1 SNP to rule them & in the darkness bind them?

Last year a group out of Australia published a paper which purported to explain eye color variation based upon a polymorphism around the OCA2 locus. The paper was A Three-Single-Nucleotide Polymorphism Haplotype in Intron 1 of OCA2 Explains Most Human Eye-Color Variation, and I blogged it here. Basically the paper showed that three SNPs arranged on several haplotypes could be plugged into a function to generate a relatively good prediction of eye color. Why does this matter? First, because eye color is one of the first things you learn about "genetics" in high school, but we're still stuck in the theoretical Mendelian land where we have to infer from inheritance patterns of putative loci instead of knowing where they are empirically in reality. I remember one girl in high school being taken a back learning that blue eyed parents could not have brown eyed children, ever, she being a brown eyed daughter of blue eyed parents and with no knowledge of adoption. Second, there are forensic uses that might be made of knowledge of the genes which control physical appearance. Finally, there are some interesting evolutionary questions which emerge out of examining salient phenotypic characteristics which vary between populations. As most of you probably know, light eye color is predominantly a European trait. I have argued before that it emerged because of its affect on skin color, with eye color being a secondary byproduct, in the recent evolutionary past (i.e., last 10,000 years).

All that being said, the same group (or some of them) of Australian researchers are now coming out with a new paper (it is provisional) in The American Journal of Human Genetics, A Single SNP in an Evolutionary Conserved Region within Intron 86 of the HERC2 Gene Determines Human Blue-Brown Eye Color. Wait, wasn't I talking about OCA2? Yes, I was, but it turns out HERC2 is right next to OCA2, the same genomic neighborhood. Remember that the SNPs were on an intron, a non-coding region of a gene, so they were useful tags or markers in the neighborhood, not the "real deal" so to speak (as a contrast, SLC2A5, the gene which accounts for 1/4 to 2/5 of the skin color difference between Africans & Europeans has a substitution on a functional base pair that changes the amino acid coded). It's not the sort of answer that would satisfy a molecular geneticist, but if you're a forensic investigator or curious about evolutionary history it'll do. In this follow up paper the researchers come closer to answering the functional questions which a molecular geneticist might pose; and HERC2 is more likely to give an answer than not.¹ Here's their abstract:

We have previously demonstrated that haplotypes of three single nucleotide polymorphisms (SNPs) within the first intron of the OCA2 gene are extremely strongly associated with variation in human eye color. In the present work, we describe additional fine association mapping of eye color SNPs in the intergenic region upstream of OCA2 and within the neighboring HERC2 (hect domain and RLD2) gene. We screened an additional 92 SNPs in 300–3000 European individuals and found that a single SNP in intron 86 of HERC2, rs12913832, predicted eye color significantly better (ordinal logistic regression R² = 0.68, association LOD = 444) than our previous best OCA2 haplotype. Comparison of sequence alignments of multiple species showed that this SNP lies in the center of a short highly conserved sequence and that the blue-eye-associated allele (frequency 78%) breaks up this conserved sequence, part of which forms a consensus binding site for the helicase-like transcription factor (HLTF).We were also able to demonstrate the OCA2 R419Q, rs1800407, coding SNP acts as a penetrance modifier of this new HERC2 SNP for eye color, and somewhat independently, of melanoma risk. We conclude that the conserved region around rs12913832 represents a regulatory region controlling constitutive expression of OCA2 and that the C allele at rs12913832 leads to decreased expression of OCA2, particularly within iris melanocytes, which we postulate to be the ultimate cause of blue eye color.

The linkage study here is showing the posterior probability of eye color conditional upon a prior genetic architecture. An R² of 0.68 is pretty good, meaning most of the variation can be attributed to the SNP in question. If an LOD is above 3 that should indicate genetic linkage, so they seem safe on that ground. If you read the previous paper and my posts on it the shell game of three SNPs, haplotypes and diplotypes (analogous to alleles and genotypes) could get a little confusing. So in the interests of verbal simplicity this seems a step forward, and, it will make high school genetics a lot easier since the physical correspondence between one base pair and the phenotypic variation is easier to present didactically.²

But the paper doesn't stop at the statistical analysis which might satisfy the crime scene investigation types. What's going on between HERC2 and OCA2? Note that the new SNP is on an intron, so we're not talking a nonsynonymous substitution here, it isn't the alteration of some protein product that HERC2 is producing. The researchers cross-referenced this region of the genome across several species and found that intron 86, where our SNP is located, is highly conserved. It seems to show up on a test for regulatory significance, in other words it might not be a producer of gene product, but it could play a large control in controlling levels of expression. Here's the critical part:

Analysis of this sequence with the MatInspector Program suggests that this may represent a HLTF (SMARCA3) binding site. Tellingly, this consensus sequence is abolished in the rs12913832*C blue-eyed allele....

HLTF is "helicase-like transcription factor." As implied by their name transcription factors up or down regulate the production of mRNA from the DNA; that is, they can crank up or shut down production of gene product by controlling the amount of RNA precursor early on in the various steps that lead to that end point. The section above is implying that on intron 86 there's a physical region that HLTF hooks onto, and this grappling is why there is a "consensus sequence" which is seen across the species in their queries. You knock that out and all of a sudden HLTF has no where to grab, and you get a nice little cascade which can abolish the expression of the gene downstream.

OCA2 does have something to do with pigmentation pathways. Its long-form name is "oculocutaneous albinism II," which is telling you that it was originally detected due to its association with a form of albinism. The inference made from this paper is that the SNP that they're fixing on is associated with a break around the region which is essential to the action of a transcription factor which modulates the expression of OCA2. As I said above, the OCA2 polymorphisms which this group associated with eye color also have a relationship to skin color (additive, with fair and olive homozygotes, and medium-complected heterozygotes). And in this study they found another SNP on OCA2 which has an independent effect on melanoma susceptibility and seemed to modulate the probability of blue or brown eyes as well (if an allele is incompletely penetrant that just means it doesn't always result in the phenotypic outcome you expect it to produce, I personally think it would be better to just admit that there's polygenic complexity which you're not accounting for, but they stick with penetrance). The likelihood of other SNPs makes if OCA2 really does have a strong functional significance when it comes to the pigmentation pathways. And that significance is also conditional upon tissue expression, after all there is a moderate impact upon skin color in these studies, but a major change in eye color. The authors conclude:

Based on the foregoing, we conclude that the conserved region around rs12913832 represents a regulatory regioncontrolling constitutive expression of OCA2, and that the C allele at rs12913832 leads to decreased expression of OCA2, particularly within iris melanocytes. We speculate that the regulatory mechanism is abrogation of the binding site for HLTF that regulates transcription of the neighboring OCA2 gene. We also confirmed that the common coding variant OCA2*R419Q acts to modify the penetrance of this locus and that this effect includes modification of the risk of malignant melanoma.

I love molecular geneticists with their terms like "abrogation." In any case, there's another paper on HERC2 out in AJHG which I'll cover next....

STURM, R. (2008). A Single SNP in an Evolutionary Conserved Region within Intron 86 of the HERC2 Gene Determines Human Blue-Brown Eye Color. The American Journal of Human Genetics DOI: 10.1016/j.ajhg.2007.11.005

1 - I'm not surprised by this really...if you pop OCA2 into Haplotter HERC2 actually has the much lower empirical p-value and comes out with a strong recent selection signature.

2 - I ran into a biological anthropologist who was actually confused by the exposition in this group's original paper, so it isn't just high school students who'll benefit.