Posts Tagged ‘GWAS’

The price of lighter skin?

Friday, February 19th, 2010

Initially posted 11 September 2009 by evomed

Chapter 3 introduced the concept of positive selection, which refers to the increase in frequency of specific traits that confer a fitness advantage. Table 3.1 also gave examples of several human genes which have been shown to be under recent positive selection (ie, within the last 10,000 years). It is thought that environmental changes act as strong selective pressures for genotypes that enable adaptation to the new local environment. An example of such a change would be the transition from hunter-gatherer to agricultural practices 10–12,000 years ago. The lactase-coding gene LCT, which enables hydrolysis of the predominant milk sugar lactose, provides an illustrative example. Positive selection for this gene, and hence lactase persistence into adulthood, has been shown in Northern European and East African populations and is attributed to the domestication of cattle and subsequent introduction of milk to the diet.

Several genes involved in skin pigmentation have also shown population-specific selective sweeps, suggesting that the evolution of human skin pigmentation is also driven by adaptation to different climates as humans migrated out of Africa towards more temperate regions. KITLG, which codes for a ligand of the tyrosine kinase receptor encoded by the KIT locus, is one such example. Among other biological properties, the Kit ligand plays a critical role in melanocyte development and migration. KITLG has been shown to be under positive selection in Europeans and East-Asians, but not Africans. Genotyping of an ancestral SNP (rs642742), located at a potentially regulatory region upstream of KITLG, demonstrated that Africans possessed the ancestral A allele while European and East Asian populations displayed a significantly higher frequency of the derived G allele that leads to lighter skin, possibly due to lower KITLG expression than that from the A allele (Miller et al. 2008). The selective pressure for lighter skin is not clear, although vitamin D requirements and sexual selection have been proposed.

Recently, Kanetsky et al. (2009) and Rapley et al. (2009) independently conducted genome-wide association studies (GWAS) to determine markers of testicular germ cell tumors (TGCT). TGCTs are the most commonly diagnosed cancer among young to mid-age males. Genotype frequencies for cases and controls of European ancestry were determined and compared. Seven of the eight SNP markers that reached genome-wide statistical significance (P < 5.0 x 10–8) in the study by Kanetsky et al. were located within the KITLG gene region on 12q22. Independent replication was then performed on two of the SNPs (rs3782179 and rs4474514) in another cohort. Using similar methodology, Rapley et al. found strong evidence of association for two SNPs located on chromosome 12 (rs995030 and rs1508595) which held up after replication. The estimated per-allele odds ratios (OR) for the susceptibility loci on chromosome 12 in the two studies ranged from 2.55 to 3.08—ratios that are remarkably high in comparison to the OR for other GWAS-identified cancer susceptibility loci.

The findings from both teams clearly demonstrate that KITLG is a risk factor for the development of TGCT. It is interesting to note, as have Kanetsky et al., that the incidence of TGCT in men of European ancestry is almost fivefold higher than in black men.  The findings, coupled with HapMap data showing significantly higher disease allele frequency in European than in African ancestry populations, suggest that the difference may be explained at least partially by inherited variation at the KITLG locus. It is important to bear in mind that GWAS can only reveal associations, not causality; however biochemical studies investigating the mechanisms by which lower KITLG expression affects melanocyte properties, and the role that KITLG plays in the development of TGCT, may provide some clues. The similar results from the two TGCT studies suggest that potentially negative consequences of positive selection cannot be overlooked.


Tags: ,




Regulating the age of puberty

Friday, February 19th, 2010

Initially posted 27 July 2009 by evomed

The age of puberty is a central feature of a species’ life history. The mechanism that controls the onset of sexual maturity evolved to allow for successful reproduction. Reproduction is energetically costly for the female and a degree of physical and psychosocial maturity is necessary for successful pregnancy and infant care. In humans, the timing of puberty evolved within ecological conditions that were radically different from ours and in which social independence was achieved at a younger age. Indeed, the gap between the sexual and social maturation continues to increase. Its steady growth is caused not just by the delay of social independence in the modern society but also by the falling age at puberty observed worldwide. It seems that the controls evolved to permit earlier puberty in the conditions of abundance are permanently switched on in our affluent society, where children are taller and heavier than ever before.

Studies published recently in Nature Genetics have begun to reveal some of the biological mechanisms behind the onset of puberty and of its link to the height and weight. A team led by John Perry conducted a meta-analysis of genome-wide association data of 17,520 women from eight different population-based cohorts. The women were all of European descent and had a self-reported age at menarche between 9 and 17 years, with the mean of 13.22 years. The SNPs that passed the significance threshold were all located at either chromosome 6 (6q21) or 9 (9q31.2). The strongest signal at 9q31.2 was observed with SNP rs2090409 (nearest genes TMEM38B, FKTN, FSD1L, TAL2 and ZNF462), where each A allele was associated with a 5-week reduction in menarcheal age. The T allele at rs7759938 within the 6q21 signal was also associated with a 5-week reduction in menarcheal age and it was found near a gene previously associated with a variation in human height, LIN28B. In the same issue of Nature Genetics, Ong at al confirmed the link between LIN28B on chromosome 6 and reduced age at menarche in several cohorts, with each copy of the major C allele at rs314276 reducing the age of menarche by 0.10-0.22 years. The same allele was then found to be associated with earlier breast development, and a higher BMI. In boys, it was found to be linked at age 15 with more advanced pubic hair stage, voice breaking status and tempo of height growth. In both sexes the allele was linked with faster tempo of growth in height between ages 7 and 11.

LIN28B shows high sequence, structural and functional homology with LIN28 on chromosome 1 and both exhibit sequence homology to lin-28 in the much-studied nematode worm Caenorhabditis elegans; deleterious mutations in that gene produce an abnormal tempo of development, while the enhancement of its expression by deletion of regulatory elements delays larval stage expression. It is thus suggested that LIN28B, the first cell marker associated with the timing of puberty, is associated with an evolutionarily old cell regulatory system of human growth and development.

While important, these studies explain only a small part of observed variation: even in homozygotes, the C allele at rs314276 upstream of LIN28B accounts for just a few months of the observed difference. These studies furthermore do not explain the ‘secular trend’ or the widely observed reduced age at puberty in the modern world. It is hoped that the location of other similar cell markers and a closer look into the regulation of their expression, especially during the plastic developmental period, will shed more light on this fundamental life phase.


Tags: , , ,