Purification of linearized template plasmid DNA decreases double-stranded RNA formation during IVT reaction

Abstract

After the COVID-19 pandemic, messenger RNA (mRNA) has revolutionized traditional vaccine manufacturing. With the increasing number of RNA-based therapeutics, valuable new scientific insights into these molecules have emerged. One fascinating area of study is the formation of double-stranded RNA (dsRNA) during in vitro transcription (IVT) which is considered a significant impurity, as it has been identified as a major trigger in the cellular immune response pathway. Therefore, there is a growing importance placed to develop and optimize purification processes for the removal of this by-product. Traditionally, efforts have primarily focused on mRNA purification after IVT through chromatographic separations, with anion exchange and reverse phase chromatography emerging as effective tools for this purpose. However, to the best of our knowledge, the influence and significance of the quality of the linearized plasmid have not been thoroughly investigated. Plasmids production involves the growth of bacterial cultures, bacterial harvesting and lysis, and multiple filtration steps for plasmid DNA purification. The inherent complexity of these molecules, along with the multitude of purification steps involved in their processing, including the subsequent linearization and the less-developed purification techniques for linearized plasmids, often result in inconsistent batches with limited control over by-products such as dsRNA. This study aims to demonstrate how the purification process employed for linearized plasmids can impact the formation of dsRNA. Several techniques for the purification of linearized plasmids based on both, resin filtration and chromatographic separations, have been studied. As a result of that, we have optimized a chromatographic method for purifying linearized plasmids using monolithic columns with C4 chemistry (butyl chains located in the surface of the particles), which has proven successful for mRNAs of various sizes. This chromatographic separation facilitates the generation of homogeneous linearized plasmids, leading to mRNA batches with lower levels of dsRNA during subsequent IVT processes. This finding reveals that dsRNA formation is influenced not only by RNA polymerase and IVT conditions but also by the quality of the linearized template. The results suggest that plasmid impurities may contribute to the production of dsRNA by providing additional templates that can be transcribed into sequences that anneal with the mRNA molecules. This highlights the importance of considering the quality of plasmid purification in relation to dsRNA generation during transcription. Further investigation is needed to fully understand the mechanisms and implications of plasmid-derived dsRNA. This discovery could shift the focus in mRNA vaccine production, placing more emphasis on the purification of linearized plasmids and potentially saving, in some instances, a purification step for mRNA following IVT.

Keywords: IVT; chromatography; dsRNA; immunogenicity; mRNA; polishing; vaccines.

FIGURE 1

dsRNA generation mechanisms during IVT. (A) Template strand of DNA is transcribed into mRNA, 3’-extreme loops due to complementarity and T7RNAP uses mRNA as template to continue transcription. (B) Polymerase might switch to the non-template strand and generate the complementary strand of the original mRNA. (C) Abortive fragments during IVT with complementary sequences anneals with the run-off transcript.

FIGURE 2

Comparison of the influence of the template purification method in the content of dsRNA in COVID transcripts. Pointed line represents threshold 1%.

FIGURE 3

Degradation of mRNA is influenced by the type of purification method employed. Electropherogram comparison between tandem Oligo-dT + AEX (blue line) vs. AEX + Oligo-dT (red line). Blue line shows more degraded profiles. Mechanism of intramolecular hydrolysis of RNA phosphodiester bond.

FIGURE 4

%dsRNA obtained for GFP, Fluc and COV. Linearized DNA was isolated by Wizard^® Megapreps DNA Purification Resin, Promega, Cat#A7361 and RNA crudes were purified by POROS™ Oligo (dT)25 Affinity column. Pointed line represents threshold 1%.

FIGURE 5

AEX-HPLC (CIMac™ pDNA 0.3 mL Analytical Column (1.4 µm)) profile comparison GFP (green line), Fluc (red line) and COV (blue line) linearized plasmids purified by resin (0.5ug each template injected). Chromatogram shows more intense impurity peaks for COV and GFP.

FIGURE 6

Template characterization. (A) AEX-HPLC (CIMac™ pDNA) analytical profile of linear DNA profiles for encoded GFP, Fluc and COV purified by Megapreps DNA Resin (blue lines) and CIMmultus C4 HLD column (red lines) (0.5ug each template injected). (B) Capillary electrophoresis for the same samples.

FIGURE 7

Comparison of dsRNA content related to the template purification method and transcript size. Corresponding dot blot membranes.

FIGURE 8

dsRNA generation mechanisms during IVT. Not removed impurities of DNA might be treated as the linearized complete plasmid producing sequences of RNA that could anneal with the original transcript. That might lead to a population of different dsRNA species.