Y-chromosome genetic variation associated with atherosclerosis and inflammation

Human Genome-Wide Association Studies (GWAS) have identified over 100 loci contributing to Coronary Artery Disease (CAD), but such studies have largely ignored the Y chromosome (chr) ¹. In this issue of ATVB, Eales and colleagues have identified a significant association between genetic variants on the Y chr and CAD in 129,133 men from the UK Biobank study ². They followed up on this finding by identifying epigenetic and gene expression patterns associated with the variants and they present evidence that the Y chr gene UTY, a histone demethylase, may contribute through its inflammatory effects in macrophages.

The Y chr is unique among chromosomes being present in one copy per genome. Whereas the 22 pairs of autosomes and the X chr undergo recombination during meiosis, thereby shuffling the genetic variants they carry, only the tips of the Y chr, that are homologous to the tips of the X chr (the pseudoautosomal regions) undergo recombination. Thus, with the exception of new mutations, the male-specific, non-recombinant region, encompassing over 95% of the Y chr, is inherited intact from father to son. The combination of genetic variants that a chromosome contains is termed a “haplotype”, and because of the conserved nature of the haplotypes, the Y chr has proven useful for deconvoluting the origins of human populations ³.

Based on some smaller previous studies, the authors hypothesized that some Y chr defined by their haplotypes may contain genetic variants predisposing to CAD as compared to other Y chr ^{4, 5}. To test this hypothesis they turned to a very large population-based cohort study from the United Kingdom, termed the UK Biobank, that includes more than half a million individuals characterized for an extensive range of health related traits. Unlike most other human population based studies, the UK Biobank is open access, allowing researchers around the world to utilize the data. All of the individuals in the study were genotyped for hundreds of thousands of single nucleotide polymorphisms (SNPs), including 162 on the Y chr. From these, there were 85 different groups of similar haplotypes (haplogroups) in the male specific region, and 7 of these were relatively common (frequency > 1%) (Figure 1). Using multivariate logistic regression, the authors found that carriers of one haplogroup, I1, exhibited significantly increased CAD among the 11,234 cases and 117,899 CAD-free controls. Thus, after correction for other known risk factors, these carriers showed 11% excess CAD, and odds ratios of 1.11, as compared to the other Y chr.

Identification of an analysis of Y-chromosome variation contributing to atherosclerosis.

Genetic variants of the male-specific region of the Y-chromosome are defined by common single nucleotide polymorphisms (SNPs) with alleles indicated by a “1” or a “2” in this made up example. A particular combination of the SNPs, say 2,1,1,2, is termed a haplotype. Altogether, 162 SNPs were examined by Eales and colleagues and closely related haplotypes were grouped together as “haplogroups”. (A) The figure shows three different Y chromosome haplogroups for purposes of illustration. (B) The 7 common haplogroups in the study were tested for association and one (haplogroup I1) was found to occur significantly more frequently in patients than in controls in the UK Biobank dataset. (C) Following additional sequencing, putative regulatory variations specific to the identified haplogroup were identified and one protein coding gene, UTY, out of 11 examined, was associated with the haplogroup. (D) Knockdown of the UTY gene in cultured macrophages using antisense oligonucleotides revealed strong effects on gene expression for genes in inflammatory and coagulation pathways.

Because of the lack of recombination, it was not possible to identify a specific locus on the Y chr associated with the increased risk. To help understand how the Y chr affects disease, the authors performed targeted sequencing of the I1 chr to identify additional associated variants. None of these variants were in coding regions of genes but many mapped to chromatin regulatory elements identified by Roadmap Epigenomics. These analyses further showed that the I1 haplogroup was enriched for variants in active chromatin states in cells and tissues relevant to atherosclerosis as compared to the other haplogroups. The authors then examined gene expression data of relevant tissues in two datasets, the GTEx study of the general population and the STAGE study of CAD patients. Consistent changes in gene expression associated with haplogroup I1 were observed, and these were enriched in several pathways including innate and adaptive immunity, lipid metabolism and coagulation.

Apart from the sex determining factor SRY, the Y chromosome carries relatively few functional genes. Of 11 ubiquitously expressed Y chr protein coding genes in blood-derived macrophages only one, Ubiquitously Transcribed Tetratricopeptide Repeat Containing Y-Linked (UTY), was associated with the I1-haplogroup. UTY is a minor histocompatibility antigen with possible contributions to antigen processing and it also exhibits histone demethylase activity. The authors performed knockdown of UTY in macrophages and observed substantial effects on gene expression, with enrichment for pathways such as immunity and coagulation, consistent with the above analyses of the I1 haplogroup.

The association of the I1 haplogroup with risk of CAD was modest, with a similar effect size as most of the loci identified on other chromosomes ⁶. Thus, although the clinical value of these findings is only an incremental, the evidence supporting pro-inflammatory and hemostatic impact through macrophages is of potential therapeutic significance. It is interesting that the I1 haplogroup is of relatively recent origin, with a most recent common ancestor that lived about 4000 years ago. It has primarily a Northern European distribution and conceivably could have a selective advantage related to immune defense ⁷. Lastly, it is noteworthy that these findings suggest that in addition to hormonal effects, sex chromosomes directly contribute to sex differences in the susceptibility to CAD.