Self-amplifying RNA vaccine against avian flu virus in a mouse model

?

Introduction

Avian flu virus is deadly to many species of birds and can also infect animals and humans. A recent BBC News feature warns that bird flu, caused by influenza A virus subtype H5N1 (“H5”) and its genetic variants, has become an unprecedented animal pandemic. Hundreds of millions of domestic poultry have died from H5 flu or have been preemptively culled, and at least 26 species of mammals have also been infected. Fortunately, human cases have been few to date. Federal health officials say there is an urgent need to develop effective vaccines against current strains of highly pathogenic avian flu virus.?

This blog summarizes how Belgian researchers (Cui et al.) have addressed this challenge by investigating self-amplifying RNA (saRNA) vaccines for H5 avian flu, predicated on earlier findings (Vogel et al.) that an saRNA vaccine for other flu types was more potent than a non-amplifying mRNA. As outlined below, the in vitro transcription (IVT)-based synthesis of the H5 saRNA candidates used TriLink’s CleanCap? Reagent AU. This report by Cui et al. has received more than 2,400 views since its publication on August 3, 2024.

Design and synthesis of H5 saRNA vaccine candidates

H5N1 is an enveloped RNA virus defined by the combination its antigenic hemagglutinin (HA) and neuraminidase (N) proteins on the viral membrane, namely, type-5 HA and type-1 N. As depicted in Figure 1, HA is a “spike” protein homotrimer wherein each monomeric unit is comprised of a globular head domain and a stalk domain, structurally akin to the spike protein in SARSCoV2.

Cui et al. note that, like the antigenic spike protein in SARSCoV2, HA plays a pivotal role in the immune response and serves as a preferred antigen for the development of influenza A vaccine platforms, including protein subunit vaccines, DNA vaccines, and mRNA vaccines.?

Based on the previously reported higher potency of saRNA compared to mRNA, and the structure of HA (Figure 1), Cui et al. designed four saRNA vaccine candidates: a secreted full-length HA (sFL-HA), a secreted HA head domain (sHD-HA), a secreted stalk domain (sSD-HA), and a full-length membrane-anchored HA (FL-HA) having a transmembrane region. The sFL-HA was constructed by replacing the transmembrane domain of FL-HA with a leucine zipper sequence that can form a trimer complex. A Flag-tag for detection by anti-Flag antibody (Ab) was added to all three secreted saRNA constructs.

Corresponding linear DNA sequences for these HA designs and a luciferase reporter were synthesized with overlaps to a plasmid vector encoding self-amplification elements from Venezuelan Equine Encephalitis Virus. After cloning, the saRNAs were prepared by IVT using TriLink’s CleanCap? Reagent AU and unmodified nucleoside 5’-triphosphates.

The purified saRNAs were formulated in lipid nanoparticles (LNPs) containing the same ionizable lipid (ALC-0315) as in the COVID-19 BNT162b2 vaccine. An optimized ratio of ionizable lipid to saRNA was determined by intramuscular (im) injection of mice with the luciferase saRNA for cumulative in vivo bioluminescence signal measurement during 21 days post-injection.?

Immunogenicity of H5 saRNA vaccines encoding secreted HA antigens

Optimized LNP formulations of the secreted HA saRNA vaccines and the luciferase saRNA as a control were i.m injected in mice using a dose of 1 μg in a prime-boost schedule with an interval of three weeks (Figure 2). Serum Ab levels were measured three weeks after the prime (day 21) and 12 days after the boost (day 33).

After the prime, the highest anti-H5 IgG levels determined by ELISA were found in the sFL-HA vaccine group. These anti-H5 IgG levels were approximately 1.4-times higher than those in the sHD-HA vaccine group and 4.5-times higher than in the sSD-HA vaccine group. After the boost, the anti-H5 IgG levels in the sFL-HA group showed a 20-fold increase and were significantly higher than the IgG levels in the other groups. As expected, the control group was negative.?

To evaluate the protective capacity of the elicited Abs, hemagglutination inhibition (HAI) titers were determined using the vaccine-matched H5N1 virus. Mice vaccinated with the sFL-HA vaccine achieved a mean HAI titer of 13 by day 21 that increased markedly to 150 by day 33 and was significantly higher than the mean HAI titer in the other groups.?

Cellular immunity can play a vital role in conferring cross-strain protection, prompting Cui et al. to analyze HA-specific T cell responses by intracellular staining of splenocytes isolated on day 33. Mice in the sHD-HA group showed the highest percentage of HA-specific T cells, with an average percentage of 3.1% IFN-γ positive CD3+ CD8+ T cells by flow cytometry, which is significantly higher than in the sFL-HA group (0.58%). No detectable T cell immunity was observed in the sSD-HA group and control group.

Dose-dependent and mucosal immune responses of H5 sFL-HA saRNA vaccine

Because of the higher Ab levels found for sFL-HA compared to the other secreted HA constructs, this vaccine was studied in more detail. To investigate vaccine dosage and the strength of the immune responses, Cui et al. immunized mice with 0.25 μg, 1 μg, or 4 μg of the sFL-HA vaccine or 4 μg of luciferase saRNA control, using a prime-boost schedule with a three-week interval (Figure 3A).

Serum Ab (total IgG, IgG1, IgG2a) (Figure 3A) and HAI titers (Figure 3D) were determined three weeks after the prime (day 21), and then 7 (day 28) and 14 (day 35) days after the boost. This revealed a positive correlation between the dose and both the Ab (total IgG, IgG1, IgG2a) and HAI titers, with the 4 μg group significantly outperforming the 0.25 μg and 1 μg groups on day 35.

A clear effect of the boost was noticed as two weeks after the booster Ab levels increased more than 10-fold in all groups (Figure 3B). Importantly, the mean HAI titers of the mice that received the 0.25 μg dose increased after the boost to a comparable high and protective mean level of about 85 as obtained with 1 μg saRNA vaccine (Figure 3D). Finally, the ratio of IgG2a/IgG1 (Figure 3C) consistently remained above 1, suggesting the immune response induced by the sFL-HA vaccine is skewed towards a Th1 response (Figure 3C), whereby Th1 cells?function as the principal mediators of immunity to eradicate intracellular pathogens.

Given the aerosol transmission route of influenza, the establishment of robust mucosal immunity is crucial in curbing the viral infection. Therefore, two weeks after the boost (day 35), HA-specific IgA levels in bronchoalveolar lavage fluid (BALF) samples were determined. All vaccinated mice had detectable antigen-specific IgA titers in their BALF, with the highest levels found in the 1 μg saRNA vaccine group, significantly (*p < 0.05) higher than other groups.

Head-to-head comparison of H5 saRNA vaccines encoding secreted or membrane-anchored HA

Genetic vaccines encoding membrane-anchored HA elicit stronger immune responses than genetic vaccines encoding secreted HA (Freyn et al.). Therefore, Cui et al. compared the efficacy of the saRNAs encoding membrane-anchored FL-HA or secreted sFL-HA. Mice were i.m vaccinated with 1 μg of either saRNA using a prime-boost schedule with a four-week interval (day 28).

Mice vaccinated with FL-HA exhibited significantly (****p < 0.0001) higher anti-H5 IgG, IgG1, and IgG2a levels than mice that received sFL-HA after the boost (day 35 and day 42). The IgG2a/IgG1 ratios in both groups were above 1, which again indicates a Th1 response.?

In this mouse model, protective titers are indicated by HAI ≥40. The boost led to protective HAI titers in both groups: seven days after the boost (day 35) the HAI titers were around 256 and 100 in the FL-HA and sFL-HA group, respectively. Two weeks after the boost (day 42), the HAI titers in both groups reached a similar level of around 200.

On day 35, the BALFs were collected to measure the anti-H5 IgA level. The saRNA encoding FL-HA induced a significantly (*p < 0.05) stronger mucosal immune response in the lungs, with anti-H5 IgA levels 3.5-fold higher than the sFL-HA group.

Concluding comments

Cui et al. concluded that, overall, the above findings provide comprehensive evidence that all of the LNP-formulated HA saRNA vaccine candidates can induce strong humoral and cellular response, with full-length FL-HA and sFL-HA versions giving protective HAI titers well above the minimal threshold for an efficient vaccine. Importantly, even when a low dose of 0.25 μg of sFL-HA was used, protective HAI titers were obtained after a boost vaccination, indicative of higher potency due to self-amplification.

Although a virus-challenge study was not performed in this work, as mice are not sensitive to the H5N1 virus, all of the findings highlight the potential of HA saRNA-LNP vaccines as a promising approach for preventing avian influenza infections.

?

Your comments are welcomed, as usual.