Sequential early-life viral infections modulate the microbiota and adaptive immune responses to systemic and mucosal vaccination

Abstract

Increasing evidence points to the microbial exposome as a critical factor in maturing and shaping the host immune system, thereby influencing responses to immune challenges such as infections or vaccines. To investigate the effect of early-life viral exposures on immune development and vaccine responses, we inoculated mice with six distinct viral pathogens in sequence beginning in the neonatal period, and then evaluated their immune signatures before and after intramuscular or intranasal vaccination against SARS-CoV-2. Sequential viral infection drove profound changes in all aspects of the immune system, including increasing circulating leukocytes, altering innate and adaptive immune cell lineages in tissues, and markedly influencing serum cytokine and total antibody levels. Beyond changes in the immune responses, these exposures also modulated the composition of the endogenous intestinal microbiota. Although sequentially-infected mice exhibited increased systemic immune activation and T cell responses after intramuscular and intranasal SARS-CoV-2 immunization, we observed decreased vaccine-induced antibody responses in these animals. These results suggest that early-life viral exposures are sufficient to diminish antibody responses to vaccination in mice, and highlight the potential importance of considering prior microbial exposures when investigating vaccine responses.

Fig 1. Sequential viral exposure creates an inflammatory host environment.

(A) Abbreviated sequential infection regimen including murine rotavirus (MRV), murine gammaherpesvirus 68 (MHV68), murine norovirus CR6 (MNV), Influenza A Virus (IAV), murine astrovirus (MAstV), and coxsackievirus B3 (CVB3). Whole blood and indicated tissues of ten-week-old mock- and sequentially-infected C57BL/6J mice were collected for analysis. (B, C) Hematologic analysis of peripheral blood using the automatic blood cell counter Element HT5. Radar plot comparing the indicated blood cell populations in mock- (n = 16) and sequentially-infected mice (n = 21) (B). White blood cells (WBC), red blood cells (RBC), platelets (PLT), lymphocytes (Lym), neutrophils (Neu), monocytes (Mon), eosinophils (EOS), and basophils (BAS). (D, E) Serum was collected from mock- (n = 16) and sequentially-infected (n = 24) mice at the age of 10 weeks. The concentration of cytokines and chemokines was determined by Bioplex assay. Radar plot comparing the serum concentrations of indicated cytokines and chemokines in the indicated groups (D). Columns show median values and error bars represent the standard error of the mean, with dotted lines indicating the limit of detection (LOD) of the assays. Undetected cytokines were given a value of LOD. Statistical analysis was performed using unpaired Mann-Whitney test for (C) and (E): *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001; ns, not significant.

Fig 2. Sequentially infected mice have altered circulating immune cell compositions.

(A) Peripheral blood T cell counts from mock- (n = 16) or sequentially-infected (n = 21) laboratory mice at the age of 10 weeks. (B, C) CD8⁺ T cell phenotypes in blood from mock- or sequentially-infected laboratory mice. Representative flow plots (n = 18–27) are shown (B). Phenotypes of CD44^lo/CD62L^hi (naive) and CD44^hi (antigen-experienced) CD8⁺ T cells from PBMCs of mock- and sequentially-infected mice compared by flow cytometry (C). (D, E and J) Frequency of CD44^hi (antigen experienced) CD8⁺ T cells from PBMCs of mock-(n = 18) or sequentially-infected mice (n = 27) from four independent experiments (D). Comparison of CD44^lo/CD62L^hi (Naive), CD44^hi/CD62L^hi (antigen-experienced central memory, CM), and CD44^hi/CD62L^lo (antigen-experienced effector memory, EM) CD8⁺ T cell between mock- and sequentially-infected mice at 10 weeks (E) or 7 months (J) of age. (F, G and K) Frequency of KLRG1⁺ CD44^hi CD8⁺ T cells from PBMCs of mock- (n = 18) or sequentially-infected mice (n = 27) from four independent experiments (F). Comparison of KLRG1⁺ CD44^hi CD8⁺ T cell between mock- and sequentially infected mice at 10 weeks (G) or 7 months (K) of age. (H, I and L) Frequency of GrzB⁺ CD44^hi CD8⁺ T cells from PBMCs of mock- mice (n = 18) or sequentially-infected mice (n = 27) from four independent experiments (H). Comparison of GrzB⁺ CD44^hi CD8⁺ T cell between mock- and sequentially-infected mice at 10 weeks (I) or 7 months (L) of age. Columns show median values and error bars represent the standard error of the mean. Vertical dotted lines indicate the first and last viral inoculations. For (A), (E), (F), (H), (I), (K) and (L) statistical analysis was performed using Mann-Whitney test: *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001; ns, not significant.

Fig 3. Sequential viral exposure shapes immune cell profiles in mucosal tissues.

(A,B) Cell types isolated and enumerated by flow cytometry from the lung (B) and ileum (C) of mock- and sequentially-infected 10 week-old mice (n = 13/group); also shown in S4 Fig. ILC, innate lymphoid cells; T_h, CD4 T helper cells; T_reg, regulatory CD4 T cells; γδ T, gamma delta T cells. Radar plot comparing the indicated immune cell populations in mock- and sequentially-infected mice. Single-positive staining for T-bet, Gata3, and Rorγt were used to determine Th1, Th2, and Th17 lineages, respectively. (C) Antigen specificity was assessed and enumerated by flow cytometry from the spleens of mock-infected (n = 10) and sequentially infected (n = 9) laboratory mice at 10 weeks of age. Columns show median values and error bars represent the standard error of the mean. Statistical analysis was performed using unpaired Mann-Whitney test: *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001; ns, not significant.

Fig 4. Sequential viral exposure modulates antibody levels and B cell lineages in tissues.

(A) Total antibody concentrations in serum of mock- (n = 16) and sequentially-infected mice (n = 24) at 10 weeks of age. (B) Virus-specific total IgG levels in serum of mock- (n = 16) and sequentially-infected mice (n = 24) at 10 weeks of age. (C, D) Cell types isolated and enumerated by flow cytometry from the spleen (C) and bone marrow (BM) (D) of mock- and sequentially-infected mice. Radar plot comparing the indicated immune cell populations in mock- and sequentially-infected mice. Columns show median values and error bars represent the standard error of the mean. Statistical analysis was performed using unpaired Mann-Whitney test: *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001; ns, not significant.

Fig 5. Sequential viral infection modulates the intestinal bacterial microbiome.

The bacterial microbiome of mock- and sequentially-infected mice was profiled for fecal samples (mock-: n = 17, sequentially-infected: n = 26), and complete nasal turbinate tissues (mock-: n = 6, sequentially-infected: n = 11) using sequencing of the 16S rRNA gene. (A) Mean values and interquartile ranges of alpha diversity indices. Bacterial richness (Observed), Shannon diversity (Shannon), and Pielou’s evenness (Pielou) are shown. Mann-Whitney test: *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001; ns, not significant. (B) Principal coordinates analysis (PCoA) using weighted UniFrac distances was performed. Differences between groups were assessed using PERMANOVA. (C) Linear discriminant analysis (LDA) effect size analysis was conducted to identify discriminant taxa between the intestinal and complete nasal turbinate microbiomes of mock- and sequentially-infected mice. In the left panel, the LDA effect size (with a 95% confidence interval) is displayed for taxa that were found to be distinct between groups. On the right panel, the false discovery rate (FDR) of each taxa is depicted, with vertical dotted line indicating an FDR of 0.05. (D) The relative abundance of discriminant taxa with low FDR identified from the intestinal microbiota at the genus level. Analyzed by Wilcoxon-Mann-Whitney test: **p < 0.01; ***p < 0.001.

Fig 6. Sequential viral infection is associated with reduced antibody response to intramuscular and intranasal ChAd-SARS-CoV-2 vaccination.

(A, B) Antibody responses in sera of immunized mice at day 28 after priming were evaluated. Anti-Spike IgG and IgA levels were measured by ELISA in intramuscular (IM) or intranasal (IN) vaccination studies. Data are pooled from two experiments (n = 8–13). (C) Neutralizing activity of serum from immunized mice against pseudoviruses expressing Wuhan-1 spike proteins (numbers below indicate Geometric Mean Titer values; dotted lines show the LOD; values at the LOD are plotted slightly below the LOD). (D, E) Cell-mediated immune responses were analyzed at day 35 post-immunization after re-stimulation with an S protein-peptide pool. Splenocytes were assayed for IFN-γ and GrzB expression in CD8⁺ T cells and GrzB only in CD4⁺ T cells by flow cytometry. Cell counts of positive splenocytes in intramuscular (D) or intranasal (E) vaccination studies (n = 8–13). (F—I) Lung CD8⁺ T cells from intramuscularly (F, G) or intranasally (H, I) vaccinated mice were assayed for IFN-γ and GrzB expression by flow cytometry after re-stimulation with an S protein peptide pool. Lung CD8⁺ T cells were phenotyped for expression of CD103 and CD69 after intramuscular (G) or intranasal (I) vaccination. Columns show median values and error bars represent the standard error of the mean. Dotted lines indicate the LOD of the assays. Mann-Whitney test: *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001; ns, not significant.