Diabetes Insipidus, Nephrogenic, X-Linked

A number sign (#) is used with this entry because of evidence that X-linked nephrogenic diabetes insipidus is caused by mutation in the gene encoding the vasopressin V2 receptor (AVPR2; 300538) on chromosome Xq28.

Description

Nephrogenic diabetes insipidus (NDI) is caused by the inability of the renal collecting ducts to absorb water in response to antidiuretic hormone (ADH), also known as arginine vasopressin (AVP; 192340). Approximately 90% of patients are males with the X-linked recessive form (type I), which is caused by a defect in the vasopressin V2 receptor in renal collecting duct cells. The remaining 10% of patients have autosomal NDI (125800) (type II), which is caused by mutations in the gene encoding the aquaporin-2 water channel (AQP2; 107777) on chromosome 12q13 (Morello and Bichet, 2001).

Neurogenic, or central, diabetes insipidus (125700) is caused by mutation in the gene encoding arginine vasopressin, located on 20p13.

Clinical Features

In a Mormon family traced to 1813, Cannon (1955) reported 3 instances of male-to-male transmission of diabetes insipidus. However, he noted reduced penetrance in females, as carriers did not show the disorder. Thus was raised the suspicion that the disorder in this family was actually X-linked. Cutler et al. (1955) proved the renal basis of the disorder in this family. Ten Bensel and Peters (1970) described hydronephrosis in affected male sibs of the family reported by Cannon (1955). They determined that the pedigree, covering 5 generations with 12 affected males, was typical of X linkage.

Nakano (1969) described familial nephrogenic diabetes insipidus in 4 generations of a Samoan family.

Van Lieburg et al. (1999) made a retrospective analysis of clinical data from 30 male nephrogenic diabetes insipidus patients, aged 1 month to 40 years, from 18 Dutch families. In 28 patients, 17 different mutations in the AVPR2 gene were found. Two patients had mutations in the AQP2 gene. Eighty-seven percent of the patients were diagnosed within the first 2.5 years of life. Main symptoms at clinical presentation were vomiting and anorexia, failure to thrive, fever, and constipation. Most patients were on hydrochlorothiazide-amiloride treatment without significant side effects. Two patients suffered from severe hydronephrosis with a small rupture of the urinary tract after a minor trauma, and 2 patients experienced episodes of acute urine retention. Height SD scores for age remained below the 50th percentile in the majority of patients, whereas weight for height SD scores showed catch-up after several years of being underweight. The majority of patients were found to have normal intelligence; this was in contrast to the belief that mental retardation is the most frequent long-term sequela of NDI. Except for a possibly milder phenotype in patients with an AVPR2 G185C mutation (300538.0003), no clear relationship between clinical and genetic data could be found.

Other Features

Bell et al. (1974) found that a subset of patients with NDI did not show increased urinary levels of cyclic AMP (cAMP) in response to ADH administration, indicating that the defect was proximal to the adenylate cyclase step. In contrast, patients with type II NDI showed normal increased urinary levels of cAMP in response to ADH, indicating that the defect in those cases was distal to the adenylate cyclase step (Zimmerman and Green, 1975).

In normal persons, 1-desamino-8D-arginine vasopressin (dDAVP), a synthetic vasopressin analog, stimulates the release of von Willebrand factor (VWF; 613160) from vascular epithelium and factor VIII (F8; 300841) from liver and other unidentified sites. Kobrinsky et al. (1985) found that the factor VIII and VWF responses to dDAVP were absent in NDI patients and about 50% of normal in carriers. The authors concluded that the vasopressin receptor defect is not confined to the kidney, and suggested that a decreased factor VIII response may be a useful carrier test.

Bichet et al. (1988) found that patients with X-linked NDI had no response to dDAVP infusion compared to healthy controls and patients with central diabetes insipidus, as measured by mean arterial pressure, pulse rate, plasma renin activity, and release of factor VIII and VWF. Obligatory carriers had minimal response. They hypothesized the existence of an extrarenal vasopressin V2-like receptor, which may also be defective in these patients. In patients with NDI, Bichet et al. (1989) found that plasma cAMP did not increase in response to dDAVP, suggesting a pre-cAMP V2 receptor defect.

Clinical Management

Based on previous observations suggesting that chlorothiazide acts in the distal portion of the nephron where it may inhibit the production of 'free water' that normally results from selective reabsorption of sodium and accompanying anions, Crawford and Kennedy (1959) used the drug in the treatment of rats in which diabetes insipidus was produced by electrolytic damage of the hypothalamus. The result of the treatment was a reduction in the spontaneous intake of water by 50% or more. They then treated 2 human subjects, one with central diabetes insipidus due to lack of vasopressin, the other with nephrogenic diabetes insipidus. With dose levels in the same range as that used for diuretic purposes, the urine volume was reduced dramatically with rise in urine osmolality.

Alon and Chan (1985) found that treatment of nephrogenic diabetes insipidus with a combination of amiloride, a potassium-sparing diuretic, and hydrochlorothiazide not only had a potentiating effect, but also helped prevent urinary potassium loss, hypokalemia, and alkalosis. They found the hydrochlorothiazide-amiloride regimen superior to hydrochlorothiazide alone and to the combination of hydrochlorothiazide and a prostaglandin synthetase inhibitor.

Libber et al. (1986) found that among the inhibitors of prostaglandin synthesis, indomethacin was much more effective than ibuprofen in treating NDI.

Mapping

Bode and Miettinen (1970) excluded close linkage of NDI with the Xg blood group.

In 5 NDI families, Knoers et al. (1987) found no recombination with DXS52, which maps to chromosome Xq28 (lod = 3.47; theta = 0.00). In 10 affected families, Knoers et al. (1988) found no crossovers between NDI and 4 markers, giving maximum lod scores of 3.23 (DXS15), 10.35 (DXS52), 2.19 (F8), and 2.09 (DXS134). Knoers et al. (1989) further corroborated the assignment of NDI to Xq28 by the study of 11 families, 10 of which had previously been described, and by the testing with 2 additional markers closely linked to DXS52. Kambouris et al. (1988) found a maximum lod score of 3.31 at theta = 0 for multipoint linkage with a factor VIII probe and DXS15.

By screening several hybrid cell lines containing different parts of human chromosomes, van den Ouweland et al. (1991) demonstrated a positive correlation between the presence of the distal part of the human X chromosome and the expression of the vasopressin renal-type V2 receptor. By the use of the V2 receptor-specific agonist dVSAVP, they demonstrated that the vasopressin binding activity of the hybrid cells was dependent on the V2-type receptor. Furthermore, the V2/V1 antagonist was able to completely inhibit the induction of vasopressin by AVP.

Bichet et al. (1992) found that carrier status of X-linked NDI could be predicted in 24 of 26 at-risk females by use of linked markers.

Molecular Genetics

In 2 unrelated patients with X-linked NDI, Rosenthal et al. (1992) identified 2 different mutations in the AVPR2 gene (300538.0001; 300538.0002). One of the patients was a 37-year-old man with a lifelong history of polyuria, polydipsia, and mental retardation resulting from repeated and prolonged episodes of dehydration in the first year of life.

In 3 unrelated patients with X-linked nephrogenic diabetes insipidus, van den Ouweland et al. (1992) identified 3 different mutations in the AVPR2 gene (300538.0003-300538.0005). All of the mutations occurred in a highly conserved extracellular domain.

Bichet (1994) indicated that a total of 30 different mutations had been identified in the AVPR2 gene, distributed among 37 families. About half of these represented missense mutations. He emphasized the usefulness of early detection of the mutation in the neonate to prevent the adverse effects of dehydration.

In 5 Arab families with nephrogenic diabetes insipidus, Carroll et al. (2006) identified 1 novel and 1 previously reported missense mutation in the AVPR2 gene as well as a novel contiguous gene deletion involving AVPR2.

Skewed X Inactivation

The possibility of skewed X inactivation was invoked by van Lieburg et al. (1995), who described 3 NDI families in which females showed clinical features resembling those of males. DNA analysis indicated that each was heterozygous for a specific AVPR2 mutation, as were also 2 asymptomatic female family members. The authors concluded that, in female NDI patients, the possibility of heterozygosity for an AVPR2 gene mutation has to be considered in addition to homozygosity for mutations in the AQP2 gene located on chromosome 12.

In 6 members of a Japanese family with X-linked NDI, Nomura et al. (1997) identified a mutation in the AVPR2 gene (300538.0013). Three heterozygous females had differences in clinical severity of NDI. The X-inactivation patterns of these females were investigated by studying the methylated trinucleotide repeat in the human androgen receptor gene. The grandmother showed extremely skewed methylation of one X chromosome, while her daughter had moderately skewed methylation. The daughter of the grandmother's sister, who had no symptoms of NDI, showed random methylation. Nomura et al. (1997) suggested that the highly skewed X-inactivation pattern of the grandmother indicated that her NDI phenotype was caused by dominant methylation of the normal allele of the AVPR2 gene.

Carroll et al. (2006) described evidence of skewed X inactivation associated with a novel deletion in the AVPR2 gene.

Population Genetics

Bode and Crawford (1969) and Bode and Miettinen (1970) proposed that patients with nephrogenic diabetes insipidus in eastern North America shared a common ancestor, an Ulster Scot who had arrived in Halifax in 1761 on the ship Hopewell. They also suggested a link between this family and the large Mormon pedigree reported by Cannon (1955). Bode and Crawford (1969) stated that 'it is likely that the Hopewell hypothesis can never be proved.' However, Bichet et al. (1992) used haplotype analysis to show that there was not likely to be a connection between the family reported by Cannon (1955) and the Hopewell kindred. Genealogical studies also seemed to exclude a connection. Bichet et al. (1992) also studied 11 other affected families of diverse ethnic backgrounds: 5 French Canadian, 1 African American, 1 Puerto Rican, 1 Iranian, 2 French, and 1 English. All of the families had similar phenotypic features and showed linkage to markers in the Xq28 region. Differences in RFLP patterns indicated that the 'Hopewell hypothesis' could not explain the origin of NDI in many of the North American families.

Holtzman et al. (1993) identified the AVPR2 mutation in the Hopewell family (W71X; 300358.0008). This finding and data on various other mutations in the AVPR2 gene found in North American pedigrees, both of Ulster Scot ancestry and others, made the founder effect as proposed in the Hopewell hypothesis invalid. Bichet et al. (1993) likewise presented evidence for multiple mutations in the AVPR2 gene, arguing against the Hopewell founder hypothesis.

History

Sawin (1998) reprinted the lecture by Armand Trousseau (1862) on polydipsia, specifically what he called nonsaccharine diabetes and what we now call diabetes insipidus. Trousseau (1801-1867) was a famous Paris physician and professor.