Two patients (IV-2 and IV-4 in Figure 1) were admitted to the hospital at the age of 2 months with fever of unknown origin. NDI was diagnosed on the basis of clinical symptoms and laboratory findings (dehydration, hypernatremia, and hypotonic urine) and failure to increase urine osmolarity in response to 1-2esamino-8-D-arginine vasopressin (dDAVP) (Table 1). The sister (IV-1) had no history of dehydration, but polyuria and polydipsia were noticed by her family members, and NDI was diagnosed on the basis of laboratory findings at the age of 2 years. None of the patients had any evidence of mental retardation or significant disease other than NDI. The pedigree of the Japanese NDI family is shown in Figure 1. Subjects indicated by a diagonal box (II-1, III-3, III-4, and II-4) have a history of polyuria and polydipsia since childhood, but NDI has not been diagnosed.

Genomic DNA from patients and available family members was subjected to PCR analysis. We first used the primer pair (additional file 1) that amplifies the genomic region between exon 1 and exon 3 of AVPR2, and no PCR products were amplified from DNA of IV-2 and IV-4 (data not shown). Therefore, we hypothesised that a large sequence around the AVPR2 gene was deleted and performed PCR using the primers listed in the additional file 1. After narrowing the deleted region, we designed the primer pair AVPR2-GAP4 and obtained 386-bp PCR products for III-2, IV-2, III-3, III-4, II-4, III-10, and IV-4 (Figure 1), suggesting that these seven family members carry a large deletion that covers AVPR2 and part of ARHGAP4. The AVPR2exon1 and AVPR2exon2 primer pairs amplify the genomic region of AVPR2 exon 1 and exon 2, respectively. PCR was successful with DNAs from III-2, II-5, III-9, III-10, and IV-3, indicating that III-2 and III-10 are asymptomatic female carriers of the deletion. The physical map and identified deleted regions are shown in Figure 2. The deleted region includes all of the AVPR2 gene locus and approximately half of the ARHGAP4 gene. Sequencing of the 386-bp product from IV-4 revealed that the 5' and 3' break points were at positions 53,385 and 64,673 bp, respectively, relative to the published sequence (GenBank U52112.2) with an insertion of the GGGTACACCTC sequence in the break point (Figure 3).

We evaluated the immune function of 3 family members lacking ARHGAP4. None of the patients in the present study had a clinical history of recurrent or opportunistic infections. Differential leukocyte counts were within normal range (data not shown), however, a reduction of CD4+ cells was observed in III-4, and a reduction of CD8+ cells was observed in IV-4 (Table 2). Normal levels of CD3+CD4+CD45RA+ and CD3+CD4+CD45RO+ T cells were observed 18 and T cell receptor excision circles (TREC) of whole blood were within normal range in three patients tested. Serum IgG levels were relatively low in adult patients with slight reduction of CD19+ cells in III-3 and III-4. However differentiation of B cells using IgD, IgM and CD27 was not different from normal controls (data not shown).

We examined the gene expression profiles of peripheral blood mononuclear cells (PBMCs) from subjects lacking ARHGAP4 (III-3 and III-4) and 6 healthy male volunteers (age, 23–39 years). Transcripts satisfying all of the following criteria were identified as up- or down-regulated by microarray analysis: (1) transcripts that were expressed in at least 2 of 8 samples, (2) transcripts that showed statistically significant differences between subjects lacking ARHGAP4 and, controls (Welch t-test, q < 0.05; multiple tests were corrected by using the Benjamini and Hochberg false discovery rate 19), and (3) transcripts that showed an average increase/decrease of more than 1.5-fold. Transcripts of ARHGAP4 in PBMCs in III-3 and III-4 were absent, whereas strong expression was observed in all of the controls. Up- and down-regulated genes in PBMCs lacking ARHGAP4 are listed in Table 3. We hypothesized that lack of ARHGAP4 could be compensated for by upregulation of other RhoGAP family members. Fifty-seven transcripts belonging to the RhoGAP family were included in the microarray analysis, and 25, including ARHGAP4 were expressed in at least 2 of 8 samples. Expression levels of the transcripts other than ARHGAP4 are shown in Table 4. None of the transcripts of the RhoGAP family members were statistically significantly up/down regulated in the PBMCs.

The pedigree of the Japanese NDI family is shown in Figure 1. A full verbal and written explanation of the study was given to family members, and informed consent for the genetic study was obtained from members who participated in this study. Informed consent for the children was provided by their parents. This study was approved by the Committee of Ethics of the University of Tsukuba and was performed in accordance with the ethical standards of the 1964 Declaration of Helsinki. DNA was extracted from peripheral blood leukocytes collected in ethylenediaminetetraacetic acid (EDTA). To identify the deleted region around AVPR2, primers were designed according to the published sequence (GenBank accession number U52112.2) with Primer3 software 22. The primer sequences are listed in the additional file 1. Sequencing was performed with the Big Dye Terminator kit (Applied Biosystems, Foster City, California, USA) on an ABI PRISM 3100 DNA Sequencer (Applied Biosystems).

Peripheral blood samples from III-3, III-4 and IV-4 in Figure 1 were labeled with monoclonal antibodies according to the manufacturer's instructions and analysed with a FACSCalibur system (BD Biosciences, San Jose, California, USA). Monoclonal antibodies were purchased from BD Biosciences or Beckman-Coulter (Fullerton, California, USA). Quantification of TRECs was performed as described previously 23. In brief, total DNA was isolated from 100 μl of peripheral blood with the use of a QIAamp DNA Micro Kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. The DNA concentration was determined with a GeneQuant pro system (GE Healthcare Bio-Sciences Corp. Piscataway, New Jersey, USA). Quantitative real-time polymerase chain reaction (PCR) for δRec-ψJα sjTRECs was performed with the following primers and probe: δRec primer, 5'-TCGTGAGAACGGTGAATGAAG-3', ψJα primer, 5'-CCATGCTGACACC'TCTGGTT-3', and δRec probe, FAM-5'-CACGGTGATGCATAGGCACCTGC-3'-TAMRA. As an internal control, RNase P genes were amplified in each sample tested. In each experiment, serial dilutions (10⁹, 10⁸, 10⁷, 10⁶, 10⁵, 10³, 10¹, and 10⁰) of the subcloned sjTREC-plasmid (pCR4-TOPO, Invitrogen Corp., Carlsbad, California, USA) and RNase P-plasmid were used as standards for absolute quantification of TRECs and RNase P copies. Each 20-μl reaction contained 1 μl DNA, and the final concentration of each component for TREC-PCR was as follows: 2× TaqMan Universal PCR Master Mix (Applied Biosystems), 500 nM of each primer, and 250 nM TaqMan probe. The final concentration of each component for RNase P-PCR was as follows: 2× TaqMan Universal PCR Master Mix and 20× TaqMan RNase P Primer Probe (VIC dye) Mix (Applied Biosystems). PCR conditions were 50°C for 2 minutes, 95°C for 10 minutes, and 40 cycles at 95°C for 15 seconds and 60°C for 1 minute. Experiments were performed and analysed with an ABI PRISM 7300 Sequence Detection System (Applied Biosystems). Results were extrapolated to the number of TRECs per μg DNA and that of RNase P copies per μg DNA. To confirm that the DNA yields were adequately purified, TRECs were also normalised to copies of RNase P genomic DNA.

PBMCs from III-3, III-4, and six healthy male volunteers (age, 23–39 years) were purified with a Ficoll-Paque™ gradient (GE Healthcare Bio-Sciences Corp.). RNA from PBMCs was extracted with an RNeasy Mini Kit (Qiagen) according to the manufacturer's instructions. cRNA was synthesized with an Illumina^® RNA Amplification Kit (Ambion, Austin, Texas, USA) according to the manufacturer's instructions. In brief, 500 ng total RNA from PBMCs was reverse transcribed to synthesise first and second strand cDNA and purified on spin columns, and in vitro transcription was performed to synthesise biotin-labeled cRNA. A total of 1500 ng biotin-labelled cRNA was hybridised to a Sentrix Human-6 Expression BeadChip version 2 (Illumina, San Diego, California, USA) at 55°C for 18 hours. The hybridised BeadChip was washed and labelled with streptavidin-Cy3 (GE Healthcare Bio-Sciences Corp.) and then scanned with the Illumina BeadStation 500 System (Illumina). The scanned image was imported into BeadStudio (Illumina) for analysis. Forty-eight thousand transcripts representing six whole-genome samples can be analysed on a single BeadChip. We included at least two technical replicates (i.e., the same cRNA samples) for each BeadChip.

Background-corrected values for each probe on the BeadChip array were extracted using BeadStudio version 1.5.1.3 (Illumina). This is based on the average of negative control genes and is called the method of background normalization by Illumina. These extracted values were exported to the software GeneSpring version 7.3.1 (Silicon Genetics, Redwood, CA), and per chip and per gene normalization were performed. The statistical significance of the microarray data was calculated using the Welch t-test, and multiple tests were corrected by the Benjamini and Hochberg false discovery rate 19. Significance was defined as q < 0.05.