**3. mRNA Structure and Implications for Use as a Vaccine**

At this stage, it is perhaps useful to review the structure of the mRNA as designed for drug and vaccine delivery, which incorporates elements to improve both stability and protein expression. The mRNA comprises a 5 cap, a 5 untranslated region (UTR) (also called leader RNA), the coding sequence with a stop signal, a 3- UTR, and a poly(A) tail. This molecule provides the template in the cytoplasm of a cell for translation by the ribosome and tRNA into the encoded protein, making multiple copies of the protein from each mRNA template. This amplification provides a quantitative advantage per molecule compared to providing individual proteins. However, offsetting that numeric advantage is that, in addition to the instability of the mRNA, it is thought that only one out of 10,000 molecules of mRNA will escape an endosome into the cytoplasm [5]. The amplification by translation of the mRNA into protein has to overcome the losses and the inefficiencies of degradation and the transduction process. Another obvious implication for this is that, compared to plasmid DNA, which must enter the nucleus of a cell, the mRNA only needs to be present in the cytoplasm, which eliminates the additional cellular (i.e., nuclear) membrane that plasmid DNA needs to cross. On the other hand, plasmid DNA is more stable than mRNA, and each DNA molecule results in the production of multiple mRNA molecules, thus the theoretical advantages of one over the other boil down to the realities of the net stability of plasmid DNA versus mRNA in their final formulation, as well as the efficiencies of targeting to the desired cell, the transduction to the cytoplasm or nucleus followed by the efficiencies of transcription of the plasmid DNA (resulting in amplification from DNA to mRNA), and the translation of mRNA, whether transcribed from DNA or in vitro-transcribed mRNA, to protein (also resulting in amplification).

Lower quantities of the (antigenic) protein are presumably needed for vaccines (due to amplification of the immune response against the antigen) compared to amounts of protein that might be needed for therapeutic disease targets. Additionally, whereas for gene therapy, where long-lasting or even permanent production of the therapeutic protein is desired, vaccines likely benefit from the transient nature of the antigen (followed by boosting). This is because, for example, the development of high affinity antibodies occurs as antigen becomes scarcer. Subsequent boosts with antigen then expand the production of these high affinity antibodies. The relatively temporary nature and presumably small amounts of protein produced by mRNA would fit with this paradigm if the mRNA is present in great enough quantities, persists, and is active long enough to produce sufficient amounts of protein antigen to stimulate the desired immune responses. DNA vaccines likewise have been demonstrated to produce the encoded protein for a limited period of time, although this is likely longer than mRNA constructs given the greater inherent stability of plasmid DNA compared to mRNA. Plasmid DNA has been shown to persist in muscle up to six months in a non-integrated fashion [18].

As noted above, the ability to make either a plasmid DNA or an mRNA construct quickly by simply knowing the genetic sequence of a desired antigen makes plasmid DNA or mRNA much faster technologies (compared to current approaches) to produce a vaccine, if needed, for an epidemic or an emerging disease. Five characteristics—rapidity of making constructs, relatively temporary presence in vivo of the encoded protein, amplification by the immune system responding to even small amounts of expressed protein, the manufacturing advantages (generic and rapid processes compared to drugs or recombinant proteins), and the intrinsic immunostimulatory properties of both plasmid DNA and mRNA—combine to make a compelling rationale for vaccines to be viewed as the best initial targets for widespread development efforts for both technologies.

## **4. Manufacture**

The manufacture of plasmid DNA has been considered to be one of its strengths, making it a platform technology where the same process could essentially be used regardless of the gene that was encoded [19]. Moreover, the process of bacterial fermentation is fairly simple, since the product is a plasmid grown in bacteria, such as *Escherichia coli*, and the plasmid DNA is relatively stable, making purification straightforward. This is in contrast to the time-consuming process of earlier generation vaccines, which required finding ways to grow the pathogen such as making it weaker or inactivating it. Historically, the process to develop vaccines, including the manufacturing process, has been long and could reach up to decades (e.g., the chicken pox vaccine). The advent of recombinant proteins provided a simpler means of making vaccine antigens, and one that eliminated the need to work with a virulent pathogen during manufacture. However, this still had drawbacks, such as ensuring that the antigen had any crucial antigenically correct post-translational modifications (such as glycosylations), which can differ between host cells (such as yeast or baculovirus compared to humans), that the antigen was properly folded, and so on. Recombinant proteins generally also need to be soluble, providing a challenge for proteins with a transmembrane domain that is needed either antigenically or for any necessary oligomerization [e.g., HIV envelope]. Recombinant proteins administered exogenously (e.g., given in an immunization) also have an inherent limitation of not stimulating Major Histocompatibility Complex (MHC) Class I-restricted Cytolytic T Lymphocytes (CTLs), as is discussed later.

mRNA is made by in vitro transcription starting from a linearized DNA template, performing in vitro transcription, then getting rid of the template by digestion with DNAses, at which point the mRNA can be purified. Manufacturing mRNA by in vitro transcription is thus even more appealing than manufacturing plasmid DNA because while it is also a generic process, (i.e., independent of the gene insert), it is essentially a chemical process with no animal or cellular components (although the cost is potentially greater [20]). A graphic detailing the various steps and suggested possible improved processes can be seen in the reference [21]. The manufacturing process might be guided by pharmaceutical product Good Manufacturing Practice (GMP) guidelines [22,23] rather than those for biologicals [24], a likely advantage. Of course, any formulations or the addition of immunomodulators, adjuvants, or delivery systems may increase the complexity and cost of the manufacture for either mRNA or plasmid DNA.

#### **5. Stability as a Product**

DNA vaccines as a manufactured entity are noted for their stability [25], particularly when supercoiled. As noted above, this stability is reflected even in vivo, since plasmid has been detected in a non-integrated form in muscle up to six months following injection [18]. Although the ubiquity of RNAses with the resulting instability of native mRNA has been a significant reason for the delay in development of mRNA, the actual manufactured mRNA is stable in liquid or lyophilized form, (reported to be stable up to two years at room temperature [26]) with an inverse relationship of stability to temperature. It has been reported that a rabies mRNA vaccine was still effective for pre-clinical protection after several months at temperatures ranging from −80 ◦C to as high as +70 ◦C [26]. The stability of mRNA as a vialed product (i.e., protected from RNAses) is a separate consideration from the stability in vivo, and any formulations or delivery devices (which may cause shearing during delivery) are key factors in the final stability as a delivered product.
