Main

A recent molecular genetic approach has demonstrated that congenital NDI is caused by at least the defects of two different genes, that is, the AVPR2 in X-linked NDI and the vasopressin-sensitive water channel gene (aquaporin-2, AQP2) in autosomal recessive NDI(1, 2). The gene coding for AVPR2 was cloned in X chromosome(3, 4), and the deduced amino acid sequence showed that the receptor contains seven putative transmembrane domains, and this receptor is a member of a large family of G protein-coupled receptors(3, 4). Since then, more than 63 putative disease-causing mutations in AVPR2 have been identified in 90 presumably unrelated families with X-linked NDI from diverse ethnic groups(5–22). Furthermore, study of the expression of several mutant receptors provided biochemical proof of receptor inactivation(13, 20, 21).

In this presentation, we analyzed the AVPR2 in two sibling cases and one sporadic case with X-linked NDI. We identified three novel mutations: a deletion of one out of three G nucleotides (nucleotide 786-788, 786delG) in kindred #1 and two missense mutations (Pro to Ser, P322S and Pro to His, P322H) at codon 322 in the seventh transmembrane in kindred #2 and another sporadic case.

METHODS

Subjects. Four male patients from two unrelated Japanese families (Fig. 1, kindred #1 and #2 family) and one sporadic male patient (patient #3) were studied. The clinical symptoms and relevant laboratory tests of these subjects are shown inTable 1.

Figure 1
figure 1

Two Japanese kindreds with X-linked NDI. (A) Kindred #1; (B) kindred #2; closed square; affected male patient, half-closed circle; obligate carrier. Those whose samples were analyzed by PCR-SSCP and direct sequencing of PCR products are denoted byasterisks.

Full size image
Table 1 Characterization of subjects with X-linked NDI
Full size table

Kindred #1. Two siblings (#1-II-1 and #1-II-2) with a history of recurrent fever, vomiting, and poor weight gain, which began 5 mo after birth, were diagnosed as having X-linked NDI. Their basal AVP levels were high, and they showed the impaired responses of urinary osmolality to exogenously administrated AVP (Table 1). They were treated with a thiazide diuretic and indomethacin. Their motor milestones had been normal. Molecular analysis was carried out in subjects #1-II-1, #1-II-2, and #1-I-2.

Kindred #2. Vomiting and fever manifested 1 mo after birth in subject #2-III-1. He was strongly dehydrated and found to be hypernatremic(161 mEq/L) with high plasma osmolality (333 mosmol/kg). He was diagnosed as having X-linked NDI based upon the high basal plasma AVP value and the unresponsiveness to exogenous AVP or desmopressin (Table 1). The mother (#2-II-2) subsequently gave birth to a second male child. At the age of 2 wk, the affected boy (#2-III-2) showed failure to gain weight. His serum osmolality was found to be 290 mosmol/kg, and urine osmolality was only 64 mosmol/kg. He was also diagnosed as having X-linked NDI due to the high basal plasma AVP value and unresponsiveness to exogenous AVP or desmopressin (Table 1). Subjects #2-III-1 and #2-III-2 were treated with thiazide diuretics and indomethacin, but their urine volume increased up to 3000 mL/day at the age of 1 y.

After water deprivation of subject #2-III-1, the osmolality of urine was 105 mosmol/kg (normal range, 666-824 mosmol/kg). His mother (#2-II-2) had partial renal resistance to vasopressin because of a low response of urinary osmolality (475 mosmol/kg: normal range, 666-824 mosmol/kg) to AVP. After a water deprivation test, the mother (#2-II-2) revealed a low ability of concentrating water, but the grandparents (#2-I-1 and #2-I-2), father(#2-II-1), and uncle (#2-II-3) showed normal responses after water a deprivation test. Subjects #2-III-1, #2-III-2, and #2-II-2 were further analyzed in this study.

Sporadic subject (patient #3). This subject was a 2-y-old boy who had a history of polyuria and polydipsia. He did not reveal any developmental delay and failure to thrive. His daily urinary volume showed approximately 3000 mL, and his basal AVP was elevated (Table 1). He was diagnosed as having X-linked NDI and treated with a thiazide and spironolactone. With this treatment, urine volume was reduced moderately.

PCR-SSCP analysis of V2 receptor gene. Genomic DNA was extracted from leukocytes according to standard procedures. We selected primers amplifying the V2 receptor gene based upon the sequences of the human V2 receptor vasopressin gene (Fig. 2 andTable 2)(4, 12). For the first PCR reaction followed by SSCP analysis, the following primer pairs were used, AP1-2, AP3-4, AP5-6, AP7-8, and AP9-10 (Fig. 2). Thirty cycles of PCR for PCR-SSCP analysis using primer pairs AP1-2, AP3-4, AP5-6, AP7-8, and AP9-10 at 94°C for 60 s for denaturation, 55°C for 60 s for annealing, and 72°C for 120 s for polymerization were run, with 100 ng of genomic DNA, 100 μM each of dNTPs, 0.5 mM each pair of primers, 1.5 mM MgCl2, 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 0.25 U of Taq polymerase, and 1μL of [α-32P] dCTP (3000Ci/mmol, 10 mCi/mL) in a volume of 20μL. The reaction mixture (1 μL) was added to the stop solution (9 μL)(23, 24). Then this solution (2 μL) was heated in boiling water for 5 min for electrophoresis on 5% polyacrylamide-TBE gels containing 5% glycerol. Gel electrophoresis was performed at 30 W for 3 h at room temperature by cooling with fans. The gels were dried on Whatman 3MM paper. Autoradiography was performed for 72 to 96 h at -30°C(23, 24).

Figure 2
figure 2

Structural organization of the vasopressin V2 receptor gene. The primers used for PCR are shown by arrows. DNA fragments amplified by the primer pairs AP1-2, 3-4, 5-6, 7-8, and 9-10 were subjected to PCR-SSCP analysis. PCR products from AP1-6 and 5-8 were directly sequenced by using the appropriate internal primer as described in “Methods.” The point mutations detected in our study are shown.

Full size image

Direct sequencing of V2 receptor gene. For direct sequencing to detect mutations responsible for different mobilities of PCR-SSCP analysis, genomic samples were amplified using primer pairs AP1-6 and AP5-8(Fig. 2 andTable 2). Amplification using primer pairs AP1-AP6 and AP5-AP8 was carried out for 35 cycles with each cycle of incubation consisting of 60 s at 94°C, for denaturation, 120 s at 55°C for annealing, and then 180 s at 72°C for extension. These amplified products were purified on 1% agarose gel and subjected to direct sequencing by the chain termination method using appropriate sequencing primers and Sequenase (U.S. Biochemical Corp., Cleveland, OH) as described previously(24).

RESULTS

PCR-SSCP analysis and direct sequencing. In all subjects whom we were able to analyze, V2 receptor gene using primer pairs AP1-2, 3-4, 5-6, 7-8, and 9-10 was successfully amplified to expected size. Thus, the V2 receptor gene from these patients did not have any gross deletions. To investigate mutations in V2 receptor genes of each subject, we performed PCR-SSCP analysis by primer pairs AP1-2, 3-4, 5-6, 7-8, and 9-10.Figure 3A shows the results of PCR-SSCP analysis using primer pairs AP5-6. The band of #1-II-1 (lane 2) migrated differently from the others. To determine the nucleotide changes responsible for the altered mobility band, we sequenced the concerned region in #1-II-1. Direct sequencing analysis revealed a deletion of one out of three guanosines (nucleotide 786-788, 786delG) at codon 239 (Fig. 4A). This deletion changes codon 239 from GGG to GGC, shifts the reading frame with an altered amino acid sequence beginning with codon 240, and introduces a premature stop codon (TGA) at codon 270. Thus, this stop codon produced the truncated mutant receptor lacking the carboxy terminus. In Figure 3B, the results of PCR-SSCP analysis using primer pairs AP7-8 is shown. The electrophoretic mobilities of #2-III-1 (lane 3) and patient #3 (lane 1) were different from others. Amplification and direct sequencing of exon 4 in#2-III-1 revealed single nucleotide change at codon 322 in the seventh transmembrane region: CCC to TCC, accompanying amino acid substitution Pro for Ser (P322S) (Fig. 4B). Patient #3 was identified with a point mutation that changes the same Pro residue at codon 322, but the amino acid substitution was different: Pro (CCC) to His (CAC) (P322H). To determine whether these mutations at codon 322 are not merely polymorphisms, 50 unrelated normal Japanese individuals were tested for these changes by PCR-SSCP analysis. The mutations were not present in any of these control samples (data not shown). In all patients, no other mutations were found in the remainder of the protein coding region as well as the intron-exon boundary.

Figure 3
figure 3

(A) PCR-SSCP analysis using primer pairs AP5-6. C, Control; lane 1, patient #3; lane 2, II-1 in kindred #1; lane 3, III-1 in kindred #2. The mobility of band from patient #1-II-1 (lane 2) can be distinguished from the others. (B) PCR-SSCP analysis using primer pairs AP7-8. C, Control; lane 1, patient #3; lane 2, #1-II-1 in kindred #1; lane 3, #2-III-1 in kindred #2. Patient #3 (lane 1) and#2-III-1 (lane 3) show different migrating bands from the normal control.

Full size image
Figure 4
figure 4

Direct sequence analysis. (A) In kindred #1, one out of three consecutive guanosine residues between nucleotides 786 and 788 (786delG) was deleted in proband II-1. According to the deletion mutation, a new reading frame starts at codon 240 (italic). (B) The results of direct sequence of kindred #2 and patient #3. Control, normal sequence. In subject III-1 from kindred #2, a C to T substitution was found at the first base of codon 322, resulting in an amino acid change from Pro (CCC) to Ser (TCC) (P322S). Patient #3 had a C to A transition at the second base of codon 322. This mutation changes a Pro (CCC) to a His (CAC) (P322H).

Full size image

Pedigree analysis of kindred #1 and #2. In kindred #1, the affected male subjects (#1-II-1 and #1-II-2, lanes 2 and 3) showed an SSCP variant in which two bands migrated lower (Fig. 5A), representing both strands from the mutant allele, whereas their mother showed four bands (Fig. 5A, #1-I-2, lane 1), implying the presence of the normal and mutant allele. The mother of kindred 2 (#2-II-2, lane 1) showed a heterozygote at the mutation site (Fig. 5B), and the affected members (#2-III-1 and #2-III-2, lanes 2 and 3) are hemizygous for this mutation (Fig. 5B). The PCR products of their mother and affected family members were subjected to direct sequencing, and the presence of the mutations was confirmed (data not shown).

Figure 5
figure 5

(A) PCR-SSCP analysis in kindred #1. PCR-SSCP screening demonstrated that II-1 and II-2 were hemizygous for this mutation. The mother of I-2 showed a heterozygous pattern for the mutation.(B) PCR-SSCP screening in kindred #2. Subjects III-1 and III-2 were hemizygous for this mutation. In this family, the mother of II-2 was heterozygous.

Full size image

DISCUSSION

Three novel mutations of the V2 vasopressin receptor gene were identified in two Japanese unrelated kindreds and one sporadic case with X-linked NDI. Up to the present, mutations in the V2 vasopressin receptor gene identified in X-linked NDI in North American, European, and Japanese patients were heterogenous(5–22).

In kindred #1, the deletion mutation in one out of three guanosine(786delG), resulted in a changed reading frame and induced a premature stop codon at position 270. This one-base deletion introduced the termination codon to the same position as that described by Rosenthal et al.(5). Thus, the mutant receptor lacks the entire carboxy-terminal third and conserves only 17 of the 42 amino acids of the third intracellular domain. The missing third intracellular domain and carboxy terminus are determinants for G protein recognition(5). This strongly indicates that this mutation causes X-linked NDI in kindred#1.

In kindred #2 and patient #3, both of the two nucleotide substitutions at codon 322 in the seventh transmembrane resulted in the loss of a proline residue, which was replaced by Ser (P322S) in kindred #2 and by His (P322H) in patient #3. Although the expression study of mutant receptors is definitely required to prove these mutations as a cause of X-linked NDI, the mutant receptors would not be able to transduce a signal. Pro at codon 322, which is replaced by a Ser or His, is a well conserved residue among rat V1 and V2 vasopressin receptors, the human oxytocin receptor, and other G protein-coupled receptors(25–29). The study of the expression of the mutation of P286R of the AVPR2 detected in one X-linked NDI individual showed an absence of vasopressin-binding ability and resulted in a failure to stimulate the adenylyl cyclase system. This indicates the importance of conserved proline residues of AVPR2(20). Furthermore, in an in vitro study of β-adrenergic receptor, the important role of this conserved Pro was identified(29). According to this report, the mutant receptor of substituting the Pro at codon 323 with Ser, which corresponds with Pro at codon 322 in the V2 receptor, resulted in an improper or incomplete processingβ-adrenergic receptor, presumably reflecting a role for this residue in the folding of the receptor(29). For this mutant, the expected molecular mass was not observed, two polypeptides of different molecular masses were detected, and this mutant was unable to bind ligand(29). These results indicate that the mutant receptors of AVPR2 derived from P322S and P322H may have lowered affinity for ligand and result in failure to stimulate adenylyl cyclase(13, 21). Further characterization of mutant genes of P322S and P322H invitro will help us to understand the role of this conserved proline residue.

Previous reports of Japanese patients with X-linked NDI demonstrated several mutations; 528delG, H80R, R143P, and a single codon deletion (from nucleotide 832 to 835)(11, 12). Those mutations are not identical to mutations we identified, implying that the mutations of AVPR2 seem to be heterogeneous even in Japanese populations.

Because the various mutations causing X-linked NDI have been described so far in North America, Europe, and Japan(5–22), direct analysis of mutations in AVPR2 will be required for each individual with X-linked NDI. This will help in the accurate diagnosis of patients and carriers of X-linked NDI.

Table 2 Sequence of primers used for PCR-SSCP analysis and direct sequencing
Full size table