Genome-wide association studies (GWAS) are large population based collaborative studies seeking genetic patterns that would explain common complex phenotypes. These large studies have given us much insight into the genetic risk profile of patients with diseases such as chronic kidney disease and hypertension. Frequently the genetic markers described as risk genotypes are single nucleotide polymorphisms (SNPs) that lie in regions of the genome yet to be ascribed a functional role. Trudu et al in Nature Medicine this month describe a beautiful set of experiments that explain how risk variants for CKD and hypertension found by GWAS effect blood pressure regulation at the molecular level.
A number of GWAS have described risk SNPs in the promoter region of UMOD (1, 2, 3,4, 5, 6, 7). The UMOD gene codes for uromodulin or Tamm-Horsfall protein, which is secreted into the urine by cells of the thick ascending loop of Henle (TAL). Uromodulin has been shown to reduce UTIs and regulate NKCC2 and ROMK, the two main channels responsible for NaCl transport in the TAL. Furthermore, UMOD mutations cause dominantly inherited CKD (MCKD2). Susceptibility variants found in the UMOD gene are at high frequency in the general population and confer a 20% increased risk of CKD and 15% risk of hypertension.
This paper set out to uncover the biological mechanism that would explain the increased CKD and hypertension in patients with these risk genotypes. They looked specifically at the 2 lead variants located in the UMOD promotor region. Briefly, in human nephrectomy samples (removed due to RCC), those with the risk genotypes had higher uromodulin expression than those with non-risk genotypes. They confirmed this association of UMOD promotor risk variants and higher urinary uromodulin in large population-based cohort (SKIPOGH). In mouse-models, mice over-expressing UMOD had higher blood pressures and more LVH than controls and had more interstitial pathology (despite normal renal function) than controls. They then showed that mice over-expressing UMOD had more active NKCC2 (furosemide sensitive channels) than controls. They also showed that the higher BP in UMOD over-expressing mice could be dropped to baseline levels by furosemide (all mice had comparable levels of ENaC and NCC). More work was done to further elucidate the mechanism of NKCC2 phosphorylation by UMOD. Finally the authors bring their attention back to humans. They used a never-treated human hypertensive cohort (MI_HPT) and stratified them according to one of the risk variant genotypes (rs4293393). Patients homozygous for the risk genotype had statistically significant higher baseline diastolic BP. Some of these patients underwent furosemide testing. Patients with the homozygous risk allele had a higher natriuretic response and diastolic BP drop than others (both statistically significant). To summarize the authors conclusions; patients with risk alleles in the UMOD promotor had greater risk of hypertension and CKD. These patients made more uromodulin. Uromodulin increases NKCC2 activity and thus salt sensitive hypertension. Also, UMOD over-expression increases interstitial kidney damage and thus increases the risk of CKD. These risk genotypes for disease are at high frequency in all ethnicities tested. The authors suggest that selective pressure for disease related variants in UMOD is similar to the APOL1 story. The protective effect of uromodulin on UTIs and ability to raise blood pressure may have lead to its selective over-expression.
This paper demonstrates the mutual importance of large-scale studies such as GWAS and studies of rare monogenetic diseases such as Bartter and Gitelman syndromes. This paper is a worthy contender of a top ten listing this year with a truly translational set of studies combining modern basic science with clinical studies.