High blood pressure affects 1 billion people world-wide and is a major risk factor for cardiovascular diseases including heart attack and stroke, the leading causes of death world-wide. We have identified mutations in genes that drive blood pressure to the highest and lowest levels seen in the human population. These genes encode mediators and regulators of renal salt reabsorption; mutations that cause severe hypertension markedly increase salt reabsorption, while severe hypotension arises from impaired salt reabsorption. These findings provide the scientific basis for strategies to prevent and treat hypertension and its morbid consequences in the general population.
Recent studies have identified the molecular basis of aldosterone-producing adenoma, adrenal tumors that are a common cause of severe hypertension. Forty-five percent of these tumors are caused by either of two single somatic mutations that alter the selectivity filter of the K+ channel encoded by KCNJ5. Another 10% are caused by single somatic mutations in the Ca2+ channel encoded by CACNA1D. KNCJ5 mutations allow the channel to conduct Na+, which results in membrane depolarization and activation of voltage-gated Ca2+ channels, resulting in Ca2+ entry. CACNA1D mutations are gain of function and cause constitutive Ca2+ entry. In both cases, Ca2+ influx provides the normal signal for aldosterone biosynthesis and for cell proliferation, accounting for the cardinal features of these tumors. Proof that these single somatic mutations are sufficient for disease pathogenesis comes from rare patients with the same mutations occurring in the germline, in which case every adrenal glomerulosa cell is proliferating and producing aldosterone. Patients with germline KCNJ5 mutations develop massive adrenal hyperplasia and severe hypertension that requires adrenalectomy in the first years of life. Based on the simple genetics of these tumors we are attempting to develop in vitro diagnostic tests for these characteristic mutations in venous blood samples.
We have gone on to show that other tumors that constitutively produce hormones- insulin-secreting pancreatic islet cell tumors and cortisol-producing tumors of the adrenal gland- also are commonly caused by single mutations (in the transcription factor YY1 and the catalytic subunit of protein kinase A (PRKACA), respectively, in each case by causing constitutive activity of pathways that promote both cell proliferation and hormone release.
A Novel Pathway Regulating the Balance Between Salt Reabsorption and K+ Secretion
We have identified mutations in four genes, WNK1, WNK4, KLHL3 and CUL3 that cause a unique syndrome of hypertension with hyperkalemia, revealing that the balance between salt reabsorption and K+ secretion is explicity regulated. WNK1 and WNK4 are novel Cl- - regulated kinases and CUL3 and KLHL3 are partners in a ubiquitin ligase. We have shown that the kelch domain of KLHL3 directly binds an acidic motif of WNK kinases, targeting them for ubiquitylation and degradation, and that disease-causing missense mutations in either WNK4 or the kelch domain of KLHL3 impair binding, leading to constitutively increased WNK4 levels. Further, angiotensin II signaling (the hormone that increases in response to volume depletion) phosphorylates a site in the kelch domain of KLHL3, increasing WNK4 levels. Thus these mutations phenocopy the normal physiologic mechanism that promotes Na-Cl reabsorption. WNK4 increases activity of the Na-Cl cotransporter in the distal convoluted tubule and promotes increased Cl- reabsorption in the collecting duct while simultaneously inhibiting K+ secretion.
These findings reveal a previously unrecognized pathway in which K+ secretion requires reduced Na-Cl reabsorption. The clinical importance of these findings is that they explain the known effect of increased dietary K+ to reduce blood pressure, findings that have implications for prevention of hypertension in the general population.
Identifying New Mendelian Traits and Genes
There are ~4000 Mendelian traits recognized from the patterns of trait recurrence in families; these have been shown to arise from mutation in ~3000 genes. Nonetheless, virtually all of the ~20,000 protein-coding genes are conserved among all vertebrates, indicating they are maintained by purifying selection. If mutation of these genes impart large effects on human traits, why have these not been previously recognized by traditional Mendelian patterns of recurrence in families? We have used exome sequencing to explore these possibilities by studying cohorts of patients with extreme phenotypes to search for greater burden of mutation in specific genes and pathways than expected by chance.
Diseases Caused by de novo Mutations
By sequencing healthy parent – affected offspring trios as part of the NIH Pediatric Cardiac Genomics Consortium, we have shown that de novo mutations contribute to ~10% of all severe congenital heart disease cases, with an even higher fraction among patients with congenital heart disease plus another congenital malformation. These are predominantly haploinsufficient mutations with greatest enrichment in genes whose products enzymatically modify histone proteins, revealing dosage-sensitivity for these genes.
We have similarly used this approach to study newborns with severe unexplained diseases. For example, we identified a novel syndrome featuring periodic fever, neonatal enterocolitis and chronic inflammation caused by a de novo gain-of-function mutation in NLRC4, which encodes a core inflammasome protein, resulting in constitutive production of IL-1B and IL-18. Use of IL-1 antagonists in affected subjects can reduce chronic inflammation and febrile episodes in these patients.
Diseases Caused by Rare Recessive Mutations
Rare recessive traits with high locus heterogeneity (in which case mutations in any single gene account for a small fraction of families) can be solved by sequencing cohorts of patients with apparently similar extreme phenotypes arising in the setting of consanguineous union. For example, hemolytic-uremic syndrome (HUS) features episodic small vessel thrombosis leading to kidney failure. Genetic forms of disease have previously been attributed to mutations that cause constitutive activity of the complement cascade. By sequencing probands with HUS who did not have complement defects, we found that 20% of cases had homozygous loss of function mutations in DGKE, encoding diacylglycerol kinase epsilon. Patients with these mutations defined a distinct subset of patients who presented in the first year of life, had persistent hypertension and proteinuria, and progressed to early renal failure. Unlike patients with complement mutations, they did not respond to anti-complement therapy, but could be cured of disease with renal transplantation.
In collaboration with Murat Gunel at Yale and Friedhelm Hildebrandt at Boston Children’s Hospital, we have performed analogous studies of malformations of the cerebral cortex and renal development, identifying a substantial number of new disease loci.
Diseases Caused by Mutations with Low Penetrance
Mendelian traits that require environmental co-factors for expression of the trait can be difficult to recognize from the pattern of familial recurrence. In collaboration with Christine Garcia at U. Texas Southwestern, we sequenced 100 probands with familial pulmonary fibrosis. The results implicated haploinsufficiency with incomplete penetrance for two genes, PARN and RTEL1 in this disease. Like other genes mutated in this disease, affected patients have significantly shortened telomeres, implicating PARN and RTEL1 in telomere maintenance. PARN has since been shown to be required for maintenance of the telomerase RNA component TERC. Mutation carriers who develop pulmonary fibrosis have significant inhalational exposures including smoking and work as miners, welders, and work with farm animals, while mutation carriers who remain free of disease are typically free of inhalant exposure.
Collectively, these results are consistent with the proposition that mutation of virtually all human genes will have large phenotypic effects in the right environmental context, and define efficient strategies for their discovery. In the last four years, our efforts with collaborators around the globe have led to identification of several hundred new Mendelian trait loci and phenotypic expansions. Identification of the consequence of all human genes will define the opportunities for new diagnostic, preventive and therapeutic strategies for both rare and common diseases.
As of March 25, 2016