L'hyperaldostéronisme primaire (HAP) est la plus fréquente cause identifiable de l'hypertension, et résulte de la production autonome d'aldostérone par les glandes surrénales. Chez la souris, la délétion génétique des canaux TASK1 et TASK3 provoque des changements biochimiques qui imitent HAP humain. Ces canaux permet la sortie de K+ et polarise le potentiel de la membrane des cellules glomérules. Nous avons étudié la variation et l'expression de KCNK3 et 9 chez l'homme. Notre étude d'association à montré aucune association d'HAP avec n'importe quel SNP au niveau de l'ensemble du génome. Le séquençage de l'ADN de la lignée germinale dans 825 cas d'HAP, et 41 échantillons d'ADN tumoral a abouti à 14 variantes différents dans KCNK3 et 9 dans la lignée germinale, dont 6 non-synonyme, 8 synonyme. Des tests in vitro n'ont montré aucune perte de la fonction du canal. Aucun changement de séquence somatique à été trouvé. L'hybridation-in-situ dans 6 glandes surrénales contrôle (CA) et 20 glandes adénomes produisant l'aldostérone (APA) a montré que KCNK3 été fortement exprimée dans les trois couches du cortex, tandis que l'expression de KCNK9 était faible et limitée au glomérule en CA. Dans les APA, l'expression de KCNK3 a été détectée, alors que l'expression KCNK9 était faible et hétérogène. Le transcriptome de 43 APA et 11 CA a révélé une légère surexpression de KCNK3 dans les APA, en corrélation avec l'expression de CYP11B2. La surexpression de TASK1 dans les APA peut être secondaire à un phénomène épigénétique. Alors que la variation de l'ADN est incompatible avec un rôle causal, il peut y avoir une possible contribution des changements d'expression de TASK1 dans HAP humain. / Hypertension is the leading cause of human mortality globally. Representing about a tenth of all patients, Primary Aldosteronism (PA) is the commonest identifiable cause of hypertension, and results from the autonomous production of aldosterone by the adrenal glands. The two principal sub-types are Bilateral Adrenal Hyperplasia (BAH), and Aldosterone Producing Adenoma (APA), which account for two-thirds and one-third of the cases respectively. The molecular etiology of primary aldosteronism has remained elusive until recently, when through an exome sequencing study, mutations in the potassium channel-coding gene KCNJ5 were found to cause PA in humans. These mutations were found in up to 40% of APAs, and only in a rare familial variety of BAH. A subsequent exome sequencing study identified mutations in ATPase famile genes in about 7% of APAs, bringing the total genetic yield to about 47%. The molecular pathology of more than half of APAs and of most BAHs remain unexplained. In mouse models, the genetic deletion of TASK-1 and TASK-3 potassium channels cause biochemical changes that resemble those seen in human PA. TASK 1 and TASK 3 are background ‘leak’ potassium channels, which by permitting the outward flow of K+ ions, polarise the adrenal glomerulosa cell membrane potential. The genetic removal of these channels therefore results in a marked depolarization of the glomerulosa cells, leading to their increased aldosterone secretory function, diagnosed as PA. In humans, the contribution of TASK-1 and TASK-3 channel dysfunction to PA has been negated by sequencing studies of the genes that code for these channels (KCNK3 and KCNK9 respectively). However, these studies have included only a small number of patients, motivating a comprehensive molecular analysis of the genes in a large patient cohort. To this end, we investigated commonly and rarely occuring genetic variation in, and expression of, KCNK 3 and KCNK9. Our Genome Wide Association Study (GWAS) showed no association of PA (either APA or BAH subtypes or both) with any single SNP at the genome-wide level of statistical significance. At sub genome-wide levels, however, SNPs of KCNK3 did associate, and the association signal strengthened when specific combinations of the SNPs were tested for association at a time. While no inherited or acquired DNA sequence variation in KCNK3 and KCNK9 have ever been detected in PA patients, on sequencing germline DNA in 825 PA cases, and 41 tumoral DNA samples, 14 different coding single nucleotide variants in KCNK3 and KCNK9 were found in the germline DNA only, of which 6 were non-synonymous, and 8 synonymous. However, on heterologous expression and electrophysiology, these did not affect channel function. No somatic sequence changes were found.Expression of KCNK3 and KCNK9 was investigated by in-situ hybridization in 6 control adrenal glands and 20 adrenals from patients with APA. In the control adrenal, the KCNK3 gene was highly expressed in all three layers of the adrenal cortex, while KCNK9 expression was barely detectable, and restricted to the zona glomerulosa. In APAs, KCNK3 expression was detected in a majority of patients, while KCNK9 expression was low and heterogeneous among samples. Strikingly, KCNK9 was highly expressed in the hyperplastic peritumoral zona glomerulosa, possibly due to a positive feed-back by high circulating aldosterone or low potassium levels on KCNK9 expression. Transcriptome profiling of 43 APA and 11 control adrenals revealed a slight, but significantly increased expression of KCNK3 in adenomas compared to controls that correlated positively with CYP11B2 expression. The quantitative changes of TASK1 expression observed in APAs may be secondary to a primary epigenetic phenomenon or be secondary to increased aldosterone production due to dysregulation of master transcription factors or upstream signaling cascades in the aldosterone biosynthetic pathway.