Fig 1.
Virologic and immunologic responses in Rh/CMV vaccination and SIV challenge.
A. Schematic of the vaccine phase of the two cohorts of RMs (n = 15 each) administered the 68–1 RhCMV/SIV vector set by either subcutaneous or oral routes at wk0 Pr and wk18 Bo, indicating time points for which whole blood samples were collected for RNAseq analysis. Repeated limiting dose SIVmac239 challenge was initiated at wk91. B,C. Assessment of the outcome of effective challenge by longitudinal analysis of the de novo development of SIV Vif-specific CD4+ and CD8+ T cell responses (B) and plasma viral load (C). RMs were challenged until the onset of any above-threshold SIV Vif-specific T cell response, with the SIV dose administered 2 or 3 weeks prior to this response detection considered the infecting challenge (week 0). RMs with sustained viremia were considered not protected; RMs with no or transient viremia were considered protected (6–8). D. Bone marrow (BM), peripheral lymph node (LN) and peripheral blood mononuclear cell (PBMC) samples from all vaccine-protected RMs and representative non-protected or unvaccinated control RMs, collected from between day 28 and day 56 post-SIV infection, were analyzed by nested, quantitative PCR/RT-PCR for cell-associated SIV DNA and RNA. The horizontal line indicates the threshold of detection (B.T. = below threshold) with data points below this line reflecting no positive reactions across all replicates. Above threshold cell-associated SIV RNA was detected in LN and BM of all protected RMs, confirming SIV infection “take”.
Fig 2.
Identification of DE genes after RhCMV/SIV vaccination in protected vs. non-protected RMs.
S3 Fig. Wk0, d0 signature comparison. A. PCA of the per-time point mean log2 fold-change (FC) values, using all expressed genes and averaged over the RMs within each treatment and outcome group, showing the mean log2(FC) of all expressed genes per time point in oral (circles) and subcutaneous (diamonds) for protected (red) and non-protected (black) RMs. B. Upper panel: number of DE genes per time point in each group. Lower panel: Heatmap showing genes most associated with each PC, for the first 7 principal components. Percent variance explained by each PC is shown at left. Prime, boost, and pre-challenge time points are shown at bottom.
Fig 3.
Gene and pathway correlates of protection.
A. Heatmap showing all DDE genes. Three clusters were defined using hierarchical clustering. B. Ingenuity pathway analysis of the three major DDE gene clusters. C. Network of direct physical interactions of encoded proteins (indicated by grey connector line) between major enriched immune pathways with red and blue arrows indicating activating and inhibitory interactions, respectively. P-values for the association of each pathway with vaccine protection are shown and are based on permutation testing. D. Network overview of JAK-STAT signaling in enriched interleukin pathways. E. Heatmap of gene expression changes for expressed interleukin genes (left), and their enrichment as upstream regulators of DDE genes using Ingenuity analysis (right).
Fig 4.
The IL-15 response links with correlates of vaccine protection.
A. Study design–RM treatment with rRh-Het-IL-15. Red arrowheads indicate IL-15 treatment. B. Heatmap of DDE gene correlates of protection regulated by IL-15 (IL-15 DE genes on d1 post-administration). P-values, shown below the heatmap, are highly significant for both the sum statistic and the contrast statistic differing across protection outcomes. The sum statistic is the average over time of the sum of the absolute change-from-baseline expression, and the contrast statistic is the average over time of the sum of the change-from-baseline expression of IL-15 d1 up-regulated genes minus the sum over the IL-15 d1 down-regulated response genes (see Methods). C. Ingenuity pathway and upstream regulator enrichment analyses for gene cluster A and B from panel B. D,E. De novo network of IL-15 response genes from Cluster A (D), and Cluster B (E) were constructed in GeneMania. Transcription factor nodes are indicated by thick borders with connections shown as gray lines and blue arrows, showing co-expression interactions (GeneMania) and direct transcription factor-target interactions (Ingenuity), respectively.
Fig 5.
Validation of the IL-15 response signature of protection.
Heatmaps comparing the IL-15 response signature (overlap of DDE and day 1 IL-15 DE genes as shown in Fig 4B) in the validation cohort (left panel; 68–1 + 68–1.2 RhCMV/SIV vaccinated; 6 of 15 protected) vs. the original subQ 68–1 RhCMV/SIV vaccinated cohort (right panel; 8 of 15 protected). P-values, shown below each heatmap, are significant for both cohorts for the permutation test evaluating the contrast statistic across protection groups, which had been pre-specified for validation. The contrast statistic is the average over time of the sum of the change-from-baseline expression of IL-15 d1 up-regulated genes minus the sum over the IL-15 d1 down-regulated response genes (see Methods).
Fig 6.
Baseline evaluation of DDE and IL-15 response signature of protection.
Box plots and heat map comparing normalized expression values at baseline across protection outcomes. A. Box plots show sum of normalized expression counts over the genes identified as DDE up-regulated or down-regulated across protection outcome groups (S2 Table) with unadjusted Wilcoxon p-values shown. B. Analogous results for the IL-15 d1 clusters A (up) and B (down) at baseline (left two panels) and IL-15 gene expression (right panel). C. Heat map showing the IL-15 response genes in the subQ and oral cohort RMs at baseline. The positions of selected genes are indicated at right. D. Scatter plot showing the IL-15 cluster A minus cluster B contrast for each individual animal, at baseline vs. the change-from-baseline average over time post-vaccination (Pearson correlation R and p-value shown); protected RMs are shown in red; non-protected in gray.