8, along with
extracted ion chromatograms that provide evidence
to achieving separations of the glycoforms. This
300 mm configuration provided the requisite column
pressures for an optimal HILIC separation and
additionally produced greater resolution between
glycoforms. Clearly, additional theoretical plates
are therefore advantageous during HILIC of even very
high molecular weight species, which supports the
significance of partitioning for such separations.
An LC method for glycan occupancy
A UPLC HILIC separation of an intact IgG can be used
for more than just an attempt to separate individual
glycoforms. Equally interesting is the use of these
new separation capabilities to resolve information
about glycan occupancy. To this end, we evaluated
the capabilities of the BEH Amide, 300Å column
to assess the glycan occupancy of an IgG. This was
exemplified by a study of reaction products resulting
from various PNGase F deglycosylation treatments.
Using an elevated 80 °C column temperature, TFA
ion pairing, and an HFIP mobile phase additive, we
have been successful in enhancing the solubility
of IgGs and collapsing the fine structure otherwise
captured for the individual, heterogenous intact
IgG glycoforms (i.e. G0F/G0F versus G0F/G1F).
Figure 9 presents HILIC fluorescence chromatograms
resulting from such a separation of native Intact
mAb Mass Check Standard (a murine IgG1 mAb) and
its partially as well as completely deglycosylated
isoforms. As can be seen, HILIC fluorescence
profiles for these three samples are dramatically
different. On-line mass spectrometric detection
has confirmed that the peaks in these profiles
correspond to different states of glycan occupancy.
0
0.1
0.2
15
16
17
18
19
20
21
22
23
24
25
A
21
4
Time (min)
0
200
400
600
800
1000
1200
15
16
17
18
19
20
21
22
23
24
25
XI
C
In
te
ns
it
y
Time (min)
(G0F)2
3023 ±0.6 m/z, 49+
G0F/G1F
3026 ±0.6 m/z, 49+ (G1F)2 or
G0F/G2F
3029 ±0.6 m/z, 49+
G1F/G2F
3033 ±0.6 m/z, 49+
(G2F)2
3062 ±0.6 m/z, 49+
Figure 8. Separation of intact trastuzumab glycoforms using coupled ACQUITY UPLC Glycoprotein
BEH Amide, 300Å, 1.7 µm, 2.1 x 150 mm Columns. A UV chromatogram and extracted ion
chromatograms for each of the major heterogenous glycoforms of trastuzumab are displayed.
The column pressure at the retention time of the mAb was approximately 7,000 psi.
Figure 9. Assaying glycan occupancy and deglycosylation by intact protein HILIC-FLR-MS. HILIC
fluorescence profiles obtained for three different samples are shown: (A) native, (B) partially
deglycosylated, and (C) completely deglycosylated Intact mAb Mass Check Standard. Samples
of this mAb (1.5 µg) were separated using two coupled Glycoprotein BEH Amide, 300Å, 1.7 µm,
2.1 x 150 mm Columns. HILIC fluorescence profiles of partially deglycosylated Intact mAb Mass
Check Standard using a (D) ACQUITY UPLC Glycan BEH Amide, 130Å, 1.7 µm, 2.1 x 150 mm
Column versus a (E) Glycoprotein BEH Amide, 300Å, 1.7 µm, 2.1 x 150 mm Column.
145000
150000
145000
150000
146786
146952
145000
150000
148240
148560
Intact
-1 Glycan
-2 Glycans
PNGase F
148.4 kDa
145.3 kDa
146.8 kDa
9
14 min
PNGase F
Untreated
Native mAb
Partially PNGase F
Deglycosylated mAb
Completely PNGase F
Deglycosylated
A
B
C
Time
8.00
10.00
12.00
14.00
EU x 10
e4
0.000
1000000.063
2000000.125
3000000.250
4000000.250
5000000.500
6000000.500
7000000.500
D
E
Glycan BEH Amide
130Å, 1.7 m
Glycoprotein BEH Amide
300Å, 1.7 m
[ 21 ]
Developing High Resolution HILIC Separations of Intact Glycosylated Proteins Using a Wide-Pore Amide-Bonded Stationary Phase
Waters Corporation
34 Maple Street
Milford, MA 01757 U.S.A.
T: 1 508 478 2000
F: 1 508 872 1990
www.waters.com
Waters, The Science of What’s Possible, ACQUITY UPLC, Oasis, and Empower are registered trademarks of Waters Corporation.
All other trademarks are the property of their respective owners.
©2015 Waters Corporation. Produced in the U.S.A. April 2015 720005380EN AG-PDF
The most strongly retained species, represented by the native mAb sample,
corresponds to the doubly (fully) glycosylated form of the intact mAb. The
partially deglycosylated mAb sample meanwhile yielded several additional peaks
with lower HILIC retention, two of which with corresponding detected molecular
weights that are indicative of once deglycosylated and fully deglycosylated mAb
species and a third with a corresponding detected molecular weight consistent
with PNGase F. In contrast, the completely deglycosylated mAb sample presented
a homogenous fluorescence profile along with an observed molecular weight
for the mAb that is in agreement with the predicted molecular weight of the
deglycosylated mAb (145.3 kDa). It is worth noting that when attempting to use
the BEH Amide, 130Å, 1.7 µm stationary phase, none of the above peaks could
be resolved (Figures 9D and 9E). So indeed, the widepore phase facilitates the
development of previously unobtainable separations.
In our hands, the above assay has been used to develop rapid enzymatic
deglycosylation protocols.3 However, it is natural to suggest that these same
methods could be applied to measure the glycan occupancy of an intact
therapeutic mAb, in which case the relative abundance of aglycosylated forms
(-2 and -1 N-glycans) could potentially be monitored by fluorescence and
corroborated by LC-MS.
C O N C LU S IO NS
HILIC of small molecules has garnered wide-spread attention and use. In
contrast, the application of the technique to large biomolecule separations has
been limited. With the development of the above mentioned amide-bonded,
wide-pore HILIC stationary phase and corresponding methods, it is now
possible to resolve the glycoforms of intact glycosylated proteins, as has
been exemplified by the resolution of the heterogenous glycoforms on intact
trastuzumab. Alternatively, the described techniques can be applied to studies
of glycan occupancy. Just as reversed phase separations are employed for
resolving protein isoforms that have varying hydrophobicities, HILIC separations
with BEH Amide 300Å can be explored for resolving protein isoforms that
exhibit varying hydrophilicities, such as isoforms differing with respect to
glycan occupancy. With the availability of these new separation capabilities,
it will be possible to perform more detailed characterization of intact
glycoproteins, whether by means of combining HILIC with optical detection or
with ESI-MS.
References
1. Ahn, J.; Yu, Y. Q.; Gilar, M., UPLC-FLR Method Development
of 2-AB Labeled Glycan Separation in Hydrophilic
Interaction Chromatography (HILIC). Waters Appication Note
720003238EN 2010.
2. Ahn, J.; Bones, J.; Yu, Y. Q.; Rudd, P. M.; Gilar, M., Separation
of 2-aminobenzamide labeled glycans using hydrophilic
interaction chromatography columns packed with 1.7 microm
sorbent. J Chromatogr B Analyt Technol Biomed Life Sci 2010,
878 (3–4), 403–8.
3. Lauber, M. A.; Brousmiche, D. W.; Hua, Z.; Koza, S. M.; Guthrie,
E.; Magnelli, P.; Taron, C. H.; Fountain, K. J., Rapid Preparation
of Released N-Glycans for HILIC Analysis Using a Novel
Fluorescence and MS-Active Labeling Reagent. Waters
Application Note 720005275EN 2015.
4. Wang, B.; Tsybovsky, Y.; Palczewski, K.; Chance, M. R.,
Reliable Determination of Site-Specific In Vivo Protein
N-Glycosylation Based on Collision-Induced MS/MS and
Chromatographic Retention Time. J Am Soc Mass Spectrom
2014, 25 (5), 729–41.
5. Shah, B.; Jiang, X. G.; Chen, L.; Zhang, Z., LC-MS/MS peptide
mapping with automated data processing for routine profiling
of N-glycans in immunoglobulins. J Am Soc Mass Spectrom
2014, 25 (6), 999–1011.
6. Houel, S.; Hilliard, M.; Yu, Y. Q.; McLoughlin, N.; Martin, S. M.;
Rudd, P. M.; Williams, J. P.; Chen, W., N- and O-glycosylation
analysis of etanercept using liquid chromatography and
quadrupole time-of-flight mass spectrometry equipped with
electron-transfer dissociation functionality. Anal Chem
2014, 86 (1), 576–84.
7. Gilar, M.; Yu, Y. Q.; Ahn, J.; Xie, H.; Han, H.; Ying, W.; Qian,
X., Characterization of glycoprotein digests with hydrophilic
interaction chromatography and mass spectrometry.
Anal Biochem 2011, 417 (1), 80–8.
8. O’Gara, J. E.; Wyndham, K. D., Porous Hybrid
Organic-Inorganic Particles in Reversed-Phase Liquid
Chromatography.
J Liq Chromatogr Relat Technol. 2006, 29, 1025–1045.
9. Alpert, A. J., Hydrophilic-interaction chromatography for
the separation of peptides, nucleic acids and other polar
compounds. J Chromatogr 1990, 499, 177–96.
10. Xie, H.; Chakraborty, A.; Ahn, J.; Yu, Y. Q.; Dakshinamoorthy,
D. P.; Gilar, M.; Chen, W.; Skilton, S. J.; Mazzeo, J. R., Rapid
comparison of a candidate biosimilar to an innovator
monoclonal antibody with advanced liquid chromatography
and mass spectrometry technologies. MAbs 2010, 2 (4).
11. Eschelbach, J. W.; Jorgenson, J. W., Improved protein
recovery in reversed-phase liquid chromatography by the use
of ultrahigh pressures. Anal Chem 2006, 78 (5), 1697–706.
12. Neue, U. D.; Hudalla, C. J.; Iraneta, P. C., Influence of
pressure on the retention of sugars in hydrophilic interaction
chromatography. J Sep Sci 2010, 33 (6–7), 838-40.
[ 22 ] Developing High Resolution HILIC Separations of Intact Glycosylated Proteins Using a Wide-Pore Amide-Bonded Stationary Phase
Authors: Matthew A. Lauber and Stephan M. Koza
Figure 1. HILIC analysis of intact trastuzumab. (A) HILIC fluorescence chromatograms for native
and partially deglycosylated trastuzumab. (B) Raw and deconvoluted ESI mass spectra for
the major species resolved in the native trastuzumab sample. (C) Raw and deconvoluted ESI
mass spectra for the low abundance species resolved in the native trastuzumab sample. HILIC
was performed on 1 µL aqueous injections of 5 mg/mL trastuzumab using an ACQUITY UPLC
Glycoprotein BEH Amide 300Å 1.7 µm Column heated to 80˚C. Sample was separated at a flow
rate of 0.4 mL/min with aqueous (A) and acetonitrile (B) mobile phases containing 0.1% (v/v) TFA
and 0.3% (v/v) HFIP. A linear gradient was applied as follows: hold at 20% A for 0.5 min, 20 to
25% in 0.5 min, 25% to 40% A in 9 min, 40% to 100% A in 0.5 min, hold at 100% A for 0.5
min, 100% to 20% A in 0.5 min, and hold at 20% A for 3.5 min for re-equilibration. Eluting
proteins were detected by intrinsic fluorescence (Ex 280 nm/Em 320 nm).
Species
%
2 N-Glycans
68.9
1 N-Glycan
26.6
0 N-Glycans
4.5
Species
%
2 N-Glycans
75.4
1 N-Glycan
21.5
0 N-Glycans
3.1
Species
%
2 N-Glycans
99.0
1 N-Glycan
1.0
0 N-Glycans
0.0
Trastuzumab
Native
5.0
10.0 min
2 N-Glycans
1 N-Glycan
2 N-Glycans
1 N-Glycan
0 N-Glycans
2 N-Glycans
1 N-Glycan
0 N-Glycans
Trastuzumab
PNGase F
3 min 23˚C
Trastuzumab
PNGase F
3 min 23˚C
1 min 50˚C
A
C
B
Trastuzumab (2 N-Glycans)
Trastuzumab (1 N-Glycan)
FA2/FA2
148,057 Da
FA2/FA2G1
148221 Da
FA2
146,614 Da
FA2G1
146,773 Da
1,000
5,000
m/z
1,000
5,000
m/z
145
149
kDa
145
149
kDa
Intrinsic Fluorescence Detection
G OA L
To demonstrate the use of HILIC with an
ACQUITY UPLC® Glycoprotein BEH Amide
300Å Column to assay the glycan occupancy
of intact mAbs.
BAC KG RO U N D
Monoclonal antibodies (mAbs) have emerged
as some of the most important therapeutics on
the market. These mAbs that are prescribed
for therapeutic use are most often expressed
from eukaryotic cell lines, such as CHO, and, as
a result, are N-glycosylated at two consensus
site asparagine residues in the Fc portion of
their heavy chains. Since glycosylation can be a
measure of efficacy, safety and manufacturing
conditions, it is often critical to characterize
and routinely monitor the N-glycan profile of a
mAb drug substance.1-2 The recent introduction
of the GlycoWorks™ RapiFluor-MS™ N-Glycan
Kit has made it significantly easier to perform
highly detailed HILIC-based, released
N-glycan analyses and to thereby elucidate the
heterogeneity of N-glycosylation.3-4 However,
it is also critical to determine the extent to
which the asparagine linkage sites are occupied
with N-glycans, particularly since incomplete
glycosylation dramatically changes the
effector functions of a mAb.5 Traditionally,
this assessment has been performed by a
sized-based Capillary Electrophoresis-Sodium
Dodecyl Sulfate (CE-SDS) separation of the
heavy chains resulting from reduction of a
mAb.6,7 To instead directly assess glycan
Unprecedented Hydrophilic Interaction
Chromatography (HILIC) separations of intact mAb
glycan occupancy variants using the ACQUITY UPLC
Glycoprotein BEH Amide 300A 1.7µm Column.
Measuring the Glycan Occupancy
of Intact mAbs using HILIC and Detection
by Intrinsic Fluorescence
[ 23 ]
A Single Column Chemistry for
Multiple Glycoprotein Analyses
The A
CQUIT
Y UPLC
®
Glycopr
ote
in B
EH Am
ide, 300Å, 1.7 µm C
olum
n offers:
■
Optim
ized wid
e-pore, HI
LIC sta
tiona
ry p
hase fo
r resolv
ing g
lycoforms fr
om
inta
ct or diges
ted glyco
prote
ins
■
Gener
atio
n of dom
ain s
pecific g
lycan l
inka
ges wit
h or w
ithou
t MS
■
Elucidatio
n of site s
pecific g
lycan occu
pancy
■
High reso
lutio
n glyco
peptid
e mappi
ng wit
hout l
imita
tions due to p
eptide/
glyc
an
size o
r com
positio
n
■
Impr
oved reso
lutio
n in sepa
ratio
ns of large, re
leased
N-g
lycans (
EPO, F
acto
r IX
)
■
QC tes
ted wit
h Water
s Glycopr
ote
in P
erform
ance T
est S
tanda
rd
©2015 W
ater
s Cor
pora
tion. W
ater
s, The S
cience of W
hat’s P
ossib
le,
and A
CQUIT
Y UPLC a
re registered tr
adem
arks of W
ater
s Cor
pora
tion.
wat
ers
.co
m/gl
yco
pro
tein
NEW
A Single Column Chemistry for
Multiple Glycoprotein Analyses
he ACQU
ITY U
PLC
®
Glycopr
ote
in B
EH Am
ide, 300Å, 1.7 µm C
olum
n offers:
■
Optim
ized wid
e-pore, HI
LIC sta
tiona
ry p
hase fo
r resolv
ing g
lycoforms fr
om
inta
ct or diges
ted glyco
prote
ins
■
Gener
atio
n of dom
ain s
pecific g
lycan l
inka
ges wit
h or w
ithou
t MS
■
Elucidatio
n of site s
pecific g
lycan occu
pancy
■
High reso
lutio
n glyco
peptid
e mappi
ng wit
hout l
imita
tions due to p
eptide/
glyc
an
size o
r com
positio
n
■
Impr
oved reso
lutio
n in sepa
ratio
ns of large, re
leased
N-g
lycans (
EPO, F
acto
r IX
)
■
QC tes
ted wit
h Water
s Glycopr
ote
in P
erform
ance T
est S
tanda
rd
©2015 W
ater
s Cor
pora
tion. W
ater
s, The S
cience of W
hat’s P
ossib
le,
and A
CQUIT
Y UPLC a
re registered tr
adem
arks of W
ater
s Cor
pora
tion.
wat
ers
.co
m/gl
yco
pro
tein
Figure 2. HILIC analysis of four intact mAbs. Bevacizumab, NIST IgG1κ candidate reference
material, and trastuzumab were injected without any sample preparation except for dilution to
5 mg/mL from their respective formulations. Prior to HILIC analysis, Intact mAb Mass Check
Standard (1 mg) was reconstituted in 500 µL of 6M guanidine HCl, filtered via 3 passes with a
100KDa MWCO polyethersulfone membrane (GE Healthcare Life Sciences, Vivaspin 500), and
concentrated to 5 mg/mL.
Bevacizumab
IgG1
Reference
Material
Trastuzumab
5.0
10.0 5.0
10.0 min
Intrinsic Fluorescence Detection
1 N-Glycan
4.9%
1 N-Glycan
0.7%
1 N-Glycan
1.3%
1 N-Glycan
1.0%
Murine IgG1
(Intact mAb
Mass Check Std)
occupancy variants for intact mAbs, we have
developed an LC separation based on HILIC. To
achieve these unprecedented separations, a new,
purposefully designed HILIC column was employed.
This new column, the ACQUITY UPLC Glycoprotein
BEH Amide 300Å 1.7 µm Column, contains a wide-
pore amide bonded stationary phase that ensures
that mAb species have access to the porous network
of the stationary phase and are not significantly
impaired by restricted diffusion.8
T H E S O LU T IO N
A high-throughput, high resolution HILIC separation
was established for intact mAbs using a 2.1 x
150 mm wide-pore BEH amide column, a 0.4
mL/min flow rate, and a column temperature of
80 ˚C. In addition, two mobile phase additives,
0.1% trifluoroacetic acid (TFA) and 0.3%
hexafluoroisopropanol (HFIP) were employed to
improve the solubility of intact mAbs in the high
organic, initial mobile phase conditions of the
HILIC gradient. To enhance the sensitivity of this
LC method, the intact proteins were detected by
means of their intrinsic fluorescence. Excitation
and emission wavelengths of 280 and 320 nm
were found to provide optimal signal-to-noise
and consistently flat chromatographic baselines
that are desirable for peak integration. A
representative set of chromatograms resulting
from this 15 minute LC method is shown in Figure
1A. Three chromatograms are displayed. The top
chromatogram shows trastuzumab as injected
from a dilution of its formulation, while the other
two chromatograms show samples of trastuzumab
after being subjected to partial PNGase F
deglycosylation. Deglycosylated samples of
trastuzumab clearly showed three distinct peaks in
their HILIC profiles, as was predicted if the glycan
occupancy variants of a mAb were to be resolved.
The unadulterated sample of trastuzumab contained
measurable levels of only the fully occupied and
singly occupied forms (1%) as confirmed by online
mass analysis. The deconvoluted mass spectrum
corresponding to the main LC peak exhibited several
masses, such as 148,057 Da and 148,221 Da,
that are within 2 Da of the theoretical molecular weights for the predominant
glycoforms of trastuzumab (Figure 1B).9 In contrast, the deconvoluted
mass spectrum for the smaller, less strongly retained peak showed reduced
heterogeneity and masses that were lighter by approximately 1000 to 2000
Da, consistent with the single occupancy form (Figure 2B). Fully aglycosylated
species were not detected in this sample. An interesting observation from these
data is that the levels of fully deglycosylated forms are lower than would be
predicted if both sites were deglycosylated at the same rate. In these examples,
the observed level of fully deglycosylated forms is approximately one-third
lower than would be predicted. This may indicate that either the digestion rate
for one of the N-glycans is slower than the other, or that upon removal of the
first N-glycan, the digestion rate of the remaining N-glycan is reduced.
To assess the applicability of this technique to other mAbs, we analyzed three
additional IgG samples. Results on trastuzumab, bevacizumab, a candidate
IgG1
κ reference material, and a murine IgG1 (Intact mAb Mass Check Standard,
p/n 186006552) are shown in Figure 2. Integrations on the HILIC-fluorescence
chromatograms indicated that these samples contain 1 N-glycan (incomplete
glycosylation) variants at relative abundances ranging from 0.7 to 4.9%. These
observations suggest that this HILIC method could be an attractive technique for
assaying incomplete glycosylation of mAbs down to levels well below 1% for
the 1 N-glycan form. Detection limits for the 0 N-glycan form may perhaps be
even lower. Of these samples, bevacizumab might be predicted to have the
highest abundance of the 0 N-glycan form. Indeed, the bevacizumab profile
presented a peak, albeit very minor, with a retention time consistent with
a 0 N-glycan species, the area of which would contribute to the overall profile
[ 24 ] Measuring the Glycan Occupancy of Intact mAbs Using HILIC and Detection by Intrinsic Fluorescence
Waters Corporation
34 Maple Street
Milford, MA 01757 U.S.A.
T: 1 508 478 2000
F: 1 508 872 1990
www.waters.com
A Single Column Chemistry for
Multiple Glycoprotein Analyses
The A
CQUIT
Y UPLC
®
Glycopr
ote
in B
EH Am
ide, 300Å, 1.7 µm C
olum
n offers:
■
Optim
ized wid
e-pore, HI
LIC sta
tiona
ry p
hase fo
r resolv
ing g
lycoforms fr
om
inta
ct or diges
ted glyco
prote
ins
■
Gener
atio
n of dom
ain s
pecific g
lycan l
inka
ges wit
h or w
ithou
t MS
■
Elucidatio
n of site s
pecific g
lycan occu
pancy
■
High reso
lutio
n glyco
peptid
e mappi
ng wit
hout l
imita
tions due to p
eptide/
glyc
an
size o
r com
positio
n
■
Impr
oved reso
lutio
n in sepa
ratio
ns of large, re
leased
N-g
lycans (
EPO, F
acto
r IX
)
■
QC tes
ted wit
h Water
s Glycopr
ote
in P
erform
ance T
est S
tanda
rd
©2015 W
ater
s Cor
pora
tion. W
ater
s, The S
cience of W
hat’s P
ossib
le,
and A
CQUIT
Y UPLC a
re registered tr
adem
arks of W
ater
s Cor
pora
tion.
wat
ers
.co
m/gl
yco
pro
tein
NEW
at a level of only 0.05%. Such an observation
suggests that the 0 N-glycan level is lower
than would be statistically predicted; however,
rigorous determination of the quantitative limits
of this analysis and the identity of this putative
0 N-glycan peak would need to be evaluated to
confirm that this is indeed the case.
This strategy for measuring mAb glycan occupancy
is most appealing in that it requires minimal,
if any, sample preparation. We have observed
that some interferences can be encountered that
are due to partially fragmented and/or reduced
mAb species. In which case, as with the Intact
mAb Mass Check Standard, a simple, centrifugal
filtration clean-up step with a 100kDa MWCO
polyethersulfone membrane was sufficient to
minimize such interferences. Given the unique
selectivity of the HILIC separation, it should also
be possible to address potential interferences
by performing offline or online 2D-LC, wherein
a size exclusion or reversed phase separation
could be coupled to the wide-pore amide HILIC
separation. Future investigations could also
include reducing glycan heterogeneity (via a
sialidase or other exoglycosidase) to allow the mAb
glycan occupancy variants to be more discretely
resolved. Similarly, there is an opportunity to use
this separation in combination with enzymes that
generate Fc subunits.
Waters, The Science of What’s Possible, and ACQUITY UPLC are registered trademarks of Waters Corporation. GlycoWorks and
RapiFluor-MS are trademarks of Waters Corporation. All other trademarks are the property of their respective owners.
©2015 Waters Corporation. Produced in the U.S.A. June 2015 720005435EN AO-PDF
S UMM A RY
In addition to profiling the heterogeneity of glycosylation, it is also critical
to assay glycan occupancy. Here, we demonstrate that a Glycoprotein BEH
Amide column, purposefully designed for large molecule HILIC separations,
can be used to directly quantify incomplete N-glycan occupancy in intact
mAb samples. Unlike a conventional CE-SDS separation of reduced mAbs, this
technique provides a non-inferred assessment on the nature (i.e. 1 N-glycan
versus 0 N-glycans) of glycan occupancy for the intact mAb. The proposed HILIC
methodology is also MS-compatible, making it possible to readily confirm the
assignments of observed peaks.
References
1. Beck, A.; Wagner-Rousset, E.; Ayoub, D.; Van Dorsselaer, A.; Sanglier-Cianferani, S.,
Characterization of therapeutic antibodies and related products. Anal Chem 2013, 85 (2),
715–36.
2. Dalziel, M.; Crispin, M.; Scanlan, C. N.; Zitzmann, N.; Dwek, R. A., Emerging principles for the
therapeutic exploitation of glycosylation. Science 2014, 343 (6166), 1235681.
3. Lauber, M. A.; Yu, Y. Q.; Brousmiche, D. W.; Hua, Z.; Koza, S. M.; Magnelli, P.; Guthrie, E.; Taron,
C. H.; Fountain, K. J., Rapid Preparation of Released N-Glycans for HILIC Analysis Using a
Labeling Reagent that Facilitates Sensitive Fluorescence and ESI-MS Detection. Anal Chem
2015, DOI: 10.1021/acs.analchem.5b00758.
4. Lauber, M. A.; Koza, S. M., Enhancing the Peak Capacity of High Molecular Weight N-Glycan
HILIC Separations with a Wide-Pore Amide Bonded Stationary Phase. Waters Tech Brief
720005381EN 2015.
5. Beck, A.; Wagner-Rousset, E.; Ayoub, D.; Van Dorsselar, A.; Sanglier-Cianferani, S.,
Characterization of Therapeutic Antibodies and Related Products. Analytical Chemistry 2013,
(85), 715–736.
6. Chen, G.; Ha, S.; Rustandi, R. R., Characterization of glycoprotein biopharmaceutical products
by Caliper LC90 CE-SDS gel technology. Methods Mol Biol 2013, 988, 199–209.
7. Lacher, N. A.; Roberts, R. K.; He, Y.; Cargill, H.; Kearns, K. M.; Holovics, H.; Ruesch, M. N.,
Development, validation, and implementation of capillary gel electrophoresis as a replacement
for SDS-PAGE for purity analysis of IgG2 mAbs. Journal of Separation Science 2010, 33 (2),
218–227.
8. Renkin, E. M., J. Gen. Physio. 1954, (38), 225.
9. Damen, C. W.; Chen, W.; Chakraborty, A. B.; van Oosterhout, M.; Mazzeo, J. R.; Gebler, J. C.;
Schellens, J. H.; Rosing, H.; Beijnen, J. H., Electrospray ionization quadrupole ion-mobility
time-of-flight mass spectrometry as a tool to distinguish the lot-to-lot heterogeneity in
N-glycosylation profile of the therapeutic monoclonal antibody trastuzumab. J Am Soc Mass
Spectrom 2009, 20 (11), 2021-33.
[ 25 ]
Measuring the Glycan Occupancy of Intact mAbs Using HILIC and Detection by Intrinsic Fluorescence
GOAL
To demonstrate the capabilities of the integrated
UPLC®-MS analysis of an intact monoclonal anti-
body with a comprehensive platform for accurate
mass measurement, data processing, and reporting
with UNIFI® Scientific Information System.
BACKGROUND
The growing biotherapeutic pipeline
means that the efficient characterization
of monoclonal antibodies (mAb) is of
growing importance, both to regulatory
authorities and to pharmaceutical
companies. Being able to perform
acquisition and processing within
the same platform, complete with an
audit trail, is an important goal for
regulated environments.
Accurately identifying post-translational
modifications such as protein
glycosylation is required as part
of guidelines as they play several key
roles in biological systems. Fast and
accurate analysis of the glycoproteins
is required in order to ensure the safety
and efficacy of the biotherapeutic.
The ACQUITY UPLC® H-Class Bio’s
high-resolution bioseparations combined
with high mass accuracy mass spectrometry
detection with the Xevo® G2 Tof provides
routine UPLC-MS applications for
biopharmaceutical laboratories.
This UNIFI-based platform addresses previous
limitations with a comprehensively integrated platform
for data acquisition by chromatography and mass
spectrometry, with automated reporting.
Intact protein characterization:
Waters Biopharmaceutical
System Solution with UNIFI
Figure 1. UNIFI’s intact protein analysis report
of an intact mAb, Trastuzumab. The report
shows, from top to bottom, the integrated
chromatographic peak, the charge state
distribution, the continuum deconvoluted data,
the integrated and processed data, and finally
the summary table assigning major glycoforms
according to mass.
[ 26 ]
Waters Corporation
34 Maple Street
Milford, MA 01757 U.S.A.
T: 1 508 478 2000
F: 1 508 872 1990
www.waters.com
Waters, The Science of What’s Possible, ACQUITY UPLC, Xevo, UNIFI, and UPLC are registered trademarks of
Waters Corporation. All other trademarks are the property of their respective owners.
©2011–2016 Waters Corporation. Produced in the U.S.A. January 2016 720003843EN LM-PDF
The intact protein analysis report demonstrates the report objects, which can be
entirely configured by the user: TIC summarized chromatogram; raw, deconvoluted,
and centroid mass spectra; and tabulated summary of the interpreted LC-MS data
(Figure 1). This detailed view shows an example of a deconvoluted spectrum within
a specified mass range and parameter settings defined in the method. Deconvolution
reveals several core glycosylated species which match the number of glucose
residues and level of fucosylation. Another report object is a table with mass
measurement of the intact mAb and accurately assigned mAb glycan variants
(Figure 2). Mass errors were reported for each Trastuzumab MS peak with
a corresponding retention time entry from the TIC chromatogram.
Such an integrated LC-MS approach provides the user flexibility to work with both
raw and processed data followed by quick and efficient data management.
SUMMARY
The capabilities of the Biopharmaceutical Platform Solution with UNIFI have been
successfully demonstrated with the example of an intact biotherapeutic mAb.
Modern instrumentation and evolving analytical techniques extend the limits
of the biopharmaceutical industry and consequently impose strict control of
manufacturing processes.
Highly efficient and cost-effective integrated UPLC-MS approaches with the UNIFI
Scientific Information System for data processing and reporting satisfies regulatory
requirements and facilitates intact protein characterization. This technology covers
the range from detailed structural protein characterization to sophisticated data
management with UPLC-MS platforms.
There is a large set of data generated during each
mAb analysis requiring interpretation of a variety of
glycosylated forms and comprehensive characterization
of the final product. This step sets productivity limits
to otherwise high-throughput procedures and hinders
automation of the process.
The UNIFI-based platform addresses these limitations
with a comprehensively integrated platform for data
acquisition by chromatography and mass spectrometry,
with automated reporting.
THE SOLUTION
To solve the problem of time-consuming data analysis
and facilitate data processing of therapeutic mAb,
the Biopharmaceutical Platform Solution with UNIFI
was configured for the study of intact proteins.
This represents a holistic approach of UPLC-MS data
acquisition followed by automatic processing and
annotation of the data in a high-throughput manner,
which are further exported for data management.
UPLC-MS analysis of the mAb Trastuzumab was
performed automatically. Aqueous solutions of 0.1%
FA and 0.1% FA solution in acetonitrile were used as
eluents A and B, respectively. Column temperature
set to 80 °C is critical for successful chromatographic
separation. The system included an ACQUITY UPLC
H-Class Bio, an ACQUITY UPLC Protein BEH C4
Column, and a Xevo G2 Tof. The UNIFI Scientific
Information System for acquisition, data processing,
and reporting completes this comprehensive
Biopharmaceutical Platform Solution.
Figure 2. Zoomed in view of the table in Figure 1 summarizing mass measurement
of the intact mAb and accurately assigned mAb glycan variants.
[ 27 ]
Intact Protein Characterization: Waters Biopharmaceutical System Solution with UNIFI
WAT E RS SO LU T IONS
Biopharmaceutical Platform Solution
with UNIFI
ACQUITY UPLC® H-Class Bio System
Xevo® G2 QTof
ACQUITY UPLC Protein BEH C4 Column
ACQUITY UPLC Tunable UV Detector
K E Y WO R DS
Intact mass analysis, mAb,
biotherapeutic characterization
A P P LIC AT ION BEN E FIT S
The Biopharmaceutical Platform Solution
with UNIFI® enables a fully integrated workflow
for intact mass analysis, including acquisition,
processing, and reporting, for organizations
in early development and those operating
under regulatory compliant environments.
The ability to automate and standardize
intact mass analysis enables laboratories
to deploy their scientific resources with
greater efficiency and effect.
INT RO DUC T ION
Intact mass analysis is a rapid and convenient method for confirming protein
identity and profiling product-related variants. In conjunction with other
analytical techniques, such as peptide mapping and released glycan analysis,
intact mass analysis can help determine if the biomolecule had been correctly
cloned, expressed, purified, and formulated during the biopharmaceutical drug
development process.
Intact mass analysis can provide a semi-quantitative view of product
heterogeneity and is often employed to determine relative composition of product
glycoforms. As a lot release test, intact protein mass analysis often provides a
quick identity test using the mass of a major variant, sometimes in conjunction
with a purity test with defined product variation for peaks corresponding
to variants displaying critical product attributes. Demonstration of process
consistency through such comparability exercises is critical to obtain initial
regulatory approval and for later process improvement studies.
Data processing and report generation often become productivity-limiting
tasks for organizations responsible for biotherapeutic protein characterization
and analysis. It is still common for LC-MS intact protein data to be manually
processed, an inefficient process that lacks standardization and is prone to human
error. Further inefficiency and sources of error result from scientists having to
reformat results into graphical and tabular formats suitable for communicating
information to their organizations.
The ability to automate and standardize data acquisition, processing, and reporting
for intact mass analysis allows laboratories to deploy their scientific resources
with greater efficiency and effect. The Waters® UNIFI Scientific Information
System enables these benefits, as well as regulatory compliance, to be realized
throughout discovery, development, and quality management organizations.
In this application note, an integrated and compliant-ready solution for intact
mass analysis is described. The combination of UPLC® separations, optimized
application-tested protein column chemistries, the Xevo G2-S QTof for mass
detection, all used under control of the UNIFI Scientific Information System,
achieves the goal of total workflow automation and standardization.
Streamlining Compliant and Non-Compliant Intact Mass Analysis of
Biotherapeutic mAbs with the Biopharmaceutical Platform Solution with UNIFI
Henry Shion and Weibin Chen
Waters Corporation, Milford, MA, USA
[ 28 ]
E X P E RIM ENTA L
LC conditions
System:
ACQUITY UPLC H-Class Bio System
Detector:
ACQUITY UPLC Tunable UV Detector
Column:
ACQUITY UPLC Protein BEH C4 Column,
300Å, 1.7 µm, 2.1 mm X 50 mm
(p/n 186004495)
Column temp.:
80 °C
Mobile phase A:
Water
Mobile phase B:
Acetonitrile
Mobile phase C:
1% formic acid
Optical detection:
UV 280 nm
LC gradient table:
Time
Flow
(min) (mL/min) %A
%B
%C
%D Curve
Initial
0.40
85.0
5.0 10.0
0 Initial
1.00
0.40 85.0
5.0 10.0 0
6
1.01
0.20 85.0
5.0 10.0 0
6
3.50
0.20
5.0
95.0 0.0
0
6
3.70
0.40
5.0
95.0 0.0
0
6
4.00
0.40
10.0
80.0 10.0 0
6
4.50
0.40
10.0
80.0 10.0 0
6
5.00
0.40 85.0
5.0 10.0 0
6
5.50
0.40 85.5
5.0 10.0 0
6
Total run time: 6.5 min
MS conditions
Mass spectrometer:
Xevo G2-S QTof
Capillary:
2.5 kV
Sampling cone:
80 V
Extraction cone:
4 V
Source temp.:
150 °C
Desolvation temp.:
350 °C
Cone gas flow:
0 L/h
Desolvation gas flow: 800 L/h
Informatics
UNIFI Scientific Information System
Results derived from an intact IgG1 mAb mass analysis are
used to illustrate how this integrated system solution can help
the biopharmaceutical laboratories to streamline a common
analytical workflow, shown in Figure 1, and more quickly and
efficiently communicate key information needed to bring better
molecules to market faster.
Biopharmaceutical System Solution with UNIFI
■
■
ACQUITY UPLC H-Class Bio System
■
■
Protein Separation Technology (PrST) Columns
■
■
Xevo G2-S QTof with an ACQUITY UPLC Tunable UV Detector
■
■
UNIFI Scientific Information System
Sample preparation
Waters Intact mAb Mass Check Standard (p/n 186006552) was
analyzed by solubilizing the standard (10 mg/mL or 67 μM,
100 µL DI water to standard vial, 5 min sonication), and diluting
20X (Final 3.3 μM, 0.50 μg/μL) with eluent A for Xevo G2 QTof
analysis or 200X (0.33 μM, 0.05 μg/μL) for Xevo G2-S analysis.
Analysis Method
A holistic UNIFI method contains information
sufficient for LC-MS acquisition, data processing,
reporting, and report sign-off.
Analysis
Data is acquired once the sample list reaches
the top of the sample queue, is subjected
to automatic post-acquisition processing, and
the assigned reporting templates are executed.
Report
Standard report elements have been optimized
for displaying antibody results. Custom calculations
and filters efficiently summarize overall findings.
Figure 1. Intact mass analysis workflow with the UNIFI Scientific
Information System.
[ 29 ]
Streamlining Compliant and Non-Compliant Intact Mass Analysis of Biotherapeutic mAbs
with the Biopharmaceutical Platform Solution with UNIFI
R E SU LT S AND DIS CUSSION
An automated mAb LC-MS analysis set of 11 injections was automatically acquired, processed, and reported as
specified in a single UNIFI method. Data are representative of a simple method development set, where the goal
of the researcher is to assess the extent of product glycovariation and determine analytical reproducibility.
For the processed results, a single injection is represented in the review panel of the UNIFI analysis center,
shown in Figure 2. This panel is configured to convey chromatographic information (integrated total ion
chromatogram), the MaxEnt™ deconvoluted MS spectrum corresponding to the summed spectra under the
detected peak, and a component summary window filtered to display the top five most intensely assigned
glycoforms (G0F/G0F, G0F/G1F, G1F/G1F, G1F/G2F, or G2F/G2F).
This combined panel enables a researcher to assess chromatographic quality, the quality of MS data processing,
and the quality of glycoform assignments in a single display. Closer examination also reveals the relative
abundance of each glycoform was automatically calculated as part of the processing.
Figure 2. The review panel in UNIFI displays automatically processed experimental results.
[ 30 ]
Streamlining Compliant and Non-Compliant Intact Mass Analysis of Biotherapeutic mAbs
with the Biopharmaceutical Platform Solution with UNIFI
Having designated one sample as the reference enables a researcher to select the comparative mode display
of the review tab. The binary comparison display, shown in Figure 3, provides a means to visually examine
the differences between the two samples, thus revealing the extent of variation between samples. In this
display, comparative chromatograms and spectra (A280 and summed m/z spectra) are depicted, along with the
component summary, now reformatted to address comparative questions. Since both injections were from the
same sample, minimal experimental result differences are predictably observed.
Figure 3. UNIFI’s review panel (compare mode) displays automatically processed experiment results, with a focus on identifying similarities and differences between
a reference sample and unknown samples.
[ 31 ]
Streamlining Compliant and Non-Compliant Intact Mass Analysis of Biotherapeutic mAbs
with the Biopharmaceutical Platform Solution with UNIFI
The summary plot tool within the review tab enables researchers to quickly compare trends and differences
within the larger data set. The variation of mAb glycoform MS response, as shown in Figure 4, would be a
common application of this capability, as would comparisons of observed retention time or mass error across
the sample set. The consistent MS response of glycoforms across all injections illustrates the expected
reproducibility of the intact mass analysis of replicated injections.
Figure 4. summary plot view of MS response for 5 mAb glycoforms (n = 11 injections).
UNIFI reporting
The reporting functionality within UNIFI Software is powerful, addressing one of the common bottlenecks
encountered by organizations when generating and managing large volumes of complex scientific data.
The ability to customize common report objects by means of filters, formatting, and the use of custom fields
and calculations enables report content to be automatically generated by an entire organization with high
quality on a consistent basis. Based on the analytical objectives, one or more report templates can be
attached to the analysis method.
[ 32 ]
Streamlining Compliant and Non-Compliant Intact Mass Analysis of Biotherapeutic mAbs
with the Biopharmaceutical Platform Solution with UNIFI
The first page of a typical intact mass analysis experimental report contains a summary of sample information
and acquisition status, as shown in Figure 5. More detailed experimental results (such as TUV and TIC
chromatograms, raw and deconvoluted MS spectra, and identified component response summary table) are
often grouped for each injection, as shown in Figure 6.
In the case of mAbs, generic report objects were tuned to account for the rapid desalting LC-MS method that
was used, the acquisition of UV and MS data, and the typical input m/z and output mass ranges encountered
during antibody ESI mass analysis.
Figure 5. Typical first page of an intact protein LC-MS report in UNIFI Software that summarizes sample and acquisition details.
[ 33 ]
Streamlining Compliant and Non-Compliant Intact Mass Analysis of Biotherapeutic mAbs
with the Biopharmaceutical Platform Solution with UNIFI
Figure 6. Example report object grouping (TUV and TIC chromatograms, raw and deconvoluted MS spectra, and identified component response summary table) from a
single injection within the analysis.
[ 34 ]
Streamlining Compliant and Non-Compliant Intact Mass Analysis of Biotherapeutic mAbs
with the Biopharmaceutical Platform Solution with UNIFI
Waters Corporation
34 Maple Street
Milford, MA 01757 U.S.A.
T: 1 508 478 2000
F: 1 508 872 1990
www.waters.com
Waters, The Science of What’s Possible, ACQUITY UPLC, UPLC, Xevo, and UNIFI are registered trademarks of Waters Corporation.
MaxEnt is a trademark of Waters Corporation. All other trademarks are the property of their respective owners.
©2013–2016 Waters Corporation. Produced in the U.S.A. January 2016 720004617EN AG-PDF
CONC LUSIONS
The intact mass analysis workflows within the Biopharamceutical Platform
Solution with UNIFI enable automated data acquisition, processing, and reporting
of a typical method validation sample set. This demonstrates UNIFI Software’s
ability to facilitate robust glycoform profiling of a recombinant mAb, removes
the necessity of manual data processing, and improves the process of data review
and reporting. The implementation of such highly automated workflows should
enable biotherapeutic development and quality organizations to handle larger
volumes of sample requests with the same resources, while improving the quality
of the information they provide.
In addition, the ability to automate reporting summary results across the sample sets eliminates the use of
external software for data aggregation, as shown in Figure 7. This not only greatly increases the timeliness of
communicating results, but avoids the human errors and validation efforts that cost analytical organizations
time and money. In the case of this typical method validation injection set, the precision of MS response and
mass accuracy is reported for one of the observed glycoforms.
Figure 7. The report object summarizes MS response and mass accuracy/precision across all injections within the sample set in table and bar chart formats.
[ 35 ]
Streamlining Compliant and Non-Compliant Intact Mass Analysis of Biotherapeutic mAbs
with the Biopharmaceutical Platform Solution with UNIFI
GLYCOPEPTIDE MAPPING AND SUBUNIT ANALYSIS
[ 36 ]
WAT E R S S O LU T IO NS
ACQUITY UPLC® Glycoprotein BEH
Amide 300Å Column (patent pending)
Glycoprotein Performance Test Standard
ACQUITY UPLC H-Class Bio System
Waters SYNAPT® G2-S HDMS
K E Y W O R D S
ACQUITY UPLC H-Class Bio System,
BEH Amide 300Å, glycans,
glycosylated proteins, glycosylation,
HILIC, mab, glycopeptide, glycoprotein
A P P L I C AT IO N B E N E F I T S
■
■
Orthogonal selectivity to conventional
reversed phase (RP) peptide mapping
for enhanced characterization of
hydrophilic protein modifications, such as
glycosylation
■
■
Class-leading HILIC separations
of IgG glycopeptides to interrogate
sites of modification
■
■
MS compatible HILIC to enable detailed
investigations of sample constituents
■
■
Enhanced glycan information that
complements RapiFluor-MS released
N-glycan analyses
■
■
Glycoprotein BEH Amide 300Å
1.7 µm stationary phase is QC tested
via a glycoprotein separation to ensure
consistent batch to batch reproducibility
I N T RO DU C T IO N
Peptide mapping of biopharmaceuticals has longed been used as a tool
for identity tests and for monitoring residue-specific modifications.1-2 In a
traditional analysis, peptides resulting from the use of high fidelity proteases,
like trypsin and Lys-C, are separated with very high peak capacities by reversed
phase (RP) separations with C18 bonded stationary phases using ion-pairing
reagents. Separations such as these are able to resolve peptides with single
amino acid differences such as asparagine; and the two potential products of
asparagine deamidation, aspartic acid and isoaspartic acid.3-4
Nevertheless, not all protein modifications are so easily resolved by RP
separations. Glycosylated peptides, in comparison, are often separated with
relatively poor selectivity, particularly if one considers that glycopeptide
isoforms usually differ in their glycan mass by about 10 to 2,000 Da. So,
while RP separations are advantageous for generic peptide mapping, they are
limited in their ability to resolve hydrophilic modifications. Previous studies
have demonstrated that hydrophilic interaction chromatography (HILIC)
with an amide-bonded stationary phase can provide complementary and
highly resolving separations of glycosylated peptides.5-6 These studies have
demonstrated that amide-bonded stationary phases are particularly effective
for these separations, because they afford high retentivity as a consequence of
their hydrophilicity and propensity for hydrogen bonding.7
Expanding upon this technology, we have developed an amide-bonded
stationary phase with a nominally larger pore diameter, a so-called “wide-
pore” material, such that amide HILIC separations can be universally applied
to separating the glycoforms of both intact and digested glycoproteins. This
stationary phase found in ACQUITY UPLC Glycoprotein BEH Amide 300Å 1.7
μm Columns ensures that glycopeptides, regardless of their size, will have
access to the majority of the porous network and be less prone to restricted
diffusion.8-9 In previous work, we have demonstrated the use of this HILIC column
to assay the glycan occupancy of an intact monoclonal antibody (mAb),10 to
map the domain-specific glycosylation of IgG subunits,11 and to improve the
resolution of tri- and tetra-antennary GlycoWorks,™ RapiFluor-MS™ labeled
N-glycans.12 Here, we explore the use of the Glycoprotein BEH Amide 300Å 1.7
μm Column to produce high resolution HILIC separations of glycopeptides from
three different monoclonal antibodies: trastuzumab, cetuximab and an IgG1K
candidate reference material from NIST.
HILIC Glycopeptide Mapping with a Wide-Pore Amide Stationary Phase
Matthew A. Lauber and Stephan M. Koza
Waters Corporation, Milford, MA, USA
[ 37 ]
E X P E R IM E N TA L
Sample description
Lys-C digest of trastuzumab and NIST candidate
reference material
An adaptation of a previously published single reaction vial,
overnight (16+ hours) procedure4 was employed to prepare
non-reduced Lys-C digests of trastuzumab and a IgG1K monoclonal
antibody candidate reference material obtained from NIST
(#8670, lot# 3F1b). TFA quenched digests were stored at -80 °C
until analyzed. In preparation for HILIC chromatography, aqueous
digests were diluted with 4 parts acetonitrile and 0.1 parts
dimethylsulfoxide and were then centrifuged at 16 x 1000 g for
10 minutes to remove any insoluble composition. Supernatant
from the centrifuged digest was thereafter injected.
Lys-C/tryptic digest of cetuximab
Reduced and alkylated cetuximab was digested with a combination
of Achromobacter protease I (Lys-C) and trypsin. Formulated
cetuximab was concentrated to 10 mg/mL and buffer exchanged
with a 10 kDa MWCO centrifugal filter (Millipore, Billerica, MA) into
a solution of 6 M GuHCl, 50 mM DTT, and 0.2 M phosphate (pH 8.1),
then incubated at 37 °C for 2 hours. Thereafter, the sample was
diluted with a solution of iodoacetamide, bringing the antibody
concentration to 8 mg/mL and the buffer composition to 4.8 M
GuHCl, 40 mM DTT, 50 mM iodoacetamide, and 0.17 M phosphate
(pH 8.1). Alkylation with iodoacetamide was allowed to proceed
under these conditions for 10 min in the dark at 37 °C, before
being quenched by the addition of cysteine, diluted with a urea-
containing buffer, and mixed with Achromobacter protease I (Lys-C)
at a 4:1 w/w ratio. The resulting digest solution of 0.8 mg/mL
cetuximab, 0.5 M GuHCl, 3 M Urea, 40 mM NH2OH, 4 mM DTT,
5 mM iodoacetamide, 6 mM cysteine, and 0.1 M phosphate (pH ~7.1)
was incubated at 37 °C. After 2 hours of incubation, this digest
solution was diluted two fold with water and an aliquot of trypsin
(Sigma T6567), such that the protein:trypsin ratio was 4:1 (w/w).
After incubation at 37 °C for another 2 hours, the digest solution
was again diluted two fold with water and a fresh aliquot of trypsin.
With a total protein:trypsin ratio of 2:1 (w/w), the digest was left to
incubate at 37 °C for 16 hours. Following this incubation, the digest
was quenched by acidification with TFA and stored at -80 °C until
analyzed. In preparation for HILIC chromatography, aqueous digests
were diluted with 4 parts acetonitrile and 0.1 parts dimethylsulfoxide
and were then centrifuged at 16 x 1000 g for 10 minutes to remove
any insoluble composition. Supernatant from the centrifuged digest
was thereafter injected.
Method conditions
(unless otherwise noted):
Column conditioning
ACQUITY UPLC Glycoprotein BEH Amide 300Å 1.7 µm Columns
(as well as other amide columns intended for glycoprotein
or glycopeptide separations) should be conditioned via two
sequential injections and separations of 40 µg Glycoprotein
Performance Test Standard (p/n 186008010; 10 µL injections
of 4 mg/mL in 0.1% TFA, 80% ACN) or with equivalent loads
of a test sample for which the column has been acquired. The
separation outlined by the following method can be employed for
conditioning with the Glycoprotein Performance Test Standard.
Column conditioning gradient
2.1 x 150 mm
Mobile phase A:
0.1% (v/v) TFA, H2O
Mobile phase B:
0.1% (v/v) TFA, ACN
Time
%A
%B
Curve
0.0 15.0 85.0
6
0.5 15.0 85.0
6
1.0 33.0 67.0
6
21.0 40.0
60.0
6
22.0 100.0
0.0
6
24.0 100.0
0.0
6
25.0 15.0
85.0
6
35.0 15.0
85.0
6
LC conditions for LC-UV-MS of mAb glycopeptides
(Figures 1–6):
LC system:
ACQUITY UPLC H-Class Bio System
Sample temp.:
10 °C
Analytical
column temp.:
30 °C (trastuzumab Lys-C digest
HILIC separations)
60 °C (cetuximab Lys-C/tryptic digest
HILIC separations)
60 °C (trastuzumab Lys-C reversed
phase separations)
[ 38 ] HILIC Glycopeptide Mapping with a Wide-Pore Amide Stationary Phase
Flow Rate:
0.2 mL/min
Mobile phase A:
0.1% (v/v) TFA, H2O
Mobile phase B:
0.1% (v/v) TFA, ACN
HILIC injection volume: 100–250 µL (Aqueous digests were
diluted with 4 parts acetonitrile and
0.1 parts dimethylsulfoxide to obtain
a miscible, HILIC compatible diluent.)
Reversed phase
injection volume:
24.2 µL (Aqueous digest)
Columns:
ACQUITY UPLC Glycoprotein BEH
Amide 300Å 1.7 µm, 2.1 x 150 mm
(p/n 176003702, with Glycoprotein
Performance Test Standard)
ACQUITY UPLC Peptide BEH C18 300 Å
1.7 µm, 2.1 x 150 mm (p/n 186003687)
Vials:
Polypropylene 12 x 32 mm Screw Neck
Vial, 300 µL Volume (p/n 186002640)
Gradient used for reversed phase separations of trastuzumab
Lys-C digests (Figure 1A):
Time
%A
%B
Curve
0.0 98.0
2.0
6
96.0 50.0
50.0
6
99.0 20.0
80.0
6
101.0 20.0
80.0
6
102.0 98.0
2.0
6
113.0 98.0
2.0
6
Gradient used for HILIC separations of trastuzumab
Lys-C digests and Lys-C/tryptic digests of cetuximab
(Figures 1B-6):
Time
%A
%B
Curve
0.0 20.0 80.0
6
60.0 50.0
50.0
6
61.0 80.0
20.0
6
63.0 80.0
20.0
6
64.0 20.0
80.0
6
75.0 20.0
80.0
6
MS conditions for IgG subunit separations
MS system:
SYNAPT G2-S HDMS
Ionization mode:
ESI+
Analyzer mode:
Resolution (~20 K)
Capillary voltage:
3.0 kV
Cone voltage:
25 V
Source temp.:
120 °C
Desolvation temp.:
350 °C
Desolvation gas flow: 800 L/Hr
Calibration:
NaI, 1 µg/µL from 100–2000 m/z
Lockspray:
300 fmol/µL Human Glufibrinopeptide B
in 0.1% (v/v) formic acid,
70:30 water/acetonitrile
every 90 seconds
Acquisition:
50–2500 m/z, 0.1 sec scan rate
Data management:
MassLynx Software (V4.1) / UNIFI V1.7
LC Conditions for a Glycopeptide Mapping of an IgG1K
with Fluorescence Detection (Figure 7):
LC system:
ACQUITY UPLC H-Class Bio System
Sample temp.:
10 °C
Analytical
column temp.:
45 °C
Fluorescence detection: Ex 280/Em 320 nm
(10 Hz scan rate, Gain =1)
Injection volume:
100 µL (DMF/ACN diluted sample)
Mobile phase A:
0.1% TFA in water
Mobile phase B:
0.1% TFA in ACN
Columns:
ACQUITY UPLC Glycoprotein BEH Amide
300Å 1.7 µm, 2.1 x 50 mm
(p/n 176003702, with Glycoprotein
Performance Test Standard)
Other columns:
Column A: 2.6 µm, 2.1 x 150 mm
Column B: 1.8 µm, 2.1 x 150 mm
Vials:
Polypropylene 12 x 32mm Screw Neck
Vial, 300 µL Volume (p/n 186002640)
Gradient (Figure 7):
Time
(min)
%A
%B
Curve
0.0 15.0 85.0
6
0.5 15.0 85.0
6
1.0 30.0 70.0
6
21.0 37.0
63.0
6
22.0 100.0
0.0
6
24.0 100.0
0.0
6
25.0 15.0
85.0
6
35.0 15.0
85.0
6
Data management:
UNIFI v1.7
[ 39 ]
HILIC Glycopeptide Mapping with a Wide-Pore Amide Stationary Phase
R E SU LT S A N D D IS C U S S IO N
Orthogonal and complementary glycopeptide mapping separations
To demonstrate a conventional approach to peptide mapping, we first performed LC-UV-MS analysis on a
Lys-C digest of a mAb using a RP chromatographic separation with a wide-pore C18 bonded stationary phase
(Peptide BEH C18 300Å 1.7 μm). Trastuzumab was selected for this study, given its prominence as a first
generation mAb drug product and a potential target for biosimilar development.13 Figure 1A shows a UPLC
chromatogram that is typical for a Lys-C digest of trastuzumab, wherein peptides are broadly resolved
across a separation with a gradient corresponding to a change of 0.5% acetonitrile per minute. The non-
glycosylated peptides of the digest spread across the extremes of the chromatogram while the glycopeptides
elute in an approximately one minute wide window at a retention time of about 60 minutes. The conditions
to produce this high resolution separation involve the use of mobile phases modified with trifluoroacetic acid
(TFA); the same mobile phases that have proven to be optimal for HILIC of proteinaceous analytes.10-11
Accordingly, an orthogonal method to the RP separation can be achieved via HILIC by simply reversing the
gradient and using the newly developed wide-pore amide bonded stationary phase (Glycoprotein BEH Amide
300Å 1.7 μm). An example of a chromatogram obtained from a column packed with this wide-pore amide
material and a gradient ramp of 0.5% acetonitrile per minute is shown in Figure 1B. Here, the peptides from the
Lys-C digested trastuzumab are very clearly segregated into early and late eluting species, corresponding to
the non-glycosylated and glycosylated species, respectively. The use of TFA ion pairing facilitates obtaining
this separation, as it masks the hydrophilicity of the peptide residues and provides improved selectivity for the
hydrophilic modifications. Also note that the glycopeptides have not only been class separated with the amide
column, but the selectivity of the peptide glycoforms is remarkably improved over the analogous RP separation.
0.0
0.5
1.0
0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
90
A
214
0.0
0.5
1.0
0
5
10
15
20
25
30
35
40
45
A
214
Time (min)
Peptide BEH C18
300Å 1.7
m
Glycoprotein BEH Amide
300Å 1.7
m
Non-glycosylated
Peptides
Non-glycosylated
Peptides
Glycopeptides
Lys-C
Glycopeptides
A
B
Figure 1. Lys-C glycopeptide mapping of trastuzumab. (A) A traditional reversed phase separation of
the Lys-C digest using a 2.1 x 150 mm ACQUITY UPLC Peptide BEH C18 300Å 1.7 µm Column.
(B) A HILIC separation of the Lys-C digest using a 2.1 x 150 mm ACQUITY UPLC Glycoprotein BEH
Amide 300Å 1.7 µm Column. In each analysis, 9.2 µg of the Lys-C digest was separated using
the same gradient slope and injecting sample from a diluent comprised of either approximately
0.2% TFA in 80:20 ACN/water (HILIC) or 100% water (reversed phase).
[ 40 ] HILIC Glycopeptide Mapping with a Wide-Pore Amide Stationary Phase
By focusing on the strongly retained peaks, one can begin to interrogate the glycosylation of the
trastuzumab molecule (Figure 2A). In particular, MS data acquired from online mass detection and a total
ion chromatogram (TIC) can be applied to identify the peptide species and its corresponding glycoforms, as
shown in Figure 2B. This Lys-C glycopeptide map presents a 29 amino acid residue peptide (K16) from the Fc
domain of trastuzumab. From an analysis of the MS data, many biantennary structures typical found on mAbs
in relatively high abundance can be readily identified (Figure 3A). Further interrogation of the MS data, also
shows that low abundance N-glycan species can likewise be detected. Figure 3B, for instance, provides
MS data supporting the identification of monosialylated and disialylated glycoforms at retention times of
approximately 34.7 minutes and 36.4 minutes, respectively. These identifications correlate extremely well
with the released N-glycan profiles of trastuzumab that have been previously reported.14-15
0.01
0.04
0.07
25
26
27
28
29
30
31
32
33
34
35
36
37
38
A
214
0E+0
1E+6
2E+6
25
26
27
28
29
30
31
32
33
34
35
36
37
38
In
te
ns
it
y
Time (min)
TIC
A1
M5
FA2G1
FA2G2
FA1G1
A2G1
A2
FA1
FA2
-17.03 Da
FA2G1
-17.03 Da
UV
FA2G1’
FA2
A
B
HC:K16
TKPREEQYNSTYRVVSVLTVLHQDWLNGK
Glycoforms
33
34
35
36
37
38
FA2G2
1380 1381 1382 1383 1384
m/z
FA2G2S1
1380.8834, 4+
-4.1 ppm
1453 1454 1455 1456 1457
m/z
FA2G2S2
1453.6593, 4+
-2.5 ppm
min
Species Modification Calculated
Mass (Da)
RT
(Min)
Intensity
Error
(ppm)
HC:K16
A1
4555.1865 26.4
15461
-1.0
HC:K16
A2
4758.2656 27.7
219414
-1.1
HC:K16
FA1
4701.2441 27.7
17931
-3.3
HC:K16
FA2
4904.3237 28.8
1502968
-1.7
HC:K16
M5
4676.2129 29.3
51057
-0.7
HC:K16
A2G1
4920.3188 29.7
89714
-0.9
HC:K16
A2G1
4920.3188 30.3
33516
-1.9
HC:K16
FA1G1
4863.2974 30.4
17005
-1.1
HC:K16
FA2G1
5066.3765 30.8
1156785
-0.7
HC:K16
FA2G1’
5066.3765 31.4
312270
-1.0
HC:K16
FA2G2
5228.4292 33.2
226743
-2.0
A
B
Figure 2. Lys-C glycopeptide mapping of trastuzumab
with HILIC and an ACQUITY UPLC Glycoprotein BEH
Amide 300Å 1.7 µm Column. (A) UV chromatogram for
the Lys-C glycopeptide retention window. (B) Total ion
chromatogram (TIC) for the same retention window.
Figure 3. Mass spectrometric data supporting trastuzumab
Lys-C glycopeptide identifications. (A) Retention times,
MS intensities and mass errors for assignments labeled in
Figure 2. (B) MS spectra supporting the identification of low
abundance Lys-C glycopeptides modified with mono and
di-sialylated N-glycans.
[ 41 ]
HILIC Glycopeptide Mapping with a Wide-Pore Amide Stationary Phase
Lot-to-lot analysis of trastuzumab glycosylation via HILIC-UV glycopeptide mapping
HILIC-MS based glycopeptide mapping clearly yields information-rich data. However, these HILIC glycopeptide
mapping separations also lend themselves to methods based only on optical detection. We have, for example,
applied a HILIC-UV method to perform lot-to-lot analysis of trastuzumab glycosylation for two drug product
samples. Representative HILIC chromatograms for glycopeptide K16 obtained from two different lots of
trastuzumab are shown in Figure 4A. Previous released glycan analyses on these lots have shown there to
be differences in glycosylation.14 Through comparison of peak areas across the glycopeptide profile, we have
found that these two lots of trastuzumab indeed differ with respect to their glycosylation. Specifically, these
lots of trastuzumab appear to have different extents of terminal galactosylation, as can be seen in the differing
abundances of FA2, FA2G1 and FA2G2 glycoforms (Figure 4B). This observation was consistent with data
obtained from previous released glycan analyses and previous HILIC based profiling of trastuzumab subunits.11
1
2
3
4 5 6
7
8
9
1
2
4 5 6
7
8
9
3
25.0
0.02
0.02
0.08
0.08
A
214
A
214
Lot 1
Lot 2
0
10
20
30
40
50
1 2 3 4 5 6 7 8 9
% Amount
Component
Batch 1
Batch 2
37.5 min
FA2
FA2G1
FA2G2
A
B
Figure 4. Lot-to-lot profiling of trastuzumab Lys-C peptide glycoforms. (A) HILIC chromatograms of trastuzumab Lys-C glycopeptides
from two different lots of drug product. (B) Relative abundances of the major sample components. Analyses were performed in
triplicate using a 2.1 x 150 mm ACQUITY UPLC Glycoprotein BEH Amide 300Å 1.7 µm Column.
Complementing GlycoWorks RapiFluor-MS N-glycan analyses with domain and peptide specific
information about mAb glycosylation
An appealing aspect of glycopeptide mapping is that it can be applied to the elucidation of domain and
peptide specific information. By inference or ETD fragmentation analyses, or both, glycopeptide mapping
can also be used to detail the exact sites of glycosylation.16 As we have noted before,11 IgGs contain one
conserved N-glycosylation site at Asn297 of the heavy chain, meaning they will be modified with
two glycans in their Fc subunit. In addition, some IgGs and even some mAb IgG therapeutics exhibit
multi-domain glycosylation. Cetuximab, for instance, is glycosylated in both its Fc and Fab domains,17
making it a very interesting case study for this work.
[ 42 ] HILIC Glycopeptide Mapping with a Wide-Pore Amide Stationary Phase
A HILIC glycopeptide map of a Lys-C/tryptic digest of cetuximab provides a clear indication of the
complicated glycan profile of this molecule (Figure 5). In the presented chromatogram, approximately thirty
chromatographic peaks are observed. Furthermore, a cursory analysis of the MS data has shown there to be at a
minimum twenty five different glycoform species with rather high relative abundances of greater than 1–2%.
Figure 6 provides the MS data supporting these assignments. As can be seen, 9 unique glycoforms could be
assigned to tryptic peptide T22 from the Fc domain, while the other 16 glycoforms could be assigned to tryptic
peptide T8 from the Fab domain of cetuximab. It is interesting to note that the majority of the Fab domain (T8)
glycans contain immunogenic epitopes, such as non-human alpa-1,3-galactose or non-human N-glycolyl-
neuraminic acid moieties.18 In previous work, these glycan species were identified through complementary
subunit mapping and RapiFluor-MS released N-glycan analyses.11 With these results on glycopeptide mapping,
we show yet another complementary technique for assessing protein glycosylation.
Species
Modification
Calculated
Mass (Da) RT (Min) Intensity
Error
(ppm)
HC:T22
FA1
2429.9592
32.2
9047
1.5
HC:T22
FA2
2633.0386
33.6
263523
1.4
HC:T22
M5
2404.9275
35.2
78988
0.0
HC:T22
FA2G1
2795.0913
36.1
186156
1.2
HC:T22
FA2G1'
2795.0913
36.7
57185
0.7
HC:T22
[Hex5HexNAc3DHex1]
2754.0647
37.8
7783
1.0
HC:T22
M6
2566.9805
38.1
5432
-1.8
HC:T22
FA2G2
2957.1443
38.9
38257
1.1
HC:T22
[Hex6HexNAc3DHex1]
2916.1177
40.5
12246
-1.3
HC:T8
FA2
3350.3689
29.4
5688
4.9
HC:T8
M5
3122.2578
30.4
6437
2.7
HC:T8
FA2G1
3512.4216
31.7
8925
3.2
HC:T8
FA2G2
3674.4746
34.1
11695
8.2
HC:T8
FA2G2Ga1
3836.5273
36.7
9657
-0.6
HC:T8
FA2G2Ga1'
3836.5273
37.0
15605
0.3
HC:T8
FA2G2Sg1
3981.5649
37.5
25532
1.9
HC:T8
FA2G2Ga2
3998.5801
39.1
212644
-0.7
HC:T8
A2G2
3471.3950
39.1
21542
2.0
HC:T8
[Hex7HexNAc5DHex1]
4144.6377
39.2
6944
-10.8
HC:T8
FA2G2Ga1Sg1
4143.6177
39.6
90905
-0.7
HC:T8
FA2G2Sg2
4288.6553
40.2
7428
-2.5
HC:T8
[Hex8HexNAc5DHex1]
4363.7124
41.8
5280
1.7
HC:T8
[Hex7HexNAc5DHex1NGNA1]
4508.7498
42.4
7114
-3.6
HC:T8
[Hex9HexNAc5DHex1]
4525.7651
43.8
20104
4.3
HC:T8
[Hex8HexNAc5DHex1NGNA1]
4670.8027
44.2
7640
1.2
*Hex: hexose, HexNAc: N-acetylate hexosamine, DHex: deoxyhexose, NGNA: N-glycoyl neuraminic acid / (‘): structural isomer
0.015
0.040
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
A
21
4
Time (min)
T8
+FA2G1
T22
+FA1
T22
+FA2
T8
+FA2G2
T22
+M5
T22
+FA1G1
T22
+FA2G1
T22
+FA2G1’
T8
+FA2G2Ga1
T8
+FA2G2Sg1
T22
+Hex5HexNAc3DHex1
T22
+M6
T22
+FA2G2
T8
+Hex8HexNAc5DHex1NGNA1
T8
+Hex9HexNAc5DHex1
T8
+Hex8HexNAc5DHex1
T8
+Hex7HexNAc5DHex1NGNA1
T8
+FA2G2Sg2
T22
+Hex6HexNAc3DHex1
T22
+FA2G2Ga1
T8 (deamidated)
+FA2G2Ga2
T8
+FA2G2Ga2
T8
+FA2G2Ga1Sg1
HC:T22
EEQYNSTYR
Glycoforms
HC:T8
MNSLQSNDTAIYYC(am)AR
Glycoforms
Figure 5. Assaying the N-linked glycan sites of cetuximab and their microheterogeneity using a combined Lys-C/tryptic digest.
Assignments for glycoforms of peptide T22 from the Fc domain of cetuximab are shown in dark gray, while assignments to the
glycoforms of peptide T8 from the Fab domain are shown in red. Analyses were performed on 9.2 µg of Lys-C/tryptic digest using
a 2.1 x 150 mm ACQUITY UPLC Glycoprotein BEH Amide 300Å 1.7 µm Column and a sample diluent of approximately 0.2% TFA
in 80:20 ACN/water.
Figure 6. Mass spectrometic data supporting
glycopeptide assignments from the HILIC-UV-MS
analysis of Lys-C/trypsin digested cetuximab. From
a cursory analysis of MS data, 9 unique glycoforms
could be assigned to the Fc domain, and 16 unique
glycoforms could be assigned to the Fab domain.
[ 43 ]
HILIC Glycopeptide Mapping with a Wide-Pore Amide Stationary Phase
Benchmarking the Capabilities of the
Glycoprotein BEH Amide 300Å 1.7 μm Column
The peak capacities obtained in these example
glycopeptide separations are particularly
noteworthy when a comparison is made to otherwise
available column technologies. To benchmark
the performance of the Glycoprotein BEH Amide
300Å 1.7 μm Column, we have analyzed a Lys-C
digest of a NIST candidate reference material, an
IgG1K mAb. In this testing, a focused gradient was
used along with intrinsic peptide fluorescence
instead of low wavelength UV detection so that
higher signal-to-noise could be achieved in the
obtained chromatograms. The glycopeptide that
originates from the Fc domain of a mAb will contain
a tryptophan residue upon Lys-C cleavage, which in
large part makes this detection mechanism feasible.
Three fluorescence chromatograms obtained for
the Lys-C glycopeptides from the NIST IgG1K are
presented in Figure 7. These three chromatograms
were obtained from the use of the ACQUITY UPLC
Glycoprotein BEH Amide 300Å 1.7 µm as well as
two commercially available alternatives (Columns
A and B). Peak capacities have been measured for
each specific column using the retention windows
demarcated by the most extreme glycopeptide
retention times (*) and the half-height peak widths
of the K16+FA2, K16+FA2G1, K16+FA2G1',
K16+FA2G2, and K16+FA2G2Ga1 peaks. This
analysis shows that these columns exhibit strikingly
different resolving power. With an effective peak
capacity of 72.8, the Glycoprotein BEH Amide
column shows a superior peak capacity and
performance increases over the alternative
amide column technologies of 40 and 96%.
Glycoprotein BEH Amide 300Å 1.7 m
Column B: 1.8 µm, 2.1 x 150 mm
Column A: 2.6 µm, 2.1 x 150 mm
HC:K16
+FA2
HC:K16
+FA2G2Ga1
Pc* = 51.9
HC:K16
+FA2G1
HC:K16
+FA2G1’ HC:K16
+FA2G2
*
*
*
*
*
*
Pc* = 37.1
Pc* = 72.8
Figure 7. Lys-C glycopeptide mapping of an IgG1K using fluorescence detection and
various 2.1 x 150 mm columns packed with amide bonded stationary phase:
a Competitor Column A: 150Å 2.6 µm, 2.1 x 150 mm (Top), a Competitor Column B:
1.8 µm, 2.1 x 150 mm (Middle), and on an ACQUITY UPLC Glycoprotein BEH Amide
300Å 1.7 µm column (Bottom). Peak capacities were calculated based on the half-height
peak widths of the labeled glycopeptides and the retention window established by the
earliest and latest eluting glycopeptide species, marked with asterisks(*). Comparative
separations may not be representative in all applications.
[ 44 ] HILIC Glycopeptide Mapping with a Wide-Pore Amide Stationary Phase
Waters Corporation
34 Maple Street
Milford, MA 01757 U.S.A.
T: 1 508 478 2000
F: 1 508 872 1990
www.waters.com
Waters, ACQUITY UPLC, SYNAPT and The Science of What's Possible are registered trademarks of Waters Corporation. GlycoWorks
and RapiFluor-MS are trademarks of Waters Corporation. All other trademarks are the property of their respective owners.
©2015 Waters Corporation. Produced in the U.S.A. September 2015 720005409EN AG-PDF
C O N C LU S IO NS
Glycopeptide mapping of glycoproteins presents a highly effective technique
that can be used to elucidate both domain and peptide-specific glycosylation.
In this work, we have demonstrated the use of an ACQUITY UPLC Glycoprotein
BEH Amide 300Å 1.7 μm Column to obtain HILIC separations of glycopeptides
that complement the chromatographic information afforded by a reversed phase
separation. In addition, our results indicate that these HILIC separations provide
exemplary peak capacity in comparison to other commercially available amide
column technologies. That the HILIC separation is MS-compatible means that
information-rich data can be readily acquired to characterize a glycopeptide
map. For instance, this work shows that it can be a relatively straightforward
exercise to characterize multidomain protein glycosylation, such as the Fc
and Fab domain glycosylation of cetuximab. Combined with other recently
developed strategies, such as HILIC subunit mapping and GlycoWorks RapiFluor-
MS released N-glycan analyses, glycopeptide mapping with the ACQUITY UPLC
Glycoprotein BEH Amide Column shows significant promise for facilitating the
characterization of protein glycosylation to unprecedented levels of detail.
References
1. Xie, H.; Gilar, M.; Gebler, J. C., Characterization of protein impurities and site-specific
modifications using peptide mapping with liquid chromatography and data independent
acquisition mass spectrometry. Anal Chem 2009, 81 (14), 5699–708.
2. Witze, E. S.; Old, W. M.; Resing, K. A.; Ahn, N. G., Mapping protein post-translational
modifications with mass spectrometry. Nat Methods 2007, 4 (10), 798–806.
3. Huang, H. Z.; Nichols, A.; Liu, D., Direct identification and quantification of aspartyl
succinimide in an IgG2 mAb by RapiGest assisted digestion. Anal Chem 2009, 81 (4),
1686–92.
4. Lauber, M. A.; Koza, S. M.; McCall, S. A.; Alden, B. A.; Iraneta, P. C.; Fountain, K. J.,
High-Resolution Peptide Mapping Separations with MS-Friendly Mobile Phases and
Charge-Surface-Modified C18. Anal Chem 2013, 85 (14), 6936–44.
5. Gilar, M.; Yu, Y. Q.; Ahn, J.; Xie, H.; Han, H.; Ying, W.; Qian, X., Characterization of glycoprotein
digests with hydrophilic interaction chromatography and mass spectrometry. Anal Biochem
2011, 417 (1), 80–8.
6. Martin Gilar; Ying-Qing Yu; Joomi Ahn; Xie, H., Analysis of Glycopeptide Glycoforms in
Monoclonal Antibody Tryptic Digest using a UPLC HILIC Column. Waters Application Note
720003363en 2010.
7. Ahn, J.; Bones, J.; Yu, Y. Q.; Rudd, P. M.; Gilar, M., Separation of 2-aminobenzamide labeled
glycans using hydrophilic interaction chromatography columns packed with 1.7 microm sorbent. J
Chromatogr B Analyt Technol Biomed Life Sci 2010, 878 (3–4), 403–8.
8. Gustavsson, P.-E.; Larsson, P.-O., Support Materials
for Affinity Chromatography. In Handbook of Affinity
Chromatography, Hage, D., Ed. Taylor & Francis: Boca Raton,
FL, 2006; pp 15–33.
9. Renkin, E. M., J. Gen. Physio. 1954, (38), 225.
10. Lauber, M. A.; Koza, S. M., Developing High Resolution
HILIC Separations of Intact Glycosylated Proteins using
a Wide-Pore Amide-Bonded Stationary Phase Waters
Application Note 720005380en 2015.
11. Lauber, M. A.; Koza, S. M., Mapping IgG Subunit Glycoforms
using HILIC and a Wide-Pore Amide Stationary Phase Waters
Application Note 720005385en 2015.
12. Lauber, M. A.; Koza, S. M., Enhancing the Peak Capacity of
High Molecular Weight N-Glycan HILIC Separations with a
Wide-Pore Amide Bonded Stationary Phase. Waters Tech
Brief 720005381en 2015.
13. Beck, A.; Sanglier-Cianferani, S.; Van Dorsselaer, A.,
Biosimilar, biobetter, and next generation antibody
characterization by mass spectrometry. Anal Chem 2012,
84 (11), 4637–46.
14. Yu, Y. Q.; Ahn, J.; Gilar, M., Trastuzumab Glycan Batch-to-
Batch Profiling using a UPLC/FLR/MS Mass Spectrometry
Platform. Waters Appication Note 720003576en 2010.
15. Xie, H.; Chakraborty, A.; Ahn, J.; Yu, Y. Q.; Dakshinamoorthy,
D. P.; Gilar, M.; Chen, W.; Skilton, S. J.; Mazzeo, J. R., Rapid
comparison of a candidate biosimilar to an innovator
monoclonal antibody with advanced liquid chromatography
and mass spectrometry technologies. MAbs 2010, 2 (4).
16. Houel, S.; Hilliard, M.; Yu, Y. Q.; McLoughlin, N.; Martin,
S. M.; Rudd, P. M.; Williams, J. P.; Chen, W., N- and
O-glycosylation analysis of etanercept using liquid
chromatography and quadrupole time-of-flight mass
spectrometry equipped with electron-transfer dissociation
functionality. Anal Chem 2014, 86 (1), 576–84.
17. Qian, J.; Liu, T.; Yang, L.; Daus, A.; Crowley, R.; Zhou, Q.,
Structural characterization of N-linked oligosaccharides
on monoclonal antibody cetuximab by the combination of
orthogonal matrix-assisted laser desorption/ionization
hybrid quadrupole-quadrupole time-of-flight tandem
mass spectrometry and sequential enzymatic digestion.
Anal Biochem 2007, 364 (1), 8–18.
18. Arnold, D. F.; Misbah, S. A., Cetuximab-induced anaphylaxis
and IgE specific for galactose-alpha-1,3-galactose.
N Engl J Med 2008, 358 (25), 2735; author reply 2735–6.
[ 45 ]
HILIC Glycopeptide Mapping with a Wide-Pore Amide Stationary Phase
WAT E R S S O LU T IO NS
ACQUITY UPLC® Glycoprotein BEH
Amide, 300Å Column
Glycoprotein Performance Test Standard
GlycoWorks™ RapiFluor-MS™ N-Glycan Kit
ACQUITY UPLC H-Class Bio System
Xevo® G2 QTof Mass Spectrometer
SYNAPT® G2-S HDMS
K E Y W O R D S
ACQUITY UPLC H-Class Bio System,
BEH Amide 300Å, Glycans, Glycosylated
Proteins, Glycosylation, HILIC, IdeS
A P P L I C AT IO N B E N E F I T S
■
■
Improved HILIC separations of IgG
subunit glycoforms.
■
■
MS-compatible HILIC to enable detailed
investigations of sample constituents.
■
■
Orthogonal selectivity to conventional
reversed-phase (RP) separations for
enhanced characterization of hydrophilic
protein modifications.
■
■
Domain-specific glycan information that
complements profiling glycosylation by
RapiFluor-MS released N-glycan analyses.
■
■
Glycoprotein BEH amide, 300Å,
1.7 µm stationary phase is QC tested via
a glycoprotein separation to ensure
consistent batch to batch reproducibility.
I N T RO DU C T IO N
Without question, the most successfully exploited protein modality for therapeutic
applications has been monoclonal antibodies (mAbs), which currently account
for nearly half of the biopharmaceutical market.1 An intriguing characteristic of
mAbs, in particular IgG-based mAbs, is that they are formed by the linking of two
identical light chains and two identical heavy chains through disulfide bonding
and non-covalent interactions. Moreover, the resulting mAb structure exhibits
functionally significant subunits, for instance one crystallizable fragment
(Fc domain) and two equivalent antigen binding fragments (Fab domains).
In what is commonly referred to as a middle-up or middle-down analysis,2-5 native
mAbs can be proteolyzed into these and other related subunits enzymatically,
as a means to perform cell-based studies and to facilitate characterization. One
increasingly popular way to produce subunit digests of mAbs is via the IdeS
protease (Immunoglobulin Degrading Enzyme of S. pyogenes).2,6 IdeS cleaves
with high fidelity at a conserved sequence motif in the hinge region of humanized
mAbs to cleanly produce, upon reduction, three 25 kDa mAb fragments that are
amenable to mass spectrometry and useful for localizing different attributes of
therapeutic mAbs (Figure 1).3 IdeS digestion combined with reversed-phase (RP)
chromatography has, in fact, been proposed as a simple identity test for mAbs
and fusion proteins, because IdeS produced subunits from different drug products
will exhibit diagnostic RP retention times.3 Additionally, RP techniques have been
shown to be useful in assaying and obtaining domain specific information about
oxidation, since RP retention can be dramatically affected by the oxidation of
protein residues, such as methionine.3
Mapping IgG Subunit Glycoforms Using HILIC and a Wide-Pore
Amide Stationary Phase
Matthew A. Lauber and Stephan M. Koza
Waters Corporation, Milford, MA, USA
Cleavage
Site
-G—G-
IdeS
Digestion
(Fab )2
2x Light Chain
2x Fd
2x Fc/2
Denaturation
Reduction
2x Fc/2
Figure 1. IdeS digestion and reduction scheme for preparing IgG LC, Fd', and Fc/2 subunits.
[ 46 ]
E X P E R IM E N TA L
Sample description
IdeS digestion and reduction of mAbs:
Formulated trastuzumab was diluted 7 fold into 20 mM
phosphate (pH 7.1) and incubated at a concentration of 3 mg/mL
with IdeS (Promega, Madison, WI) for 30 minutes at 37 °C at
a 50:1 w/w ratio of trastuzumab to IdeS. The resulting
IdeS-digested antibody was denatured and reduced by the
addition of 1M TCEP (tris(2-carboxyethyl)phosphine) and solid
GuHCl (guanidine hydrochloride). The final buffer composition for
the denaturation/reduction step was approximately 6 M GuHCl,
80 mM TCEP, and 10 mM phosphate (pH 7.1). IdeS-digested
trastuzumab (1.5 mg/mL) was incubated in this buffer at 37 °C
for 1 hour. An IdeS digested, reduced sample of an IgG1K mAb
obtained from NIST as candidate reference material #8670
(lot #3F1b) was prepared in the same manner.
Cetuximab IdeS/carboxypeptidase B digestion
and reduction:
Prior to digestion with IdeS,10 cetuximab was treated with
carboxypeptidase B to complete the partial removal of the
lysine-C-terminal residues that is typical of the antibody.4
Formulated cetuximab was mixed with carboxypeptidase B
(223 µ/mg, Worthington, Lakewood, NJ) at a ratio of 100:1
(w/w), diluted into 20 mM phosphate (pH 7.1), and incubated
at a concentration of 1.8 mg/mL for 2 hours at 37 °C. The
carboxypeptidase B treated cetuximab was then added to
100 units of IdeS and incubated for 30 minutes at 37 °C. The
resulting IdeS digest was denatured and reduced by the addition
of 1 M TCEP and solid GuHCl. The final buffer composition for
the denaturation/reduction step was approximately 6 M GuHCl,
80 mM TCEP, and 10 mM phosphate (pH 7.1). IdeS-digested
cetuximab (0.9 mg/mL) was incubated in this buffer at 37 °C
for 1 hour.
Preparation of RapiFluor-MS Labeled N-Glycans
from Cetuximab:
RapiFluor-MS labeled N-glycans were prepared from
cetuximab using a GlycoWorks RapiFluor-MS N-Glycan Kit
(p/n 176003606) according to the guidelines provided in
its Care and Use Manual (715004793).
Method conditions
(unless otherwise noted)
Column conditioning
ACQUITY UPLC Glycoprotein BEH Amide, 300Å, 1.7 µm columns
(as well as other amide columns intended for glycoprotein
separations) should be conditioned via two sequential injections
and separations of 40 µg Glycoprotein Performance Test
Standard (p/n 186008010; 10 µL injections of 4 mg/mL in
0.1% trifluoroacetic acid [TFA], 80% acetonitrile [ACN]) or with
equivalent loads of a sample for which the column has been
acquired. The separation outlined by the following method can
be employed for conditioning with the Glycoprotein Performance
Test Standard.
Column conditioning gradient:
Column dimensions: 2.1 x 150 mm
Mobile phase A:
0.1% (v/v) TFA, water
Mobile phase B:
0.1% (v/v) TFA, ACN
Time
(min)
%A
%B
Curve
0.0 15.0 85.0 6
0.5 15.0 85.0 6
1.0 33.0 67.0 6
21.0 40.0 60.0 6
22.0 100.0 0.0
6
24.0 100.0 0.0
6
25.0 15.0 85.0 6
35.0 15.0 85.0 6
LC conditions for IgG subunit separations
LC system:
ACQUITY UPLC H-Class Bio System
Sample temp.:
5 °C
Analytical
column temp.:
45 °C (trastuzumab and NIST IgG1K
subunit HILIC separations)
60 °C (cetuximab subunit HILIC
separations)
80 °C (trastuzumab reversed phase (RP)
subunit separations)
Flow rate:
0.2 mL/min
Mobile phase A:
0.1% (v/v) TFA, water
Mobile phase B:
0.1% (v/v) TFA, ACN
UV detection:
214 nm, 10 Hz
[ 47 ]
Mapping IgG Subunit Glycoforms Using HILIC and a Wide-Pore Amide Stationary Phase
Injection volume:
≤1.2 µL (aqueous diluents). Note: It might
be necessary to avoid high organic
diluents for some samples due to the
propensity for proteins to precipitate
under ambient conditions. A 2.1 mm
I.D. column can accommodate up to
a 1.2 µL aqueous injection before
chromatographic performance is
negatively affected.
Waters columns:
ACQUITY UPLC Glycoprotein BEH
Amide, 300Å, 1.7 µm, 2.1 x 150 mm
(p/n 176003702, with Glycoprotein
Performance Test Standard);
ACQUITY UPLC Glycan BEH Amide,
130Å, 1.7 µm, 2.1 x 150 mm
(p/n 186004742);
ACQUITY UPLC Protein BEH C4, 300Å,
1.7 µm, 2.1 x 150 mm (p/n 186004497)
Other columns:
Agilent® AdvanceBio Glycan Mapping,
1.8 μm, 2.1 x 150 mm;
Thermo Scientific® Accucore™ Amide
150 HILIC, 150Å 2.6 μm, 2.1 x 150 mm
Vials:
Polypropylene 12 x 32 mm Screw Neck
Vial, 300 µL volume (p/n 186002640)
Gradient used for reversed-phase (RP) separations of
trastuzumab subunits (Figure 2A):
Time
(min)
%A
%B
Curve
0.0 95.0 5.0 6
1.0 66.7 33.3 6
21.0 59.7 40.3 6
22.0 20.0 80.0 6
24.0 20.0 80.0 6
25.0 95.0 5.0
6
35.0 95.0 5.0
6
Gradient used for HILIC separations of IgG subunits (Figures 2–7):
Time
(min) %A
%B Curve
0.0 20.0 80.0 6
1 30.0 70.0 6
21 37.0 63.0 6
22 100.0 0.0 6
24 100.0 0.0 6
25 20.0 80.0 6
35 20.0 80.0 6
MS conditions for IgG subunit separations
MS system:
Xevo G2 QTof or SYNAPT G2-S HDMS
Ionization mode:
ESI+
Analyzer mode:
Resolution (~20 K)
Capillary voltage:
3.0 kV
Cone voltage:
45 V
Source temp.:
150 °C
Desolvation temp.:
350 °C
Desolvation gas flow: 800 L/Hr
Calibration:
NaI, 2 µg/µL from 500–5000 m/z
Acquisition:
500–4000 m/z, 0.5 sec scan rate
Data management:
MassLynx® Software (v4.1)/UNIFI V1.7
LC conditions for RapiFluor-MS Released N-Glycan
HILIC separations:
LC system:
ACQUITY UPLC H-Class Bio System
Sample temp.:
10 °C
Analytical
column temp.:
60 °C
Fluorescence detection: Ex 265/Em 425 nm (RapiFluor-MS)
(5 Hz scan rate [50 mm column], Gain =1)
Injection volume:
10 µL (DMF/ACN diluted sample)
Mobile phase A:
50 mM ammonium formate, pH 4.4
(LC-MS grade; from a 100x concentrate,
p/n 186007081)
Mobile phase B:
ACN (LC-MS grade)
Columns:
ACQUITY UPLC Glycan BEH Amide, 130Å,
1.7 µm, 2.1 x 50 mm (p/n 186004740)
Vials:
Polypropylene 12 x 32mm, 300 μL,
Screw Neck Vial, (p/n 186002640)
Gradient used for RapiFluor-MS N-Glycan HILIC Separations
(Figure 7B):
Time Flow Rate
(min) (mL/min) %A
%B Curve
0.0
0.4 25 75 6
11.7
0.4
46 54
6
12.2
0.2 100 0
6
13.2
0.2 100 0
6
14.4
0.2
25 75
6
15.9
0.4
25 75
6
18.3
0.4
25 75
6
[ 48 ] Mapping IgG Subunit Glycoforms Using HILIC and a Wide-Pore Amide Stationary Phase
MS conditions for RapiFluor-MS N-Glycan HILIC
separations
MS system:
SYNAPT G2-S HDMS
Ionization mode:
ESI+
Analyzer mode:
TOF MS, resolution mode (~20 K)
Capillary voltage:
3.0 kV
Cone voltage:
80 V
Source temp.:
120 °C
Desolvation temp.:
350 °C
Desolvation gas flow: 800 L/Hr
Calibration:
NaI, 1 µg/µL from 500–2500 m/z
Lockspray
(ASM B-side):
100 fmol/µL Human Glufibrinopeptide B
in 0.1% (v/v) formic acid,
70:30 water every 90 seconds
Acquisition:
500–2500 m/z, 1 Hz scan rate
Data management:
MassLynx Software (v4.1)
It should, however, be kept in mind that many IgG modifications
more strongly elicit changes in the hydrophilicity of a mAb
along with its capacity for hydrogen bonding. A very obvious
example of this type of modification is glycosylation. Glycans
released from a mAb are very often profiled by hydrophilic
interaction chromatography (HILIC), in which case an amide
bonded stationary phase has historically been used, because it
affords high retentivity as a consequence of its hydrophilicity and
propensity for hydrogen bonding.7 Here, we propose that HILIC
with an amide bonded stationary phase also be considered for
IgG subunit separations. For such an application, a stationary
phase with a wide average pore diameter is critical, so that large
subunit structures will have access to the majority of the porous
network and be less prone to restricted diffusion while eluting
through a column.8-9 Through the development of a sub-2-μm
wide-pore amide stationary phase, we have facilitated a novel
and complementary workflow to RP based subunit analyses. In
this application note, we demonstrate the use of a glycoprotein
BEH amide, 300Å, 1.7 μm column to develop LC-MS and LC-UV
techniques that can be used to rapidly profile domain specific
information about the N-linked glycosylation of IgG molecules.
R E SU LT S A N D D IS C U S S IO N
Orthogonal, complementary IgG subunit separations
To demonstrate a conventional approach to IgG subunit mapping,
we first analyzed a reduced\IdeS digest of an IgG1 mAb using
a RP chromatographic separation with a wide-pore C4 bonded
stationary phase (Protein BEH C4, 300Å, 1.7 μm). The IgG1 mAb
selected for this work was trastuzumab, given its prominence as
a first generation mAb drug product and a potential target for
biosimilar development.11 Figure 2A shows a UPLC chromatogram
that is typical for reduced, IdeS-digested trastuzumab, wherein
three peaks are near equally spaced with an elution order
corresponding to the Fc/2, LC and Fd' subunits, respectively. The
conditions to produce this high resolution separation entail the
use of TFA for ion-pairing. Interestingly, the same mobile phases
have proven to be optimal for protein HILIC, as they reduce
the hydrophilicity of protein residues by masking them via a
hydrophobic ion pair. This, in turn, leads to improved selectivity
for hydrophilic modifications.12 That is, an orthogonal method to
the RP separation can be achieved via HILIC by simply reversing
a gradient and using a newly developed wide-pore amide bonded
stationary phase (glycoprotein BEH Amide, 300Å, 1.7 μm).
[ 49 ]
Mapping IgG Subunit Glycoforms Using HILIC and a Wide-Pore Amide Stationary Phase
An example of a chromatogram obtained from a
column packed with this wide-pore amide material
is shown in Figure 2B. Here, the same reduced,
IdeS digested trastuzumab is separated into
approximately 10 peaks. The first two eluting
peaks correspond to the Fd' and LC subunits, while
the remaining, more strongly retained peaks
correspond to the glycoforms of the Fc/2 subunit.
By focusing on the more strongly retained peaks,
an analyst can elucidate information about the
heterogeneity of glycosylation (Figure 3A). Given
that this is a method with volatile mobile phases,
the glycoform peaks can be readily interrogated
by ESI-MS. Deconvoluted mass spectra and
molecular weights corresponding to species in the
glycoform profile are presented in Figures 3B and
3C. In Figure 3, chromatographic peaks are labeled
with the same color as their corresponding mass
spectra. Notice that this HILIC separation facilitates
producing deconvoluted mass spectra for individual
glycoforms with limited interference between
similar molecular weight species, for instance the
Fc/2+A2G1 versus the Fc/2+FA2 species (orange
versus blue spectrum). In a first pass analysis, all
glycan species from trastuzumab that are known
to be present at a relative abundance greater than
2% are readily detected.13 It should be noted that
lower abundance species, such as Fc/2+M5 (Man5),
are also detected and can be observed by extracted
ion chromatograms (XICs). This indicates there is a
possibility to perform selected reaction monitoring
(SRM) MS analyses when and if there is a need to
monitor particular low abundance structures. While
it is not resolved under these conditions, the M5
Fc/2 glycoform is resolved in a different example
separation (see below, Figure 7A).
0.00
0.04
0.08
0.12
0.16
6
7
8
9
10
11
12
13
14
15
16
A
214
Time (min)
0.00
0.06
0.12
0.18
0.24
6
7
8
9
10
11
12
13
14
15
16
A
214
Time (min)
Fc/2 Glycoforms
Fd’
LC
Protein BEH C4, 300Å, 1.7 µm
LC
Fd’
Glycoprotein BEH Amide, 300Å, 1.7 µm
Fc/2
A
B
+Glycans
Figure 2. Trastuzumab subunit separations. (A) 1 µg of reduced, IdeS digested separated using
an ACQUITY UPLC Protein BEH C4, 300Å, 1.7 µm Column (0.7 µL aqueous injection). (B) 1 µg of
reduced, IdeS digested separated using an ACQUITY UPLC Glycoprotein BEH Amide, 300Å, 1.7 µm
Column (0.7 µL aqueous injection).
Figure 3. Profiling trastuzumab Fc/2 subunit glycoforms. (A) Retention window from Figure 2B
corresponding to the glycoform separation space. (B) Deconvoluted ESI mass spectra for the
HILIC chromatographic peaks. Chromatographic peaks are labeled with the same color as their
corresponding mass spectra. (C) Molecular weights for the observed trastuzumab subunits.
Species
MWAvg
Theoretical
(Da)
MWAvg
Observed
(Da)
Fd
25383.6
25383.3
LC
23443.1
23443.1
Fc/2+A2
25090.2
25091.0
Fc/2+FA2
25236.3
25236.9
Fc/2+A2G1
25252.3
25248.7
Fc/2+FA2G1
25398.5
25398.5
Fc/2+FA2G1
25398.5
25399.1
Fc/2+FA2G2
25560.6
25561.8
B
C
0.01
0.02
0.03
0.04
0.05
A
21
4
Time (min)
Fc/2
+FA2
Fc/2
+FA2G1
Fc/2
+FA2G2
Fc/2
+FA2G1’
Fc/2
+A2G1
Fc/2
+A2
A
24900
25300
25700
Molecular Weight Da
25091.0 Da
25236.9 Da
25248.7 Da
25398.5 Da
25399.1 Da
255561.8 Da
[ 50 ] Mapping IgG Subunit Glycoforms Using HILIC and a Wide-Pore Amide Stationary Phase
Batch-to-batch analysis of trastuzumab
Fc/2 glycosylation by HILIC-UV profiling
Clearly, data generated by subunit-level HILIC-MS
are very information-rich. Optical detection
based subunit HILIC separations can be equally
informative. To this end, we have applied a HILIC-UV
method to perform batch-to-batch analysis of
trastuzumab Fc/2 glycosylation, as exemplified
in Figure 4. Two example HILIC chromatograms for
Fc/2 glycoforms obtained from two different lots
of trastuzumab are shown in Figure 4A. Previous
testing on these lots has demonstrated differences
in glycosylation at the released glycan level.14
Here, by integration of peaks across the profile,
we have found that the two lots of trastuzumab
indeed differ with respect to their Fc domain
glycosylation profiles, in ways consistent with the
mentioned released glycan assays. In particular,
these lots of trastuzumab differ with respect to their
extents of terminal galactosylation, as estimated
from the abundances of FA2, FA2G1, and FA2G2
Fc/2 subunits (Figure 4B). This is an informative
observation, since the extent of galactosylation can
affect complement-dependent cytotoxicity (CDC).15
Lifetime testing of glycoprotein BEH amide
300Å, 1.7μm columns for profiling IgG
subunit glycoforms
The ability of a BEH amide, 300Å, 1.7 μm column
to robustly deliver the above mentioned separations
over time was tested by performing a series of
experiments involving a single column being
subjected to 300 sequential injections of a reduced,
IdeS digested trastuzumab sample. This was a
potentially challenging use scenario given that
the reduced, IdeS digested mAb sample contains
both high concentrations of guanidine denaturant
and TCEP reducing agent. Total ion chromatograms
corresponding to the 20th, 180th, and 300th
injections of this experiment are displayed in Figure
5A. In these analyses, particular attention was paid
to the half-height resolution of the Fc/2+A2 and
Fc/2+FA2 species, which was assessed every 20th
separation using extracted ion chromatograms (XICs).
Batch 1
Batch 2
0
10
20
30
40
50
1
2
3
4
5
6
7
8
% Amount
Component
Batch 1
Batch 2
FA2
FA2G1
FA2G2
A
B
Figure 4. Batch-to-batch profiling of trastuzumab Fc/2 subunit glycoforms. (A) HILIC
chromatograms of trastuzumab Fc/2 subunit glycoforms from two different lots of drug product.
(B) Relative abundances of the major sample components. Analyses were performed in triplicate
using an ACQUITY UPLC Glycoprotein BEH Amide, 300Å, 1.7 µm, 2.1 x 150 mm Column.
Figure 5. Lifetime testing of an ACQUITY UPLC Glycoprotein BEH Amide, 300Å, 1.7 µm,
2.1 x 150 mm Column for sequential injections of reduced, IdeS digested trastuzumab. (A) Total
ion chromatograms (TICs) from the 20th, 180th, and 300th injections. Example extracted ion
chromatograms (XICs) for Fc/2+A2 and Fc/2+FA2 that were used to measure resolution.
(B) Chromatographic parameters observed across the 300 injection lifetime test. Each panel
shows results for each 20th injection, including retention time (RT) of the FA2 glycoform,
Rs between A2 and FA2 glycoforms, maximum pressure across the run, and % carryover as
measured by a repeat gradient and XICs.
3000
4000
5000
6000
7000
0
1
2
3
4
5
6
2
4
6
8
10
0.0
0.1
0.2
0.3
0.4
0
100
200
300
Injection
Fc/2+FA2
1010.0-1011.0
m/z
Fc/2+A2
1004.2-1005.2
m/z
XIC
Rs 2.17
20th Injection
A
B
RT
(Fc/2+FA2
, min)
R
s
(Fc/2+
A2
/FA
2)
Max Pressur
e
(psi)
% Carryove
r
(Fc/2+FA2
)
Fc/2
+A2
Fc/2
+FA2
Fc/2
+A2
Fc/2
+FA2
Fc/2
+A2
Fc/2
+FA2
[ 51 ]
Mapping IgG Subunit Glycoforms Using HILIC and a Wide-Pore Amide Stationary Phase
In this testing, several additional chromatographic parameters were also monitored, including the retention
time of the Fc/2+FA2 species, the maximum system pressure observed during the chromatographic run, and
the percent (%) carryover of the most abundant glycoform, the Fc/2+FA2 species (Figure 5B). Plots of these
parameters underscore the consistency of the subunit separation across the lifetime of the column. With
noteworthy consistency, the column produced relatively stable retention times, a consistent resolution of the
A2 and FA2 glycoforms (Rs≈2), a maximum system pressure consistently at only ~6 Kpsi, and a remarkably
low carryover between 0.1 and 0.2%. This latter aspect of the HILIC separations is particularly noteworthy
since it indicates that carryover with these methods is almost an order of magnitude lower than analogous
C4 based RP methods.
Benchmarking the capabilities of the glycoprotein BEH amide, 300Å, 1.7 μm column
We have benchmarked the performance of this new wide-pore column technology against not only its standard
pore diameter analog but also its two most closely related, commercially-available alternatives. Figure 6
presents chromatograms obtained for a reduced, IdeS digested sample of an IgG1K mAb acquired from NIST
using these different column technologies. In a visual comparison, it is clear that the glycoprotein BEH amide
300Å column significantly outperforms the other three columns. To quantify this assessment, peak-to-
valley ratios were calculated for the separation of the FA2 glycoform away from the FA2G1 glycoform. The
glycoprotein BEH amide 300Å column was found to demonstrate improvements of 48%, 152%, and 261%
over the 130Å glycan BEH amide, the Accucore amide, and AdvanceBio glycan mapping columns. This mAb
sample also has a particularly interesting attribute in that it has a reasonably high relative abundance of an
immunogenic alpha-1,3-galactose containing glycan (an FA2G2Ga1 structure).16-17 As shown in Figure 6, this
Fc/2+FA2G2Ga1 species can be readily visualized with the wide-pore amide column. This represents a sizeable
improvement in the peak capacity of large molecule HILIC separations for this emerging application.
0E+0
2E+8
7.5
12.5
In
te
ns
it
y
7.0
12.0 6.5
11.5 9.5
14.5
Glycoprotein BEH Amide,
300Å, 1.7 m
Glycan BEH Amide,
130Å, 1.7 m
Accucore Amide,
150Å, 2.6 m
AdvanceBio
Glycan Mapping,
1.8 µm
Fc/2
+ FA2
Fc/2
+ FA2G1
p/v
8.3
p/v
5.6
p/v
3.3
p/v
2.3
Fc/2
+ FA2G2Ga1
Figure 6. Subunit glycoform profiles of an IgG1K obtained with various 2.1 x 150 mm columns: ACQUITY UPLC Glycoprotein BEH
Amide, 300Å, 1.7 µm Column, ACQUITY UPLC Glycan BEH Amide, 130Å, 1.7 µm Column, Thermo Scientific Accucore Amide,
150Å, 2.6 µm column, and Agilent AdvanceBio Glycan Mapping, 1.8 µm column. Peak-to-valley (p/v) ratios for the Fc/2+FA2
versus FA2G1 glycoforms are provided. An alpha gal containing Fc/2+FA2G2Ga1 is readily visualized with the glycoprotein BEH
amide, 300Å, 1.7 µm column.
[ 52 ] Mapping IgG Subunit Glycoforms Using HILIC and a Wide-Pore Amide Stationary Phase
Figure 7. HILIC Profiling of
cetuximab glycosylation. (A)
HILIC-UV chromatogram of reduced,
IdeS/carboxy peptidase B-digested
cetuximab obtained using an ACQUITY
UPLC Glycoprotein BEH Amide, 300Å,
1.7 µm, 2.1 x 150 mm Column.
Species corresponding to Fc/2 and
Fd' subunits are labeled in gray and
red, respectively. Subunit glycan
assignments based on deconvoluted
mass spectra are provided. (B) HILIC-
fluorescence chromatograms of
RapiFluor-MS labeled N-glycans
from cetuximab obtained using an
ACQUITY UPLC Glycan BEH Amide,
130Å, 1.7 µm, 2.1 x 50 mm Column.
Mass spectral data supporting the
assignments of the RapiFluor-MS
labeled N-glycans are provided.
0.01
0.02
0.03
0.04
0.05
0.06
8
9
10
11
12
13
14
15
16
17
18
A
21
4
Time (min)
Fc/2
+FA2
Fc/2
+FA2G2
Fc/2
Glycosylated
Fc/2
+M5
Fd’ pE
+ (FA2G2Ga2)
Fd’ pE
+ (FA2G2Ga1Sg1)
Fd’ pE
+ (Hex9HexNAc5DHex1)
Fd’
Glycosylated
N-term pE
Fd’ pE
+ (FA2G2Ga1)
Fd’ pE
+ (FA2G2Sg1)
Fc/2
+FA2G1
0E+0
4E+6
4
5
6
7
8
9
10
11
12
EU
Time (min)
M5
FA2
FA2G1
FA2G2
Hex9HexNAc5DHex1
FA2G2Ga1
FA2G2Ga2
FA2G2Ga1Sg1
FA2G2Sg1
Domain-Specific
Glycan
Information
High Resolution
High Sensitivity
Released
N-Glycan Profile
A
B
Species
MW
Avg
Theoretical
(Da)
MWAvg
Observed
(Da)
Mass
Error
(Da)
LC
23427.0
23427.1
0.1
Fc/2-K
+ FA2
25236.3
25237.4
1.1
Fc/2-K
+ M5
25008.1
25008.8
0.7
Fc/2-K
+ FA2G1
25398.5
25399.8
1.3
Fc/2-K
+ FA2G2
25560.6
25562.0
1.4
Fd' pE
+ FA2G2Ga1
27385.5
27386.8
1.3
Fd' pE
+ FA2G2Sg1
27530.6
27531.8
1.2
Fd' pE
+ FA2G2Ga2
27547.6
27548.2
0.6
Fd' pE +
FA2G2Ga1Sg1
27692.7
27693.1
0.4
Fd' pE +
Hex9HexNAc5DHex1
28075.1
28075.3
0.2
Species
MW
Mono
Theoretical
(Da)
MW
Mono
Observed
(Da)
Mass
Error
(ppm)
FA2
1545.6080
1545.6136
3.6
M5
1773.7190
1773.7242
2.9
FA2G1
1935.7719
1935.7834
5.9
FA2G2
2097.8247
2097.8136
-5.3
FA2G2Ga1
2259.8775
2259.8860
3.8
FA2G2Sg1
2404.9150
2404.9150
0.0
FA2G2Ga2
2421.9303
2421.9320
0.7
FA2G2Ga1Sg1
2566.9678
2566.9792
4.4
Hex9HexNAc5DHex1 2949.1154
2949.1424
9.2
Complementing RapiFluor-MS N-glycan analyses with
domain specific information about mAb glycosylation
One of the key advantages to profiling IgG subunits by HILIC
is being able to elucidate domain specific information about
glycosylation. In an IgG structure, there exists one conserved
N-glycosylation site at Asn297 of the heavy chain. As a
consequence, most IgGs will be modified with two glycans in
the CH2 domains (constant heavy chain 2 domains) of the Fc
subunit. However, it is estimated that 20% of human IgGs are also
modified in their CH1 domains, which reside in the Fab subunits,
and more specifically the IdeS generated Fd' subunit.18-19 For
example, it is known that cetuximab, a chimeric mAb expressed
from a murine cell line, is glycosylated in both its CH1 and
CH2 domains.20 Characterization of this mAb has thus proven
to be an interesting case study for the application of our newly
developed techniques. HILIC separations obtained for a reduced,
IdeS digested sample of carboxypeptidase B treated cetuximab
showed only one weakly retained subunit species, which could
be easily assigned to the LC subunit by online ESI-MS (data not
shown). Furthermore, and as shown in Figure 7A, the glycoform
retention window for cetuximab was populated with twice as
many peaks as had been observed for trastuzumab and its
glycosylated Fc/2 subunit. Deconvoluted ESI-MS data from these
HILIC-MS separations confirmed that the first grouping of peaks
(labeled in gray) corresponded to Fc/2 glycoforms and typical mAb
glycan species, such as FA2, FA2G1, M5, and FA2G2. Meanwhile,
the second grouping of peaks were found to be distinctively
related to glycoforms of the Fd' subunit given their unique masses.
Curiously, each of the identified Fd’ glycoforms (labeled in red) are
immunogenic in nature, containing either non-human alpha-1,3-
galactose or non-human N-glycolyl-neuraminic acid epitopes.21
The identification of these glycan species has been confirmed
through released N-glycan analyses. Using the newly developed
GlycoWorks RapiFluor-MS N-Glycan Kit,22 cetuximab N-glycans
were rapidly prepared and labeled with the novel fluorescence and
MS-active labeling reagent, RapiFluor-MS. The resulting labeled
N-glycans were subsequently separated using a glycan BEH amide,
130Å, 1.7 μm column and detected by fluorescence and positive
ion mode ESI-MS, as portrayed in Figure 7B. The sensitivity gains
afforded by the RapiFluor-MS label facilitated making confident
assignments of the released N-glycan structures. The species
that have been assigned as a result of both this released glycan
analysis as well as the subunit HILIC-UV-MS method are supported
by previous reports on cetuximab glycosylation.6,20
[ 53 ]
Mapping IgG Subunit Glycoforms Using HILIC and a Wide-Pore Amide Stationary Phase
With the combination of released glycan and subunit-derived glycan
information, cetuximab glycosylation has been characterized with significant
detail. With the RapiFluor-MS released glycan analysis, a very high resolution
separation has been achieved with an LC-MS compatible method in which
glycans can even be subjected to detailed MS/MS analyses. With an equally
MS-compatible subunit HILIC separation, domain-specific glycan information
has been readily obtained with minimal sample preparation. Each method has
therefore provided complementary information on the glycosylation of the
mAb. Nevertheless, the widepore amide HILIC method stands out as a useful
technique for rapidly screening mAbs for multidomain glycosylation.
C O N C LU S IO NS
Subunit analyses of mAbs represent a useful strategy for rapidly investigating
domain-specific modifications. The combination of high fidelity IdeS proteolysis
with high resolution LC-UV-MS has presented a new approach to mAb identity
testing and assaying oxidation.3 The current subunit mapping strategies have
exclusively relied upon reverse phase chromatography. However, since N-linked
glycosylation of IgG proteins elicit dramatic changes in hydrophilicity and
hydrogen bonding characteristics, a separation by hydrophilic interaction
chromatography (HILIC) can be effectively used for this application or as a
complementary method to reversed-phase separations since the same mobile
phases can be employed. For this reason, we have proposed the use of HILIC with
an amide bonded stationary phase that has been optimized for large molecule
separations, the wide-pore glycoprotein BEH amide, 300Å, 1.7 μm stationary
phase. Along with new developments in released N-glycan analysis afforded
by RapiFluor-MS,22 the glycoprotein BEH amide, 300Å, 1.7 μm column enables
new possibilities for routine monitoring and detailed characterization of mAb
glycosylation, including elucidation of domain-specific glycan information.
[ 54 ] Mapping IgG Subunit Glycoforms Using HILIC and a Wide-Pore Amide Stationary Phase
Waters Corporation
34 Maple Street
Milford, MA 01757 U.S.A.
T: 1 508 478 2000
F: 1 508 872 1990
www.waters.com
Waters, The Science of What’s Possible, ACQUITY UPLC, Oasis, MassLynx, SYNAPT, Xevo, and Empower are registered trademarks
of Waters Corporation. GlycoWorks and RapiFluor-MS are trademarks of Waters Corporation. All other trademarks are the property
of their respective owners.
©2015 Waters Corporation. Produced in the U.S.A. April 2015 720005385EN AG-PDF
References
1. Aggarwal, R. S., What’s fueling the biotech engine – 2012 to 2013.
Nat Biotechnol 2014, 32 (1), 32–9.
2. Wang, B.; Gucinski, A. C.; Keire, D. A.; Buhse, L. F.; Boyne, M. T., 2nd,
Structural comparison of two anti-CD20 monoclonal antibody drug products
using middle-down mass spectrometry. Analyst 2013, 138 (10), 3058–65.
3. Gucinski, A. C., Rapid Characterization and Comparison of Stressed
anti-CD20 Drugs using Middle Down Mass Spectrometry. In 61st ASMS
Conference on Mass Spectrometry and Allied Topics, Minneapolis, MN, 2013.
4. Ayoub, D.; Jabs, W.; Resemann, A.; Evers, W.; Evans, C.; Main, L.; Baessmann,
C.; Wagner-Rousset, E.; Suckau, D.; Beck, A., Correct primary structure
assessment and extensive glyco-profiling of cetuximab by a combination
of intact, middle-up, middle-down and bottom-up ESI and MALDI mass
spectrometry techniques. MAbs 2013, 5 (5), 699–710.
5. Alain Beck; Hélène Diemer; Daniel Ayoub; François Debaene; Elsa Wagner-
Rousset; Christine Carapito; Alain Van Dorsselaer; Sanglier-Cianférani, S.,
Analytical characterization of biosimilar antibodies and Fc-fusion proteins.
Trends in Analytical Chemistry 2013, 48, 81–95.
6. Janin-Bussat, M. C.; Tonini, L.; Huillet, C.; Colas, O.; Klinguer-Hamour, C.;
Corvaia, N.; Beck, A., Cetuximab Fab and Fc N-glycan fast characterization
using IdeS digestion and liquid chromatography coupled to electrospray
ionization mass spectrometry. Methods Mol Biol 2013, 988, 93–113.
7. Ahn, J.; Bones, J.; Yu, Y. Q.; Rudd, P. M.; Gilar, M., Separation of
2-aminobenzamide labeled glycans using hydrophilic interaction
chromatography columns packed with 1.7 microm sorbent. J Chromatogr B
Analyt Technol Biomed Life Sci 2010, 878 (3–4), 403–8.
8. Gustavsson, P.-E.; Larsson, P.-O., Support Materials for Affinity
Chromatography. In Handbook of Affinity Chromatography, Hage,
D., Ed. Taylor & Francis: Boca Raton, FL, 2006; pp 15–33.
9. Renkin, E. M., J. Gen. Physio. 1954, (38), 225.
10. Tetaz, T.; Detzner, S.; Friedlein, A.; Molitor, B.; Mary, J. L., Hydrophilic
interaction chromatography of intact, soluble proteins. J Chromatogr A
2011, 1218 (35), 5892–6.
11. Beck, A.; Sanglier-Cianferani, S.; Van Dorsselaer, A., Biosimilar, biobetter,
and next generation antibody characterization by mass spectrometry.
Anal Chem 2012, 84 (11), 4637–46.
12. Lauber, M. A.; Koza, S. M., Waters Application Note ****to be updated 2015.
13. Xie, H.; Chakraborty, A.; Ahn, J.; Yu, Y. Q.; Dakshinamoorthy, D. P.; Gilar,
M.; Chen, W.; Skilton, S. J.; Mazzeo, J. R., Rapid comparison of a candidate
biosimilar to an innovator monoclonal antibody with advanced liquid
chromatography and mass spectrometry technologies. MAbs 2010, 2 (4).
14. Yu, Y. Q.; Ahn, J.; Gilar, M., Trastuzumab Glycan Batch-to-Batch Profiling
using a UPLC/FLR/MS Mass Spectrometry Platform. Waters Application Note
720003576en.
15. Raju, T. S.; Jordan, R. E., Galactosylation variations in marketed therapeutic
antibodies.
MAbs 2012, 4 (3), 385–91.
16. Schiel, J.; Wang, M.; Formolo, T.; Kilpatrick, L.; Lowenthal, M.; Stockmann,
H.; Phinney, K.; Prien, J. M.; Davis, D.; Borisov, O. In Biopharmaceutical
Characterization: Evaluation of the NIST Monoclonal Antibody Reference
Material, 62nd Conference on Mass Spectrometry and Allied Topics,
Baltimore, MD, Baltimore, MD, 2014.
17. Bosques, C. J.; Collins, B. E.; Meador, J. W., 3rd; Sarvaiya, H.; Murphy, J. L.;
Dellorusso, G.; Bulik, D. A.; Hsu, I. H.; Washburn, N.; Sipsey, S. F.; Myette,
J. R.; Raman, R.; Shriver, Z.; Sasisekharan, R.; Venkataraman, G., Chinese
hamster ovary cells can produce galactose-alpha-1,3-galactose antigens on
proteins. Nat Biotechnol 2010, 28 (11), 1153–6.
18. Jefferis, R., Glycosylation of Recombinant Antibody Therapeutics.
Biotechnol Prog 2005, (21), 11–16.
19. Huang, L.; Biolsi, S.; Bales, K. R.; Kuchibhotla, U., Impact of variable
domain glycosylation on antibody clearance: an LC/MS characterization.
Anal Biochem 2006, 349 (2), 197–207.
20. Qian, J.; Liu, T.; Yang, L.; Daus, A.; Crowley, R.; Zhou, Q., Structural
characterization of N-linked oligosaccharides on monoclonal antibody
cetuximab by the combination of orthogonal matrix-assisted laser
desorption/ionization hybrid quadrupole-quadrupole time-of-flight tandem
mass spectrometry and sequential enzymatic digestion. Anal Biochem 2007,
364 (1), 8–18.
21. Arnold, D. F.; Misbah, S. A., Cetuximab-induced anaphylaxis and IgE specific
for galactose-alpha-1,3-galactose. N Engl J Med 2008, 358 (25), 2735;
author reply 2735–6.
22. Lauber, M. A.; Brousmiche, D. W.; Hua, Z.; Koza, S. M.; Guthrie, E.; Magnelli,
P.; Taron, C. H.; Fountain, K. J., Rapid Preparation of Released N-Glycans for
HILIC Analysis Using a Novel Fluorescence and MS-Active Labeling Reagent.
Waters Application Note 720005275EN 2015.
[ 55 ]
Mapping IgG Subunit Glycoforms Using HILIC and a Wide-Pore Amide Stationary Phase
WAT E RS SO LU T IONS
Biopharmaceutical Platform Solution
with UNIFI®
ACQUITY UPLC® H-Class Bio System
ACQUITY UPLC Tunable UV Detector
UNIFI Scientific Information System
ACQUITY UPLC Protein BEH C4 Column
K E Y WO R DS
Biosimilar, intact mass analysis, intact
mass subunit analysis, light chain,
heavy chain, glycosylation, glycoprofile,
infliximab, mAb, biotherapeutic
characterization
A P P LIC AT ION BEN E FIT S
A streamlined workflow within an integrated
UPLC®-MS/MS system solution that features
automated data acquisition, processing, and
reporting, and that is deployable to both
regulated and non-regulated laboratories,
enables efficient structural analysis and
comparative analysis of multiple batches of an
innovator product and its biosimilar candidate
within a comparability study.
INT RO DUC T ION
The expiration of patents and other intellectual property rights for originator
biologics over the next decade opens up ample opportunities for biosimilars to
enter the market and push industry competition to a high level.1-4
Compared to small molecule drugs, biopharmaceuticals have much more
complex structures and are more expensive to develop. The complexity of the
biopharmaceutical molecular entity puts greater challenges on organizations
seeking to manufacture safe and effective biosimilar products for patients.
Regulatory bodies such as the U.S. FDA and EMA5-8 require a demonstration of
comprehensive characterization for the drug substance: Confirming primary
sequence and identifying post-translational modifications (PTMs), establishing
biophysical and functional comparability for the innovator and candidate
biosimilar, and performing studies that establish expected variation within an
innovator biotherapeutic.
Infliximab (Remicade) is a monoclonal antibody (mAb) used to treat autoimmune
diseases; it was first approved by the FDA for the treatment of Crohn’s disease in
1998, and in 2013 two biosimilars have been submitted for approval in Europe.
In this application note, we characterize infliximab and a biosimilar candidate,
produced in a different cell line, using Waters® Biopharmaceutical Platform
Solution with UNIFI Scientific Information System. The objective is to screen
multiple lots of both the innovator and biosimilar products at the subunit level
(light chain (LC) and heavy chain (HC)) to establish comparability at this higher
level of structure. Lot-to-lot and batch-to-batch comparisons will show product
variation, illustrating the range of quality attributes to be considered in a
candidate biosimilar.
Structural Comparison of Infliximab and a Biosimilar via Subunit Analysis
Using the Waters Biopharmaceutical Platform with UNIFI
Henry Shion and Weibin Chen
Waters Corporation, Milford, MA, USA
[ 56 ]
E X P E RIM ENTA L
Biopharmaceutical Platform Solution with UNIFI
■
■
ACQUITY UPLC H-Class Bio System
■
■
Xevo G2-S QTof
■
■
ACQUITY UPLC Tunable UV Detector
■
■
UNIFI Scientific Information System
Intact protein LC-MS conditions
Column:
ACQUITY UPLC Protein BEH C4 Column,
300Å, 1.7 µm, 2.1 mm x 50 mm
(p/n 186004495)
Column temp.:
80 °C
Mobile phase A:
Water
Mobile phase B:
Acetonitrile
Mobile phase C:
Not used
Mobile phase D:
0.5% TFA (in water)
Detection:
UV 280 nm
LC gradient table:
Time
Flow
(min) (mL/min) %A
%B
%C
%D
Curve
Initial
0.20
65.2
29.8
0
5.0
Initial
12.0
0.20
63.5
31.5
0
5.0
6
14.0
0.20
63.5
31.5
0
5.0
6
14.1
0.20
10.0 85.0 0
5.0
6
15.1
0.20
10.0 85.0 0
5.0
6
15.2
0.20
65.2 29.8
0
5.0
6
18.0
0.20
65.2 29.8
0
5.0
6
Total run time: 20.0 min
MS conditions
Capillary:
3.0 kV
Sampling cone:
80 V
Extraction cone:
4 V
Source temp.:
125 °C
Desolvation temp.:
350 °C
Cone gas flow:
0 L/Hr
Desolvation gas flow: 800 L/Hr
Data acquisition and processing
MaxEnt™ 1 for MS spectra deconvolution
UNIFI Scientific Information System
Sample preparation
Three batches of innovator infliximab were acquired from
Jenssen Biotech, Inc. (Horsham, PA, USA). The batches were
produced by the SP2/0 mouse cell line. Three batches of
candidate biosimilar infliximab produced by an alternative
mammalian cell line (Chinese hamster ovary (CHO) were
obtained from a third-party collaborator. All of the samples
were stored at -80 °C before analysis.
A reduction buffer solution containing 25 mM NaCl, 25 mM Tris,
1 mM EDTA (pH 8.0) was made to prepare mAb subunits. For
each of the six batches, 10 μL of formulated mAb solution
(21 mg/mL, the commonly used concentration level for patient
injection) was mixed with 180 μL of reduction buffer in a
1.5 mL Eppendorf tube for a protein concentration of 1.0 mg/mL.
A concentrated dithiothreitol (DTT) solution (100 mM in H2O) was
then added to each solution to obtain a final DTT concentration
of 1.0 mM. The samples were incubated at 37 °C for 20 minutes.
The samples were briefly centrifuged, then 105 μL of each sample
was mixed with an equal volume of aqueous solution containing
3% acetonitrile and 0.1% formic acid. The final concentration of
the mAb was about 0.5 mg/mL. Triplicate injections of each sample
were made onto an ACQUITY UPLC Protein BEH C4 Column,
300Å, 1.7 µm, 2.1 mm x 50 mm (p/n 186004495) LC-MS
analysis of the mAb subunit.
[ 57 ]
Structural Comparison of Infliximab and a Biosimilar via Subunit Analysis Using the Waters Biopharmaceutical Platform with UNIFI
R E SU LT S AND DIS CUSSION
Figure 1. Reversed-phase (C4) chromatograms of the innovator infliximab (top) and a biosimilar
infliximab (bottom). The signal trace is the Total Ion Current (TIC) from the mass spectrometer.
Figure 2. The combined raw MS spectra and deconvoluted spectra in mirror image plots. The MS
spectra of the light chain (eluting around 3.5 min in Figure 1) from the innovator sample (top)
and the biosimilar sample (bottom) are displayed.
Subunit characterization for Infliximab
from two cell lines
Figure 1 shows the reversed-phase LC-MS
chromatograms from the analysis of reduced
infliximab from both the innovator and biosimilar
products. There are two major components to
each chromatogram, a peak at ~3.5 minutes
and a complex set of peaks at ~10 minutes. The
chromatographic peaks eluting around 3.5 minutes
have ESI-MS measurements of 23434.0 Da,
respectively, in full agreement with the calculated
mass of the light chain of infliximab (23434.0 Da).
The complex peak eluting at ~10 minutes
is comprised of several species with MW in
the 51,000 Da range, corresponding to the
glycosylated heavy chain.
Figure 2 shows the comparison of the light chain
spectra in a mirror plot using UNIFI Scientific
Information System’s software, with the combined
raw MS spectra shown on the left panel (as
demonstrated by multiple charged spectrum
envelopes) and the MaxEnt 1 deconvoluted spectra
displayed on the right.
The results indicate that there is only one isoform
and no noticeable difference in the light chains
between the innovator and biosimilar samples.
This observation is consistent with other IgG1
biosimilar studies9 that show little or no
post-translational modifications of LC subunits.
The chromatographic profile of the heavy chains of
infliximab was more complicated than that of the
light chain. A cluster of peaks is observed around
10 minutes in Figure 1, corresponding to different
isoforms of the heavy chains, and they exhibit
significantly different chromatographic behavior
from that of the light chain. Similarly, as shown
by Figure 1 and 3, the heavy chains of infliximab
from the two cell lines show quite distinct
chromatographic and spectral differences.
[ 58 ] Structural Comparison of Infliximab and a Biosimilar via Subunit Analysis Using the Waters Biopharmaceutical Platform with UNIFI
The two major chromatographic peaks (at 9.5 min and 10.5 min) from the analysis of reduced innovator
infliximab (Figure 1A) come from the heavy chains and appear to have multiple isoforms (Figure 3C).
Mass spectrometry analysis of these peaks (Figure 3) shows that variation in both the polypeptide sequence
(+/- lysine) and glycosylation contribute to the heterogeneity of the innovator HC.
This is in contrast to the biosimilar sample, which displays a more homogeneous peak at 9.8 minutes
(Figure 1B) and fewer mass variants (Figure 3D).
Several major glycoforms (e.g. G0, G0F, G1F, G2F, and Man5) were identified for the innovator heavy chain as
shown in Figure 4, demonstrating a high degree of heterogeneity of the innovator infliximab. The biosimilar has
three major glycoforms (G0, G0F, and G1F) and no apparent amino acid variations. All of the MS peaks in the
deconvoluted spectrum can be automatically identified in UNIFI Scientific Information System’s software based
on the mAb’s reported sequence and the suspected PTM, and annotated, as displayed in Figure 4.
Figure 3. The combined raw MS
spectra mirror image comparison
between the heavy chains (eluted
around 10 min in Figure 1) for the
innovator and the biosimilar.
Figure 4. The innovator heavy chain
deconvoluted spectra reveals the
presence of glycoforms G0, G0F,
G1F, G2F, and Man5 as well as lysine
variations. Incomplete removal of
lysine from the C-terminus of the HC
is a known variant for IgG1.
[ 59 ]
Structural Comparison of Infliximab and a Biosimilar via Subunit Analysis Using the Waters Biopharmaceutical Platform with UNIFI
Assessment of batch-to-batch variability
The analysis of reduced IgG is a straightforward, high-sensitivity method that provides valuable information
on the identity and amount of related variants of mAb structure. Analysis of the reduced infliximab indicates
that its structural heterogeneity resides within the heavy chain of the antibody, and includes variation in both
glycosylation and amino acid sequence. The incomplete removal of C-terminal lysine residues is a known
structural variant, so it can be surmised that this PTM is occurring in the innovator infliximab.
As demonstrated by the spectra of the HC (Figure 4), the biantennary oligosaccharides G0F, G1F, and G2F, along
with smaller amounts of the high mannose forms, are the major glycoforms of infliximab. Since there is only
one N-glycosylation site on the HC, the intensity of peaks for the various oligosaccharide structures can be
used to quantify the relative abundance of the various glycoforms. The MaxEnt 1 algorithm used for generating
the deconvoluted spectra preserves the intensity information from the raw spectra, for quantitative assessment
of structural variation.
This measurement establishes a foundation upon which structural comparison for multiple batches of infliximab
can be performed, thus making the analysis at the subunit level an attractive approach to establishing
development requirements for biosimilars.
On the basis of the analysis of reduced infliximab subunits, we compared the structure differences among
multiple batches of infliximab from the two cell lines. Regulatory guidelines for biosimilar development
recommend that any analytical characterization first establish the structural variation range of the reference
product. As such, analysis of multiple lots of reference products (infliximab from SP2/0 cells) as well as
biosimilar products (inflixmab from CHO cells) is necessary to establish the range of values for critical
structural features. In the meantime, replicate analysis is also performed for each sample to demonstrate the
reproducibility of the LC-MS method itself.
The analysis of multiple samples in triplicate helps establish a vigorous analytical procedure to provide
sound analytical support for biosimilar development. However, this approach generates a high volume of data
that requires efficient informatics tools to process data and produce meaningful results. The UNIFI Scientific
Information System automatically acquires and processes the data and generates reports on the results,
demonstrating the great power and flexibility available for such data analysis tasks.
Next, we demonstrate how UNIFI Scientific Information System’s software can be utilized to streamline the
structural comparison of reduced infliximab from two cell lines.
Structural comparison
Figure 5A displays the MS response summary plot for glycoform G0 in percentage. This UNIFI Scientific
Information System plot offers a simple and direct view to demonstrate the variation in relative abundance
of the G0 glycoform across the injections of innovator and biosimilar batches. This functionality removes the
scientific and compliance burden of summarizing reports of such data in Excel or other data analysis tools that
are not core features of the instrument’s software. By including both automated processing statistical reporting
within UNIFI, the software also prevents human transcription errors that may require significant time and effort
to identify and correct. Similar plots can be readily generated within UNIFI Scientific Information System for
other glycoforms identified in the analysis, such as G1F and Man5, as shown in Figure 5B and 5C.
[ 60 ] Structural Comparison of Infliximab and a Biosimilar via Subunit Analysis Using the Waters Biopharmaceutical Platform with UNIFI
The triplicate analysis for each sample shows a highly reproducible measurement. There is some minor batch-
to-batch variability, notably in the abundance of G1F in the innovator (5B) as well as the Man5 content in the
biosimilar (5C). On the other hand, it appears that the biosimilar, produced in CHO cells, has approximately 10
times more non-fucosylated G0 glycoform compared to that of the innovator (SP2/0 cell line) product. It is also
observed that there are about twice as many G1F glycoforms (by percentage) in the biosimilar batches than in
the innovator, and there is about 30% more Man5 glycoform in the innovator batches than in the biosimilar
sample batches.
Figure 5. Relative abundance of the G0 (A), G1F (B), and Man5 (C) glycoforms in infliximab HC, from all the injections of the innovator (left, blue) and biosimilar
(right, Red) batches.
As this example shows, the glycoforms of infliximab from two cell lines can be readily analyzed and
information on the glycosylation variation can be quickly obtained via UNIFI Scientific Information System
software’s automated workflow covering data acquisition, processing, and reporting. Additionally, the
workflow can be deployed in both non-regulated and regulated environments, so a common analytical
platform can be employed and consistent information acquired across the entire development process.
[ 61 ]
Structural Comparison of Infliximab and a Biosimilar via Subunit Analysis Using the Waters Biopharmaceutical Platform with UNIFI
Another major source of HC heterogeneity is lysine variants. Depending on the cell line and other production
conditions, a lysine residue may remain on the C-terminus of the polypeptide chain. Figure 6 displays the
percentage of clipped-lysine variants, automatically calculated in UNIFI Scientific Information System
software, for both the innovator and biosimilar batches. As can be seen, the percentage was much smaller
for the biosimilar sample batches (from CHO cell line) as compared to that observed in the innovator (SP2/0
cell line) batches. This experimental result confirms that there was a much lower level of C-terminal lysine in
antibodies derived from the CHO cell line, and the lysine content is more consistent from batch-to-batch. The
innovator infliximab has a lower overall abundance for variants with the complete removal of lysine, and the
amount does vary from batch to batch.
Figure 6. The percentage of clipped-lysine (0K) variants, automatically calculated in UNIFI Scientific Information System, is shown for
the innovator (left, blue) and biosimilar (right, red) batches.
[ 62 ] Structural Comparison of Infliximab and a Biosimilar via Subunit Analysis Using the Waters Biopharmaceutical Platform with UNIFI
Waters Corporation
34 Maple Street
Milford, MA 01757 U.S.A.
T: 1 508 478 2000
F: 1 508 872 1990
www.waters.com
Waters, The Science of What’s Possible, ACQUITY UPLC, UPLC, Xevo, and UNIFI are registered trademarks of Waters Corporation.
MaxEnt is a trademark of Waters Corporation. All other trademarks are the property of their respective owners.
©2013–2016 Waters Corporation. Produced in the U.S.A. January 2016 720004796EN AG-PDF
CONC LUSIONS
In this work, the extent of comparability was established between multiple batches
of innovator and candidate biosimilar infliximab, using an integrated analytical
platform with capabilities for automated data processing and reporting. The
Biopharmaceutical Platform Solution with UNIFI Scientific Information System
was applied to study these samples at the level of reduced heavy and light chain
subunits, and to report on several biotherapeutic structural differences between
these preparations.
Overall, the innovator molecule exhibited more heterogeneity with respect to
PTM’s (glycosylation and C-terminal lysine) compared to the candidate biosimilar.
Potentially significant differences were found between the innovator and the
biosimilar samples, particularly in regard to the presence of fucosylated glycans.
We found that the biosimilar had a much higher abundance of the non-fucosylated
glycoform G0, and less of the fucosylated G1F, in comparison to the innovator.
Some batch-to-batch variability was observed among both the innovator batches
and the biosimilar batches.
The power to universally deploy high resolution analytics to address these
important questions, combined with the ability to quickly communicate these
results, enables organizations to make rapid and confident decisions in the race
to market with safe and effective innovator and biosimilar therapeutics.
References
1. Biosimilars and Follow-On Biologics Report: The Global
Outlook 2010–2025, Visiongain Ltd. 2010.
2. Lawrence, S. Billion dollar babies: biotech drugs as
blockbusters. Nature Biotech. 2007; 25, 380–382.
3. Pharmaceutical Research & Manufacturers of America.
Medicines in Development: Biologics. 2013. http://www.
phrma.org/sites/default/files/pdf/biologics2013.pdf
4. Erickson, BE. Untangling biosimilars. Chem. Eng. News,
2010; 88, 25–27.
5. Guidance for Industry, Quality Considerations in
Demonstrating Biosimilarity to a Reference Protein Product,
FDA website [online]. 2012. http://www.fda.gov/downloads/
Drugs/GuidanceComplianceRegulatoryInformation/
Guidances/UCM291134.pdf
6. European Medicines Agency. Guideline on similar biological
medicinal products containing biotechnology-derived
proteins as an active substance: quality issues. EMA
website [online]. 2011. http://www.ema. europa.eu/docs/
en_GB/document_library/Scientific_ guideline/2009/09/
WC500003953.pdf
7. Chakraborty A. Chen W. Gebler J. Characterization of reduced
monoclonal antibody by on-line UPLC-UV/ESI-TOF MS,
Waters Application note. 2009; 720002919en.
8. European Medicines Agency, European Medicines Agency
recommends approval of first two monoclonal-antibody
biosimilar. [online]. 06/2013. http://www.ema.europa.eu/
docs/en_GB/document_library/Press_release/2013/06/
WC500144941.pdf
9. Xie H. Chakraborty A. Ahn J. Yu Y. Dakshinamoorthy D. Gilar
M. Chen W. Skilton SJ. Mazzeo JR. Rapid comparison of a
candidate biosimilar to an innovator monoclonal antibody
with advanced liquid chromatography and mass spectrometry
technologies, MAbs. 2010; Jul–Aug; 2(4): 379–394.
[ 63 ]
Structural Comparison of Infliximab and a Biosimilar via Subunit Analysis Using the Waters Biopharmaceutical Platform with UNIFI
RELEASED N-GLYCANS
[ 64 ]
WAT E R S S O LU T IO NS
GlycoWorks™ RapiFluor-MS™
N-Glycan Kit
GlycoWorks HILIC µElution Plate
ACQUITY UPLC® Glycan BEH Amide 130Å
Column
ACQUITY UPLC H-Class Bio System
ACQUITY® QDa® Mass Detector
Xevo® G2-XS QTof MS
SYNAPT® G2-S HDMS
K E Y W O R D S
GlycoWorks, RapiFluor-MS, RapiGest™ SF,
Rapid Tagging, PNGase F, Deglycosylation,
ACQUITY UPLC H-Class Bio System, BEH
Amide 130Å, Glycans, Glycoproteins,
Glycosylation, HILIC, Fluorescence
A P P L I C AT IO N B E N E F I T S
■
■
Preparation of labeled N-glycans
(from glycoprotein to analysis ready sample)
in 30 minutes
■
■
Complete deglycosylation to produce
unbiased results
■
■
Simple, streamlined protocol provided with
the GlycoWorks RapiFluor-MS N-Glycan Kit
■
■
Unprecedented sensitivity for labeled
N-glycans with at least 2 and 100 fold
increases to fluorescence and MS
detection, respectively
■
■
Accurate profiling based on robust SPE
for neutral to tetrasialylated N-glycans
I N T RO DU C T IO N
The N-glycan profile of a biopharmaceutical is commonly defined as a critical
quality attribute, since it can be a measure of efficacy, safety, and manufacturing
conditions.1-2 Therefore, it is important that approaches for the glycan analysis of
clinical and commercial biotherapeutic formulations exhibit high sensitivity and
facilitate detailed characterization. Additionally, it would be highly advantageous
if such an analysis could also be performed with rapid turnaround times and high
throughput capacity to expedite product development. Most analytical strategies
for evaluating N-glycans from glycoproteins involve deglycosylation via PNGase
F and the labeling of the resulting N-glycans with a chemical moiety that imparts
a detectable attribute. In one, highly effective approach, labeled glycans are
separated by hydrophilic interaction chromatography (HILIC) and detected by
fluorescence (FLR) and sometimes mass spectrometry (MS).3-10
Unfortunately, conventional approaches to the preparation of N-glycans for
HILIC-FLR-MS are either laborious, time-consuming, or require compromises in
sensitivity.11 For instance, a conventional deglycosylation procedure requires
that a glycoprotein sample be incubated for about 1 hour, while many analysts
generically employ an overnight (16 hour) incubation. Combined with this
process is a lengthy, 2 to 3 hour labeling step that relies on reductive amination
of reducing, aldehyde termini that form on N-glycans only after they hydrolyze
from their glycosylamine forms. And in the case of one of the most frequently
employed labeling compounds, 2-aminobenzamide (2-AB), the resulting glycans
can be readily detected by fluorescence but are rather challenging to detect by
electrospray ionization mass spectrometry (ESI-MS).
Variations to conventional approaches for N-glycan sample preparation have been
explored, but have not, as of yet, presented a solution that combines the desired
attributes of simplicity, high MS sensitivity, and high throughput. Alternative
labeling reagents, for example procainamide, that have functional groups to
enhance electrospray ionization efficiency have been used,12 but this does not
address the cumbersome, time consuming nature of relying on a reductive
amination labeling step. Rapid tagging procedures that yield labeled glycans in a
matter of minutes have consequently been investigated. In fact, two rapid tagging
glycan labels were recently introduced, including a rapid tagging analog
Rapid Preparation of Released N-Glycans for HILIC Analysis Using
a Novel Fluorescence and MS-Active Labeling Reagent
Matthew A. Lauber,1 Darryl W. Brousmiche,1 Zhengmao Hua,1 Stephan M. Koza,1 Ellen Guthrie,2 Paula Magnelli,2
Christopher H. Taron,2 Kenneth J. Fountain1
1
Waters Corporation, Milford, MA, USA
2
New England BioLabs, Ipswich, MA, USA
[ 65 ]
E X P E R IM E N TA L
Method conditions (unless otherwise noted):
LC conditions
LC system:
ACQUITY UPLC H-Class
Bio System
Sample temp.:
5 °C
Analytical column
temp.:
60 °C
Flow rate:
0.4 mL/min
Fluorescence detection: Ex 265/Em 425 nm
(RapiFluor-MS)
Ex 278/Em 344 nm
(Instant AB)
Ex 330/Em 420 nm (2-AB)
(2 Hz scan rate [150 mm
column]/5 Hz scan rate
[50 mm column], Gain =1)
Injection volume:
≤1 µL (aqueous diluents
with 2.1 mm I.D. columns)
≤30 µL (DMF/ACN diluted
samples with 2.1 mm
I.D. columns)
Columns:
ACQUITY UPLC Glycan
BEH Amide 130Å, 1.7 µm,
2.1 x 50 mm
(p/n 186004740)
ACQUITY UPLC Glycan
BEH Amide 130Å, 1.7 µm,
2.1 x 150 mm
(p/n 186004742)
Sample collection/
vials:
Sample Collection Module
(p/n 186007988)
Polypropylene 12 x 32 mm
Screw Neck Vial, 300 µL
volume (p/n 186002640)
Gradient used with 2.1 x 50 mm columns:
Mobile phase A:
50 mM ammonium
formate, pH 4.4 (LC-MS
grade; from a 100x
concentrate,
p/n 186007081)
Mobile phase B:
ACN (LC-MS grade)
Time
Flow rate
%A
%B
Curve
(mL/min)
0.0
0.4 25 75 6
11.7
0.4 46 54 6
12.2 0.2 100 0 6
13.2 0.2 100 0 6
14.4 0.2 25 75 6
15.9 0.4 25 75 6
18.3 0.4 25 75 6
Gradient used with 2.1 x 150 mm columns:
Mobile phase A:
50 mM ammonium
formate, pH 4.4 (LC-MS
grade; from
a 100x concentrate,
p/n 186007081)
Mobile phase B:
ACN (LC-MS grade)
Time
Flow rate
%A
%B
Curve
(mL/min)
0.0
0.4 25 75 6
35.0 0.4 46 54 6
36.5 0.2 100 0 6
39.5 0.2 100 0 6
43.1 0.2 25 75 6
47.6 0.4 25 75 6
55.0 0.4 25 75 6
[ 66 ] Rapid Preparation of Released N-Glycans for HILIC Analysis Using a Novel Fluorescence and MS-Active Labeling Reagent
of aminobenzamide (AB).13 In a rapid reaction, the precursor glycosylamines
of reducing, aldehyde terminated glycans are modified via a urea linked
aminobenzamide. Although such a rapid tagging reagent accelerates the
labeling procedure, it does not provide the enhanced ionization efficiencies
needed in modern N-glycan analyses.
To address the above shortcomings, we have developed a sample preparation
solution that enables unprecedented FLR and MS sensitivity for glycan detection
while also improving the throughput of N-glycan sample preparation. A novel
labeling reagent has been synthesized that rapidly reacts with glycosylamines
upon their release from glycoproteins. Within a 5 minute reaction, N-glycans
are labeled with RapiFluor-MS, a reagent comprised of an N-hydroxysuccinimide
(NHS) carbamate rapid tagging group, an efficient quinoline fluorophore, and
a highly basic tertiary amine for enhancing ionization. To further accelerate the
preparation of N-glycans, rapid tagging has been directly integrated with a
Rapid PNGase F deglycosylation procedure involving RapiGest SF surfactant
and a HILIC µElution SPE clean-up step that provides highly quantitative
recovery of the released and labeled glycans with the added benefit of not
requiring a solvent dry-down step prior to the LC-FLR-MS analysis of samples.
SAM P L E D E S C R I P T IO N
N-glycans from Intact mAb Mass Check Standard (p/n 186006552), bovine
fetuin (Sigma F3004), and pooled human IgG (Sigma I4506) were prepared
according to the guidelines provided in the GlycoWorks RapiFluor-MS N-Glycan
Kit Care and Use Manual (715004793).
To compare the response factors of Instant AB™ and RapiFluor-MS labeled
glycans, labeling reactions were performed with equivalent molar excesses of
reagent, and crude reaction mixtures were directly analyzed by HILIC-FLR-MS in
order to avoid potential biases from SPE clean-up procedures. Response factors
were determined as ratios of the FA2 N-glycan (Oxford notation) chromatographic
peak area to the mass of glycoprotein from which the glycan originated.
To compare the response factors of 2-AB labeled versus RapiFluor-MS labeled
glycans, equivalent quantities of labeled N-glycans from pooled human IgG
were analyzed by HILIC-FLR-MS. Column loads were calibrated using external
quantitative standards of 2-AB labeled triacetyl chitotriose and RapiFluor-MS
derivatized propylamine (obtained in high purity; confirmed by HPLC and
1H NMR). Response factors were determined as ratios of the FA2
chromatographic peak area to the mole quantity of glycan.
The procedure for extracting labeled RapiFluor-MS glycans after derivatization
was evaluated using a test mixture containing N-glycans released and labeled
from a 1:1 mixture (by weight) of pooled human IgG and bovine fetuin. The test
mixture was prepared and then reconstituted in a solution equivalent in
MS conditions
MS system:
SYNAPT G2-S HDMS
Ionization mode:
ESI+
Analyzer mode:
TOF MS, resolution mode
(~20 K)
Capillary voltage:
3.0 kV
Cone voltage:
80 V
Source temp.:
120 °C
Desolvation temp.:
350 °C
Desolvation gas flow: 800 L/Hr
Calibration:
NaI, 1 µg/µL from
500–2500 m/z
Lockspray
(ASM B-side):
100 fmol/µL Human
Glufibrinopeptide B in
0.1% (v/v) formic acid,
70:30 water/acetonitrile
every 90 seconds
Acquisition:
500–2500 m/z,
1 Hz scan rate
Data management:
MassLynx Software (V4.1)
[ 67 ]
Rapid Preparation of Released N-Glycans for HILIC Analysis Using a Novel Fluorescence and MS-Active Labeling Reagent
composition to the solution glycans are subjected to when following the protocol of the RapiFluor-MS
N-Glycan Kit. All other sample preparation techniques are described in the GlycoWorks RapiFluor-MS
N-Glycan Kit Care and Use Manual (715004793).
R E SU LT S A N D D IS C U S S IO N
Rational design of a new N-Glycan labeling reagent
A new labeling reagent for facilitating N-glycan analysis has been synthesized based on rational design
considerations (Figure 1) that would afford rapid labeling kinetics, high fluorescence quantum yield,
and significantly enhanced MS detectability. Conventional N-glycan sample preparation is dependent
on reductive amination of aldehyde terminated saccharides, a process that requires glycans to undergo
multiple chemical conversions and a lengthy high temperature incubation step.11 Moreover, glycans must be
reductively aminated in anhydrous conditions in order to minimize desialylation. Sample preparations are
therefore burdened with transitioning a sample from aqueous to anhydrous conditions. For these reasons, the
newly designed labeling reagent foregoes reductive amination and instead takes advantage of an aqueous
rapid tagging reaction. To this end, Waters has drawn upon its experience with rapid fluorescence labeling
of amino acids to develop a new reagent that meets the needs of modern, N-glycan analysis. More than 20
years ago, Waters introduced a rapid tagging labeling reagent, known as AccQ•Fluor™, that is now widely
use to accurately profile the amino acid composition of protein samples via fluorescence detection.14-15
O
NH
O
N
O
NH
N
N
O
O
Rapid
Fluorescence
MS
NHS Carbamate Rapid
Tagging Group
Quinolinyl
Fluorophore
Tertiary Amine
Charge Tag
Figure 1. RapiFluor-MS Molecular Structure.
Features of the chemical structure that enable
rapid tagging, efficient fluorescence, and
enhanced ionization efficiency are highlighted.
Figure 2. Reaction Schematic for RapiFluor-MS
Derivatization of an N-glycosylamine. The
pathway on the left shows the derivatization of
a glycosylamine, which produces an N-glycan
with a urea (NH-CO-NH) linked RapiFluor-MS
label. Hydrolysis of RapiFluor-MS is shown in
the pathway on the right.
+
+
H2O
+
+ CO2
[ 68 ] Rapid Preparation of Released N-Glycans for HILIC Analysis Using a Novel Fluorescence and MS-Active Labeling Reagent
AccQ•Fluor possesses two important chemical characteristics: an
NHS-carbamate rapid tagging reactive group and a highly efficient
quinolinyl fluorophore. These features of AccQ•Fluor form the basis
of the new glycan labeling reagent. The NHS-carbamate reactive
group of this reagent enables glycosylamine bearing N-glycans
to be rapidly labeled following their enzymatic release from
glycoproteins. Within a 5 minute reaction, N-glycans are labeled
with the new reagent under ambient, aqueous conditions to yield a
highly stable urea linkage (Figure 2). In addition to rapid tagging
capabilities, the new labeling reagent also supports high sensitivity
for both MS and flourescence detection. A quinoline fluorophore
serves as the central functionality of the new reagent that, as with
AccQ•Fluor, facilitates high sensitivity fluorescence detection.
In addition to AccQ•Fluor, however, the new reagent has been
synthesized with a tertiary amine side chain as a means to enhance
MS signal upon positive ion mode electrospray ionization (ESI+).
In summary, the resulting N-glycan labeling reagent is built upon
our expertise in chemical reagents and three important chemical
attributes, a rapid tagging reactive group, an efficient fluorophore,
and a highly basic MS active group. To describe these noteworthy
attributes, the new labeling reagent has accordingly been
named RapiFluor-MS.
300x zoom
4.5
5
5.5
6
6.5
7
4.5
5
5.5
6
6.5
7
FLR
MS
(BPI)
FLR
MS
(BPI)
342.8
179.9
233.7
0.3
0
100
200
300
400
Compound 4
Compound 1
A
FA2
FA2
FA2
FA2
B
C
FLR
MS
(BPI)
Re
sponse
Fa
ctors
(F
A2
Pe
ak Area per Sample of N-Glycans
from 1
g of Intact mAb Mass Check Standard/1000)
Rapi Fluor-MS Labeled N- Glycans
from Intact mAb Mass Check Standard (0.4 µg)
Instant AB Labeled N-Glycans
from Intact mAb Mass Check Standard (0.4 µg)
min
min
0.0E+0
2.6E+6
1.7E+6
0.0E+0
2.6E+6
0.0E+0
1.7E+6
0.0E+0
RapiFluor-MS
Labeled
Instant AB
Labeled
NH
O
NH
N
NH
O
N
Glycan
NH
O
NH
Glycan
O
NH2
RapiFluor-MS enables high sensitivity detection
RapiFluor-MS N-glycan labeling has been extensively studied. In
particular, the response factors of RapiFluor-MS labeled glycans
have been benchmarked against those observed for glycans labeled
with alternative reagents. The most closely related, commercially
available alternative to RapiFluor-MS is an NHS carbamate analog
of aminobenzamide, known as Instant AB.13 Figures 3A and 3B
present HILIC fluorescence and base peak intensity (BPI) MS
chromatograms for equivalent quantities of N-glycans released from
a murine monoclonal antibody (Intact mAb Mass Check Standard,
p/n 186006552) and labeled with RapiFluor-MS and Instant AB,
respectively. Based on the observed chromatographic peak areas,
response factors for fluorescence and MS detection were determined
for the most abundant glycan in the IgG profile, the fucosylated,
biantennary FA2 glycan (Figure 3C). Our results for the FA2 glycan
indicate that RapiFluor-MS labeled glycans produce 2 times higher
fluorescence signal and, more astoundingly, 780 times greater
MS signal than N-glycans labeled with Instant AB.
Figure 3. HILIC-FLR-MS of (A) RapiFluor-MS
and (B) Instant AB Labeled N-Glycans from
Intact mAb Mass Check Standard. Fluorescence
(FLR) chromatograms are shown in orange and
base peak intensity (BPI) MS chromatograms
are shown in blue. Labeled glycans (from 0.4
µg of glycoprotein, 1 µL aqueous inection)
were separated using a ACQUITY UPLC BEH
Amide 130Å, 1.7 µm, 2.1 x 50 mm Column.
(C) Response factors for RapiFluor-MS and
Instant AB labeled glycans (measured as
the FA2 peak area per sample of N-glycans
resulting from 1 µg of Intact mAb Mass Check
Standard). Fluorescence (FLR) and MS (base
peak intensity) response factors are shown in
orange and blue, respectively. Analyses were
performed in duplicate.
[ 69 ]
Rapid Preparation of Released N-Glycans for HILIC Analysis Using a Novel Fluorescence and MS-Active Labeling Reagent
In a similar fashion, RapiFluor-MS labeling has also been compared to conventional 2-AB labeling. To draw such a comparison, N-glycans
released from pooled human IgG and labeled with either RapiFluor-MS or 2-AB were analyzed by HILIC-FLR-MS at equivalent mass loads
(Figures 4A and 4B, respectively). Given that rapid tagging and reductive amination are performed by significantly different procedures,
external calibrations were established using quantitative standards in order to determine the amounts of FA2 glycan loaded and eluted
from the HILIC column. Response factors determined using these calibrated amounts of FA2 glycan are provided in Figure 4C. Again, it
is found that RapiFluor-MS labeled glycans are detected with significantly higher signal (14 times higher fluorescence and 160 times
greater MS signal versus 2-AB labeled glycans.)
3
4
5
6
7
8
9
4.0E+6
3
4
5
6
7
8
9
3.3E+6
100x zoom
FA2
(2.17 pmol)
FA2
FA2
(2.61 pmol)
FA2
105.8
7.4
126.9
0.8
0
50
100
150
A
B
RapiFluor-MS
Labeled
FLR
MS
(BPI)
FLR
MS
(BPI)
2-AB
Labeled
C
RapiFluor-MS Labeled N-Glycans from Pooled Human IgG
FLR
MS
(BPI)
Response Factors
(Peak Area per
pmol of Labeled FA2 Glyca
n
/ 1000)
2-AB Labeled
N -Glycans from Pooled Human IgG
min
min
0.0E+0
0.0E+0
4.0E+6
3.3E+6
0.0E+0
0.0E+0
NH
O
NH
N
NH
O
N
Glycan
NH
Glycan
NH2
O
Figure 4. HILIC-FLR-MS of (A) RapiFluor-MS and (B) 2-AB
Labeled N-Glycans from Pooled Human IgG. Fluorescence (FLR)
chromatograms are shown in orange and base peak intensity (BPI)
MS chromatograms are shown in blue. Labeled glycans
(~14 pmol total glycan, 1 µL aqueous injection) were separated
using a ACQUITY UPLC Glycan BEH Amide 130Å, 1.7 µm,
2.1 x 50 mm Column. The quantities of FA2 glycan were calibrated
via two-point external calibrations with quantitative standards.
(C) Response factors for RapiFluor-MS and 2-AB labeled glycans
(measured as the FA2 peak area per picomole of FA2 determined
by the external calibration). Fluorescence (FLR) and MS (BPI)
response factors are shown in orange and blue, respectively.
Analyses were performed in duplicate.
Figure 5. Relative Performance of Glycan Labels. Response
factors normalized to the fluorescence and MS response
factors of RapiFluor-MS labeled N-glycans. (*) Comparative
result extrapolated from a published comparison of N-glycans,
wherein it was found that procainamide provided comparable
fluorescence and up to 50 fold greater ESI-MS sensitivity when
compared to 2-AB (Klapoetke et al. 2010).
To summarize the above observations, we have plotted the response factors of Instant AB and 2-AB as percentages of the response
factors of RapiFluor-MS (Figure 5). The gains in fluorescence and MS sensitivity are apparent in this plot, since it portrays response
factors for Instant AB and 2-AB normalized to those for RapiFluor-MS. In this plot, the relative performance of reductive amination with
another alternative labeling reagent, procainamide, is also provided. Procainamide is a chemical analog to aminobenzamide that has
recently been exploited to enhance the ionization of reductively aminated glycans when they are analyzed by HILIC-ESI(+)-MS. Previous
studies have shown that procainamide labeled glycans yield comparable fluorescence signal and up to 50 times greater MS signal when
compared to 2-AB labeled glycans.12,16 Compared to procainamide, RapiFluor-MS will therefore provide sizeable gains in MS sensitivity.
That is to say, the novel RapiFluor-MS labeling reagent not only supports rapid tagging of glycans but it also provides analysts with
unmatched sensitivity for fluorescence and MS based detection.
0
10
20
30
40
50
60
70
80
90
100
0
10
20
30
40
50
60
70
80
90
100
Fluorescence
MS (BPI)
Rapi
Fluo
r-MS
2-
AB
Instant AB
Instant AB Labeled
RapiFluor-MS Labeled
2-AB Labeled
Re
lative
Pe
rformance (%)
52.5
7.0
0.1
0.6
Procainamide Labeled
Procainamide
*
Rapi
Fluo
r-MS
2-
AB
Instant AB
Procainamide
*
30.0*
7.0*
NH
O
NH
N
NH
O
N
Glycan
NH
O
NH
Glycan
O
NH2
NH
Glycan
NH2
O
NH
Glycan
O
NH
N
[ 70 ] Rapid Preparation of Released N-Glycans for HILIC Analysis Using a Novel Fluorescence and MS-Active Labeling Reagent
0.0E+0
1.2E+6
10
12
14
16
18
20
22
24
26
28
30 min
0.0E+0
1.2E+6
10
12
14
16
18
20
22
24
26
28
30 min
MS
FLR
Rapid deglycosylation with a novel formulation of Rapid PNGase F and RapiGest SF Surfactant
RapiFluor-MS labeling revolutionizes N-glycan sample preparation and can be readily adopted in the laboratory with the GlycoWorks
RapiFluor-MS N-Glycan Kit. This complete solution from Waters and New England BioLabs was purposefully designed to remove the
bottlenecks from all aspects of N-glycan sample preparation. The optimized N-glycan sample preparation workflow requires a minimum
of three steps, including deglycosylation (to release glycans from a glycoprotein), labeling (to impart a detectable chemical entity to
glycans), and a clean-up step (to remove potential interferences from the sample) (Figure 6). Conventional approaches to N-glycan
sample preparation can be very time consuming due to not only lengthy labeling procedures but also lengthy deglycosylation steps
that range from 1 to 16 hours. To ensure rapid labeling with RapiFluor-MS was not encumbered by a time-consuming deglycosylation
process, Waters partnered with New England BioLabs to co-develop a Rapid PNGase F deglycosylation procedure specifically designed
for integration with rapid tagging labeling reagents.
Figure 6. Workflow for the Rapid Preparation of N-glycans Using the RapiFluor-MS N-Glycan Kit.
DATA PROCESSING
UNIFI,, MassLynx, and
Empower Software
ANALYSIS ON LC
WITH FLR DETECTOR
Alliance HPLC or
ACQUITY UPLC Systems
CHARACTERIZATION OR
MASS CONFIRMATION
ACQUITY® QDa® Detector
Xevo G2-XS QTof MS
SYNAPT G2-Si HDMS
Results
RELEASED
N-GLYCANS BY
DEGLYCOSYLATION
GlycoWorks™ Rapid
PNGase F and Buffer, and
RapiGest™ SF Surfactant
LABELED
GLYCANS
GlycoWorks
RapiFluor-MS
Labeling
PURIFIED GLYCANS
GlycoWorks HILIC µElution
Plate or Cartridges
INJECTION ON COLUMN
FOR LC ANALYSIS
Glycan BEH Amide Column and
Glycan Performance Test Standard
Dextran Calibration Ladder
GLYCOPROTEIN
Intact mAb Mass
Check Standard
[ 71 ]
Rapid Preparation of Released N-Glycans for HILIC Analysis Using a Novel Fluorescence and MS-Active Labeling Reagent
The GlycoWorks RapiFluor-MS N-Glycan Kit provides a novel formulation of Rapid PNGase F and RapiGest SF
Surfactant that can be used to completely deglycosylate a diverse set of glycoproteins in an approximately
10 minute procedure. This fast deglycosylation procedure is facilitated by the use of RapiGest, an anionic
surfactant, that is used to ensure that N-glycans are accessible to Rapid PNGase F and that glycoproteins
remain soluble upon heat denaturation. Most importantly, RapiGest is an enzyme-friendly reagent and can
therefore be used at high concentrations without hindering the activity of Rapid PNGase F. In the developed
fast deglycosylation technique, a glycoprotein is subjected to a high concentration of RapiGest (1%) and
heated to ≥80˚C for 2 minutes. Subsequently and without any additional sample handling, Rapid PNGase F is
added to the solution and the mixture is incubated at an elevated, 50˚C temperature for 5 minutes to achieve
complete, unbiased deglycosylation.
The effectiveness of this rapid deglycosylation process has been evaluated using sodium dodecyl sulfate
polyacrylamide gel electrophoresis (SDS-PAGE). SDS-PAGE is an effective technique for separating proteins
based on their size in solution and can often be used to separate the glycosylated and de-glycosylated forms
of proteins.17-18 A diverse set of glycoproteins were deglycosylated according to the rapid deglycosylation
procedure and analyzed by SDS-PAGE along with negative controls containing no PNGase F and positive
controls, in which the glycoproteins were subjected to conventional multiple step deglycosylation with
SDS based denaturation and PNGase F incubation for 30 minutes at 37 ˚C. Figure 7 shows the results of
this study, where it can be seen that for each of the tested proteins there is a significant decrease in protein
apparent molecular weight after they are subjected to the rapid deglycosylation procedure. Moreover, the
apparent molecular weight decreases are visually comparable to those observed for proteins deglycosylated
by the control method. These results demonstrate that the fast deglycosylation approach facilitated by a
unique formulation of Rapid PNGase F and RapiGest SF Surfactant produces deglycosylation comparable to a
conventional approach but in only a fraction of the time required.
- + R - + R - + R - + R
- + R - + R - + R
- + R
Std
Fibrinogen
RN
as
e
B
Fetui
n
Intact
mA
b
Mass Check Std
Lactoferrin
Ovalbumi
n
-Acid
Glycoprotein
Asialofeutin
(-) Neg Control / Conventional / 2 Steps without PNGase F: SDS + DTT, 95˚C, 2 min / NP-40 + Reaction Buffer, 37˚C, 30 min
(+) Pos Control / Conventional / 2 Steps with PNGase F: SDS + DTT, 95˚C, 2 min / NP-40 + Reaction Buffer + PNGase F, 37˚C, 30 min
(R) 2 Step Rapid Deglycosylation / Reaction Buffer + RapiGest ≥80˚C, 2 min / + PNGase F, 50˚C, 5 min
PNGase F inactivated after the deglycosylation step via heat denaturation at 70˚C for 20 min
Figure 7. Gel Electrophoresis Assay
for Deglycosylation of Glycoproteins.
A negative control (-) shows the migration
distance and apparent molecular weight
of the native glycoproteins, and a positive
control (+) shows the migration distance
and decreased apparent molecular weight
of deglycosylated proteins as obtained
by conventional two step deglycosylation
using SDS denaturation and a subsequent
30 minute incubation with PNGase F
at 37 ˚C. Results demonstrating the
complete deglycosylation of these
glycoproteins with a fast procedure
involving a two step approach with
RapiGest-assisted heat denaturation and
a subsequent 5 minute incubation with
Glycoworks Rapid PNGase F at 50˚C are
also shown (R). Coomassie blue staining
was used for band visualization.
[ 72 ] Rapid Preparation of Released N-Glycans for HILIC Analysis Using a Novel Fluorescence and MS-Active Labeling Reagent
Robust, quantitative HILIC SPE
As mentioned earlier, the final step in an N-glycan sample preparation aims to extract the labeled glycans in preparation for their analysis.
An effective approach for extraction of labeled glycans from reaction byproducts has been devised using solid phase extraction (SPE). In
particular, this SPE is designed to selectively extract RapiFluor-MS labeled N-glycans from a mixture comprised of deglycosylated protein,
PNGase F, buffer/formulation components, RapiGest Surfactant, and labeling reaction byproducts, which otherwise interfere with analysis
of the labeled glycans by HILIC column chromatography (Figure 8). For the RapiFluor-MS N-Glycans kit, a GlycoWorks µElution plate is
provided that contains a silica based aminopropyl sorbent specifically selected for this application.19-22 Due to its highly polar nature, the
GlycoWorks SPE sorbent readily and selectively retains polar compounds such as glycans. In addition, this sorbent possesses a weakly
basic surface that provides further selectivity advantages based on ion exchange and electrostatic repulsion. It is also worth noting that the
GlycoWorks µElution Plate is designed for minimal elution volumes such that samples can be immediately analyzed without a dry down step.
Moreover, the GlycoWorks µElution Plate is constructed as a 96 well format, meaning it can be used to perform high throughput experiments
or used serially (with appropriate storage, see the Care and Use Manual) for low throughput needs.
In this HILIC SPE process, the sorbent is first conditioned with water and then equilibrated to high acetonitrile loading conditions.
Thereafter, glycan samples that have been diluted with acetonitrile are loaded and washed free of the sample matrix using an acidic
wash solvent comprised of 1% formic acid in 90% acetonitrile. This washing condition achieves optimal SPE selectivity by introducing
electrostatic repulsion between the aminopropyl HILIC sorbent and reaction byproducts and by enhancing the solubility of the matrix
components. After washing, the labeled, released glycans are next eluted from the HILIC sorbent. Since the GlycoWorks SPE sorbent
has a weakly basic surface, and the capacity for anion exchange, just as it has the capacity for cation repulsion, it is necessary to elute
the labeled glycans with an eluent of significant ionic strength. We have, as a result, developed an elution buffer comprised of a pH 7
solution of 200 mM ammonium acetate in 5% acetonitrile (p/n 186007991). Upon their elution, the RapiFluor-MS labeled glycans can
be diluted with a mixture of organic solvents (acetonitrile and dimethylformamide) and directly analyzed by UPLC or HPLC HILIC column
chromatography using fluorescence and/or ESI-MS detection.
0.0E+0
1.9E+6
0
5
10
15
0.0E+0
2.3E+6
After SPE (1x SPE)
No SPE
Crude Reaction Mixture
Figure 8. HILIC SPE to Remove Chromatographic Interference.
(A) A test mixture comprised of RapiFluor-MS labeled glycans
from pooled human IgG and bovine fetuin separated on an
ACQUITY UPLC BEH Amide 130Å, 1.7 µm, 2.1 x 50 mm
Column and detected via fluorescence (labeled N-glycans
from 0.4 µg glycoprotein, 1 µL injection crude reaction
mixture). (B) The test mixture after extraction by HILIC SPE
(labeled N-glycans from 0.4 µg glycoprotein, 10 µL injection
of ACN/DMF diluted SPE eluate).
A
B
[ 73 ]
Rapid Preparation of Released N-Glycans for HILIC Analysis Using a Novel Fluorescence and MS-Active Labeling Reagent
Figure 9. Extraction of RapiFluor-MS Labeled N-glycans by SPE with a GlycoWorks HILIC µElution Plate. (A) A test mixture comprised
of RapiFluor-MS labeled glycans from pooled human IgG and bovine fetuin separated on an ACQUITY UPLC Glycan BEH Amide, 130Å,
1.7 µm, 2.1 x 150 mm Column and detected via fluorescence (labeled N-glycans from 0.4 µg glycoprotein, 10 µL injection of
ACN/DMF diluted sample). (B) The test mixture after extraction by HILIC SPE. (C) Relative abundances determined for a set of
RapiFluor-MS labeled glycans before and after GlycoWorks HILIC SPE.
As with other aspects of the RapiFluor-MS N-Glycan Kit, this
GlycoWorks SPE step has been extensively evaluated. A test
mixture was created to assess the recovery of a diverse set of
RapiFluor-MS labeled N-glycans, ranging from small neutral
glycans to high molecular weight, tetrasialylated glycans. Such a
mixture was prepared by releasing and labeling N-glycans from
both pooled human IgG and bovine fetuin. An example analysis of
this test mixture by HILIC column chromatography and fluorescence
detection is shown in Figure 9A. Species representing extremes in
glycan properties are labeled, including an asialo FA2 glycan and a
glycan with a tetrasialylated, triantennary structure (A3S1G3S3).
To evaluate the effects of the SPE process, this mixture was
subjected to a second pass of GlycoWorks SPE and again analyzed
by HILIC chromatography and fluorescence detection, as shown
in Figure 9B. It can be observed that the sample processed twice
by SPE presents a labeled glycan profile comparable to the profile
observed for the sample processed only once by SPE. Indeed, this SPE
step has been found to exhibit an absolute recovery of approximately
70-80% for all purified glycans and more importantly highly
accurate relative yields. Figure 9C shows the relative abundances
for four glycans (FA2, FA2G2S1, A3G3S3, and A3S1G3S3) as
determined for samples subjected to one pass versus two passes of
SPE. The largest deviation in relative abundance was observed for
the tetrasialylated A3S1G3S3, in which case relative abundances of
5.7% and 6.1% were determined for samples processed by one and
two passes of SPE, respectively. With these results, it is demonstrated
that GlycoWorks SPE provides a mechanism to immediately analyze a
sample of extracted, labeled glycans and does so without significant
compromise to the accuracy of the relative abundances determined for
a wide range of N-glycans.
0.0E+0
1.2E+6
10
15
20
25
30
35
Positive Control
0.0E+0
1.2E+6
10
15
20
25
30
35
0
5
10
15
20
25
30
Re
la
ti
ve
Ab
und
an
ce
(%
)
Positive
Control
SPE
Processed
After SPE (2x SPE)
FA2
FA2G2S1
A3G3S3
A3S1G3S3
FA
2
FA2G2S1
A3G3S3
A3S1G3S3
C
B
A
Before SPE (1x SPE)
Before
SPE
(1x SPE)
After
SPE
(2x SPE)
[ 74 ] Rapid Preparation of Released N-Glycans for HILIC Analysis Using a Novel Fluorescence and MS-Active Labeling Reagent
Waters Corporation
34 Maple Street
Milford, MA 01757 U.S.A.
T: 1 508 478 2000
F: 1 508 872 1990
www.waters.com
Waters, The Science of What’s Possible, ACQUITY UPLC, ACQUITY, QDa, Xevo, and SYNAPT are registered trademarks
of Waters Corporation. GlycoWorks, RapiFluor-MS, RapiGest, and AccQ•Fluor are trademarks of Waters Corporation.
All other trademarks are the property of their respective owners.
©2015 Waters Corporation. Produced in the U.S.A. January 2015 720005275EN AG-PDF
C O N C LU S IO NS
Conventional approaches to the preparation of N-glycans
for HILIC-FLR-MS are either laborious, time-consuming, or
require compromises in sensitivity. With the development of
the GlycoWorks RapiFluor-MS N-Glycan Kit, we address these
shortcomings by enabling unprecedented sensitivity for glycan
detection while also improving the throughput of N-glycan sample
preparation. With this approach, glycoproteins are deglycosylated
in approximately 10 minutes to produce N-glycosylamines. These
glycans are then rapidly reacted with the novel RapiFluor-MS
reagent within a 5 minute reaction and are thereby labeled with a
tag comprised of an efficient fluorophore and a highly basic tertiary
amine that yields enhanced sensitivity for both fluorescence and
MS detection. In a final step requiring no more than 15 minutes, the
resulting RapiFluor-MS labeled glycans are extracted from reaction
byproducts by means of µElution HILIC SPE that has been rigorously
developed to provide quantitative recovery of glycans (from neutral
to tetrasialylated species) and to facilitate immediate analysis of
samples. Accordingly, an analyst can complete an N-glycan sample
preparation, from glycoprotein to ready-to-analyze sample, in just
30 minutes when using the sensitivity enhancing RapiFluor-MS
labeling reagent.
References
1. Beck, A.; Wagner-Rousset, E.; Ayoub, D.; Van Dorsselaer, A.; Sanglier-
Cianferani, S., Characterization of therapeutic antibodies and related
products. Anal Chem 2013, 85 (2), 715–36.
2. Dalziel, M.; Crispin, M.; Scanlan, C. N.; Zitzmann, N.; Dwek, R. A., Emerging
principles for the therapeutic exploitation of glycosylation. Science 2014,
343 (6166), 1235681.
3. Kaneshiro, K.; Watanabe, M.; Terasawa, K.; Uchimura, H.; Fukuyama, Y.;
Iwamoto, S.; Sato, T. A.; Shimizu, K.; Tsujimoto, G.; Tanaka, K., Rapid
quantitative profiling of N-glycan by the glycan-labeling method using
3-aminoquinoline/alpha-cyano-4-hydroxycinnamic acid. Anal Chem 2012,
84 (16), 7146–51.
4. Ahn, J.; Bones, J.; Yu, Y. Q.; Rudd, P. M.; Gilar, M., Separation of
2-aminobenzamide labeled glycans using hydrophilic interaction
chromatography columns packed with 1.7 microm sorbent. J Chromatogr B
Analyt Technol Biomed Life Sci 2010, 878 (3–4), 403–8.
5. Yu, Y. Q.; Ahn, J.; Gilar, M., Trastuzumab Glycan Batch-to-Batch Profiling
using a UPLC/FLR/MS Mass Spectrometry Platform. Waters Appication Note
720003576en 2010.
6. Hilliard, M.; Struwe, W.; Carta, G.; O’Rourke, J.; McLoughlin, N.; Rudd, P.; Yu,
Y. Q., A Systematic Approach to Glycan Analysis Using HILIC-UPLC and
an Online Database of Standardized Values. Waters Application Note
720004203en 2012.
7. Hilliard, M.; Struwe, W.; Adamczyk, B.; Saldova, R.; Yu, Y. Q.; O’Rourke,
J.; Carta, G.; Rudd, P., Development of a Glycan Database for Waters ACQUITY
UPLC Systems. Waters Application Note 720004202en 2012.
8. Yu, Y. Q., Analysis of N-Linked Glycans from Coagulation Factor IX,
Recombinant and Plasma Derived, Using HILIC UPLC/FLR/QTof MS.
Waters Appication Note 720004019en 2011.
9. Marino, K.; Bones, J.; Kattla, J. J.; Rudd, P. M., A systematic approach
to protein glycosylation analysis: a path through the maze. Nat Chem Biol
2010, 6 (10), 713–23.
10. Gillece-Castro, B.; Tran, K. v.; Turner, J. E.; Wheat, T. E.; Diehl, D. M., N-Linked
Glycans of Glycoproteins: A New Column for Improved Resolution. Waters
Appication Note 720003112en 2009.
11. Mechref, Y.; Hu, Y.; Desantos-Garcia, J. L.; Hussein, A.; Tang, H., Quantitative
glycomics strategies. Mol Cell Proteomics 2013, 12 (4), 874–84.
12. Klapoetke, S.; Zhang, J.; Becht, S.; Gu, X.; Ding, X., The evaluation of a
novel approach for the profiling and identification of N-linked glycan with a
procainamide tag by HPLC with fluorescent and mass spectrometric detection.
J Pharm Biomed Anal 2010, 53 (3), 315–24.
13. Cook, K. S.; Bullock, K.; Sullivan, T., Development and qualification of an
antibody rapid deglycosylation method. Biologicals 2012, 40 (2), 109–17.
14. Cohen, S. A.; Michaud, D. P., Synthesis of a fluorescent derivatizing reagent,
6-aminoquinolyl-N-hydroxysuccinimidyl carbamate, and its application
for the analysis of hydrolysate amino acids via high-performance liquid
chromatography. Anal Biochem 1993, 211 (2), 279–87.
15. Cohen, S. A.; DeAntonis, K.; Michaud, D. P., Compositional Protein Analysis
Using 6-Aminoquinolyl-N-Hydroxysuccinimidyl Carbamate, A Novel
Derivatization Reagent. In Techniques in Protein Chemistry IV, 1993; pp
289–298.
16. Ivleva, V.; Yu, Y. Q., UPLC/FLR/QTof MS Analysis of Procainamide-Labeled
N-Glycans. Waters Appication Note 720004212en 2012.
17. Kung, L. A.; Tao, S. C.; Qian, J.; Smith, M. G.; Snyder, M.; Zhu, H., Global
analysis of the glycoproteome in Saccharomyces cerevisiae reveals new roles
for protein glycosylation in eukaryotes. Mol Syst Biol 2009, 5, 308.
18. Schwalbe, R. A.; Wang, Z.; Bianchi, L.; Brown, A. M., Novel sites of
N-glycosylation in ROMK1 reveal the putative pore-forming segment H5 as
extracellular. J Biol Chem 1996, 271 (39), 24201–6.
19. Lauber, M. A.; Koza, S. M.; Fountain, K. J., Optimization of GlycoWorks HILIC
SPE for the Quantitative and Robust Recovery of N-Linked Glycans from mAb-
Type Samples. Waters Appication Note 720004717EN 2013.
20. Yu, Y. Q.; Gilar, M.; Kaska, J.; Gebler, J. C., A Deglycosylation and Sample
Cleanup Method for Mass Spectrometry Analysis of N-linked Glycans. Waters
Application Note 720001146EN 2007.
21. Yu, Y. Q.; Gilar, M.; Kaska, J.; Gebler, J. C., A rapid sample preparation method
for mass spectrometric characterization of N-linked glycans. Rapid Commun
Mass Spectrom 2005, 19 (16), 2331–6.
22. Lauber, M. A.; Fournier, J. L.; Koza, S. M.; Fountain, K. J., GlycoWorks
HILIC SPE Robust Glycan Sample Preparation. Waters Technology Brief
720005116EN 2014.
[ 75 ]
Rapid Preparation of Released N-Glycans for HILIC Analysis Using a Novel Fluorescence and MS-Active Labeling Reagent
WAT E R S S O LU T IO NS
GlycoWorks™ RapiFluor-MS™ N-Glycan Kit
GlycoWorks HILIC µElution Plate
RapiFluor-MS Glycan Performance
Test Standard
RapiFluor-MS Dextran Calibration Ladder
ACQUITY UPLC® Glycan BEH Amide,
130Å Column
XBridge® Glycan BEH Amide,
130Å Column
ACQUITY UPLC H-Class Bio System
ACQUITY® QDa® Mass Detector
Xevo® G2-XS QTof MS
SYNAPT® G2-Si HDMS
K E Y W O R D S
GlycoWorks, RapiFluor-MS, RapiGest™ SF,
Rapid Tagging, PNGase F, Deglycosylation,
ACQUITY UPLC H-Class Bio System, BEH
Amide 130Å, Glycans, Glycoproteins,
Glycosylation, HILIC, Fluorescence
A P P L I C AT IO N B E N E F I T S
■
■
High yield sample preparation with
quantitative recovery to ensure accurate
and repeatable profiling of N-glycans
■
■
Comparability to historical
2-AB based released glycan
analysis approaches
■
■
RapiFluor-MS Glycan Performance Test
Standard for method familiarization,
troubleshooting, and benchmarking
■
■
Robust Glycan BEH Amide HILIC separations
supported by GU calibration with the novel
RapiFluor-MS Dextran Calibration Ladder
I N T RO DU C T IO N
N-glycosylation of proteins is routinely characterized and monitored because
of its significance to the detection of disease states1-3 and the manufacturing
of biopharmaceuticals.4-5 Glycosylation profiles are most often assessed by
means of released glycan analyses, wherein samples are often prepared by
techniques that are notoriously time-consuming or lead to compromises in MS
sensitivity.6-7 With the development of the GlycoWorks RapiFluor-MS N-Glycan
Kit, we have addressed these shortcomings by enabling unprecedented sensitivity
for glycan detection while also improving the throughput of N-glycan sample
preparation.8 Using the GlycoWorks RapiFluor-MS N-Glycan Kit, glycoproteins are
deglycosylated in 10 minutes to produce N-glycosylamines that are then rapidly
reacted with the novel RapiFluor-MS labeling reagent (Figure 1). In a final step, the
resulting labeled glycans are extracted from reaction byproducts by means of an
SPE method that facilitates immediate analysis of samples. As a result, an analyst
can now complete an N-glycan sample preparation, from glycoprotein to ready-
to-analyze sample, in just 30 minutes and be poised to perform high sensitivity
N-glycan profiling using hydrophilic interaction chromatography (HILIC) and mass
spectrometric (MS) or fluorescence (FLR) detection.
Equally important as the efficiency and sensitivity gains afforded by this new
sample preparation approach is its robustness and its ability to produce results
consistent with historical N-glycan profiling. Within this application note, we
will discuss these attributes of the RapiFluor-MS based sample preparation and
the corresponding HILIC-based LC analyses.
Robustness of
RapiFluor-MS N-Glycan Sample Preparations
and Glycan BEH Amide HILIC Chromatographic Separations
Matthew A. Lauber,1 Michael F. Morris,1 Darryl W. Brousmiche,1 and Stephan M. Koza1
1
Waters Corporation, Milford, MA, USA
[ 76 ]
E X P E R IM E N TA L
Method conditions
(unless otherwise noted)
LC conditions
LC system:
ACQUITY UPLC H-Class Bio
Sample temp.:
10 °C
Analytical
column temp.:
60 °C
Flow rate:
0.4 mL/min
Fluorescence
detection:
Ex 265/Em 425 nm (RapiFluor-MS)
Ex 330 / Em 420 nm (2-AB)
(2 Hz scan rate [150 mm column], Gain =1)
(5 Hz scan rate [50 mm column], Gain=1)
Injection volume:
≤1 µL (aqueous diluents with
2.1 mm I.D. columns)
≤30 µL (DMF/ACN diluted samples
with 2.1 mm I.D. columns)
Columns:
ACQUITY UPLC Glycan BEH Amide, 130Å,
1.7 µm, 2.1 x 50 mm
(p/n 186004740)
XBridge Glycan BEH Amide XP,
130Å, 2.5 µm, 2.1 x 150 mm
(p/n 186007265)
Agilent AdvanceBio Glycan Mapping Rapid
Resolution HD, 1.8 µm, 2.1 x 150 mm
Thermo Scientific Accucore™ 150 Amide
HILIC, 2.6 µm, 2.1 x 150 mm
Sample
collection/vials:
Sample Collection Module
Polypropylene 12 x 32 mm Screw Neck
Vial, 300 µL Volume (p/n 186002640)
Gradient used with 2.1 x 50 mm columns:
Mobile phase A:
50 mM ammonium formate, pH 4.4
(LC-MS grade; from a 100x concentrate,
p/n 186007081)
Mobile phase B:
Acetonitrile (LC-MS grade)
Time Flow rate
%A
%B
Curve
(mL/min)
0.0 0.4
25 75 6
11.7 0.4
46 54
6
12.2 0.2
100 0
6
13.2 0.2
100 0
6
14.4 0.2
25 75
6
15.9 0.4
25 75
6
18.3 0.4
25 75
6
Gradient used with 2.1 x 150 mm columns:
Mobile phase A:
50 mM ammonium formate, pH 4.4
(LC-MS grade; from a 100x concentrate,
p/n 186007081)
Mobile phase B:
Acetonitrile (LC-MS grade)
Time Flow rate
%A
%B
Curve
(mL/min)
0.0 0.4
25 75 6
35.0 0.4
46 54
6
36.5 0.2
100 0
6
39.5 0.2
100 0
6
43.1 0.2
25 75
6
47.6 0.4
25 75
6
55.0 0.4
25 75
6
Intact mass analysis was performed by LC-MS with a Xevo G2-QTof
MS conditions
MS system:
Xevo G2 QTof
Ionization mode:
ESI+
Analyzer mode:
TOF MS, resolution mode (~20 K)
Capillary voltage:
3.0 kV
Cone voltage:
45 V
Source temp.:
120 °C
Desolvation temp.:
350 °C
Desolvation gas flow: 800 L/Hr
Calibration:
NaI, 1 µg/µL from 500–5000 m/z
Acquisition:
500–5000 m/z, 1 Hz scan rate
Data management:
UNIFI® 1.7/MassLynx® Software (v4.1)
[ 77 ]
Robustness of RapiFluor-MS N-Glycan Sample Preparations and Glycan BEH Amide HILIC Chromatographic Separations
Sample description
RapiFluor-MS labeled N-glycans were prepared from
glycoproteins, including Intact mAb Mass Check
Standard (p/n: 186006552), using a GlycoWorks
RapiFluor-MS N-Glycan Kit (p/n: 176003606)
according to the guidelines provided in its Care
and Use Manual (715004793).
2-AB labeled N-glycans were prepared using a
Prozyme GlykoPrep® Rapid N-Glycan Preparation
with 2-AB Kit according to the manufacturer’s
recommended protocol. In addition, 2-AB labeled
N-glycans were also prepared using an approach
combining the use of a Prozyme GlykoPrep
Digestion Module, an in-house optimized 2-AB
labeling protocol, and a GlykoPrep Cleanup Module
(Prozyme, Hayward, CA).
1.5 x 107 pg IgG
1 pmol
150,000 pg
2 pmol glycan
1 pmol IgG
0.45 pmol FA2
1 pmol total
glycan pool
10 µL injection
400 µL
sample
prepared
X
X
X
X
= 2.3 pmol
15 min
5 min
10 min
Figure 1. GlycoWorks RapiFluor-MS N-Glycan Kit sample preparation workflow and
the chemical structure of the RapiFluor-MS Reagent.
RapiFluor-MS Glycan Performance Test Standard (p/n: 186007983) was reconstituted in 50 µL of water and
injected as a 1 µL volume for chromatographic benchmarking and lifetime testing experiments. RapiFluor-MS
Dextran Calibration Ladder (p/n: 186007982) was reconstituted in 100 µL of water and injected as a 1 µL
volume for retention time calibrations.
Percent yields for the sample preparation workflows were determined by means of quantitative analyses.
Column loads were calibrated using external quantitative standards of 2-AB labeled triacetyl chitotriose
and RapiFluor-MS derivatized propylamine obtained in high purity (confirmed by HPLC and 1H NMR).
To determine percent yields, the measured quantities of FA2 glycan from Intact mAb Mass Check Standard
(p/n: 186006552) were compared to theoretical yields calculated for the preparation. For example,
the theoretical yield for the FA2 glycan resulting from the GlycoWorks RapiFluor-MS N-Glycan Kit was
calculated as follows:
*This calculation is based on the assumption that the sample of Intact mAb Mass Check Standard was 15 µg, that the mAb has a
molecular weight of 150 kDa, that there are only 2 N-glycans per one mAb, that the N-glycan profile of the mAb contains the FA2
glycan at a relative abundance of 45%, and that only 2.5% of the sample was analyzed.
[ 78 ] Robustness of RapiFluor-MS N-Glycan Sample Preparations and Glycan BEH Amide HILIC Chromatographic Separations
R E SU LT S A N D D IS C U S S IO N
Robust sample preparation: Deglycosylation
Each procedural step in the GlycoWorks RapiFluor-MS
N-Glycan Kit has been optimized to be high yielding
and to minimize the introduction of bias to an
N-glycan profile. Previous work based on SDS
PAGE gel shift assays has demonstrated that the
rapid deglycosylation procedure developed for
this kit produces complete deglycosylation of a
diverse set of glycoproteins.8 This completeness of
deglycosylation is also supported by intact mass
analysis using LC-MS, where the deglycosylation of
a monoclonal antibody (mAb) can be readily tracked.
Figure 2 presents deconvoluted ESI mass spectra
for Intact mAb Mass Check Standard, a murine IgG1
mAb. The top spectrum shows the mAb before it had
been subjected to rapid deglycosylation (Figure 2A).
The bottom spectrum meanwhile presents the mAb
after it was processed according to the approach
specified in the GlycoWorks RapiFluor-MS N-Glycan
Kit, wherein glycoproteins are subjected to 1% (w/v)
RapiGest SF Surfactant-assisted heat denaturation
followed by incubation with Rapid PNGase F at
50 °C for 5 minutes (Figure 2B). The masses
observed in these spectra confirm that these
samples differ in terms of glycan occupancy. The
control sample contains the mAb in its doubly
glycosylated, native form (one glycan on each heavy
chain). In contrast, the sample subjected to the
proposed 2-step rapid deglycosylation procedure is
homogenous with an observed molecular weight that
is in agreement with the predicted molecular weight
of the fully deglycosylated mAb (145.3 kDa). And
although high temperatures are employed in this
method for the purpose of heat denaturation, no
detrimental effects on an N-glycan profile have
been observed. To this point, notice that there are
no differences in an N-glycan profile prepared
from pooled human IgG when using an excessive 20
minute heat denaturation at 90 °C versus the rapid
3 minute procedure (Figure 3).
145000
150000
145000
150000
148.4 kDa
145.3 kDa
No PNGase F
3 min
≥90˚C
5 min
50˚C
A
B
Pooled hIgG
90 ˚C
3 min denaturation
Pooled hIgG
90 ˚C
20 min denaturation
Recommended procedure
No significant change to the
glycan profile
A
B
Figure 2. Intact mass analysis of Intact mAb Mass Check Standard (A) before
and (B) after rapid deglycosylation with the GlycoWorks RapiFluor-MS
N-Glycan Kit.
Figure 3. Testing the effects of subjecting human IgG and its N-glycans to heat denaturation.
(A) The RapiFluor-MS N-glycan profile as observed using the recommended 3-minute heat
denaturation versus (B) the RapiFluor-MS N-glycan profile as observed using a 20-minute
heat denaturation.
[ 79 ]
Robustness of RapiFluor-MS N-Glycan Sample Preparations and Glycan BEH Amide HILIC Chromatographic Separations
Robust sample preparation: Rapid labeling
The efficiency of the sample preparation carries over
from deglycosylation to RapiFluor-MS labeling. A
primary concern in this step is the relative stability
of the PNGase F released N-glycosylamines, which
are required for RapiFluor-MS labeling, in the pH
7.9 GlycoWorks Rapid Buffer. A time-course study
involving varying delays between deglycosylation
and RapiFluor-MS labeling steps has shown that the
N-glycosylamines have a relatively long half-life of
approximately 2 hours at 50 °C (Figure 4). That is,
with our 5 minute deglycosylation step, there should
be little concern over sample loss (< 3% loss) due to
hydrolysis of the glycosylamine. In addition, sample
losses from the labeling reaction are minimal. Many
experimental parameters were explored during the
development of the rapid labeling reaction specified
in the RapiFluor-MS N-Glycan Kit, including pH,
temperature, ionic strength, time, buffer components,
and reagent molar excess. Figure 5 shows an example
of optimizing the reagent molar excess as needed to
maximize labeling yield. Fluorescence chromatograms
for labeled, released N-glycans from Intact mAb
Mass Check Standard are stacked on the left (Figure
5A). Note that with the GlycoWorks RapiFluor-MS
N-Glycan Kit proteins are purposely not depleted from
the sample after deglycosylation to save time and to
give better control over the labeling. The RapiFluor-MS
Reagent is therefore used in a molar excess over all
of the nucleophiles from the glycoprotein, which for
an IgG corresponds to approximately seventy five
protein amines and two N-glycosylamines. Each of the
corresponding samples was obtained from labeling
a fixed glycoprotein concentration of 0.36 mg/mL
with RapiFluor-MS Reagent at concentrations varying
from 18 to 108 mM. As shown in Figure 5B,
plotting of the fluorescence peak areas for the
resulting N-glycan profile indicates that labeling
is maximized near a RapiFluor-MS Reagent
concentration of 36 mM, the conditions designed
into the GlycoWorks RapiFluor-MS N-Glycan Kit.
Moreover, molar excess conditions both higher and
lower than the 36 mM reagent condition produced
comparable fluorescence profiles, underscoring the
robustness of RapiFluor-MS labeling.
FA2
FA2
5 m in
1 0 m in
2 0 m in
4 0 m in
1st Order
2nd Order
50 ˚C
100 min
120 min
37 ˚C
200 min
240 min
RT
400 min
480 min
E s tim a ted
t1/2
I ncubation tim e prior
to rapid labeling
A
B
Figure 4. Estimating the half-life of N-glycosylamine hydrolysis through a time-course on
deglycosylation incubation. (A) Fluorescence traces for RapiFluor-MS labeled FA2 from
Intact mAb Mass Check Standard observed after implementing varying incubation times for
deglycosylation (50˚C incubations). (B) Approximation of the N-glycosylamine half-life assuming
1st or 2nd order reaction kinetics.
Figure 5. Optimization of labeling reagent molar excess for the GlycoWorks RapiFluor-MS
N-Glycan Kit. (A) Fluorescence chromatograms for labeled glycans obtained by titration of
0.36 mg/mL deglycosylated Intact mAb Mass Check Standard with varying concentrations
of RapiFluor-MS Reagent. Separations were performed with labeled glycans from 0.4 µg of
glycoprotein and an ACQUITY UPLC Glycan BEH Amide, 130Å, 1.7 µm, 2.1 x 50 mm Column. (B)
Fluorescence peak area as a function of RapiFluor-MS Reagent concentration.
Reagent Concentration (mM)
RapiFluor-MS
Concentration
0.0E+0
1.6E+5
0
50
100
FA
2 Fluorescence
Pe
ak Area
RapiFluor-MS Reagent Concentration (mM)
18 mM
36 mM
54 mM
72 mM
108 mM
4
9 min
A
B
Maximized yield
[ 80 ] Robustness of RapiFluor-MS N-Glycan Sample Preparations and Glycan BEH Amide HILIC Chromatographic Separations
Robust sample preparation: μElution HILIC SPE
The last step in the sample preparation involves extraction of the
RapiFluor-MS labeled glycans from reaction byproducts using
HILIC SPE. This technique has been routinely used for preparations
of 2-AB labeled N-glycans and has now been optimized for
RapiFluor-MS labeled species.8-9 Previous studies have shown
that RapiFluor-MS labeled glycans are obtained through this SPE
processing at relatively high yields of approximately 74%.8 Nearly
all of the observed sample losses in this step are non-specific.
Figure 6 plots fluorescence peak areas for preparations of
N-glycans from Intact mAb Mass Checked Standard, in which the
final SPE elution volume was either 30, 90 or 180 µL. This plot
shows that SPE recovery is a function of elution volume and that
highest recoveries are achieved when employing large elution
volumes. To facilitate direct analyses, however, a compromise is
made such that a 90 µL elution volume is used in order to obtain
a relatively concentrated glycan eluate. Regardless of the elution
volume and absolute yield of glycans from the SPE sorbent, the
most important characteristic of this clean-up is that the observed
sample losses have been determined to be non-specific with no
significant bias being introduced to a glycan profile for a wide
range of glycans with diverse chemical properties, including
small, neutral glycans up to large, tetrasialylated glycans (see
Reference 8 for more details about GlycoWorks HILIC SPE).
Yield of RapiFluor-MS labeled N-glycans
In another measurement of robustness, it is worth looking at the
yield of N-glycans through the entire workflow. This was evaluated
in order to measure the collective efficiency of combining fast
deglycosylation, rapid labeling, and HILIC SPE extraction of
RapiFluor-MS labeled glycans (Figure 7). RapiFluor-MS labeled
N-glycans from Intact mAb Mass Check Standard were prepared,
analyzed by HILIC-FLR, and quantified by means of an external
calibration. Based on a calculated theoretical yield (see
experimental) and duplicate analyses, it was determined that
the percent yield through the entire RapiFluor-MS N-Glycan Kit
sample preparation was approximately 73%. To provide
perspective, we evaluated the yield of 2-AB labeled N-glycans
from an alternative sample preparation workflow involving the
use of a GlykoPrep Rapid N-Glycan Preparation with 2-AB Kit.
0E+0
1E+8
2E+8
3E+8
4E+8
5E+8
6E+8
0
30 60 90 120 150 180
Rapi
Fluor-MS
FA
2
Fluorescence
Pe
ak Area
SPE Elution Volume (
L)
Specified elution
volume
Figure 6. Fluorescence peak area as a function
of SPE elution volume. The specified elution
volume in the GlycoWorks RapiFluor-MS
N-Glycan Kit is 90 µL.
Figure 7. Percent yield for the preparation of RapiFluor-MS labeled N-glycans
with the GlycoWorks RapiFluor-MS N-Glycan Kit. Testing that has been
performed to confirm minimal sample loss and quantitative recovery is listed
for each procedural step. The percent yield that has been measured for the
preparation of 2-AB labeled N-glycans with a GlykoPrep Rapid N-Glycan
Preparation with 2-AB kit is also provided. These results may not be
representative of all applications.
FA2
Rep #1
1.6 pmol
Rep #2
1.7 pmol
100% Theoretical Yield = 2.3 pmol
Step
Yield
Testing to confirm minimal bias
Deglycosylation
Complete Intact mass analysis/subunit LC-MS
Gel shift assays
Labeling
>95%
Released glycan profile vs. subunit derived
glycan information
SPE
~74%
Recovery measurements
Glycan profile before vs. after SPE
GlycoWorks
RapiFluor-MS
N-Glycan Kit
~73% Yield
GlykoPrep® Rapid
N-Glycan Preparation
with 2-AB
~35% Yield
1
2
3
Quantitative analyses showed that 2-AB labeled N-glycans
are prepared using this kit with a relatively low yield of
approximately 35%, though it has been found that the yield
of this kit can be dramatically improved by optimization and
lengthening of the labeling step. Comparatively speaking,
though, these results show that not only does the RapiFluor-MS
approach quicken a historically time-consuming sample
preparation, it also exhibits reasonably high yields.
[ 81 ]
Robustness of RapiFluor-MS N-Glycan Sample Preparations and Glycan BEH Amide HILIC Chromatographic Separations
Minimal impact to glycan profiling
with reagent batch variation
Lastly, sample preparations with the GlycoWorks
RapiFluor-MS N-Glycan Kit have proven to be robust
with respect to reagent manufacturing. A robustness
study was performed to test the impact of changing
the batches of each reagent that plays a critical
role in the preparation of RapiFluor-MS labeled
N-glycans, namely RapiGest SF, GlycoWorks Rapid
Buffer, GlycoWorks Rapid PNGase F, RapiFluor-MS
Reagent, DMF Reagent Solvent, GlycoWorks µElution
SPE Plate, and the SPE Elution Buffer. Three sets of
these reagents, each varying by batch, were tested
in their application to profiling the N-glycans from
Intact mAb Mass Check Standard. Average relative
abundances observed for the glycan species in
this standard with the three different reaction sets
are presented in Figure 8. Relative abundances of
N-glycans were observed to be largely comparable
across the different preparations with an average
RSD for the labeled N-glycan species being 2.3%.
Comparability to 2-AB N-glycan profiling
Another critical aspect to the RapiFluor-MS
N-glycan sample preparation is that it can be
used in place of legacy 2-AB methods without
requiring significant adaptations to existing
analytical techniques. With the speed of the
sample preparation and the enhanced method
sensitivity afforded by the RapiFluor-MS tag,8 the
task of analyzing N-glycosylation is in fact made
significantly easier.
Just like 2-AB labeled glycans, RapiFluor-MS
labeled glycans are ideally suited for HILIC
separations with an amide bonded stationary
phase, such as that found in the Waters Glycan
BEH Amide Columns. Figure 9 shows example
separations for 2-AB and RapiFluor-MS labeled
glycans obtained from Intact mAb Mass Check
Standard. The 2-AB labeled glycans, in this case,
were prepared using the previously mentioned
GlykoPrep Kit and an approximately 3.5 hour
protocol, whereas the RapiFluor-MS labeled
glycans were prepared in less than 30 minutes
using a GlycoWorks RapiFluor-MS N-Glycan Kit.
Reacon set
Species
1
2
3
Avg
Std Dev
RSD
A2
1.23
1.25
1.21
1.23
0.021
1.70
FA2
44.54
43.95
43.29
43.93
0.622
1.42
FA2G1
19.81
20.01
19.95
19.92
0.101
0.51
FA2G1
22.54
22.86
22.91
22.77
0.199
0.87
FA2G2
8.93
9.02
9.42
9.12
0.264
2.90
FA2G2Sg1/FA2G2Ga2
0.92
0.82
0.94
0.89
0.060
6.71
Percent Abundance (%)
50
45
40
35
30
25
20
15
10
5
0
Reacon set 1
Reacon set 2
Reacon set 3
Pe
rcent Abundance (%)
A2
FA
2
FA
2G1
FA
2G1’
FA2G2
FA
2G2Sg1/
FA2G2Ga2
For each reaction set: unique batches of RapiGest SF, Rapid Buffer,
Rapid PNGase F, RapiFluor-MS Reagent, DMF Reagent Solvent,
GlycoWorks Elution SPE plate, SPE Elution Buffer.
2-AB
Rapi Fluor-MS
1
2 3
4
5 6
7
8
9 10
11
12
13 14 15 16 17 18
Peak
Glycan Name
1
A1
2
FA1
3
A2
4
FA2
5
A2G1
6
A2G1
7
FA2G1
8
FA2G1
9
FA2BG1
10
FA2BG1
11
FA2G2
12
FA2G1Ga1
13
FA2G2Ga1
14
FA2G2Sg1
15
FA2G2Sg1
16
FA2G2Ga2
17
FA2G2GaSg1
18
FA2G2Ga1Sg1
Figure 8. Characterization of batch-to-batch variation in the RapiFluor-MS sample preparation.
Percent abundances were measured for the preparation of RapiFluor-MS labeled N-glycans
from Intact mAb Mass Check Standard using three different sets of materials. Each reaction set
was represented by unique batches of RapiGest SF, Rapid Buffer, Rapid PNGase F, RapiFluor-MS
Reagent, DMF Reagent Solvent, GlycoWorks µElution SPE plate, SPE Elution Buffer. Testing was
performed in triplicate. FA2G1’ denotes the structural isomer of FA2G1.
Figure 9. Similarity between 2-AB and RapiFluor-MS N-glycan HILIC profiles for a typical mAb.
Fluorescence chromatograms for labeled glycans from Intact mAb Mass Check Standard using
an ACQUITY UPLC BEH Amide, 130Å, 1.7 µm, 2.1 x 150 mm Column. Peak identifications for
the RapiFluor-MS labeled N-glycans are provided. 2-AB labeled N-glycans were prepared using
a GlykoPrep Rapid N-Glycan Preparation with 2-AB kit. (') denotes a structural isomer.
[ 82 ] Robustness of RapiFluor-MS N-Glycan Sample Preparations and Glycan BEH Amide HILIC Chromatographic Separations
So that chromatograms exhibiting equivalent signal-to-noise could be compared, the RapiFluor-MS sample was
analyzed in this study at a significantly lower mass load than the 2-AB labeled sample. Despite being prepared
by different approaches, it can be seen that the labeled N-glycans are resolved by the HILIC separation into very
similar profiles. For a typical mAb profile, RapiFluor-MS and 2-AB labeling both yield HILIC glycan separations
with similar selectivity. However, as a result of its additional hydrogen bonding donors/acceptors,
the RapiFluor-MS label introduces a slight shift of the mAb N-glycan profile to higher retention times. This
change in the absolute retention window of an N-glycan profile is predictable and can therefore be easily
accounted for when transitioning from 2-AB to RapiFluor-MS based methods.
Consistency in results observed for the RapiFluor-MS-based approach compared to historical 2-AB
techniques was also evaluated. N-glycan profiling of the same monoclonal IgG1 reference sample has
been studied between these different methodologies. Figure 10 displays N-glycan information obtained
for this mAb sample throughout 160 different profiling experiments involving 2-AB labeling and HPLC
chromatography. Likewise, Figure 10 provides data from 12 recent experiments using RapiFluor-MS labeling
and UPLC® chromatography. Comparable relative abundances are observed for this sample in a direct
comparison (Figure 10, center panel) and a control chart demonstrates the ability to transition between
these two methods (Figure 10, right panel). This consistency in N-glycan profiling makes it possible to
replace time-consuming 2-AB/HPLC methods with RapiFluor-MS/UPLC techniques.
N-Glycan
2-AB
with HPLC
(n=160)
RapiFluor-MS
with UPLC
(n=12)
Avg (%)
Avg (%)
A1
0.33
0.28
FA1
0.23
0.42
A2
5.98
5.76
FA2
45.44
45.88
Man5
1.20
1.08
A2G1
1.91
1.86
A2G1
0.95
0.79
FA2G1
27.36
26.29
FA2G1
9.66
9.23
A2G2
0.29
0.32
FA2G2
6.00
5.58
FA2G2S1
0.31
0.51
FA2
FA2G1
FA2G1
A2
A2G1
FA2G2S1
2-AB
HPLC
RapiFluor-MS
UPLC
Area
%
50
45
40
35
30
25
20
15
10
5
0
Figure 10. Consistency between UPLC-based RapiFluor-MS N-glycan profiling and HPLC-based 2-AB N-glycan profiling of a humanized monoclonal IgG1. Comparison of
relative abundances for N-glycans detected using a method combining the GlycoWorks RapiFluor-MS N-Glycan Kit with a UPLC separation (n=12) versus a historical 2-AB
sample preparation combined with an HPLC separation (n=160). Trending data for the N-glycans from the human monoclonal IgG1 (light colored lines = 2-AB/HPLC, dark
colored lines = RapiFluor-MS/UPLC). FA2G1' and A2G1' denote the structural isomers of FA2G1 and A2G1, respectively.
[ 83 ]
Robustness of RapiFluor-MS N-Glycan Sample Preparations and Glycan BEH Amide HILIC Chromatographic Separations
Robustness of RapiFluor-MS N-glycan separations with glycan BEH amide columns
The robustness and resolving power of the HILIC column chromatography is critically important to successfully
implementing this methodology. To this end, a test standard called RapiFluor-MS Glycan Performance Test
Standard is available for method familiarization, system suitability, troubleshooting, and benchmarking
studies. This standard contains a complex mixture of RapiFluor-MS N-glycans from human IgG that has been
isolated from pooled human serum. Its composition of approximately 20 different major constituents makes it
useful for evaluating the resolving power of a separation and the sensitivity of detection methods (Figure 11).
1
2
3
4 5
6
7
8
9 10
11
12
13
14
15 16 17
Glycan
Labeled Glycan
RapiFluor-MS
Composition
Mi (Da)
Mavg (Da)
1
A2
C67H105O37N9
1627.6611 1628.5887
2
FA2
C73H115O41N9
1773.7190
1774.7299
3
FA2B
C81H128O46N10 1976.7984
1977.9224
4
A2G1
C73H115O42N9
1789.7140
1790.7293
5
A2G1
C73H115O42N9
1789.7140
1790.7293
6
FA2G1
C79H125O46N9
1935.7719
1936.8705
7
FA2G1
C79H125O46N9
1935.7719
1936.8705
8
FA2BG1
C87H138O51N10 2138.8512
2140.0630
9
FA2BG1
C87H138O51N10 2138.8512
2140.0630
10
A2G2
C79H125O47N9
1951.7668
1952.8699
11
FA2G2
C85H135O51N9
2097.8247
2099.0111
12
FA2BG2
C93H148O56N10 2300.9041
2302.2036
13 FA2G1S1
C90H142O54N10 2226.8673
2228.1251
14 FA2G2S1
C96H152O59N10 2388.9201
2390.2657
15 FA2BG2S1 C104H165O64N11 2591.9995 2593.4582
16 FA2G2S2
C107H169O67N11 2680.0150
2681.5203
17 FA2BG2S2 C115H182O72N12 2883.0949 2884.7128
RapiFluor-MS Glycan Performance Test Standard
Figure 11. RapiFluor-MS Glycan Performance Test Standard. An example fluorescence chromatogram obtained from an 8 pmole load of the standard and a separation
with an ACQUITY UPLC Glycan BEH Amide, 130Å, 1.7 µm, 2.1 x 150 mm Column. Peak identifications are provided. FA2G1', A2G1', and FA2BG1' denote the structural
isomers of FA2G1, A2G1, and FA2BG1 respectively.
In line with its intended purpose, we have used the RapiFluor-MS Glycan Performance Test Standard to
benchmark the chromatographic performance of four different columns containing amide bonded stationary
phases designed for glycan separations. Two of the columns were UPLC-based and contained sub-2-µm
particles while the remaining two were intended for use on HPLC instrumentation and contained 2.5 µm
and 2.6 µm particles. Figure 12 shows representative chromatograms obtained with each of these columns
run under equivalent conditions and linear velocities. Four glycan species spread across these separations
were monitored to measure retention windows, average peak widths, and peak capacities. Notice that
whether performing a separation with a phase intended for UPLC or HPLC chromatography, Glycan BEH
Amide Columns provide exemplary resolving power and comparable selectivities thereby enabling the
seamless transfer of this glycan separation between HPLC and UPLC platforms.10
[ 84 ] Robustness of RapiFluor-MS N-Glycan Sample Preparations and Glycan BEH Amide HILIC Chromatographic Separations
Figure 12. Chromatographic
benchmarking of HILIC columns
containing amide bonded
stationary phases designed
for glycan separations.
Fluorescence chromatograms
of the RapiFluor-MS Glycan
Performance Test Standard
were obtained from an
8 pmole load of the standard
and separations with
2.1 x 150 mm columns. All
separations were performed
at the same linear velocity on
an ACQUITY UPLC H-Class Bio
System. Four glycan species
spread across the separations
were monitored to measure
retention windows, average
peak widths, and
peak capacities.
Agilent AdvanceBio
Glycan Mapping, 1.8 m
ACQUITY UPLC
Glycan BEH Amide,
130Å, 1.7 µm
Thermo Accucore Amide,
150Å, 2.6 m
XBridge Glycan BEH Amide XP ,
130Å, 2.5 m
Pc* = 94
RT1,4 = 12.39 min
Wh,avg = 7.87 sec
Pc* = 83
RT1,4 = 12.80 min
Wh,avg = 9.28 sec
Pc* = 82
RT1,4 = 12.91 min
Wh,avg = 9.48 sec
Pc* = 76
RT1,4 = 10.57 min
W
Comparative separations may not be representative of all applications.
h,avg = 8.40 sec
1
2
3
4
1
2
3
4
1
2
3
4
4
3
2
1
3 - FA2G2S1
2 – FA2G1
1 – A2
4 – FA2BG2S2
3 - FA2G2S1
2 – FA2G1
4 – FA2BG2S2
Separations of RapiFluor-MS labeled glycans with glycan BEH amide columns have also proven to be very
robust. In demonstration of this, a single Glycan BEH Amide, 130Å, 1.7 μm Column was subjected to lifetime
testing and 300 sequential runs. At every 20th run, RapiFluor-MS Glycan Performance Test Standard was
separated in order to track any changes in the retentivity and selectivity of the column.
Chromatograms corresponding to the 1st and 300th runs are provided in Figures 13A and 13B,
respectively. Quite clearly, near identical separations were obtained at the onset as well as at the end of
this approximately 2-week constant use scenario, with no significant shifts in retention times of the labeled
N-glycans having been observed throughout the testing (Figure 13C).
y = 6E-5x + 5.4777
R = 0.09
y = 4E-5x + 6.5806
R = 0.03
y = 2E-5x + 8.6452
R = 0.01
y = -3E-5x + 9.9527
R = 0.02
4
5
6
7
8
9
10
11
0
100
200
300
GU
Va
lu
e
Run
5
10
15
20
25
30
0
100
200
300
Re
te
nt
io
n
Ti
me
(m
in
)
Run
300th Run
RapiFluor-MS Glycan
Performance Test Standard
(after 12 days, 10° C)
1st Run
RapiFluor-MS Glycan
Performance Test Standard
(Initial)
1
2
3
4
3 – FA2G2S1
2 – FA2G1
1 – A2
4 – FA2BG2S2
3 – FA2G2S1
2 – FA2G1
1 – A2
4 – FA2BG2S2
1
2
3
4
A
B
C
D
Glycan BEH Amide, 130Å, 1.7
m
Figure 13. Robustness testing
of an ACQUITY UPLC Glycan
BEH Amide, 130Å, 1.7 µm
2.1 x 150 mm Column for
separations of RapiFluor-
MS labeled N-glycans.
Fluorescence chromatograms
of the RapiFluor-MS Glycan
Performance Test Standard
were obtained at every 20th
run from an 8 pmole load of the
standard. Four glycan species
spread across the separations
were monitored to track the
retentivity of the stationary
phase and column. Fluorescence
chromatograms are shown
for the (A) 1st run and the
(B) 300th run with the column.
(C) Retention times as a function
of run. (D) Glucose unit (GU)
values as a function of run.
[ 85 ]
Robustness of RapiFluor-MS N-Glycan Sample Preparations and Glycan BEH Amide HILIC Chromatographic Separations
In this testing, LC calibrations were performed after every
separation of the glycan mixture through application of a dextran
ladder and assignment of glucose unit (GU) values. Separations
with glycan BEH amide columns can be used in conjunction with
glucose unit (GU) values as a means to calibrate HILIC-based
glycan separations. Use of GU values minimizes subtle retention
time variations between runs and between different instruments
by expressing chromatographic retention in terms of standardized
GU values.11 To assign GU values, a dextran ladder (comprised of
glucose multimers of increasing length) is used as an external
calibrant. The retention times of the glucose multimers are then
used via cubic spline fitting to convert glycan retention times
into GU values.
The development of a dextran calibration ladder suitable for use
with RapiFluor-MS labeled glycans was essential yet technically
challenging. Given that dextran is a reducing sugar without a
strong nucleophile, it cannot, unlike N-glycosylamines, be readily
labeled with RapiFluor-MS Reagent. Because of the distinctive
urea linkage imparted to N-glycans upon their derivatization
with rapid tagging reagents, RapiFluor-MS labeled N-glycans
have very unique fluorescence maxima at approximately 265
nm (excitation) and 425 nm (emission) (Figure 14A). In a novel
labeling approach, we have prepared a RapiFluor-MS Dextran
Calibration Ladder by first reductively aminating dextran with
ethanolamine and then labeling it with RapiFluor-MS.
The resulting urea-linked dextran derivatives exhibit identical
fluorescence properties to those of RapiFluor-MS labeled
N-glycans. Furthermore, the obtained dextran is tuned for desired
HILIC retention because of the hydroxyl group being incorporated
through ethanolamine. A representative fluorescence
chromatogram for this novel dextran ladder is provided in Figure
14B, and an example cubic spline fit of the retention data is
shown in Figure 14C.
The impact of implementing GU value calibration is exemplified
in Figure 13D, where the retention time data throughout the
Glycan BEH Amide lifetime testing are reported in GU values.
In comparing the retention time data shown in Figure 13C to the
GU data in Figure 13D, one can see that the subtle fluctuations
in retention times across the 2-week lifetime testing are
compensated for by the GU calibration. In fact, RSDs for the GU
value data are reduced by a factor of 2 compared to the RSDs in
the retention time data.
Also, linear regression analysis of the GU value data shows
that there is essentially no drifting in the HILIC retention data
once calibrated using a dextran ladder. This analysis therefore
clearly demonstrates the value of GU calibration with respect to
improving the quality of reported data.
Emission Wavelength (nm)
235 nm
240 nm
245 nm
250 nm
255 nm
260 nm
265 nm
270 nm
275 nm
280 nm
285 nm
290 nm
295 nm
Excitation,Max=265 nm
Emission,Max=425 nm
Ex 265 nm
Em 425 nm
Reductive Amination
With an Analog
400
500 nm
400
500 nm
400
500 nm
x
x
!#$%%$&'()*+,-,'./0(1'#2
&'$()*
+&$()*
+,!()*
x
!#$%%$&'()*+,-,'./0(1'#2
RapiFluor-MS
Labeled N-Glycans
Ex 265 nm
Em 425 nm
Ex 370 nm
Em 480 nm
Dextran
Novel Dextran Ladder
Dextran
Glycan
A
B
C
GU
3
GU
7
GU
11
GU
15
Cubic Spline Fit
Figure 14. Assignment of Glucose Unit (GU) values with
RapiFluor-MS labeling. (A) Chemical structures and
fluorescence spectra of RapiFluor-MS labeled N-glycans
versus dextrans derivatized with RapiFluor-MS-like labels.
The novel dextran ladder that has been commercialized as the
RapiFluor-MS Dextran Calibration Ladder is highlighted. (B)
An example fluorescence chromatogram for the RapiFluor-MS
Dextran Calibration obtained for a 0.5 µg mass load with
an ACQUITY UPLC Glycan BEH Amide, 130Å, 1.7 µm,
2.1 x 150 mm Column. (C) Calibration curve resulting
from cubic spline fitting.
[ 86 ] Robustness of RapiFluor-MS N-Glycan Sample Preparations and Glycan BEH Amide HILIC Chromatographic Separations
Waters Corporation
34 Maple Street
Milford, MA 01757 U.S.A.
T: 1 508 478 2000
F: 1 508 872 1990
www.waters.com
Waters, The Science of What’s Possible, ACQUITY UPLC, XBridge, ACQUITY, UPLC, QDa, Xevo, SYNAPT, UNIFI, and MassLynx are registered
trademarks of Waters Corporation. GlycoWorks, RapiFluor-MS, and RapiGest are trademarks of Waters Corporation. All other trademarks are
property of their respective owners.
©2015 Waters Corporation. Produced in the U.S.A. May 2015 720005370EN AG-PDF
C O N C LU S IO NS
In this application note, we have demonstrated the robustness of RapiFluor-MS
N-glycan preparations and Glycan BEH Amide HILIC Column chromatography.
The RapiFluor-MS N-Glycan Kit enables analysts to perform a high yielding
sample preparation with quantitative recovery that ensures accurate and
repeatable profiling of N-glycans that is highly comparable to HPLC, 2-AB based
methodologies. Moreover, it has been demonstrated that Glycan BEH Amide
Columns afford exemplary resolving power and ruggedness for separations of
RapiFluor-MS labeled N-glycans. Additionally, this separation can be readily
transferred between UPLC and HPLC platforms. To further ensure success with
these new methodologies, two standards have been commercialized, and their
use to facilitate RapiFluor-MS analyses has been demonstrated. The RapiFluor-
MS Glycan Performance Test Standard has been used for benchmarking studies,
while the novel RapiFluor-MS Dextran Calibration Ladder has been employed
to enhance the reproducibility of chromatographic retention time data. In
summary, the GlycoWorks RapiFluor-MS N-Glycan Kit and supporting standards
and columns can significantly reduce the burdens associated with N-glycan
profiling while providing accurate, reproducible, and sensitive analyses.
Acknowledgement
We would like to thank Peter De Vreugd and Mark Eggink from Synthon
Biopharmaceuticals BV for providing data on the comparability of released
N-glycan analyses with RapiFluor-MS and UPLC versus 2-AB and HPLC.
References
1. Ohtsubo, K.; Marth, J. D., Glycosylation in cellular
mechanisms of health and disease. Cell 2006, 126 (5),
855–67.
2. Mechref, Y.; Hu, Y.; Garcia, A.; Zhou, S.; Desantos-Garcia, J.
L.; Hussein, A., Defining putative glycan cancer biomarkers
by MS. Bioanalysis 2012, 4 (20), 2457–69.
3. Ruhaak, L. R.; Miyamoto, S.; Lebrilla, C. B., Developments in
the identification of glycan biomarkers for the detection of
cancer. Mol Cell Proteomics 2013, 12 (4), 846–55.
4. Dalziel, M.; Crispin, M.; Scanlan, C. N.; Zitzmann, N.;
Dwek, R. A., Emerging principles for the therapeutic
exploitation of glycosylation. Science 2014, 343 (6166),
1235681.
5. Beck, A.; Wagner-Rousset, E.; Ayoub, D.; Van Dorsselaer,
A.; Sanglier-Cianferani, S., Characterization of therapeutic
antibodies and related products. Anal Chem 2013, 85 (2),
715–36.
6. Mechref, Y.; Hu, Y.; Desantos-Garcia, J. L.; Hussein, A.;
Tang, H., Quantitative glycomics strategies.
Mol Cell Proteomics 2013, 12 (4), 874–84.
7. Ruhaak, L. R.; Zauner, G.; Huhn, C.; Bruggink, C.;
Deelder, A. M.; Wuhrer, M., Glycan labeling strategies and
their use in identification and quantification. Anal Bioanal
Chem 2010, 397 (8), 3457–81.
8. Lauber, M. A.; Brousmiche, D. W.; Hua, Z.; Koza, S. M.;
Guthrie, E.; Magnelli, P.; Taron, C. H.; Fountain, K. J., Rapid
Preparation of Released N-Glycans for HILIC Analysis
Using a Novel Fluorescence and MS-Active Labeling
Reagent. Waters Application Note 720005275EN 2015.
9. Yu, Y. Q.; Gilar, M.; Kaska, J.; Gebler, J. C., A rapid sample
preparation method for mass spectrometric characterization
of N-linked glycans. Rapid Commun Mass Spectrom 2005,
19 (16), 2331–6.
10. Koza, S. M.; Lauber, M. A.; Fountain, K. J., Successful
Transfer of Size-Exclusion Separations between HPLC and
UPLC. Waters Application Note 720005214EN 2015.
11. Campbell, M. P.; Royle, L.; Radcliffe, C. M.; Dwek, R. A.;
Rudd, P. M., GlycoBase and autoGU: tools for HPLC-based
glycan analysis. Bioinformatics 2008, 24 (9), 1214–6.
[ 87 ]
Robustness of RapiFluor-MS N-Glycan Sample Preparations and Glycan BEH Amide HILIC Chromatographic Separations
Matthew A. Lauber, Jennifer L. Fournier, Stephan M. Koza, and Kenneth J. Fountain
G OA L
To demonstrate the benefits of using HILIC as
an ideal separation mode for glycans as well
as the robustness of HILIC in an SPE format.
BAC KG RO U N D
Reversed-phase (RP) chromatography,
though universally accepted for a majority
of compounds, is not particularly well suited
for analytes that are hydrophilic in nature.
When subjected to RP chromatography, polar
compounds are often poorly retained or separated
with non-optimal selectivity. Alternatively,
Hydrophilic Interaction Chromatography (HILIC)
can be used to successfully improve retention
of very polar species, such as the glycans
encountered during the characterization of
protein therapeutics. There are over 40 highly
cited published papers using HILIC chemistry for
the separation of glycans (search on SciFinder
June 2014). One reason for this is that this
mode of separation is ideal for these types of
compounds due to their highly polar nature.
Polar analytes can be strongly and selectively
retained onto a HILIC stationary phase when a low
polarity mobile phase is used. The concept can be
simplified by describing it as like-attracts-like.2
Specifically, the retention of glycans onto a HILIC
stationary phase can be explained in terms of the
hydrogen bonding as well as ionic and dipole-
dipole interactions that occur while the glycans
partition into an immobilized water layer.
This technique can be exploited as a form of solid
phase extraction (SPE) with the goal to clean up
and concentrate the analyte of interest,
in this case glycans.
Glycans are highly polar biomolecules, making them
amenable to Hydrophilic Interaction Chromatography
(HILIC) based Solid Phase Extraction (SPE).
GlycoWorks HILIC SPE
Robust Glycan Sample Preparation
Through sequential load wash
and elution steps, one can
successfully purify glycans
while removing many of the
less polar “contaminants”
found in deglycosylation and
post-derivatization mixtures.
Overall, HILIC provides many
benefits as an attractive mode
of separation for glycans.1
Figure 1. Benefits of HILIC SPE.
T H E S O LU T IO N
HILIC method robustness
In the case of GlycoWorks™ SPE, a silica-based aminopropyl sorbent was chosen
from several tested due to its highly polar nature. This stationary phase readily and
selectively retains polar compounds such as glycans. In addition, this sorbent
possesses a weakly basic surface allowing the added potential to exploit ion
exchange/repulsion properties.
The eluent pH and ionic strength can impact retention on this HILIC phase and
cannot be ignored. For instance, as the ionic strength of the eluent is increased,
ionic interactions between the stationary phase and the solutes are disrupted,
resulting in ion exchange playing a lesser role in retention.1
HILIC
Enhanced
Retention for
Polar
Compounds
Orthogonal
Selectivity
to RP
Multi-modal retention
mechanisms result
in high retention
Compounds elute with
increasing polarity
[ 88 ]
Figure 1. Benefits of HILIC SPE.
Figure 2. Recovery of
2-AB labeled N-glycans
(Figure 2) from
GlycoWorks HILIC SPE
96-well µElution Plates
(30 pmol of glycans
processed). Percent
recoveries as a function
of experimentally
determined glucose
units (GU) are shown
for various elution
conditions. Values are
based on the average of
three replicate analyses.
Trisialylated
25 mM
NH4HCO3
100 mM
NH4OAc
25 mM
NH4OAc
10 mM
NH4OAc
Lyophilization
Control
10%
ACN
5%
ACN
Lyophilization
Control
20%
ACN
25 mM
NH4HCO3
100 mM
NH4HCO3
Lyophilization
Control
10 mM
NH4HCO3
0 mM
NH4HCO3
Monosialylated
Neutral
Elution with 25 mM NH4HCO3
Various Concentrations of ACN
Elution with 5% ACN
4
Various Concentrations of N HOAc
Elution with 20% ACN
Various Concentrations of NH4HCO3
A
B
C
5
7
9
11
5
5
7
9
11
Experimental GU
20
40
60
80
100
7
9
11
%
Re
covery
0
In addition, as the mobile phase pH is altered, the charge state of the stationary
phase surface is impacted (by nature of the aminopropyl ligand and the base
particles’ silanol activity). At higher pH values the ionization of basic analytes and
the aminopropyl ligand is reduced, however the ionization of the surface silanols
is increased. Because of this, it is critical to choose an an eluent that will have a
fixed, stable pH every time an assay is performed. Method development taking
into account all of these factors is crucial for successful assay reproducibility.1
The first step to HILIC SPE is conditioning with aqeuous mobile phase, which establishes
a layer of polar solvent on the stationary phase surface. It is into this aqueous layer that
the glycans will partition when they are loaded under low polarity solvent conditions.
Subsequently, the adsorbed glycans are washed with solvent to ensure less polar
compounds are removed from the sample. Thereafter, the glycans are eluted off the
stationary phase with a strongly polar solution, in this case a high concentration of
water with a buffer in order to minimize the ionic interactions. The use of ammonium
salts of formic acid or acetic acid (ammonium formate or ammonium acetate) are
preferentially used due to their volatility. Refer to the application note Optimization
of GlycoWorks HILIC SPE for the Quantitative and Robust Recovery of N-Linked
Glycans from mAb-Type Samples and one of its corresponding figures (Figure 2) for
a demonstration on how an eluent for GlycoWorks HILIC SPE was optimized for the
reproducible elution of 2-AB labeled glycans.3
In this work, 2-AB labeled glycans were loaded onto a 96-well HILIC µElution plate
according to the protocol provided in the GlycoWorks High-throughput Sample
Preparation Kit Care and Use Manual. Various eluents were then employed for elution
of the labeled glycans and recoveries for each major species in the test mixture were
subsequently determined. These data were compared alongside the recoveries of the
glycans from just the lyophilization and reconstitution steps that were performed after
the HILIC SPE procedure in preparation of the samples for HILIC-FLR. By means of this
development work, a 100 mM NH4OAc, 5% ACN eluent was selected as the optimal
elution condition, since it is a pH stable solution and it provided high as well
as relatively unbiased analyte recoveries.
SPE robustness
Many times, due to the nature of solid phase extraction
devices, especially in the case of micro-elution plates,
well to well flow rate variation can occur. At times, it
can be observed that one well will exhibit a faster flow
than another. Such well-to-well flow rate differences
can be further amplified by more viscous solvent such
as highly aqueous solvents. Other variables that can
contribute to flow variability include improper seating
of the plate on the vacuum manifold, introduction of
air when pipetting into a well and changes in operating
pressures when vacuum driven SPE is performed.
Optimization of individual vacuum manifolds can
be critical to achieving consistent well-to-well flow
rates. Nevertheless, it is almost impossible to entirely
eliminate well to well flow rate differences so the
question was asked, what impact does this have on the
resulting reproducibility of oligosaccharide recovery?
If the method is robust, this mechanical variability
should be inconsequential.
To this end, the GlycoWorks Care and Use
protocol was employed along with the Dextran
Calibration Ladder (PN 186006841) to test the
impact of variable flow rates on sample recovery
(GlycoWorks High-throughput Sample Preparation
Kit Care and Use Manual).4 As an initial step,
500 µL of water was pulled through wells on a
GlycoWorks HILIC SPE uElution plate. A vacuum
setting of 5 inches of Hg was used to drive flow
through one well at a time.
[ 89 ]
GlycoWorks HILIC SPE Robust Glycan Sample Preparation
Waters Corporation
34 Maple Street
Milford, MA 01757 U.S.A.
T: 1 508 478 2000
F: 1 508 872 1990
www.waters.com
Waters and The Science of What’s Possible are registered trademarks of Waters Corporation. GlycoWorks
is a trademark of Waters Corporation. All other trademarks are the property of their respective owners.
©2014 Waters Corporation. Produced in the U.S.A. September 2014 720005116EN LM-PDF
Meanwhile, the time to draw the 500 µL through each
well was recorded, enabling a flow rate metric for
each well to be determined. Sample was thereafter
processed, without the use of any special practices.
Figure 3A shows the measured flow rates. The color
gradient denotes faster flow rates with darker colors
and slower flow rates with lighter colors.
Figure 3B illustrates the fluorescence chromatograms
obtained for oligosaccharide samples processed
through wells with varying flow rate characteristics.
Overall, the GlycoWorks HILIC SPE plate exhibited
less than a 3-fold variation in flow rate across a set
of 48 wells when operated with a vacuum manifold
(Figure 3A).
When these data were viewed in the context of
oligosaccharide recovery, however, it became
clear that any differences that do exist in flow rate
characteristics from well-to-well do not negatively
impact recovery, whether absolute or relative, as
demonstrated in Figure 4.
S UMM A RY
GlycoWorks HILIC SPE offers a robust, reliable
solution for cleanup and concentration of glycans
from complex matrices. Its application during the
sample preparation of 2-AB labeled glycans helps
ensure that an analyst reliably obtains successful
results when studying protein glycosylation.
0
2
2
16
0
2
2
16
0
2
2
16
0
2
2
16
A7
(14 µL/sec)
B7
(23 µL/sec)
C7
(27 µL/sec)
D7
(16 µL/sec)
GlycoWorks HILIC µElution Plate (µL/sec)
EU
EU
EU
EU
A
B
GU 4
GU 11
Dextran Test Mixture
Load, Wash, Elute
Analysis by HILIC-FLR
%RSD of 26
R² = 0.01
R² = 0.06
R² = 0.05
0
10
20
30
10
20
30
FLR
Pe
ak Area (GU 4 through 11)
R² = 0.09
0
3000
6000
10
20
30
Flow Rate Aqueous
(µL/sec, -5 in Hg)
Flow Rate Aqueous
(µL/sec, -5 in Hg)
R2 = 0.09
Absolute recoveries are highly similar
with no correlation to flow rate
Individual recoveries are highly similar
with no correlation to flow rate
GU 11
GU 8
GU 5
FLR
Pe
ak Area (GU 4 through 11)
A
B
Figure 3. Well-to-well reproducibility testing of the GlycoWorks HILIC µElution Plate. (A) Flow rates
(µL/sec) observed during an aqueous conditioning step. (B) HILIC-Fluorescence (FLR) chromatograms
obtained using eluate of the dextran test mixture from wells A7, B7, C7, and D7.
Figure 4. Well-to-well reproducibility of the GlycoWorks HILIC µElution Plate. (A) Combined
fluorescence peak area for dextran test mixture eluted from 48 different wells displayed as
a function of flow rate. (B) Peak areas of Glucose Unit (GU) 5, 8, and 11 from the test mixture
plotted as a function of flow rate.
References
1. Comprehensive Guide to HILIC - Hydrophilic Interaction Chromatography. Waters Primer
2. Wuhrer, M.; de Boer, A. R.; Deelder, A. M., Structural glycomics using hydrophilic interaction
chromatography (HILIC) with mass spectrometry. Mass Spectrom Rev 2009, 28 (2), 192–206.
3. Optimization of GlycoWorks HILIC SPE for the Quantitative and Robust Recovery of N-Linked
Glycans from mAb-Type Samples. Waters Application Note 720004717EN.
4. GlycoWorks High-throughput Sample Preparation Kit Care and Use Manual. Waters User Manual
715004079.
[ 90 ] GlycoWorks HILIC SPE Robust Glycan Sample Preparation
WAT E RS SO LU T IONS
GlycoWorks™ RapiFluor-MS N-Glycan Kit
Intact mAb Mass Check Standard
ACQUITY UPLC® Glycan BEH Amide,
130Å, 1.7 μm, Column
ACQUITY UPLC H-Class Bio System
Xevo® G2-XS QTof Mass Spectrometer
MassLynx® 4.1 Software
GlycoWorks Rapid Buffer, 5 mL
K E Y WO R DS
HILIC Chromatography, UPLC,®
HPLC, method transfer, N-glycans,
RapiFluor-MS, GlycoWorks
A P P LIC AT ION BEN E FIT S
■
■
RapiFluor-MS™ glycan labeling procedure
with larger volume, simplified liquid transfer
to improve ease of use and automatability.
INT RO DUC T ION
Recently, Waters® introduced a novel labeling reagent, RapiFluor-MS (RFMS),
that provides a fast, efficient, and reproducible sample preparation workflow and
unsurpassed fluorescent and MS sensitivity for released N-glycan profiling.1,2
This initial methodology was designed to accommodate the lowest possible
glycoprotein sample concentration and, as result, calls for several low volume
(1.2 to 7 µL) liquid transfers. Looking to minimize the impact of pipetting volume
inaccuracies, we have redesigned this sample preparation to make pipetting
volumes larger (≥10 µL) and thereby reduce the variation in the absolute
quantities of analytes and reagents that get delivered during the denaturation,
PNGase F deglycosylation, and RFMS labeling steps of the procedure.
In the following work, we demonstrate that RFMS labeled glycan samples
prepared using this alternative sample preparation scheme are comparable
to those produced by the previously published flexible volume procedure. By
virtue of its simplification and use of larger volumes, this protocol should be an
excellent fit for adoption of RFMS into Quality Control (QC) environments and
automated platforms.
Quality Control and Automation Friendly GlycoWorks RapiFluor-MS
N-Glycan Sample Preparation
Stephan M. Koza, Scott A. McCall, Matthew A. Lauber, and Erin E. Chambers
Waters Corporation, Milford, MA, USA
[ 91 ]
E X P E RIM ENTA L
Method conditions
LC system:
ACQUITY UPLC H-Class Bio System
Detection:
ACQUITY UPLC FLR Detector with
analytical flow cell
Wavelength:
265 nm Excitation,
425 nm Emission
Column:
ACQUITY UPLC Glycan BEH Amide,
130Å, 1.7 µm, 2.1 mm x 150 mm
(p/n 186004742)
Column temp.:
60 °C
Sample temp.:
10 °C
Injection volume:
10.0 µL
Mobile phase A:
50 mM ammonium formate (pH 4.4)
LC-MS grade water,
from a 100 x concentrate
(p/n 186007081)
Mobile phase B:
LC-MS grade acetonitrile
Gradient:
Flow rate
Time (mL/min.)
%A
%B
Curve
0.0
0.4
25
75
6
35.0
0.4
46
54
6
36.5
0.2
100
0
6
39.5
0.2
100
0
6
43.1
0.2
25
75
6
47.6
0.4
25
75
6
55.0
0.4
25
75
6
Sample vials:
Polypropylene 12 x 32 mm Screw Neck
Vial, 300 µL (p/n 186002640)
Data management:
MassLynx 4.1 Software
MS conditions for RapiFluor-MS released N-glycans
MS system:
Xevo G2-XS QTof
Ionization mode:
ESI+
Analyzer mode:
Resolution (~ 40,000)
Capillary voltage:
2.2 kV
Cone voltage:
75 V
Source temp.:
120 °C
Desolvation temp.:
500 °C
Source offset:
50 V
Desolvation gas flow: 600 L/Hr
Calibration:
NaI, 1 µg/µL from 100–2000 m/z
Acquisition:
700–2000 m/z, 0.5 sec scan rate
Lockspray:
300 fmol/µL Human Glufibrinopeptide B
in 0.1% (v/v) formic acid,
70:30 water/acetonitrile
every 90 seconds
Data management:
MassLynx 4.1 Software
[ 92 ] Quality Control and Automation Friendly GlycoWorks RapiFluor-MS N-Glycan Sample Preparation
R E SU LT S AND DIS CUSSION
Comparison of GlycoWorks RapiFluor-MS protocols
The GlycoWorks RapiFluor-MS sample preparation procedure
(Table 1) was developed to allow for maximum flexibility with
respect to the concentration of the sample being prepared for
released N-glycan analysis.3 By altering the addition of water, this
method is capable of preparing samples with concentrations as low
as 0.66 mg/mL. While this procedure was designed with significant
molar excesses of the critical reagents, such as denaturant, enzyme,
and the RFMS label, to produce reproducible results,4 a potential
drawback of this procedure is that several of the aliquoted
volumes are well below 10 µL. As such, the methodology is not
as amenable for adoption into certain QC laboratories, depending
on their internally imposed method requirements, or for use in
specific robotic platforms. Pipetted volume accuracy and precision
increases with volume and, based on the International Organization
for Standardization (ISO) requirements for mechanical pipette
accuracy and precision, it is at 10 µL or more that the lowest
permissible systematic and random errors are obtained
(Figure 1, Adapted from Reference 5). It should be noted that
these maximum permissible errors are doubled for the use of
multi-channel pipettes. For this reason, some laboratories prefer
to avoid procedures with pipetted volumes lower than 10 µL.
0
1
2
3
4
5
6
0
2
4
6
8
10
12
14
16
18
20
±%
Erro
r
Pipetted Volume (µL)
Permissable Systematic Error (±%)
Permissable Random Error (±%)
Figure 1. Trends in the
maximum permissible
pipette volume errors
based on ISO 8655.
Component
Flexible volume
standard tube
(1 mL tube)
Flexible volume
PCR tube
(200 µL tube)
2.0 mg/mL sample
7.5 µL
7.5 µL
5% RapiGest1
6.0 µL
3.0 µL
Water
15.3 µL
3.3 µL
PNGaseF
1.2 µL
1.2 µL
Total volume of
released N-glycan
sample
30 µL
15 µL
RFMS2
12.0 µL
6.0 µL
Total volume of the
labeled N-glycan
sample
42 µL aliquot
21 µL aliquot
ACN dilution
358 µL
179 µL
Total volume of
HILIC SPE Load
400 µL
200 µL
Table 1. Aliquoted volumes for GlycoWorks RapiFluor-MS Kit flexible-volume
protocols for 1 mL and 200 µL tubes.
RapiGest reconstitution:
10 mg with 200 µL buffer
or 3 mg w/60 µL buffer.
RFMS reconstitution: 23 mg in 335 µL DMF and
9 mg in 131 µL DMF (68.7 µg/µL) for Standard
Protocol or 23 mg in 168 µL DMF and 9 mg in
66 µL DMF (136.4 µg/µL) for PCR Protocol.
[ 93 ]
Quality Control and Automation Friendly GlycoWorks RapiFluor-MS N-Glycan Sample Preparation
To provide a protocol with transfer volumes of 10 µL or more, the
standard RFMS sample preparation procedure was reconsidered.
The primary changes made to the deglycosylation procedure
were that the addition of water was removed, and more dilute
solutions of RapiGest™ SF surfactant and PNGase F were used.
The final recommended conditions are presented in Table 2. This
revised procedure is designed to give optimal results using a
10 µL sample of glycoprotein at a concentration of 1.5 mg/mL
(15 µg of glycoprotein), however, samples that are more or less
concentrated can still potentially produce quality results.
It should be noted that at significant deviations from the optimal
sample quantity, i.e. <5 µg and >30 µg, the labeling reaction
can produce undesirable side reactions or low yields, so it is
recommended to assess these situations on a case-by-case basis.
The only fundamental difference in this revised procedure is that
the concentration of RapiGest SF is slightly higher (1.5%) during
the denaturation step versus the previous procedure (1.0%).
This increase is not predicted to cause any deleterious effects
and may provide some benefit for certain glycoprotein samples
that are particularly resistant to denaturation. In the following
step, PNGase F deglycosylation, concentrations of the principal
components (glycoprotein, RapiGest SF, and PNGase F) are
equivalent to the standard, flexible volume procedure.
In modifying the labeling step, the aliquoted amount of the RFMS
label solution was decreased from 12 to 10 µL to be consistent
with the other lowest pipetted volumes of the procedure. To
account for this volume change, the concentration of the RFMS
reagent was increased proportionally such that the final ratio
of RFMS to glycoprotein remains equivalent. In addition to the
protocol using the 1 mL reaction tubes provided in the kit, this
revised procedure, like the previous procedure, has also been
adapted for use with a 200 µL thermocycler tube. If using a
thermocycler with this new QC and automation friendly protocol,
it is necessary to divide the final released and labeled glycan
sample into two aliquots, or to transfer sample to a larger tube,
prior to dilution and SPE clean-up.
Component
Automated /QC
standard tube
(1 mL tube)
Automated/QC
PCR tube
(200 µL tube)
1.5 mg/mL sample
10 µL
10 µL
3% RapiGest1
10 µL
10 µL
Water
0 µL
0 µL
PNGaseF (diluted)2
10 µL
10 µL
Total volume of
released N-glycan
sample
30 µL
30 µL
RFMS3
10 µL
10 µL
Total volume of the
labeled N-glycan
sample
40 µL aliquot
Divide into
2 x 20 µL aliquots
ACN dilution
360 µL
2 x 180 µL
Total volume of
HILIC SPE Load
400 µL
200 + 200 = 400 µL
Table 2. Aliquoted volumes for GlycoWorks RapiFluor-MS Kit automation
and QC volume protocols for 1 mL and 200 µL tubes.
1
RapiGest reconstitution: 10 mg with 200 µL buffer + 135 µL water or
3 mg w/60 µL buffer + 40 µL water.
2
PNGase F dilution (contents of vial 30 µL + 220 µL water).
3
RFMS reconstitution: 23 mg in 280 µL DMF or 9 mg in 110 µL DMF
(82.5 µg/µL).
[ 94 ] Quality Control and Automation Friendly GlycoWorks RapiFluor-MS N-Glycan Sample Preparation
Comparing RapiFluor-MS labeled N-glycan profiles
To compare the standard, flexible volume procedure with its newly designed, QC and automation-friendly analog, analyses of the Waters
Intact mAb Mass Check Standard were performed. Samples from this murine monoclonal antibody were prepared and analyzed following
the two different protocols along with 1 mL sample tubes and single-channel pipettes. A comparison of representative chromatograms
for each of these sample preparations are presented in Figure 2. The labeled peaks were integrated and the quantitative results are
presented in Figure 3. The glycan species observed in this profile were assigned using online ESI-MS detection with a Xevo G2-XS QTof
Mass Spectrometer (Table 3). As can be clearly seen, the two procedures produce both qualitative and quantitative results that are
comparable and reproducible for peaks with relative abundances as low as 0.06%.
Peak #
Peak ID
1
FA1
2
A2
3
FA2
4
M5
5
FA1G1 A2G1
6
A2G1 (iso)
7
FA2G1
8
FA2G1 (iso)
10
FA2G1B
11
FA2G1B (iso)
12
FA2G2
13
FA2G1Ga1
14
FA2BG2
18
FA2G2Ga1
19
FA2G2Ga1 (iso)
22
FA2G2Sg1
23
FA2G2Ga2
24
Fa2G2GaSg1
25
Fa2G2GaSg1 (iso)
9, 15, 16, 17,
20, 21, 26
unidentified
Table 3. Figure 1 peak identifications based on mass
(Xevo G2-S QTof).
0.00
5.00
10.00
15.00
20.00
25.00
30.00
35.00
40.00
45.00
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
FLR Percent Peak Area
Peak #
Flexible Volume Protocol
Automated/QC Protocol
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40
1.60
1.80
2.00
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
FLR Percent Peak Area
Peak #
Figure 2. Quantitative comparison of RFMS labeled glycans prepared using GlycoWorks RapiFluor-MS Kit
flexible volume protocol and proposed automation and QC volume protocol. Inset shows zoomed view of results.
[ 95 ]
Quality Control and Automation Friendly GlycoWorks RapiFluor-MS N-Glycan Sample Preparation
Waters Corporation
34 Maple Street
Milford, MA 01757 U.S.A.
T: 1 508 478 2000
F: 1 508 872 1990
www.waters.com
Waters, The Science of What’s Possible, MassLynx, Xevo, UPLC, and ACQUITY UPLC are registered trademarks of Waters Corporation.
RapiFluor-MS, RapiGest, and GlycoWorks are trademarks of Waters Corporation. All other trademarks are the property
of their respective owners.
©2015–2016 Waters Corporation. Produced in the U.S.A. January 2016 720005506EN AG-PDF
CONC LUSIONS
The GlycoWorks RapiFluor-MS N-glycan sample preparation procedure has
been successfully adapted to be more amenable to automation and QC use by
adjusting pipetted volumes to ≥10 µL. This supplemental procedure requires a
glycoprotein sample concentration of 1.5 mg/mL to obtain optimal results. As
an added benefit, the dispensed aliquots of the sample and principal reagents
are equivalent in volume (10 µL), thereby providing greater assurance that
the relative amounts of these components will be equivalent regardless of the
systematic accuracy of the pipetting device that is used, which should result in
greater intra-laboratory and inter-laboratory reproducibility.
References
1. MA Lauber, DW Brousmiche, Z Hua, SM Koza, E Guthrie,
P Magnelli, CH Taron, KJ Fountain. Rapid Preparation of
Released N-Glycans for HILIC Analysis Using a Novel
Fluorescence and MS-Active Labeling Reagent,
Application Note [serial on the Internet]. 2015;
(720005275en).
2. MA Lauber, YQ Yu, DW Brousmiche, Z Hua, SM Koza,
P Magnelli, E Guthrie, CH Taron, KJ Fountain.
Rapid Preparation of Released N-Glycans for HILIC
Analysis Using a Labeling Reagent that Facilitates
Sensitive Fluorescence and ESI-MS Detection.
Analytical Chemistry 2015.
3. GlycoWorks RapiFluor-MS N-Glycan Kit Care and Use Manual
4. MA Lauber, MF Morris, DW Brousmiche, SM Koza.
Robustness of RapiFluor-MS N-Glycan Sample Preparations
and Glycan BEH Amide HILIC Chromatographic Separations,
Application Note [serial on the Internet].
2015; (720005370en).
5. ISO 8655-2:2002, pg. 6.
[ 96 ] Quality Control and Automation Friendly GlycoWorks RapiFluor-MS N-Glycan Sample Preparation
WAT E R S S O LU T IO NS
GlycoWorks™ RapiFluor-MS N-Glycan Kit
Intact mAb Mass Check Standard
RapiFluor-MS High Mannose Standard
ACQUITY UPLC® Glycan BEH Amide,
130Å, 1.7 µm
ACQUITY UPLC H-Class Bio System
Xevo® G2-XS QTof Mass Spectrometer
UNIFI® 1.7 Software
MassLynx® 4.1 Software
K E Y W O R D S
ACQUITY UPLC H-Class Bio System,
ACQUITY UPLC Glycan BEH Amide Column,
glycans, glycosylated protein, glycoprotein,
Glycosylation, N-Linked glycans, HILIC,
RapiFluor-MS labeling, Intact mAb
Mass Check Standard, IgG, monoclonal
antibody (mAb), high mannose N-glycans,
RapiFluor-MS High Mannose Standard
A P P L I C AT IO N B E N E F I T S
■
■
Optimized LC method to improve the HILIC
profiling of common mAb N-glycan species.
■
■
Improved resolution of high mannose
structures, in addition to sialylated
species and N-glycan structures
containing alpha-linked galactose units.
■
■
The use of RapiFluor-MS™ High Mannose
Standard in system suitability
testing and peak identification.
I N T RO DU C T IO N
Glycan characterization is at the forefront of the biopharmaceutical industry, since
most protein therapeutics possess N-glycosylation, if not also O-glycosylation.
The most common therapeutic modality, an IgG monoclonal antibody (mAb), is,
for instance, typically N-glycosylated at two conserved sites in its Fc domain.
Not surprisingly, the nature of the N-glycans on a mAb can impact its circulation
half-life and efficacy. As a result, it is particularly important that the N-glycans of
a mAb be well characterized and routinely monitored.
A powerful strategy for such an analysis involves releasing N-glycans
enzymatically, labeling them with a reagent to improve their detectability,
and profiling them using hydrophilic interaction chromatography (HILIC).1
Recent developments in released N-glycan profiling, made possible by the
novel RapiFluor-MS labeling reagent, have eclipsed conventional N-glycan
techniques by simplifying the steps in the procedure, reducing the overall
sample preparation times, and enabling unprecedented sensitivities for both
fluorescence and mass spectrometric detection.2 However, regardless of how
quickly a sample can be prepared or with what sensitivity it can be detected, it is
imperative for N-glycan profiling to exhibit optimal chromatographic resolution.
With there being many different sample types, it is often necessary to consider
tailoring liquid chromatographic (LC) methods to ensure that robust and optimal
chromatographic performance can indeed be achieved.
In this application note, we highlight the development of an LC method that
optimizes the chromatographic resolution for the released N-glycans that are
commonly found on mAbs, including the high mannose N-glycan structures
that are known to negatively affect circulation half-life as well as indicate
aberrant cell culture conditions. In this work, N-glycans from a mAb were
rapidly released with PNGase F, labeled with RapiFluor-MS and profiled by HILIC
using sensitive fluorescence and mass spectrometric detection. The separation
method was optimized to improve the resolution of high mannose glycans, those
terminated with N-glycolyl neuraminic acid (Sg) (a member of the broad class of
sialic acids), and species containing alpha-linked galactose monosaccharides.
Within this work, the utility of this newly developed LC method is demonstrated
by means of system suitability testing with high mannose spiked samples.
Profiling Released High Mannose and Complex N-Glycan Structures from
Monoclonal Antibodies Using RapiFluor-MS Labeling and Optimized
Hydrophilic Interaction Chromatography
Scott A. McCall, Matthew A. Lauber, Stephan M. Koza, and Erin E. Chambers
Waters Corporation, Milford, MA, USA
[ 97 ]
E X P E R IM E N TA L
Sample description
The Intact mAb Mass Check Standard (p/n 186006552) was
reconstituted in water to a concentration of 2 mg/mL. N-glycans
were released from a 15-μg aliquot of this murine mAb and labeled
with RapiFluor-MS using a GlycoWorks RapiFluor-MS N-Glycan Kit
(p/n 176003606) following the instructions provided in its care
and use manual (715004793). RapiFluor-MS-labeled N-glycans
were prepared for injection at a concentration of 0.5 pmol/μL
(as a mixture in a solvent composed of 90 µL SPE eluate,
100 µL dimethylformamide, and 210 µL acetonitrile).
RapiFluor-MS High Mannose Standard (p/n 186008317) was
reconstituted in water to produce a 5 pmol/μL solution. A series
of spiked samples were then prepared by mixing RapiFluor-MS
labeled glycans from Intact mAb Mass Check Standard with the
RapiFluor-MS High Mannose Standard and water. In this way,
four spiked samples containing 0.45 pmol/µL of sample-derived
N-glycans were prepared along with varying concentrations
of the high mannose glycans. Spiking produced samples
with mannose-5 (M5) at a relative abundance ranging from
approximately 0.2% to 2%.
RapiFluor-MS Dextran Calibration Ladder (p/n 186007982)
was reconstituted with 100 µL of water to produce a 0.5 µg/µL
solution. The GU assignments were calculated using a cubic spline
fitting method and UNIFI 1.7 Software.
LC conditions for RapiFluor-MS Released N-Glycans
Chromatographic separations were performed using the following
conditions, unless otherwise noted:
Universal N-Glycan Profiling Method
LC system:
ACQUITY UPLC H-Class Bio System
Sample temp.:
10 °C
Analytical
column temp.:
60 °C
Flow rate:
0.4 mL/min
Injection volume:
10 µL for DMF/ACN-diluted samples
or 1 μL for aqueous samples
Column:
ACQUITY UPLC Glycan BEH Amide,
1.7 µm, 2.1 x 150 mm (p/n 186004742)
Fluorescence
detection:
Ex 265 nm / Em 425 nm, 2 Hz
Mobile phase A:
50 mM aqueous ammonium formate,
pH 4.4 (LC-MS grade water; from
a 100x ammonium formate concentrate
(p/n 186007081)
Mobile phase B:
ACN (LC-MS grade)
Flow rate
Time (mL/min)
%A
%B
Curve
0.0
0.4
25
75
35.0
0.4
46
54
6
36.5
0.2
100
0
6
39.5
0.2
100
0
6
43.1
0.2
25
75
6
47.6
0.4
25
75
6
55.0
0.4
25
75
6
mAb N-Glycan Profiling Method
LC system:
ACQUITY UPLC H-Class Bio System
Sample temp.:
10 °C
Analytical
column temp.:
45 °C
Flow rate:
0.5 mL/min
Injection volume:
10 µL (DMF/ACN-diluted samples),
1 μL (aqueous samples)
Column:
ACQUITY UPLC Glycan BEH Amide 1.7
µm, 2.1 x 150 mm (p/n 186004742)
Fluorescence detection: Ex 265 nm / Em 425 nm, 2 Hz
Mobile phase A:
50 mM ammonium formate, pH 4.4
(LC-MS grade; from a 100x concentrate
(p/n 186007081)
Mobile phase B:
ACN (LC-MS grade)
Flow rate
Time
(mL/min)
%A
%B
Curve
0.00
0.5
20
80
3.00
0.5
27
73
6
35.0
0.5
37
63
6
36.5
0.2
100
0
6
39.5
0.2
100
0
6
43.1
0.2
20
80
6
47.6
0.5
20
80
6
55.0
0.5
20
80
6
[ 98 ]
Profiling Released High Mannose and Complex N-Glycan Structures from Monoclonal Antibodies
Using RapiFluor-MS Labeling and Optimized Hydrophilic Interaction Chromatography
MS conditions for RapiFluor-MS Released N-Glycans
MS system:
Xevo G2-XS QTof
Ionization mode:
ESI+
Analyzer mode:
Resolution (~40 K)
Capillary voltage:
2.2 kV
Cone voltage:
75 V
Source temp.:
120 °C
Desolvation temp.:
500 °C
Source offset:
50 V
Desolvation gas flow: 600 L/Hr
Calibration:
NaI, 0.1 µg/µL from 100–2000 m/z
Acquisition:
700–2000 m/z, 0.5 sec scan rate
Lockspray™:
100 fmol/µL human Glu-fibrinopeptide
B prepared in a solution composed of
70:30:0.1% water/acetonitrile/formic
acid, sampled every 90 seconds
Data management:
MassLynx 4.1 Software,
UNIFI 1.7 Software
R E SU LT S A N D D IS C U S S IO N
A Universal N-Glycan Profiling Method
The GlycoWorks RapiFluor-MS N-Glycan Kit facilitates the robust
analysis of many, very diverse N-glycans. Accordingly, we first
aimed to establish a separation method for an ACQUITY UPLC
Glycan BEH Amide Column that can be universally applied to
all types of N-glycans, from small biantennary structures up
to highly sialylated, tetraantennary species. It is very useful to
run this so-called ‘universal N-glycan profiling method’ when
analyzing new samples. Given that this method is the basis of an
upcoming glucose unit (GU) database for RapiFluor-MS labeled
glycans, it will also be the technique recommended for future
workflows involving GU based peak assignments.
Being a generic tool, the universal N-glycan profiling method is
not optimized for any particular N-glycan sample, including the
N-glycans obtained from the Intact mAb Mass Check Standard
that is provided as a control sample in each GlycoWorks RapiFluor-
MS N-Glycan Kit. Since this standard is a mAb expressed from
a murine cell line, it is a relevant surrogate to many mAb
therapeutics and will, in fact, produce a highly similar N-glycan
profile. The universal N-glycan profiling method produces
chromatograms where the mAb glycans elute in only
the first half of the analytical gradient, as shown in Figures 1A and
1B. Using online, mass spectrometric detection, at least
14 different N-glycans can be readily identified (Table 1). In
addition to having been eluted in a narrower retention window,
several of these glycans are only partially resolved and
consequently are difficult to monitor, and reliably quantitate,
on an LC system that is not fully optimized for low extra-column
dispersion. Such peaks, or those that are unresolved, require
the use of MS detection and an analysis of extracted ion
chromatograms (Figures 1C and 1D) to be deciphered.
Notable critical pairs exhibiting at least partial co-elution include
M5/A2G1 and FA2G2Sg1/FA2G2Ga2. Given the significance of
monitoring M5 and immunogenic glycans, like those containing
the noted N-glycolyl neuraminic acid (Sg) and alpha-linked
galactose monosaccharides, these separations were optimized
for increased resolution between species containing these types
of sugars.
[ 99 ]
Profiling Released High Mannose and Complex N-Glycan Structures from Monoclonal Antibodies
Using RapiFluor-MS Labeling and Optimized Hydrophilic Interaction Chromatography
FA2G2Sg1
FA2G2Ga2
A2G1
A
B
C
D
FLR
FLR
10x zoom
BPI
XIC
6.0
22.0 min
6.0
22.0 min
M5
10.0
22.0 min
1
2
3
4
5
6
7
8
9
10 11
12
13
1
2
3 4 5
6
7
8
9
10 11
12 13
13.5
14.5
21.7
22.7
Figure 1. (A) Fluorescence (FLR) chromatogram of RapiFluor-MS labeled N-glycans from Intact mAb Mass Check Standard obtained
using the universal N-glycan profiling method and a 2.1 x 150 mm ACQUITY UPLC Glycan BEH Amide 130Å 1.7 µm Column.
(B) Fluorescence chromatogram scaled with a 10x zoom to show low abundance N-glycans (C) Base peak intensity (BPI) chromatogram
of RapiFluor-MS labeled N-glycans obtained with the universal N-glycan profiling method (D) Extracted ion chromatograms (XICs)
of critical pairs to highlight issues with partial co-elution. N-glycan samples corresponding to 0.38 µg of the Intact mAb Mass Check
Standard were analyzed in each experiment.
Table 1. RapiFluor-MS labeled N-glycans from the Intact mAb Mass Check Standard, a murine monoclonal antibody.
Glycan
RapiFluor-MS labeled
glycan composition
Mi (Da)
2+
3+
1
A2
C67H105O37N9
1627.66
814.84
543.56
2
FA2
C73H115O41N9
1773.72
887.87
592.25
3
M5
C63H99O37N7
1545.61
773.81
516.10
4
FA1G1
C71H112O41N8
1732.69
867.35
578.57
A2G1
C73H115O42N9
1789.71
895.86
597.58
5
A2G1
C73H115O42N9
1789.71
895.86
597.58
6
FA2G1
C79H125O46N9
1935.77
968.89
646.26
7
FA2G1
C79H125O46N9
1935.77
968.89
646.26
8
FA2G2
C85H135O51N9
2097.82
1049.92
700.28
9
FA2G1Ga1
C85H135O51N9
2097.82
1049.92
700.28
10
FA2G2Ga1
C91H145O56N9
2259.88
1130.95
754.30
11
FA2G2Ga1
C91H145O56N9
2259.88
1130.95
754.30
12
FA2G2Sg1
C96H152O60N10
2404.92
1203.46
802.65
13
FA2G2Ga2
C97H155O61N9
2421.93
1211.97
808.32
14
FA2G2Ga1Sg1
C102H162O65N10
2566.97
1284.49
856.66
[ 100 ]
Profiling Released High Mannose and Complex N-Glycan Structures from Monoclonal Antibodies
Using RapiFluor-MS Labeling and Optimized Hydrophilic Interaction Chromatography
M5
FA2G2Sg1
FA2G2Ga2
A2G1
A
B
C
D
Universal N-glycan
profiling method
mAb N-glycan
profiling method
FLR
1
2
3
4
5
6
7
8
9
10 11
12 13
14
XIC
Rs HH
1.61
Rs HH
1.13
Developing a Higher Resolution mAb N-Glycan Analysis Method
The significant co-elution of the members of these critical glycan pairs necessitated the development of an LC method specifically tailored
for N-glycans released from mAbs. Modifying a HILIC separation by reducing the slope of the gradient can produce more resolution between
labeled N-linked glycans with similar partition coefficients. As a result, the first change made to the method was to reduce the gradient
slope while retaining the overall run time. A new gradient running from 26%–37% mobile phase A (versus 25% to 46%) indeed showed
improvement in peak resolution. In another step, the retention time of the glycans was reduced by increasing the flow rate of the separation
from 0.4 to 0.5 mL/min. This flow rate adjustment shifted the mAb N-glycan profile to be well within the gradient window, while at the
same time, the maximum pressure of the analysis was maintained well below the pressure limit of the system. Increasing the flow rate of the
separation also yielded an improvement in the resolution of the critical pairs (Figures 2A and 2B), but caused an average of a 90% increase
in peak widths. This peak broadening was attributed to poor band formation when the injected sample approached the head of the column. To
improve band formation, the eluent strength at the onset of the separation was reduced. The gradient was changed to have two segments, an
initial ramp from 20% to 27% mobile phase A over 3.2 minutes followed by 27% to 37% mobile phase A over 31.8 minutes. Indeed, these
changes facilitated better band formation at the head of the column and obtaining correspondingly sharper glycan peaks (Figure 2C). Despite
improving the method in multiple ways, the two noted critical pairs of RapiFluor-MS labeled N-glycans remained only partially resolved.
One final adjustment to the running conditions proved highly effective in improving the resolution of these critical pairs.
Lowering the column temperature from 60 °C to 45 °C increased their separation to the point that near baseline resolution could be
achieved (Figure 2D). With these final conditions, half-height resolution of the M5/A2G1+FA1G1 and the FA2G2Sg1/FA2G2Ga2 peaks
were found to be 1.61 and 1.13, respectively. The extent of separation has been improved for all major species throughout the mAb
profile, except for the alpha-linked galactose isomers of the FA2G2Ga1 glycan eluting at approximately 25 minutes.
Figure 2. (A) Fluorescence (FLR) chromatogram and extracted ion chromatograms (XICs) obtained with the universal N-glycan profiling method and
a 2.1 x 150 mm ACQUITY UPLC Glycan BEH Amide 130Å 1.7 µm Column. (B) FLR chromatogram and XICs obtained with a 0.5 mL/min flow rate,
60 °C column temperature, and a 35 minute gradient from 26% to 37% H2O (C) FLR chromatogram and XICs obtained with a 0.5 mL/min flow rate,
60 °C column temperature, and a two-step gradient of 20% to 27% H2O in 3.2 min followed by 27% to 37% H2O in 31.8 min. (D) FLR
chromatogram and XICs obtained as listed in (C) except with a column temperature of 45 ˚C (the mAb N-glycan profiling method). N-glycan
samples corresponding to 0.38 µg of the Intact mAb Mass Check Standard were analyzed in each experiment. RsHH denotes peak resolution
measured at half-height (HH).
[ 101 ]
Profiling Released High Mannose and Complex N-Glycan Structures from Monoclonal Antibodies
Using RapiFluor-MS Labeling and Optimized Hydrophilic Interaction Chromatography
Using the RapiFluor-MS High Mannose Standard to Demonstrate System Suitability
The resolution gains afforded by the new mAb N-glycan profiling method allows for better monitoring
of high mannose structures. To this end, we have employed a new proficiency standard, the RapiFluor-MS
High Mannose Standard, to demonstrate its ability to precisely monitor high mannose structures. This new
standard contains RapiFluor-MS-labeled M5, M6, M7, M8, and M9. When separated on its own, it is evident
that the RapiFluor-MS High Mannose Standard is a relatively simple mixture (Figure 3).
B
BPI
A
M6*
M7*
M5
M 6
M7
M8
M9
FLR
6.0
34.0
6.0
34.0 min
Mannose-5
(M5)
Mannose-6
(M6)
Mannose-7
(M7)
Mannose-8
(M8)
Mannose-9
(M9)
M5 2+
M6 2+
M7 2+
M8 2+
M9 2+
Mi (Da)
tR
(min)
Relative
Abun (%)
M5* 1545.6080 14.51
1.26
M5 1545.6080 14.82
40.97
M6* 1707.6609 18.21
1.97
M6 1707.6609 18.67
19.89
M7* 1869.7137 21.99
2.14
M7 1869.7137 22.74
5.93
M8* 2031.7665 26.12
1.62
M8 2031.7665 26.63
18.89
M9 2193.8193 29.66
7.33
M5*
M8*
C
D
Figure 3. (A) Fluorescence (FLR) and base peak intensity (BPI) chromatograms (B) of the Waters RapiFluor-MS High Mannose Standard as obtained using a 2.1 x 150 mm
ACQUITY UPLC Glycan BEH Amide 130Å 1.7 µm Column and the mAb N-glycan profiling method. Asterisks denote isomers of M5, M6, and M7. (C) ESI mass spectra
of each major component. (D) Representative chromatographic data. Approximately 5 pmoles of high mannose N-glycans were analyzed in these experiments.
An asterisk (*) denotes a linkage isomer.
[ 102 ]
Profiling Released High Mannose and Complex N-Glycan Structures from Monoclonal Antibodies
Using RapiFluor-MS Labeling and Optimized Hydrophilic Interaction Chromatography
As a result, it can potentially be used to support the identification of high mannose species when MS detection
may not be available. Moreover, this high mannose glycan mixture is well suited for use in spiking studies,
which can be performed to establish system suitability. In this work, RapiFluor-MS labeled N-glycans from
Intact mAb Mass Check Standard were spiked with varying concentrations of the RapiFluor-MS High Mannose
Standard (as outlined in the experimental section). Four RapiFluor-MS labeled glycan samples were prepared
with M5 relative abundances ranging from 0.2% to 2.0% and analyzed as illustrated in Figure 4A. In these
samples, M5, M6, and M8 are readily detected, while M7 and M9 are not due to their lower relative abundances
in the spiking standard. The high resolution of the method allows for better integration of the high mannose
glycan species. This can be clearly demonstrated by plotting the fluorescence peak areas of M5, M6, and M8
as functions of the spiking level. The linearity of these data (R2≥0.974) underscores the suitability of this
technique for monitoring high mannose structures (Figure 4B). These spiking experiments also demonstrate
that the Intact mAb Mass Check Standard is effectively free of high mannose species and that the M5 previously
monitored during the development of the separation is near the limit of quantitation of this method.
= 0.980
0.00% 1.00% 2.00% 3.00% 4.00% 5.00% 6.00%
Spiking Level
Minutes
6.00
8.00
10.00
12.00
14.00
16.00
18.00
20.00
22.00
24.00
26.00
28.00
30.00
32.00
34.00
M5
M6
M8
M5
M6
M8
A
High Mannose
Spiking Level
5x
B
FLR
= 0.974
0
0.06
Spiking Level
R2
R2
R2
= 0.994
0
0.06
Spiking Level
High Mannose
Spiking Level
3x
High Mannose
Spiking Level
2x
High Mannose
Spiking Level
1x
1
2
3
5x
Rel. Abun M5: 1.38%
1
2
3
5x
1
2
3
5x
Rel. Abun M5: 0.47%
Figure 4. (A) Fluorescence (FLR) chromatogram obtained for RapiFluor-MS N-glycans prepared from Intact mAb Mass Check Standard spiked with varying
concentrations of the RapiFluor-MS High Mannose Standard. Top to bottom: spiking levels are 5x, 3x, 2x, and 1x. (B) Peak area of high mannose structures at different
spike concentrations. N-glycan samples corresponding to 0.34 µg of the Intact mAb Mass Check Standard were analyzed in each experiment. Separations were
performed using the mAb N-glycan profiling method and an ACQUITY UPLC Glycan BEH Amide 130Å 1.7 µm Column.
[ 103 ]
Profiling Released High Mannose and Complex N-Glycan Structures from Monoclonal Antibodies
Using RapiFluor-MS Labeling and Optimized Hydrophilic Interaction Chromatography
GU Values from the Universal N-Glycan Profiling Method Versus the mAb N-Glycan Profiling Method
As mentioned earlier, the universal N-glycan profiling method is a generic tool for all N-glycan sample types.
It will also be the method recommended for workflows involving assignment of new glycan peaks based on
matching GU values to data in an upcoming RapiFluor-MS GU database, which is currently being constructed
in collaboration with the National Institute for Bioprocessing Research and Training (NIBRT).
With this in mind, it is important to recognize that GU values, regardless of labeling strategy, are method
specific. So although the mAb N-glycan profiling method can be used with GU values, it will not generate
GU values that are meaningful for searching a NIBRT database based on the universal N-glycan profiling
method. GU values still have merit for glycan analyses, even if they are not used for database matching. Use
of GU values minimizes subtle retention time variations between runs and between different instruments by
expressing chromatographic retention in terms of standardized GU values.3 To assign GU values, a dextran
ladder, consisting of glucose multimers of increasing length, is used as an external calibrant. The retention
times of the glucose multimers are then used via cubic spline fitting to convert glycan retention times into GU
values. Chromatographic data collected from separations of the RapiFluor-MS Dextran Calibration Ladder are
provided in Table 2, one set of data obtained with the universal N-glycan profiling method and the other with
the new, mAb N-glycan profiling method. Not surprisingly, differences between the methods led to shifts
in the retention times of the individual glucose multimers. Therefore, GU values derived for the mAb N-glycans
are also shifted, as shown in (Table 2). For the most strongly retained species, the FA2G2Ga1Sg1 glycan,
there is, in fact, a GU shift of +0.37. Clearly, it is important to give consideration to how GU values are
generated and how they are to be used. The universal N-glycan profiling method is the appropriate method
for GU database searching. Nevertheless, GU values can be used along with the mAb N-glycan profiling
method as replacements to standard retention times to improve the robustness of data reporting.
Universal N-Glycan
Profiling
mAb N-Glycan
Profiling
Component
Name
Glucose Units
Glucose Units
1
A2
5.49
5.54
2
FA2
5.82
5.91
3
M5
6.19
6.24
4
FA1G1 + A2G1
6.23
6.37
5
A2G1
6.38
6.49
6
FA2G1
6.69
6.72
7
FA2G1
6.85
6.86
8
FA2G2
7.43
7.69
9
FA2G1Ga1
7.55
7.81
10
FA2G2Ga1
8.25
8.57
11
FA2G2Ga1
8.30
8.60
12
FA2G2Sg1
9.06
9.39
13
FA2G2Ga2
9.11
9.49
14
FA2G2Ga1Sg1
9.88
10.25
Table 2. Glucose unit values for the RapiFluor-MS labeled N-glycans from the Intact mAb Mass Check Standard and the RapiFluor-
MS High Mannose Standard, as obtained with the universal N-glycan profiling method versus the mAb N-glycan profiling method.
Glucose unit (GU) values were assigned using cubic spline fitting and UNIFI 1.7 Software.
[ 104 ]
Profiling Released High Mannose and Complex N-Glycan Structures from Monoclonal Antibodies
Using RapiFluor-MS Labeling and Optimized Hydrophilic Interaction Chromatography
Waters Corporation
34 Maple Street
Milford, MA 01757 U.S.A.
T: 1 508 478 2000
F: 1 508 872 1990
www.waters.com
Waters, The Science of What’s Possible, Xevo, UNIFI, MassLynx, UPLC, and ACQUITY UPLC, are registered trademarks
of Waters Corporation. LockSpray, RapiFluor-MS and GlycoWorks, are trademarks of Waters Corporation. All other
trademarks are the property of their respective owners.
©2015 Waters Corporation. Produced in the U.S.A. October 2015 720005516EN AG-PDF
C O N C LU S IO NS
The N-linked glycosylation of mAbs can impact their circulation half-life and
efficacy. Therefore, it is particularly important for the N-glycans of a mAb to be
well characterized and routinely monitored. By labeling mAb N-glycans with
RapiFluor-MS, high sensitivity detection by both fluorescence and MS is made
possible. The sample loading condition, gradient steepness, flow rate, and
separation temperature of the universal N-glycan profiling method were adjusted
to create a mAb N-glycan profiling method that was able to better resolve the
Man5/A2G1+FA1G1 and FA2G2Sg1/FA2G2Ga2 critical pairs. The mAb N-glycan
profiling method yielded a half-height resolution of 1.61 for M5/A2G1+ N-glycans
and FA2G2Sg1/FA2G2Ga2 of 1.13. By improving the resolution of these critical
pairs of N-glycans, we have provided additional separation space for monitoring
high mannose structures. To this end, the RapiFluor-MS High Mannose Standard
was used in a series of spiking experiments to demonstrate the quantitative
performance of this new gradient for analyzing high mannose N-glycan
structures. Using the mAb N-glycan profiling method in conjunction with
the new RapiFluor-MS High Mannose Standard and the RapiFluor-MS Dextran
Ladder allows for the easier adoption of this system solution.
References
1. Rudd, P. M.; Guile, G. R.; Kuster, B.; Harvey, D. J.;
Opdenakker, G.; Dwek, R. A., Oligosaccharide sequencing
technology. Nature 1997, 388 (6638), 205–7.
2. Lauber, M. A.; Yu, Y. Q.; Brousmiche, D. W.; Hua, Z.; Koza,
S. M.; Magnelli, P.; Guthrie, E.; Taron, C. H.; Fountain,
K. J., Rapid Preparation of Released N-Glycans for HILIC
Analysis Using a Labeling Reagent that Facilitates
Sensitive Fluorescence and ESI-MS Detection.
Anal Chem 2015, 87 (10), 5401–9.
3. Campbell, M. P.; Royle, L.; Radcliffe, C. M.; Dwek, R. A.;
Rudd, P. M., GlycoBase and autoGU: tools for HPLC-based
glycan analysis. Bioinformatics 2008, 24 (9), 1214–6.
[ 105 ]
Profiling Released High Mannose and Complex N-Glycan Structures from Monoclonal Antibodies
Using RapiFluor-MS Labeling and Optimized Hydrophilic Interaction Chromatography
Matthew A. Lauber and Stephan M. Koza
G OA L
To demonstrate the enhanced resolving power
of the ACQUITY UPLC® Glycoprotein BEH
amide, 300Å, Column for separations of high
molecular weight, RapiFluor-MS™ labeled
N-glycans.
BAC KG RO U N D
Protein glycosylation is frequently profiled
by removing glycans from their counterpart
glycoprotein and imparting them with a
detectable chemical moiety, such as the
fluorescence and MS-active RapiFluor-MS
label.1 High resolution separations of these
released and labeled N-glycans can be
obtained by UPLC® hydrophilic interaction
chromatography (HILIC) with purposefully
designed glycan BEH amide, 130Å columns.2
Interestingly, glycosylation of proteins can
be extremely diverse. While monoclonal
antibodies tend to be modified with relatively
low molecular weight (1 to 3 kDa) biantennary
structures, numerous biotherapeutic proteins
are expressed with comparatively high
molecular weight (3 to 6 kDa) tri- and tetra-
antennary structures. Such large and highly
branched glycan structures exhibit large radii
of hydration.
Wide-pore glycoprotein BEH amide, 300Å, 1.7 µm
columns for enhancing the resolution of tri- and
tetra-antennary,
RapiFluor-MS labeled N-glycans.
Enhancing the Peak Capacity of High
Molecular Weight N-Glycan HILIC
Separations with a Wide-Pore Amide
Bonded Stationary Phase
Consequently, the application of chromatography columns containing particles
with standard average pore diameters (80 to 150Å) can limit the resolution
with which these species can be separated. It is therefore advantageous to
employ a stationary phase with a wide average pore diameter, wherein large
structures will have access to the majority of the porous network and the surface
area of the stationary phase. In addition, the large labeled glycan structures
are less likely to experience restricted diffusion while migrating through the
pores of a wide-pore material.3-4 In this technology brief, we demonstrate the
utility of an amide bonded stationary phase with an average pore diameter of
300Å (glycoprotein BEH amide, 300Å, 1.7 μm) and its ability to enhance the
resolution of RapiFluor-MS labeled tri and tetra-antennary N-glycans derived
from recombinant human Factor IX.
[ 106 ]
Figure 1. HILIC fluorescence chromatograms for RapiFluor-MS labeled N-glycans from pooled human IgG (orange) and recombinant Factor IX (blue). (A) Chromatograms
obtained for glycans from 0.4 µg protein using an ACQUITY UPLC Glycan BEH Amide, 130Å, 1.7 µm, 2.1 x 150 mm Column. (B) Chromatograms obtained for
glycans from 0.4 µg protein using an ACQUITY UPLC Glycoprotein BEH Amide, 300Å, 1.7 µm, 2.1 x 150 mm Column. Separations were performed according to the
conditions in the GlycoWorks™ RapiFluor-MS Care and Use Manual (p/n 715004793). Peak capacities were calculated from half-height widths and retention windows
derived from the labeled peaks.
5
10
15
20
25
30
35 min
5
10
15
20
25
30
35 min
Retention Time
1
2
3
4
5
6
7
8
Glycoprotein BEH Amide, 300Å, 1.7 µm
Glycan BEH Amide, 130Å, 1.7 µm
1
2
3
4
5
6
7
8
Pc*= 104
ΔRT1,4 = 12.74 min
Wh, avg= 7.43 sec
Pc*= 53
ΔRT5,8 = 6.45 min
Wh, avg= 7.47 sec
Pc*= 101
ΔRT1,4 = 11.54 min
Wh, avg= 6.89 sec
Pc*= 62
ΔRT5,8 = 6.70 min
Wh, avg= 6.60 sec
17% Increase
in Peak Capacity
A
B
0E+0
3E+6
0E+0
3E+6
Fucose
GlcNAc
Manoses
Galactose
NeuAc
5. FA3G3S3
6. FA4G4S4
7. FA4G4Lac1S4
8. FA4G4Lac2S4
1. A2
2. FA2G1
3. FA2G2S1
4. FA2BG2S2
Figure 2. HILIC fluorescence chromatograms for RapiFluor-MS labeled N-glycans from recombinant Factor IX as obtained using
2.1 x 150 mm columns with various amide bonded stationary phases. Peak capacities were calculated from half-height widths
and retention windows derived from the labeled peaks.
24
25
26
27
28
29
30
31
32
33
34 min
5
6
7
8
Glycan BEH Amide, 130Å, 1.7 µm
5
6
7
8
FA4G4S2
3E+5
0E+0
22
23
24
25
26
27
28
29
30
31 min
Retention Time (min)
Glycoprotein BEH Amide, 300Å, 1.7 µm
FA4G4S3
FA4G4Lac1S3
3E+5
0E+0
Fucose
GlcNAc
Manoses
Galactose
NeuAc
5. FA3G3S3
6. FA4G4S4
7. FA4G4Lac1S4
8. FA4G4Lac2S4
[ 107 ]
Enhancing the Peak Capacity of High Molecular Weight N-Glycan HILIC Separations with a Wide-Pore Amide Bonded Stationary Phase
Waters Corporation
34 Maple Street
Milford, MA 01757 U.S.A.
T: 1 508 478 2000
F: 1 508 872 1990
www.waters.com
T H E SO LU T ION
N-glycans were prepared from both pooled human IgG
as well as recombinant human Factor IX, a glycoprotein
known to be modified with large, highly sialylated
N-glycans.5 Specifically, these samples were prepared
using rapid deglycosylation, RapiFluor-MS labeling,
and μElution HILIC SPE, as described in the GlycoWorks
RapiFluor-MS N-Glycan Kit Care and Use Manual.1, 6
Glycan mapping of the resulting RapiFluor-MS
labeled N-glycans was first performed with a glycan
BEH amide, 130Å, 1.7 μm column, given that it is
intended for general purpose glycan analyses and for
use with GlycoBase database searching.7 Figure 1A
displays the fluorescence chromatograms obtained
when HILIC-based chromatography is performed on
a sample containing less hydrophilic IgG N-glycans
(orange) versus a sample containing later eluting,
more hydrophilic N-glycans from Factor IX (blue).
In addition, Figure 1A displays the effective peak
capacities for the retention windows of the two sample
types. These results can be compared to Figure
1B, which presents the chromatograms and peak
capacities obtained with a glycoprotein BEH amide
column containing the 300Å, wide-pore stationary
phase. While the effective peak capacities for the IgG
N-glycans are comparable, a marked improvement in
peak capacity of approximately 17% is apparent in the
separations of the Factor IX N-glycans when the wide-
pore glycoprotein BEH amide 300Å, 1.7 μm column is
used. The resolving power of the wide-pore, BEH amide
column for the large N-glycans is noteworthy, in that
it facilitates resolving several low abundance species.
Figure 2 highlights some of the impacted regions of the
chromatogram where there are improvements in the
resolution of FA4G4S2, FA4G4S3, and FA4G4Lac1S3
N-glycans. In summary, when smaller, biantennary
N-glycans are to be separated, a glycan BEH amide,
130Å, 1.7 μm stationary phase is an ideal choice
due to its high surface area and high retentivity. For the characterization of large,
tri- and tetra-antennary N-glycans, it is, however, advantageous to use the wide-pore
amide stationary phase. Moreover, the wide-pore amide column is intended specifically
for large biomolecule separations: ACQUITY UPLC Glycoprotein BEH Amide, 300Å,
1.7 μm stationary phase is ensured to have consistent batch-to-batch performance
through stringent quality control testing involving a separation of ribonuclease B
(RNase B) glycoforms at the intact protein level (see reference 8 for an example of
this chromatography).8
SUMMA RY
High molecular weight, tri- and tetra-antennary N-glycans are highly branched
structures that adopt relatively large radii of hydration in solution. To achieve
optimal HILIC separations of these large structures, we propose a column with
a wide-pore amide bonded stationary phase, a glycoprotein BEH amide, 300Å,
1.7 μm column. For large glycan species, this column provides increases in peak
capacity over a conventional pore diameter column of approximately 17%. Improved
resolving power is particularly useful in this separation space as it is typified by
highly complex glycan profiles. Most notably, these improvements in resolution
should be of significant utility in the characterization and routine monitoring of
biopharmaceuticals that are expressed with large, highly complicated N-glycan
structures, such as coagulation Factor IX, erythropoietin, and darbepoetin.
References
1. Lauber, M. A.; Brousmiche, D. W.; Hua, Z.; Koza, S. M.; Guthrie, E.; Magnelli, P.; Taron, C. H.;
Fountain, K. J., Rapid Preparation of Released N-Glycans for HILIC Analysis Using a Novel
Fluorescence and MS-Active Labeling Reagent. Waters Application Note 720005275EN 2015.
2. Ahn, J.; Yu, Y. Q.; Gilar, M., UPLC-FLR Method Development of 2-AB Labeled Glycan Separation in
Hydrophilic Interaction Chromatography (HILIC). Waters Application Note 720003238EN 2010.
3. Gustavsson, P.-E.; Larsson, P.-O., Support Materials for Affinity Chromatography. In Handbook
of Affinity Chromatography, Hage, D., Ed. Taylor & Francis: Boca Raton, FL, 2006; pp 15–33.
4. Renkin, E. M., J. Gen. Physio. 1954, (38), 225.
5. Yu, Y. Q., Analysis of N-Linked Glycans from Coagulation Factor IX, Recombinant and Plasma
Derived, Using HILIC UPLC/FLR/QTof MS. Waters Application Note 720004019EN 2011.
6. GlycoWorks RapiFluor-MS N-Glycan Kit Care and Use Manual 715004793EN.
7. Campbell, M. P.; Royle, L.; Radcliffe, C. M.; Dwek, R. A.; Rudd, P. M., GlycoBase and autoGU: tools
for HPLC-based glycan analysis. Bioinformatics 2008, 24 (9), 1214–6.
8. Lauber, M. A.; Koza, S. M., Developing High Resolution HILIC Separations of Intact Glycosylated
Proteins using a Wide-Pore Amide-Bonded Stationary Phase Waters Application Note
720005380EN 2015.
Waters, The Science of What’s Possible, ACQUITY UPLC, and UPLC are registered trademarks of Waters Corporation. GlycoWorks
and RapiFluor-MS are trademarks of Waters Corporation. All other trademarks are the property of their respective owners.
©2015 Waters Corporation. Produced in the U.S.A. April 2015 720005381EN KP-PDF
[ 108 ] Enhancing the Peak Capacity of High Molecular Weight N-Glycan HILIC Separations with a Wide-Pore Amide Bonded Stationary Phase
WAT E R S S O LU T IO NS
GlycoWorks™ Rapi Fluor-MS™ N-Glycan Kit
ACQUITY® QDa® Mass Detector
ACQUITY UPLC® H-Class Bio System
ACQUITY UPLC Autosampler with FTN
ACQUITY UPLC Fluorescence Detector (FLR)
ACQUITY UPLC Glycan BEH Amide Column
Empower® 3 FR2 CDS
K E Y W O R D S
Glycan, mass detection, H-Class,
ACQUITY, QDa, Rapi Fluor-MS
A P P L I C AT IO N B E N E F I T S
■
■
Reduced sample preparation times
for released N-glycan analyses
■
■
Increased confidence in glycan monitoring
by obtaining fluorescence and mass
detection for every peak
I N T RO DU C T IO N
Glycosylation is one of the most complex post-translational modifications
of protein-based biotherapeutics. The efficacy of glycosylated therapeutics
is directly related to the glycoprofile. The presence of undesired structures can
lead to changes in PK/PD profiles, either positively or negatively, and have been
associated with immunogenic responses. For these reasons glycosylation is often
designated as a critical quality attribute (CQA). During the development process,
the glycoprofile of candidate molecules is extensively studied and characterized.
Characteristic profiles are then monitored through process development,
commercialization, and post-approval studies to maintain product efficacy
and safety.
In this application note, we present a streamlined approach to label
released N-glycans with RapiFluor-MS and analyze the labeled N-glycans
with the ACQUITY UPLC H-Class Bio System with fluorescent (FLR) and
ACQUITY QDa Mass Detectors. This new monitoring workflow allows researchers
to prepare samples from glycoprotein to UPLC-FLR/MS analysis in 30 minutes.
In addition to reduced sample preparation times, RapiFluor-MS yields
14 times greater fluorescence response and 160 times greater MS response
when compared to 2-AB. These improvements enable the use of FLR and mass
detection with the ACQUITY QDa for routine analysis. In this application note
we present the utility of RapiFluor-MS coupled with UPLC®-FLR-MS for
monitoring labeled glycans ranging across a range of properties, masses,
and relative abundance.
Exploiting
RapiFluor-MS Labeling to Monitor Diverse N-Glycan Structures
via Fluorescence and Mass Detection
Eoin F.J. Cosgrave, Robert Birdsall, and Sean M. McCarthy
Waters Corporation, Milford, MA, USA
[ 109 ]
E X P E R IM E N TA L
LC conditions
LC system:
ACQUITY UPLC H-Class Bio
Detectors:
ACQUITY UPLC FLR and
ACQUITY QDa Mass Detector
Column:
ACQUITY UPLC Glycan BEH Amide,
130Å, 1.7 µm, 2.1 x 150 mm
(p/n 186004742)
Column temp.:
60 °C
Sample temp.:
10 °C
Injection volume:
2 µL
Data management:
Empower 3 FR2 CDS
FLR settings
Data rate:
5 points/sec.
Excitation wavelength: 265 nm
Emission wavelength: 425 nm
QDa settings
Sample rate:
5 points/sec
Mass range:
500–1250 Da
Cone voltage:
15 V
Capillary voltage:
1.5 kV
Probe temp.:
500 °C
Ionization mode:
ESI+
Mobile phase A:
Acetonitrile (Pierce, LC-MS Grade)
Mobile phase B:
50 mM ammonium formate, pH 4.4,
(LC-MS Grade, Waters ammonium
formate concentrate)
Mobile phase C:
Acetonitrile (LC-MS grade)
Mobile phase D:
Acetonitrile (LC-MS grade)
Time
Flow rate
% A
% B
%C
% D
(mL/min)
Initial 0.400
75 25 0
0
35.0 0.400
54 46 0 0
36.5 0.200 0 100 0 0
39.5 0.200 0 100 0 0
42.5 0.200
75 25 0 0
47.4 0.400
75 25 0 0
55.0 0.400
75 25 0 0
A sample of murine IgG1 mAb N-Glycans was prepared from
Waters Intact mAb Mass Check Standard (p/n 186006552),
which is included in the GlycoWorks RapiFluor-MS N-Glycan Kit
(p/n 176003606). N-Glycans were also prepared from RNase B
and bovine fetuin (Sigma Aldrich). Released and labeled N-glycan
pools were generated using the GlycoWorks RapiFluor-MS N-Glycan
Kit following the protocol provided in the Care and Use Manual
(715004793). Following release and labeling, samples were
dried using a CentriVap™ and reconstituted in 25 µl of a mixture
of ACN/Water/DMF at a ratio of 22.5%:55.5%:22%, respectively.
In each case the targeted mass load was 30 pmoles of released
glycan. The ammonium formate mobile phase was prepared using
Waters ammonium formate concentrate (p/n 186007081).
R E SU LT S A N D D IS C U S S IO N
N-glycosylation is a non-template driven process that generates
a vast array of glycan structures that vary in size, charge,
and extent of branching depending on the protein and expression
system. To evaluate the capacity of the ACQUITY QDa to detect
glycans both within and beyond its mass range, three glycoproteins
(human IgG, RNAse B, and bovine fetuin) were selected to provide
typically observed glycans ranging from neutral bi-antennary
structures to tetra-sialylated structures. N-glycans from each
protein were released using Rapid PNGase F and labeled with
RapiFluor-MS following the provided sample preparation protocol.
Labeled glycans were separated via UPLC-HILIC and detected
using both an ACQUITY FLR and ACQUITY QDa.
As is evident in Figure 1, each glycan structure is
chromatographically resolved using a single gradient method.
In addition, each glycan structure observed in fluorescence
(top panel) is also observed by the ACQUITY QDa Mass Detector
(bottom panel), indicating the ability of the ACQUITY QDa to
detect glycans across a range of possible structures and attributes
when labeled with RapiFluor-MS. For traditional labeling
technologies this is not possible due to poor ionization efficiency.
[ 110 ] Exploiting RapiFluor-MS Labeling to Monitor Diverse N-Glycan Structures via Fluorescence and Mass Detection
EU
0
20
40
60
Minutes
5
10
15
20
25
30
35
40
Intensity
1x10
6
2x10
6
3x10
6
4x10
6
5x10
6
6x10
6
Minutes
5
10
15
20
25
30
35
40
ACQUITY FLR
ex 260 nm
em 425 nm
ACQUITY QDa
TIC
300 to 1250 m/z
FA2
M6
M7
M8
M9
M5
FA2G1
FA2G2
A3G3S3a
A3G3S3b
A3G3S3c
A3G3S4
A2
A2G2S2
Figure 1. The ACQUITY QDa can detect an array of RapiFluor-MS labeled N-glycans. Glycans from
human IgG (red trace ), RNAse B (black trace), and bovine fetuin (blue trace) were released with
Rapid PNGase F, labeled with RapiFluor-MS reagent. Individual glycan pools were then separated
via HILIC and detected with both fluorescence (A) and mass detection (B).
Figure 2. Spectra of selected RapiFluor-MS labeled glycans detected with an ACQUITY QDa Mass
Detector. Glycans from murine IgG1 mAb, RNase B, and bovine fetuin were released with Rapid
PNGase F and labeled with RapiFluor-MS. Shown are representative spectra for selected glycan
structures separated in Figure 1.
While it is useful that glycan structures can be
observed by mass detection, it is important to
understand the quality of the resulting spectra
and the charge states of the glycan ions obtained
within. To understand this aspect, we integrated
peaks spanning a range of glycan properties and
measured the relative abundances of species in
each sample using FLR integrated data. The spectra
shown in Figure 2 demonstrate the ability of the
ACQUITY QDa to generate high quality spectra for
glycan structures across a wide range of properties
and masses. The data also demonstrate that both
high and low abundance glycan structures can
be readily detected. Our data indicates that high
quality spectra are generated for structures
present in the fluorescence profiles at abundances
as low as 0.5% highlighting the sensitivity of
ACQUITY QDa mass detection combined with
the improved ionization efficiency afforded by
RapiFluor-MS. Our data also demonstrate how the
improved charging of glycan structures by the
use of RapiFluor-MS allows small structures
such as A2, as well as very large structures,
such as the tetrasialylated A3G3S4, to be
detected with the QDa.
Intensity
0
12000
24000
36000
48000
m/z
500
750
1000
1250
Intensit
y
0
2.50x10 5
5.00x10 5
7.50x10 5
1.00x10 6
m/z
500
750
1000
1250
m/z
500
750
1000
1250
Intensity
0
25000
50000
75000
100000
Intensity
0
4.0x10
5
8.0x10
5
1.2x10
6
1.6x10
6
m/z
500
750
1000
1250
Intensit
y
0
150000
300000
450000
600000
m/z
500
750
1000
1250
Intensity
0
200000
400000
600000
800000
m/z
500
750
1000
1250
Intensity
0
30000
60000
90000
120000
m/z
500
750
1000
1250
Intensity
0
25000
50000
75000
100000
125000
m/z
500
750
1000
1250
Intensity
0
90000
180000
270000
360000
m/z
500
750
1000
1250
815.0
774.0
845.8
969.2
855.0
1064.8
1050.2
936.0
1161.8
IgG
RNase B
Bovine Fetuin
A2
[M+2H]2+
A2G2
[M+2H]2+
F(6)A2G2
[M+2H]2+
M5
[M+2H]2+
M6
[M+2H]2+
M7
[M+2H]2+
A2G2S2
[M+2H]2+
A3G3S3
[M+3H]3+
A3G3S4
[M+3H]3+
[ 111 ]
Exploiting RapiFluor-MS Labeling to Monitor Diverse N-Glycan Structures via Fluorescence and Mass Detection
Waters Corporation
34 Maple Street
Milford, MA 01757 U.S.A.
T: 1 508 478 2000
F: 1 508 872 1990
www.waters.com
Waters, The Science of What’s Possible, ACQUITY, QDa, ACQUITY UPLC, Empower, and UPLC are registered trademarks
of Waters Corporation. GlycoWorks and RapiFluor-MS are trademarks of Waters Corporation. All other trademarks are
the property of their respective owners.
©2015 Waters Corporation. Produced in the U.S.A. March 2015 720005353EN AG-PDF
C O N C LU S IO NS
Glycosylation of is a complex and critical aspect of most therapeutic proteins
which must be well characterized. Often, the profile of N-glycans is identified as
a critical quality attribute and as a result is monitored throughout the lifecycle
of products. As discussed in this application note, preparation of samples with
a GlycoWorks RapiFluor-MS N-Glycan Kit can dramatically reduce sample
preparation time and complexity. In addition, the use of RapiFluor-MS yields
improved FLR sensitivity and dramatically improved MS sensitivity. Through
improving glycan MS sensitivity RapiFluor-MS labeling permits the use of mass
detection with the ACQUITY QDa and thereby affords greater confidence in peak
monitoring across the range of structures encountered during biopharmaceutical
development. Taken together, RapiFluor-MS labeling and HILIC-FLR-MS with
ACQUITY UPLC H-Class Bio System and the ACQUITY QDa Mass Detector offer
an unparalleled solution for monitoring the N-glycan profiles of biotherapeutics.
[ 112 ] Exploiting RapiFluor-MS Labeling to Monitor Diverse N-Glycan Structures via Fluorescence and Mass Detection
WAT E R S S O LU T IO NS
Rapi Fluor-MS™ Glycan Performance
Test Standard
GlycoWorks™ Rapi Fluor-MS N-Glycan Kit
ACQUITY® QDa® Mass Detector
ACQUITY UPLC® H-Class Bio System
ACQUITY UPLC FTN
ACQUITY UPLC Fluorescence Detector
(FLR)
ACQUITY UPLC Glycan BEH
Amide Column
Empower® 3 CDS Software
K E Y W O R D S
Glycans, glycoforms, labeled glycans,
peak monitoring, mass detection,
fluorescence detection, IgG
A P P L I C AT IO N B E N E F I T S
■
■
Reduce sample preparation times
for released N-glycan analyses
■
■
Increase confidence in glycan
monitoring by routinely obtaining
mass information and fluorescence
for every peak
I N T RO DU C T IO N
During the development of biopharmaceuticals, it is important to characterize
and monitor glycoprofiles as they are often implicated as a product critical quality
attributes due to their impact on safety, efficacy, and potency among other factors.
It is well accepted that structural characterization of the glycoforms present is
necessary, and that mass spectrometry (MS) often plays a large role
in the identification of glycans.
Often, once the profile has been established, methods are transferred downstream
which incorporate fluorescence detection. In many cases, there is a desire to
obtain mass information for each detected peak even after characterization.
These data have been difficult to obtain for a number of reasons, including
a scarcity of mass spectrometers due to their cost and the requirement that
MS specialized analysts are needed to generate meaningful and useful data.
In this application note, we present the combined use of Rapi Fluor-MS
labeling reagent, ACQUITY UPLC H-Class Bio System, and serial
fluorescence/ACQUITY QDa Mass Detector for the monitoring of released
N-glycan profiles from IgGs. Overall, this new workflow allows scientists to
rapidly prepare samples, from glycoprotein to analysis in 30 minutes.
In addition, Rapi Fluor-MS labeling yields unprecendented MS response,1
which enables the use of the ACQUITY QDa for mass detection. We will discuss
the improved sensitivity and charge state profile afforded by Rapi Fluor-MS,
its general utility for fluorescence and mass detection, and the quality of
ACQUITY QDa mass spectra obtained for a range of IgG glycan structures.
New Capabilities for Monitoring Released N-Glycans through the Combined
Use of
RapiFluor-MS Labeling, ACQUITY UPLC H-Class Bio System,
and Serial Fluorescence/ACQUITY QDa Mass Detection
Eoin F.J. Cosgrave, Matthew A. Lauber, Robert Birdsall, and Sean M. McCarthy
Waters Corporation, Milford, MA, USA
[ 113 ]
E X P E R IM E N TA L
LC conditions
LC system:
ACQUITY UPLC H-Class Bio
Detectors:
ACQUITY UPLC FLR and
ACQUITY QDa Mass Detector
Column:
ACQUITY UPLC Glycan BEH Amide,
130Å, 1.7 µm, 2.1 x 150 mm
(p/n 186004742)
Column temp.:
60 °C
Sample temp.:
10 °C
Injection volume:
2 µL
FLR settings
Data rate:
5 points/sec
Excitation wavelength: 265 nm
Emission wavelength: 425 nm
QDa settings
Sample rate:
5 points/sec
Mass range:
500–1250 Da
Cone voltage:
15 V
Capillary voltage:
1.5 kV
Probe temp.:
500 °C
Ionization mode:
ESI+
Mobile phase A:
Acetonitrile (Pierce, LC/MS Grade)
Mobile phase B:
50 mM ammonium formate, pH 4.4,
(LC/MS grade, Waters Ammonium
Formate Concentrate)
Mobile phase C:
Acetonitrile (LC/MS grade)
Mobile phase D:
Acetonitrile (LC/MS grade)
Time
Flow rate
%A
%B
%C
%D
(mL/min)
Initial 0.400
75 25 0 0
35.0 0.400
54 46 0 0
36.5 0.200
0 100 0 0
39.5 0.200
0 100 0 0
42.5 0.200
75 25 0 0
47.4 0.400
75 25 0 0
55.0 0.400
75 25 0 0
SYNAPT® G2-S was used for assessment of Rapi Fluor-MS
versus 2-AB N-glycan charge states. See Reference 1 for
experimental details.
The Rapi Fluor-MS Glycan Performance Test Standard
(p/n 186007983) was reconstituted in 25 µL of a mixture of
DMF/acetonitrile/water at a ratio of 22.5%:55.5%:22%,
respectively and used directly. For each analysis the injection
volume was 2 µL, which corresponds to 32 pmol of released and
labeled N-glycan on column. LC/MS-grade acetonitrile and water
were purchased from Pierce. Ammonium formate was prepared
using Waters Ammonium Formate Solution-Glycan Analysis
(p/n 18600708) by pouring the entire contents of the solution
into 1 L of water and mixed. The UPLC® System used was
dedicated for applications which do not require non-volatile salts
to reduce the likelihood of adduct formation in the mass detector.
R E SU LT S A N D D IS C U S S IO N
Addition of mass detection to an existing analytical workflow
permits rapid and unambiguous identification of glycans.
Historically, this has been a difficult task due to the need for
high resolution instruments with appropriate sensitivity to
obtain meaningful mass data. To overcome this issue, the novel
labeling reagent, Rapi Fluor-MS, can been used. Rapi Fluor-MS
dramatically increases both the MS sensitivity and charging of
released N-glycans.
To demonstrate this, we compared the mass spectra of RapiFluor-MS
labeled glycans to those of glycans labeled with a more traditional
fluorescent label, 2-AB. This analysis was performed using time-of-
flight mass spectrometry, which characteristically has a very wide
mass range. The charge state characteristics of the different labeling
technologies could thereby be objectively observed.
As shown in Figure 1, signal intensity improves dramatically
when using Rapi Fluor-MS. Equally interesting is the shift in the
charge states of the detected glycan ions that results from use of
Rapi Fluor-MS labeling. As shown, Rapi Fluor-MS labeled FA2 near
exclusively adopts an [M+2H]2+ charge state, while more complex
structures begin to adopt even higher [M+3H]3+ charge states. In
each case, at least one highly populated charge state falls well
within the mass range of the ACQUITY QDa. Accordingly,
Rapi Fluor-MS makes it feasible to use the cost effective,
user-friendly ACQUITY QDa Mass Detector for N-glycan
monitoring experiments.
[ 114 ]
New Capabilities for Monitoring Released N-Glycans through the Combined Use of RapiFluor-MS Labeling,
ACQUITY UPLC H-Class Bio System, and Serial Fluorescence/ACQUITY QDa Mass Detection
FA2BG2S2
5.3e2
500
2000 m/z
500
2000 m/z
2.9e4
1.2e5
3.5e6
[M+2H]2+
FA2
2-AB
RapiFluor-MS
[M+2H]2+
[M+3H]3+
[M+2H]2+
[M+H]+
ACQUITY QDa
Mass Detection
Window
ACQUITY QDa
Mass Detection
Window
Figure 1. Charge States of RapiFluor-MS Labeled N-Glycans. Time-of-flight ESI+ mass spectra for two N-glycans labeled with
RapiFluor-MS and 2-AB, respectively. The detected, protonated charge states that are within the ACQUITY QDa acquisition window
are highlighted in green. The upper mass range of the ACQUITY QDa is indicated by the dashed line in each spectrum.
Figure 2. Fluorescence (top trace) and smoothed total ion chromatograms (bottom trace, 5 point
mean smooth) of IgG glycans. Each peak is labeled with component name and base peak mass
natively in Empower Data Management Software.
As discussed above, routine detection of
N-glycans with the ACQUITY QDa is made
possible by Rapi Fluor-MS labeling. Importantly,
ACQUITY QDa mass detection can be paired with
fluorescence detection to facilitate obtaining
optical-based quantification along with
corroborating data on peak homogeneity and mass
information. To enable this data to be collected
routinely, the design characteristics of the
ACQUITY QDa are such that users without extensive
mass spectrometry training are able to generate
meaningful mass data easily.
To demonstrate this ability, we separated a sample
of IgG released N-glycans labeled with RapiFluor-MS
and monitored the eluting glycans with both FLR
and ACQUITY QDa detectors. As shown in Figure 2,
high quality data were obtained for both detector
channels, with each species identified in the FLR also
represented with ACQUITY QDa MS data such that
peak assignments can be readily confirmed. Within
Empower Software, it is possible to annotate peaks
with component and mass information, which makes
reviewing data simple, as exemplified in Figure 2.
A2 - 815.1
FA
2 - 888.2
FA
2B - 989.6
A2G1a - 896.2
A2G1b - 896.0
FA
2G1a - 969.3
FA
2G1b - 969.2
FA
2BG1a - 1070.7
FA
2BG1b - 1070.6
A2G2 - 977.1
FA
2G2 - 1050.3
FA
2BG2 - 1151.8
FA
2G1S1 - 1114.6
FA
2G2S1 - 1196.0
FA
2BG2S1 - 865.3
FA
2G2S2 - 894.7
FA
2BG2S2 - 962.2
0.00
240.00
480.00
EU
0.0
1.6x10
6
3.2x10
6
Retention Time (min)
10
15
20
25
30
8.0x10
5
2.4x10
6
Intensity
[ 115 ]
New Capabilities for Monitoring Released N-Glycans through the Combined Use of RapiFluor-MS Labeling,
ACQUITY UPLC H-Class Bio System, and Serial Fluorescence/ACQUITY QDa Mass Detection
While the ability to detect N-glycan structures with the
ACQUITY QDa is impressive, spectral quality is paramount
for N-glycan monitoring, particularly when there is a need to
interrogate the data in detail. We therefore reviewed the quality
of MS data associated with peaks observed in the previously shown
chromatograms. Figure 3 illustrates the spectra for each assigned
peak in Figure 2. Notice that the ACQUITY QDa produced clean,
easily interpretable mass spectra for the Rapi Fluor-MS labeled
glycans, regardless of their relative abundance, molecular weight,
or sialic acid content. Clearly, the ACQUITY QDa together with
Rapi Fluor-MS can provide highly informative data that can be used
to increase the confidence of assignments made during routine
detection of N-glycans.
815.3
0
6000
12000
18000
24000
m/z
600
800 1000 1200
A2
[M+2H]
2+
0.80% RPA
896.0
0
15000
30000
45000
60000
m/z
600
800 1000 1200
A2[6]G1
[M+2H]2+
1.23% RPA
969.2
0
90000
180000
270000
360000
m/z
600
800 1000 1200
FA2[3]G1
[M+2H]2+
0.85% RPA
888.2
0
150000
300000
450000
600000
m/z
600
800 1000 1200
FA2
[M+2H]2+
20.51% RPA
896.0
0
6000
12000
18000
24000
m/z
600
800 1000 1200
A2[3]G1
[M+2H]2+
0.54% RPA
1070.7
0
50000
100000
150000
200000
m/z
600
800 1000 1200
FA2B[6]G1
[M+2H]2+
5.21% RPA
989.6
0
50000
100000
150000
200000
m/z
600
800 1000 1200
FA2B
[M+2H]2+
4.43% RPA
969.3
0
120000
240000
360000
480000
m/z
600
800 1000 1200
FA2[6]G1
[M+2H]2+
19.17% RPA
1070.6
0
6000
12000
18000
24000
m/z
600
800 1000 1200
FA2B[3]G1
[M+2H]2+
0.85% RPA
977.1
0
9000
18000
27000
36000
m/z
600
800 1000 1200
A2G2
[M+2H]2+
1.14% RPA
1151.7
0
6000
12000
18000
24000
m/z
600
800 1000 1200
FA2BG2
[M+2H]2+
2.20% RPA
1114.7
0
20000
40000
60000
80000
m/z
600
800 1000 1200
FA2G1S1
[M+2H]2+
2.86% RPA
1050.2
0
150000
300000
450000
600000
m/z
600
800 1000 1200
FA2G2
[M+2H]2+
14.56% RPA
865.3
0
20000
40000
60000
80000
m/z
600
800 1000 1200
FA2BG2S1
[M+3H]3+
2.34% RPA
894.6
0
15000
30000
45000
60000
m/z
600
800 1000 1200
FA2G2S2
[M+3H]3+
1.72% RPA
1196.0
0
50000
100000
150000
200000
m/z
600
800 1000 1200
FA2G2S1
[M+2H]2+
9.16% RPA
962.4
0
20000
40000
60000
80000
m/z
600
800 1000 1200
FA2BG2S2
[M+2H]2+
1.87% RPA
Figure 3. Combined spectra for each glycan structure identified in the chromatograms shown in Figure 2. Identified structures span from simple structures, such as
A2, to more complex structures present in IgG samples, such as FA2G2S2. RPA = Relative Peak Area based on FLR integration.
[ 116 ]
New Capabilities for Monitoring Released N-Glycans through the Combined Use of RapiFluor-MS Labeling,
ACQUITY UPLC H-Class Bio System, and Serial Fluorescence/ACQUITY QDa Mass Detection
Waters Corporation
34 Maple Street
Milford, MA 01757 U.S.A.
T: 1 508 478 2000
F: 1 508 872 1990
www.waters.com
Waters, The Science of What’s Possible, ACQUITY, QDa, ACQUITY UPLC, Empower, SYNAPT, and UPLC are registered trademarks
of Waters Corporation. GlycoWorks and RapiFluor-MS are trademarks of Waters Corporation. All other trademarks are the property
of their respective owners.
©2015 Waters Corporation. Produced in the U.S.A. March 2015 720005352EN AG-PDF
C O N C LU S IO NS
Glycosylation is a complex and critical aspect of most therapeutic proteins that
must be well characterized and monitored throughout product development and
commercialization. As discussed in this application note, RapiFluor-MS can be
used to dramatically reduce sample preparation times and complexity, to enhance
FLR sensitivity, and to dramatically improve MS sensitivity. By improving glycan
MS sensitivity, RapiFluor-MS labeling permits the use of mass detection with the
ACQUITY QDa and thereby affords greater confidence in peak monitoring across
the range of structures encountered during biopharmaceutical development.
Taken together, Rapi Fluor-MS labeling and HILIC-FLR-MS with the ACQUITY
UPLC H-Class Bio System and the ACQUITY QDa Mass Detector offer an
unparalleled solution for monitoring the N-glycan profiles of biotherapeutics.
References
1. Lauber, M. A.; Brousmiche, D. W.; Hua, Z.; Koza, S. M.;
Guthrie, E.; Magnelli, P.; Taron, C. H.; Fountain, K. J., Rapid
Preparation of Released N-Glycans for HILIC Analysis
Using a Novel Fluorescence and MS-Active Labeling
Reagent. Waters Application Note 720005275EN 2015.
[ 117 ]
New Capabilities for Monitoring Released N-Glycans through the Combined Use of RapiFluor-MS Labeling,
ACQUITY UPLC H-Class Bio System, and Serial Fluorescence/ACQUITY QDa Mass Detection
WAT E R S S O LU T IO NS
RapiFluor-MS™ Glycan Performance Test
Standard (p/n 186007983)
ACQUITY UPLC® H-Class Bio System (FTN)
ACQUITY UPLC Fluorescence Detector (FLR)
ACQUITY UPLC Glycan BEH Amide Columns
Empower®3 Chromatography Data Software
Waters® Fraction Manager – Analytical
K E Y W O R D S
Glycans, mass detection, H-Class,
ACQUITY, QDa, RapiFluor-MS, IgG
A P P L I C AT IO N B E N E F I T S
■
■
Rapid feedback on glycoprofiles during
production to ensure product quality
■
■
Reduced sample preparation times for
released N-glycans
■
■
Increased throughput for N-glycan analysis
■
■
Specificity for N-glycan species by
incorporating mass detection
I N T RO DU C T IO N
As glycosylated biotherapeutics move through the development pipeline, the
glycoprofile and N-glycan species present are characterized. In addition, as new
protein therapeutics progress through development, manufacturing conditions
are carefully studied and evaluated during scale-up to ensure consistent safety
and efficacy in preparation for clinical studies, and eventual commercialization.
As part of this process, the critical quality attributes are often monitored closely
to ensure production batches remain within defined acceptance criteria, and to
identify those parameters that are critical, often as part of a quality-by-design
(QbD) approach. In particular, the N-glycan profile is often monitored closely due
to the importance of glycans on the safety and efficacy of protein biotherapeutics.
Monitoring of released N-glycan profiles has historically been burdened with
labor intensive sample preparation, which often takes several hours to days.
This makes monitoring of the impact of manufacturing conditions on N-glycan
profiles challenging. In addition, the analysis of released and labeled N-glycans
frequently requires long analysis times. When monitoring of specific structures is
desired, users often rely on optical detection for identification and quantification.
In this application note we present the use of RapiFluor-MS, a novel reagent for
rapidly labeling released N-glycans. RapiFluor-MS dramatically reduces overall
released N-glycan sample preparation times to 30 minutes, while improving
fluorescence signal by up to 14x and MS signal by up to 1000x, compared to
traditional labeling techniques. In conjunction with reduced sample preparation
times, we geometrically scaled a highly resolving chromatographic method to
one having a total cycle time of 10 minutes. Finally, we incorporated the
ACQUITY QDa Detector to monitor specific glycan species using selected ion
recording (SIR), which provides a selective means of monitoring species, even
if they co-elute. We will discuss how the combination of RapiFluor-MS and the
ACQUITY QDa Detector provides a powerful solution for obtaining meaningful
data rapidly and efficiently.
Rapidly Monitoring Released N-Glycan Profiles During Process Development
Using
RapiFluor-MS and the ACQUITY QDa Detector
Eoin F.J. Cosgrave, Robert Birdsall, and Sean M. McCarthy
Waters Corporation, Milford, MA, USA
[ 118 ]
E X P E R IM E N TA L
Released N-glycans were prepared from commercially
available trastuzumab following the protocol provided within
the RapiFluor-MS sample preparation kit. High mannose species
used in spiking studies were isolated from RNase-B following
release and labeling with RapiFluor-MS. Mannose species were
chromatographically separated and collected using the Waters
Fraction Manager – Analytical. Collected samples were dried
down using a CentriVap and reconstituted in water. For each
analysis the mass load was approximately 32 pmol of released
and labeled N-glycan on column. LC-MS grade acetonitrile and
water were purchased from Pierce. Ammonium formate was
prepared using Ammonium Formate Solution-Glycan Analysis
(p/n 186007081) by pouring the entire contents of the solution
into 1 L of water and mixed. The UPLC® system used was
dedicated for applications which do not require non-volatile salts
to reduce the likelihood of adduct formation in the mass detector.
LC conditions
LC system:
ACQUITY UPLC H-Class Bio
Detectors:
ACQUITY UPLC FLR
ACQUITY QDa
Columns:
High resolving method:
ACQUITY UPLC Glycan BEH Amide,
130Å, 1.7 µm, 2.1 mm x 150 mm
(p/n 186004742)
High throughput method:
ACQUITY UPLC Glycan BEH Amide,
130Å, 1.7 µm, 2.1 mm x 50 mm
(p/n 186004740)
Column temp.:
60 °C
Sample temp.:
10 °C
FLR settings
Data rate:
5 points/sec
Excitation wavelength: 265 nm
Emission wavelength: 425 nm
QDa settings
Sample rate:
5 points/sec
Mass range:
500 – 1250 Da
Cone voltage:
15 V
Capillary voltage:
1.5 kV
Probe temp.:
400 °C
Mode:
Positive ion
Mobile phase A:
Acetonitrile (Pierce, LC-MS Grade)
Mobile phase B:
50 mM ammonium formate, pH 4.4,
(LC-MS Grade, ammonium
formate concentrate)
Mobile phase C:
Acetonitrile (LC-MS Grade)
Mobile phase D:
Acetonitrile (LC-MS Grade)
Gradient table high resolution method:
Flow
Time (mL/min)
%A
%B
%C
%D
Initial
0.400
75
25
0
0
35.0
0.400
54
46
0
0
36.5
0.200
0
100
0
0
39.5
0.200
0
100
0
0
42.5
0.200
75
25
0
0
47.4
0.400
75
25
0
0
55.0
0.400
75
25
0
0
Gradient table high throughput method:
Flow
Time (mL/min)
%A
%B
%C
%D
Initial
0.800
75
25
0
0
5.8
0.800
54
46
0
0
6.1
0.400
0
100
0
0
6.6
0.400
0
100
0
0
7.1
0.400
75
25
0
0
8.0
0.800
75
25
0
0
10.0
0.800
75
25
0
0
Data management
Empower 3 Chromatography Data Software (CDS)
[ 119 ]
Rapidly Monitoring Released N-Glycan Profiles During Process Development Using RapiFluor-MS and the ACQUITY QDa Detector
R E SU LT S A N D D IS C U S S IO N
EU
0
10
20
30
40
Minutes
2
3
4
5
10 minute
method
EU
0
10
20
Minutes
5
10
15
20
55 minute
method
*
*
Figure 1. A high resolution separation (top chromatogram) was scaled
to a high throughput method by scaling the gradient geometrically
while reducing column length and flow rate. While resolution is
reduced, selectivity remains constant.
Figure 2. Selected ion chromatograms for N-Glycan species
separated using high throughput method. Co-eluting M5 and
A2G(4)1are easily discriminated by mass detection.
During characterization of released N-glycans, a
highly resolving method is often used to provide
accurate identification and quantification of the
species present in samples. While these methods
can be effectively scaled, there is a corresponding
loss in resolution as overall run time decreases
when using the same chromatograph and particle
size column. Often, some loss in resolution will
be tolerated if the benefit of speed is achieved,
however critical structures must remain clearly
identifiable. As shown in Figure 1, moving from
a higher resolving 55 min method to a 10 min
high throughput method preserves much of the
resolution between N-glycan species, however
there is loss of resolution between the indicated
peaks when moving to the shorter method. This
loss of resolution complicates accurate monitoring
by optical detection as there is no ability to
discriminate between two species.
Due to the improved MS response, we introduced
the use of the ACQUITY QDa Detector as part of
the detector stream to selectively monitor each of
the species present in the sample. By using the SIR
function of the ACQUITY QDa Detector we were able
to collect independent chromatographic traces for
each of the components to overcome the challenge
of using optical only detection. As shown in
Figure 2, we can clearly discriminate between
different glycoforms by using selected ion recording.
For each species the corresponding peak, or peaks
for species with resolved positional isomers, can be
clearly identified and integrated for quantification.
Intensit
y
0
50000
Intensit
y
0
200000
Intensit
y
0
1x10
6
2x10
6
Intensit
y
0
50000
100000
Intensit
y
0
1x10
6
Intensit
y
0
200000
400000
Intensit
y
0
20000
Minutes
1
2
3
4
5
6
A2G(4)1
896.4 m/z
F(6)A2
888.4 m/z
F(6)A2G(4)1
969.4 m/z
F(6)A2G(4)2
1050.5 m/z
A2
815.3 m/z
F(6)A2G(4)2S1
865.5 m/z
M5
774.1 m/z
[ 120 ] Rapidly Monitoring Released N-Glycan Profiles During Process Development Using RapiFluor-MS and the ACQUITY QDa Detector
With a high throughput method developed and the ability to
selectively monitor particular species with mass detection, we
investigated the linearity of response for the target N-glycan
species Mannose 5 (M5). As described in the experimental section,
the RapiFluor-MS labeled M5 species was isolated from the labeled
N-glycan pool of RNAse-B. After collection, the collected material
was dried and reconstituted in water. The reconstituted sample
was added to a RapiFluor-MS labeled released N-Glycan sample
from trastuzumab at various levels. We investigated the linearity
of the response by selectively monitoring the peak area of M5 in
relation to the volume added to the sample. As shown in Figure 3,
the chromatographic reproducibility was quite good. In addition,
the peak area for each volume added was highly linear (Figure 4),
Intensit
y
0
50000
100000
150000
200000
250000
Minutes
1.00
2.00
3.00
4.00
5.00
SIR: M5
Figure 3. Overlay of chromatograms over a range of M5 spike levels. Spiked amounts ranged
from 1-6 µL of reconstituted M5. Absolute concentrations were not determined.
Figure 4. Plot of amount of M5 added vs. peak area for spike
M5 samples (data shown in Figure 3).
Figure 5. SIR for co-eluting glycan structures.
Left: fluorescence profiles of trastuzumab
N-glycans with increasing M5 (A to F).
The indicators denote the retention times
for co-eluting glycans M5 and A2G1.
Middle: SIR of M5 for each of the glycan
samples A to F. Right: SIR for the co-eluting
structure, A2G1. Use of ACQUITY QDa SIR
enables the quick determination of glycan
structure responsible for changing peak area
in fluorescence profiles.
R = 0.9929
0.0E+00
1.0E+05
2.0E+05
3.0E+05
4.0E+05
5.0E+05
6.0E+05
7.0E+05
8.0E+05
0
2
4
6
8
Peak area (uV sec)
Amount M5 added (µL)
EU
0
20
40
60
EU
0
20
40
EU
0
20
40
EU
0
20
40
EU
0
20
40
EU
0
20
40
Retention time (min)
1
2
3
4
5
Intensity
0
100000
200000
Intensity
0
100000
200000
Intensity
0
100000
200000
Intensity
0
100000
200000
Intensity
0
100000
200000
Intensity
0
100000
200000
Retention time (min)
1
2
3
4
5
Intensity
0
50000
100000
150000
Intensity
0
50000
100000
150000
Intensity
0
50000
100000
150000
Intensity
0
50000
100000
150000
Intensity
0
50000
100000
150000
Intensity
0
50000
100000
150000
Retention time (min)
1
2
3
4
5
FLR
A
B
D
E
F
C
ACQUITY QDa
M5
SIR 774.3 m/z
A2G1
SIR 896.4 m/z
AM5
BM5
DM5
EM5
FM5
CM5
AA2G1
BA2G1
DA2G1
EA2G1
FA2G1
CA2G1
*
*
*
*
*
*
strongly indicating that the mass detector provides a response
suitable for quantification.
After determining the linearity of response for spiked M5 species,
we simulated a bioreactor process in which the relative amount of
M5 was increasing. For this study we selected the A2G1 species as
the reference for relative quantification and spiked in increasing
amounts of M5. As shown in Figure 5, the abundance of A2G1
(right column) remains largely constant over the course of the study
while the M5 species (middle column) increases as expected with
increased spiking levels. In addition, the FLR trace (left column)
demonstrates an increase in peak area for these two species
(labeled peak), however in the absence of mass information the
precise cause of this increase cannot be determined.
[ 121 ]
Rapidly Monitoring Released N-Glycan Profiles During Process Development Using RapiFluor-MS and the ACQUITY QDa Detector
Waters Corporation
34 Maple Street
Milford, MA 01757 U.S.A.
T: 1 508 478 2000
F: 1 508 872 1990
www.waters.com
Waters, The Science of What’s Possible, ACQUITY, QDa, ACQUITY UPLC, UPLC, and Empower are registered trademarks of Waters
Corporation. RapiFluor-MS is a trademark of Waters Corporation. All other trademarks are the property of their respective owners.
©2015 Waters Corporation. Produced in the U.S.A. August 2015 720005438EN AG-PDF
C O N C LU S IO NS
For routine high throughput assays, RapiFluor-MS with the ACQUITY UPLC
H-Class Bio System and the ACQUITY QDa Detector provides a novel approach for
accurately monitoring released N-glycan species. Reduced sample preparation
times and greatly improved MS response when coupled with the ACQUITY UPLC
and ACQUITY QDa enable the ability to more closely monitor released N-glycan
profiles, something which has previously not been possible. While FLR detection
was used in this example, for high throughput methods requiring only relative
quantification this may not be needed as each species can be monitored with the
ACQUITY QDa Detector. As discussed here, complete sample preparation and
analysis can be completed in 40 minutes. In addition to reproducible sample
preparation, separation and quantification are reproducible and quantitative,
allowing scientists to make meaningful decisions rapidly.
Reference
1. Lauber MA, Brousmiche DW, Hua Z, Koza SM, Guthrie E,
Magnelli P, Taron CH, and Fountain KJ. Rapid Preparation
of Released N-Glycans for HILIC Analysis Using a Novel
Fluorescence and MS-Active Labeling Reagent. 2015,
Waters Application Note P/N 720005275EN.
[ 122 ] Rapidly Monitoring Released N-Glycan Profiles During Process Development Using RapiFluor-MS and the ACQUITY QDa Detector
WAT E R S S O LU T IO NS
GlycoWorks™ RapiFluor-MS N-Glycan Kit
Biopharmaceutical Platform Solution
with UNIFI®
ACQUITY UPLC® H-Class System
ACQUITY UPLC Glycan BEH
Amide Column
ACQUITY UPLC FLR Detector
Xevo G2-XS Mass Spectrometer
K E Y W O R D S
Automated N-Glycan analysis
A P P L I C AT IO N B E N E F I T S
■
■
A novel glycan labeling reagent,
RapiFluor-MS,™ significantly enhances
both FLR and MS signals. Improvement
from MS detection allows better detection
for minor glycan forms.
■
■
The Xevo® G2-XS QTof Mass Spectrometer
combines an off-axis ion guide, StepWave,™
with a novel collision cell design to provide
significant increases in sensitivity for
RapiFluor-MS labeled glycans.
I N T RO DU C T IO N
UPLC-FLR/MS(MS) analysis of released N-glycans labeled with a fluorescent
tag has become routine with high-performance LC and MS instrumentations.
Glycans labeled with commonly used fluorescent tags, such as 2-AB and 2-AA,
can be detected by fluorescent (FLR) detection with ultra-high sensitivity.
Unlike an FLR detector, mass spectrometry is known to be less sensitive to
detect native or tagged glycans, especially low abundant ones, due to their poor
ESI performance. The limited dynamic range of this approach has restricted the
use of this combined workflow for glycan characterization.
To overcome the low MS ionization efficiency associated with conventional
labels and confidently assign lower-level glycans, a novel tag, RapiFluor-MS
has been developed by Waters. RapiFluor-MS contains a rapid tagging reactive
group, an efficient fluorophore, and a functional group that imparts high
ionization efficiency.1 Complete tagging of glycans can be achieved in less than
5 minutes using this novel reagent.
Initial results with this glycan label show significant enhancement in both FLR
and MS(MS) signals compared to 2-AB.1 The increased sensitivity enables the
detection and identification of very low level glycans, at 0.1%, with sufficient
MS signal. In this study, we demonstrate the benefits of combining RapiFluor-MS
with an integrated UPLC-FLR/QTof MS system for detailed characterization of
the minor glycoforms from the human IgG and mouse IgG1 samples.
Applying a Novel Glycan Tagging Reagent,
RapiFluor-MS, and
an Integrated UPLC-FLR/QTof MS System for Low Abundant N-Glycan Analysis
Ying Qing Yu
Waters Corporation, Milford, MA, USA
[ 123 ]
E X P E R IM E N TA L
MS conditions
System:
Xevo G2-XS QTof MS:
ESI+ in sensitivity mode
(resolution ~ 30,000)
Capillary voltage:
3.0 kV
Cone voltage:
80 V
Source temp.:
120 °C
Desolvation temp.:
300 °C
Desolvation gas flow: 800 L/h
LockSpray
Capillary voltage:
3.0 V
Cone voltage:
40 V
Scan time:
0.5 s
Interval:
20 s
GFP solubilized in 0.1% formic acid with 50:50 (MeCN: H20) at
200 fmol/µL was infused, m/z = 785.8421 (z = 2) was used for
lock mass calibration.
Collision induced dissociation
MS/MS analyses were performed in continuum mode from
100–2000 m/z with collision induced dissociation (CID) to
generate glycan fragmentation data. Ions with 2+ and 3+ charge
states were selected for fragmentation. Customized collision
energy tables that were charge state and mass specific were used
for optimized fragmentations; the approximated CE range was
between 15 to 40 eV. Data Dependent Acquisition (DDA) was
used with duty cycle times of 1.6 sec and 0.5 sec for MS and
MS/MS modes. The two most abundant precursors were selected
for fragmentation.
Data management
UNIFI Scientific Information System v1.7.1
Sample preparation
The GlycoWorks RapiFluor-MS N-Glycan Kit Care and Use
manual (p/n 715004793en) contains a detailed sample
preparation procedure for the deglycosylation of N-glycans from
biotherapeutics, followed by the RapiFluor-MS labeling step
and glycan extraction using an SPE device. The entire sample
preparation procedure took 30 minutes.
LC conditions
All chromatographic mobile phases are prepared using LC/MS
compatible solvents and reagents.
System:
ACQUITY UPLC H-Class
Detector:
ACQUITY UPLC FLR
Column:
ACQUITY UPLC Glycan BEH Amide
Column, 130Å, 1.7 μm, 2.1 mm x 150 mm
(p/n 186004742)
Column temp.:
60 °C
Mobile phase A:
50 mM ammonium formate (pH 4.4)
Mobile phase B:
100% acetonitrile
UPLC HILIC LC gradient table:
Time
Flow rate
%A
%B Curve
(min) (mL/min)
0.0
0.4
25 75 6
40.0
0.4
49
51
6
41.5
0.2
100
0
6
44.5
0.2
100
0
6
48.1
0.2
25
75
6
52.6
0.4
25
75
6
60.0
0.4
25
75
6
FLR settings:
[ 124 ]
Applying a Novel Glycan Tagging Reagent, RapiFluor-MS, and
an Integrated UPLC-FLR/QTof MS System for Low Abundant N-Glycan Analysis
R E SU LT S
Previous work showed that the RapiFluor-MS
labeling reagent improves N-glycan MS ionization
in positive ion mode. More than two order of
magnitude MS sensitivity increase was observed
when compared to 2-AB label.1 Combined
with highly sensitive Xevo G2-XS QTof Mass
Spectrometer, we are now able to detect minor
glycoforms with high confidence.
Figure 2 shows an example of analyzing the
RapiFluor-MS labeled N-Glycans released from
0.5 µg of human IgG on UPLC/FLR/QTof MS system.
Comparable FLR and MS response across a broad
range of glycans was easily achieved.
The MS and MS/MS fragmentation spectra were
also shown as an example in Figure 2 for a minor
glycoform, A2G2S1, which is present at 0.1% level.
The MS spectrum shows doubly charged ions with
minor sodium adduct ions in the raw MS spectrum.
We observed a similar fragmentation pathway for
the RapiFluor-MS labeled glycans compared to the
2-AB labeled glycans. The MS/MS fragmentation
of A2G2S1 showed that glycosidic bond cleavage
from both reducing and non-reducing end was the
dominant fragmentation pathway. The observed
sequential neutral losses from the non-reducing
end stops at the first GlcNAc residue at the reducing
end with the RapiFluor-MS label attached. Also, the
counter fragment ions from the non-reducing end,
oxonium ions, were readily observed.
In addition to human IgG, we also tested the
RapiFluor-MS labeled glycans released from
a mouse IgG1 sample. It is well known that
N-glycolyneuraminic acid and alpha (1-3) galactose
containing N-glycans on mAbs generated from
murine cell lines are glycans with immunogenic
epitopes. These glycans present analytical
challenges, due to 1) their low abundance in the
glycan mixture, and 2) difficulty to characterize
them structurally due to poor MS and MS/MS
signals from using the conventional labels.
Figure 1. Biopharmaceutical
System Solution with UNIFI
for glycan analysis.
Figure 2. UPLC/FLR/MSMS analysis of RapiFluor labeled human IgG N-glycans. A) FLR data
channel. B) BPI MS data channel. The MS spectrum of a low intensity ion was inserted (A2G2S1).
The dominant ions were doubly charged with minor sodium adduct ions. C) Deconvoluted MS/MS
spectrum of A2G2S1 was displayed.
FLR
A
B
C
BPI MS
[ 125 ]
Applying a Novel Glycan Tagging Reagent, RapiFluor-MS, and
an Integrated UPLC-FLR/QTof MS System for Low Abundant N-Glycan Analysis
Figure 3 shows an example of a UPLC/FLR/QTof MS analysis of the mouse IgG1 glycans that contain these
immunogenic epitopes. Structural informative fragments (with asterisks) are observed for a low abundant
immunogenic glycan, FA2Gal1Sg1, which is present at about 0.1% level. The fragment ion at m/z of
528.2 suggests this glycan contains alpha-gal when this ion was the most dominant fragment ion in the
entire spectrum; also another diagnostic ion at m/z of 2260.8 was generated from losing one NeuGc from
the precursor ion. This glycan was also observed in FLR chromatogram of 2-AB labeled glycans without
sufficient MS signals to obtain good quality CID fragmentation (data not shown). With RapiFluor-MS labeling
chemistry, sufficient amount of precursor ions were obtained for subsequent MS/MS fragmentation.
Overall, we demonstrated that RapiFluor-MS labeling chemistry enhances MS and MS/MS sensitivity to
obtain high quality precursor and fragmentation ion spectra. Therefore, rich structural information for low
abundant glycan species are achieved using this approach.
FLR
A
BPI MS
B
Figure 3. UPLC/FLR/MSMS analysis of RapiFluor-MS labeled mouse IgG1 N-glycans. A) FLR data channel. B) BPI MS data channel. One of the last eluting glycans were
selected for MS/MS fragmentation. The deconvoluted fragmentation data from FA2Ga1Sg1 was displayed in C); “Ga” stands for galatose and “Sg” stands for NeuGc.
Structurally informative fragments (with asterisks) are observed for this low abundant ion (< 0.1% relative abundance). Fragment ion at m/z of 528.2 suggests this
glycan contains alpha-gal when this ion was the most dominant fragment ion in the entire spectrum; also another diagnostic ion at m/z of 2260.8 was generated
from the loss of one NeuGc from the precursor ion.
[ 126 ]
Applying a Novel Glycan Tagging Reagent, RapiFluor-MS, and
an Integrated UPLC-FLR/QTof MS System for Low Abundant N-Glycan Analysis
Waters Corporation
34 Maple Street
Milford, MA 01757 U.S.A.
T: 1 508 478 2000
F: 1 508 872 1990
www.waters.com
Waters, The Science of What’s Possible, Xevo, ACQUITY UPLC, and UNIFI are registered trademarks of Waters Corporation. RapiFluor-MS,
StepWave, and GlycoWorks are trademarks of Waters Corporation. All other trademarks are the property of their respective owners.
©2015 Waters Corporation. Produced in the U.S.A. April 2015 720005383EN AG-PDF
C O N C LU S IO NS
LC/FLR analysis of N-glycans released from protein therapeutics is performed
routinely in analytical laboratories around the world. For scientists who want
to add MS characterization capability to their glycan analysis, they often
struggle with low MS signals and poor quality MS/MS fragmentation for
mass confirmation and structure elucidation using conventional FLR labels
such as 2-AB and 2-AA. To address these challenges, Waters offers enabling
technologies that include the novel RapiFluor-MS labeling chemistry for rapid
glycan sample preparation, and a UPLC/FLR/QTof MS system controlled by UNIFI
Scientific Information System. The improved FLR and MS sensitivity from the
RapiFluor-MS label and the QTof MS with StepWave Technology allow confident
identification and characterization of minor but critical glycoforms from mAbs.
References
1. Rapid Preparation of Released N-Glycans for HILIC Analysis
Using a Novel Fluorescence and MS-Active Labeling
Reagent. Waters and New England Biolabs application note
(p/n 720005275en.)
2. GlycoWorks RapiFluor-MS Kit Care and Use Manual
[ 127 ]
Applying a Novel Glycan Tagging Reagent, RapiFluor-MS, and
an Integrated UPLC-FLR/QTof MS System for Low Abundant N-Glycan Analysis
Ying Qing Yu
Waters Corporation, Milford, MA, USA
GOA L
To demonstrate two fit-for-purpose glycan
analysis workflows for comprehensive N-linked
glycan profiling and structural elucidation
within the Waters® Glycan Application Solution
with UNIFI®.
BAC KG ROUND
The vast majority of biotherapeutics are
glycosylated. Glycans attached to the proteins
play a critical role in the serum half-life,
efficacy, and safety of the biotherapeutic drug.
In recent years, Waters has launched a series
of innovative analytical tools to address the
challenges faced in N-glycan analysis. This
began with the launch of the ACQUITY UPLC®
BEH Amide Column (1.7 µm particle size) for
enhanced chromatographic separations of
glycans under HILIC mode.1 Early in 2015,
Waters then introduced a new GlycoWorks™
sample preparation kit that provides fast, easy
N-glycan sample preparation from enzymatic
glycan release to labeling and clean up. This kit
includes a novel fluorescent labeling reagent,
RapiFluor-MS™, enabling highly sensitive mass
spectral detection of the labeled glycans.2,3
The advancements in sample preparation,
chromatographic separation, and enhanced
ESI MS is now further complemented with an
equally enterprising Informatics solution – the
UNIFI Scientific Information System – to
streamline glycan data acquisition, processing,
and reporting.4
Released N-linked Glycan Analysis Using the
Glycan Application Solution with UNIFI
■
■
ACQUITY UPLC H-Class Bio System
■
■
ACQUITY UPLC Column Manager
■
■
ACQUITY UPLC FLR Detector
■
■
Xevo® G2-XS QTof MS
■
■
UNIFI Scientific Information System
■
■
GlycoWorks RapiFluor-MS N-Glycan Kit
■
■
ACQUITY UPLC Glycan BEH Amide Column
Comprehensive N-linked glycan analysis using the
Glycan Application Solution with UNIFI.
Figure 1. Glycan Application Solution with UNIFI for RapiFluor-MS labeled glycan analysis.
[ 128 ]
T H E SO LU T ION
Two workflows available with the Glycan Application
Solution with UNIFI are featured: 1) Glycan FLR with
MS confirmation for profiling and mass confirmation;
2) Glycan DDA workflow via exporting of processed
MS/MS data to SimGlycan (Premier Biosoft) for
identification and structural elucidation.
Workflow 1: Glycan FLR with MS confirmation
The heart of this workflow is a scientific library
containing calibrated chromatographic retention
times (in glucose units, GU5) and accurate mass
values for fluorescently labeled glycan structures.
N-glycan identification using the scientific glycan
library is illustrated in Figure 2. The assignment is
based on accurately matched retention times in GU
(calibrated using a fluorescently labeled dextran
ladder) and accurate mass measurements from a
Xevo G2-XS QTof MS.6 Currently, a comprehensive
2AB-glycan GU library containing 319 unique
N-glycan structures from therapeutic proteins is
available with the Glycan Application Solution
with UNIFI. A new scientific library based on the
RapiFluor-MS labeling technology is currently
under joint development by Waters and NIBRT.
UNIFI software also allows users to create
customized glycan scientific libraries which can
be constructed directly by entering experimental
GU values and importing structures from
GlycoWorkbench. In addition, this workflow
automatically calculates relative percentage value
for each glycan component based on integrated
fluorescent intensity for robust quantitation.
Workflow 2: Glycan DDA
Figure 3 shows the Glycan DDA workflow: Glycan
MS/MS information using a Data Dependent
Acquisition (DDA) mode was first acquired, followed
by peak processing to convert all ions to singly
charged “ion sticks.” The processed data can then
be exported in either .mzML or .LCS file format into
SimGlycan software for identification and fragment
ion annotation.
Figure 2. Workflow 1: Glycan FLR with MS confirmation.
Figure 3. Workflow 2: Glycan DDA workflow in UNIFI Software. Collision-induced dissociation (CID)
of glycans in data dependent acquisition (DDA) mode is processed and exported to SimGlycan.
[ 129 ]
Released N-linked Glycan Analysis Using the Glycan Application Solution with UNIFI
Waters Corporation
34 Maple Street
Milford, MA 01757 U.S.A.
T: 1 508 478 2000
F: 1 508 872 1990
www.waters.com
Waters, The Science of What’s Possible, UNIFI, ACQUITY UPLC, and Xevo are registered trademarks of Waters Corporation.
GlycoWorks and RapiFluor-MS are trademarks of Waters Corporation. All other trademarks are the property of their respective owners.
©2016 Waters Corporation. Produced in the U.S.A. January 2016 720005598EN AW-PDF
Analytical method for Glycan Application Solution in UNIFI
Sample preparation:
N-glycans were prepared using the GlycoWorks
RapiFluor-MS N-Glycan Kit (p/n 176003713)
System:
Biopharmaceutical Platform with UNIFI
LC settings for RapiFluor-MS labeled glycans
Column:
ACQUITY UPLC Glycan BEH Amide, 130Å, 1.7µm,
2.1 mm x 150 mm (p/n 186004742)
Column temp.:
60 °C
Mobile phase A:
50 mM ammonium formate (pH 4.4, LC-MS grade)
Mobile phase B:
100% acetonitrile (LC-MS grade)
Gradient:
Time Flow rate
(min) (mL/min) A (%)
B (%) Curve
0.0 0.4 25% 75% 6
35.0 0.4 46% 54% 6
36.5 0.2 80% 20% 6
39.5 0.4 25% 75% 6
43.1 0.4 25% 75% 6
55.0 0.4 35% 75% 6
Fluorescent:
λex = 265 nm, λem = 425 nm
Xevo G2-XS QTof MS settings
Capillary voltage:
3.0 kV
Sample cone:
30 V
Source temp.:
120 °C
Desolvation temp.:
300 °C
Desolvation gas:
800 L/hr
Recommend settings for DDA
Charge state recognition: 2+, 3+, and 4+
Collision energy ramping
Low mass start:
10 V, low mass end: 15 V
High mass start:
44 V, high mass end: 50 V
MS scan:
0.5 sec, MS/MS scan: 0.5 sec
SUMMA RY
Released glycan analyses are traditionally done
using either optical or MS only analytical systems,
and the data interpretation can be very challenging
due to a lack of integrated analytical systems. The
Glycan Application Solution with UNIFI features two
independent analytical glycan workflows. These
workflows allow scientists to characterize and
profile glycans using both optical (fluorescent) and
MS (MS) data within an integrated UPLC/FLR/QTof
MS system. Combined with the novel RapiFluor-MS
glycan labeling technology and the sophisticated
UNIFI Software, scientists are now able to identify
and quantify low abundant, potentially immunogenic
glycan structures with higher confidence.
References
1. “Separation of 2-aminobenzamide labeled glycans using
hydrophilic interaction chromatography columns packed with
1.7 um sorbent.” J. Chrom. B. 878 (2010) 403–408.
2. “Rapid Preparation of Released N-Glycans for HILIC
Analysis Using a Labeling Reagent that Facilitates Sensitive
Fluorescence and ESI-MS Detection.” Anal. Chem. 2015, 87
(10), 5401–5409.
3. “Applying a Novel Glycan Tagging Reagent, Rapi Fluor-MS,
and an Integrated UPLC-FLR/QTof MS System for Low
Abundant N-Glycan Analysis.” Waters Application Note
(p/n 720005383EN).
4. “UNIFI Scientific Information System.” Waters literature.
5. Mattew P. Campbell, Louise Royle, Catherine M. Radcliffe,
Raymond A. Dwek and Pauline M. Rudd. “GlycoBase
and autoGU: tools for HPLC-based glycan analysis.”
Bioinformatics Applications Note, vol. 24, no, 9, 2008,
1214–1216.
6. “Biopharmaceutical Platform Solution with UNIFI: A
Holistic Workflow for Acquiring, Processing, and Reporting
Fluorescent-Labeled Glycans.” Waters Application Note
(p/n 720004619en).
[ 130 ] Released N-linked Glycan Analysis Using the Glycan Application Solution with UNIFI
WAT E RS SO LU T IONS
Biopharmaceutical Platform Solution
with UNIFI
ACQUITY® UPLC BEH Glycan Column
UNIFI® Scientific Information System
Glycobase Database
2-AB Dextran Calibration Ladder
GlycoWorks™ Reductive Amination Single
Use Sample Preparation Kit
K E Y WO R DS
2-AB labeled glycans, dextran ladder,
Glycan Units
A P P LIC AT ION BEN E FIT S
We present a fully integrated HILIC UPLC® FLR
detection workflow that is carried out within a
comprehensive compliant-ready platform that
integrates informatics and instrument control.
This platform enables laboratories to perform
routine biotherapeutic glycan analysis with
greater speed, accuracy, and consistency than
by using disparate laboratory processes.
INT RO DUC T ION
One of the challenges of managing the routine use of analytics in
biopharmaceutical laboratories is the number of approaches used
to characterize a biomolecule – and the complexity of controlling
instruments and processing data collected from different structural levels.
The Waters® Biopharmaceutical System Solution with UNIFI integrates high-
resolution biotherapeutic analyses with bioinformatics that are designed to
support routine workflows used in the development process. The system enables
researchers to acquire, process, report, and share mass spectrometry and
chromatography data including intact mass analysis of proteins, peptide mapping,
and glycan profiling – with a single software platform that can also manage
multiple systems in a networked workgroup.
In this application note, we illustrate a dedicated workflow for the acquisition,
processing, and automated reporting of data from fluorescent-labeled (such as
2-aminobenzamide, or 2-AB) released glycan samples. Sample data was acquired
using a HILIC UPLC® separation and fluorescent (FLR) detection methodology.
Normalized retention time for the various glycans was achieved using a calibrated
RapiFluor-MS Dextran Calibration Ladder (p/n 186007982).
Data acquisition, processing, reporting and management were achieved under
control of the UNIFI Scientific Information System. The HILIC UPLC FLR System
used for glycan data acquisition was part of a compliant-ready UNIFI workgroup,
consisting of two UPLC/QTof MS systems and multiple UPLC optical detection
(FLR and UV) systems. This workgroup configuration enabled centralized
instrument control, processing, and data integration on a common server for the
methods, data, and reports acquired from the networked systems. We demonstrate
that such an integrated laboratory and data workflow provides exceptional
efficiencies for routine released glycan profiling of biotherapeutics.
A Holistic Workflow for Acquisition, Processing, and Reporting Fluorescent-
Labeled Glycans Using the Biopharmaceutical Platform Solution With UNIFI
Ying Qing Yu
Waters Corporation, Milford, MA, USA
[ 131 ]
E X P E RIM ENTA L
FLR settings
UPLC HILIC LC gradient table
Biopharmaceutical Platform Solution with UNIFI
■
■
ACQUITY UPLC H-Class System
■
■
ACQUITY UPLC BEH Glycan Column
■
■
ACQUITY UPLC FLR Detector
■
■
UNIFI Scientific Information System
■
■
2-AB Dextran Calibration Ladder (p/n 186006841)
■
■
GlycoWorks Reductive Amination Single Use Sample
Preparation Kit (p/n 176003119)
Sample preparation
The 2-AB Dextran Calibration Ladder (p/n 186006841) and the
GlycoWorks Reductive Amination Single Use Sample Preparation
Kit (p/n 176003119) are glycan standards available from the
Waters Corporation. The dextran ladder is used to calibrate
and normalize labeled glycan retention times for exceptional
day-to-day, system-to-system, and lab-lab reproducibility. The
retention time for polyglucose 4–12 peaks were used to produce
a fifth order polynomial calibration curve (Glucose Unit or GU
vs. Retention Time). All analyte peaks are reported and searched
using this calibrated GU value.
LC conditions
System:
ACQUITY UPLC H-Class System
Column:
ACQUITY UPLC Glycan BEH Amide Column,
130Å, 1.7 µm, 2.1 mm x 150 mm
(p/n 186004742)
Column temp.:
40 °C
Mobile phase A:
50 mM ammonium formate (pH 4.4)
Mobile phase B:
Acetonitrile
Informatics
UNIFI Scientific Information System
[ 132 ]
A Holistic Workflow for Acquisition, Processing, and Reporting Fluorescent-Labeled Glycans
using the Biopharmaceutical Platform Solution with UNIFI
R E SU LT S AND DIS CUSSION
Workflow for routine released glycan determination
The majority of the therapeutic proteins are glycosylated, and the attached glycans have significant impact on
the efficacy and safety of the biotherapeutic. The International Conference on Harmonization Guideline Q6B
requires the analysis of carbohydrate content, structural profiles, and characterization of the glycosylation
site(s) within the polypeptide chain(s).
The most widely adopted analytical workflow for routine N-linked glycan characterization involves
labeling the enzymatically released glycans with a fluorescent tag (typically 2-aminobenzamide, or 2-AB),
resolving the labeled glycans by hydrophilic interaction liquid chromatography (HILIC UPLC), and detecting
the labeled glycan peaks with a fluorescence detector. The assignment of glycan peaks during routine
analysis is fundamentally based on matching their retention time to established values. For non-routine
analysis, glycosidase arrays or MS analysis are employed to give tentative assignments or resolve
ambiguous peak assignments.
In order to best control method variation (between runs, days, instruments, scientists, and labs) glycan profiles
from the HILIC separation are always calibrated and normalized against a 2-AB Dextran Calibration Ladder
(glucose homopolymer). Glycan peaks in an unknown sample can be assigned a Glucose Unit (GU) value from
the GU vs. Retention Time calibration curve using the dextran ladder, which is typically fitted with a fifth-order
polynomial or cubic spline calibration line.1
The new streamlined workflow in UNIFI Software, v. 1.6, (Figure 1) enables users to automatically collect one
or more dextran ladder standard data sets, process the chromatograms, generate dextran calibration curve, and
apply curves directly to unknown samples. Assigned GU values for peaks in an unknown are searched through
Glycobase 3.1+, an integrated UPLC GU released glycan database, developed by Waters in collaboration with
the National Institute for Bioprocessing Research and Technology (NIBRT)
(https://glycobase.nibrt.ie/glycobase/browse_glycans.action).
Targeted component search is enabled within a more information-rich integration of Glycobase in the next
version of UNIFI Software, v. 1.6.1. The integrated Glycobase database will reference hundreds of glycan
GU entries along with supporting structural, mass, and exoglycosidase digestion pathway information.
[ 133 ]
A Holistic Workflow for Acquisition, Processing, and Reporting Fluorescent-Labeled Glycans
using the Biopharmaceutical Platform Solution with UNIFI
Figure 1. Glycan analysis workflow. A) 2-AB Dextran Calibration Ladder was used as a separation standard. B) The Glucose Unit (4-12)
calibration from the separation standard was fitted by a fifth order polynomial curve. C) Glycan peaks from an unknown sample
were assigned GU values from this curve. The measured GU value was searched using the Glycobase entries to assign structures
from GU values.
[ 134 ]
A Holistic Workflow for Acquisition, Processing, and Reporting Fluorescent-Labeled Glycans
using the Biopharmaceutical Platform Solution with UNIFI
UNIFI method generation
Figure 2 depicts the key elements of producing an automated and holistic glycan acquisition, processing,
and reporting method within UNIFI. The instrument settings define the instrument modules required for the
analysis, and define method parameters for HILIC UPLC separation and fluorescence detection. The processing
setting defines peak integration and dextran ladder calibration settings. A comprehensive report can be
generated automatically following acquisition and processing using one or more report templates that can
be readily modified. Modifications to the processing and reporting settings within the method are possible
post-acquisition, and are audit trailed for simplified documentation and compliance purposes.
Figure 2. The UNIFI released glycan analysis method.
[ 135 ]
A Holistic Workflow for Acquisition, Processing, and Reporting Fluorescent-Labeled Glycans
using the Biopharmaceutical Platform Solution with UNIFI
Figures 3 through 5 detail how a scientist can use UNIFI Software to navigate through processed and raw chromatographic results. A typical
dextran calibration curve shown in Figure 3 is applied to the GlycoWorks Reductive Amination Single Use Sample Preparation Kit (p/n
176003119), a mixture of N-glycans from mAb (Figure 4). Reproducibility of the results is depicted in Figure 5. The ability of UNIFI to
automatically summarize data from multiple injections of standards and unknown samples is highlighted in these displays.
Figure 3. In UNIFI Software,
the review panel summarizes
GU values (GU 4–12) obtained
across 5 injections of the 2-AB
Dextran Calibration Ladder
standard. The chromatogram
of the first injection and
calibration curve results for all
injections are displayed below
the summary table.
Figure 4. A component
summary table in the review
page that compares the GU
value for all the identified
glycans across a triplicate
injection of a glycan mixture
sample. Summary calculations
for individual glycans across
triplicate analysis of the glycan
test standard are presented in
the component summary table.
[ 136 ]
A Holistic Workflow for Acquisition, Processing, and Reporting Fluorescent-Labeled Glycans
using the Biopharmaceutical Platform Solution with UNIFI
Figure 5. The Investigate
panel of the experiment
demonstrates glycan profile
reproducibility for five HILIC
UPLC FLR analyses of the
glycan mixture.
Figure 6 provides a snapshot of a typical glycan
analysis report; information such as the sample list,
retention time calibration curve fitting for 2-AB
Dextran Calibration Ladder, and calculated GU
values for unknown glycan samples are organized
within single report template. Valuable information
such as the relative abundance and the GU value for
each glycan component were also summarized with
appropriate statistical analysis.
Figure 6. A typical UNIFI released glycan report.
[ 137 ]
A Holistic Workflow for Acquisition, Processing, and Reporting Fluorescent-Labeled Glycans
using the Biopharmaceutical Platform Solution with UNIFI
Waters Corporation
34 Maple Street
Milford, MA 01757 U.S.A.
T: 1 508 478 2000
F: 1 508 872 1990
www.waters.com
Waters, The Science of What’s Possible, ACQUITY UPLC, UPLC, and UNIFI are registered trademarks of Waters Corporation.
RapiFluor-MS and GlycoWorks are trademarks of Waters Corporation. All other trademarks are the property of their respective owners.
©2013–2016 Waters Corporation. Produced in the U.S.A. January 2016 720004619EN AG-PDF
CONC LUSIONS
UNIFI is the first comprehensive software that seamlessly integrates UPLC
chromatography, optical detection, high resolution mass spectrometry, and
integrated informatics within one platform. Released glycan analysis is one
of the latest application workflows to be offered in UNIFI Software, v. 1.6.
This new UNIFI Application Solution enables a scientist in regulated or
unregulated laboratory environments to acquire, process, and report qualitative
and quantitative glycan information along with high confidence and minimal
user intervention.
Reference
1. Campbell, M.P.; Roylez, L.; Radcliffe, C.M.; Swek, R.A.; Rudd,
P.M. GlycoBase and autoGU: tools for HPLC-based glycan
analysis. Bioinformatics. 2008, 24:9, 1214–1216.
[ 138 ]
A Holistic Workflow for Acquisition, Processing, and Reporting Fluorescent-Labeled Glycans
using the Biopharmaceutical Platform Solution with UNIFI
WAT E RS SO LU T IONS
UNIFI Scientific Information System
ACQUITY UPLC® H-Class Bio System
ACQUITY UPLC Glycan BEH Amide Column
GlycoWorks™ Reductive Amination
Single Use Sample Preparation Kit
2-AB Dextran Calibration Ladder
2-AB Glycan Performance Test Standard
K E Y WO R DS
Biosimilar, etanercept, Waters Glycan GU
Library, glucose units
A P P LIC AT ION BEN E FIT S
The integrated UPLC/FLR/QTof MS analytical
technologies available with the Waters
Biopharmaceutical Platform Solution with UNIFI®
improve a biopharmaceutical organization’s
ability to deliver well-characterized glycosylated
biotherapeutics to market, from discovery
through QC. The solution allows routine
assignment of N-linked glycan structures using
data from time-aligned FLR and MS channels
along with database-driven assignment of
glycans based on retention time. This enables
the profiling of released glycans for individual
analysis or to facilitate multi-batch or
biosimilar/innovator comparability studies.
INT RO DUC T ION
The Waters® Biopharmaceutical Platform Solution with UNIFI is comprised of
industry-leading UPLC bioseparations columns and analytical instrumentation,
along with optical detection and mass spectrometry, for comprehensive
biopharmaceutical characterization and analysis. Data acquisition, processing,
bioinformatics, and reporting tools are integrated and automated within UNIFI
Scientific Information System’s compliant-ready architecture.
In this application note, we detail a new workflow for a glycan assay, using FLR
with mass confirmation, available in the latest version of the Biopharmaceutical
Platform Solution with UNIFI. The practical use of this workflow for fluorescent
labeled (2-AB) N-linked released glycan characterization is illustrated using a
biosimilar/innovator biotherapeutic comparability study.
The analytical platform used for this study is comprised of an ACQUITY UPLC
H-Class Bio System and an ACQUITY UPLC Fluorescent Detector in-line with a
Xevo G2-S QTof Mass Spectrometer. This Glycan Application Solution with UNIFI
enables the assignment and profiling of 2-AB labeled released N-linked glycans
based on searches of calibrated retention time in glucose units (also known
as GU) and accurate mass data within the Waters Glycan GU Library, which is
integrated within UNIFI Scientific Information System version 1.7 and higher.
Accurate mass analysis proves a valuable technique for confirming GU based
assignments and distinguishing cases where multiple glycan structures could
be assigned to a single peak. Other complementary data for confirming these
assignments (e.g. glycan DDA MS/MS data and exoglycosidase array studies)
can also be collected on the Biopharmaceutical Platform Solution with UNIFI,
and will be addressed in future application notes.
N-linked Glycan Characterization and Profiling:
Combining the Power of Accurate Mass, Reference Glucose Units,
and UNIFI Software for Confident Glycan Assignments
Ying Qing Yu
Waters Corporation, Milford, MA, USA
[ 139 ]
E X P E RIM ENTA L
LC conditions
System:
ACQUITY UPLC H-Class Bio System
Column:
ACQUITY UPLC Glycan BEH Amide
Column, 130Å, 1.7 µm,
2.1 mm x 150 mm (p/n 186004742)
Column temp.:
40 °C
Mobile phase A:
50 mM Ammonium Formate (pH 4.4)
Mobile phase B:
Acetonitrile
Note:
LC-MS grade water and acetonitrile
was used for this experiment
MS conditions
MS system:
Xevo G2-S QTof MS
Mode:
ESI+ in sensitivity mode
Capillary voltage:
3.0 kV
Cone:
80 V
Source temp.:
120 °C
Desolvation temp.:
300 °C
Desolvation gas flow: 800 L/h
Scan time:
0.5 s
Interval:
20 s
Data acquisition, processing, and reporting
UNIFI Scientific Information System
In this work, we illustrate the features of the platform for
glycan analysis:
■
■
The analytical workflow moves seamlessly from acquisition
through data processing, with FLR and MS data channels
being acquired and time-aligned automatically, for a routine
and repeatable approach to data processing and reporting.
■
■
The Waters Glycan GU Library allows confident assignment
of the glycan structures based on retention time (in GU)
with accurate mass confirmation.
■
■
The streamlined workflow continues through reporting of
quantitative (relative %) and qualitative analysis of N-glycan
profiles, enabling scientists to easily communicate this critical
information without exporting information to external data
packages and thus reducing sources of data manipulation error.
As a result, the laboratory’s ability to maintain compliance and
data integrity is enhanced.
ACQUITY UPLC FLR Detector settings
UPLC HILIC gradient table
[ 140 ]
N-Linked Glycan Characterization and Profiling: Combining the Power of Accurate Mass,
Reference Glucose Units, and UNIFI Software for Confident Glycan Assignments
Sample preparation and retention time calibration in GU values
The 2-AB Dextran Calibration Ladder (p/n 186006841) and the 2-AB Glycan Performance Test Standard
(p/n 186006349) are glycan standards available from Waters Corporation. The 2-AB Dextran Calibration
Ladder is used to calibrate and normalize labeled glycan retention times for exceptional day-to-day, system-
to-system, and lab-to-lab reproducibility. This enables routine use of the Waters Glycan GU Library to produce
primary glycan assignments.
The retention times for polyglucose 4–12 peaks were used to produce a fifth order polynomial calibration
curve of GU vs. retention time. Peaks in experimental samples are automatically assigned and reported using
this calibrated GU value. The 2-AB Glycan Performance Test Standard (p/n 186006349) contains a set of
biantennary glycans, including high mannose and sialated structures, typical of many therapeutic mAbs
commercializaed and in development today.
The GlycoWorks Reductive Amination Single Use Sample Preparation Kit (p/n 176003119) was used to
generate 2-AB labeled released N-linked glycans from the innovator and a candidate biosimilar version
of etanercept. The instructions were followed as detailed in the documentation package for the kit.
Fluorescent and MS chromatogram alignment
The fluorescent and MS chromatograms were aligned automatically during data acquisition using an
experimentally derived value entered on the instrument console page. The time offset value depends on the
length of the peak tubing (connection between the FLR and MS inlet) and the flow rate, and may vary system
to system.
Critical settings
In the UNIFI processing method, settings are made for retention time calibration using 2-AB Dextran
Calibration Ladder (p/n 186006841) (GU 4-12 is the typical range of calibration for mAb derived glycan
samples). Both the fifth order and the cubic spine curve fit are applicable for retention time calibration.
[ 141 ]
N-Linked Glycan Characterization and Profiling: Combining the Power of Accurate Mass,
Reference Glucose Units, and UNIFI Software for Confident Glycan Assignments
R E SU LT S AND DIS CUSSION
In this work, we provide specific details about the glycan UPLC-FLR/MS workflow (Figure 1) used with the
Biopharmaceutical Platform Solution with UNIFI, including details of the analytical methods employed, the
data review workflows employed, and reporting schemes required for efficient analysis of individual glycan
samples and for more complex comparability studies.
Data acquision
Data processing
and confirmaon through
scienfic library search
Reporng
Figure 1. The glycan assay workflow, using FLR with MS confirmation.
Step 1: Data acquisition
Fluorescent-labeled glycans were separated using an ACQUITY UPLC H-Class Bio System with both FLR and
MS detection, the latter using the Xevo G2-S QTof MS. The ACQUITY UPLC FLR Detector was directly interfaced
with the QTof MS without any fluidic path modifications. The MS chromatogram is automatically time aligned
with the FLR chromatogram as described above. An example of the UPLC-FLR/QTof MS chromatogram is shown
in Figure 2.
Figure 2. UPLC-FLR/MS chromatogram of 2-AB Glycan Performance Test Standard. The FLR chromatogram is shown at the top,
and the BPI MS chromatogram is shown at the bottom. The BPI MS trace was time aligned with the FLR during data acquisition.
[ 142 ]
N-Linked Glycan Characterization and Profiling: Combining the Power of Accurate Mass,
Reference Glucose Units, and UNIFI Software for Confident Glycan Assignments
A 2-AB Dextran Calibration Ladder (p/n 186006841) was used as a retention time calibration standard.
Typically, the samples are sandwiched in between dextran ladder injections. A fifth order curve, or cubic
spline curve, for retention times vs. glucose unit values was automatically calculated using the average of all
dextran ladders analyzed, and subsequently applied to the experimental glycan chromatograms during data
processing. The benefit of using retention time calibration is to adjust the retention time shift to accommodate
any variations in mobile phase preparation, instrument configuration, and other aspects of user and
laboratory variability.
Figure 3 reviews the dextran ladder standard calibration result.
Figure 3. An example of dextran ladder calibration is shown. The top chromatogram shows the overlay of four injections of dextran
ladder collected before and after two experimental sample runs. The bottom plot is the fifth order curve generated from these
injections. The R2 (0.999963) and overlayed data points highlight the excellent retention time correlations across these injections.
[ 143 ]
N-Linked Glycan Characterization and Profiling: Combining the Power of Accurate Mass,
Reference Glucose Units, and UNIFI Software for Confident Glycan Assignments
Step 2: Data processing and scientific library search
The Waters Glycan GU Library contains retention times (in GU) and mass information for 2-AB labeled N-linked
glycans from a list of diverse glycoproteins as well as bulk human serum. The total number of unique glycans is
currently 300. GU value, molecular formula, glycan structure, and monoisotopic mass are associated with each
glycan entry. Scientists can search for a particular glycan or focus a search on specific classes of glycans in the
library search.
An example is shown in Figure 4. The reagent selected is 2-aminobenzamide (2-AB), since the experimental
GU values in the Waters Glycan GU Library are from 2-AB labeled proteins. The search criteria functions as a filter
to narrow the range of GU search tolerance and by using restrictions for many types of glycan attributes.
The Waters Glycan GU Library is the default library delivered with the Biopharmaceutical Platform Solution
with UNIFI; however, a user can create their own GU library to search instead or in addition to the Waters library.
Figure 4. Library search settings for Waters Glycan GU Library.
[ 144 ]
N-Linked Glycan Characterization and Profiling: Combining the Power of Accurate Mass,
Reference Glucose Units, and UNIFI Software for Confident Glycan Assignments
Waters Glycan GU Library
UNIFI Scientific Information System’s automated data processing encompasses calculation of the GU value
for integrated FLR peaks, determination of accurate mass values associated with the peaks, and the resulting
scientific library search. The assignment is based on the following logic:
1. All glycans with experimental GU values (experimental vs. library) within the database GU search tolerance
are associated with an FLR peak.
2. Among the potential assignments, those with accurate mass confirmation are given priority of assignment,
with closest GU value assigned as the default candidate.
3. Since FLR is more sensitive than mass spectrometry, the very low-level glycans may have good FLR signal,
but no or low MS signals. The assignment of these glycans may only be based on the GU value difference
from the database.
4. In the case of coeluting glycans or glycans with identical GUs, the glycan that is most abundant (by mass
spectrometry signal) gets the default assignment. Less abundant glycans (if present) are still represented
in the alternative assignments (with mass confirmed checked).
5. When a GU value is not found in the library within the given search tolerance, such peaks are marked as
“Discovered” components. Further investigation is needed to identify these peaks, and once the structures
of these glycans are verified, a new library entry can be created.
6. Glycans that are structural isomers tend to have close GUs and identical mass. In such cases, the matching
isomers will be marked as mass confirmed; the one that has the closest GU value will be highlighted as the
top assignment. These may require glycosidase treatment or MS/MS analysis for direct assignment.
Figure 5 is a screen capture from UNIFI Scientific Information System’s review tab, detailing the processed
library search results. The FLR peaks are integrated and assigned with the best match. After reviewing the
search result, an assignment can be changed to another glycan that has a similar GU value. This change is
tracked by audit trail within the software.
Figure 5. Waters Glycan GU Library search result from the review window. On the left is the processed FLR chromatogram and
XIC MS of a highlighted glycan; on the right is the library search result of the selected glycan peak. Information such as the structure
(with linkage assignment), expected GU, expected mass, ΔGU and Δmass are listed. In addition, the “Mass Confirmed” box is checked
off if the mass of any of the candidate glycans is observed.
[ 145 ]
N-Linked Glycan Characterization and Profiling: Combining the Power of Accurate Mass,
Reference Glucose Units, and UNIFI Software for Confident Glycan Assignments
Practical application: Using UNIFI Scientific Information System to compare N-glycan profiles of an
innovator biotherapeutic and a biosimilar candidate
Etanercept (trade name Enbrel) is a biotherapeutic mAb fusion protein for the treatment of rheumatoid arthritis
and other autoimmune diseases; it is also one of the highest revenue biotherapeutics on the market today.
Many biotechnology companies are actively working to creating biosimilar versions of etanercept.
In this study, we compared the 2-AB labeled N-glycan profile from one biosimilar candidate to that of the
innovator using the Glycan Application Solution with UNIFI and its FLR/MS workflow. We observed that the
biosimilar’s N-glycan profile is highly similar to that of the innovator’s, but some points of difference can
be detected.
For example, high mannose structures were observed at higher abundance in the biosimilar candidate,
including some extended mannose (e.g. Man 6 and Man 8) structures detected only in the biosimilar candidate
(Figure 6). We also observed that the biosimilar candidate contains the following glycans, F(6)A2[6]BG(4)1,
F(6)A2[3]G(4)1S(3)1, F(6)A3G(4)3S(3,3)2, and A2G(4)2S(6)1 in relative abundance that is greater than
0.1%, however, these sialylated glycans are either absent or below the 0.1% threshold in etanercept.
The cause of the N-glycan profile differences is most likely due to the variations in cell culture conditions.
Bioassays and clinical experience are likely required to establish the extent to which these differences
would affect the safety or efficacy of the biosimilar candidate.
Figure 6. A) Overlay chromatogram of the N-glycans from the innovator and a biosimilar candidate etanercept. The blue arrow
highlights the Man8 glycan that is only observed in the biosimilar candidate. B) The Waters Glycan GU Library search result shows that
the marked peak from (A) is assigned to Man8. The library search result for the highlighted peak, Man8, XIC of Man8 provides further
evidence for the correct structural assignment display in the chromatogram window.
[ 146 ]
N-Linked Glycan Characterization and Profiling: Combining the Power of Accurate Mass,
Reference Glucose Units, and UNIFI Software for Confident Glycan Assignments
Reporting
UNIFI Scientific Information System includes application-specific reporting templates. The default glycan
assay report templates provide a sound basis for reporting details on individual samples as well as
comparative data. The templates can also be readily modified to suit the reporting needs of a specific glycan
analysis project.
Figure 7 gives an example of the type of information captured by a generic glycan analysis report using the
innovator/biosimilar etanercept N-glycan analysis as an example.
Figure 7. An example of a UNIFI report for the N-glycan profiling and comparison between innovator/biosimilar etanercept glycans. Key information such as sample list,
retention time calibration curve, the scientific library search result for each identified glycan, summary table on relative % amount for innovator/biosimilar N-glyans,
UPLC-FLR/MS chromatograms, and component tables. The report template is customized to fit the analysis type.
[ 147 ]
N-Linked Glycan Characterization and Profiling: Combining the Power of Accurate Mass,
Reference Glucose Units, and UNIFI Software for Confident Glycan Assignments
Waters Corporation
34 Maple Street
Milford, MA 01757 U.S.A.
T: 1 508 478 2000
F: 1 508 872 1990
www.waters.com
Waters, The Science of What’s Possible, ACQUITY UPLC, UPLC, Xevo, and UNIFI are registered trademarks of Waters Corporation.
GlycoWorks is a trademark of Waters Corporation. All other trademarks are the property of their respective owners.
©2013–2016 Waters Corporation. Produced in the U.S.A. January 2016 720004845EN AG-PDF
CONC LUSIONS
Glycan characterization has remained a challenging aspect of biotherapeutic
characterization compared to techniques such as intact mass or peptide map
analysis, which most labs consider routine today. The addition of the glycan
UPLC-FLR/MS workflow and use of the experimentally derived Waters Glycan
GU Library within the Glycan Application Solution with UNIFI have addressed
the desire for compliant-ready, automated, high-confidence glycan structure
assignments by enabling rapid acquisition, review, and communication of
individual glycan profile results, and the larger sets of glycan analyses used
for comparability studies.
Additional capabilities with the Biopharmaceutical Platform Solution with UNIFI,
such as glycan/glycopeptide DDA MS/MS analysis and the ability to execute
exoglycosidase arrays, certainly complement this new workflow, providing
additional orthogonal results that enable the characterization of even the most
complex biotherapeutic glycoproteins.
[ 148 ]
N-Linked Glycan Characterization and Profiling: Combining the Power of Accurate Mass,
Reference Glucose Units, and UNIFI Software for Confident Glycan Assignments
WAT E RS SO LU T IONS
ACQUITY UPLC Glycan BEH Amide,
130Å, 1.7 μm Columns
XBridge® Glycan BEH Amide X P,
130Å, 2.5 µm Columns
ACQUITY UPLC H-Class Bio System
Rapi Fluor-MS Glycan Performance
Test Standard
K E Y WO R DS
HILIC Chromatography, UPLC, HPLC,
method transfer, N-glycans, RapiFluor-MS
A P P LIC AT ION BEN E FIT S
■
■
Seamless scalability and transfer of
RapiFluor-MS™ labeled glycan separations
between UPLC and HPLC instrumentation
■
■
ACQUITY UPLC® Glycan BEH Amide,
130Å, 1.7 μm and XBridge® Glycan
BEH Amide X P, 130Å, 2.5 μm Columns
provide high resolution UPLC® and HPLC
glycan separations
INT RO DUC T ION
In 2009, Waters introduced a revolutionary UPLC HILIC Column designed
specifically for achieving high resolution glycan separations. This column
technology was based on stationary phase constructed from 1.7 µm diameter,
130Å pore size, ethylene bridged hybrid (BEH) particles with an optimized amide
ligand bonding that has exhibited exceptional resolution for a broad range of
N-glycans ranging from small neutral structures to highly sialylated extended
structures.1 In addition to this UPLC-based column, Waters has also introduced
HPLC-based XBridge Glycan BEH Amide Columns based on 2.5 µm particles
and has demonstrated that these columns provide selectivity for 2-AB labeled
N-glycans comparable to that observed in UPLC separations.2 Most recently,
Waters has introduced a novel labeling reagent, RapiFluor-MS, that provides
both a fast and efficient sample preparation workflow and unsurpassed
fluorescent and MS sensitivity.3
In the following work, we demonstrate that Glycan BEH Amide Columns
packed with 1.7 μm and 2.5 μm particle sizes afford scalability between
RapiFluor-MS labeled glycan separations performed under UPLC and
HPLC-compatible conditions. Using standard LC method transfer principles to
account for differences in particle diameter (dp), column length, and column
internal diameter, we show that comparable chromatographic profiles and
relative quantitation can be achieved with the larger particle size column
at HPLC-compatible pressures, albeit with an increase in sample load,
mobile phase use, and most importantly, analysis time.
Transferring
RapiFluor-MS Labeled N-Glycan HILIC Separations
Between UPLC and HPLC
Stephan M Koza, Matthew A Lauber, and Kenneth J Fountain
Waters Corporation, Milford, MA, USA
[ 149 ]
E X P E RIM ENTA L
Time
UPLC/HPLC
(min)
Flow rate
UPLC/HPLC
(mL/min)
%A
(50 mM amm.
formate, pH 4.4)
%B
(Acetonitrile)
0.0/0.0
0.40/0.56
25
75
35.0/77.2
0.40/0.56
46
54
36.5/80.5
0.20/0.28
100
0
39.5/87.1
0.20/0.28
100
0
43.1/95.1
0.20/0.28
25
75
47.6/105.0
0.40/0.56
25
75
55.0/121.3
0.40/0.56
25
75
Sample vials:
Polypropylene 12 x 32 mm Screw Neck
Vial, with Cap and PTFE/silicone Septum,
300 µL Volume (p/n 186002640)
Data management:
MassLynx 4.1 Software
Empower Pro 3.0 Software
Sample description
The RapiFluor-MS Glycan Performance Test Standard
(p/n 186007983) was diluted in water to a concentration
of 20 pmole/μL.
Table 1.
Method conditions
LC system:
Alliance HPLC or
ACQUITY UPLC H-Class Bio System
Detection:
Alliance HPLC 2475 Fluorescence (FLR)
Detector ACQUITY UPLC FLR Detector with
analytical flow cell
Wavelength: 265 nm excitation,
425 nm emission
Columns:
ACQUITY UPLC Glycan BEH
Amide Column, 130Å, 1.7 µm,
2.1 mm x 150 mm, (p/n 186004742)
XBridge Glycan BEH Amide X P Column,
130Å, 2.5 µm, 3.0 x 150 mm
(p/n 186008040) and
XBridge Glycan BEH Amide XP Column,
130Å, 2.5 µm, 3 mm x 75 mm
(p/n 186008039) in series. Columns
connected by 0.005 x 1.75 UPLC SEC
Connection Tubing (p/n 186006613)
Column temp.:
60 °C
Sample temp.:
10 °C
Injection volume:
1.2 µL UPLC, 3.7 µL HPLC
Mobile phase A:
50 mM Waters Ammonium Formate
Solution – Glycan Analysis
(p/n 186007081), pH 4.4 (LC-MS-grade
water, from a 100X concentrate)
Mobile phase B:
LC-MS-grade acetonitrile
Gradients:
[ 150 ] Transferring RapiFluor-MS Labeled N-Glycan HILIC Separations Between UPLC and HPLC
R E SU LT S AND DIS CUSSION
Calculating the transfer of glycan HILIC methods between UPLC and HPLC columns
There are two primary considerations to be made when transferring a HILIC-based N-glycan separation
method from one LC system and column to another. Most importantly, the surface chemistry and pore size
of the particles in the two columns must be comparable. Once appropriate columns have been chosen, the
separation must then be appropriately scaled with respect to particle size. Generally this can be accomplished
by maintaining a comparable ratio between the length of the column and the size of the particle, L/dp. Once
determined, alterations to the gradient can be calculated. In this example, the transfer between a 1.7 µm
particle size, 2.1 mm x 150 mm ACQUITY UPLC Glycan BEH Amide Column to an XBridge Glycan BEH Amide
Column with a 2.5 µm particle size required a column approximately 50% greater in length (225 mm) since
the ratio of the particle sizes is 1.47 (i.e. 2.5/1.7). In practice, a 225 mm length can be easily constructed by
combining 150 mm and 75 mm length columns with a suitable column connector. In addition to column length,
it is also important to consider the optimal column I.D. HPLC systems invariably exhibit higher dispersion than
UPLC systems (bandspread ~30 μL versus ~10 μL), so it is advisable to perform separations with relatively
larger I.D. columns to ensure that the effect of extra-column band broadening is minimized. With a 3.0 mm
HPLC column I.D. format, near optimal resolution separations can be achieved on an HPLC system, without the
high mobile phase consumption rates typical of 4.6 mm I.D. column formats.
Having selected a 3.0 mm x 225 mm effective column dimension, we next calculated the appropriate
gradient for the HPLC separation using general method transfer principals (refer back to Table 1 for the
gradient).4 Table 2 outlines column lengths, analysis times, and mobile phase as well as sample consumption
corresponding to the use of various scaled methods and potential Glycan BEH Amide Column configurations.
Clearly, this exercise highlights two of the significant advantages that UPLC separations provide: shorter analysis
times (≥55% decrease) and decreased mobile phase usage (≥68% decrease). The UPLC separations also benefit
from lower required sample loads (≥68% decrease), which can prove useful in cases where an analyst is sample
limited. For these comparisons, mobile phase use was determined based on the gradient shown in Table 1. Based
on these calculated results, the advantages of the UPLC format is evident as is the use of the XBridge Glycan BEH
Amide X P Column, 2.5 µm, 3 mm I.D. Columns on an low band spread (29 µL) HPLC.
Particle size
(µm)
Column length
(mm)
Column I.D.
(mm)
Flow rate
(mL/min)
Run time
(min)
Mobile phase
(mL)
Sample
(µL)
1.7
150
2.1
0.4
55
20
1.2
2.5
225
3.0
0.56
121.3
62
3.7
2.5
225
4.6
1.32
121.3
146
8.8
3.5
300
4.6
0.93
229.2
194
11.6
Table 2. Particle size, column lengths, flow rate, analysis times, and mobile phase as well as sample consumption corresponding to
the use of various scaled methods and potential Glycan BEH Amide Column configurations.
[ 151 ]
Transferring RapiFluor-MS Labeled N-Glycan HILIC Separations Between UPLC and HPLC
Figure 1. Comparison of UPLC and HPLC HILIC separations of RapiFluor-MS labeled N-glycans from the Glycan Performance Test
Standard (p/n 186007983) diluted in water to a concentration of 20 pmole/µL. Injection volumes of 1.2 µL and 3.7 µL for the UPLC
and HPLC analyses.
Comparing UPLC and HPLC Rapi Fluor-MS labeled N-glycan profiles
The effectiveness of scaling from a 2.1 x 150 mm, 1.7 µm particle size, Glycan BEH Amide Column using
an ACQUITY H-Class UPLC System to a total 225 mm length (150 mm + 75 mm) 2.5 µm particle size,
3.0 mm I.D., XBridge Column run on an Alliance HPLC System is demonstrated qualitatively in Figure 1.
Both pairs of chromatograms show comparable profiles over normalized time ranges for the RapiFluor-MS
Glycan Performance Test Standard (p/n 186007983), which represents the N-glycans released from a pooled
human IgG sample. In this example, the analysis time difference is approximately 2.2-fold.
30.00
10.00
12.00 14.00 16.00 18.00 20.00 22.00 24.00 26.00 28.00
64.00
24.00
28.00
32.00
36.00
40.00
44.00
48.00
52.00
56.00
60.00
Alliance HPLC
XBridge Glycan BEH Amide XP, 130Å, 2.5 m
3.0 mm x 225 mm total length (150 mm + 75 mm)
ACQUITY UPLC H-Class Bio
ACQUITY UPLC Glycan BEH Amide, 130Å, 1.7 m
2.1 mm x 150 mm
Total Run Time = 55 min
Total Run Time = 121.3 min
Fluorescence (relative)
Time (min)
1
2
3
4 5
6
7
8
9 10
11
12
13
14 15
16
17
18
19
20 21
[ 152 ] Transferring RapiFluor-MS Labeled N-Glycan HILIC Separations Between UPLC and HPLC
These separations were also compared
quantitatively. Shown in Figure 2 is a comparison
of the relative retention times for 21 of the most
abundant N-linked oligosaccharides observed.
Relative retention times were calculated off of
Peak 1 (Figure 1) and corrected for the increased
system and column dwell volumes of the HPLC
separation (~1.2 minutes). Figure 3 illustrates the
general comparability of the two separations with
respect to the relative quantitation for the same
21 peaks evaluated for retention time. The majority
of these values are well within 5% of each other
with the most significant difference (~35%) being
observed for Peak 19, which has a relative abundance
of ~0.2% as determined by the HPLC analysis.
If more precise quantitation of these low abundance
species is required it would be advantageous to
report these results relative to a reference material.
Overall, these data demonstrate that the HILIC-based
separation of Rapi Fluor-MS labeled glycans can be
readily transferred between UPLC and HPLC formats.
The comparability of the observed chromatographic
profiles underscores the chemical comparability of
the particle surfaces, as well as the comparability
in pore characteristics.
1
1.2
1.4
1.6
1.8
2
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15 16 17 18 19 20 21
Relative Retention Time
Peak Number
UPLC
HPLC
Figure 2. Comparison of UPLC and HPLC HILIC relative retention times (n=2) of RapiFluor-MS
labeled N-glycans. Peak numbers are as labeled in Figure 1. Relative retention times were
determined based on Peak 1.
Figure 3. Comparison of UPLC and HPLC HILIC relative peak areas (n=2) of RapiFluor-MS labeled
N-glycans. Peak numbers are as labeled in Figure 1.
0
5
10
15
20
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15 16 17 18 19 20 21
Percent Peak Area (FLR)
Peak Number
UPLC
HPLC
[ 153 ]
Transferring RapiFluor-MS Labeled N-Glycan HILIC Separations Between UPLC and HPLC
Waters Corporation
34 Maple Street
Milford, MA 01757 U.S.A.
T: 1 508 478 2000
F: 1 508 872 1990
www.waters.com
Waters, The Science of What’s Possible, UPLC, ACQUITY UPLC, XBridge, Alliance, MassLynx, and Empower are registered trademarks of
Waters Corporation. RapiFluor-MS is a trademark of Waters Corporation. All other trademarks are the property of their respective owners.
©2015 Waters Corporation. Produced in the U.S.A. December 2015 720005344EN AG-PDF
CONC LUSIONS
These results demonstrate that a HILIC separation of RapiFluor-MS labeled
N-linked oligosaccharides can be seamlessly transferred between UPLC and
HPLC platforms when using the appropriate Glycan BEH Amide Columns. The
advantage in using the UPLC-based separation is the capability to dramatically
improve sample throughput while decreasing mobile phase use. Sample load
requirements are also lowered. However, in the event that a laboratory encounters
instrumentation limitations, it is beneficial to be able to easily transfer between
UPLC and HPLC separations. Additionally, scaling from a UPLC to an HPLC
platform can be useful if glycans must be fractionated and purified for structural
analysis or to generate materials for method validation spiking studies.
References
1. Beth Gillece-Castro, Kim van Tran, Jonathan E. Turner,
Thomas E. Wheat, Diane M. Diehl, N-Linked Glycans of
Glycoproteins: A New Column for Improved Resolution.
Waters Application Note 720003112en. 2009.
2. Matthew A. Lauber, Stephan M. Koza, Jonathan E. Turner,
Pamela C. Iraneta, and Kenneth J Fountain, Amide-Bonded
BEH HILIC Columns for High Resolution, HPLC-Compatible
Separations of N-Glycans. Waters Application Note
720004857en. 2013.
3. Matthew A. Lauber, Darryl W. Brousmiche, Zhengmao Hua,
Stephan M. Koza, Ellen Guthrie, Paula Magnelli, Christopher
H. Taron, Kenneth J. Fountain, Rapid Preparation of Released
N-Glycans for HILIC Analysis Using a Novel Fluorescence
and MS-Active Labeling Reagent, Waters Application Note
720005275en. 2015.
4. Neue, U. D.; McCabe, D.; Ramesh, V.; Pappa, H.; DeMuthc, J.
In Transfer of HPLC procedures to suitable columns of
reduced dimensions and particle sizes, Pharmacopeial Forum,
2009; pp 1622–1626.
[ 154 ] Transferring RapiFluor-MS Labeled N-Glycan HILIC Separations Between UPLC and HPLC
RELEASED N- AND O-LINKED GLYCANS
[ 155 ]
Rajasekar R. Prasanna, Rajiv Bharadwaj, Matthew A. Lauber, Stephan M. Koza, and Erin E. Chambers
Waters India Pvt. Ltd, Bangalore, India
Waters Corporation, Milford, MA, USA
Figure 1. HILIC-FLR-MS of RapiFluor-MS labeled N-glycans. N-glycans from 0.5 µg of an rhEPO were
separated using a 2.1 x 150 mm, ACQUITY UPLC® Glycan BEH Amide, 130Å, 1.7 µm Column, mobile
phases comprised of 50 mM ammonium formate (pH 4.4) (A) and acetontrile (B), and a column
temperature of 60 °C. The separation was performed using a 35 minute gradient from 25% A
to 46% A. Additional details on the method can be found in the GlycoWorks RapiFluor-MS
Care and Use Manual (p/n 715004793).
Erythropoeitin_Rapifluor
%
0
100
EU x 10
e4
0.000
25000.002
50000.004
75000.008
100000.008
125000.008
150000.016
175000.016
200000.016
225000.016
250000.016
275000.031
FA
2G2S2
FA
3G3S3
FA
4G4S4
(1Acetylation)
FA
4G4S3
FA
4G4S4
FA
4G4Lac1S3
FA
4G4Lac1S4
FA
4G4Lac2S3
FA
4G4Lac2S4
FA
4G4Lac3S3
FA
4G4Lac3S4
FA
4G4Lac4S4
FA
4G3S3
(1Acetylation)
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38 min
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38 min
MS trace
FLR trace
G OA L
To elucidate the complex N-glycans from
erythropoietin using the GlycoWorks™
RapiFluor-MS™ N-Glycan Kit and hydrophilic
interaction chromatography (HILIC).
BAC KG RO U N D
Erythropoietin (EPO) is a highly glycosylated
protein hormone that stimulates the production
of red blood cells. EPO exhibits significant
heterogeneity due to its multiple sites of
glycosylation (three N-glycosylation sites at
Asn 24, 38 and 83 and one O-glycosylation
site at Ser 126) and the fact that each of
these sites can bear various highly branched
sialylated N-glycan structures.1-2 As a
consequence of these post-translational
modifications, an EPO will have an apparent
SDS-PAGE molecular weight between 30
and 40 kDa, 40% of which corresponds
to glycan content. Not surprisingly, the
glycosylation of an EPO has been found to
impact its therapeutic characteristics, namely
its stability, efficacy and potency.1, 3-5 In vivo
studies using glyco-engineered EPO have
shown links between the safety and efficacy of
a therapeutic EPO and several glycosylation
associated critical quality attributes (CQA),
perhaps with sialic acid content being the most
important. Sialylated oligosaccharides have
been shown to be associated with increased
half-life in plasma as compared to desialylated
forms which tend to be cleared within
minutes.1, 3-5
The analysis of EPO N-glycans is facilitated by the
high fluorescence and MS sensitivity afforded by
RapiFluor-MS labeling.
Characterization of EPO N-Glycans
using RapiFluor-MS and HILIC Profiling
Recombinant human EPO (rhEPO) expressed using Chinese hamster ovary
(CHO) cells has been used efficiently in the treatment of anemia, since
Epogen was approved by the FDA in 1989.6 As patents for EPO therapeutics
approach expiration, the market for biosimilar rhEPO is expected to increase
exponentially. Accordingly, there is a need for efficient and accurate methods
that can be used for the characterization of EPO glycosylation.
[ 156 ]
Figure 2. Mass spectrometric analysis of the RapiFluor-MS labeled FA4G4Lac4S4 glycan eluting at retention time 36.94 minutes.
(A) Background subtracted ESI-MS mass spectrum showing an [M+4H]4+ ion of FA4G4Lac4S4. (B) Charge deconvoluted and deisotoped
mass spectrum. (C) Diagram of the identified FA4G4Lac4S4 glycan. (Proposed structure based on a previous study of Lac repeats.7)
02032015_RapiFluor_EPO_001
mass
5430
5440
5450
5460
5470
%
0
100
5454.0273
5425.9600 5434.9634
5455.0659
5457.9824
5466.9751
02032015_RapiFluor_EPO_001
1350 1355 1360 1365 1370 1375 1380 1385 1390 1395 1400 1405 1410 1415 1420 1425 1430 1435 1440 1445 1450 m/z
%
0
100
1364.7704
1364.5137
1364.2721
1369.2742
1369.5314
1376.2714
1376.5292
1381.2650 1385.2483
1387.5161
1393.0488 1394.9719
FA4G4Lac4S4
B
A
C
[M+4H]4+
[M+3H+NH4]4+
02032015_RapiFluor_EPO_001
1364
1365
1366
1367 m/z
0
100
1364.7704
1364.5137
1364.2721
1365.0120
1365.2687
1365.5255
1365.7672
1366.0240 1366.2657
Zoomed View of m/z 1364.7704 [M+4H]4+
%
FA4G4Lac4S4
T H E S O LU T IO N
N-glycans from rhEPO were released through fast enzymatic deglycosylation and rapidly labeled using a Waters
GlycoWorks RapiFluor-MS N-Glycan Kit (p/n 176003635).7 The new RapiFluor-MS reagent has been designed to
facilitate rapid labeling, improve fluorescence quantum yields and greatly enhance MS sensitivity.7 In this sample
preparation, the complex N-glycans of EPO were first made accessible for enzymatic deglycosylation by the use of
RapiGest™ SF, an anionic surfactant. Subsequently, its N-glycans were released in approximately 5 minutes using
Rapid PNGase F and an elevated incubation temperature of 50 °C. The resulting deglycosylation mixture, containing
free N-glycans (glycosylamines), was then subjected to a 5 minute labeling reaction with RapiFluor-MS. Labeled
N-glycans were thereafter efficiently extracted from the reaction mixture using a GlycoWorks HILIC µElution plate
(p/n 186002780) and GlycoWorks SPE Elution Buffer (p/n 186007992). This process of going from glycoprotein
to extracted, labeled N-glycans was accomplished in 30 minutes. In addition, this sample preparation allowed for
the immediate analysis of the RapiFluor-MS labeled N-glycans via a HILIC separation with a 2.1 mm x 150 mm,
ACQUITY UPLC Glycan BEH Amide, 130Å, 1.7 µm Column (p/n 186004742) and an ACQUITY UPLC I-Class System.
RapiFluor-MS N-glycan species eluting during these separations were serially detected by their fluorescence (FLR)
and by positive ion-mode ESI-MS with a Xevo® G2-S QTof Mass Spectrometer.
Figure 1 presents chromatograms from the HILIC-FLR-MS analysis of EPO N-glycans as labeled with RapiFluor-MS.
Notably, both the fluorescence and base peak intensity (BPI) MS chromatograms showed high signal-to-noise such
that the presence of different N-glycan species could be readily confirmed. The major N-glycans species in this profile
were identified using the accurate mass information in combination with data from previous observations of EPO
N-glycans.7 Previously, multidimensional chromatography strategies combining anion exchange chromatography and
HILIC had been required to comprehensively characterize the N-glycans of EPO.7 In this work, we have been successful in
identifying EPO N-glycans by employing a one dimensional HILIC separation along with online ESI-Q-Tof MS detection.
This is an approach that is facilitated by the improved fluorescence and MS sensitivity afforded by RapiFluor-MS
labeling.8 These new developments in N-glycan analysis aided in identifying tetra-antennary glycans with multiple
sialic acids (three and four) as the most abundant species present on the analyzed rhEPO. The GlycoWorks RapiFluor-MS
approach also helped in determining that tetra-antennary glycans with poly-N-acetyl lactosamine extensions were
[ 157 ]
Characterization of EPO N-Glycans using RapiFluor-MS and HILIC Profiling
Waters Corporation
34 Maple Street
Milford, MA 01757 U.S.A.
T: 1 508 478 2000
F: 1 508 872 1990
www.waters.com
also present in relatively high abundance. Relative
quantification from the fluorescence profile, in
fact, showed that that the tetra-sialylated, tetra-
antennary glycan species (FA4G4S4) represents
approximately 20% of the total N-glycan
pool, while tetra-antennary glycans carrying
one (FA4G4Lac1S4) and two (FA4G4Lac2S4)
lactosamine extensions constitute approximately
12 and 4.5% of the total N-glycans, respectively.
More interestingly, GlycoWorks RapiFluor-MS
approach has yielded identifications of tetra-
antennary structures with three (FA4G4Lac3S4)
and four (FA4G4Lac4S4) lactosamine extensions
at levels of 0.75% and 0.25%, respectively.
Although some previous studies on EPO have
reported one and two N-acetyl lactosamine
extensions, few studies have reported detailed
information on species containing four or more
lactosamine repeats.7 In this work, we have been
able to successfully identify up to four repeats
of poly-N-acetyl lactosamine using only a single
dimension of separation and a gradient time of just
35 minutes (Figure 2). Moreover, it was possible
to make confident identifications throughout this
HILIC profile because of the enhanced fluorescence
yields and the improvements in the ionization
efficiencies of complex N-glycans that result from
the use of the novel RapiFluor-MS labeling reagent.
S UMM A RY
An approach combining the advantages of GlycoWorks RapiFluor-MS N-glycan
sample preparation with the separation capabilities of UPLC HILIC has enabled
us to perform a comprehensive analysis of the complex N-glycans present on
a recombinant human erythropoietin (rhEPO). With the GlycoWorks RapiFluor-
MS workflow, N-glycan samples were prepared in just 30 minutes. Most
importantly, the samples were amenable to direct analysis by HILIC-ESI-QTof-
MS analysis. RapiFluor-MS labeling not only reduced the burden of the sample
preparation, but also enhanced the sensitivity of N-glycan detection, making
it possible to obtain information-rich data and to elucidate the complicated
N-glycan profile of an rhRPO. Because of these benefits, this new approach to
N-glycan analysis could be used to hasten the development of EPO biosimilars.
References
1. Dube, S.; Fisher, J. W.; Powell, J. S., Glycosylation at specific sites of erythropoietin is essential
for biosynthesis, secretion, and biological function. J Biol Chem 1988, 263 (33), 17516–21.
2. Sasaki, H.; Bothner, B.; Dell, A.; Fukuda, M., Carbohydrate structure of erythropoietin expressed
in Chinese hamster ovary cells by a human erythropoietin cDNA. J Biol Chem 1987, 262 (25),
12059–76.
3. Hashii, N.; Harazono, A.; Kuribayashi, R.; Takakura, D.; Kawasaki, N., Characterization
of N-glycan heterogeneities of erythropoietin products by liquid chromatography/mass
spectrometry and multivariate analysis. Rapid Commun Mass Spectrom 2014, 28 (8), 921–32.
4. Takeuchi, M.; Inoue, N.; Strickland, T. W.; Kubota, M.; Wada, M.; Shimizu, R.; Hoshi, S.;
Kozutsumi, H.; Takasaki, S.; Kobata, A., Relationship between sugar chain structure and
biological activity of recombinant human erythropoietin produced in Chinese hamster ovary
cells. Proc Natl Acad Sci USA 1989, 86 (20), 7819–22.
5. Goldwasser, E.; Kung, C. K.; Eliason, J., On the mechanism of erythropoietin-induced
differentiation. 13. The role of sialic acid in erythropoietin action. J Biol Chem 1974, 249 (13),
4202–6.
6. Bennett, C. L.; Spiegel, D. M.; Macdougall, I. C.; Norris, L.; Qureshi, Z. P.; Sartor, O.; Lai,
S. Y.; Tallman, M. S.; Raisch, D. W.; Smith, S. W.; Silver, S.; Murday, A. S.; Armitage, J. O.;
Goldsmith, D., A review of safety, efficacy, and utilization of erythropoietin, darbepoetin, and
peginesatide for patients with cancer or chronic kidney disease: a report from the Southern
Network on Adverse Reactions (SONAR). Semin Thromb Hemost 2012, 38 (8), 783–96.
7. Bones, J.; McLoughlin, N.; Hilliard, M.; Wynne, K.; Karger, B. L.; Rudd, P. M., 2D-LC analysis
of BRP 3 erythropoietin N-glycosylation using anion exchange fractionation and hydrophilic
interaction UPLC reveals long poly-N-acetyl lactosamine extensions. Anal Chem 2011, 83(11),
4154–62.
8. Lauber, M. A.; Yu, Y. Q.; Brousmiche, D. W.; Hua, Z.; Koza, S. M.; Magnelli, P.; Guthrie, E.; Taron,
C. H.; Fountain, K. J., Rapid Preparation of Released N-Glycans for HILIC Analysis Using a
Labeling Reagent that Facilitates Sensitive Fluorescence and ESI-MS Detection. Anal Chem
2015, 87 (10), 5401–9.
Waters, The Science of What’s Possible, ACQUITY UPLC, and Xevo are registered trademarks of Waters Corporation. GlycoWorks, RapiGest,
and RapiFluor-MS are trademarks of Waters Corporation. All other trademarks are the property of their respective owners.
©2015 Waters Corporation. Produced in the U.S.A. July 2015 720005444EN KP-PDF
[ 158 ] Characterization of EPO N-Glycans using RapiFluor-MS and HILIC Profiling
WAT E R S S O LU T IO NS
ACQUITY UPLC Glycoprotein BEH Amide,
300Å Column (patent pending)
Glycoprotein Performance Test Standard
GlycoWorks™ RapiFluor-MS™ N-Glycan Kit
ACQUITY UPLC H-Class Bio System
Xevo® G2-XS QTof
Mass Spectrometer
SYNAPT® G2-S HDMS
K E Y W O R D S
ACQUITY UPLC H-Class Bio System, BEH
Amide 300Å, glycans, glycosylated
protein, glycosylation, O-Linked,
N-Linked, HILIC, RapiFluor-MS Labeling
A P P L I C AT IO N B E N E F I T S
■
■
Two facile strategies to elucidate
information about both the N and
O-linked glycosylation of EPO
■
■
Unprecedented HILIC separations of high
antennarity released N-glycans and intact
protein glycoforms
■
■
MS compatible HILIC to enable detailed
investigations of sample constituents
■
■
ACQUITY UPLC® Glycoprotein BEH Amide
Column (300Å, 1.7 µm stationary phase)
is QC tested via a glycoprotein
separation to ensure consistent
batch-to-batch reproducibility
I N T RO DU C T IO N
The immunoglobulin G (IgG) modality has paved the way for many efficacious
protein-based therapies.1 At the same time, numerous highly effective patient
therapies have also been made possible by the production of recombinant,
human hormones and enzymes. For example, erythropoesis stimulating
therapeutics, like epoetin (EPO) alpha, have long been available for the
treatment of anemia. Such a therapy for increasing patient red blood cell counts
was first made possible by the commercialization of Epogen,® which has been
available in the US market since its approval by the FDA in 1989.2 And now,
because the landscape of the biopharmaceutical industry continues to evolve
and Epogen patents expired in 2013,3 EPO drug products are targets for being
developed into both international and domestic-market biosimilars.
Epoetin alpha has a relatively small primary structure, yet it has 3 sites of
N-glycosylation and 1 site of O-glycosylation (Figure 1).4 Because of its
glycosylation, epoetin alpha has a molecular weight between 30 and 40
kDa even though its protein mass amounts to only 18 kDa. Interestingly, the
glycosylation of epoetin is very much tied to its potency and serum half life.
Two attributes of its glycan profile that are known to show positive correlations
with in vivo activity include antennarity and sialylation.5-7 As a result, it is
critical for the glycosylation of an epoetin therapeutic to be well characterized.
In addition, the significance of epoetin glycosylation suggests that detailed
glycan profiling would be a path toward establishing a viable epoetin biosimilar.
Comprehensive Characterization of the N and O-Linked Glycosylation
of a Recombinant Human EPO
Matthew A. Lauber, Stephan M. Koza, and Erin E. Chambers
Waters Corporation, Milford, MA, USA
APPRLICDSR
VLERYLLEAK
EAENITTGCA
EHCSLNENIT
VPDTKVNFYA
WKRMEVGQQA VEVWQGLALL SEAVLRGQAL LVNSSQPWEP LQLHVDKAVS
GLRSLTTLLR
ALGAQKEAIS
PPDAASAAPL RTITADTFRK
LFRVYSNFLR
GKLKLYTGEA
CRTGD
N-Linked Glycans
O-Linked Glycan
O-Linked
Glycan
N-Linked
Glycans
Figure 1. Sequence and structural information for recombinant, human epoetin alpha (rhEPO).
[ 159 ]
E X P E R IM E N TA L
Sample description
A recombinant, human epoetin alpha expressed from CHO cells
(PeproTech, Rocky Hill, NJ) was reconstituted in 50 mM HEPES
NaOH pH 7.9 buffer to a concentration of 2 mg/mL.
N-glycans were released from rhEPO and labeled with
RapiFluor-MS using a GlycoWorks RapiFluor-MS N-Glycan Kit
and the instructions provided in its care
and use manual (p/n 715004793). RapiFluor-MS labeled
N-glycans were injected as a mixture of 90 µL SPE eluate,
100 µL dimethylformamide, and 210 µL acetonitrile.
To facilitate analysis of O-glycosylation, rhEPO was
N-deglycosylated using the rapid deglycosylation technique
outlined in the care and use manual of the GlycoWorks
RapiFluor-MS N-Glycan Kit (p/n 715004793).
Method conditions (unless otherwise noted)
Column Conditioning
New (previously unused) ACQUITY UPLC Glycoprotein BEH Amide,
300Å, 1.7 µm Columns should be conditioned via two or more
sequential injections and separations until a consistent profile is
achieved. The care and use manual of the column can be referred
to for more information (p/n 720005408EN).
LC conditions for RapiFluor-MS Released N-Glycans
LC system:
ACQUITY UPLC H-Class Bio System
Sample temp.:
10 °C
Analytical
column temp.:
60 °C
Flow rate:
0.4 mL/min
Injection volume:
10 µL
Column:
ACQUITY UPLC Glycoprotein BEH
Amide, 300Å, 1.7 µm, 2.1 x 150 mm
(p/n 176003702, with Glycoprotein
Performance Test Standard)
Fluorescence detection: Ex 265 nm / Em 425 nm, 2 Hz
Sample collection/
Vials:
Sample Collection Module
(p/n 186007988)
Polypropylene 12 x 32 mm Screw Neck
vial, 300 µL volume (p/n 186002640)
Mobile phase A:
50 mM ammonium formate, pH 4.4
(LC-MS grade; from a 100x concentrate,
p/n 186007081)
Mobile phase B:
ACN (LC-MS grade)
Flow Rate
Time (mL/min)
%A
%B
Curve
0.0 0.4 25 75 6
35.0 0.4
46 54
6
36.5 0.2 100 0
6
39.5 0.2 100 0
6
43.1 0.2
25 75
6
47.6 0.4
25 75
6
55.0 0.4
25 75
6
MS conditions for RapiFluor-MS Released N-Glycans
MS system:
Xevo G2-XS QTof
Ionization mode:
ESI+
Analyzer mode:
Resolution (~40 K)
Capillary voltage:
2.2 kV
Cone voltage:
75 V
Source temp.:
120 °C
Desolvation temp.:
500 °C
Source offset:
50 V
Desolvation gas flow: 600 L/Hr
Calibration:
NaI, 1 µg/µL from 100–2000 m/z
Acquisition:
700–2000 m/z, 0.5 sec scan rate
Lockspray:
300 fmol/µL Human glufibrinopeptide B
in 0.1% (v/v) formic acid, 70:30 water/
acetonitrile every 90 seconds
Data management:
MassLynx® Software v4.1
LC conditions for Intact Protein HILIC of N-Deglycosylated
rhEPO
LC system:
ACQUITY UPLC H-Class Bio System
Sample temp.:
10 °C
Analytical
column temp.:
45 °C
Flow rate:
0.2 mL/min
Fluorescence detection: Ex 280 nm/Em 320 nm
(Intrinsic fluorescence), 10 Hz
[ 160 ] Comprehensive Characterization of the N and O-Linked Glycosylation of a Recombinant Human EPO
Mobile phase A:
0.1% (v/v) TFA, H2O
Mobile phase B:
0.1% (v/v) TFA, ACN
HILIC injection volume: 1.3 µL (A 2.1 mm I.D. HILIC column can
accommodate up to an ~1 µL aqueous
injection before chromatographic
performance is negatively affected)
Columns:
ACQUITY UPLC Glycoprotein BEH Amide,
300Å, 1.7 µm, 2.1 x 150 mm Column
(p/n 176003702, with Glycoprotein
Performance Test Standard)
Vials:
Polypropylene 12 x 32 mm Screw Neck,
300 µL volume (p/n 186002640)
Gradient:
Time
%A
%B
Curve
0.0 15.0 85.0 6
0.5 15.0 85.0 6
1.0 25.0 75.0 6
21.0 35.0 65.0 6
22.0 100.0 0.0
6
24.0 100.0 0.0
6
25.0 15.0 85.0 6
35.0 15.0 85.0 6
MS conditions for for Intact Protein HILIC
of N-Deglycosylated rhEPO
MS system:
SYNAPT G2-S HDMS
Ionization mode:
ESI+
Analyzer mode:
Resolution (~20 K)
Capillary voltage:
3.0 kV
Cone voltage:
45 V
Source offset:
50 V
Source temp.:
150 °C
Desolvation temp.:
500 °C
Desolvation gas flow: 800 L/Hr
Calibration:
NaI, 1 µg/µL from 500–5000 m/z
Acquisition:
700–4800 m/z, 1 sec scan rate
Data management:
MassLynx Software v4.1
In this application note, we demonstrate the use of two facile
strategies that can be used to detail the N and O-linked
glycosylation of a recombinant, human epoetin (rhEPO). In this
work, rhEPO N-glycans were rapidly released, labeled with
GlycoWorks RapiFluor-MS and profiled by hydrophilic interaction
chromatography (HILIC) using sensitive fluorescence and mass
spectrometric detection. Then, in a second, parallel analysis,
N-deglycosylated rhEPO was interrogated by intact protein HILIC
to elucidate information on O-glycosylation.
R E SU LT S A N D D IS C U S S IO N
Released N-Glycan analysis of rhEPO using RapiFluor-MS
labeling and HILIC profiling
The glycosylation of recombinant, human epoetin (rhEPO) has
been investigated many times before.4-5, 8-13 In large part, these
previous studies have required relatively involved techniques.
With this work, it was our objective to establish two facile and
complementary, LC based approaches for the analysis of EPO,
one capable of providing information about N-glycosylation
and the other information about O-glycosylation.
A profile of the N-glycans from rhEPO can be readily obtained
with a new sample preparation strategy involving the novel
glycan labeling reagent, RapiFluor-MS. This sample preparation,
based on the GlycoWorks RapiFluor-MS N-Glycan Kit, allows
an analyst to rapidly release N-glycans and label them with
a tag that provides enhanced sensitivity for fluorescence and
electrospray ionization mass spectrometric (ESI-MS) detection.14
In previous applications, RapiFluor-MS has been predominately
used in the analysis of different IgG samples.14-16 Nevertheless,
using the protocol from the GlycoWorks RapiFluor-MS N-Glycan
Kit, an analyst can successfully prepare samples from even
heavily glycosylated proteins, such as rhEPO.
RapiFluor-MS labeled N-glycans have proven to be amenable to
hydrophilic interaction chromatography (HILIC). Accordingly,
HILIC-fluorescence-MS of RapiFluor-MS has emerged as a very
powerful tool for detailing the N-glycosylation of proteins.14
[ 161 ]
Comprehensive Characterization of the N and O-Linked Glycosylation of a Recombinant Human EPO
To this end, a sample of RapiFluor-MS N-glycans derived from rhEPO was profiled using HILIC. A recently
introduced widepore amide column, the ACQUITY UPLC Glycoprotein BEH Amide, 300Å, 1.7µm Column, was
selected for this work to obtain high resolution N-glycan separations. This column was purposefully designed
to facilitate HILIC separations of large molecules, such as glycopeptide and glycoproteins. However, the
widepore particle architecture has also been shown to increase the peak capacity of highly branched, tri- and
tetra-antennary N-glycans by 10–20%,17 making it an ideal choice for the HILIC profiling of EPO N-glycans,
which typically exhibit high antennarity. Figure 2A shows the HILIC fluorescence and base peak intensity
(BPI) MS chromatograms of the RapiFluor-MS N-glycans resulting from 0.4 µg of rhEPO. Even with this
relatively limited amount of sample, high signal-to-noise chromatograms are obtained. The sensitivity of
the fluorescence trace allows for accurate, relative quantitation across the profile. The signal-to-noise of the
MS chromatogram is also particularly noteworthy, though it should be noted that MS sensitivity decreases as
N-glycan structures become larger. Nevertheless, the quality of these particular data is made possible by use
of the RapiFluor-MS reagent in combination with the Xevo G2-XS QTof, a new generation MS instrument with
improved transmission efficiency and sensitivity. This QTof technology provides unprecedented sensitivity
as well as high mass resolution, as can be observed in the collection of mass spectra in Figure 2B that have
been used to support the assignment of various N-glycan species.
0E+0
2E+5
10
15
20
25
30
35
0E+0
3E+6
N-Linked
Glycans
FA4G4Lac2S4
FA4G4Lac1S4
FA4G4S4
FA4G4Lac1S4+Ac
FA2G2S2
FA4G4Lac4S4
FA4G4Lac3S4
FA4G4Lac3S3
FA4G4Lac2S3
FA4G4Lac1S3
FA3G3S3
FA2G2S1
FA4G4S4+Ac
FA4G4S3
FA2G2S2
FA4G4Lac2S4
FA4G4Lac4S4
FA4G4S4+Ac
891.3492, 3+
-3.8 ppm
1345.8372, 3+
-2.1 ppm
1181.6909, 4+
0.1 ppm
1364.2560, 4+
0.9 ppm
890
895
m/z
1344
1349
m/z
1180
1185
m/z
1363
1368
m/z
Fluorescence
Intensity
BPI
FLR
A
B
C
Figure 2. HILIC profiling of released N-glycans from rhEPO. (A) Fluorescence and (B) base peak intensity (BPI) chromatograms for
RapiFluor-MS labeled N-glycans from rhEPO. Chromatograms obtained for glycans from 0.4 µg protein using an ACQUITY UPLC
Glycoprotein BEH Amide, 300Å, 1.7 µm, 2.1 x 150 mm Column. (C) MS spectra for four example N-glycan species. N-glycan
assignments are listed according to Oxford notation. “+Ac” denotes an acetylation, such as the previously reported O-acetylation
of sialic acid residues (Neu5Ac).8
[ 162 ] Comprehensive Characterization of the N and O-Linked Glycosylation of a Recombinant Human EPO
The chromatographic and MS-level selectivity afforded by this analysis simplifies making N-glycan
assignments such that the species of the rhEPO N-glycan profile were easily mapped (Figure 3).
The rhEPO analyzed in this study exhibits an N-glycan profile comprised primarily of tetra-antennary,
tetrasialylated N-glycans (FA4G4S4) with varying N-acetyl lactosamine extensions. However, the profile also
shows a highly abundant peak that corresponds to a disialylated, biantennary N-glycan (FA2G2S2). Given that
the ratio of tetra-antennary to biantennary N-glycans has a positive correlation with the in vivo activity of an
EPO,6 this analysis has clearly produced valuable information. Other information that can be readily obtained
from this N-glycan analysis includes the degree of sialylation and the extent to which structures are modified
with lactosylamine extensions. Overall, these results demonstrate that a very-information rich N-glycan profile
can indeed be obtained from a comparatively simple RapiFluor-MS N-glycan preparation and a corresponding
HILIC-fluorescence-MS analysis.
RT
(min)
Species
MWMono, Theo
(Da)
Observed
m/z
z
MWMono, Obs
(Da)
Mass error
(ppm)
16.21
FA2G2S1
2388.9201
1195.4659
2
2388.9172
1.2
18.12
FA2G2S2
2680.0155
894.3492
3
2680.0258
-3.8
22.24
FA3G3S3
3336.2432
1113.0924
3
3336.2554
-3.7
23.68
FA4G4S4 + Ac
4034.4813
1345.8372
3
4034.4898
-2.1
24.15/24.60
FA4G4S3
3701.3754
1234.7966
3
3701.368
2.0
25.52
FA4G4S4
3992.4708
1331.8309
3
3992.4709
0.0
25.7
FA4G4Lac1S4 + Ac
4399.6135
1467.5425
3
4399.6057
1.8
26.16/26.66
FA4G4Lac1S3
4066.5076
1356.5104
3
4066.5094
-0.4
27.34
FA4G4Lac1S4
4357.6030
1090.4097
4
4357.6097
-1.5
27.95
FA4G4Lac2S3
4431.6397
1108.9143
4
4431.6281
2.6
28.97
FA4G4Lac2S4
4722.7352
1181.6909
4
4722.7345
0.1
29.66
FA4G4Lac3S3
4796.7719
1200.2004
4
4796.7725
-0.1
30.50
FA4G4Lac3S4
5087.8674
1272.976
4
5087.8749
-1.5
31.77
FA4G4Lac4S4
5452.9996
1364.256
4
5452.9949
0.9
Figure 3. LC-MS data supporting the identification of various released N-glycan species. “+Ac” denotes an acetylation, such as the
previously reported O-acetylation of sialic acid residues (Neu5Ac).8
[ 163 ]
Comprehensive Characterization of the N and O-Linked Glycosylation of a Recombinant Human EPO
Profiling the O-Glycosylation of Intact rhEPO using a Widepore Amide HILIC Separation
O-linked glycans can be challenging to characterize due to the paucity of high fidelity mechanisms to
release them from their counterpart proteins. Released glycan analysis is an attractive approach for the
characterization of N-glycans because of the simplicity and effectiveness of PNGase F deglycosylation.
In place of using an analogous, universal glycosidase, analysts have resorted to releasing O-linked glycans
by chemical means, such as alkaline beta elimination18 or hydrazinolysis.19 These release mechanisms can
be challenging to implement and can very often produce artifacts, known as peeling products.
Rather than attempt a released O-glycan analysis of rhEPO, we looked to develop an alternative
characterization strategy. A novel workflow was devised that first involved subjecting the rhEPO to rapid
deglycosylation using GlycoWorks Rapid PNGase F and 1% RapiGest™ SF surfactant. In a 10-minute
preparation, a sample of N-deglycosylated intact rhEPO was obtained that could then be profiled via a HILIC
separation with an ACQUITY UPLC Glycoprotein BEH Amide Column. Figure 4 presents the chromatogram
obtained in this analysis using intrinsic fluorescence detection and intact protein HILIC techniques that
have been described in previous work.20 The N-deglycosylated rhEPO analyzed in this study resolved
into a series of approximately 10 peaks. Online ESI-MS provided highly detailed information, allowing
for proteoforms of rhEPO to be assigned to the various chromatographic peaks. The two most abundant
LC peaks were found to be represented by deconvoluted masses of 18893.8 and 19185.3 Da, which are
consistent with N-deglycosylated rhEPO that has a C-terminal arginine truncation as well as trisaccharide
and tetrasaccharide O-linked glycan modifications, respectively. More specifically, the mass shift observed
for the lighter species is indicative of a glycan modification comprised of 1 hexose, 1 N-acetylhexosamine,
and 1 N-acetylneuraminic acid. Meanwhile, the mass shift observed for the heavier species suggests a glycan
modification comprised of the same structure with an additional N-acetyl neuraminic acid.
0E+0
1E+7
Aglycosylated
PNGase F
18237.4 Da
+TFA
18893.8 Da
19185.3 Da
+TFA
+TFA
7
15 min
20 kDa
18
8
9
10
11
12
13
14
O-Linked
Glycan
O-Linked
Glycan
Intrinsic Fluorescence Detection
A
B
Figure 4. HILIC-fluorescence-MS analysis of N-deglycosylated, intact rhEPO. (A) Fluorescence chromatogram demonstrating
O-linked glycan heterogeneity and occupancy. Chromatograms obtained from 0.7 µg protein using a 2.1 x 150 mm ACQUITY UPLC
Glycoprotein BEH Amide, 300Å, 1.7 µm, 2.1 x 150 mm Column. (B) Deconvoluted mass spectra corresponding to three of the major
rhEPO proteoforms. Peak identifications, in addition to those denoted here, are tabulated in Figure 5.
[ 164 ] Comprehensive Characterization of the N and O-Linked Glycosylation of a Recombinant Human EPO
Further investigation of the LC-MS data also showed that the proteoform of rhEPO that is aglycosylated with
respect to the O-linked glycan eluted with a retention time of approximately 8.2 min. Moreover, these LC-MS
data indicated there to be at least two additional O-linked glycoforms and even more C-terminal truncation
proteoforms (Figure 5). Here, it is seen that this workflow can indeed be used to rapidly profile the O-linked
glycosylation of an rhEPO, such that information is gained about both occupancy and heterogeneity.
RT
(min)
Species
MWAvg, Theo
(Da)
MWAvg, Obs
(Da)
Mass Error
(Da)
8.0
N-deglycosylated, –C-term GDR
18066.5
18065.2
-1.3
N-deglycosylated, –C-term DR
18123.6
18122.4
-1.2
8.2
N-deglycosylated, –C-term R
18238.7
18237.4
-1.3
9.3
N-deglycosylated, –C-term R
+Hex1HexNAc1Neu5Ac1+Ac
18937.3
18936.2
-1.1
9.5
N-deglycosylated, –C-term GDR
+Hex1HexNAc1Neu5Ac1
18723.1
18722.3
-0.8
N-deglycosylated, –C-term DR
+Hex1HexNAc1Neu5Ac1
18780.1
18779.1
-1.0
9.7
N-deglycosylated, –C-term R
+Hex1HexNAc1Neu5Ac1
18895.2
18893.8
-1.4
9.9
N-deglycosylated, –C-term R
+Hex1HexNAc1Neu5Ac2+Ac
19228.5
19227.3
-1.2
10.0
N-deglycosylated, –C-term R
+Hex1HexNAc1Neu5Ac1 + O
18911.2
18910.0
-1.2
10.2
N-deglycosylated, –C-term GDR
+Hex1HexNAc1Neu5Ac2
19014.3
19013.7
-0.6
10.5
N-deglycosylated, –C-term R
+Hex1HexNAc1Neu5Ac2
19186.5
19185.3
-1.2
10.8
N-deglycosylated, –C-term R
+Hex1HexNAc1Neu5Ac2 + O
19202.5
19201.2
-1.3
Figure 5. LC-MS data supporting the identification of various N-deglycosylated rhEPO proteoforms. “–C-term” denotes the C-terminal
truncation of the rhEPO; losses of different residues are noted. Hex, HexNAc, and Neu5Ac stand for hexose, n-acetylhexosamine,
and N-acetylneuraminic acid. For example, Hex1HexNAc1NeuN5Ac1 corresponds to O-glycosylation involving 1 hexose,
1 N-acetylhexosamine, and 1 N-acetylneuraminic acid. “+O” denotes a mass shift indicative of the addition of an oxygen atom,
such as an oxidation or an exchange of Neu5Ac for Neu5Gc.8 Data supporting identifications of the most abundant rhEPO sequence
variant (–C-term R) and its glycoforms are highlighted with bold text. “+Ac” denotes an acetylation, such as the previously reported
O-acetylation of sialic acid residues (Neu5Ac).8
[ 165 ]
Comprehensive Characterization of the N and O-Linked Glycosylation of a Recombinant Human EPO
C O N C LU S IO NS
Several powerful tools have recently emerged
for the analysis of glycans that are built upon
LC-MS compatible hydrophilic interaction
chromatography (HILIC). At the heart of these new
glycan analysis workflows is a HILIC column that
has been purposefully designed for large molecule
separations. With this ACQUITY UPLC Glycoprotein
BEH Amide Column, an analyst can achieve higher
resolution separations of large, released N-glycans.
And when this analysis is paired with RapiFluor-MS
labeling, a technique is established that affords
not only high resolution but also unprecedented
sensitivity. This approach has been successfully
applied to obtain highly detailed information about
the N-glycosylation of a recombinant, human
epoetin alpha (rhEPO). Given that N-glycosylation
correlates with the half life and activity of an EPO,
such information, with its unparalleled quality, would
be invaluable in developing a new EPO therapeutic.
EPO is also O-glycosylated; the occupancy and
heterogeneity of which could also be critical to
demonstrate comparability among different drug
substances. Using the ACQUITY UPLC Glycoprotein
BEH Amide Column, we have outlined a simple
sample preparation and subsequent HILIC separation
that is capable of profiling these O-glycan attributes
on intact rhEPO. In summary, we have demonstrated
the use of two facile strategies that can be used
to detail both the N and O-linked glycosylation of
recombinant, human epoetin (rhEPO), a molecule
which has been perceived to be challenging to
characterize due to its relatively complicated
glycosylation. Collectively, these tools could be used
to accelerate the development of new biosimilars.
References
1. An, Z., Monoclonal antibodies – a proven and rapidly expanding therapeutic modality for
human diseases. Protein Cell 2010, 1 (4), 319–30.
2. Bennett, C. L.; Spiegel, D. M.; Macdougall, I. C.; Norris, L.; Qureshi, Z. P.; Sartor, O.; Lai,
S. Y.; Tallman, M. S.; Raisch, D. W.; Smith, S. W.; Silver, S.; Murday, A. S.; Armitage, J. O.;
Goldsmith, D., A review of safety, efficacy, and utilization of erythropoietin, darbepoetin,
and peginesatide for patients with cancer or chronic kidney disease: a report from the Southern
Network on Adverse Reactions (SONAR). Semin Thromb Hemost 2012, 38 (8), 783–96.
3. US$67 billion worth of biosimilar patents expiring before 2020 2012. http://www.gabionline.
net/Biosimilars/General/US-67-billion-worth-of-biosimilar-patents-expiring-before-2020.
4. Rush, R. S.; Derby, P. L.; Smith, D. M.; Merry, C.; Rogers, G.; Rohde, M. F.; Katta, V.,
Microheterogeneity of erythropoietin carbohydrate structure. Anal Chem 1995, 67 (8),
1442–52.
5. Dube, S.; Fisher, J. W.; Powell, J. S., Glycosylation at specific sites of erythropoietin is essential
for biosynthesis, secretion, and biological function. J Biol Chem 1988, 263 (33), 17516–21.
6. Takeuchi, M.; Inoue, N.; Strickland, T. W.; Kubota, M.; Wada, M.; Shimizu, R.; Hoshi, S.;
Kozutsumi, H.; Takasaki, S.; Kobata, A., Relationship between sugar chain structure and
biological activity
of recombinant human erythropoietin produced in Chinese hamster ovary cells.
Proc Natl Acad Sci USA 1989, 86 (20), 7819–22.
7. Takeuchi, M.; Takasaki, S.; Shimada, M.; Kobata, A., Role of sugar chains in the in vitro
biological activity of human erythropoietin produced in recombinant Chinese hamster
ovary cells. J Biol Chem 1990, 265 (21), 12127–30.
8. Harazono, A.; Hashii, N.; Kuribayashi, R.; Nakazawa, S.; Kawasaki, N., Mass spectrometric
glycoform profiling of the innovator and biosimilar erythropoietin and darbepoetin by
LC/ESI-MS. J Pharm Biomed Anal 2013, 83, 65–74.
9. Sasaki, H.; Bothner, B.; Dell, A.; Fukuda, M., Carbohydrate structure of erythropoietin expressed
in Chinese hamster ovary cells by a human erythropoietin cDNA. J Biol Chem 1987, 262 (25),
12059–76.
10. Takeuchi, M.; Takasaki, S.; Miyazaki, H.; Kato, T.; Hoshi, S.; Kochibe, N.; Kobata, A., Comparative
study of the asparagine-linked sugar chains of human erythropoietins purified from urine and
the culture medium of recombinant Chinese hamster ovary cells. J Biol Chem 1988, 263 (8),
3657–63.
11. Rush, R. S.; Derby, P. L.; Strickland, T. W.; Rohde, M. F., Peptide mapping and evaluation of
glycopeptide microheterogeneity derived from endoproteinase digestion of erythropoietin by
affinity high-performance capillary electrophoresis. Anal Chem 1993, 65 (14), 1834–42.
12. Gimenez, E.; Benavente, F.; Barbosa, J.; Sanz-Nebot, V., Analysis of intact erythropoietin and
novel erythropoiesis-stimulating protein by capillary electrophoresis-electrospray-ion trap
mass spectrometry. Electrophoresis 2008, 29 (10), 2161–70.
13. Gimenez, E.; Benavente, F.; de Bolos, C.; Nicolas, E.; Barbosa, J.; Sanz-Nebot, V., Analysis of
recombinant human erythropoietin and novel erythropoiesis stimulating protein digests by
immunoaffinity capillary electrophoresis-mass spectrometry. J Chromatogr A 2009, 1216
(12), 2574–82.
14. Lauber, M. A.; Yu, Y. Q.; Brousmiche, D. W.; Hua, Z.; Koza, S. M.; Magnelli, P.; Guthrie, E.;
Taron, C. H.; Fountain, K. J., Rapid Preparation of Released N-Glycans for HILIC Analysis
Using a Labeling Reagent that Facilitates Sensitive Fluorescence and ESI-MS Detection.
Anal Chem 2015, 87 (10), 5401–9.
15. Lauber, M. A.; Brousmiche, D. W.; Hua, Z.; Koza, S. M.; Guthrie, E.; Magnelli, P.;
Taron, C. H.; Fountain, K. J., Rapid Preparation of Released N-Glycans for HILIC Analysis
Using a Novel Fluorescence and MS-Active Labeling Reagent. Waters Application Note
720005275EN 2015.
16. Eoin F.J. Cosgrave, M. A. L., Robert Birdsall, and Sean M. McCarthy, New Capabilities for
Monitoring Released N-Glycans through the Combined Use of RapiFluor-MS Labeling,
ACQUITY UPLC H-Class Bio System, and Serial Fluorescence/ACQUITY QDa Mass Detection.
Waters Application Note 720005352EN 2015.
[ 166 ] Comprehensive Characterization of the N and O-Linked Glycosylation of a Recombinant Human EPO
Waters Corporation
34 Maple Street
Milford, MA 01757 U.S.A.
T: 1 508 478 2000
F: 1 508 872 1990
www.waters.com
Waters, The Science of What’s Possible, MassLynx, ACQUITY UPLC, SYNAPT, and Xevo are registered trademarks of Waters Corporation.
GlycoWorks, RapiGest, and RapiFluor-MS are trademarks of Waters Corporation. All other trademarks are the property of their
respective owners.
©2015 Waters Corporation. Produced in the U.S.A. August 2015 720005462EN AG-PDF
17. Lauber, M. A.; Koza, S. M., Enhancing the Peak Capacity of High Molecular Weight N-Glycan
HILIC Separations with a Wide-Pore Amide Bonded Stationary Phase. Waters Technology Brief
720005381EN 2015.
18. Huang, Y.; Mechref, Y.; Novotny, M. V., Microscale nonreductive release of O-linked glycans
for subsequent analysis through MALDI mass spectrometry and capillary electrophoresis.
Anal Chem 2001, 73 (24), 6063–9.
19. Patel, T. P.; Parekh, R. B., Release of oligosaccharides from glycoproteins by hydrazinolysis.
Methods Enzymol 1994, 230, 57–66.
20. Lauber, M. A.; Koza, S. M., Developing High Resolution HILIC Separations of Intact
Glycosylated Proteins Using a Wide-Pore Amide-Bonded Stationary Phase Waters
Application Note 720005380EN 2015.
[ 167 ]
Comprehensive Characterization of the N and O-Linked Glycosylation of a Recombinant Human EPO
MONOSACCHARIDES
[ 168 ]
WAT E R S S O LU T IO NS
ACQUITY UPLC® H-Class Bio System
Empower® 3 Chromatography
Data Software
XBridge® HPLC Columns
ACQUITY UPLC BEH Column
K E Y W O R D S
Glycosylation, glycoproteins
therapeutics, N-linked glycans,
monosaccharide, 2AA-derivatized
monosaccharides, glycan composition,
HPLC, UPLC
A P P L I C AT IO N B E N E F I T S
■
■
Geometric scaling of an HPLC method for
determining monosaccharide composition
in glycoprotein samples to UPLC®
■
■
Application of scalable column chemistries
for monosaccharide analysis
■
■
Guidance for scaling HPLC gradient
methods to UPLC
■
■
A high sample throughput and high
resolution UPLC approach for
determining monosaccharide analysis
I N T RO DU C T IO N
Evidence for reliable and consistent glycosylation of glycoprotein therapeutics
is typically obtained through LC-based analysis of N-linked glycans. On
occasion, regulatory agencies request information of monosaccharide
content as an orthogonal technique for confirming the glycan composition.
Beyond profiling changes in total glycan composition, monosaccharide
analysis can also be used as an exploratory technique for identifying various
monosaccharide modifications, including phosphorylation and sulfation, which
can be particularly challenging to discern when analyzing at the released
glycan level in the absence of mass spectrometry.
Initial investigation of monosaccharide analyses resulted in a reliable and
informative HPLC-based approach using 2-aminobenzoic acid (2AA)-derivatized
glycan hydrolysates where each of the individual monosaccharides were
successfully separated using reversed-phase chromatography. Since this time,
the analytical demands placed on development and QC labs that are tasked
with characterizing and monitoring glycoprotein therapies have increased
dramatically. Assays used to monitor changes in glycosylation therefore
need to be updated in order to maximize productivity through improved
analytical efficiency.
In this application note, we illustrate the transfer of a robust HPLC
monosaccharide method to the ACQUITY UPLC H-Class Bio System running
as an HPLC, using monosaccharide standards as well as bovine fetuin and the
commercially available monoclonal antibody cetuximab as analytes. Geometric
scaling of the presented method is then performed in a stepwise manner to
demonstrate the advantages of transferring legacy HPLC monosaccharide
analyses to UPLC technology as a means of decreasing assay time.
Future Proofing the Biopharmaceutical QC Laboratory:
Chromatographic Scaling of HPLC Monosaccharide Analyses
Using the ACQUITY UPLC H-Class Bio System
Eoin F.J. Cosgrave and Sean M. McCarthy
Waters Corporation, Milford, MA, USA
[ 169 ]
E X P E R IM E N TA L
LC conditions
Waters ACQUITY UPLC
H-Class Bio System,
comprised of:
ACQUITY UPLC H-Class
Bio Quaternary Solvent
Manager (QSM)
ACQUITY UPLC H-Class
Bio Sample Manager
(FTN)
ACQUITY UPLC H-Class
Bio Column Heater
(CH-A)
ACQUITY UPLC
FLR Detector
Extension loop: 100 μL
(p/n 430002625)
Waters columns:
XBridge C18 5 μm,
130 Å, 4.6 x 100 mm
(p/n 186003115)
XBridge C18 3.5 μm,
130 Å, 2.1 x 100 mm
(p/n 186003033)
XBridge C18 XP 2.5 μm,
130 Å, 2.1 x 100 mm
(p/n 186006031)
ACQUITY UPLC BEH C18
1.7 μm, 130 Å,
2.1 mm x 100 mm
(p/n 186002352)
Mobile phase A:
0.2% N-butylamine,
0.5% phosphoric acid,
and 1% THF in H2O
Mobile phase B:
50% mobile phase A
in acetonitrile
Excitation wavelength: 360 nm
Emission wavelength: 425 nm
Column temp.:
30 °C
Injection vol.:
4.6 mm x 100 mm format,
4.8 μL, 2.1 mm x 100 mm
format, 1 μL
[ 170 ]
Future Proofing the Biopharmaceutical QC Laboratory: Chromatographic Scaling of
HPLC Monosaccharide Analyses Using the ACQUITY UPLC H-Class Bio System
Following the described approach, typical run times for HPLC-based monosaccharide analysis are reduced
from 45 minutes to just 17 minutes. Importantly, chromatographic resolution between measured critical
peak pairs is observed to improve with migration to smaller column particle sizes. Selectivity is unaffected
due to the availability of reversed-phase column chemistries in a number of particle sizes and dimensions.
The data presented here indicate migration to UPLC technology offers significant advantages for improving
monosaccharide chromatographic quality.
Sample preparation
Derivatization of monosaccharides was performed as previously described,1, 2 with a number of minor
modifications as recommended by Stepan and Staudacher.3 Monosaccharides from bovine fetuin were released
by acid hydrolysis using 2 M TFA with hydrolysis occurring for 3 h at 100 °C. Resulting hydrolysates were then
dried by centrifugal evaporation followed by reconstitution in 5 μL of 80 mg/mL sodium acetate trihydrate.
A 2AA labeling solution was prepared by dissolving 30 mg of 2AA in 1 mL of 2% (w/v) boric acid
in methanol. This suspension was then used to dissolve 30 mg of sodium cyanoborohydride. Of this
preparation, 10 μL was added to each of the monosaccharide mixtures. Monosaccharides were labeled at
80 °C for 60 min. Upon completion of labeling, serial dilutions were performed to generate a 1000-fold
dilution of the labeled material. For preparation of monosaccharide standards, labeling was performed as
outlined above with the omission of acid hydrolysis.
Method details (flow rate and time)
Step
%B1
5 µm
3.5 µm
2.5 µm
1.7 µm
Flow
(mL
min-1)
Time
(min)
Flow
(mL
min-1)
Time
(min)
Flow
(mL min-1)
Time
(min)
Flow
(mL min-1)
Time
(min)
1
7
0.480
0.00
0.685
0.00
0.200
0.00
0.294
0.00
2
7
0.480
7.78
0.685
5.45
0.200
3.89
0.294
2.64
3
17
0.480
27.78
0.685
19.47
0.200
13.88
0.294
9.44
4
100
0.480
28.89
0.685
20.24
0.200
14.43
0.294
9.82
5
100
0.480
40.00
0.685
28.03
0.200
19.99
0.294
13.60
6
7
0.480
41.11
0.685
28.81
0.200
20.54
0.294
13.97
7
7
0.480
50.00
0.685
35.04
0.200
24.98
0.294
17.00
Calculations
Flow rate scaling:
F refers to flow rate, d refers to column I.D., and dp
refers to particle diameter. In each case, 2 refers to
the new column and 1 refers to the original column.
Injection volume scaling:
Vi refers to injection volume, r refers to column
internal radius, and L refers to column length.
In each case, 2 refers to the new column and
1 refers to the original column.
[ 171 ]
Future Proofing the Biopharmaceutical QC Laboratory: Chromatographic Scaling of
HPLC Monosaccharide Analyses Using the ACQUITY UPLC H-Class Bio System
R E SU LT S A N D D IS C U S S IO N
ACQUITY UPLC H-Class Bio System for HPLC monosaccharide analysis
To verify the ability of the ACQUITY UPLC H-Class Bio to run legacy analyses of 2AA-derivatized
monosaccharides, we first established an HPLC separation using method conditions based on previously
described chromatographic conditions1,2 To evaluate the proposed method, a reference standard mix was
prepared by combining individual monosaccharides into a common mix. This mix included the typical
monosaccharides expected in biologically relevant samples, namely N-acetylglucosamine (GlcNAc),
N-acetylgalactosamine (GalNAc), glucose (Glc), mannose (Man), galactose (Gal), xylose (Xyl), and fucose
(Fuc). In addition to these standards, two glycoprotein samples were also selected to determine the accuracy
of this approach in determining monosaccharide composition. The first glycoprotein selected was bovine
fetuin, a protein known to contain both N- and O-glycosylation sites. The second glycoprotein selected
was the commercial monoclonal antibody cetumixab.
Following hydrolysis from the glycoprotein samples and 2AA derivatization, monosaccharides were
separated using the aforementioned method. The resulting HPLC chromatogram (Figure 1) acquired on the
ACQUITY UPLC H-Class Bio is consistent with previously published data.2 In terms of chromatographic
performance, peak capacity was measured together with selectivity and resolution between critical peak pairs.
These data are summarized in Table 1. Consistent peak area was observed across separations using all particle
sizes (Figure 2d). These data confirmed the separation of all relevant components and, therefore, established
a suitable method for monosaccharide method scaling to UPLC technology.
Figure 1. HPLC analysis of monosaccharides. A separation performed with
a Waters XBridge 5 µm C18 Column using a previously described method1.
Monosaccharides are identified as follows: 1) N-acetylglucosamine (GlcNAc),
2) N-acetylgalactosamine (GalNAc), 3) Galactose (Gal), 4) Mannose (Man),
5) Glucose (Glc), and 6) Fucose (Fuc).
Table 1. Summary data for chromatographic analysis.
Measurement
Particle size (µm)
5
3.5
2.5
1.7
Column ID (mm)
4.6
4.6
2.1
2.1
Column Length (mm)
100
100
100
100
Average W1/2h
0.349
0.189
0.125
0.076
Tg
20.00
9.99
14.02
6.80
Pc
58
75
81
90
GlcNAc
8.71
5.84
4.40
2.85
GalNAc
9.69
6.49
4.87
3.15
Gal
20.92
14.32
10.87
7.08
Man
22.01
15.09
11.46
7.46
Glc
22.83
15.66
11.92
7.78
Fuc
28.12
19.39
14.76
9.74
GlcNAc
3.65
3.83
3.69
3.67
GalNAc
8.26
8.52
8.23
8.31
Gal
22.17
22.48
22.45
22.06
Man
23.04
23.07
23.05
22.72
Glc
23.76
23.89
23.88
23.61
Fuc
19.11
18.21
18.71
19.62
GlcNAc, GalNAc
1.12
1.13
1.13
1.15
GalNAc, Gal
2.26
2.38
2.47
2.67
Gal, Man
1.05
1.06
1.06
1.06
Man, Glc
1.04
1.04
1.04
1.05
Glc, Fuc
1.24
1.25
1.26
1.28
GlcNAc, GalNAc
2.52
3.10
3.16
3.32
GalNAc, Gal
21.05
27.12
30.69
33.25
Gal, Man
1.62
2.10
2.48
2.62
Man, Glc
1.20
1.53
1.90
2.23
Glc, Fuc
7.34
9.59
11.41
12.78
[ 172 ]
Future Proofing the Biopharmaceutical QC Laboratory: Chromatographic Scaling of
HPLC Monosaccharide Analyses Using the ACQUITY UPLC H-Class Bio System
Figure 2. Geometric scaling of a monosaccharide separation.
(A) 5 µm particle (B) 3.5 µm particle, (C) 2.5 µm particle, and
(D) 1.7 µm particle. 1) GlcNAc, 2) GalNAc, 3) Gal, 4) Man,
5) Glc, and 6) Fuc.
Migration of monosaccharide analysis from HPLC to UPLC improves resolution
In an effort to improve throughput of analyses and general chromatographic quality, the above described
method was geometrically scaled in a stepwise manner to UPLC column technology. This involved scaling
the flow rate to the new column dimensions and adjusting individual steps in the gradient method to deliver
equivalent column volumes as itemized in the original recommended HPLC method. Details of the UPLC
method can be found in the experimental section of this application note.
Several particle sizes of identical chemistry ranging from 1.7 to 5 μm were used in the scaling exercise. In
the case of the 1.7 μm and 2.5 μm particles, a 2.1 mm x 100 mm column dimension was used while 4.6 mm
x 100 mm column dimensions were used for the 3.5 μm and 5 μm particles. Flow rate and injection volume
scaling calculations specific for individual column dimensions were determined using the appropriate
equations defined in the experimental section of this application note. The duration for each step in the gradient
table was subsequently modified based on the new flow rate and column volume to ensure consistent delivery
of equivalent column volumes per change in organic composition when compared to the original method. The
results of these calculations are summarized in the experimental section of this application note where flow
rates and gradient step durations are itemized with respect to each column particle size.
To evaluate the results of method scaling, the monosaccharide mix was separated under the new gradient
conditions for each particle size and column dimensions (Figure 2). Method scaling to UPLC column
technology reduced the total required run time of the method from 50 min (in the case of the 5 μm column,
Figure 2a) to just 17 min (in the case of the 1.7 μm column, Figure 2d), an improvement in efficiency of
approximately 66%. Details of chromatographic performance are presented in Table 1.
A
B
C
D
1
2
3 4 5
6
[ 173 ]
Future Proofing the Biopharmaceutical QC Laboratory: Chromatographic Scaling of
HPLC Monosaccharide Analyses Using the ACQUITY UPLC H-Class Bio System
In general, peak capacity was shown to increase with decreasing particle size (Figure 3a), an expected
outcome based on the narrower peaks achieved with UPLC technology. The reduction in run time was not at
the cost of resolution, where a general improvement was observed across all critical peak pairs as column
particle size decreased (Figure 3b). Selectivity remained unaffected mainly due to the availability of
identical column chemistry across multiple particle sizes (Figure 3c). Changes in column particle size did not
impact relative peak area determination, evidenced by averaged peak areas for each monosaccharide across
all column formats (Table 1 and Figure 3d). Taken together, scaling of the original monosaccharide method
produced improved resolution in a shorter amount of time, with negligible impact to selectivity.
0
5
10
15
20
25
30
Relative Peak
Ar
ea
(%
)
Monosaccharide
0
20
40
60
80
100
5
3.5
2.5
1.7
Peak
Capacity
Column Particle Size (µm)
0
1
2
3
4
Resolution
Peak Pair
5 µm
3.5 µm
2.5 µm
1.7 µm
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
Selecti
vity
Peak Pair
5 µm
3.5 µm
2.5 µm
1.7 µm
A
B
C
D
Figure 3. Quantitative analysis of geometric scaling. (A) Relative peak area measurements for individual monosaccharides, averaged
across all columns with standard deviation denoted by error bars; (B) Resolution measurements from 3 monosaccharide peak pairs;
(C) Peak capacity measurements of geometrically scaled methods across all column dimensions, and; (D) selectivity measured
between 3 monosaccharide peak pairs.
[ 174 ]
Future Proofing the Biopharmaceutical QC Laboratory: Chromatographic Scaling of
HPLC Monosaccharide Analyses Using the ACQUITY UPLC H-Class Bio System
Figure 4. Monosaccharide analysis of fetuin
and cetuximab using the developed UPLC
separation. (A) monosaccharide standard,
(B) cetuximab monosaccharides, (C) fetuin
monosaccharides. In all chromatograms,
monosaccharides are identified as:
1) GlcNAc, 2) GalNAc, 3) Gal, 4) Man,
5) Glc, and 6) Fuc.
Determination of monosaccharide content in glycoprotein samples
To verify this approach in determining monosaccharide composition, we analyzed both fetuin and
cetuximab 2AA-labeled monosaccharides. Fetuin is known to contain O-glycosylation and therefore
should demonstrate the presence of GalNAc within its profile. Cetuximab, on the other hand, contains no
O-glycosylation and should therefore be absent of any GalNAc.
To perform the analysis, the 1.7-μm BEH C18 particle chemistry was used and results of the separations
were compared to the monosaccharide standard (Figure 4a). GalNAc is clearly detected in the fetuin sample
(Figure 4b), as evidenced by a chromatographic peak with similar retentivity to the GalNAc standard. Also
consistent with literature is the absence of core fucosylation in fetuin, largely evidenced by the absence of a
peak with similar retentivity to the fucose standard. This finding is consistent with previous reports of bovine
fetuin N-glycosylation characterization.4,5 Analysis of cetuximab reveals all individual monosaccharides with
the exception of xylose, a monosaccharide not typically associated with mammalian N-linked glycosylation
(Figure 4c). Mannose was found to be in higher quantity than galactose, a result consistent with previous
literature indicating the presence of several high mannose structures in cetuximab.6-8
1
2
3 4
5
6
A
B
C
1
3
4
6
1
2
3
4
[ 175 ]
Future Proofing the Biopharmaceutical QC Laboratory: Chromatographic Scaling of
HPLC Monosaccharide Analyses Using the ACQUITY UPLC H-Class Bio System
Waters Corporation
34 Maple Street
Milford, MA 01757 U.S.A.
T: 1 508 478 2000
F: 1 508 872 1990
www.waters.com
Waters, The Science of What’s Possible, UPLC, ACQUITY UPLC, XBridge, and Empower are registered trademarks of Waters Corporation.
All other trademarks are the property of their respective owners.
©2015 Waters Corporation. Produced in the U.S.A. March 2015 720005255EN AG-PDF
C O N C LU S IO NS
As a complementary approach to released glycan analysis,
monosaccharide profiling allows analysts to verify glycan
composition determined in traditional HILIC-based separations.
A well-established assay for monosaccharide analysis uses
reversed-phase chromatography to separate 2AA-derivatized
monosaccharides. In this application note, we have demonstrated
the ability of the ACQUITY UPLC H-Class Bio System for running
both HPLC and UPLC methods for monosaccharide analyses.
This HPLC assay provided sufficient resolution of individual
monosaccharides but was restricted in part by the time required
to perform the separation. A significant reduction in runtime
was obtained by transferring the legacy HPLC method to
UPLC technology. With the modernized, UPLC-based separation,
a higher throughput assay for monosaccharide analysis was
thereby achieved.
References
1. Ludger, Product Guide for LudgerTag 2-AA (2-Aminobenzoic Acid)
Monosaccharide Release and Labeling Kit. In Ludger Ltd: Oxford, UK,
2012; pp 1–17.
2. Anumula, K. R., Quantitative determination of monosaccharides in
glycoproteins by high-performance liquid chromatography with highly
sensitive fluorescence detection. Anal Biochem 1994, 220, (2), 275–83.
3. Stepan, H.; Staudacher, E., Optimization of monosaccharide determination
using anthranilic acid and 1-phenyl-3-methyl-5-pyrazolone for gastropod
analysis. Anal Biochem 2011, 418, (1), 24–9.
4. Zhou, H.; Froehlich, J. W.; Briscoe, A. C.; Lee, R. S., The GlycoFilter: a simple
and comprehensive sample preparation platform for proteomics, N-glycomics
and glycosylation site assignment. Mol Cell Proteomics 2013, 12, (10),
2981–91.
5. Green, E. D.; Adelt, G.; Baenziger, J. U.; Wilson, S.; Van Halbeek, H., The
asparagine-linked oligosaccharides on bovine fetuin. Structural analysis
of N-glycanase-released oligosaccharides by 500-megahertz 1H NMR
spectroscopy. J Biol Chem 1988, 263, (34), 18253–68.
6. Wiegandt, A.; Meyer, B., Unambiguous characterization of N-glycans of
monoclonal antibody cetuximab by integration of LC-MS/MS and (1)H
NMR spectroscopy. Anal Chem 2014, 86, (10), 4807–14.
7. Janin-Bussat, M. C.; Tonini, L.; Huillet, C.; Colas, O.; Klinguer-Hamour, C.;
Corvaia, N.; Beck, A., Cetuximab Fab and Fc N-glycan fast characterization
using IdeS digestion and liquid chromatography coupled to electrospray
ionization mass spectrometry. Methods Mol Biol 2013, 988, 93–113.
8. Qian, J.; Liu, T.; Yang, L.; Daus, A.; Crowley, R.; Zhou, Q., Structural
characterization of N-linked oligosaccharides on monoclonal antibody
cetuximab by the combination of orthogonal matrix-assisted laser
desorption/ionization hybrid quadrupole-quadrupole time-of-flight tandem
mass spectrometry and sequential enzymatic digestion. Anal Biochem 2007,
364, (1), 8–18.
[ 176 ]
Future Proofing the Biopharmaceutical QC Laboratory: Chromatographic Scaling of
HPLC Monosaccharide Analyses Using the ACQUITY UPLC H-Class Bio System
Waters Corporation
34 Maple Street
Milford, MA 01757 U.S.A.
T: 508 478 2000
F: 508 872 1990
www.waters.com
Waters, The Science of What’s Possible, Alliance, ACQUITY, ACQUITY UPLC, UPLC, Xevo,
XBridge, XSelect, MassLynx, Oasis, Empower, UNIFI, SYNAPT, UltraPerformance LC,
and QDa are registered trademarks of Waters Corporation. GlycoWorks, RapiFluor-MS,
RapiGest, MaxEnt, and AccQ•Fluor are trademarks of Waters Corporation. All other
trademarks are the property of their respective owners.
©2016 Waters Corporation. Printed in the U.S.A.
February 2016 720005532EN AW-SIG
Similar PDF
Key words
Key words
Key words
