Quality Control in IVT RNA Vaccine Workflows

While in vitro transcribed (IVT) RNA can be used for a myriad of purposes from biochemical analyses to genetic studies, it has garnered significant attention recently for its role as the active portion of several very successful COVID-19 vaccines. Although fairly straightforward in principle—essentially just RNA in a delivery vehicle—there are several steps in the production of mRNA vaccines that should be monitored to assure that the correct product, in sufficient quantity, with the desired purity, is being made.

Here we review the role of parallel capillary electrophoresis as a quality control (QC) tool that can help to optimize and streamline the IVT RNA vaccine workflow and ultimately save time and money, and maximize resources.

What is IVT?

In vitro transcription is the process by which RNA is synthesized by an RNA polymerase from a DNA template outside of a cellular environment. This allows for a much more controlled process, and readier access to the RNA product, than would be the case in vivo.

The DNA template can theoretically be a PCR fragment, but for vaccine production “this is not done because PCR is not perfect every replication,” says Steve Siembieda, Product Marketing Director, Parallel Capillary Electrophoresis Instruments, at Agilent. “In the pharmaceutical industry they want to clone exactly one fragment and let the plasmids be replicated in E. coli, which do a much better job of proofreading.”

The plasmids contain the gene of interest (GOI) along with the upstream and downstream material needed for the GOI to be translated in the vaccine’s animal recipient. Plasmids are extracted from their bacterial factory, linearized, and transcribed into what will be the drug substance.

Why QC?

“Though at times, people may see QC as an extra cost, it truly can help people make good decisions on whether they should proceed to the next step of what they're doing,” Siembieda explains. Because if they do not know the quality of the material that they are working on, and they move to the next step, they may manufacture something at the end that does not have the quality they want, or be built at the efficiency they want, he adds. “In this case with IVT RNA, it makes a big difference.”

Siembieda points to four steps in the RNA-based vaccine workflow that QC might be most helpful. At these junctures it should be decided whether the desired quality and quantity has been achieved. If not, perhaps the process can be revisited and optimized before proceeding.

QC, step 1

The plasmid needs to be linearized to allow it to act as a template for the RNA-based vaccine, otherwise the RNA polymerase will keep circling around the plasmid, producing transcripts far longer than desired. Knowing that the restriction enzyme has cleaved the plasmid “to a high percentage of linear fragments—so no supercoiled or open circular DNA—will assure the person doing the production that they are actually going to get large volumes of the transcript they want,” Siembieda says.

An electrophoresis-based tool such agarose gel electrophoresis can be used to detect the DNA in a sample, but it is low throughput with relatively low resolution and low sensitivity when compared with an instrument such Agilent’s parallel capillary electrophoresis-based Fragment Analyzer. Here DNA fractions are separated by size, and these compared with the expected size. The shape of each peak in an electropherogram indicates the purity of that fraction, while its height is a measure of its quantity. The presence of peaks in addition to the main (expected) one may represent plasmids that have not been, or have been only partially, linearized.

Orthogonal techniques, such as high-performance liquid chromatography (HPLC), can similarly be used to ascertain size, purity, and integrity of DNA in a sample, but typically separate only a single sample at a time (compared with the Fragment Analyzer’s up to 96).

Making RNA

After cleaning up the linearized plasmid, the next step in making an RNA vaccine is to add RNA polymerase and allow it to transcribe the DNA. “But the RNA polymerase has a habit of falling off. So, you’re making full-length transcripts but also truncated transcripts. And there may be some degradation of the transcripts as well,” Siembieda notes.

So his second QC recommendation is to see how much IVT RNA was made, and how much unwanted material is also present. An electropherogram will show full transcript along with truncated and degraded products. It will also show if there is any double-stranded DNA still present, indicating that insufficient DNAse was added during RNA cleanup. This particular QC analysis step can be performed at-line, if desired, using a quicker but lower-resolution instrument such as Agilent’s TapeStation.

There are two ways to add a poly(A) tail to an IVT RNA: it can be engineered into the DNA template, or it can be added after transcription using poly-adenylase. Regulators prefer the former because it will yield a defined length for the transcript, “so most [biopharmaceutical companies] work with poly(T) in the plasmid,” Siembieda says. His third QC recommendation is to examine the transcript including the poly(A) tail. Engineering the tail into the plasmid should result in a single discrete peak in the electropherogram.

Siembieda’s fourth QC recommendation is to look at the drug substance—the IVT RNA—both as naked RNA and within the lipid nanoparticle (LNP) delivery vehicle, “so they know they haven’t lost anything. It hasn’t changed. It hasn’t degraded.”

Other attributes, other techniques

Parallel capillary electrophoresis excels at determining the size, purity, and integrity of nucleic acids up to 10,000 nucleotides long. Yet there are other critical quality attributes (CQAs) of IVT RNA that other techniques and instrumentation are more suited to addressing—such as determining the sequence of the RNA itself or its ability to express the target protein. Similarly, various flavors of chromatography, mass spectrometry, spectrophotometry, or a pH meter, all can query different characteristics of the product. Ultimately, it is up to the manufacturer—and regulators— to determine the questions to be asked and how to answer them.