Genome-Wide Association Studies in Gout

Hello, my name is Tony Merriman. I'm a professor in the division of clinical immunology and rheumatology at University of Alabama Birmingham and I'm speaking from my office there. I'm going to talk to the RheumNow folks today about genome-wide association studies and gout and the advances in knowledge we've got from them over the past 20 years or so.

So a genome-wide association study has really revolutionized the understanding of complex diseases like gout, which we know are caused by dozens if not hundreds of inherited genetic variants, but also there's the environmental aspect to it as well, in order to identify these genetic variants and then look at the pathways these variants are operating in, which will open up new therapeutic target possibilities.

These genetic variants have a weak effect, so we need very large cohorts to do GWAS studies. So the first thing is very large cohorts, and I'm talking the size of hundreds of thousands of people. The second thing is being able to measure throughout the genome as much of the genetic variation as you can, and we measure these days millions of single nucleotide polymorphisms — SNPs — or I probably call them common genetic variants. So we need technology to do that. Those are the two important things: large cohorts and technology to measure these variants. And you need very large cohorts because you're simultaneously testing millions of genetic variants, so you have to account for those simultaneous millions of variants in your p-value at the end of the day. We have to have a p-value that's in the order of 10 to the minus 8 to identify a locus or a gene as associated with gout.

Initially the first GWAS was done not in gout but in serum urate levels, and we know how critical and causal hyperuricemia is for gout. So GWAS done about 10 years ago in serum urate levels, led by Professor Anna Köttgen from Germany, yielded a lot of insights. In particular, it emphasized genetic variation and genetic control of the expression and activity of urate transporters, particularly in the kidney. The genes with the largest effect on urate levels and also gout are genes like SLC2A9, which is a re-uptake transporter in the kidney. It has an extremely large effect on urate levels, explaining about 3% of variation in urate levels in the general population, and that's a really large effect for a gene in a common condition like hyperuricemia. Another transporter is ABCG2, which is an efflux transporter working in both the gut and the kidney. There are other transporters found as well, but that emphasized the importance of genetic control of excretion of urate as central to the genetic basis of gout.

More recently, genome-wide association studies have been done in gout itself, and I have led them along with colleagues in China, Japan, and Europe. We were essentially looking for and addressing the question: not everyone with hyperuricemia goes on to get gout — what contribution does genetics have to that, and what genes might they be?

We published a very large study two years ago including 120,000 gout patients and 2.5 million controls, and we found a total of about 400 loci or genetic variants. But what was kind of surprising was only 10% of them seem to be exclusively involved in the inflammatory aspect of gout. In other words, these 10% of genetic variants didn't associate with urate levels, so therefore we hypothesize that they associate with inflammation and gout. These genetic variants are in genes that weren't surprising, like IL-1 beta, the IL-1 receptor, and other cytokine genes. There was also another very interesting gene called colony stimulating factor 1, which is important in controlling the differentiation of monocytes into macrophages and controlling their response to crystals.

I bring up this gene as an example because we've studied it further and it epitomizes the type of complex genetic control — genomic genetic control — that's happening at loci. So this gene, CSF1, is transcribed when monocytes see monosodium urate crystals that cause gout. But the genetic control is complex; it involves long non-coding RNAs, two of them being initially transcribed and then connecting up with the promoter of CSF1, and you get transcription there.

I just want to finish off by highlighting another pathway that we found in the genome-wide association study in gout. What we noticed was that genes controlling the pathway called clonal hematopoiesis of indeterminate potential were coming up quite regularly. This pathway, which I'll call CHIP, was identified with somatic mutations in these genes in older people that increase the chance of hematological cancer. So we were able to — we noticed
that genes like TET2 or TET2 were coming up and TET2 is very — very commonly mutated somatically in clonal hematopoiesis of indeterminate potential, and this may reflect — the genes that are involved in the CHIP pathway generally are involved in remodeling the genome epigenomically, and it may be involved in — what we do know is that when monocytes are exposed to high levels of urate, you get epigenetic epigenomic remodeling that increases the responsiveness of these cells to urate crystals. So we think it may be involved in that aspect — in other words, sort of training the immune system to be more responsive to urate crystals. So that was another finding from the gout genome-wide association study.

So to summarize, the genetic control of urate and gout is very complex — many, many, you know, hundreds of genes — and we expect we will find more as the cohorts get larger. The genes have weak effect size but collectively they modulate a person's proclivity to develop gout under the right environmental conditions.

Thank you very much for listening, and if you have any more questions, head to the RheumNow site. Thank you.