Automated High-Throughput Data Processing for Targeted Multiplexed Insulin Analog Detection and Quantification

Significance of the Topic

The accurate and high-throughput quantification of insulin and its analogs in biological matrices underpins therapeutic monitoring, pharmacokinetic studies, and quality control in biopharmaceutical production. Traditional immunoassays and UV methods lack the selectivity to discriminate co-eluting peptide variants. Integration of mass spectrometric immunoassay (MSIA) with high-resolution, accurate-mass (HRAM) detection offers a solution for multiplexed, targeted quantification at picomolar levels.

Study Objectives and Overview

This work presents an automated workflow combining MSIA, liquid chromatography-HRAM mass spectrometry on an Orbitrap-based Q Exactive platform, and Pinpoint software processing. The goal is to detect and quantify five insulin variants (Humulin S, Apidra, Lantus, NovoRapid, and bovine insulin) spiked into bovine serum albumin (PBS/BSA) and human plasma at 1.5–960 pM, using porcine insulin as an internal standard.

Methodology and Instrumentation

Preparation and Extraction:

• Automated affinity capture via antibody-coated MSIA D.A.R.T.S and a 96-channel liquid handler.
• Sequential rinses and elution with ACTH 1-24 in acetonitrile/TFA, drying, and reconstitution.

Chromatography and Detection:

• UltiMate 3000 RSLCnano system with ProSwift RP-4H column (1 × 250 mm) at 50 °C.
• Gradient elution (10–50% acetonitrile/0.1% formic acid) over 10 min.
• Q Exactive mass spectrometer: full-scan HRAM (70,000 FWHM) and data-dependent MS/MS (17,500 FWHM), m/z 800–2000.

Data Processing:

• Pinpoint software v1.4 for automated extraction of six isotopes across three charge states per variant with ±7 ppm tolerance.
• Qualitative scoring via isotopic distribution, charge-state patterns, and LC peak profile.
• Quantitation curves built with 1/x weighting and normalization to porcine insulin internal standard.

Main Results and Discussion

• Chromatographic separation and isotopic resolution of co-eluting insulin analogs were achieved, with minimal interference even at low concentrations.
• Linear regression coefficients exceeded 0.98 across the dynamic range (1.5–960 pM) for all variants.
• Average dot-product correlation scores for isotopic patterns were >0.97 at 50 pM, declining modestly to ~0.58 at 1.5 pM, yet still diagnostic.
• Inter- and intra-assay precision (%RSD) remained below 20% for most measurements; only a minority of values at the lowest level exceeded this threshold.

Benefits and Practical Applications

• High selectivity and sensitivity for multiplexed peptide hormone analysis in complex matrices.
• Automated sample preparation and data processing reduce hands-on time and enable throughput compatible with routine clinical or research laboratories.
• Post-acquisition flexibility allows retrospective data mining for novel or modified insulin sequences without re-running samples.

Future Trends and Opportunities

• Extension to broader peptide and protein biomarker panels using MSIA coupled with HRAM detection.
• Integration with clinical decision-support systems for real-time therapeutic monitoring.
• Development of universal MSIA workflows for large-scale population proteomics and pharmacokinetic studies.
• Optimization of data-processing algorithms incorporating machine learning for improved qualitative and quantitative scoring.

Conclusion

The presented automated MSIA-LC-HRAM workflow successfully quantifies intact insulin analogs at picomolar levels in serum and plasma with high accuracy, precision, and throughput. The combination of affinity enrichment, Orbitrap-level resolution, and advanced software processing provides a robust platform for targeted peptide analysis, overcoming limitations of traditional assays.

References

1. Krastins B. et al. Clinical Biochemistry 2013,46(6):399–410.
2. Niederkofler E.E. et al. Application Note, Thermo Fisher Scientific 2013,1001.
3. Oran P.E. et al. Proteomics Clinical Applications 2011,5(7–8):454–459.
4. Couchman L. et al. Application Note, Thermo Fisher Scientific 2014,0614.
5. Peterman S.M. et al. Proteomics 2014,14(12):1445–1456.