mRNA-LNP vaccines and adjuvanted recombinant protein vaccines elicit SARS-CoV-2 IgG responses in mouse pups

We first evaluated antibody responses to multiple SARS-CoV-2 vaccines in weanling mice in the absence of maternal antibodies. We vaccinated matAb-negative C57BL/6 mouse pups at weaning (~21 days of life) with SARS-CoV-2 mRNA-LNP (1 µg, 5 µg, or 10 µg dose), recombinant spike protein (RP) diluted in PBS (RP-PBS; 5 µg protein), RP adjuvanted with empty LNP (RP-LNP, 5 µg dose), or RP adjuvanted with the MF59-like adjuvant AddavaxTM (RP-Ad; 5 µg protein). We obtained serum samples and measured SARS-CoV-2 full-length spike protein-specific IgM and IgG by ELISA over time (Fig. 1a). Vaccine immunogens in our study consisted of full-length spike protein from the ancestral Wuhan Hu-1 strain of SARS-CoV-2, with diproline substitution to maintain pre-fusion conformation.

Fig. 1: mRNA-LNP vaccines and adjuvanted protein subunit vaccines elicit SARS-CoV-2 IgG responses in mouse pups.

a The experimental design is shown; Mouse pups were inoculated at weaning with 5 µg SARS-CoV-2 recombinant protein (RP) vaccine b, c adjuvanted with PBS (RP-PBS), AddavaxTM (RP-Ad), or LNP (RP-LNP) or SARS-CoV-2 mRNA-LNP vaccine d, e at 1 µg, 5 µg, or 10 µg. b–e Sera were collected from mouse pups at indicated time points after weaning/vaccination and SARS-CoV-2 full length Spike (S) protein-specific IgG concentrations b, d and IgM concentrations c, e were measured by ELISA. Each point represents the geometric mean with error bars indicating 95% confidence interval (CI) of the geometric mean. Mouse groups were n = 5 (RP vaccinated groups) or n = 6 (mRNA vaccinated groups). Sample IgG and IgM concentrations were plotted to a standard curve of known concentrations of a S-specific IgG monoclonal antibody and are reported as arbitrary units (AU)/mL. Data are shown as geometric mean concentrations with 95% confidence intervals. All panels show results of one experiment that is representative of two independent biological replicates. Group geometric mean IgG concentrations were compared at 4 weeks and 8 weeks post-vaccination to corresponding within-group 1 week concentrations by repeated measures (RM) one-way ANOVA with Tukey’s post hoc test (* indicates p < 0.05, ** indicates p < 0.01). Schematic for a created with Biorender.

The RP vaccine was not immunogenic in mice without co-administering adjuvant (Fig. 1b, c). In mice receiving RP-PBS, spike-specific IgG was not detected in any mice at 1 week post-vaccination and was detected in only a single mouse at 4 and 8 weeks post-vaccination (Fig. 1b). Both adjuvanted RP vaccines (RP-Ad and RP-LNP) as well as mRNA-LNP vaccines (at all three doses) elicited spike-specific IgG responses in weanling mice (Fig. 1b, d, respectively). In mice receiving RP-Ad, spike-specific IgG was detected in most mice at 1 week post-vaccination and increased significantly compared to this 1 week baseline in all mice at 4 and 8 weeks post-vaccination (Fig. 1b). In mice receiving RP-LNP, spike-specific IgG was detected in most mice at 1 week post-vaccination and increased significantly in all mice at 4 and 8 weeks post-vaccination (Fig. 1b). In groups receiving mRNA-LNP, spike-specific IgG was detectable at 1 week and increased significantly compared to each respective baseline at 4 and 8 weeks post vaccination (Fig. 1d).

In mice receiving RP-PBS or RP-Ad, spike-specific IgM was not detected at 1, 4, or 8 weeks post-vaccination (Fig. 1c). In most mice receiving RP-LNP, spike-specific IgM was detectable at 1 week post-vaccination before waning below the threshold of detection at 4 and 8 weeks post-vaccination (Fig. 1c). Spike-specific IgM was detected at 1 week post-vaccination in most mice receiving 1 µg and 5 µg mRNA-LNP and was detectable in all mice receiving 10 µg mRNA-LNP before waning to undetectable levels at 4 and 8 weeks post-vaccination (Fig. 1e).

These experiments show that all SARS-CoV-2 mRNA-LNP vaccines and adjuvanted recombinant protein vaccines elicit antibody responses in weanling mouse pups. For ongoing experiments, 1 µg mRNA-LNP was used for vaccination of weanling pups as it was the lowest dose of mRNA-LNP eliciting comparable spike-specific IgM and IgG responses compared to adjuvanted RP vaccines.

Establishment of a SARS-CoV-2 matAbs mouse model

We next established a mouse model of SARS-CoV-2 matAb transfer. We determined the transfer efficiency and longevity of SARS-CoV-2 spike-specific matAbs following maternal vaccination. For these experiments, we vaccinated adult pregnant female C57BL/6 mice with SAR-CoV-2 spike mRNA-LNP at varying doses (0.1-5 µg) 5 days after introducing unvaccinated male breeding partners. All dams were vaccinated 5 days after introduction with a male without checking plugs, with vaccination of dams occurring up to 5 days gestation. These dams were then allowed to deliver pups. Because matAbs are transferred to mouse offspring both in utero and via milk9, we collected serum from pups at weaning (~ 21 days of life) and at intervals over 9 weeks post-weaning and measured SARS-CoV-2 full-length spike protein-specific IgG by enzyme-linked immunosorbent assay (Fig. 2a). Vaccinated dams transferred SARS-CoV-2 spike-specific matAbs to pups in a dose-dependent manner. All pups born to vaccinated dams had detectable spike-specific IgG and these spike-specific matAbs waned to undetectable levels over time in the 0.1 µg, 1 µg, and 5 µg vaccine dose groups (Fig. 2b, 0.1 µg AUC 25.00 AU, 95% CI 23.03 - 26.97 AU; 1 µg AUC 156.4 AU, 95% CI 124.1 - 188.8 AU; 5 µg AUC 382.6 AU, 95% CI 306.3 - 458.8 AU). MatAb half-life was similar in all groups (Fig. 2b). On day of weaning, dams (50.41 AU/mL, 95% CI 19.09–133.1 AU/mL) had higher spike-specific IgG levels than pups (21.94 AU/mL, 95% CI 20.06 – 28.56 AU/mL, Fig. 2c). Offspring:dam pairs had placental IgG transfer ratios of ~ 0.57 (mean 0.57, range 0.20 to 1.53). These experiments demonstrate that SARS-CoV-2 spike-specific maternal antibodies are transferred to pups and wane over time.

Fig. 2: MatAbs against SARS-CoV-2 are transferred to mouse pups and wane over time.

a The experimental design is shown; Female mice were mated with males and then were inoculated at day 5 after introduction of males with SARS-CoV-2 mRNA-LNP vaccine at 0.1 µg, 1 µg, or 5 µg. b, c. b Sera were collected from mouse pups at indicated time points after weaning/vaccination and SARS-CoV-2 full length Spike (S) protein-specific IgG concentrations were measured by ELISA. One phase-decay was fitted to IgG data (R2 > 0.90 for all groups) with each point representing a single mouse pup and each line representing the decay curve for one dose group (0.1 µg n = 2; 1 µg n = 12; 5 µg n = 7). c Dams were vaccinated with 5 µg mRNA-LNP and serum was collected from dams (n = 6) and pups (n = 48) at day of weaning. Each point represents the geometric mean with error bars indicating 95% confidence interval (CI) of the geometric mean. Sample IgG concentrations were plotted to a standard curve of known concentrations of a S-specific IgG monoclonal antibody and are reported as arbitrary units (AU)/mL. Data are shown as geometric mean concentrations with 95% confidence intervals (* indicates p < 0.05 by unpaired two-tailed Student’s T test). Schematic for Figure a created with Biorender.

SARS-CoV-2 mRNA-LNP vaccine and adjuvanted protein subunit vaccine IgG responses are not inhibited by SARS-CoV-2-specific matAbs

We tested different SARS-CoV-2 vaccines in our newly established SARS-CoV-2 matAb mouse model. We vaccinated matAb-positive and matAb-negative C57BL/6 mouse pups at weaning (~21 days of life) with SARS-CoV-2 mRNA-LNP (1 µg dose), RP adjuvanted with LNP (RP-LNP; 5 µg protein), RP adjuvanted with the MF59-like adjuvant AddavaxTM (RP-Ad; 5 µg protein), or PBS and measured SARS-CoV-2 full-length spike-specific IgG by ELISA over time (Fig. 3a). We did not test RP without adjuvant in these experiments since we previously found that the unadjuvanted RP vaccine was poorly immunogenic even in the absence of matAbs (Fig. 1b, c).

Fig. 3: SARS-CoV-2 mRNA-LNP vaccines and adjuvanted protein subunit vaccine-elicited IgG responses are not inhibited by SARS-CoV-2-specific matAbs.

a The experimental design is shown; Female mice were mated with males and then were inoculated at day 5 after introduction of males with SARS-CoV-2 mRNA-LNP vaccine (5 µg) or PBS and pups were vaccinated at weaning with PBS, SARS-CoV-2 recombinant protein (RP) with Addavax (RP-Ad) or empty lipid nanoparticle (RP-LNP), or SARS-CoV-2 mRNA vaccine (mRNA-LNP). b–e Sera were collected from mouse pups at indicated time points after weaning/vaccination and SARS-CoV-2 full length Spike (S) protein-specific IgG concentrations were measured by ELISA. For each vaccine condition (b, PBS; c, mRNA-LNP; d, RP-Ad; e RP-LNP; n = 4-6 mice per group) at each time point, geometric mean IgG concentrations were compared between the matAb+ and matAb- groups by one-way ANOVA with post hoc Šidák test. Each point represents the geometric mean with 95% confidence intervals. f, g Column graphs depict SARS-CoV-2 IgG concentrations for matAb- f and matAb+ g mice at 18 weeks post-vaccination for each vaccine condition, with each point representing a single mouse and central bars representing geometric mean concentration and 95% CI. Group geometric mean IgG concentrations were compared by one-way ANOVA with Tukey’s post hoc test with significance level of group comparison indicated by nested brackets. For all panels, sample IgG concentrations were plotted to a standard curve of known concentrations of a SARS-CoV-2 IgG monoclonal antibody and are reported as arbitrary units (AU)/mL with samples for which IgG was below the threshold of detection (TOD, represented by horizontal dotted lines) imputed to half the TOD (represented by the x axis). Central points/bars and error bars for all panels represent geometric mean concentrations with 95% confidence intervals. Significance levels for statistical hypothesis testing are indicated (* indicates p < 0.05, ** indicates p < 0.01, ns indicates p ≥ 0.05). All panels show results of one experiment that is representative of two independent biological replicates. Schematic for a created with Biorender.

In the matAb-negative (matAb-) group vaccinated with PBS, spike-specific IgG was not detected at any time point. In the matAb-positive (matAb + ) group vaccinated with PBS, spike-specific IgG waned to near the threshold of detection at 18 weeks (126 days) post-weaning (Fig. 3b), consistent with the matAb decay rates that we observed in earlier experiments (Fig. 2). In matAb+ mice prior to vaccination, spike-specific IgG was detected on day 0 (at weaning) in all mice and was significantly elevated compared to within-group matAb- mice, none of which had detectable IgG on day 0 (Fig. 3c–e). At 7 days post-vaccination, matAb- mice receiving SARS-CoV-2 mRNA-LNP vaccines developed detectable IgG antibodies at concentrations similar to matAb+ mice (Fig. 3c) and these remained detectable at similar levels until the last point of sampling at 126 days post-vaccination (Fig. 3c). Mice vaccinated with adjuvanted recombinant protein SARS-CoV-2 vaccines exhibited similar trends albeit with antibody concentrations rising to similar levels between matAb+ and matAb- groups occurring at 28 days post-vaccination. At 28 days post-vaccination, matAb- mice receiving either SARS-CoV-2 RP-LNP or SARS-CoV-2 RP-Ad vaccines developed detectable IgG antibodies at concentrations similar to matAb+ mice (Fig. 3e, d) and these remained detectable at similar levels until the last point of sampling at 126 days post- vacation (Fig. 3e, d).

At 18 weeks (126 days) post-vaccination, spike-specific IgG concentrations in all vaccinated mouse groups were significantly higher than PBS control in both matAb+ and matAb- arms (Fig. 3, g, respectively). Comparing between matAb- groups, spike-specific IgG concentrations were not significantly different between mRNA-LNP, RP-Ad, and RP-LNP. Similarly, in matAb+ groups, spike-specific IgG concentrations were not significantly different between mRNA-LNP, RP-Ad, and RP-LNP.

To determine the neutralization efficiency of vaccine-elicited spike-specific IgG, sera from pups at 4 weeks post-vaccination were analyzed by focus reduction neutralization tests (FRNTs) using SARS-CoV-2 pseudotyped VSV∆G-RFP virus (Fig. 4). Neutralization titers (expressed as FRNT50) were significantly higher for mRNA-LNP vaccinated pups than for pups receiving PBS, RP-Ad, or RP-LNP in the absence (Fig. 4a) or presence (Fig. 4b) of SARS-CoV-2-specific matAbs. Comparison within vaccine conditions of matAb+ and matAb- pups demonstrated that serum from matAb+ pups did not have lower neutralization titers than serum from the corresponding matAb- pups (Fig. 4c).

Fig. 4: Neutralization efficiency of SARS-CoV-2 mRNA-LNP vaccines and adjuvanted protein subunit vaccine-elicited antibodies in the presence of absence of SARS-CoV-2-specific matAbs.

Female mice were mated with males and then were inoculated at day 5 after introduction of males with SARS-CoV-2 mRNA-LNP vaccine (5 µg) or PBS and pups were vaccinated at weaning with PBS, SARS-CoV-2 recombinant protein (RP) with Addavax (RP-Ad) or empty lipid nanoparticle (RP-LNP), or SARS-CoV-2 mRNA vaccine (mRNA-LNP). Sera were collected from mouse pups at 4 weeks after weaning/vaccination and neutralization efficacy measured by focus reduction neutralization test (FRNT) of pseudotyped vesicular stomatitis virus (VSV) expressing SARS-CoV-2 D614G spike glycoprotein. Column graphs depict FRNT50 for matAb- a and matAb+ b mice for each vaccine condition, with each point representing a single mouse and central bars representing geometric mean concentration and 95% CI. c Presents all groups on a single column graph for ease of depiction of within-vaccine condition comparisons of matAb- and matAb- mice. Group geometric mean IgG concentrations were compared by one-way ANOVA with Tukey’s post hoc test with significance level of group comparison indicated by nested brackets. Samples for which no neutralization was detected at the lowest dilution were imputed to half the TOD (dotted line). Significance levels for statistical hypothesis testing are indicated (* indicates p < 0.05, ** indicates p < 0.01, ns indicates p ≥ 0.05).

Together,these experiments demonstrate that mRNA-LNP and adjuvanted recombinant SARS-CoV-2 vaccines elicit robust IgG responses in weanling mouse pups in the presence of SARS-CoV-2 spike-specific matAbs at the time of vaccination, and that SARS-CoV-2 mRNA-LNP vaccines elicit higher overall neutralizing antibody titers relative to adjuvanted recombinant vaccines.