A New Purification Strategy For The Removal Of dsRNA Impurities From IVT mRNA Products

By Kyle Tynan, NRC Postdoctoral Researcher, National Institute of Standards and Technology (NIST)

One of the major challenges facing messenger RNA (mRNA) therapeutics is the presence of double-stranded RNA (dsRNA) impurities. Although these impurities typically represent only a small fraction of the total RNA produced during in vitro transcription (IVT), they can have a disproportionate impact on the safety and performance of mRNA therapeutics. dsRNA is recognized by the innate immune system as a molecular signature of viral infection, triggering inflammatory signaling pathways that can reduce protein expression, decrease therapeutic efficacy, and contribute to adverse immune responses.^1-3 Consequently, effective removal and characterization of dsRNA impurities has become one of the most important quality control challenges in the manufacturing of mRNA vaccines and therapeutics.

The difficulty in dsRNA identification and removal from IVT products is the fact that dsRNA impurities closely resemble the desired mRNA product in many of its physicochemical properties. Both molecules are composed of nucleic acids, possess similar charge densities, absorb ultraviolet light at 260 nm, and often exist over overlapping size ranges. Unlike protein impurities that may differ substantially in charge or hydrophobicity, dsRNA and mRNA frequently behave similarly during conventional purification techniques.

Furthermore, dsRNA impurities are not necessarily a single molecular species. They may consist of partially duplexed transcripts, self-complementary regions, antisense products, or complexes involving other IVT byproducts. This heterogeneity makes both purification and analytical characterization challenging.^4,5 While several purification strategies have been introduced over the past decade, each possesses limitations.

Cellulose chromatography selectively retains dsRNA through interactions with cellulose under ethanol-containing conditions.⁶ While effective, the technique presents practical challenges related to column preparation and scalability for manufacturing applications.

Low-pH denaturation represents another strategy. Under acidic conditions, hydrogen bonding within dsRNA is disrupted, after which oligo(dT) chromatography can separate the desired polyadenylated mRNA from denatured impurities.⁷ Although effective, exposing mRNA to pH values near 3.5 raises concerns regarding RNA stability and potential degradation.

Enzymatic digestion using RNase III provides a more targeted approach by selectively digesting dsRNA. However, enzymatic treatments introduce additional purification requirements because the enzyme itself must subsequently be removed. There is also the possibility of unintended digestion of therapeutic mRNA.⁸

Reverse-phase ion-pairing high-performance liquid chromatography (RP-IP-HPLC) can remove dsRNA but typically requires elevated temperatures, organic solvents, and often yields relatively low recovery of the target mRNA.⁹ Anion-exchange chromatography performs well for smaller RNA molecules but becomes less effective for large mRNA constructs.¹⁰ More recently, affinity-based dsRNA scavenger resins have demonstrated excellent scalability and efficient removal, although some nonspecific binding of mRNA has been reported.¹¹

Because no single purification strategy is ideal, there remains considerable interest in alternative approaches that combine simplicity, high recovery, and compatibility with standard laboratory instrumentation.

Size-exclusion chromatography (SEC) is among the simplest chromatographic separation techniques because it relies primarily on molecular size rather than chemical interactions. SEC columns are packed with porous particles containing an interconnected network of pores. As molecules travel through the column, smaller molecules diffuse into these pores and therefore follow longer pathways before exiting the column. Larger molecules cannot fully enter the pores and instead travel around them, resulting in earlier elution. SEC is considered a relatively gentle purification method that can preserve sensitive biomolecules while achieving high recoveries, around 80 %.¹²

The work by Tynan et al. introduces an SEC-based approach for dsRNA impurity removal, demonstrating that a relatively simple chromatographic technique can effectively separate detectable dsRNA impurities from IVT mRNA while maintaining high recovery of the desired product.¹³ Rather than positioning SEC as an immediate manufacturing solution, the study highlights its value as a laboratory-scale purification method that may a facilitate improved characterization of dsRNA impurities.

Initial chromatographic experiments identified two principal peaks. The later peak corresponded to the desired mRNA transcript, while an earlier-eluting impurity peak was hypothesized to contain dsRNA impurities. Enzyme-linked immunosorbent assay (ELISA) measurements confirmed that the IVT product contained 1.68% dsRNA by mass, establishing the presence of measurable dsRNA impurities prior to purification.

Although the initial chromatographic separation was encouraging, complete baseline resolution had not yet been achieved. We investigated whether modest changes in mobile phase conditions could improve separation.

Interestingly, reducing the buffer pH only slightly — from 6.44 to 6.33 — improved chromatographic resolution, increasing the resolution factor from 1.2 to 1.6, producing baseline resolution between the impurity peak and the mRNA peak. Fractions corresponding to each peak were subsequently collected and analyzed using the dsRNA ELISA. All collected impurity fractions contained detectable dsRNA above the assay limit of detection, whereas none of the purified mRNA fractions contained detectable dsRNA.

While these findings demonstrate successful separation, the mechanism responsible remains uncertain. We originally hypothesized that lowering the pH might compact the mRNA structure while leaving dsRNA relatively unchanged, thereby altering hydrodynamic size. However, the observed chromatographic behavior did not fully support this explanation.

Alternative possibilities include subtle reductions in analyte-column interactions or structural changes affecting the impurity fraction rather than the mRNA itself. Importantly, the study emphasizes that all proposed mechanisms remain speculative and require further investigation. Understanding precisely why the separation occurs could enable future optimization of buffer composition, pore size, and chromatographic conditions to improve both resolution and recovery.

The method also demonstrated encouraging analytical performance. Approximately 81.8% of the target mRNA was recovered following purification, although variability between experiments remained relatively high because fractions were collected manually. We suggest that automated fraction collection could substantially improve reproducibility and increase recovery consistency in future implementations.

To ensure that purification itself did not damage the therapeutic RNA, purified samples were analyzed using RP-IP-HPLC. Compared with crude material, purified mRNA exhibited narrower chromatographic peaks, disappearance of higher molecular weight impurity peaks, and preservation of the principal mRNA peak, suggesting that SEC removed contaminants without compromising transcript integrity.

Although the SEC method presented in this study is currently intended for laboratory-scale purification, its broader implications extend beyond simple cleanup of IVT products.

First, the work demonstrates that conventional SEC instrumentation can effectively remove detectable dsRNA impurities without relying on harsh solvents, enzymatic digestion, or acidic denaturation. Because SEC is already widely available in many analytical laboratories, implementation requires relatively little specialized equipment.

Secondly, the approach provides a valuable research tool for studying dsRNA impurities themselves. One of the persistent obstacles in the field has been the inability to isolate impurity fractions for detailed characterization. By physically separating the impurity fraction from the therapeutic mRNA, SEC creates opportunities to analyze the composition of these fractions using orthogonal analytical methods such as sequencing or mass spectrometry. Such studies could reveal whether the impurity peak contains distinct classes of dsRNA impurities. Better understanding of these impurities could ultimately improve IVT process design and reduce impurity formation at the source.

Overall, this work represents an important contribution to the evolving field of mRNA purification. Rather than replacing existing purification technologies, it introduces SEC as a complementary platform that combines gentle operating conditions, relatively high recovery, and effective removal of detectable dsRNA impurities. Perhaps more importantly, the method provides researchers with a practical means of isolating dsRNA impurity fractions for further study. As the mRNA therapeutic landscape continues to mature, improved understanding of dsRNA impurities — and the development of accessible tools to remove and characterize them — will remain central to advancing the safety, efficacy, and manufacturability of RNA medicines.

References

1. Jo, S.; Lee, J.; Park, S. I.; Kim, Y.; Yoon, S.; Lee, S.; Cho, S.; Cho, Y.; Oh, A.; Ha, D.; Choi, H.; Kim, J.; Lee, Y.; Lee, S.; Lee, S. Y.; Choi, E. J.; Roh, G.; Lee, Y. S.; Bae, S. H.; Jeon, S.; Park, H. J.; Nam, J. H. A Systemic Evaluation of the Effect of DsRNA Contamination on MRNA Vaccine Expression and Immunogenicity. Journal of Controlled Release 2026, 389. https://doi.org/10.1016/j.jconrel.2025.114471.
2. Quan, Y.; Yang, H.; Li, W.; Li, L. MRNA Vaccines: Immunogenicity and Quality Characteristics. Journal of Nanobiotechnology 2026, 24 (1). https://doi.org/10.1186/s12951-025-03800-5.
3. Nowak, C. J. A.; Liu, S.; Falconer, R. J.; Gerstweiler, L. Process and Analytical Strategies for the Safe Production of MRNA Vaccines and Therapeutics. Mol. Biol. Rep. 2026, 53 (1). https://doi.org/10.1007/s11033-026-11455-0.
4. Liu, X.; Hu, C.; He, Q.; Bai, Y.; Zhang, X.; Fu, Z.; Ma, X.; Xu, M.; Liang, Z.; Mao, Q. Research Progress on Immune Mechanism and Control Strategy of DsRNA Impurities in MRNA Vaccine. Expert Rev. Vaccines 2025, 24 (1), 457–469. https://doi.org/10.1080/14760584.2025.2510335.
5. Clark, N. E.; Schraut, M. R.; Winters, R. A.; Kearns, K.; Scanlon, T. C. An Immuno-Northern Technique to Measure the Size of DsRNA Byproducts in in Vitro Transcribed RNA. Electrophoresis 2024. https://doi.org/10.1002/elps.202400036.
6. Baiersdörfer, M.; Boros, G.; Muramatsu, H.; Mahiny, A.; Vlatkovic, I.; Sahin, U.; Karikó, K. A Facile Method for the Removal of dsRNA Contaminant from In Vitro-Transcribed mRNA. Molecular Therapy - Nucleic Acids 2019, 15, 26–35. https://doi.org/10.1016/j.omtn.2019.02.018.
7. Puc, J.; Mencin, N.; Krušič, A.; Nett, E.; Perković, M.; Sahin, U.; Štrancar, A.; Sekirnik, R. pH Denaturation of dsRNA: A Novel Approach to mRNA Purification. Separation and Purification Technology 2026, 390, 136864. https://doi.org/10.1016/j.seppur.2026.136864.
8. Siew, Y. Y.; Zhang, W. Removing Immunogenic Double-Stranded RNA Impurities Post in Vitro Transcription Synthesis for mRNA Therapeutics Production: A Review of Chromatography Strategies. Journal of Chromatography A 2025, 1740, 465576. https://doi.org/10.1016/j.chroma.2024.465576.
9. Weissman, D., N.; Pardi, H. M.; Karikó, K.; HPLC Purification of In Vitro Transcribed Long RNA. Methods in Molecular Biology 2013, 969, 43-54. https://doi.org/10.1007/978-1-62703-260-5_3
10. Peršic, Š., U. Cernigoj, D. Dolenc, and P. S. Gagnon. 2021. A Method of Single-­Stranded RNA Purification. International Patent Application WO2021209595A2.
11. Clark, N. E.; Kozarski, M.; Asci, S. D.; Van den Heuvel, J.; Schraut, M. R.; Winters, R. A.; Kearns, K.; Scanlon, T. C.; Dillen, S. Removal of dsRNA Byproducts Using Affinity Chromatography. Molecular Therapy Nucleic Acids 2025, 36 (2), 102549. https://doi.org/10.1016/j.omtn.2025.102549.
12. Miklavčič, R.; Megušar, P.; Kodermac, Š. M.; Bakalar, B.; Dolenc, D.; Sekirnik, R.; Štrancar, A.; Černigoj, U. High Recovery Chromatographic Purification of mRNA at Room Temperature and Neutral pH. International Journal of Molecular Sciences 2023, 24 (18), 14267. https://doi.org/10.3390/ijms241814267.
13. Tynan, K. J.; Mouchahoir, T.; Lowenthal, M. S.; Phinney, K. W. Purification of Double‐Stranded RNA Impurities From In Vitro‐Transcribed mRNA Using Size‐Exclusion Chromatography. Biomedical Chromatography 2026, 40 (7). https://doi.org/10.1002/bmc.70528.

About The Author

Kyle J. Tynan, Ph.D., is a national research council postdoctoral fellow at the National Institute of Standards and Technology (NIST), where he is a member of the bioanalytical science group in the Materials Measurement Laboratory. His research focuses on developing analytical methods to improve the characterization, quality, and standardization of nucleic acid therapeutics. His current work includes advancing analytical approaches for the characterization of dsRNA impurities in in vitro transcribed mRNA, investigating protein corona formation on mRNA lipid nanoparticles, and developing reference materials to support the RNA therapeutics community. He earned his doctorate from Carnegie Mellon University, where he developed novel analytical techniques for the rapid assessment of critical quality attributes in mRNA and mRNA lipid nanoparticles. He can be reached at kyle.tynan@nist.gov.