Skewed X-chromosome inactivation causing diagnostic misinterpretation in congenital nephrogenic diabetes insipidus.

OBJECTIVE

To establish the clinical phenotype and genetic background in a family with diabetes insipidus.

MATERIAL AND METHODS

The subjects were a sister and brother, aged 34 and 27 years, respectively, with a history of polyuria since infancy. Clinical testing confirmed a diagnosis of congenital nephrogenic diabetes insipidus (CNDI) in both. Samples of purified genomic DNA were analysed.

RESULTS

The sequence of the entire coding region of the AQP2 gene as well as the AVPR2 gene was determined. Sequence analysis revealed no variations in the AQP2 gene. A missense variation in exon 2 of the AVPR2 gene (g.685G>A), predicting a p.Asp85Asn substitution, was identified in the X-chromosome of the affected male and one allele in the sister and the asymptomatic mother. The p.Asp85Asn variation in AVPR2 is known to cause CNDI, and has previously been described as inducing a partial phenotype treatable with dDAVP. However, in this family dDAVP had no influence on urine osmolality, whereas combination therapy with indomethacin and hydrochlorothiazide increased urine osmolality to 299 mosm/l in the proband. A skewed X-inactivation pattern (93%) occurring in the normal X allele was recognized in the sister.

CONCLUSIONS

This study demonstrates the effect of skewed X-chromosome inactivation associated with X-linked CNDI. Polydipsia in early childhood could be due to X-linked CNDI despite affecting both genders. The significant heterogeneity in the clinical phenotype in CNDI carries a risk of diagnostic misinterpretation and emphasizes the need for genetic characterization. Treatment combining indomethacin and hydrochlorothiazide results in a marked response on both urine output and urine osmolality.