Last year, Katie Pollard and colleagues published a couple of papers in which they identified regions of the human genome that had recently undergone an acceleration in their rate of evolution and characterized the expression pattern of an RNA gene located in one of those regions. The RNA gene is expressed in the developing brain, which lead people to speculate that it played some important role in making humans smarter than chimps (my round-up and stab at speculation can be seen here). Their approach toward identifying those regions is quite simple, but the cause of accelerated evolution in those regions is unclear. Are they evolving rapidly because of natural selection fixing beneficial mutations, or is evolution sped up because of higher mutation rates?
Pollard and colleagues started by searching for sequences that are highly conserved (ie, have very few changes) amongst all mammals, excluding humans. They then searched within the highly conserved sequences for those with the fastest rates of evolution along the lineage leading to humans, after the divergence with chimpanzees (see here for a review of relative rate tests). They dubbed these rapidly evolving human sequences “human accelerated regions” (HARs). The fastest evolving HAR contained a sequence encoding an RNA that is expressed in the developing neocortex, but not translated into a protein. The authors also attempted to detect signatures of positive selection using polymorphism data (see here and here), but they did not find much evidence for adaptive evolution from either the publicly available SNP data or their own resequencing data.
In their interpretation of the results, Pollard et al suggest that it’s unclear whether the HARs are rapidly evolving because of natural selection or higher rates of mutation in humans. But in an opinion piece published in Trends in Genetics, Nicolas Galtier and Laurent Duret argue that variation in mutation rate is a more likely explanation for many of the HARs. Their argument is based upon the types of changes that have occurred within HARs and their location in the genome.
One nice feature of Pollard et al’s study is the fact that they exclude relaxed selective constraint as an explanation for accelerated evolution along the human lineage. They do so by comparing the rates of evolution of HARs against that of four-fold synonymous sites (third positions in codons which do not change the amino acid encoded by the codon), which are assumed to evolve neutrally. If the HARs evolved at a rate equivalent to that of the synonymous sites, then the accelerated evolution would most likely be the result of relaxed selective constraint along the human lineage. But the HARs evolve much faster than the neutral rate, suggesting that the changes along the human lineage were driven by natural selection.
The changes along the human lineage are also biased towards going from adenine (A) and thymine (T) to guanine (G) and cytosine (C). These types of changes are referred to as weak to strong because A and T are held together by two hydrogen bonds, while G and C are held together by three hydrogen bonds. Throughout the majority of a genome, weak to strong and strong to weak changes should occur at approximately equal frequencies. At recombination hotspots, however, weak to strong changes will dominate. That’s because the recombination events are initiated by double strand breaks (DSBs) in DNA, which are then repaired by one of a few DSB repair pathways. If an individual is heterozygous for an A/T and a G/C nucleotide at a site near a DSB, one of the DSB repair pathways is known to preferentially replace the A/T with G/C during the repair event (known as biased gene conversion, or BGC).
That’s the mechanism by which mutational bias would allow for accelerated evolution in one region of a genome but not another. This would not explain accelerated evolution along any particular lineage unless the recombination hotspots differed between species. And they do! Gil McVean and colleagues used polymorphism data to study recombination in the human genome (reviewed here). They found that recombination hotspots are ephemeral, changing often since the divergence of humans and chimps. If recombination hotspots have changed, and those hotspots induce weak to strong mutations, and HARs are biased for weak to strong mutations, it follows that HARs can be explained by mutational bias.
Galtier and Duret use that logic to argue that many of the HARs identified by Pollard et al evolved rapidly because of mutation bias, not natural selection. Pollard et al do acknowledge the possibility that BGC could be responsible for the HARs, but nowhere near as aggressively as Galtier and Duret. Not only do the HARs have an excess of weak to strong substitutions, they are also located near recombination hotspots, further supporting the mutation explanation. Additionally, if natural selection were responsible for the HARs, you would expect the signature of selection spread widely around the region under selection. BGC, on the other hand, will produce a much more localized signature, which is what Pollard et al observed.
If Galtier and Duret’s hypothesis is correct, then these HARs represent an “Achilles’ heel” of the genome — essentially the opposite of the natural selection hypothesis. Rather than being shaped by natural selection, the HARs may be highly mutable regions which must be tolerated despite the fact that they occur in functional regions (remember, they are highly conserved amongst all mammals studied except for humans). If a recombination hotspot were to appear in a functional sequence (or a functional sequence were to migrate into a highly recombining region, like Fxy in mice), you would observe accelerated evolution in that region. Natural selection would need to be even stronger to maintain the function of that sequence, but many of those sequences would rapidly evolve despite being under strong selective constraint.
Galtier N and Duret L. 2007. Adaptation or biased gene conversion? Extending the null hypothesis of molecular evolution. Trends Genet 23: 273-277 doi:10.1016/j.tig.2007.03.011
Pollard KS, Salama SR, King B, Kern AD, Dreszer T, et al. 2006. Forces shaping the fastest evolving regions in the human genome. PLoS Genet 2: e168 doi:10.1371/journal.pgen.0020168
Pollard KS, Salama SR, Lambert N, Lambot M-A, Coppens S, et al. 2006. An RNA gene expressed during cortical development evolved rapidly in humans. Nature 443: 167-172 doi:10.1038/nature05113
Spencer CCA, Deloukas P, Hunt S, Mullikin J, Myers S, et al. 2006. The influence of recombination on human genetic diversity. PloS Genet 2: e148 doi:10.1371/journal.pgen.0020148