| |||||||||||||||||||||||
Congenital nephrogenic diabetes insipidus is a rare hereditary disorder in which renal responsiveness to arginine vasopressin is impaired. In some families, the disorder is associated with the inability of the hormone to stimulate cAMP production2. The disorder is characterized by the excretion of urine that is hypotonic as compared with plasma, even when plasma osmolality and plasma vasopressin concentrations are increased3. In affected persons, this renal abnormality is manifested soon after birth as repeated episodes of severe dehydration and hypernatremia that may result in damage to the central nervous system4. The clinical disorder occurs primarily in males, and the description of a subclinical as well as a very rare clinical form in carrier females suggests that it is an X-linked trait5. Linkage studies have demonstrated that the gene responsible for congenital nephrogenic diabetes insipidus is located on the long arm of the X chromosome in region 28 (Xq28)6,7 and that the gene for the vasopressin V2 receptor is also localized to a region on the X chromosome between Xq27 and Xq288.
Twenty-four years ago Bode and Crawford9 proposed that most North American patients with congenital nephrogenic diabetes insipidus were descendants of Ulster Scot immigrants who settled in Nova Scotia and northern New England in the mid-1700s, initially arriving on the ship Hopewell. We have studied 40 members of a large Ulster Scot family described by Bode and Miettinen in 197010; in 22 of these subjects we determined the structure of the gene for the vasopressin V2 receptor. We report a novel defect of the vasopressin V2 receptor, found only in affected male subjects and obligate carrier females and shown by linkage analysis to be the mutation responsible for congenital nephrogenic diabetes insipidus in this family. These results, coupled with the recent demonstration of different defects in the vasopressin V2-receptor gene in other North American families with the disorder,11,12 rule out the original Hopewell hypothesis.
Methods
Subjects
We studied a large family (Figure 1) with congenital nephrogenic diabetes insipidus, some members of which were evaluated 24 years ago at the Pediatric Endocrinology Unit of the Massachusetts General Hospital (Boston). The study protocol was approved by the institutional review board of Children's Hospital (Boston), and informed consent was obtained from all study subjects.
|
Genomic DNA was isolated from peripheral-blood samples obtained from 22 members of the family, according to standard techniques13. Initial screening for mutations was carried out by sequencing subcloned products of the polymerase chain reaction (PCR) from the genomic DNA of two affected family members (Subject II-10 and Subject II-11) and one normal member (Subject II-7) as described below. DNA sequences that deviated from the reported sequence of the human vasopressin V2 receptor14 were confirmed by the sequencing of at least two additional independent clones. After the mutation in these family members was characterized, 15 other members were screened by direct sequencing of PCR products amplified from genomic DNA15. We also analyzed the DNA of 21 members by PCR amplification of a 3'-mismatched primer16.
PCR Amplification and Subcloning
Four sets of oligonucleotide primers were used to amplify overlapping segments of the entire coding sequence of the vasopressin V2-receptor gene. Each reaction mixture contained 1 µg of genomic DNA, 100 pmol of each primer, and solutions containing standard buffer and deoxynucleotide triphosphates. The PCR was carried out at 95 °C for five minutes and then for 28 cycles at the following temperatures: 94 °C for one minute, 62 °C for two minutes, and 72 °C for two minutes, with a final extension of 72 °C for seven minutes. Purified and subcloned PCR products were sequenced according to standard methods13.
PCR-Based Mutation Analysis
A recently developed approach to the identification of mutations -- amplification refractory mutation16 -- was used in the analysis of all affected and normal subjects in the study. This is a general method for the analysis of point mutations or deletions; it is based on the mismatch at the 3' hydroxyl residue in the PCR primer, which does not function properly in PCR reactions under the appropriate conditions. We designed two oligonucleotide pairs to amplify either the wild-type DNA sequence or the mutant sequence and used them in PCR reactions. The wild-type primers were (forward) 5'GGGCCAGATGCCCTGTGTC3' and (reverse) 5'CATTCTAGATCACGATGAAGTGTCCTTGGCCAGGGA3'; the mutant primers were (forward) 5'GGGCCAGATGCCCTGTGTT3' and (reverse) 5'CATTCTAGATCACGATGAAGTGTCCTTGGCCAGGGA3'. The PCR conditions were the same as described above except that the annealing temperature was 64 °C.
Direct PCR Sequencing
To determine the genotype of each affected subject and verify the results of the PCR-based amplification-refractory-mutation analysis, direct sequencing of PCR products was performed for each of the family members indicated in Figure 1. After 1 µg of genomic DNA was amplified with the PCR, the resulting nucleic acid was extracted with phenol-chloroform-isoamyl alcohol, concentrated, and purified. The purified PCR product was used as a template for dideoxy-DNA sequencing with internal sequencing primers.
Linkage Analysis
The genotypes of the 22 subjects whose DNA was sequenced or determined by PCR (or both) were analyzed for genetic linkage to congenital nephrogenic diabetes insipidus with use of the LIPIN and LIPED programs,17,18 under the assumption that the trait was X-linked and recessive, with complete penetrance and a 1 percent frequency of the affected allele in a normal population.
Results
Figure 1 shows three generations of a family with congenital nephrogenic diabetes insipidus; most members are longtime residents of a town in central Maine. This family was described10 as descendants of an Ulster Scot clan that originally settled in Nova Scotia in the mid-1700s. The renal function and urinary concentrating capacity of three second-generation male subjects -- Subject II-9 (now 37 years old), Subject II-10 (35 years old), and Subject II-11 (33 years old) -- had been determined when they were 15 years old, 7 years old, and 9 months old, respectively. All three subjects had severe polydipsia and polyuria (urine volume greater than 4 liters per square meter of body-surface area per 24 hours), and the specific gravity of their urine was less than 1.004, even after they received vasopressin (1 U of Pitressin tannate in oil, by intramuscular injection). All other tests of renal function were normal for age. Subject II-14 died at the age of two years of "underdeveloped kidneys" after repeated episodes of dehydration and hyperthermia, and is presumed to have had congenital nephrogenic diabetes insipidus. Previous reports have emphasized the vulnerability of infants with this disorder to episodes of dehydration, hypernatremia, hyperthermia, seizures, and death3,4.
The three affected second-generation family members have continued to have polydipsia and polyuria as adults. On evaluation in 1992, Subjects II-9 and II-10 drank 6.5 and 5.5 liters of water per square meter per 24 hours, respectively, and both drank approximately 2 liters of water each night. End-stage renal disease of unknown cause developed in Subject II-11 in 1988, for which he received a cadaveric renal transplant in 1991. However, the finding of an enlarged urinary bladder and megaureters suggests that his renal failure was caused by functional obstruction due to his large urinary output.
The male family members in the third generation were affected in the same way as those in the second generation. For example, Subject III-7 (30 years old) drank more than 7 liters of water per square meter per 24 hours. The one female member found to have symptoms (Subject III-10, 29 years old) drank 3.5 liters of water per square meter per 24 hours and awakened two to three times each night to drink water. On the basis of clinical information only, we suspected that she was also a carrier with partial expression of the defect for congenital nephrogenic diabetes insipidus because of nonrandom X-chromosome inactivation6. These results provide convincing evidence of X-linked inheritance of the gene for the disorder in this kindred. The unaffected family members were clinically normal.
DNA Sequencing and PCR Analysis of the Mutation
The affected subjects had a cytosine-to-thymine transition (C-to-T) that resulted in a missense mutation, CGG to TGG, which caused a tryptophan to be substituted for an arginine in the expressed protein. This mutation was found in all five affected male family members studied (Subjects II-9, II-10, II-11, III-5, and III-8) (Figure 2). Three obligate heterozygotes (Subjects I-2, II-2, and II-3) were genotypically heterozygous (had both a cytosine and a thymine) at this site. Subject III-10 had symptoms of congenital nephrogenic diabetes insipidus and was heterozygous at this position. Thirteen normal family members had wild-type DNA sequences at this position14. The genotypes of 22 subjects were determined through DNA sequencing (18 subjects) or PCR analysis (21 subjects); for the 17 subjects whose genotypes were determined with both methods, the results were in 100 percent agreement.
|
Linkage analysis18 of the 22 genotypes yielded a lod score of 3.00 with a theta of 0.
Discussion
This family was initially evaluated 23 years ago as a large pedigree with congenital nephrogenic diabetes insipidus10. We have now identified the molecular nature of the genetic defect in the vasopressin V2 receptor that is consistent with the production of a dysfunctional vasopressin receptor in this pedigree. In addition, because we studied a large number of affected subjects and carriers, we were able to demonstrate linkage of this point mutation with congenital nephrogenic diabetes insipidus.
The mutation in the vasopressin V2-receptor gene of affected subjects is located one amino acid toward the carboxyl terminus from a conserved cysteine residue (cysteine-112); it results in the substitution of a tryptophan residue for the arginine in the normal sequence14. This cysteine residue is proposed to be involved in an essential extracellular disulfide bond in G protein-coupled receptors19. The visual pigment rhodopsin is a G protein-coupled receptor that binds 11-cis retinal as a ligand and is activated by light. Mutations in the rhodopsin gene similar to the mutation in the V2-receptor gene in this family cause rhodopsin to be abnormally retained in the endoplasmic reticulum and Golgi complex19. We have not yet demonstrated the functional consequences of the substitution of tryptophan for arginine in this pedigree. However, we speculate that a change from a positively charged arginine residue to an amino acid with a large side chain with both hydrophobic and hydrophilic characteristics, such as tryptophan, might interfere with the formation of disulfide bonds, thereby rendering the receptor dysfunctional because it remains trapped in intracellular organelles. The formation of disulfide bonds between the first and second extracellular loops is a repeated theme in all G protein-coupled receptors like the vasopressin V2 receptor, and tryptophan is not found adjacent to the cysteines responsible for disulfide-bond formation in any such receptor (unpublished data). To date, mutations in only one other class of G protein-coupled receptors (opsins -- e.g., human rhodopsin) have been demonstrated to cause disease in humans. All are specific degenerative diseases of the retina (autosomal dominant retinitis pigmentosa, autosomal recessive retinitis pigmentosa, and color blindness)20,21.
Other defects in the vasopressin V2-receptor gene have been found in patients with congenital nephrogenic diabetes insipidus in several North American pedigrees11,12. These results now rule out the founder effect of the Hopewell hypothesis. In addition, these results and those associated with various retinopathies involving opsins suggest that the structure of G protein-coupled receptors is susceptible to numerous point mutations resulting in clinically observable defects.
Supported by grants (DK-19406 and DK-38874) from the National Institutes of Health.
We are indebted to Drs. Hans Bode, John D. Crawford, and John T. Herrin (Pediatric Nephrology/Endocrine Service, Massachusetts General Hospital) for information about the study family, to Ms. Elena Fiamma for assistance in the preparation of the manuscript, and to Ms. Andrea I. McClatchey and Dr. James Trofatter (Molecular Neurogenetics Laboratory, Massachusetts General Hospital) for helpful suggestions and calculation of the lod scores.
Source Information
From the Renal Unit, Massachusetts General Hospital, and the Department of Medicine, Harvard Medical School, Boston (E.J.H., L.F.K., D.A.A.), and the Division of Nephrology, Children's Hospital, and the Department of Pediatrics, Harvard Medical School, Boston (H.W.H., L.M.G.-W., B.B.).
Address reprint requests to Dr. Harris at the Department of Pediatrics, Harvard Medical School, 300 Longwood Ave., Boston, MA 02115.
References
| |||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
