Functional comparison of innate immune signaling pathways in primates

Abstract

Humans respond differently than other primates to a large number of infections. Differences in susceptibility to infectious agents between humans and other primates are probably due to inter-species differences in immune response to infection. Consistent with that notion, genes involved in immunity-related processes are strongly enriched among recent targets of positive selection in primates, suggesting that immune responses evolve rapidly, yet providing only indirect evidence for possible inter-species functional differences. To directly compare immune responses among primates, we stimulated primary monocytes from humans, chimpanzees, and rhesus macaques with lipopolysaccharide (LPS) and studied the ensuing time-course regulatory responses. We find that, while the universal Toll-like receptor response is mostly conserved across primates, the regulatory response associated with viral infections is often lineage-specific, probably reflecting rapid host-virus mutual adaptation cycles. Additionally, human-specific immune responses are enriched for genes involved in apoptosis, as well as for genes associated with cancer and with susceptibility to infectious diseases or immune-related disorders. Finally, we find that chimpanzee-specific immune signaling pathways are enriched for HIV-interacting genes. Put together, our observations lend strong support to the notion that lineage-specific immune responses may help explain known inter-species differences in susceptibility to infectious diseases.

Figure 1. LPS-mediated innate immune response to infection in humans, chimpanzees, and rhesus macaques.

(A) Venn-diagram showing the number of genes whose expression levels were altered following stimulation with LPS in humans, chimpanzees and rhesus macaques at any time point (see Figure S4 for data from specific time points). (B) Gene ontology (GO) enrichment analysis using the subset of genes that responded to the treatment in all three species. GO terms related to immunity processes are plotted (see Table S2 for results including all GO terms). (C) KEGG pathway enrichment analysis for genes that responded to the treatment in all three species. (D) Transcription factor binding site enrichment analysis in the promoters of genes that responded to the treatment in all three species (see Table S3 for complete results).

Figure 2. Universal TLR response is more conserved than the immune responses to specific bacterial or viral infections.

(A) The proportion of genes that responded to the treatment in all three species, in any two species, or in only one species are plotted for the subsets of genes classified as part of the universal TLR response, the immune responses specific to bacterial infections, or the immune responses specific to viral infections (B) The proportion of genes classified as part of the universal TLR response, the immune responses specific to bacterial infections, or the immune responses specific to viral infections among genes that responded to the treatment in all three species or exclusively in one species. Genes were classified as part of the universal, bacterial, or viral TLR response, based on the findings of Amid and colleagues .

Figure 3. Species-specific responses to infection.

Examples of (A) human-specific, (B) chimpanzee-specific, and (C) rhesus-specific immune responses to the treatment. In all panels, the log₂ fold difference in expression levels () following the treatment (y-axis) is plotted for each species at the different time points following infection (x-axis).

Figure 4. An example of a functional network of genes that responded to stimulation with LPS exclusively in humans.

Genes involved in apoptosis and/or cancer pathways are highlighted.

Figure 5. Co-expression regulatory networks.

Heatmaps illustrating the correlations of expression profiles for genes responding to stimulation with LPS exclusively in (A) humans, (B) chimpanzees, and (C) rhesus macaques are plotted. Blocks of genes with highly correlated expression profiles correspond to regulatory modules determined by the MMC algorithm. We found 33, 17 and 32 modules in humans, chimpanzees and rhesus macaques, respectively, with an average connectivity (|r|) higher than 0.5. In addition to the modules discussed in the text, regulatory modules that merit particular attention include (D) module 13 in chimpanzees, which is significantly enriched for immune response genes (highlighted in yellow), and (E) regulatory module 18 in rhesus macaques, which is significantly enriched for genes involved in immune-related pathways; in particular in MAP kinase signaling pathways (highlighted in yellow), which control a range of cellular activities related to innate immune responses and are particularly important in regulating cytokine gene expression levels and pathways related to programmed cell death .

Figure 6. Species-specific immune responses and disease susceptibility.

In both panels, the ‘all genes’ category (gray bar) refers to the set of genes that were classified as expressed (based on the array data) in at least one of the conditions (i.e., at any time point in either the treated or untreated samples). (A) The proportions of genes associated with infectious diseases or immune related disorders (y-axis) among genes that responded to stimulation with LPS exclusively in humans, chimpanzees and rhesus macaques (x-axis). (B) The proportion of HIV-1 interacting genes (y-axis) among the subsets of genes that responded to stimulation with LPS exclusively in each of the three species (x-axis). The observed pattern is robust with respect to the cutoffs used to classify genes as differentially expressed following the treatment (Figure S5).