In the present episode I will give some background on mRNA vaccines, with some emphasis on a possible extension towards Influenza. I will also come back on some interesting data with regard to COVID vaccines.
After the present Episode, I intend to take a break of a few weeks to focus on another challenge.
Review on mRNA vaccines
Ep 184-1: Pilkinton in Acta Biomaterialia (June 2021) provides a nice didactic insight in the mechanism of action and recent advances in the technology that made mRNA so successful.
As I know very well from experience of working with mRNA, it is intrinsically unstable, because both the extra- and intracellular environment is full of RNAses (as our most primitive defense against RNA viruses). Therefore it is difficult to deliver and it induces innate immune responses (including type 1 IFN) that can hamper the generation of adaptive (T and B) responses, when induced too much too early.
Table 1 provides a comparison between DNA and mRNA (encoding a particular antigen) as a vaccination strategy. There are many differences, but one can safely state that the major advantage of DNA is it’s stability and the major disadvantage that it has a potential risk of host DNA modification.
The strategy to optimize mRNA includes:
- Addition of untranslated regions of genes with long-term expression (alpha and beta globins) + a long poly-A tail and efficient capping
- Nucleoside modification: replacement of common nucleosides (for instance uracil) by analogues (1-methyl-pseudo-uridine) which lower the induction of type 1 IFN and reduction of sensitivity to RNAse-L and protein kinase R (acts as a translation inhibitor).
- Another possibility is use of “self-amplifying RNA” (SAM): the vaccine gene is inserted after the viral replicase of an alpha virus. This will prolong the expression of the mRNA, but it is quite possible that immune responses against the viral replicase will limit the response to a booster or to a 2nd, 3rd … vaccine encoding a different pathogen, but using the same replicase in a SAM format.
Fig 1 provides a nice overview of how mRNA stimulates the adaptive immune system.
Equally (or even more) important is the formulation: after a long history of trial-and-error the format of “lipo-nano particles” (LNPs) shown in Fig 2 is now the standard:
- An important key is the choice of the best “ionizable” lipid, that can bind the negatively charged mRNA. Fig 4 shows the evolution from the historically used cationic and ionizable DOTAP etc to the “modern” next generation; optimized lipids.
- Poly-ethylene glycol (PEG) provides stability and prevents interaction with serum proteins, including avoidance of “opsonization” = interaction with complement
- A “helper” phosphatidylcholine and cholesterol are needed to stabilize the LNP in the extra-cellular milieu, but also enhance the release of the mRNA from the endosomes into the cytoplasm (where translation to the actual immunogenic protein can proceed).
As you can see in Fig 3, it is quite a delicate process to mix these hydrophilic and hydrophobic components in such a way that you get stable and reproducible LNPs in the end !
Size matters! As illustrated in Fig 5 and 6, LNP diameter should be between 10 and 100 nm to travel to lymph nodes and penetrate there into the cortical area, where T and B responses are induced.
Moreover intradermal (ID) administration may have an advantage as compared to intramuscular (IM) injection. The skin is richer in antigen-presenting cells (dendritic cells or DC) than the muscle and therefore immune responses start already at the site of injection. The antigen-bearing skin DC then travel to lymph nodes and trigger T and B responses.
Other routes (e.g. intranasal) or targeting strategies are being developed as well.
The remainder of the review is on Influenza mRNA (discussed in more detail below) and COVID
mRNA for Influenza in primates
Two advantages as compared to classical “egg-grown” inactivated vaccines:
- Fixing the right genetic sequence in mRNA avoids potential “genetic drift” during culture in eggs
- mRNA preparation is quicker: one could observe the evolution of Flu in Southern hemisphere longer before starting the production
→ Better chance that the vaccinal sequence will protect against the actual epidemic Flu in the North
Ep 184-2: Gustav Lindgren Front Immunol (2017): 50 µg mRNA encoding hemagglutinin H10 (from H10N8 2012 avian Flu with “pandemic potential”) in LNP in macaques. Produced by Moderna
See Fig 1 p. comparing ID and IM:
- Two injections (4 weeks interval) needed to obtain hemagglutinin inhibition (HAI) titers > 40 (the level accepted to be protective).
- ID significant higher titers than IM
- A third injection with GLA (TLR4 agonist) provided additional rise.
Evidence for high quality of the response:
- Increasing avidity over time (Fig 1 C)
- Clear presence of memory B cells; antibody-secreting plasma cells and follicular T helper cells in lymph nodes.
Ep 184-3: Follow-up by Feldman Vaccine (2019): two mRNA vaccines by Moderna, encoding H10 (from H10N8) and H7 (from another potential pandemic avian Flu H7N9) tested in humans (phase 1-2):
- IM: rather high doses (100µg) and 2 injections needed to reach “protective” titers
- ID lower dose (25-50 µg) but also 2 injections needed
Two interesting mouse studies
Ep 184-4: Freyn in Mol Ther (2020) evaluate single dose intradermal multicomponent mRNA vaccine, focusing on the more conserved parts of Influenza: the hemagglutinin stalk (mini HA, excluding the more variable head); the neuraminidase (NA) , the nucleoprotein (NP) and the M2 ion channel in a LNP. As a comparator, the mRNA for individual proteins tested:
- Vaccination with Mini HA, M2, or NP alone conferred only partial protection at 50 LD50 (lethal dose for 50 % of unvaccinated mice) and did not protect at 500 LD50.
- The NA-only vaccine prevented mortality in mice at high-dose challenges.
- A trend toward improved protection with the combination vaccine compared to NA only
Moreover: a broad protective potential of a single dose of combination vaccine was confirmed by challenge with a panel of group 1 influenza A viruses.
Is this the pathway towards the long attempted “universal influenza vaccine”?
Ep 184-5: Xinyu Zhuang et al in Vaccines MDPI develop their own H1 mRNA + LNP and show intranasal applications induce protection in mice against 10 LD50 of H1N1.
Perspective for human Influenza mRNA vaccines
Ep 184-6: A commentary in Nature, showing that several companies are developing this technology, focusing on hemagglutinin mRNA and IM injection with a discussion of potential advantages and limitations.
Ep 184-7: A very nice visual overview of vaccine efficacy data by Johns Hopkins and WHO. The first graphs are a bit difficult to understand, but p. 11-15 are more interesting:
- Protection against death is excellent for most vaccines, similar for Adeno’s (Astra-Zeneca and Janssen) as far mRNA (Pfizer and Moderna) but maybe less for CoronaVac
- Similar trend for severe disease, but with sometimes wider confidence intervals
- Protection against symptomatic disease is somewhat weaker by Astra-Zeneca and CoronaVac as compared to mRNA.
- Protection against variants is quite good in most cases.
Ep 184-8: A nice Nature comment on “hybrid immunity” = the observation that subjects who were infected first and months later vaccinated):
- higher titers of neutralizing antibodies than “naïve” (not previously infected) subjects, who receive the full vaccination,
- Also neutralize variant viruses better = broader high quality response!
In addition, it seems that naïve subjects, who received their second dose with a greater delay than prescribed actually also make more and broader neutralizing antibodies.
All these observations (with a lot of new references) extend what I described in the Episode 183: the B cell response seems to mature after infection and vaccination over time.
Therefore, we can certainly hope that the 3rd jab (with the same vaccine) may provide a more robust protection against variants, at least in immunocompetent individuals.
5 Oct 2022 Episode 289: Omicron BA.2.75 revisited and the outlook for new variants, including BQ.1.1
> More info