NEW YORK (GenomeWeb News) – Investigators from the US, France, and Cameroon brought together information on genome-wide ancestry patterns, signals of selection, and more in their search for genetic factors behind diminutive stature of Western African Pygmy populations — work that they described online last night in PLoS Genetics.
"This was a first attempt to see what we might actually find using a genome-wide set of markers and correlating this data with information about ancestry, natural selection, and height," senior author Sarah Tishkoff, a genetics researcher at the University of Pennsylvania, told GenomeWeb Daily News.
The researchers used SNP arrays to profile genetic patterns in dozens of individuals from three Western African Pygmy populations, known for their short stature, and from nearby Bantu-speaking populations, which are typically much taller. The search led to regions of the genome under selection in the Pygmy populations, including some that contain genes suspected of influencing not only height but also immune and neuroendocrine-related traits.
"Everything is interconnected," Tishkoff said. "So there may be a suite of co-adapted traits in Pygmies. It may not even be just [adaptations related to] short stature or just immunity or just reproduction — maybe it's all of the above."
Human populations living in dense tropical forests around the world appear to have undergone convergent adaptations to their environments that include especially short stature. A wide range of theories have been proposed in terms of possible advantages of being smaller in such environments, Tishkoff explained, from reduced caloric requirements and better thermal regulation to faster maturation and earlier reproduction in conditions associated with shorter lifespan.
So far, though, the genetics of height have been best studied in European populations, where past studies point to 180 or more genes influencing the complex trait.
"Who's to say that other populations have the same genetic architecture?" Tishkoff said. "What if it's completely different? What if it's not hundreds of genes but a handful of genes? What if there's very strong selection for the extremes of the [height] phenotype?"
To explore such questions, she and her colleagues relied on Illumina 1M SNP array data generated for 67 individuals from three Western Pygmy populations (the Baka, Bakola, and Bedzan) in Cameroon and 58 individuals from three neighboring Bantu-speaking populations (the Ngumba, Southern Tikar, and Lemande).
Tishkoff credits first author Joseph Jarvis with leading the analysis of this data, which involved efforts to discern genome-wide ancestry in Pygmy individuals as well as searches for signals of selection and height-associated genetic factors.
Jarvis, a postdoctoral researcher at the University of Pennsylvania when the study was performed, is now a researcher at the Coriell Institute for Medical Research.
West African Pygmy populations are thought to have diverged from Bantu-speaking populations in Africa some 60,000 to 70,000 years ago, the study's authors explained. Even so, a good deal of gene flow from the Bantu populations into the Pygmy populations has occurred more recently, as Bantu populations moved into formerly forested areas.
Consequently, virtually all of the Pygmy individuals tested had admixed genomes, Tishkoff explained, complicating efforts to discern genome-wide ancestry patterns in these individuals.
In an effort to tease apart ancestry tracts in the absence of information on allele frequency in ancestral Pygmy populations, researchers came up with new computational approaches for trying to tease apart Pygmy and Bantu ancestry.
While the ability to resolve ancestry is expected to improve in the future as more individuals are tested at more sites in the genome, Tishkoff said, ancestry patterns found so far point to Bantu gene flow into the West African Pygmy populations stretching back 1,000 years or more.
The ancestry analyses also supported the notion that Pygmy height is genetically determined, since Pygmy individuals with higher levels of Bantu ancestry in their genomes tended to be taller.
But while Bantu ancestry was generally peppered throughout the genome, researchers found one area where that was not the case: a stretch of DNA on chromosome 3 where ancestry tended to be either exclusively Pygmy or exclusively Bantu.
Investigators found the same chromosome 3 region when they searched for sites under selection in the Pygmy genome, focusing on regions that were specifically differentiated in the Pygmy populations compared to Bantu populations and the Maasai populations from East Africa.
Within other parts of the genome showing signs of selection in the Pygmy populations, meanwhile, the team found genes involved in immune-related pathways and neuroendocrine signaling pathways involving oxytocin, serotonin, and the thyrotropin-releasing hormone receptor.
The latter pathway was of particular interest, Tishkoff said, since the thyrotropin-releasing hormone receptor mediates the release of thyroid hormone, which itself contributes to thermoregulation, reproduction, immunity, and stature.
Coupled with the observation that Western Pygmy populations are less prone to goiter than neighboring Bantu populations, researchers speculated that the same thyroid-related features that may have helped Pygmy populations adapt to their iodine poor diet could also have contributed to other physiological changes, including height.
"It's intriguing to think that an adaptation to an aspect of their diet might also influence other aspects of their metabolism and other aspects of their phenotype," Tishkoff said.
Because they were underpowered to look for associations with height genome-wide, owing to their small sample size, researchers started by looking at height-related variants within regions under selection in the Pygmy genomes.
Again, that search brought them back to the same chromosome 3 region that caught their eye in earlier analyses.
In Pygmy and other human populations, the region contains eight SNPs in complete linkage disequilibrium with one another, Tishkoff explained, perhaps due to an ancient inversion polymorphism that hinders recombination at that site in the genome.
The frequency with which variants in the chromosome 3 region are found in Pygmy populations is of interest, since this area contains a gene called CISH, which codes for an immune regulatory protein that also curbs human growth hormone receptor activity. Also found in the region: DOCK3, one of the genes falling near variants that influence height in European populations.
Of the 180 or more genes associated with height in European populations, though, DOCK3 was the only one that was also a height-related candidate gene in Western African Pygmies, suggesting these populations have distinct genetic architecture for height compared to Europeans.
Meanwhile, a broader search for variants with ties to height in the Pygmy populations pointed to SNPs in and around genes in growth hormone and insulin-related pathways.
Because the array used for the study was designed to target SNPs in other populations, those involved in the study say there are likely functionally important variants that were missed in the current analysis. For instance, Tishkoff suspects there may be undetected regulatory variants affecting the growth hormone and insulin-related pathways identified in their current analysis.
"We're going to need whole-genome studies and gene expression studies in the right tissues to be able to identify these variants," she said.