Understanding the fundamentals of the on-off retention mechanism of oligonucleotides and their application to high throughput analysis
Journal of Chromatography, A Volume 1739: Understanding the fundamentals of the on-off retention mechanism of oligonucleotides and their application to high throughput analysis
The goal of this study is to deepen the understanding of the on-off retention behavior of oligonucleotides in ion-pair reversed-phase liquid chromatography (IP-RPLC), where small variations in acetonitrile (ACN) proportion significantly influence retention. This is achieved by systematically measuring S values (slopes of log k vs. %ACN) under diverse mobile phase conditions, including variations in ion-pair (IP) hydrophobicity, IP concentration, column temperature, buffering acid, and mobile phase pH. The study identifies the hierarchy of these factors in influencing S values, with IP hydrophobicity having the most significant effect.
Additionally, the insights gained are applied to optimize ultra-fast separations of two therapeutic oligonucleotides: a 20-mer antisense oligonucleotide (ASO) and a large single guide RNA (sgRNA). By optimizing mobile phase conditions to achieve high S values and prevent diastereomer separation, the study demonstrates successful separations of these compounds and their impurities in under one minute using a 5 mm column, paving the way for efficient high-throughput oligonucleotide analysis.
The original article
Understanding the fundamentals of the on-off retention mechanism of oligonucleotides and their application to high throughput analysis
Honorine LARDEUX, Selin BAGCI, Mimi GAO, Wiebke HOLKENJANS, Reinhard PELL, Davy GUILLARME
Journal of Chromatography A, Volume 1739, 4 January 2025, 465523
https://doi.org/10.1016/j.chroma.2024.465523
licensed under CC-BY 4.0
Selected sections from the article follow. Formats and hyperlinks were adapted from the original.
Highlights
- Oligonucleotides show on-off retention behaviour, with small ACN variations significantly altering retention.
- S-values (log k vs. %ACN slopes) were measured across diverse mobile phase conditions.
- Mobile phase effects on S-values: IP hydrophobicity > IP concentration > temperature > buffering acid > pH.
- Ultra-fast ASO and sgRNA separation completed in under one minute using a 5 mm column.
Abstract
Ion-pair reversed-phase liquid chromatography (IP-RPLC) is clearly recognized as the gold standard for analyzing therapeutic oligonucleotides (ONs). Recent studies have shown that ONs exhibit an on-off retention behavior in IP-RPLC, meaning that minor changes in acetonitrile (ACN) proportion can significantly impact retention. However, this behavior was initially demonstrated with only a single mobile phase condition. The aim of this study is to gain a deeper understanding of ON elution behavior by measuring the S values (slope of the retention model, log k vs. %ACN) across a broad range of mobile phase conditions. We systematically calculated the S values for both a 20-mer and 100-mer model ON under various conditions, including different IP reagents, IP concentrations, mobile phase pH, column temperatures, and two different buffering acids. We demonstrated that these mobile phase conditions impact the S values in the following order: IP hydrophobicity > IP concentration > column temperature > buffering acid > mobile phase pH. The main explanation for this trend is that mobile phase conditions that reduce the ion-pair retention mechanism (such as low IP hydrophobicity or concentration) will enhance the impact of % ACN on retention, leading to higher S values.
In the second part of the study, this knowledge was used to develop ultra-fast separations for two therapeutic oligonucleotides: a 20-mer antisense oligonucleotide (ASO) without phosphorothioate (PS) modifications and a large single guide RNA (sgRNA) that includes certain PS modifications. The mobile phase conditions were optimized to maximize S values, while preventing the separation of diastereomers. It is important to notice that an S-value of at least 30 is required to benefit from the use of ultra-short columns. This approach allows the successful separation of the main species (ASO and sgRNA) and related impurities in less than one minute using a 5 mm length column.
1. Introduction
The pharmaceutical industry has recently shifted attention from small molecules to biopharmaceuticals, including monoclonal antibodies (mAbs), for targeting specific diseases. While these molecules have achieved significant success, challenges remain in treating complex diseases, such as neurodegenerative and rare genetic diseases [1,2]. To overcome this limitation, there is increasing interest in oligonucleotide-based therapeutics to broaden the range of druggable targets [3]. Therapeutic oligonucleotides (ONs) primarily modulate (e.g. inhibit) gene expression or protein production. The development of ONs began in the 1990s, with fomivirsen as the first FDA-approved ON drug in 1998 [1]. Today, ON drugs are used to treat a variety of diseases, including neurodegenerative diseases, respiratory disorders, diabetic retinopathy, and more recently, cancers [4]. Notably, Spinraza®, developed for treating spinal muscular atrophy in children, became the first blockbuster ON therapy, achieving global sales of $1.7 billion in 2018 [5]. The ON market, valued at $6.3 billion in 2021, is projected to exceed $14.1 billion by 2026 [6]. Increased ON research is attracting investors and pharmaceutical companies, with experts predicting that 10 % of new FDA-approved drugs will be ON-based in the future [7]. This growth is driven by the development of CRISPR-Cas technology, a revolutionary gene-editing tool that modifies DNA sequences within the genome [8]. The better knowledge of the human genome, improvements in pharmacokinetic properties of ONs and the development of innovative delivery materials and methods all play an important role in the rising interest in ONs [7,9].
ONs used as therapeutic drugs require strict quality control throughout the production process. The synthesis of ONs can result in various impurities, including shortmers, longmers, and more complex variants, necessitating precise and selective analytical techniques to separate these impurities from the ON [10,11]. Accurate assessment of the purity of both intermediates and final active pharmaceutical ingredients is crucial [12]. Among the strategies available, liquid chromatography is a key strategy, and various modes have been used, such as anion-exchange chromatography (AEX), hydrophilic interaction liquid chromatography (HILIC), size-exclusion chromatography (SEC), and ion-pair reversed-phase liquid chromatography (IP-RPLC) [3,13]. Among these different modes, IP-RPLC is currently considered the gold standard [14].
When analyzing large molecules in liquid chromatography, they experience a unique elution mode [15,16]. Indeed, the analyte strongly interacts with the stationary phase at the column inlet, leading to its retention. A slight increase in the mobile phase strength, can totally disrupt these interactions, resulting in the complete elution of the analyte with no further interaction with the stationary phase. This transition from infinite retention to complete elution is characteristic of the “on-off” or “bind-elute” elution mechanism [17]. The strength of this on-off retention mechanism can be assessed by the slope of the retention model, which is the plot of the logarithm of the retention factor of an analyte against the volume fraction or proportion of the organic solvent. The slope of the retention model, denoted as S, describes the sensitivity of an analyte retention to changes in the mobile phase composition. For small molecules, S typically ranges from 3 to 5, whereas for large molecules, it can exceed 100. A larger S value indicates a stronger effect of the organic solvent on the retention behavior of the analyte. Due to this on-off mechanism in large molecules, the column length has minimal impact on retention, making short columns suitable for the high-throughput analysis of large molecules [17]. Such on-off retention behavior has been demonstrated for proteins, particularly mAbs in RPLC and IEX modes [18,19]. In a recent study, we demonstrated that ONs also exhibit on-off elution behavior in IP-RPLC, confirming the applicability of the linear solvent strength (LSS) model to 10- to 100-mer ONs [20]. This study served as a proof of concept using a single mobile-phase condition.
The goal of the present work is to gain a comprehensive understanding of the on-off elution behavior of ONs in IP-RPLC. Therefore, we will investigate the impact of various mobile phase conditions, including the hydrophobicity and concentration of the ion-pairing reagent, type of acid buffer counter ion, mobile-phase pH, and column temperature, on the S values. Based on this fundamental understanding of the on-off retention mechanism, we have defined mobile phase conditions to promote on-off behavior. By combining these conditions with ultra-short columns and ballistic gradients, we can achieve ultra-fast separation of a 20-mer antisense oligonucleotide (ASO) and a single guide RNA (sgRNA). The final objective of this work is to demonstrate the feasibility of rapidly separating therapeutic ONs from their impurities using very short columns.
2. Materials and methods
2.2. Sample and mobile phase preparation
Oligonucleotide stock solutions with a concentration of 100 μM are prepared by reconstituting the lyophilized material in the appropriate volume of RNase-free water and stored at −20 °C. Oligonucleotide samples are prepared by diluting the stock solutions to the desired concentration using RNase-free water. To prevent potential degradation by RNase enzymes, all Eppendorf tubes and pipette tips used during the sample preparation are RNase-free. Mobile phase A consisted of an IP reagent and acidic modifier in water, which is always AA, except in Section 3.1.4 where HFIP was tested. The pH of the aqueous mobile phase A is adjusted with IP reagent or AA. The amount of the IP reagent and buffering acid required to prepare the mobile phase A is recorded and used to prepare the mobile phase B, which is a 50:50 mixture of water and ACN.
2.3. Apparatus and methodology
All chromatographic measurements are performed on a Waters ACQUITY UPLC I-Class system (Milford, MA) equipped with a binary solvent delivery pump, a flow-through-needle (FTN) autosampler, and a UV detector. A Waters ACQUITY Premier Oligonucleotide BEH C18 of 50 × 2.1 mm 1.7 μm, 130 Å column was used for the comprehensive understanding of on-off elution mechanism. A Waters ACQUITY UPLC BEH C18 5 × 2.1 mm, 1.7 μm, 130 Å VanGuard pre-column was used for the separation of ASO and sgRNA samples. Data acquisition and instrument control were performed using Empower 3 Software (Waters). The freely available Excel Spreadsheet “Calculation of LSS parameters in LC” was used to obtain LSS parameters (log k0 and S) of these large ON molecules [15,21]. S values were obtained from two gradient times of 5 and 15 min. Elution programs have been optimized by changing % ACN until log (k) > 2.1 and 0.1 < b < 0.5 (with b the LSS gradient steepness). It was important to account for these two constraints to obtain good predictions with this newly developed approach. For all separations, linear gradient modes are used. Unless stated otherwise, the following conditions are employed for all separations: the sample concentration is fixed at 5 μM, the sample injection volume is set to 1 μL with a flow rate of 0.4 mL/min, the concentration of the IP reagent in the mobile phases is maintained at 25 mM, the column temperature is kept at 60 °C, the pH of the aqueous mobile phase A is adjusted to 8, a column with a length of 50 mm is used, and ACN is used as the organic solvent in the mobile phase.
Peak capacities were experimentally determined from the gradient time (tgrad) and the average measured peak width at 50 % height (W50%). The following equation was applied to estimate peak capacity on the basis of the peak width at 13.4 %, corresponding to a resolution of 1 [22]: \(\mathrm{P=1+}\mathrm{t}_\mathrm{grad}\left(\mathrm{1.7\ x\ }\mathrm{W}_{\left(\mathrm{50%}\right)}\mathrm{\ }\right)\mathrm{\ \ }\)
3. Results and discussion
3.1. Comprehensive understanding of on-off elution mechanism of oligonucleotides
3.1.1. Hydrophobicity of IP reagents
3.1.2. Concentration of IP reagents
3.1.3. Mobile phase pH
Journal of Chromatography, A Volume 1739: Fig. 3. Experimental S values as a function of mobile phase pH (pH 7, 8 and 9 were tested) for dT20 (blue) and dT100 (green). The S values were obtained using 25 mM TEA-acetate or HA-acetate in both mobile phases A and B.
3.1.4. Buffering acid selection
3.1.5. Column temperature
3.2. Application to fast analysis of ONs using ultra-short columns
3.2.1. Characterization of a 20-mer ASO sample
3.2.2. Characterization of a large sgRNA product
4. Conclusions
The aim of this study was to better understand the on-off retention mechanism of ONs under IP-RPLC conditions. To achieve this goal, S values of the retention models were calculated for two model samples: 20-mer and 100-mer poly(dT). Firstly, it was observed that S values were consistently higher for the larger ON, as previously demonstrated with small vs. large proteins. Various IP reagents with differing hydrophobicities were tested, confirming that the hydrophobicity of the IP reagent significantly impacts the S value, with variations up to 12-fold. S values strongly decreased with increasing IP reagent hydrophobicity, with DEA and TEA showing more favorable on-off retention mechanisms compared to HA and OA. However, DEA and TEA are not suitable for PS-modified ONs due to partial separation of numerous diastereomers. Secondly, IP concentration also influenced S values, to a lesser extent than IP nature. Higher IP reagent concentrations (100 mM vs. 10 mM) reduced S values by up to 200 % in some cases. On the contrary, the mobile phase pH had a minor impact on S values, with a maximum variation of 30 % between pH 7 and 9. Similarly, the choice of buffering (HFIP or acetate) and mobile phase temperature slightly affected S values, but this behavior was highly dependent on the ON size and the IP reagent used.
The observed behavior can be explained by the retention mechanism of ONs in IP-RPLC, which is influenced by both the%ACN in the mobile phase and the strength of the ion pair. One mechanism may dominate depending on the mobile phase conditions. In all cases, minimizing the impact of the IP on ON retention resulted in higher S values, likely because %ACN becomes the primary driver of retention under these conditions.
Through systematic experiments, optimal mobile phase conditions promoting high S values were identified for a non-PS-modified ASO and a sgRNA product containing some PS modifications. In the latter case, certain compromises were required to ensure the use of a sufficiently hydrophobic IP reagent, which is essential to prevent diastereomers separation of PS-modified ONs. Under these optimized conditions, effective separation of the main species from impurities was accomplished in approximately one minute using ultra-short 5 mm columns, facilitated by the on-off retention mechanism.
