Authors: Alessandra Boletta (corresponding author) 
ADPKD: genetics and proposed mechanisms of cystogenesis
Autosomal dominant polycystic kidney disease (ADPKD) is a frequent genetic disease, affecting between 1:500 and 1:1000 of the general population [1, 2, 3]. The hallmark of the disease is bilateral renal cyst formation, although a plethora of extra-renal manifestations have been reported. These include liver and pancreatic cysts as well as cardiovascular complications (for review of the clinical aspects of the disease, see ). Two genes have been linked to this disease: PKD1 , which is mutated in 85% of cases, and PKD2 , which is mutated in most of the remaining cases. Numerous mutations have been reported along the length of these two genes; most are loss-of-function mutations predicted to inactivate the affected allele [1, 2, 3].
Genetic mechanism of cystogenesis
ADPKD is inherited in an autosomal dominant manner, but it has been suggested that it is recessive at the molecular level . Analysis of the epithelia derived from renal or liver cysts has revealed that the normal inherited allele undergoes a somatic mutation, resulting in homozygous loss of either the PKD1 or PKD2 gene [4, 5]. This 'second hit' causes a single affected epithelial cell to over-grow and generate a clonal outpouching diverticulum, which eventually disconnects from the renal tubule. In some of the cysts in which homozygous inactivation of one PKD gene (either 1 or 2) is not observed, mutations in the other PKD gene have been found, suggesting that trans-heterozygosity of the two genes is sufficient to induce cyst formation . Based on these findings, a 'threshold model' of cystogenesis has been proposed, in which heterozygosity is not sufficient to induce renal cystogenesis, but a minimum level of activity of the PKD1 and PKD2 gene products, the polycystins, is required to prevent cystogenesis. Any dysfunction causing polycystin activity to fall below such a threshold would lead to cyst formation, even though a complete loss of function does not appear to be necessary. Several findings from studies in animal models appear to support this model: (i) heterozygosity of the Pkd1 or Pkd2 gene is not sufficient to cause renal cyst formation in mice, possibly due to a low rate of second hits in the murine kidney, whereas homozygous inactivation causes renal cystogenesis in utero [7, 8]; (ii) conditional inactivation of the second allele in the kidney results in renal cystogenesis with a variably severe phenotype depending on the time of inactivation [9, 10]; (iii) a mouse model carrying an unstable Pkd2 allele that spontaneously undergoes somatic inactivation displays the ADPKD phenotype when crossed with a Pkd2 -null mouse ; (iv) trans-heterozygous Pkd1 +/- :Pkd2 +/- mice display markedly more severe cystic kidney disease than single heterozygous mutants ; and (v) a mouse model carrying an aberrantly spliced variant of the Pkd1 gene, lowering its expression to 13-20% of normal levels, suffers renal cystogenesis .
Intriguingly, a recent study has identified ADPKD families carrying homozygous inheritance of mild mutations in the PKD1 gene, which resulted in PKD; this finding...