DNA and RNA Chromatography: Enabling mRNA Vaccine Validation

The development of messenger RNA (mRNA) vaccines revolutionized virology, especially during the coronavirus disease 2019 (COVID-19) pandemic. Deoxyribonucleic acid (DNA) stores genetic information, while ribonucleic acid (RNA) helps use this information to make proteins. mRNA carries genetic instructions from DNA to the cell's protein-making machinery. Because mRNA is unstable and difficult to purify, important breakthroughs were needed to produce stable, uncontaminated RNA. Bonn and coworkers developed a stationary phase for separating DNA, while Gjerde and his team made the technology commercially viable and free of metal contamination. Hornby and colleagues then adapted this stationary phase for RNA chromatography. These advances supported the work of Karikó and Weissman and helped Pfizer-BioNTech and Moderna rapidly develop and deploy effective COVID-19 mRNA vaccines.

The advent of mRNA vaccines has brought about a creative shift in vaccinology and infectious disease control. In silico–encoded mRNA vaccines are synthesized far more rapidly than conventional protein antigens. In less than one year from the start of the COVID-19 pandemic, companies such as Moderna and Pfizer-BioNTech responded to COVID-19 by producing a stable, pure supply of mRNA that was successfully formulated into vaccines and met regulatory requirements through robust clinical trials.

In living cells, mRNA is inherently unstable and transient. For life to exist, mRNA molecules carrying old cellular instructions must be recycled (degraded) and replaced by newly synthesized mRNA. Cells use ribonuclease (RNase) enzymes to turn over "spent" RNA. Because these enzymes are ubiquitous, RNA exists only fleetingly before being recycled into new molecules. Prior to this work, mRNA was difficult to manipulate, separate, and purify.

Chromatography was first described by Mikhail Tswett in 1903.¹ Because mRNA is a large macromolecule that is highly susceptible to degradation both in vivo and during in vitro manipulation, a new type of chromatography had to be developed.

Setting the Stage

Guenther K. Bonn was a professor at the University of Linz, Austria, and later at the University of Innsbruck, Austria. In 1988, Bonn was a visiting professor at Yale University, USA, working with Professor Csaba Horváth, considered by many to be the father of high performance liquid chromatography (HPLC).

Along one research path, Horváth designed uniform, nonporous silica beads and, in 1969, was the first to describe a pellicular bead support structure for chromatography media.² The pellicular support concept employs an impermeable core coated with a thin, porous active layer. This structure was proposed to minimize the analyte diffusion limitations present in the fully porous support materials available at the time. Bonn's research focused on developing media and columns for ion chromatography (IC). Columns for IC had to be acid- and base-stable and therefore used polymer-based substrates.

Douglas T. Gjerde's first chromatography work took place when he was a graduate student at Iowa State University, USA, where he worked with his advisor, Professor James Fritz, and Gabriella Schmuckler, a visiting professor from Technion – Israel Institute of Technology, Israel. Their collaboration led to the development of the first non-suppressor IC technology. At Gjerde's first company, Sarasep, Inc., he developed polymer beads and commercialized polymer columns for IC. Gjerde and Bonn first met at a conference hosted by Bonn at the University of Linz and subsequently collaborated on several projects. Bonn later returned to Innsbruck, becoming department head and professor at the University of Innsbruck, where he began work on a new type of nonporous polymer chromatography media.

The Discovery of DNA Chromatography

Drawing on conversations with Horváth, Bonn reasoned that uniform, nonporous polymers could yield high-performance ion chromatography. Professor John Ugelstad at the Norwegian Institute of Technology, Norway, had developed a method to prepare monodisperse polymer particles.³ Ugelstad incorporated metal into the interior of the beads and used affinity ligands for magnetic separation in Dynabeads.

With students Christian Huber and Peter Oefner, Bonn produced similar monodisperse polymer "seeds" and incorporated divinylbenzene monomer into them. The beads were cured and then C18-alkylated. The resulting media were uniform, pH-stable, and significantly harder than other polymer beads, which facilitated packing into stainless-steel HPLC columns.

Initial results showed either low retention or broad peaks on the new polymer columns. It was postulated that the separation of small analytes was limited by the low surface area. Bonn and coworkers reasoned that larger organic molecules would bind more strongly to the resin. A double-stranded DNA (dsDNA) sample was prepared by cutting a plasmid with a sequence-specific enzyme. DNA carries anion-exchange groups, and the ion-pairing reagent triethylammonium acetate (TEAA) was added to a water–acetonitrile gradient system. The ammonium cation pairs with DNA to form a neutral complex that adsorbs to the neutral reversed-phase polymer column. Acetonitrile was used to selectively elute DNA, beginning with smaller fragments and progressing to longer sequences as the acetonitrile concentration increased.

Bonn and coworkers discovered that the desorption isotherms were sharp and that small changes in acetonitrile concentration triggered rapid DNA desorption. This reflected the instantaneous partitioning of DNA between the stationary and mobile phases when the ion pair was released from the resin. Single-stranded DNA (ssDNA) could be resolved to single-base-pair resolution, while dsDNA could be resolved to 5–10 base pairs (bp), depending on fragment size range. Elevated temperature reduced back pressure and improved peak resolution. Figure 1 demonstrates size-based separation of dsDNA fragments.

This work was first published in 1993^4–5 along with a patent.⁶ It was fortunate, though unknown at the time, that the work was conducted in Austria, where the water supply was especially pure.

In 1993, Gjerde's company licensed the patent from Bonn, marking Gjerde's first entry into large-molecule biological separation technology. Gjerde coined the term DNASep to refer to a column designed to separate DNA single and double-stranded DNA. 

Initially, Gjerde had no success using columns produced either by Bonn's laboratory or fabricated by his own team. With the help of Robert Haefele, a student sent by Bonn to Gjerde's laboratory, Gjerde was able to resolve the column and instrument issues and subsequently filed several patents. Gjerde and coworkers identified the critical importance of metal-free reagents, solvents, and equipment for successful DNA separation.

The required water quality approached that used in silicon chip manufacturing. The effects of metal contamination in reagents and polymers are shown in Figures 2a and 2b. When metals were present and accumulated on column frits and media, a dual stationary-phase interaction formed. A weak interaction arose from the C18 alkyl groups interacting with the nonpolar DNA–TEAA ion pair, while a much stronger interaction occurred between metals and the phosphate groups of DNA, effectively destroying the separation. Commercialization of the technology therefore required ultraclean instruments and reagents beyond the capabilities of existing commercial systems.

During this work, Gjerde and coworkers developed a substantial portfolio of intellectual property for the company.^7–11 Temperature-control technology for the eluent stream was introduced, and new column-packing methods were discovered that reduced back pressure while enabling high flow rates and high column capacities.

Transgenomic was formed to commercialize the technology. Dave Hornby, professor of biochemistry at the University of Sheffield, UK, became a scientific advisor. Working independently in the UK but in collaboration with Gjerde and coworkers, Hornby and his team developed numerous applications, later described in the book DNA Chromatography, written with Gjerde as lead author.¹⁴

RNA Chromatography

Professor Hornby first proposed using the technology to separate, collect, and utilize biological RNA. Unlike DNA, which is relatively rugged, RNA is a far more fragile molecule. RNA issues thousands of distinct commands to cells to produce proteins. Each time a new mRNA command is generated, the old RNA must be destroyed to prevent conflicting instructions. Because mRNA is transient, it is continually degraded by ribonucleases present virtually everywhere.

Before Hornby's proposed work began, Transgenomic had employed several RNA-expert molecular biologists. Their RNA studies could not be performed in standard Transgenomic research and development (R&D) chromatography laboratories because RNA degraded too rapidly. They insisted that a dedicated clean room specifically designed for RNA work would be required.

Hornby purchased a commercial transfer RNA (tRNA) mixture from Sigma and injected it onto the column designed to separate DNA single and double-stranded DNA installed on a Transgenomic WAVE HPLC instrument (SpectraLab). The results were remarkable. As shown in Figure 3, an entire pool of E. coli tRNA was separated and purified. Eluted RNA fractions were collected and analyzed by reverse transcription polymerase chain reaction (RT-PCR). The amplified fractions were then sequenced to identify and quantify individual tRNAs. A typical tRNA population contains approximately 60–80 distinct species, each present at a different concentration and specifying different amino acids and codons. Hornby successfully identified every fraction. He further discovered that purified RNA fractions could be reinjected and manipulated without the extraordinary laboratory precautions previously required, which he found both surprising and gratifying.

Hornby and Conroy¹² analyzed bovine ribonuclease by 500-MHz nuclear magnetic resonance (NMR) in the presence and absence of acetonitrile. Chemical shift data showed structural changes consistent with irreversible unfolding of RNase under HPLC conditions. In addition, Gjerde reasoned that RNases are not retained on the column designed to separate DNA single and double-stranded DNA and elute with the solvent front. Thus, collected RNA fractions are free of active RNase.

Gjerde, Hornby, and co-workers were granted a patent covering RNA processes.¹³ A chapter on RNA purification was published in a book,¹⁴ followed by RNA Purification and Analysis: Sample Preparation, Extraction, Chromatography¹⁵ by Gjerde, Hoang, and Hornby, which further explained the technology and its applications. Hornby and Dickman later expanded on RNA separation principles.¹⁶

Transgenomic staff members Azarani and Hecker published the company's first report describing RNA work,¹⁷ while internal R&D efforts continued. Hornby recognized that the column designed to separate DNA single and double-stranded DNA not only enabled RNA separation and purification but also stabilized RNA so that virtually any laboratory could easily manipulate it. Experiments could be performed without RNA species "disappearing." RNA samples stable for only minutes or hours prior to injection remained stable for days, weeks, or even months after separation, even at room temperature.

RNA separations were found to differ from DNA. Figures 4a and 4b show how increasing column temperature reduced RNA secondary structure and sharpened fragment peaks. Work was also performed on double-stranded RNA (dsRNA) (Figure 5).

Using Pure and Stable RNA to Develop the First mRNA Vaccine in Mice

Katalin Karikó focused on RNA research for several decades to develop transcribed mRNA for protein therapy. Karikó and Drew Weissman discovered that nucleoside modifications suppress RNA immunogenicity, making mRNA suitable for vaccines. Their groundbreaking work replaced one of the four nucleotides in mRNA, greatly enhancing its therapeutic potential. In their experiments, mRNA encoding viral spike proteins triggered an immune response in mice of sufficient potency to combat viral infection.^18–20 By modifying mRNA and applying published lipid technologies to stabilize RNA in lipid nanoparticles, Karikó and Weissman made mRNA vaccines feasible. A key patent demonstrated that base modification of mRNA enhanced immune responses.²¹

Karikó emphasized the critical importance of mRNA purification. In a personal communication with Hornby, she noted her exclusive use of the column designed to separate DNA single and double-stranded DNA for the pioneering purification experiments required to demonstrate proof of concept for mRNA vaccines.

Karikó and Weissman discovered that impure mRNA caused excessive immunological responses, rendering vaccines potentially toxic. The antigen-encoding mRNA sequences typically exceed 1500 nucleotides and must be substantially free of RNA contaminants, including incomplete transcripts, spurious longer mRNAs, and dsRNA. In their early work, Karikó and Weissman demonstrated that the column designed to separate DNA single and double-stranded DNA met these stringent purification requirements. They acquired columns from Transgenomic to collect mRNA free of dsRNA contamination.

Karikó and Weissman licensed the modified mRNA technology to Pfizer-BioNTech and Moderna, enabling development of their respective COVID-19 vaccines. In one patent,²¹ Karikó cited the book RNA Purification and Analysis: Sample Preparation, Extraction, Chromatography by Gjerde, Hoang, and Hornby.

Using traditional vaccine development approaches would have required several years. However, enabled by mRNA technology, a COVID-19–specific vaccine was developed and introduced to the market in less than one year.

Conclusion

Establishing a tool to separate and collect pure, stable RNA enabled the development of safe and effective mRNA vaccines. Bonn and coworkers produced and demonstrated the first column. Gjerde and coworkers identified metal–DNA interactions in chromatographic systems and developed a stable, metal-free chromatography platform. Hornby and coworkers demonstrated methods to produce stable RNA under standard laboratory conditions. Karikó and Weissman used the column designed to separate DNA, together with a system to purify modified mRNA and detect and remove dsRNA, to demonstrate practical mRNA vaccines without adverse immunological effects. These advances directly enabled commercial mRNA vaccines for COVID-19 to be developed in less than one year in response to the pandemic.

Acknowledgments

Bonn thanks Peter Oefner and Christian Huber. Gjerde thanks Robert Haefele and Ron Jones. Hornby thanks Mark Dickman, Maryam Matin, Qaiser Sheikh, and Paul Brown. The authors thank Katalin Karikó for her review of the manuscript and for acknowledging the RNASep column's contribution to the development of an mRNA vaccine.

References

1. Tswett, M. S. On a New Category of Adsorption Phenomena and Its Application to Biochemical Analysis. Proc Warsaw Soc Nat Biol Sect 1905, 14 (6), 20–39. [Presented 1903.]
2. Horváth, C.; Lipsky, S. R. Column Design in High Pressure Liquid Chromatography. J Chromatogr Sci 1969, 7 (2), 109–116. DOI: 10.1093/chromsci/7.2.109
3. Ugelstad, J. Process for Preparing Latex. U.S. Patent 4,113,687, 1978.
4. Huber, C. G.; Oefner, P. J.; Bonn, G. K. Rapid Analysis of Biopolymers on Modified Non-Porous Polystyrene–Divinylbenzene Particles. Chromatographia 1993, 37 (11–12), 653–658. DOI: 10.1007/BF02274118
5. Huber, C. G.; Oefner, P. J.; Preuss, E.; Bonn, G. K. High-Resolution Liquid Chromatography of DNA Fragments on Non-Porous Poly(styrene-divinylbenzene) Particles. Nucleic Acids Res 1993, 21 (5), 1061–1066. DOI: 10.1093/nar/21.5.1061
6. Bonn, G. K.; Huber, C.; Oefner, P. Nucleic Acid Separation on Alkylated Nonporous Polymer Beads. U.S. Patent 5,585,236, 1996.
7. Gjerde, D. T.; Haefele, R. M.; Togami, D. W. System and Method for Performing Polynucleotide Separations Using Liquid Chromatography. U.S. Patent 5,772,889, 1998.
8. Gjerde, D. T.; Haefele, R. M.; Togami, D. W. Liquid Chromatography Systems for Performing Polynucleotide Separations. U.S. Patent 5,997,742, 1999.
9. Gjerde, D. T.; Haefele, R. M.; Togami, D. W. Method for Performing Polynucleotide Separations Using Liquid Chromatography. U.S. Patent 6,017,457, 2000.
10. Gjerde, D. T.; Haefele, R. M.; Togami, D. W. Apparatus for Performing Polynucleotide Separations Using Liquid Chromatography. U.S. Patent 6,030,527, 2000.
11. Gjerde, D. T.; Haefele, R. M.; Togami, D. W. Process for Performing Polynucleotide Separations. U.S. Patent 6,156,206, 2000.
12. Hornby, D. P.; Conroy, M. J. University of Sheffield, U.K. Unpublished work, 2000.
13. Gjerde, D. T.; Hornby, D. P.; Hanna, C. P.; Kuklin, A. I.; Haefele, R. M.; Taylor, P. D. Method and System for RNA Analysis by Matched Ion Polynucleotide Chromatography. U.S. Patent 6,576,133, 2003.
14. Gjerde, D. T.; Hanna, C. P.; Hornby, D. P. DNA Chromatography; Wiley-VCH, 2002.
15. Gjerde, D. T.; Hoang, L.; Hornby, D. P. RNA Purification and Analysis: Sample Preparation, Extraction, Chromatography; Wiley-VCH, 2009.
16. Waghmare, S. P.; Pousinis, P.; Hornby, D. P.; Dickman, M. J. Studying the Mechanism of RNA Separations Using RNA Chromatography and Its Application in the Analysis of Ribosomal RNA and RNA:RNA Interactions. J Chromatogr A 2009, 1216 (9), 1377–1382. DOI: 10.1016/j.chroma.2008.12.077
17. Azarani, A.; Hecker, K. H. RNA Analysis by Ion-Pair Reversed-Phase High-Performance Liquid Chromatography. Nucleic Acids Res 2001, 29 (2), e7. DOI: 10.1093/nar/29.2.e7
18. Karikó, K.; Buckstein, M.; Ni, H.; Weissman, D. Suppression of RNA Recognition by Toll-like Receptors: The Impact of Nucleoside Modification and the Evolutionary Origin of RNA. Immunity 2005, 23 (2), 165–175. DOI: 10.1016/j.immuni.2005.06.008
19. Karikó, K.; Muramatsu, H.; Welsh, F.; et al. Incorporation of Pseudouridine into mRNA Yields Superior Nonimmunogenic Vector with Increased Translational Capacity and Biological Stability. Mol Ther 2008, 16 (11), 1833–1840. DOI: 10.1038/mt.2008.200
20. Karikó, K.; Muramatsu, H.; Ludwig, J.; Weissman, D. Generating the Optimal mRNA for Therapy: HPLC Purification Eliminates Immune Activation and Improves Translation of Nucleoside-Modified, Protein-Encoding mRNA. Nucleic Acids Res 2011, 39 (21), e142. DOI: 10.1093/nar/gkr695
21. Karikó, K.; Weissman, D. RNA Containing Modified Nucleosides and Methods of Use Thereof. U.S. Patent 8,278,036, 2012.