A history of recurrent positive selection at the toll-like receptor 5 in primates

Abstract

Many genes involved in immunity evolve rapidly. It remains unclear, however, to what extent pattern-recognition receptors (PRRs) of the innate immune system in vertebrates are subject to recurrent positive selection imposed by pathogens, as suggested by studies in Drosophila, or whether they are evolutionarily constrained. Here, we show that Toll-like receptor 5 (TLR5), a member of the Toll-like receptor family of innate immunity genes that responds to bacterial flagellin, has undergone a history of adaptive evolution in primates. We have identified specific residues that have changed multiple times, sometimes in parallel in primates, and are thus likely candidates for selection. Most of these changes map to the extracellular leucine-rich repeats involved in pathogen recognition, and some are likely to have an effect on protein function due to the radical nature of the amino acid substitutions that are involved. These findings suggest that vertebrate PRRs might show similar patterns of evolution to Drosophila PRRs, in spite of the acquisition of the more complex and specific vertebrate adaptive immune system. At shorter timescales, however, we found no evidence of adaptive evolution in either humans or chimpanzees. In fact, we found that one mutation that abolishes TLR5 function is present at high frequencies in many human populations. Patterns of variation indicate that this mutation is not young, and its high frequency suggests some functional redundancy for this PRR in humans.

FIG. 1.—

Lineage-specific d_N/d_S values of TLR5 in primates (A) for the entire gene and (B) for the extracellular domain. Estimated d_N/d_S values from the branch-based model are shown above branches and the estimated number of nonsynonymous and synonymous changes are shown below branches. Branches with d_N/d_S values greater than 1 are shown in red. Mating systems categorized as “less promiscuous” (polygyny + monogamy) are indicated with a blue circle, whereas “more promiscuous” (promiscuous + dispersed) mating systems are indicated with a red circle. Arrows show the six unambiguous independent transitions between less and more promiscuous mating systems. For the Old World and New World monkey clades, “circled-pointed arrows” indicate additional transitions between low and high promiscuity according to alternative but equally parsimonious reconstructions.

FIG. 2.—

Haplotype network showing the two divergent haplogroups shared between P.t. verus and P.t. troglodytes. Each circle represents a different haplotype and its size is proportional to its frequency in the sample. Mutations distinguishing haplotypes are shown as marks along the lines, whereas missing haplotypes are shown as black dots.

FIG. 3.—

Distribution of TLR5^392STOP around the world. The frequency of the allele is shown in red.