Brain genes evolving at the origin of Homo?

Aside from the two papers finding very recent evolution in human brain-expressed genes, there is another paper (subscription required) in this week's science that documents an older genetic change in the brain. This paper is by Toshiyuki Hayakawa and colleagues from Ajit Varki's lab.

The lab studies sialic acid metabolism; sialic acids are part of the array of sugar chains (or "glycans") that adhere to the surfaces of cells, and are involved in cell signalling, among many other things. Thus, they are important in the metabolism of the brain (they are also involved in the malarial infection pathways discussed in this earlier post).

Here's the abstract:

Recent studies have shown multiple differences between humans and apes in sialic acid (Sia) biology, including Siglecs (Sia-recognizing-Ig-superfamily lectins). Comparisons with the chimpanzee genome indicate that human SIGLEC11 emerged through human-specific gene conversion by an adjacent pseudogene. Conversion involved 5' untranslated sequences and the Sia-recognition domain. This human protein shows reduced binding relative to the ancestral form but recognizes oligosialic acids, which are enriched in the brain. SIGLEC11 is expressed in human but not in chimpanzee brain microglia. Further studies will determine if this event was related to the evolution of Homo (Hayakawa et al. 2005:1163).

SIGLEC16 is a pseudogene – an inactive copy of a gene – shared by humans (hSIGLEC16) and chimpanzees (cSIGLEC16). It was generated by a gene duplication of the gene SIGLEC11 in some ancient primate, ancestral to both chimpanzees and humans, and possibly other extant species. </p>

This means that human SIGLEC11 ought to look more like chimpanzee SIGLEC11 than either looks like human (or chimpanzee) SIGLEC16 -- the two homologous genes should be more similar.

But instead, human SIGLEC11 looks more like human SIGLEC16 than it looks like chimpanzee SIGLEC11. This is a case of gene conversion -- genetic material has been exchanged from the hSIGLEC16 inactive pseudogene into functional hSIGLEC11.

The converted allele appears to be fixed in living humans, so it likely happened a long time ago (i.e. more than a few hundred thousand years). The paper suggests that it "may be related to the evolution of genus Homo" (1163). Maybe so, although that's a pretty general statement. It would be helpful to know how long ago the conversion might have happened, but that will take a lot of study of flanking sequence and may still be difficult, depending on how the conversion happened. There's no evidence that this occurred at the origin of the genus, although the loss of human Neu5Gc occurred around 3 million years ago; if this is related to that genetic change, it wouldn't be too surprising if it happened shortly thereafter.

What does the new converted gene do? The chimpanzee form of the protein binds more closely to the Sia type Neu5Gc, which is missing from humans (part of that ancient malaria defense). The human form of the protein is expressed in microglia cells in the brain, and binds to oligosialic acids in the brain.

And that's about what's in the paper, hopefully in slightly clearer language.

It's been a long time since a paleoanthropologist could write a paper with only six references!


Hayakawa T, Angata T, Lewis AL, Mikkelsen TS, Varki NM, Varki A. 2005. A human-specific gene in microglia. Science 309:1163.