Feature: How we got our big brains

This feature appeared in the November/December 2009 issue of Australian Life Scientist. To subscribe to the magazine, go here.

It’s the biggest mystery in evolutionary genetics: what genetic changes drove the divergence of large-brained humans from their African great ape relatives, chimpanzees and bonobos, just 6 million years ago?

Dr James Sikela and his colleagues in the Human Medical Genetics and Neuroscience programmes at the University of Colorado, in Denver, have potentially linked the extraordinarily rapid increase in human brain size and cognitive capabilities to an unstable region of chromosome 1, and a striking increase in copy number of a genetic element, DUF1220, which appears to have had a major role in primate, and particularly human, evolution. Sikela is an invited international speaker at ComBio 2009 in Christchurch in December.

Using comparative genomics, Sikela’s team has shown that DUF1220 domains are far more common in humans than in any other primate: humans have 212; chimpanzees and bonobos only 34; orangutans and rhesus macacques have 30.

Sikela says the number of DUF1220 domains in primates declines with increasing evolutionary distance from humans. Rodents and all other non-primates studied to date have only one DUF1220 domain and they are absent in all other vertebrates.

Large brains and high-order cognitive abilities set humans apart from their great ape cousins and all other primates. A recently published analysis of 4.4 million year old fossils of Ardipitheus ramidus, the earliest direct ancestor of humans, shows a species in the process of exiting Eden. A. ramidus had a 400cc brain, the same size as a chimpanzee’s. It was making the transition from an arboreal in the tropical existence in the tropical rainforests of north-eastern Africa to life on the ground in Africa’s savannah woodland.

It seems A. ramidus had not yet come under the selection pressures favouring rapid amplification of DUF1220 domains that may have set pre-humans on the fast track for increased head and brain size, and the dramatic increase in cognitive capabilities that followed.

In a paper published in Science in September 2006, Sikela et al provided clues to what happened. The Japanese-American geneticist Susumu Ohno postulated in 1970 that gene duplication is a prime mover of evolution. The large and rapid evolutionary increase in DUF1220 copy number in primates and particularly humans fits nicely with this view, says Sikela.

This observation, and the fact that the sequences encoding DUF1220 domains show signs of positive selective, strongly points to DUF1220 playing an important role in human evolution. DUF1220 domains have increased in copy number both by gene duplication and domain amplification.

---PB---

Makes you think

Sikela describes DUF1220 domain gene family, also called NBPF, as containing virtually nothing but DUF1220 domains arrayed “wall-to-wall” in these proteins – an arrangement consistent with the view that there was something clearly evolutionarily advantageous about generating more copies of DUF1220 in the genome.

Currently the Sikela lab estimates there may be over 200 copies of DUF1220 in the human genome, encoded by 20-45 genes. The amplification process that spawned humans’ 30-plus NBPF genes can be traced back 60 million years, to the duplication in an ancestral primate of the original, mammalian prototype, O75042. It spun off duplicates and some of its MGC8902 clones in turn produced duplicates.

Perhaps the Denver team’s most exciting finding was that, while DUF1220 proteins are expressed in most human tissues, in the brain they are only found in neurons, and they are preferentially expressed in the cell bodies and dendrites of specific populations of neurons in areas of the brain involved in higher cognition, like the neocortex. While Sikela is intrigued by the coincidental increase in brain capacity, spectacular gains in higher cognitive function and the dramatic expansion of DUF1220 domain copy number in the human genome, he remains “agnostic” on the question of causality until more evidence comes in.

His team has been involved for almost a decade in searching for human lineage-specific genes that might reveal what selection pressures have made us so different, physically and cognitively, when we have 95 percent of our DNA in common with the other African great apes.

In 2004 his team, in collaboration with Jon Pollack at Stanford, published the first genome-wide gene array study of gene copy number differences between humans and great apes. Surveying over 24,000 genes, the study identified 134 genes that showed human lineage-specific increases in copy number. It is from this list that they identified DUF1220 domain amplification.

---PB---

All in the mind

Researchers still have no idea what DUF1220-domain proteins do, but Sikela suggests they have played an important role in human evolution, and possibly brain evolution. This is due to the rapid increase in copy number in primates and particularly humans, as well as the fact they exhibit strong signatures of positive selection, and their neuron-specific expression in important brain regions.

Sikela says he and his colleagues were the first to call attention to the fact that these CNVs either encompassed, or were immediately flanked by, DUF1220 domains. Recently they looked at DUF1220 copy number in individuals who have 1q21.1 CNVs and either microcephaly or macrocephaly, and found a strong correlation (p<0.0001) between DUF1220 domain copy number and brain size (head circumference). While Sikela says this is suggestive, there are other genes in the region that could be at play and further work on this is underway.

The Sikela lab has also recently shown that DUF1220 copy-number expansions commonly associate with macrocephaly (abnormally large head) and autism, while DUF1220 copy-number deletions tend to associate with microcephaly (reduced neocortex volume) and schizophenia.

Interestingly, he says, autism and schizophrenia have been proposed to be genomic sister disorders, conditions in which the phenotypes are direct opposites of one another and caused by reciprocal duplications and deletions of the same sequences. Variation at 1q21.1 has also been associated with idiopathic (of unknown origin) mental retardation, congenital heart disorders and neuroblastoma, the most common cancer of infancy. Neuroblastoma cancers affect the peripheral nervous system, appearing in organs like the adrenal glands or in neural tissues in the neck, chest, abdomen and pelvis.

---PB---

The cost of intelligence

Should DUF1220 domains be shown, even partly, to underlie the rapid increase in size of the human brain, Sikela has proposed a positive-feedback model to explain how this process may have come about.

“It is likely that the increase in DUF1220 copies was an ongoing process that, while rapid from an evolutionary perspective, occurred over many millions of years,” he says.

“The majority of DUF1220 sequences map to 1q21.1, which is an extremely complex and dynamic genomic region. Making the 1q21.1 region more unstable would increase the chance that DUF1220 duplications would occur, which would have increased brain size and as a result conferred a strong advantage.

“Selection of those individuals with more DUF1220 copies would result in retention of the 1q21.1 instability, allowing further DUF1220 copy number increases to more readily occur. But the downside of this process is that the instability is generally random or undirected, and would have also often led to deleterious effects and disease.”

In a paper published last month in the proceedings of the Cold Spring Harbour Symposia on Quantitative Biology, Sikela and his University of Colorado colleague L. Dumas summarised their findings thus: “… the high number of 1q21.1 CNVs that are disease-causing may be the price that our species paid, and continues to pay, for the adaptive benefit of large numbers of DUF1220 domains.”