Article Text

Extended report
Population-specific influence of SLC2A9 genotype on the acute hyperuricaemic response to a fructose load
  1. Nicola Dalbeth1,
  2. Meaghan E House1,
  3. Gregory D Gamble1,
  4. Anne Horne1,
  5. Bregina Pool1,
  6. Lauren Purvis1,
  7. Angela Stewart1,
  8. Marilyn Merriman2,
  9. Murray Cadzow2,
  10. Amanda Phipps-Green2,
  11. Tony R Merriman2
  1. 1Department of Medicine, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand
  2. 2Department of Biochemistry, University of Otago, Dunedin, New Zealand
  1. Correspondence to Dr Nicola Dalbeth, Department of Medicine, Faculty of Medical and Health Sciences, University of Auckland, 85 Park Rd, Grafton, Auckland 1, New Zealand; n.dalbeth{at}auckland.ac.nz

Abstract

Background SLC2A9 is a strong genetic risk factor for hyperuricaemia and gout. SLC2A9 (GLUT9) is a high capacity urate transporter and reportedly transports glucose and fructose. Intake of fructose-containing beverages is associated with development of hyperuricaemia and gout.

Objective To determine whether genetic variation in SLC2A9 influences the acute serum urate response to a fructose load.

Methods Following an overnight fast, 76 healthy volunteers (25 Māori, 26 Pacific, 25 European Caucasian) drank a solution containing 64 g fructose. Serum and urine were obtained immediately before and then 30, 60, 120 and 180 min after ingestion. The SLC2A9 single nucleotide polymorphism (SNP) rs11942223 was genotyped and data were analysed based on the presence or absence of the gout protective minor allele (C).

Results The rs11942223 C allele was present in 17 participants (22%). In the entire group, fructose intake led to an increase in serum urate, which peaked 60 min following fructose ingestion (analysis of variance p=0.006). The presence of the C allele was associated with an attenuated hyperuricaemic response (p(SNP)<0.0001) and increased fractional excretion of uric acid (FEUA) (p(SNP)<0.0001) following the fructose load. The effects of rs11942223 variants on serum urate and FEUA in response to fructose were present only in Caucasian ancestral subgroups but not in the Māori and Pacific ancestral subgroup.

Conclusions Variation in SLC2A9 influences acute serum urate and FEUA responses to a fructose load. SLC2A9 genotype may influence the development of gout on exposure to fructose-containing beverages, particularly in European Caucasian populations.

  • Gout
  • Gene Polymorphism
  • Arthritis

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

Ingestion of fructose is associated with development of gout and hyperuricaemia.1–3 Intervention studies have shown that fructose ingestion can acutely increase serum urate concentrations.4–6 This has been attributed to hepatic metabolism of fructose which induces rapid depletion of ATP and increased urate production by degradation of preformed purine nucleotides.7 Fructose-induced hyperuricaemia is accentuated in patients with gout and first degree relatives of patients with gout. In a small European study of fructose-induced hyperuricaemia, increases in serum urate were significantly higher in children of those with gout compared with healthy controls without a family history of gout.4 These data suggest that fructose-induced hyperuricaemia has a genetic component in European Caucasian populations.

The genome-wide association studies of hyperuricaemia have consistently reported a strong association between a group of common variants of the SLC2A9 gene in strong linkage disequilibrium (LD) and serum urate concentrations.8 ,9 It is likely that these genetic variants mark a single functional effect. Case–control studies in Caucasians have reported ORs for hyperuricaemia for SLC2A9 variants ranging from 1.7–1.9 and 1.3–2.1 for gout.10 Māori and Pacific people have the highest reported prevalence of gout worldwide.11 In Māori and Pacific people, the ORs conferred by the major allele of the SLC2A9 variants for gout are even higher.12 The SLC2A9 gene encodes solute carrier family 2, member 9 protein (SLC2A9, also known as GLUT9), a high capacity urate transporter expressed in the proximal renal tubular cell and liver, which reportedly also transports glucose and fructose.13 In in vitro assays, the presence of fructose promotes urate transport via this receptor.13

Together, these observations raise the possibility that SLC2A9 variants may influence the increase in serum urate concentrations and promote the risk of gout on exposure to fructose. The aim of this study was to determine the influence of SLC2A9 variants on acute changes in serum urate and renal uric acid transport following ingestion of fructose in healthy volunteers.

Participants and methods

This acute intervention study was designed to examine the influence of SLC2A9 variation on acute fructose-induced hyperuricaemia. The primary endpoint was change in serum urate concentration to 3 h following a fructose challenge.

Participants

Seventy-six healthy participants were recruited by public advertisement. There were 25 participants of Māori ancestry, 26 participants of Pacific ancestry (all Western Polynesian; 17 Samoan, 8 Tongan, 1 Niuean), and 25 participants of European Caucasian ancestry. Participants were not included in the study if any of the following criteria were present; history of gout, history of diabetes mellitus, history of fructose intolerance, diuretic use, and fasting capillary glucose >6 mmol/l. As the primary goal of this study was to examine the effect of SLC2A9 variants on the hyperuricaemic response to fructose, we elected to exclude patients with gout, noting that most patients with established gout will have an increased hyperuricaemic response to the fructose challenge,4 and that the frequency of the protective allele at rs11942223 will be considerably lower.12

Potential participants attended a screening visit where a general health questionnaire was completed and baseline measurements and physical examination were performed. Screening blood tests were also obtained. No prospective participants were excluded following the screening visit. The study was approved by the New Zealand Ministry of Health Multiregional Ethics Committee, and each participant gave written informed consent. The study was registered by the Australian Clinical Trials Registry (ACTRN12610001036000).

Protocol

The study visit occurred within 2 weeks of the screening visit. At the study visit, a venous catheter was inserted for blood collection. Following an overnight fast, participants consumed a sugar solution between 08:00 and 09:00, and blood was obtained for urate, creatinine and glucose, prior to ingestion and then 30 min, 60 min, 120 min and 180 min after ingestion. Blood was also obtained for DNA extraction. Urine volume was measured and urine was obtained at each time point for testing of urate and creatinine.

The sugar solution of 300 kcal/300 ml was consumed within 10 min, according to the protocol of Akhaven and Anderson for fructose-induced hyperuricaemia.5 This solution contained 80% fructose and 20% glucose (64 g fructose and 16 g glucose).

Laboratory testing

Serum and urine chemistry was tested using the Roche Modular P (Hitachi) analyser. The fractional excretion of uric acid (FEUA) was calculated; this is the ratio between the renal clearance of uric acid and the renal clearance of creatinine, expressed as a percentage. Genotyping of SLC2A9 single nucleotide polymorphism (SNP) rs11942223 was done using TaqMan SNP genotyping assay technology (Applied Biosystems).12 L-lactate and fructose concentrations in the 60-min serum samples were measured using fluorescence-based assays (both Cayman Chemical Company, Ann Arbor, Michigan, USA) according to the manufacturer's instructions. This time point was chosen as both serum lactate and fructose concentrations peak 60 min after intake of an oral fructose load.14

Statistical analysis

The analysis plan for this study specified change in serum urate concentration as the primary endpoint. The key secondary endpoint was change in FEUA. Sample size calculations were based on the previous study of fructose-induced hyperuricaemia by Stirpe et al.4 This study showed significant differences between healthy volunteers with and without a family history of gout with 5–6 participants per group (all of European Caucasian ancestry). We selected the SLC2A9 SNP rs11942223 for primary analysis, and assumed that approximately 30% of participants would have the protective C allele based on our previous work.12 Subgroup power calculations are shown in the online supplementary text. Sample size calculations were performed using PASS 2002 (J Hintze, 2006, Kaysville, Utah).

Data are presented as mean (SD) or n (%) for descriptive purposes; however, measures of effect are presented with the appropriate 95% CI. The primary analysis was a comparison of the hyperuricaemic response to a fructose challenge in the entire group (n=76), based on the presence or absence of the rs11942223 protective C allele. Secondary and exploratory analyses were comparison of the FEUA following a fructose challenge in the entire group (n=76), based on the presence or absence of the rs11942223 protective C allele, and comparison of the hyperuricaemic response to a fructose challenge based on subgroup analysis (within each ancestral subgroup).

Data were analysed using a mixed models approach to repeated measures. Significant main and interaction effects were explored using the method of Tukey. Where the interaction term was not statistically significant, pairwise comparison p values are presented for completeness. Where indicated, sex and ancestry were adjusted for within the models. For change in serum urate, a mixed models analysis of covariance (ANCOVA) approach to repeated measures was used. For ANCOVA, the dependent variable was change from baseline, and baseline level was included as a covariate in the analysis. Differences in participant characteristics between those with and without the rs11942223 protective C allele were analysed using t tests for normally distributed data and Fisher's exact tests. Mixed model one way analysis of variance with Dunnett's multiple comparison test was used to analyse changes from baseline over time in the entire group. The percentage of participants in whom serum urate increased above saturation levels (≥0.41 mmol/l) was also analysed in the entire group (n=76), using generalised estimating equations logistic regression with unadjusted pairwise comparisons to compare between groups at various time points; 95% CIs were calculated using Wilson's methods. All analyses were performed using SAS V.9.2; p<0.05 was considered significant and all tests were two-tailed.

Results

Participant characteristics

Overall, the mean (SD) age of the participants was 31 (16) years, and 45 (59%) were male. The rs11942223 protective C allele was present in 17/76 participants (22%). The C allele was present in 8/25 (32%) European Caucasian participants, and in 8/25 (32%) Māori participants. Only one (4%) participant of Pacific ancestry had the protective C allele (p=0.02 compared with other ancestral subgroups). Therefore, for all further analysis, the Māori and Pacific ancestral subgroups were pooled. At baseline, the serum urate and FEUA differed between those with and without the protective C allele (table 1). However, after adjustment for sex and ancestry, there was no difference in serum urate and FEUA between those with and without the protective C allele (p=0.11 and 0.15, respectively).

Table 1

Characteristics of participants at baseline*

The effects of a fructose load in the entire group

In the entire group (n=76), fructose ingestion lead to a rapid increase in serum glucose, which peaked at 30 min and reduced below baseline at 120 min (figure 1A). Serum urate also increased acutely following the fructose load, peaking at 60 min and gradually reducing to baseline over the observation period (figure 1B). FEUA increased significantly from baseline 180 min after fructose intake (figure 1C).

Figure 1

The effects of a fructose load in the entire group. (A) Serum glucose. (B) Serum urate. (C) Fractional excretion of uric acid (FEUA). Dunnett's post hoc p values refer to comparison to baseline value. Data are presented as mean (95% CI).

The effect of ancestry on serum urate, serum glucose, and FEUA responses to a fructose load

In all ancestral subgroups, serum urate concentration increased in response to a fructose load (figure 2A,B). Serum urate concentrations were higher at baseline in the Māori and Pacific ancestral subgroup compared with the European Caucasian subgroup (figure 2A). However, following a fructose load, no significant differences were observed in serum urate concentrations between these two groups at subsequent time points (figure 2A). Analysis of the change in serum urate concentration following a fructose load indicated that the overall increase in serum urate was higher in the European Caucasian subgroup than the Māori and Pacific ancestral subgroup (sex adjusted p(ancestry)=0.002, figure 2B). At all time points measured, more participants in the Māori and Pacific ancestral subgroup had serum urate concentrations above saturation levels (sex adjusted p(ancestry)=0.0001, figure 2C).

Figure 2

The effect of ancestry on serum urate concentrations, serum glucose and fractional excretion of uric acid (FEUA) following a fructose load in the entire group. (A) Serum urate concentration. (B) Change in serum urate concentration. (C) Percentage (95% CI) of participants with serum urate above saturation level (≥0.41 mmol/l). (D) Serum glucose concentration. (E) FEUA. Unless stated, data are presented as mean (95% CI). Pairwise comparisons p values refer to the comparison between the different ancestral subgroups at each time point. Sex adjusted p values are shown throughout.

The overall serum glucose response to the fructose load was higher in the Māori and Pacific ancestral subgroup (sex adjusted p(ancestry)=0.0005, figure 2D). Serum fructose concentrations were lower in the Māori and Pacific ancestral subgroup at 60 min (sex adjusted p(ancestry)=0.040), with no difference in lactate concentrations (see online supplementary figure S1). There were also differences in renal handling of uric acid following the fructose load between the ancestral subgroups, with higher FEUA observed in the European Caucasian subgroup at all time points from 60 min following fructose ingestion (sex adjusted p(ancestry)<0.0001, figure 2E).

The effect of SLC2A9 genotype on serum urate concentration following a fructose load in the entire group: primary endpoint

Following the fructose load, serum urate increased in those with and without the rs11942223 protective C allele in the entire study group (figure 3A). Participants with the rs11942223 protective C allele had lower serum urate concentrations at all time points after fructose intake, compared to participants without the protective C allele (sex and ancestry adjusted p(SNP)<0.0001, figure 3A). Furthermore, the change in serum urate was lower in participants with the protective allele at all time points following fructose intake (sex and ancestry adjusted ANCOVA p(SNP)<0.0001, figure 3B). Participants with the protective C allele were less likely to have a serum urate concentration above saturation levels (≥0.41 mmol/l) at baseline and throughout the study period following a fructose load (sex and ancestry adjusted p(SNP)=0.0001, figure 3C).

Figure 3

The effect of SLC2A9 genotype on serum urate concentrations, serum glucose and fractional excretion of uric acid (FEUA) following a fructose load in the entire group. (A) Serum urate concentration. (B) Change in serum urate concentration. (C) Percentage (95% CI) of participants with serum urate above saturation level (≥0.41 mmol/l). (D) Serum glucose concentration. (E) FEUA. Unless stated, data are presented as mean (95% CI). Pairwise comparisons p values refer to the comparison between those with and without the protective C allele at each time point. Sex and ancestry adjusted p values are shown throughout.

The effect of SLC2A9 genotype on serum glucose and FEUA following a fructose load in the entire group: secondary endpoints

No difference was observed in the glucose response to the fructose load between participants with and without the rs11942223 protective C allele (sex and ancestry adjusted p(SNP)=0.37, figure 3D). Similarly, serum fructose and lactate concentrations at 60 min were not influenced by the presence of the rs11942223 protective C allele (see online supplementary figure S1). However, those with the protective allele had a greater increase in the FEUA in response to the fructose load, compared with those without the protective allele (sex and ancestry adjusted p(SNP)<0.0001, figure 3E). This effect was most apparent 180 min after fructose intake. Consistent effects of SLC2A9 genotype on serum urate and FEUA were observed in those with creatinine clearance values above and below the median value of 130 ml/min (see online supplementary figure S2). The effects of SLC2A9 genotype on serum urate and FEUA responses were observed in those participants with BMI ≥25 kg/m2, but not in those with BMI <25 kg/m2 (see online supplementary figure S3).

The effect of SLC2A9 genotype on serum urate concentration and FEUA following a fructose load in ancestral subgroups: prespecified subgroup analysis

In the European Caucasian participants, the presence of the rs11942223 protective C allele was associated with a markedly attenuated hyperuricaemic response and increased FEUA following the fructose load (sex adjusted p(SNP)<0.0001 for all) (figure 4). In contrast, the SLC2A9 genotype did not influence the change in serum urate concentrations (p(SNP)=0.13) in response to fructose in the Māori and Pacific ancestral subgroup, although there was a trend for a genotype difference in serum urate concentrations per se (p(SNP)=0.063). The presence of the protective C allele was associated with an increased FEUA after 180 min following fructose intake, but at no other time points. The presence of the protective C allele did not influence the glucose response to fructose in either ancestral subgroup (p(SNP)>0.44 for both).

Figure 4

The effect of SLC2A9 genotype on serum urate concentration and fractional excretion of uric acid (FEUA) following a fructose load in each ancestral subgroup. (A) Serum urate concentrations. (B) Change in serum urate concentrations. (C) FEUA. Top panel shows European Caucasian subgroup, bottom panel shows Māori and Pacific subgroup. Data are presented as mean (95% CI). Pairwise comparison p values refer to the comparison between those with and without the protective C allele at each time point. Sex adjusted p values are shown throughout.

Discussion

This study has demonstrated that variation in SLC2A9 genotype can influence both serum urate and FEUA responses to a fructose load. This effect was strongly observed in the European Caucasian ancestral subgroup, but not in the Māori and Pacific subgroup, which is of particular interest, given the high rates of hyperuricaemia and gout that is well documented in Polynesian populations,11 ,15–17 and the observation that Polynesian people have reduced FEUA.18 ,19

The observation that rs11942223 variants influence the acute serum urate responses to a fructose load suggests a link between SLC2A9 genotype and exposure to fructose-containing beverages in the development of gout. This concept is further supported by the finding that participants without the protective allele were more likely to have a sustained serum urate concentration above saturation levels at baseline and following fructose intake. The underlying mechanism for the effects of SLC2A9 variants on fructose-induced hyperuricaemia may be multifactorial; our data showed early differences in serum urate concentrations consistent with increased production of urate, and also later differences in the FEUA. No genotype-specific differences were observed in the serum glucose, fructose, or lactate response to fructose loading, suggesting that the effects are probably not related to fructose absorption or a global alteration in fructose metabolism.

This study has also identified differences between ancestral subgroups in the urate response to a fructose load. Although both European Caucasian and Māori and Pacific subgroups increased serum urate concentrations in response to the fructose load, this effect was less prominent in the Māori and Pacific subgroup. This is despite reduced clearance of uric acid in this group as identified by the FEUA data. A potential explanation for these observations is that the serum urate response to fructose may be limited by a threshold effect, and that the higher baseline serum urate in the Māori and Pacific subgroup limited further increases in serum urate. Alternatively, there may be differences in hepatic metabolism of fructose between ancestral groups, a possibility that is supported by the serum glucose and fructose analysis. It should be noted that despite the reduced serum urate response to the fructose load in the Māori and Pacific ancestral subgroup, fructose loading did lead to increases in serum urate above saturation in almost two thirds of this group, suggesting that fructose intake may still be an important risk factor for gout in Māori and Pacific populations.

A key observation of this study was that the influence of SLC2A9 on serum urate concentrations was limited to the European Caucasian ancestral subgroup. These results are consistent with previous reports that the hyperuricaemic response to fructose is accentuated in first degree relatives of European Caucasian patients with gout, and provide a potential mechanism for this observation.4 It is unclear why these relationships were not observed in the Māori and Pacific ancestral subgroup, particularly since our previous work has shown that SLC2A9 genotype (including rs11942223 and other SLC2A9 SNPs) is also associated with risk of gout in these populations.12 Potential explanations may include other population-specific genetic factors influencing either urate production or excretion that may interact epistatically with SLC2A9, non-genetic differences in renal uric acid transporter function, or other factors such as insulin resistance. It is possible that there is a Polynesian-specific genetic variant in LD with rs11942223 that encodes a functional effect that overrides the genotype-specific FEUA effect seen in European Caucasians.

Limitations of this study include the relatively small sample size, which limits the ability to examine the effects of multiple covariates in detail. The study was powered based on our previous case–control study of SLC2A9 variants in gout,12 which reported a higher frequency of the rs11942223 protective C allele than was observed in this study (30% vs 22%). Nevertheless, clear differences were observed over time and between groups. We did observe more men in the group without the protective allele. This may have influenced our results as men have higher baseline serum urate concentrations. Importantly, sex was included as a covariate in all analyses to address this potential confounder. This study was designed to examine the acute effects of a high dose of fructose on serum urate concentrations, and it is possible that SLC2A9 may also influence serum urate responses to long-term fructose intake in a different manner.

In summary, this study has shown that variation in SLC2A9 influences the acute serum urate response to fructose and the renal clearance of uric acid following a fructose load in people of European Caucasian ancestry. This effect was not observed in those ancestral subgroups at very high risk of hyperuricaemia and gout. These data suggest a complex interaction between known risk factors for gout, including specific genetic variants in urate transporters, ancestry and environmental factors.

Acknowledgments

This study was funded by the Health Research Council of New Zealand.

References

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:

Footnotes

  • Handling editor Tore K Kvien

  • Contributors ND (the guarantor) accepts full responsibility for the work and the conduct of the study, had access to the data, and controlled the decision to publish. ND conceived of the study, contributed to the data interpretation, and drafted the manuscript. MEH, AH, LP and AS recruited participants and coordinated study visits. MEH also managed clinical data entry. BP completed laboratory testing. MM, MC and APG contributed to data acquisition and genetic data entry. GDG analysed the data. TRM conceived of the study, contributed to the data interpretation and drafted the manuscript. All authors read and approved the final manuscript.

  • Funding Health Research Council of New Zealand, grant no. 08/075.

  • Competing interests None.

  • Patient consent Obtained.

  • Ethics approval New Zealand Ministry of Health Multiregional Ethics Committee.

  • Provenance and peer review Not commissioned; externally peer reviewed.