Systems Vaccinology Identifies an Early Innate Immune Signature as a Correlate of Antibody Responses to the Ebola Vaccine rVSV-ZEBOV

Abstract

Predicting vaccine efficacy remains a challenge. We used a systems vaccinology approach to identify early innate immune correlates of antibody induction in humans receiving the Ebola vaccine rVSV-ZEBOV. Blood samples from days 0, 1, 3, 7, and 14 were analyzed for changes in cytokine levels, innate immune cell subsets, and gene expression. Integrative statistical analyses with cross-validation identified a signature of 5 early innate markers correlating with antibody titers on day 28 and beyond. Among those, IP-10 on day 3 and MFI of CXCR6 on NK cells on day 1 were independent correlates. Consistently, we found an early gene expression signature linked to IP-10. This comprehensive characterization of early innate immune responses to the rVSV-ZEBOV vaccine in humans revealed immune signatures linked to IP-10. These results suggest correlates of vaccine-induced antibody induction and provide a rationale to explore strategies for augmenting the effectiveness of vaccines through manipulation of IP-10.

Keywords: Ebola vaccine; IP-10; RNA sequencing; emerging infections; innate immunity; rVSV-ZEBOV; systems vaccinology; viral immunity.

Graphical abstract

Figure 1

A Strong Increase in Cytokine Levels as Early as Day 1 after Vaccination (A) Graphs showing box and whisker plots with the minimum, first quartile, median, third quartile, and maximum of the cytokine levels on the different days. Two-sided Wilcoxon signed-rank test was performed for within-group comparisons of post-vaccination time points to baseline (d0) across both dose groups (n = 20 trial participants). Significant p values adjusted for multiple testing by adaptive FDR are shown for the day 1 to day 0 and day 3 to day 0 comparisons. (B) Heatmap reflecting the plasma concentration of 13 cytokines on days 0, 1, 3, 7, and 14 for both vaccine dose groups (group 1 [n = 10 trial participants] with 3 × 10⁶ PFUs in orange symbols, group 2 [n = 10 trial participants] with 20 × 10⁶ PFUs in blue symbols on the top part of the figure). For generation of the heatmap, plasma concentration levels below the range of the standard curve were replaced with a value of 0 pg/mL, and all values per cytokine were then standardized. (C) Box and whisker plots with the minimum, first quartile, median, third quartile, and maximum of the cytokine levels of MCP-1 and IP-10 for each dose group on day 1 (n = 10 participants per dose group). Two-sided Wilcoxon rank-sum test was performed for between-group comparisons. The p values are adjusted for multiple testing by adaptive FDR. Data were generated from one experiment. Samples were measured in duplicates.

Figure 2

Early Vaccine-Induced Changes in Frequency and Activation Status of Innate Immune Cell Subsets (A and B) Graphs show box and whisker plots with the minimum, first quartile, median, third quartile, and maximum. Shown are the changes in frequency of selected innate cell subsets (A) and a summary of the modulated MFI on selected subsets of DCs, (inflammatory CD16+) monocytes, and NK cells between day 0 and day 14 (B). Two-sided Wilcoxon signed rank-test was performed for within-group comparisons of post-vaccination time points to baseline (d0) across both dose groups (n = 20 trial participants). Significant p values, adjusted for multiple testing by adaptive FDR, are shown for the day 1 to day 0 and day 3 to day 0 comparisons. Data were generated from one experiment. See also Table S1.

Figure 3

Individual Changes in Gueckedou-GP-Specific Antibody Levels Over Time Each line corresponds to an individual trial participant. Participants in the 3 × 10⁶ PFU group are shown in blue, and participants in the 20 × 10⁶ PFU group are shown in orange. Negative optical density (OD) values were imputed as the lowest positive observed OD value (0.003). See also Figure S1.

Figure 4

Correlation Matrix (Pearson Correlation Coefficients) between Each of the Five Early Cell Populations, Cytokine Immune Response Markers Selected by sPLS, and the Antibody Responses between day 28 and day 180 (n = 20 Trial Participants) The matrix shows positive correlations with the corresponding antibody responses in blue and negative correlations in red. The stronger the intensity of the color, the stronger is the correlation. See also Figure S1, Tables S2 and S3, and Table S4.

Figure 5

Differentially Expressed Genes after Vaccination (n = 18 Trial Participants with Available RNA-Seq Data) (A) Venn diagram of the number of differentially expressed genes per post-vaccination time point. (B) Distribution of log fold changes on day 1. Each dot corresponds to a gene, plotted according to its mean log expression on the x axis and its log fold change on the y axis. Red dots indicate statistically significant differentially expressed genes. See also Figures S2 and S3.

Figure 6

IP-10 Gene Expression Pathway (A) Pathway analyses of the TIFA gene, detected in the early gene expression signature, with regards to the IP-10 (CXCL-10) pathway. (B) Observed correlations between TIFA gene expression on day 1, representative intermediate genes of the CXCL-10 pathway on day 1, CXCL10 gene expression on day or 3, and plasma IP-10 cytokine concentration on day 1 or 3; Pearson correlation coefficients are shown. TIFA, TRAF-interacting protein with forkhead-associated domain; NFKB1, NF-κB subunit 1; NFKB2, NF-κB subunit 2; CHUK, conserved helix-loop-helix ubiquitous kinase; IKBKB, inhibitor of NF-κB kinase subunit beta; IKBKE, inhibitor of NF-κB kinase subunit epsilon; IKBKG, inhibitor of NF-κB kinase subunit gamma; TNFRSF1A, TNF receptor superfamily member 1A; TRAF2, TNF receptor-associated factor 2; CXCL10, C-X-C motif chemokine ligand 10; IP-10, plasma IP-10 cytokine concentration; IKK complex, I_KB kinase complex; NfkB complex, nuclear factor κB complex. See also Figures S4 and S5.