Oxalate Transport as Contributor to Primary Hyperoxaluria: The Jury Is Still Out

Hyperoxaluria is a major risk factor for renal stone disease and can arise from exogenous sources, such as dietary excess and enteric hyperabsorptive states, or from endogenous overproduction. In the healthy individual, most urinary oxalate arises from endogenous production and is excreted almost entirely through the kidney. In hyperoxaluric states, urine becomes supersaturated with respect to calcium oxalate, leading to crystalluria, stone formation, and/or nephrocalcinosis. When end-stage renal failure ensues, systemic oxalate deposition follows in the heart, bone, and blood vessels, leading to significant morbidity and mortality. The accompanying paper by Monico et al1 in this issue of the American Journal of Kidney Diseases explores the hypothesis that hyperoxaluria may be caused by mutations in oxalate transport, in addition to better-known mutations in oxalate production.

The inherited hyperoxalurias, primary hyperoxaluria type 1 and 2 (PH1 and PH2), are rare diseases which have helped to elucidate the metabolic pathways leading to endogenous hyperoxaluria (Fig 1). Both diseases are brought about by failure to remove glyoxylate: PH1 due to deficiency of peroxisomal alanine-glyoxylate aminotransferase (AGT; encoded by the AGXT gene) and PH2 as a result of cytosolic glyoxylate reductase/hydroxypyruvate reductase (GRHPR) deficiency. In both cases, the excess glyoxylate is metabolized to oxalate by lactate dehydrogenase (LDH). These diseases are at the severe end of the spectrum for hyperoxaluria, with 50% of those presenting in childhood developing end-stage renal failure by age 15.2 Another group of patients exists in whom an inherited cause is suspected, but PH1 and PH2 have been excluded.3, 4 These individuals with atypical PH (non-PH1/PH2) may represent a single entity, but equally may include several underlying causes.

View Large Image Download to PowerPoint

Figure 1. The metabolic pathways in primary hyperoxaluria. Abbreviations: AGT, alanine-glyoxylate aminotransferase; GO, glycolate oxidase; GRHPR, glyoxylate reductase/hydroxypyruvate reductase; LDH, lactate dehydrogenase; NAD, nicotinamide adenine dinucleotide; NADH, reduced nicotinamide adenine dinucleotide.

The genes for both AGT and GRHPR have been mapped,5, 6, 7 allowing extensive mutation analysis to be carried out, and attempts have been made to look for genotype-phenotype relationships.8, 9, 10 In addition, the crystal structure of both proteins is now available,11, 12 which has provided explanations for some of the phenotypic heterogeneity at the enzyme level. However, in spite of all this information, we are no nearer to an explanation for the wide phenotypic differences. For example, we have shown that individuals with null alleles of AGXT and GRHPR can present across a wide age range, from neonate to fifth decade,10 and that wide variation, from early onset to cryptic adult, can be seen within the same family.10, 13

One possible explanation for such variability may be the natural variation in each of the contributing enzymes of the metabolic pathway. We have shown in vitro that both AGT and GRHPR exert a protective effect in minimizing glyoxylate toxicity in cultured CHO (Chinese Hamster Ovary) cells transfected with all the relevant enzymes of the glyoxylate pathway.14 These experiments illustrate a clear role for glycolate oxidase (GO; encoded by the HAO1 gene) in the production of glyoxylate, a key contributor to endogenous oxalate, and one could hypothesize a situation where variation in GO activity could lead to substantial interindividual variation in glyoxylate production. In addition, differences undoubtedly exist in the amount of urinary crystallization inhibitors present, the hydration status of an individual, and the presence or absence of infections as well as differences in intestinal oxalate uptake and renal oxalate handling.

Oxalate uptake and renal oxalate excretion have been given a boost by the identification of the SLC26 (solute-linked carrier) family of transporters. This family of anion exchangers includes 10 protein products with multifunctional properties and a range of substrate specificities, 3 of which, SLC26A2, A3, and A4, have disease associations.15 Several have been shown to transport oxalate in exchange for either sulfate or chloride, with 2 of them, SLC26A6 and SLC26A3, showing high affinity for oxalate. SLC26A6 is expressed in the apical membrane of intestinal and renal epithelia, including the pancreatic duct cells16 and duodenum17 and the proximal tubule,18 more specifically the distal proximal tubule, distal convoluting tubule, and intercalating cells of the collecting ducts.19 SLC26A3 is not found in the kidney but is present in the apical membrane of cells in the small intestine and colon.20 SLC26A6 and A3 facilitate transcellular transport of anions across the apical membrane; in some cases this exchange is electrogenic and in others electroneutral.

The Slc26a6 null mouse showed a significant increase in intestinal oxalate absorption which was insensitive to the anion exchange inhibitor DIDS (4,4′ diisothiocyanostilbene-2,2′-disulfonic acid) compared to the wildtype mouse, in which net DIDS-sensitive oxalate excretion occurred.21, 22 These results were interpreted as unopposed oxalate uptake by an as-yet unknown apical transporter, with a reduction in the apical efflux of oxalate (in exchange for chloride) and increased export across the basal membrane. In both mouse knockout models, oxalate excretion was reduced by decreasing dietary oxalate, supporting the theory that the hyperoxaluria is due to a hyperabsorptive state. It is possible that SLC26A3, which is involved in oxalate-sensitive sulfate uptake into the intestinal cells, could be the other apical membrane transporter and that one could reduce oxalate uptake by blocking this transporter. Preliminary studies with an Slc26a3 knockout mouse do indeed show a significant decrease in mucosal to serosal flux of oxalate in both the distal ileum and distal colon.23 The interplay of these 2 transporters therefore appears to have major importance for the regulation of oxalate uptake in the gut, at least in mice.

However, mice are not humans and mice are quite resistant to renal stone formation even in the presence of marked hyperoxaluria.21, 22, 23, 24 Comparative studies of the human and mouse SLC26A6 showed similar bidirectional oxalate flux and comparable exchange rates for chloride/bicarbonate and chloride/hydroxide.25 However, there were significant differences in the affinity for chloride, which appeared to reside primarily, but not exclusively, in the transmembrane domain.25, 26 The significance of the lower affinity of human SLC26A6 for chloride may be that the enteric secretion of oxalate is less in humans than mouse, leading to a greater degree of oxalate absorption with a concomitant increased risk of renal stone formation.

An adaptive mechanism has been shown in animal studies whereby in hyperoxaluric states, eg, renal failure, increasing amounts of oxalate can be secreted into the gut,27 possibly via SLC26A6. Treatment of hyperoxaluric individuals with oxalate-degrading bacteria, specifically Oxalobacter formigenes, can potentially capitalize on this response by enhancing the enteric secretion of oxalate28 and thus has the potential to act as a generic treatment for all forms of hyperoxaluria.

The possibility of a role of SLC26A6 in human forms of hyperoxaluria is therefore one that needs to be considered, whether as a cause of disease itself or as a disease modifier. Monico et al1 explore this hypothesis by analyzing the SLC26A6 gene in patients with PH1, PH2, and atypical PH. Five missense changes were found, including 2 known polymorphisms, p.Val206Met (a valine to methionine change at amino acid 206)29 and p.Val185Met (the same amino acid change at position 185).26 Expression studies have shown that p.Val185Met slightly reduces oxalate influx and efflux in response to extracellular chloride concentration.26 The p.Val206Met polymorphism, which occurred at a relatively high frequency in controls and patients, albeit mainly in the heterozygous state, reduced oxalate influx by 30% relative to the wild type transporter,1 although the effect on oxalate efflux was not described. The authors were unable to show any effect of this variant on urine or plasma oxalate concentration, although it may be that subtle differences in oxalate excretion are overlooked in such gross hyperoxaluric states or that heterozygosity has no effect on transporter function. The results would therefore suggest that mutations in SLC26A6 do not account for atypical PH, at least in the cases examined here, but its role as a modifier of oxalate excretion cannot be ruled out at this stage.