Summary
-
- Methodology for Purity Determination: The article outlines a workflow for determining the purity of synthetic oligonucleotide samples using Byos software, based on the ion-pair HPLC method developed by Rentel et al.
- Advanced Data Analysis: This Byos workflow leverages advanced deconvolution algorithms within the Oligo Intact module and performs quantification using extracted ion chromatogram (XIC) areas in the Byologic module, providing an intuitive platform for analyzing and quantifying impurities in oligonucleotide products.
- Custom Reporting and Quantification: The software's reporting capabilities allow for custom calculations and easy export of results, streamlining the analysis, saving time and ensuring accurate, GMP-compliant quality control for oligonucleotide therapeutics.
Introduction
Synthetic oligonucleotides are gaining prominence as therapeutics, with more than 10 such drugs receiving regulatory approval in recent years. Ensuring the purity and identifying potential impurities in these drugs are critical components of quality control, directly impacting patient safety. Various analytical methods have been developed to detect and quantify product-related impurities, including polyacrylamide gel electrophoresis and capillary zone electrophoresis. In recent years, ion-pair (IP) reversed-phase high-performance liquid chromatography (HPLC) has emerged as the dominant technique for separating the oligonucleotide drug from its impurities. However, chromatographic methods alone are often insufficient for complete impurity detection due to the chemical similarities between the drug and its impurities, making full resolution of all species challenging. To address this, Rentel et al. (2022) [1] developed an IP-HPLC method coupled with mass spectrometry (MS), which allows for the identification and quantification of co-eluting impurities. This assay is robust, GMP-compliant, and universally applicable to oligonucleotides. However, a tailored data analysis tool is still lacking, meaning that data analysis is often performed manually. In this work, we present a workflow built in Byos from ProteinMetrics, which automates the data analysis process from raw sample to final report, streamlining the process and improving efficiency.
Material and Methods
A detailed description of the materials and methods can be found in the original work by Rentel et al. (2022) [1]. While the materials remained unchanged, the methods used in this study differ. Specifically, only two MS methods were utilized in the original study, whereas we introduced a third MS method here. The same liquid chromatography (LC) method was applied across all conditions (Table 1).
MS Method # | Target Chromatographic Peak | Mass Range | MS Condition |
1 | Main Peak | Narrow | Standard |
2 | Main Peak | Narrow | Harsh |
3 | Early Eluting Peaks | Wide | Standard |
Table 1. Overview of the MS conditions.
The ion-pairing agent tributylammonium acetate (TBuAA) was included in the mobile phases to influence the charge state distribution. The presence of TBuAA shifts the distribution from a typical bell-shaped curve to one with fewer charge states, predominantly favoring a single charge state. A typical charge state distribution with TBuAA is shown in Figure 1. This reduction in charge states minimizes interference, allowing for more accurate impurity detection.
For two aliquots of each sample, a narrower mass range was used, assuming all detected peaks correspond to a charge state of -4. These analyses were performed under two ion source conditions: standard and harsh. Under harsh conditions, additional energy was applied to enhance the differentiation of impurities from ion adducts, which are more unstable under these settings.
The third aliquot was analyzed with a wider mass range under standard MS conditions. Impurities that elute earlier than the main product typically have lower molecular weights, and thus a broader mass range was used to ensure their detection.
An important consideration in this method is the use of a low-resolution single-quadrupole mass spectrometer. While other systems can provide spectra of higher mass resolution, this system was selected for its robustness, simple design, and ease of GMP qualification.
Figure 1. Typical charge state distribution after the addition of the ion pairing agent TBuAA.
Analysis in Byos
This Byos workflow utilizes the Oligo module for species identification and Byologic for quantification, following a streamlined three-step process: Project Creation, Data Inspection, and Results Reporting. A customized workflow designed specifically for this application is available upon request by contacting support@proteinmetrics.com.
An overview of the entire workflow is provided in Figure 2, with detailed descriptions of each step included here. In summary:
- On project creation, three Byos files are generated: two oligo intact files and one empty Byologic file. The intact files separately handle the main peak and early-eluting peaks due to differences in deconvolution settings (for further details see this article).
- Species identification is conducted within the Oligo Intact project for both the main and early-eluting peaks. The species are identified through deconvolution of mass spectra. Results from these analyses are exported as CSV files.
- These CSV files can be modified as needed to include undetected species that require monitoring. The modified CSV files are then imported into the empty Byologic project file.
- In Byologic, all species are quantified, and the final data is presented in a UVxMS report.
Figure 2. Overview of the workflow.
Analysis in Byologic
For each species in Byologic, an XIC is generated to facilitate quantification. Additionally, Byologic provides tools to review results and validate each species individually, as illustrated in Figure 3.
Figure 3. Validating species in Byologic.
Reviewing Impurities and Stacked Plots
Stacking XIC plots allows for the comparison of profiles acquired under different MS conditions (see Materials and Methods). For example, as shown in Figure 4, one species remains intact, indicating it is not an adduct, whereas another exhibits reduced signal intensity under harsher conditions, suggesting it is likely an adduct.
Figure 4. Differentiating adduct from chemical impurities.
Purity Calculation
Once data inspection is complete, further processing occurs in the report section. The purity of the sample is calculated using the UVxMS value. Because multiple species can co-elute to form a single chromatographic peak, UV area alone is insufficient for accurate purity calculation. Instead, by quantifying the fractions of each species based on XIC intensity, a more precise purity calculation is made. The report performs this calculation and outputs the results in a table (see Figure 12).
Figure 12. UVxMS report.
Conclusion
This article describes a streamlined approach to assess the purity of a synthetic oligonucleotide sample using a customized workflow in Byos, based on the approach outlined by Rentel et al. By leveraging our software tools, we take advantage of the advanced deconvolution algorithm provided by the Intact Oligo module, and enable precise quantification in the user-friendly Byologic environment. The software’s reporting capabilities further enhance this process, allowing for custom calculations and seamless result reporting.
Acknowledgments
With thanks to Edward Wilkinson and David Benstead of AstraZeneca for contributing data and discussion. The contributions do not constitute an endorsement of the software by AstraZeneca. (Chemical Development, Pharmaceutical Technology & Development, AstraZeneca, Macclesfield, UK)
References
[1] Rentel, C.; Gaus, H.; Bradley, K.; Luu, N.; Kolkey, K.; Mai, B.; Madsen, M.; Pearce, M.; Bock, B.; Capaldi, D. Assay, Purity, and Impurity Profile of Phosphorothioate Oligonucleotide Therapeutics by Ion Pair–HPLC–MS. Nucleic Acid Therapeutics 2022, 32 (3), 206–220. https://doi.org/10.1089/nat.2021.0056.