Molecular cloning and characterization of novel tissue-specific isoforms of the human vacuolar H⁺-ATPase C, G and d subunits, and their evaluation in autosomal recessive distal renal tubular acidosis

Abstract

Several of the 13 subunits comprising mammalian H⁺-ATPases have multiple isoforms, encoded by separate genes and with differing tissue expression patterns, which may play an important role in the intracellular localization and activity of H⁺-ATPases. Here we report the cloning of three previously uncharacterized human genes, ATP6V1C2, ATP6V1G3 and ATP6V0D2, encoding novel H⁺-ATPase subunit isoforms C2, G3 and d2, respectively. We demonstrate that these novel genes are expressed in kidney and few other tissues, and confirm previous reports that the C1, G1 and d1 isoforms are ubiquitously expressed, while G2 is brain-specific. Previously we have shown that mutations in two kidney-specific genes, ATP6V1B1 and ATP6V0A4, encoding the H⁺-ATPase B1 and a4 subunit isoforms, cause recessive distal renal tubular acidosis (dRTA). As the genes reported here are expressed mainly in kidney, we assessed their candidacy as causative genes for recessive dRTA in eight kindreds unlinked to either known disease locus. Although no potential disease-causing mutations were seen in this cohort, this does not rule out a role for these genes in other families. The identification of these three novel tissue-specific isoforms supports the hypothesis that subunit differences may play a key role in the structure, site and function of H⁺-ATPases within the cell.

Introduction

The multi-subunit vacuolar-type proton pump (H⁺- or V-ATPase) is essential for acidification of diverse intracellular compartments in all eukaryotic cells, including vacuoles, clathrin-coated and synaptic vesicles, endosomes, lysosomes and chromaffin granules (Nishi and Forgac, 2002). In addition, H⁺-ATPases are found at high density in the plasma membrane of specialized epithelial cells such as the renal intercalated cell, the osteoclast and a subpopulation of cells in the male genital tract, where they play an important role in urinary acidification, bone resorption and sperm maturation, respectively. Thus in each individual cell, H⁺-ATPases may function in a variety of distinct but essential cellular processes. However, the mechanisms by which cells regulate the intracellular trafficking, final destination and activity of these proton pumps are unclear. One important question is whether there are any differences in the structure of the proton pumps found in different locations that might provide an insight into these processes.

The general structure of H⁺-ATPases comprises two functional sectors, V₁ and V₀. The peripheral V₁ domain hydrolyzes ATP, providing the energy for H⁺ transport across the integral membrane V₀ domain. The complete identity of all the pump components has yet to be elucidated, but the structural model put forward by Nishi and Forgac (2002), which is based on studies of H⁺-ATPases from various sources, suggests that there are at least 13 different subunits. In this model, as depicted schematically in Fig. 1, the V₁ domain comprises subunits A–H, in a proposed stoichiometry of A₃B₃C₁D₁E₁F₁G₂H₁, while V₀ contains five subunits in a possible complex of a₁d₁c″₁(c, c′)₆. Whether an ortholog of the c′ subunit exists in mammals is currently unclear (Smith, Borthwick and Karet, unpublished observations).

In a variety of species, a number of subunits have been shown to have multiple isoforms encoded by different genes, including the B, E, G and a subunits (Nelson et al., 1992, Crider et al., 1997, Smith et al., 2001, Imai-Senga et al., 2002) (it must be noted that the two previously reported bovine H subunit isoforms are in fact splice variants of the same gene (Zhou et al., 1998)). The existence of different subunit isoforms may play an important role in the localization and activity of proton pumps in specific cell types and subcellular compartments. Indeed, it is possible that each type of proton pump has its own unique subunit identity.

In support of this hypothesis, the proton pumps expressed at high density at the cell surface of intercalated cells in the distal nephron have been found to contain the predominantly kidney-expressed B1 and a4 isoforms of the B and a subunits, respectively (Nelson et al., 1992, Smith et al., 2000). Notably, the renal cell surface B1 and the ubiquitously expressed (and usually intracellular) B2 isoforms differ at their C-termini, in a manner predicted to affect membrane targeting (Breton et al., 2000). Furthermore, it appears that the B2 subunit cannot substitute for B1 at the intercalated cell surface, since we have shown that mutations in the B1-encoding ATP6V1B1 gene cause the syndrome of recessive distal renal tubular acidosis with sensorineural hearing loss (dRTA with SNHL) (Karet et al., 1999). This disorder results in severe metabolic acidosis, accompanied by disturbances of potassium balance, urinary calcium solubility, bone physiology and growth (Batlle et al., 2001). Likewise, mutations in the ATP6V0A4 gene, encoding the a4 subunit, cause recessive dRTA (Smith et al., 2000) and therefore the presence of this particular a subunit isoform also appears to be essential for normal functioning of the specialized H⁺-ATPases found at the cell surface in the distal nephron. In a similar vein, it has recently been shown that the osteoclast-specific a3 subunit is essential for the maintenance of normal bone turnover in both mouse and human (Li et al., 1999, Frattini et al., 2000).

Interestingly, a number of the recessive dRTA families that we have studied do not support linkage to either ATP6V1B1 or ATP6V0A4, and this implies the existence of at least one additional dRTA locus. We hypothesized that the identification of genes that encode further kidney-specific proton pump subunit isoforms would provide good candidate genes for dRTA, and might also provide new insights into proton pump structure and function in the kidney.

In this study, we report the identification and characterization of three genes that encode novel isoforms of the H⁺-ATPase C, G and d subunits in man. Since we have demonstrated that these genes are expressed in kidney and few other tissues, we have also assessed their candidacy as causative genes of recessive dRTA in our cohort of families that are unlinked to the two known disease loci.