Molecular diagnosis of distal renal tubular acidosis in Tunisian patients: proposed algorithm for Northern Africa populations for the ATP6V1B1, ATP6V0A4 and SCL4A1genes

Distal renal tubular acidosis (dRTA) is defined by the inability of the kidney to acidify urine, in presence of metabolic acidosis [1, 2]. Patients present with hyperchloraemic metabolic acidosis, hypokalaemia [3], bone disease and, commonly, nephrocalcinosis. Some affected individuals also exhibit hearing impairment [3, 4]. Malfunction of the H + −adenosine triphosphatase (ATPase) pumps, or Cl/HCO3- (anion exchanger AE1) at the α-intercalated cells of the distal nephron causes failure to excrete acid or reabsorb bicarbonate, respectively, and leads to primary dRTA [1, 5].

Primary dRTA caused by mutations of H + −ATPase pump subunits is wide genetic and clinically heterogeneous [6]. In 1999, Karet et al demonstrated that, in families with classic dRTA and early sensorineural hearing loss (SNHL), which became clinically evident from birth to late childhood, the disease was caused by mutations in the ATP6V1B1 gene, encoding the B subunit of the H + ATPase pump [7]. Later, in 2000, Smith et al found that some families with dRTA and normal audiometry presented with mutations in the ATP6V0A4 gene, which encodes for the A4 subunit of the H + ATPase pump [8].

Despite those original molecular descriptions distinguished between ATP6V1B1 and ATP6V0A4 gene mutations based on the presence or absence of early SNHL, phenotype overlapping between both subgroups of patients makes that classification inaccurate. Indeed, Stover et al demonstrated that young adult patients with ATP6V0A4 mutations could develop mild SNHL in the long term [9]. Further, some dRTA patients without early SNHL were found to present mutations in the ATP6V1B1 gene as well [10, 11].

Further, autosomal dominant (AD) and autosomal recessive (AR) dRTA have also been associated with mutations in the SLC4A1 gene encoding the human AE1 [12–15], which also plays a central role in the maintenance of acid–base balance [16, 17]. Overall, lack of phenotype-genotype correlation supports the need of genetic analysis in dRTA patients.

Up to now, small series of patients with dRTA from different ethnic backgrounds had been genetically studied [7–9, 11, 18]. Despite a few data on Tunisians have been reported [8, 9, 11], to our knowledge large cohorts of Tunisian patients with dRTA have not yet been analyzed. In order to establish the genetic diagnosis of primary dRTA in Tunisia, this study aimed to identify the molecular defects that occur in ATP6V1B1, ATP6V0A4 and SLC4A1 genes using a proposed algorithm.

This current study is, to our knowledge, the first large Tunisian dRTA cohort that has been genetically studied so far. As a whole, we identified three different ATP6V1B1 gene mutations (in 13 families), and four ATP6V0A4 mutations (in 7 families) in 25 Tunisian children. Remarkably, one novel mutations was observed, c.1102G > A; p.Glu368Lys in the ATP6V1B1 gene. On the other hand, most studied patients were deaf since early childhood and, remarkably, 75% of them (12/16) had the ATP6V1B1 gene c.1155dupC mutation [7] in homozygosis, a finding that may improve genetic diagnosis in Tunisia and close populations.

Despite primary dRTA is a rare disease of unknown prevalence, we found a surprising large number of affected Tunisian patients, a country with an estimated population of 10,4 million inhabitants and a mean endogamy rate of 26% [24]. This can be explained, almost in part, by the high rate of consanguinity found in our study cohort, up to 80%. Patients’ clinical presentation was consistent with the prototypic picture of primary dRTA, and every patient showed inability to acidify urine in presence of non-gap metabolic acidosis [1, 3, 19]. In our series, ATP6V1B1 gene causative mutations were 2 time-fold more frequent than ATP6V0A4 gene mutations: 13 vs. 7. This finding may be related not only to ethnicity or the studied population’s characteristics, but also to an easier recognizable pattern of severe acidosis associated with early onset SNHL, leading straightforward to the disease clinical diagnosis [1, 7].

Similar presentation with early onset of non-gap metabolic acidosis, hypokalemia and nephrocalcinosis was observed in almost all patients despite carrying different mutations. No phenotype-genotype correlation in dRTA with SNHL was detected in this cohort, as previously described [10]. On the other hand, this study confirms the association between ATP6V1B1 mutations and dRTA with early onset SNHL since childhood, and ATP6V0A4 mutations with dRTA but normal hearing (at least until young adulthood), as reported in the original disease genetic descriptions [7, 8, 18, 25, 26], but conflicts with some other reports [9, 11], which described ATP6V0A4 gene mutations to be more frequent than ATP6V1B1 ones.

However, with more recent clinical follow up data, we have become aware that hearing impairment may represent a spectrum in dRTA, and hearing loss may also develop at an older age in a number of those affected patients with ATP6V0A4 mutations, but likely less severe than in those with ATP6V1B1 mutations [9]. Indeed, two siblings (cases 16.1 and 16.2) carrying ATP6V0A4 mutations had dRTA with SNHL since childhood, as described [11]. Reversely, patients with ATP6V1B1 mutations may not present hearing loss [11]. One of our patients with the ATP6V1B1 gene c.1155dupC mutation (case 7.2) developed hypoacusia after diagnosis. In general, these findings raise the question regarding potential mechanisms involved in the effect of different subunit mutations of the same proton pump over hearing ability [9].

Importantly, one of the most interesting findings of this study was the detection of ATP6V1B1 c.1155dupC mutation in homozygosis in 12 patients from 9 families. In all probands, SNHL was evident since childhood, as classically described [7]. This mutation is present in 12 out of 16 (75%) patients with SNHL. This c.1155dupC mutation had previously been described in 12 families, of which 7 are from North Africa [7, 9, 19]. Therefore, if we consider these studies, performed in North African patients with dRTA , it turns out that approximately 55% of patients suffering from dRTA with early onset SNHL (before 4 years of age) carry this mutation.

Anyhow, this finding may facilitate and help for rapid genetic diagnosis of dRTA in North African patients; and therefore, we propose that in patients with dRTA and SNHL from Northern Africa countries, c.1155dupC mutation in the ATP6V1B1 gene should be screened first and, then, analysis should continue with additional mutation screening only in those affected subjects in whom c.1155dupC is not found.

Regarding the origin of the other mutations, the splice-site (c.175-1G > C; p.?) defect was described for the first time in three families from Algeria and Tunisia [11]. The nonsense c.16C > T mutation was previously described in a Pakistani family [11]. The two missense c.2035G > T and c.1739 T > C mutations were previously described in one Pakistani [11] and one Turkish family [8], respectively (Table 3).

On the other hand, the newly identified mutation (c.1102G > A; p.Glu368Lys) may provide in the future some insights into A1 subunit structure and function, which are presently unknown in mammals.

The diagnostic strategy based on the presence of SNHL was successfully used in the current study to determine which gene to test first. Vargas-Poussou et al performed linkage studies to identify the disease locus in consanguineous families, and then used the age of onset of SNHL to determine which gene to test first in the non-consanguineous families [11]. Comparing with the algorithm followed by these authors, our strategy resulted in a higher mutation detection global rate: 95% versus 79.5%.

Nevertheless, our mutation screening partially failed in three patients (two of them are siblings). We only identified two already described heterozygous mutations (c.2035G > T and c.1739 T > C) in the ATP6V0A4 gene. We were unable to find the second mutation responsible for the recessive inheritance of the disease, as described by other authors [11], even after screening the SLC4A1 gene [12, 27, 28]. We admit that SLC4A1 gene was a poor candidate for screening as our patients presented severe metabolic acidosis early in life, but we wanted to exclude the possibility of presenting any mutation in the SLC4A1 gene, as compound heterozygosity has been described [14, 29]. Anyway, we cannot rule out the possibility of a potential undetected deletion involving one or more exons of the analyzed genes [30]. Other authors have hypothesized about the possibility of mutations occurring in an unknown regulatory element of either gene, like promoters [31] or that some intronic variants, especially those affecting the normal splicing sites, may also be involved in the pathogenesis of dRTA [11]. More plausible, there might be other genes involved in tubular acidification ability not yet discovered. Haplotype linkage studies should be performed in those affected patients in whom the second mutation is missed and we plan to continue with that project in the future.

In this report, the genetic analysis of a Tunisian cohort of dRTA patients as well as a new diagnostic strategy to facilitate the genetic testing in North Africans with dRTA and SNHL is presented. The putative dRTA mutation was found in 90% of the index cases (18/20, the other two being heterozygous), thus genetically confirming the clinical diagnosis.

Our findings confirm the association of ATP6V1B1 mutations and early onset severe SNHL since childhood, and identify delayed hearing loss as a feature associated with ATP6V0A4 mutations in some patients as well. Based on our results, we propose that in patients of Northern Africa origin with dRTA and SNHL, c.1155dupC mutation in the ATP6V1B1 gene should be screened first. Further studies with more patients would be needed to confirm this result.