Introduction

Type 1, distal renal tubular acidosis (dRTA) is a rare disorder characterized by persistent hyperchloremic, normal plasma anion gap, and metabolic acidosis in the presence of inappropriately high urinary pH and low urinary excretion of ammonium [1]. dRTA is caused by inability of the α-intercalated cells of the collecting tube to acidify the urine. Proximal leak of bicarbonate does not occur, and glomerular filtration rate is characteristically normal [2].

In children, dRTA is usually primary, common presenting manifestations being growth retardation, vomiting and dehydration, loss of appetite, diarrhea or constipation, and polyuria [3]. Hypokalemia is often found and may lead to weakness and paralysis [4]. The association of hypocitraturia and elevated urine calcium excretion leads to nephrocalcinosis and increased risk of urolithiasis [5].

Primary dRTA results from genetic defects, the most frequently implicated genes being ATP6V1B1, located at chromosome 2 (2p13.3), and ATP6V0A4, at chromosome 7 (7q33-34), which, respectively, encode the B1 and A4 subunits of the H+-ATPase of the α-intercalated cells [6]. The B1 isoform of the H+-ATPase V1 domain is also expressed in the inner ear cells [7]. Loss of function mutations in ATP6V1B1 and ATP6V0A4 genes causes early onset and severe forms of autosomal recessive dRTA [8]. The SLC4A1 gene, located at chromosome 17 (17q31.21), codifies the exchanger Cl−/HCO3− (AE1) placed on the basolateral surface of α- intercalated cells and in erythrocyte membrane. Mutations in SLC4A1 gene cause a milder form of dRTA that follows an autosomal dominant inheritance and often presents in adulthood [9,10,11], thereby few pediatric cases have been reported. Autosomal recessive dRTA caused by mutations in SLC4A1 gene have particularly been described in Asian people in association with ovalocytosis and spherocytosis [12,13,14]. Recently, recessive missense mutations in FOXI1 gene, encoding the transcription factor FOXI1 that regulates a group of membrane transport proteins in the collecting duct, have been found in two unrelated consanguineous families as responsible of sensorineural deafness and dRTA [15].

We used the RenalTube database [16] to better characterize the phenotypical spectrum of primary dRTA caused by ATP6V1B1, ATP6VOA4, or SLC4A1 gene defects and find out whether clinical and/or biochemical manifestations might help differentiate these three types of dRTA.

Material and methods

Patients

Twenty-seven pediatric patients with primary dRTA aged from 1 month to 15 years (18 males) and corresponding to 25 families were selected from the RenalTube database and grouped according to the underlying genetic defect as follows: ATP6V1B1, ATP6V0A4, and SLC4A1 groups. Sex, age, presenting manifestations, family history, biochemical data, and image studies at diagnosis were analyzed and compared.

Metabolic acidosis was diagnosed by a blood pH < 7.35 and/or serum bicarbonate concentration < 22 mEq/L. Hyperchloremia was defined by serum chloride values > 105 mEq/L, and hypokalemia was defined by serum potassium < 3.5 mEq/L. A urinary acidification defect was diagnosed by urinary pH > 5.5 in the presence of metabolic acidosis. Hypercalciuria was defined by calcium/creatinine ratio > 0.8 mg/mg in infants aged between 1 and 6 months, > 0.6 mg/mg in those aged 6–12 months, > 0.47 mg/mg in children of 1 year of age and > 0.22 mg/mg in those aged > 2 years [17]. The diagnosis of hypocitraturia was made when first morning urine citrate/creatinine ratio was < 400 mg/g. Fractional excretion of potassium (FEK) was calculated by the formula: (urine potassium × serum creatinine/serum potassium × urine creatinine) × 100.

Mutations of ATP6V1B1, ATP6V0A4, and SLC4A1 genes were identified by next-generation sequencing (NGS) and validated by Sanger sequencing, as described [18].

Statistical analysis

Comparisons among the three groups were performed using SPSS software (SPSS V15.0 Windows). Age at diagnosis and Ca/Cr ratio were expressed as median and interquartile range for being not normal quantitative variables and were compared by non-parametric Kruskal-Wallis test. The normal quantitative variables, expressed as mean and standard deviation (X ± SD), were compared using one-way Anova. χ2 Pearson or the Fisher exact test was used to relate categorical variables. P values lower than 0.05 were considered statistically significant. Height and weight were represented graphically as boxplots (median and interquartile).

Results

Table 1 shows the demographic, clinical, and genetic data of the three groups of patients: 10, 12, and 5 cases harboring mutations in ATP6V1B1, ATP6V0A4, SLC4A1 genes, respectively. Patients with ATP6V1B1 or ATP6V0A4 gene mutations (I.1-XX.1) were diagnosed earlier (p < 0.002) than patients with SLC4A1 mutations (XXI.1-XXV.1), median ages being 7 (30) months for ATP6V1B1, 3 (9) months for ATP6V0A4, and 120 (60) months for SLC4A1 patients. No differences were found in height and weight among groups (Figs. 1 and 2).

Table 1 Demographic and clinical features of the three groups of patients at diagnosis
Full size table
Fig. 1
figure 1

Boxplot showing the weight SDS at diagnosis. ATP6V1B1 means patients with mutation in ATP6V1B1 gene. ATP6V0A4 means patients with mutation in ATP6V0A4 gene. SLC4A1 means patients with mutation in SLC4A1 gene

Full size image
Fig. 2
figure 2

Boxplot showing the height SDS at diagnosis. ATP6V1B1 means patients with mutation in ATP6V1B1 gene. ATP6V0A4 means patients with mutation in ATP6V0A4 gene. SLC4A1 means patients with mutation in SLC4A1 gene

Full size image

Hearing loss was recognized in eight children with ATP6V1B1 mutations and two children with ATP6V0A4 mutations. None of the SLC4A1 patients had sensorineural hearing impairment.

Diagnostic laboratory tests are shown in Table 2. Differences were only found for serum potassium, higher inSLC4A1 than ATP6V0A4 patients (p = 0.046).

Table 2 Biochemical features of the three groups of patients at diagnosis
Full size table

Discussion

This study provides interesting findings useful for the diagnosis and phenotypical characterization of primary dRTA. Few publications [18,19,20,21], such as those of Palazzo et al. [20] and Besouw et al. [22] recently reported, have compared the clinical manifestations of pediatric patients with dRTA classified according to the underlying genetic defect. Among the patients here presented, ten had mutations in theATP6V1B1 gene. Five of these children were Africans and harbored the same mutation (c.1228 in. C; p.I386Hfs.) in homozygosis, according to the founder effect of this variant proposed by Nagara et al. for dRTA patients from North-African geographical origin [22]. In this group, the five remaining patients were Caucasian and three (V.1, VI.1, and VII.1) carried the same mutation (g.70960079; Intron 6 + 1 G > A, Splicing), in spite of no known familiar relationship between them.

As for the SLC4A1 group, it is of note that almost no data are available in the literature on children of Occidental origin with this variant form of autosomal dominant dRTA not associated with hemolytic anemia [23,24,25,26,27]. SLC4A1 variants found in our patients have already been related with a late clinical onset of dRTA [18, 28, 29]. The median age at diagnosis of our patients with SLC4A1 mutations was 10 years. Therefore, patients with this type of dRTA may present before adulthood, in the late childhood. There was no overlap in the age of diagnosis between dRTA caused by SLC4A1 gene mutations and the other two types of dRTA. Patients with ATP6V1B1 and ATP6V0A4 gene defects in our series debut in infancy at a mean age of 7 and 3 months, respectively. However, it should be noted that two children having ATP6V1B1mutations and one child with ATP6V0A4 mutations were diagnosed after the second year of life indicating that these forms of dRTA may not be detected during early infancy. Patients with SLC4A1 gene mutations had less severe forms of dRTA and tended to have less marked growth retardation, less severe metabolic acidosis, and significantly milder hypokalemia which might someway justify the later diagnosis. Besouw et al. [21] found that children with H+-ATPase pump defects needed higher alkali doses to correct acidosis than children with SCL4A1 mutations. It has also been reported that the degree of acidosis or hypokalemia varies depending on whether the dRTA is autosomal recessive (ATP6V0A4 and ATP6V1B1 genes) or dominant (SLC4A1 gene). Battle et al. [29] showed that individuals with autosomal-recessive pattern had serum potassium levels lower than those with an autosomal-dominant inheritance.

Neither the frequency of nephrocalcinosis or urolithiasis nor the urinary calcium excretion was different among the three groups.

The patient XXIII.1 of the SLC4A1 group had a height Z score equal to − 2.82 DS, much greater growth retardation than the other group’s cases. It is worth commenting that this patient, additionally to the de novo p.R589C mutation, had a polymorphism (rs148170067 SNP; c.889 C > T, p.V245 M) inherited from the father, who never manifested any symptom related to dRTA. Thus, this polymorphism could induce a synergistic negative effect, enhancing the harmful impact of the mutation and explaining why this patient had more severe metabolic acidosis and greater growth retardation.

Another noticeable finding of our study was that eight patients with mutations in the ATP6V1B1 gene were deaf, out of ten in whom the symptom was sought, by contrast with only 2 out of 14 children with ATPV0A4 gene mutations. As expected, none of patients harboring SLC4A1 gene defects had deafness because the Cl−/HCO3− anion exchanger does not express in the ears. It was classically assumed that dRTA caused by defective ATP6V1B1 gene was associated with early nerve hearing loss [7, 28, 30,31,32,33], while ATP6V0A4 mutations were related with either late-onset deafness or normal hearing, [34,35,36,37,38,39,40]. Vargas-Poussou et al. [41] challenged this assumption demonstrating genetic heterogeneity in dRTA associated with deafness and emphasizing the importance of mutational gene analysis for recessive forms of dRTA independent of hearing loss. However, although the early presence of neurosensorial deafness does not fully discriminate between the two types of dRTA caused by a loss of function of the H+ ATPase pump, our results indicate that, at early age, the detection of deafness in patients with dRTA is highly suggestive of an underlying mutation in the ATP6V1B1 gene. It should be mentioned that the occurrence of deafness in dRTA has been related with the expansion of the vestibular aqueduct [42, 43], a finding unfortunately not explored in our series of patients.

In summary, we here presented clinical and biochemical data at diagnosis of non-oriental patients with different genetic forms of primary dRTA. At diagnosis, the patient’s age, the severity of hypokalemia, and the presence of hypoacusia might be useful to differentiate the underlying molecular defect which needs to be confirmed by gene analysis.