Recent genome-wide association scans identified and replicated association between SNPs at the SLC2A9
gene locus and serum urate and with gout [6
]. In this study we have shown that both human SLC2A9 splice variants, SLC2A9a and SLC2A9b, can mediate urate fluxes at a very high rate and significantly faster than their facilitated transport of either glucose or fructose. The kinetics indicate that the transporter's apparent capacity for substrate, or Km
value (~ 1 mM), is above the basal, physiologic plasma concentrations of urate. Moreover, the Vmax
value indicates a high-capacity transporter. All of these data suggest that this membrane protein plays an important role in the handling of urate in the proximal nephron, which completely fits with the findings from genome-wide scans of common allelic variation elevating urate by 20 μmol/l per allele [7
]. In the context of everyday clinical practice this genetic influence on urate is equivalent to 5%–10% of the normal range of serum urate (180 μmol/l to 420 μmol/l), which is not trivial. These findings are confirmed by complementary functional studies on SLC2A9 in relation to urate handling and gout [25
Current models of urate handling in the proximal convoluted tubule indicate that several types of transporter are involved in the fluxes of urate across both the apical and the basolateral membranes of the epithelial cells. In the apical membrane, these transporters include URAT1, a urate/lactate exchanger that mediates urate movement from urine to epithelium [28
]; OATv1, a putative voltage dependent organic anion transporter [29
]; MRP4, an ATP driven pump [30
]; and UAT, a postulated urate channel [31
]. At the other pole of the cells two of the organic anion exchangers, OAT1 and OAT2, present in basolateral membrane, are thought to be able to handle urate, but their physiological role remains to be confirmed [32
]. Therefore at present there is a well-defined absorptive route across the apical membrane via UAT1 and a secretory route via MRP4, while the means by which urate can either leave the renal epithelium and enter the blood or move in the opposite direction across the basolateral membrane remains to be confirmed.
Recently, it has been proposed that both SLC2A9 proteins are high-affinity glucose/fructose transporters. However, when compared with the principal members of the SLC2A9
gene family their transport capacity (Vmax
) is very low [11
]. We now have evidence that urate is a preferred substrate for both human SLC2A9 variants and for the mouse orthologue. The ability of SLC2A9 to exchange urate with glucose in the absence of competition between these two substrates when present on the same side of the membrane indicates that the protein has separate binding sites. This is not a unique phenomenon in exchange proteins, as the glycerol-6-phosphate transporter exchanges glycerol-6-phosphate for inorganic phosphate [35
This ability of SLC2A9 to exchange glucose and, to a lesser degree, fructose for urate may be physiologically important. The renal proximal nephron plays a major role in the reabsorption of glucose from the urine using a combination of sodium-coupled hexose transporters, SGLT1 and SGLT2, and members of the SLC2A family, GLUTs 2, 5, and possibly 9 [36
]. Furthermore, the proximal convoluted tubular epithelium is a major site of gluconeogenesis, converting pyruvate to glucose, which is then released across the basolateral membrane into the blood [41
]. Our data showing that SLC2A9a can exchange glucose for urate suggest that this protein might play an important role in the secretion of urate from the blood into the urine. Glucose in the urine could exchange for urate in the proximal convoluted tubule epithelial cells across the apical membrane, resulting in the release of urate back into the urine. In addition, glucose in the epithelial cells resulting both from reabsorption and neogenesis could exchange for plasma urate across the basolateral membrane, promoting the accumulation of urate in the cells. Furthermore, such a mechanism could explain the known correlation between the glycosuria seen in diabetes and the reduction in plasma urate levels. The increased glucose in the urine could accelerate the SLC2A9-mediated urate efflux across the apical membrane of the proximal convoluted tubule, but further work is needed to confirm this hypothesis.
Therefore, our findings, in combination with epidemiologic data showing correlation of elevated serum urate with diabetes, metabolic syndrome, obesity, and hyperinsulinaemia, provide a potential mechanism for these associations that warrants further investigation [44
]. In this context it is of particular interest that recent data from the Health and Nutrition Survey shows correlation between consumption of sugar-based soft drinks with serum urate levels and gout, which might be partly explained by sugar-facilitated uptake of urate by SLC2A9 isoforms [47
Probenecid and benzbromarone are uricosuric drugs that inhibit renal uptake of urate via URAT 1 on the apical proximal nephron membrane [5
]. We found that probenecid had no effect on urate uptake into SLC2A9a-expressing oocytes, whereas benzbromarone showed a dose-dependent inhibition. The significant inhibition of urate transport by 10 and 100 μM benzbromarone implies that there are common features in the binding sites for urate in both URAT1 and SLC2A9a. When in clinical use benzbromarone promoted the loss of urate in the urine; this is believed to be a consequence of an inhibition of uptake across the apical membrane mediated by URAT1 [2
]. In contrast, the high Ki
of 27 μM for benzbromarone action on SLC2A9 suggests that at therapeutic doses this drug would have a minimal effect on urate transport mediated by this protein and so does not affect urate secretion.
The Olivetti Heart Study has previously found a strong longitudinal relationship of serum urate levels with blood pressure [48
]. The authors also reported a correlation of urate with reduced fractional clearance of lithium as an index of sodium reabsorption in the proximal nephron. This finding could reflect sodium retention and offer a mechanism for previously reported associations of urate and blood pressure [12
]. The two SNPs genotyped in the Olivetti cohort validate association of the SLC2A9
locus with serum urate and demonstrate the association of both SNPs with reduced urinary urate excretion at two time points 8 y apart. This provides additional support for our hypothesis that SLC2A9
variants reduce urinary urate loss and is further confirmed by recent cross-sectional findings for this gene in gout [25
In view of the additive influence of alleles of SLC2A9 on serum uric acid and urinary urate excretion, we explored a relationship between SLC2A9 SNPs with systolic and diastolic blood pressure in 11,629 individuals under an additive model. This analysis showed no significant association between SNP rs13113918 and blood pressure. Furthermore, we did not find any evaluated association of the same SLC2A9 SNP with hypertension under an additive model in 5,249 hypertensive cases and 6,648 normotensive controls drawn from all the studies. In the context of this large-scale case–control and population-based cohorts we conclude that there is no support for an association with blood pressure using the most biologically plausible genetic model.
Limitations of This Study
We have not defined precisely the causative variant of SLC2A9 responsible for elevated serum urate and reduced urinary urate clearance. There are several known SNPs within the gene region that might influence function of the SLC2A9 protein. Such studies will be facilitated by detailed re-sequencing of the SLC2A9 gene to establish a comprehensive inventory of genetic variation across this locus. In addition to dietary and metabolic influences on uric acid levels there will be other, as-yet unidentified genetic influences on serum urate level that may contribute to epidemiologic correlations with metabolic syndrome, diabetes, gout, and cardiovascular disease.
In this paper we have translated the genetic association of the SLC2A9 locus with serum urate derived from genome-wide scanning into a functional confirmation that SLC2A9 splice variants acts as a high-capacity urate transporter that can be facilitated by exchange with hexoses and inhibited by high concentrations of some uricosurics and siRNA technology. These findings offer novel potential pathogenic mechanisms and new drug targets for diseases such as gout.