Abstract
Mutations in the ATP6V1B1 and ATP6V0A4 genes, encoding subunits B1 and 4 of apical H+ ATPase, cause recessive forms of distal renal tubular acidosis (dRTA). ATP6V1B mutations have been associated with early sensorineural hearing loss (SNHL), whereas ATP6V0A4 mutations are classically associated with either late-onset SNHL or normal hearing. The phenotype and genotype of 39 new kindreds with recessive dRTA, 18 of whom were consanguineous, were assessed. Novel and known loss-of-function mutations were identified in 31 kindreds. Fourteen new and five recurrent mutations of the ATP6V0A4 gene were identified in 21 families. For the ATP6V1B1 gene, two new and two previously described mutations were identified in 10 families. Surprisingly, seven probands with ATP6V0A4 gene mutations developed severe early SNHL between the ages of 2 mo and 10 yr. No mutation was detected in eight families. These data extend the spectrum of disease-causing mutations and provide evidence for genetic heterogeneity in SNHL. The data also demonstrate that mutations in either of these genes may cause early deafness, and they highlight the importance of genetic screening for recessive forms of dRTA independent of hearing status.
Primary distal renal tubular acidosis (dRTA) is a rare genetic disease in which the intercalated cells in the collecting duct fail to secrete the H+ required for final urinary excretion of fixed acids. Clinical and biologic features include hyperchloremic metabolic acidosis, impaired growth, hypokalemia, nephrocalcinosis, nephrolithiasis, hypercalciuria, hypocitraturia, and rickets or osteomalacia (1,2). Both autosomal dominant and autosomal recessive forms have been described. The autosomal dominant form (OMIM 179800) is caused by mutations in the gene encoding the basolateral Cl−/HCO3− exchanger (SLC4A) (3). Autosomal recessive forms have been associated with mutations in the B1 subunit of the apical H+ ATPase gene (ATP6V1B1, initially known as ATP6B1) in individuals who display sensorineural hearing loss (SNHL; OMIM 267300) (4) or with mutations in the a4 subunit of the apical H+ ATPase gene (ATP6V0A4, initially known as ATP6N1B or ATP6N2) in individuals who do not have SNHL or display hearing loss only after the age of 10 yr (OMIM 602722) (5,6). However, at least two cases of ATP6V1B1 mutation without SNHL have been described (4,7). The spectrum of severity of SNHL and the range of ages over which hearing loss occurs in patients with ATP6V0A4 mutations are unclear. Finally, there is evidence that the recessive forms are genetically more heterogeneous (6).
In this study, we determined the frequency of ATP6V1B1 and ATP6V0A4 mutations in a group of 39 new families with autosomal recessive dRTA and evaluated the frequency of SNHL for each gene. We found a high frequency of ATP6V0A4 mutations (54% of all cases), some of which were associated with early SNHL. We propose a strategy for genetic screening for recessive dRTA, based on the frequency of mutations in the two genes and their associated phenotypes.
Materials and Methods
Patients
We studied 43 probands from 39 families who had diagnoses of dRTA (Tables 1 through 3) after we obtained written informed consent. Diagnosis was based on the presence of the following necessary criteria: Metabolic acidosis with a normal anion gap and urinary pH >5.5 in a context of acidosis, hypercalciuria, and/or nephrocalcinosis. Additional optional criteria were hypokalemia, hypocitraturia, polyuria, and failure to thrive. Hearing was assessed by pure-tone audiometry and/or auditory evoked responses. The severity of hearing loss was graded according to “Bureau International d’audiophonologie” recommendations (www.biap.org) as mild (20 to 40 dB), moderate first degree (41 to 55 dB), moderate second degree (56 to 70 dB), severe first degree (71 to 80 dB), and severe second degree (81 to 90 dB) in all cases for which appropriate data were available. Venous blood samples were obtained from the probands and from six affected and 30 unaffected family members.
Clinical characteristics of dRTA patients with ATP6VOA4 mutationsa
Linkage Studies
Peripheral blood samples were obtained and genomic DNA was extracted by standard methods. Haplotype analysis was carried out, and the disease locus was identified in consanguineous families by PCR amplification of polymorphic microsatellite markers for each gene. For the ATP6V1B1 gene, we used three flanking markers—D2S443, D2S291, and D2S2977 (GenBank accession nos. G08201, G27287, and Z23530)—with the following genetic map: D2S443—0.4 cM—ATP6V1B1—0.7 cM—D2S291—0.9 cM—D2S977. For the ATP6V0A4 gene, we used the markers D7S2560, D7S684, and D7S1824 (GenBank accession nos. G20131, Z24317, and G08617). The genetic map used was as follows: D7S2560—0.2 cM—ATP6V0A4—0.06 cM—D7S684—1.5 cM—D7S1824. We evaluated the evidence for linkage at each locus qualitatively by looking for homozygosity.
Analysis of the DNA Sequence of the Two Genes
The coding exons and intron-exon junctions were amplified with specific primers. For the ATP6V1B1 gene, we used the primers described by Karet et al. (4). The primers used for the ATP6V0A4 gene were selected on the basis of the published sequence of the gene (available upon request). We carried out direct sequencing, using the dideoxy chain termination method on an automated Perkin Elmer/Applied Biosystems (Foster City, CA) Division 373A Stretch DNA capillary sequencer and evaluated sequences with Sequencher software.
Results
Genetic Screening
On the basis of family history and clinical presentation, 39 families were classified as having autosomal recessive dRTA. Their clinical and biologic characteristics are indicated in Tables 1 through 3. Most of these families were of North African origin.
Eighteen of the families were consanguineous. In 10 families, the disease locus haplotypes showed homozygosity, implicating the ATP6V0A4 gene. These data were confirmed by the detection of one homozygous mutation in each index case. In another seven families, haplotype analysis suggested compatibility with linkage to the ATP6V1B1 gene. Homozygous mutations also were detected in five of the probands but not in the sixth, from family 21. Therefore, overall, this strategy, based on homozygosity mapping in the 18 consanguineous families, resulted in a mutation detection rate of 89%.
In the 21 nonconsanguineous families, age at onset of SNHL was used to determine which gene to test first. We initially screened for ATP6V0A4 mutation in patients without SNHL or in whom SNHL began after the age of 10 yr. Conversely, we initially screened for ATP6V1B1 mutation in patients who had SNHL that began before the age of 10 yr.
On the basis of this strategy, we began by screening for mutations in the ATP6V0A4 gene in 15 sporadic cases. Mutations were detected in seven patients (Table 1). We then investigated the ATP6V1B1 gene in the remaining 10 patients with no mutations. Mutations were detected in two probands with normal hearing (patients 8 and 28). We first screened for mutations in the ATP6V1B1 gene in five sporadic cases with early-onset SNHL. Mutations were detected in two of these cases (Table 2). In the remaining three patients, mutations were found in the ATP6V0A4 gene.
Clinical characteristics of dRTA patients with ATP6V1B1 mutationsa
On the basis of the strategy described above, the initial screening of nonconsanguineous families gave a mutation detection rate of 43%, and this rate increased to 67% with the second screening. No mutation was detected in seven nonconsanguineous families.
Mutations in the ATP6V0A4 Gene
Analysis of the nucleotide sequence of the entire coding region of the ATP6V0A4 gene from 24 unrelated index cases revealed the presence of 18 different mutations (Table 1). We identified one missense, nine nonsense, six frameshift, and two splice site mutations that were distributed evenly throughout the gene. Fourteen of these mutations have not been described before. Sixteen individuals were homozygous for one mutation, and five were compound heterozygotes.
In family 25, we detected two homozygous nonsense mutations (R6X and Y450X) in two sisters. Both parents were heterozygous for the two mutations.
On the basis of the similar geographic origins of the patients, we hypothesized that there might be a founder effect for mutations Y129X and R770X. The Y129X mutation was detected in six families, in the homozygous state (n = 4) or associated with a second severe mutation (n = 2). Five of these families originated from Algeria, and the other originated from Morocco. All shared the same haplotype at the disease locus (Figure 1), suggesting that this mutation probably was inherited from a common ancestor. The novel R770X mutation was detected in two families from Mali. These two families also shared the same haplotype (Figure 2). To our knowledge, these two families were not related, suggesting but not proving a founder effect.
Haplotypes of six North African families who carry the Y129X mutation. Families 13 and 34 were consanguineous; families 5 and 10 have no history of consanguinity, but the mutation was homozygous; and in families 2 and 9, the Y129X mutation was associated with a second mutation in the ATP6V0A4 gene. The mutation therefore is carried on the haplotype with the alleles 6 and 5 for markers D7S2560 and D7S684, respectively. The marker allele size is given in the supplementary table (available online at www.jasn.org).
Haplotypes of families 7 and 38 who are from Mali and carry the R770X mutation in the ATP6V0A4 gene. The marker allele size is given in the supplementary table (available online at www.jasn.org).
The M580T change initially was described as a mutation (5) but was reported recently to be a polymorphism in the Japanese population (8). Patient 22-1 had a homozygous M580T genotype (Table 1) and also carried a homozygous frameshift mutation. One of the two girls with dRTA in family 21 was found to have a heterozygous M580T genotype (Table 3). These observations are consistent with the M580T amino acid change corresponding to a rare polymorphism rather than a mutation.
Clinical characteristics of patients with dRTA without mutations
A brother and a sister in family 10 had dRTA, but only the brother had SNHL. Both were homozygous for the Y129X mutation. Six additional patients who had SNHL that began in the first 10 yr of life (patients 7, 17, 26, 30, 33, and 40) had nonsense or frameshift mutations in the ATP6V0A4 gene.
Mutations in the ATP6V1B1 Gene
Analysis of the nucleotide sequence of the entire coding region of the ATP6V1B1 gene from 21 probands revealed the presence of four different mutations in 10 families (Table 2). The I386fsX441 mutation was observed in six families: in the homozygous state in four (three from Algeria and one from Tunisia) and in a compound heterozygous state in two families from Algeria. These families did not all have the same haplotype at the ATP6V1B1 locus. This I386fsX441 mutation was previously described in six families from Saudi Arabia, Sicily, Morocco, Sweden, and Spain (4,6). A novel mutation in intron 2, affecting the splice acceptor site (IVS2-1G>C), was identified in three probands. Two were homozygous, and the other was a compound heterozygote. These families did not all have the same haplotype at the ATP6V1B1 locus. Finally, a heterozygous missense mutation (R394Q) was present in two probands from France. This missense mutation probably is a loss-of-function mutation, modifying an amino acid that is highly conserved in proton ATPases. Early SNHL was detected in seven probands with mutations in the ATP6V1B1 gene (Table 2).
Discussion
In this study, we evaluated the two genes that are implicated in recessive dRTA in 39 families. We detected loss-of-function mutations in 31 families, with a global detection rate of 79.5%. In these families, mutations in the ATP6V0A4 gene were twice as frequent as mutations in the ATP6V1B1 gene: 21 versus 10. This finding conflicts with previous studies, in which mutations in the ATP6V1B1 gene were more frequent than mutations in the ATP6V0A4 gene (4–7,9,10). Most of the mutations in the two genes that are considered here were nonsense and frameshift mutations, resulting in unstable mRNA or truncated proteins.
We also detected a large number of new mutations, most of which were in the ATP6V0A4 gene (n = 14) rather than the ATP6V1B1 gene (n = 2). The missense mutations probably were pathogenic, on the basis of both their presence in affected individuals and their predicted biologic consequences. For example, the D679Y mutation in the ATP6V0A4 gene leads to the replacement of an acidic residue by an uncharged polar residue. The aspartic acid residue in position 679 is moderately conserved among a4 subunits of the vacuolar H+ ATPases of various species (family ID ENSF00000000479). Similarly, the R394Q mutation in the ATP6V1B1 gene modifies the polarity of a residue that is highly conserved among the B subunits of vacuolar H+ ATPases from different species until the bacteria (family ID ENSF00000001603).
It is interesting that our haplotype and geographic data were consistent with a founder effect for the recurrent mutations Y129X and R770X in the ATP6V0A4 gene. Y129X was detected in six North African kindreds, and R770X was detected in two Malian families. In contrast, six other kindreds who harbor the recurrent I386fsX441 mutation had different haplotypes at the ATP6V1B1 locus. This mutation was described previously in six families from North Africa, Saudi Arabia, and Sicily (4,6). These data confirm that I386fsX441 is a frequent mutation.
Our mutation screening was unsuccessful in 10 cases. In two cases, we identified only one heterozygous mutation in the ATP6V1B1 gene. We were unable to find the second mutation, despite direct sequencing of all exons and intron-exon junctions, suggesting that there may be another mutation in a regulatory element in the 5′ or 3′ flanking region or an intronic variant leading to aberrant splicing. We also cannot rule out the possibility of a deletion that is limited to one or a few exons, which would have remained undetectable as a result of amplification of the normal alternative allele in these heterozygous individuals. We also detected no mutations in the probands of two consanguineous families and in six sporadic cases. These patients had similar clinical profiles to the other patients, but clinical and biologic data were scarce for three of these patients, and incorrect diagnosis therefore was possible. In one consanguineous family, linkage to the two loci was excluded. Therefore, these data are consistent with genetic heterogeneity of autosomal recessive dRTA (6,11), although mutations in the ATP6V0A4 and the ATP6V1B1 (n = 31) genes probably account for most cases.
Is SNHL a reliable clinical parameter for identification of the gene affected? Previous studies showed that early SNHL (before the age of 10 yr) occurred in 37 of 40 patients with mutations in the ATP6V1B1 gene and that late-onset SNHL (between the ages of 10 and 40 yr) was observed in patients with ATP6V0A4 gene mutations (5–7,9,10). In our group, seven of the 10 patients with ATP6V1B1 gene mutations had early-onset SNHL. The other three were younger than 10 yr and therefore still may go on to develop early SNHL. Nine of the 23 patients with ATP6V0A4 gene mutations had SNHL. Age at onset was unknown for two patients, and in the other seven patients, SNHL was detected surprisingly early, after only 2 mo in one case and before the age of 10 yr in all cases (Table 1, patients 7-1, 10-1, 17-1, 26-1, 30-1, 33-1, and 40-1). Only one of the two affected siblings in family 10 had SNHL, demonstrating that intrafamilial variability of SNHL can occur. Considerable variation in the severity of SNHL was observed for patients with ATP6V0A4 mutations (Table 1). Twelve of the patients in this group were younger than 10 yr and therefore may develop SNHL in the future. Overall, we found no evidence for an association between early-onset SNHL and the disease gene. Early-onset SNHL was observed in 70% of cases with ATP6V1B1 gene mutations and in 39% of cases with ATP6V0A4 gene mutations. A genetic screening strategy based on this phenotype therefore would often be misleading and is not recommended.
No mechanism has yet been proposed to account for the variability of SNHL in terms of both severity and age at onset, depending on the gene affected. In this study, variability in SNHL was observed both for genes and for missense and nonsense mutations. The maintenance of acidic conditions in the endolymphatic sac seems to be important for cell integrity in the inner ear. This function is fulfilled in part by vacuolar-type H+ ATPases (12–14). However, the topology and the interactions of the various subunits of the inner ear vacuolar H+ ATPase are unclear. Evidence is accumulating that vacuolar H+ ATPase binds numerous regulatory proteins and undergoes complex regulation by several factors, such as acid-base status, electrolyte disturbances, and hormones (12–14). It also has been suggested that the ubiquitous a1 isoform may be able to compensate for a4 function in the inner ear (6). This would account for SNHL’s occurring later in patients with ATP6V0A4 mutations but cannot account for the phenotypic variability that is observed in patients with mutations in this gene. On the basis of current knowledge, we can only speculate that another regulating factor or gene is involved in the production or activity of the inner ear proton pump or that an environmental factor may affect pump function in some patients.
On the basis of our data, we propose a new algorithm for the genetic screening of autosomal recessive dRTA. It is based on the presence or absence of consanguinity and does not take into account SNHL. For consanguineous families, the gene to be investigated is chosen by homozygosity mapping, using three informative microsatellite markers at the ATP6V1B1 and ATP6V0A4 loci. If inconclusive results are obtained, then we propose the same strategy as for sporadic cases: analysis of the ATP6V0A4 gene independent of hearing status, as this gene accounts for most cases (68%) in our experience. If no mutations are detected, then analysis of the ATP6V1B1 gene should be undertaken. This algorithm should detect almost 80% of mutations and is particularly useful for genetic screening in young patients, in whom hearing status is difficult to evaluate, or before the detection of deafness.
Conclusion
The study of 39 new families with recessive dRTA led to the identification of new and recurrent mutations in the two genes implicated in this disease in 79.5% of cases. Phenotypic analysis showed that early SNHL also may occur in patients with mutations in the ATP6V0A4 gene. For this recessive disease, genetic screening should be based on consanguinity and gene mutation frequency, independent of hearing loss. The study of additional dRTA families and the follow-up of such families should make it possible to evaluate the usefulness of this strategy prospectively and to elucidate genotype-phenotype relationships.
Acknowledgments
This work was supported by the Groupement d’intérêt Scientifique Institute (Grant GIS Rare Diseases).
We thank Dr. Isabelle Amstutz Montadert for assistance with hearing evaluation, and we thank the patients and their families, without whom this study would not have been possible.
Footnotes
Published online ahead of print. Publication date available at www.jasn.org.
- © 2006 American Society of Nephrology
References
In this issue
Jump to section
More in this TOC Section
Cited By...
- Mice deficient in H+-ATPase a4 subunit have severe hearing impairment associated with enlarged endolymphatic compartments within the inner ear
- A mouse model for distal renal tubular acidosis reveals a previously unrecognized role of the V-ATPase a4 subunit in the proximal tubule
- Atp6v0a4 knockout mouse is a model of distal renal tubular acidosis with hearing loss, with additional extrarenal phenotype
- V-ATPase V1 Sector Is Required for Corpse Clearance and Neurotransmission in Caenorhabditis elegans