Determining Causality in Gene Discovery: A Nephrologist's Perspective with Insights from ATP6V0A4.

Introduction The discovery of novel gene–disease associations remains a cornerstone of advances in medicine,1 including in nephrology, where genetic insights have reshaped our understanding of kidney diseases.2 Accordingly, reports proposing new monogenic kidney disorders often generate substantial interest. To determine if such associations are truly causal requires systematic evaluation of phenotype, genotype, and, where available, experimental data. Frameworks such as that developed by the Clinical Genome Resource emphasize several core principles: clear and reproducible phenotyping, concordant genetic evidence across unrelated individuals, allele frequencies compatible with disease prevalence and inheritance, and experimental data that are biologically consistent with the proposed mechanism.3 Importantly, functional evidence alone, no matter how elegant, cannot establish causality when the supporting human genetic data are not robust.4 Here, we briefly outline these principles and apply them to a recent report by Peng and colleagues, using it as a case study to illustrate the insights gained when gene–disease claims are evaluated rigorously. Applying These Principles: The Case of ATP6V0A4 p.V512L With these principles in mind, we turn to the recent report by Peng et al., which describes a father–son dyad with hypochloremic metabolic alkalosis and hypokalemia attributed to a gain-of-function (GOF) variant in ATP6V0A4.5 This proposal is striking because it is well known that recessive loss-of-function ATP6V0A4 variants cause distal renal tubular acidosis (dRTA) type 3, characterized by impaired urinary acidification, hypochloremic metabolic acidosis, and hypokalemia.6 A GOF mechanism producing the opposite phenotype—metabolic alkalosis—would represent a reversal of established ATP6V0A4 biology, and therefore, it warrants especially careful clinical and genetic evaluation. In-Depth Genetic Analyses Are Critical Establishing a novel gene–disease association begins with a rigorous and transparent assessment of the genetic evidence. This includes demonstrating that the proposed causal variant is both sufficiently rare and appropriately concordant with the observed phenotype, requirements emphasized in Clinical Genome Resource's clinical validity framework. Against this backdrop, the genetic analyses presented by Peng et al. warrant closer scrutiny. Using whole-exome sequencing, the authors identified a heterozygous variant in ATP6V0A4 (p.V512L) that they describe as novel. However, they do not describe how novelty was assessed, nor how other candidate variants were excluded. A standard first step is to query large population databases, such as gnomAD, to determine if it was previously reported in patients and if its minor allele frequency (MAF) is compatible with the estimated prevalence and inheritance pattern. While the prevalence of the new disorder is unknown, it is presumably rarer than recessive dRTA, with an estimated prevalence of 1:100,000.7 There are 14 heterozygous individuals among 807,097 gnomAD participants who harbor the variant p.V512L (rs752903587). This corresponds to a MAF of 1.6×10−5 (approximately 1:50,000). When these data are stratified by ancestry, we find an even higher MAF among East Asian individuals: 1.33×10−4 (approximately 1:10,000). MAFs in this range are difficult to reconcile with a rare autosomal dominant condition. Five of these 14 carriers also appear in the All of Us (AoU) Research Program database, which provides genotype-linked clinical data extracted from electronic medical records. Although detailed clinical data were available for only two of five individuals, neither has documented overt kidney disease, reinforcing the population-level observation that p.V512L is probably not strongly associated with the expected kidney phenotype. Finally, it is important to note that the authors' detailed analysis showing that in silico predictions label p.V512L as a pathogenic variant is not helpful, as these predictions primarily inform loss-of-function mechanisms and offer little insight into GOF effects.8 When Dealing with Renal Tubulopathies, Accurate Physiologic Phenotyping Is Key Both individuals had mild alkalemia with serum bicarbonate at the upper limit of normal and acidic urine (Table 1). The father has a slightly negative urine anion gap, which is compatible with inappropriate net acid excretion while alkalotic. Table 1 - Biochemical data for the patients from Peng et al.5 and the two subjects from All of Us Characteristics Proband Father All of Us Participant 1 All of Us Participant 2 Reference Ranges Sex Male Male Male Female Age 32 ? 32 29 pH 7.47 7.46 N/A N/A 7.35–7.45 PaCO2 40 39 N/A N/A 35–45 mm Hg HCO3 28 27 N/A N/A 21–28 mmol/L Na 132 138 140 140 137–147 mmol/L K 1.7 3.3 3.6 3.6 3.5–5.5 mmol/L Cl 89 95 105 105 99–110 mmol/L SAG 12 12 11 14 10–20 mmol/L TCO2 [28] [27] 28 25 22–27 mmol/L Creatinine 4.5 0.7 1.1 0.7 0.6–1 μmol/L Urine pH 5 5.5 N/A 6 and 6.5 Depends on acid load Urine K (24 h) N/A 32 N/A N/A <30 mmol/d when hypokalemic Urine Na (24 h) N/A 132 N/A N/A 130–260 mmol/d Urine Cl (24 h) N/A 168 N/A N/A 170–250 mmol/d UAG (24 h) N/A −4 N/A N/A mmol/d SAG, serum anion gap; UAG, urine anion gap. However, if excessive urinary proton secretion were the primary mechanism, as suggested by the acidic urine samples, one would expect serum bicarbonate levels to be clearly elevated to account for the alkalosis. However, both patients demonstrate a combined acid–base disturbance: a mild metabolic alkalosis without the expected respiratory compensation (a rise in PaCO2). For a bicarbonate of 28 mmol/L, one would predict a PaCO2 of approximately 42–44 mm Hg to normalize pH (based on Winter's formula9), yet both individuals have values of 39–40 mm Hg. This pattern suggests suboptimal respiratory compensation rather than a pure renal process. There is no clear physiological mechanism by which mutant ATP6V0A4 could impair respiratory compensation as it is expressed only in the kidneys and salivary glands. Of note, the two p.V512L carriers from AoU with clinical data show high-normal serum bicarbonate without hypokalemia or hypochloremia, and urinary pH above 6.0 in one participant (Table 1). Although limited in scope, these data are inconsistent with a severe renal tubular alkalosis phenotype unless one invokes incomplete penetrance. Elegant Biology Requires Equally Strong Genetic Evidence The authors must be commended for the beautiful experimental work presented in the paper. Their functional studies clearly show that p.V512L increases V-ATPase activity compared with wild-type. However, it is critical to highlight that strong experimental data provide only limited support and cannot, alone, establish that a variant causes the reported disorder. A major problem with the design and interpretation of functional assays for novel gene–disease associations is the lack of a bona fide, calibrated positive control, which limits assay validation and certainty of effect direction and increases vulnerability to protocol and analysis flexibility. In this specific case, the authors could have leveraged well-documented genetic data to provide a few useful comparators. They could have bolstered their claim by testing one established ATP6V0A4 loss-of-function variant known to cause dRTA and one benign variant (based on a high MAF). These controls would provide confidence that their assays are reliable when applied to known contexts. The authors could also have tested another variant at the same codon, p.V512M, which is predicted to be damaging by multiple algorithms. This same-site comparator could help determine if this is an allele-specific GOF phenotype rather than a general perturbation at position V512 and could strengthen their conclusions about the directionality of effects. Integration of New Findings with Known Physiologic Mechanisms Is Key The authors present a compelling model in which enhanced ATP6V0A4 activity at the luminal surface of α-intercalated cells increases urinary proton secretion, producing acidic urine and contributing to metabolic alkalosis. However, it may be too simplistic because it does not account for compensatory responses within the kidney triggered during metabolic alkalosis that can oppose sustained acidification of the urine.10 In metabolic alkalosis, the kidney can reduce proximal net HCO3− reabsorption. Reduction in bicarbonate reclamation can occur by decreasing luminal NHE3-mediated H+ secretion. In parallel, distal bicarbonate secretory capacity may increase through increased abundance and/or apical localization of pendrin in type B and non–type A intercalated cells. These cells are relevant because apical pendrin (a Cl−/HCO3− exchanger) is coupled to basolateral V-ATPase activity (including ATP6V0A4) that extrudes H+ to the interstitium, facilitating intracellular HCO3− availability for apical secretion when luminal chloride is present. Over time, these adaptive mechanisms would be expected to favor normalization of systemic pH and, in settings where distal chloride delivery permits pendrin-mediated exchange, could blunt sustained urinary acidification. Thus, the net effect of hyperactive apical ATP6V0A4 in α-intercalated cells may be offset by concurrent proximal and distal adaptations. Importantly, the near-normal 24-hour urinary chloride excretion documented in the father (but not in the proband, unfortunately; see Table 1) argues against luminal chloride depletion, suggesting that pendrin-mediated bicarbonate secretion was unlikely to be constrained. Conclusions The study by Peng et al. provides valuable biological insights into V-ATPase function, and additional genetic evidence may clarify whether GOF ATP6V0A4 variants contribute to a broader phenotypic spectrum. However, current data challenge the authors' claim that p.V512L causes a rare form of renal tubular alkalosis. The evidence instead suggests that this variant is likely benign or contributes modestly to the phenotype. Given these discrepancies, it would be important to re-evaluate other candidate variants in the proband that may better explain his unusual clinical presentation. In the era of precision medicine, rigor—not novelty—must guide the acceptance of new gene–disease associations. This is especially important when powerful experimental tools can easily outpace the genetic evidence required to support causal claims.