Research ArticleGenetics

Common Defects of ABCG2, a High-Capacity Urate Exporter, Cause Gout: A Function-Based Genetic Analysis in a Japanese Population

See allHide authors and affiliations

Science Translational Medicine  04 Nov 2009:
Vol. 1, Issue 5, pp. 5ra11
DOI: 10.1126/scitranslmed.3000237


Gout based on hyperuricemia is a common disease with a genetic predisposition, which causes acute arthritis. The ABCG2/BCRP gene, located in a gout-susceptibility locus on chromosome 4q, has been identified by recent genome-wide association studies of serum uric acid concentrations and gout. Urate transport assays demonstrated that ABCG2 is a high-capacity urate secretion transporter. Sequencing of the ABCG2 gene in 90 hyperuricemia patients revealed several nonfunctional ABCG2 mutations, including Q126X. Quantitative trait locus analysis of 739 individuals showed that a common dysfunctional variant of ABCG2, Q141K, increases serum uric acid. Q126X is assigned to the different disease haplotype from Q141K and increases gout risk, conferring an odds ratio of 5.97. Furthermore, 10% of gout patients (16 out of 159 cases) had genotype combinations resulting in more than 75% reduction of ABCG2 function (odds ratio, 25.8). Our findings indicate that nonfunctional variants of ABCG2 essentially block gut and renal urate excretion and cause gout.


Gout is a common disease resulting from tissue deposition of monosodium urate crystals as a consequence of hyperuricemia, which shows elevated serum uric acid (SUA) concentrations (1, 2) and has long been known to have a heritable component (3). Since Hippocrates first reported the disease about 2500 years ago, humans have long struggled with gout; for example, the Holy Roman Emperor Charles V (4, 5), Sir Isaac Newton, Charles Darwin, and Leonardo da Vinci (6) were all affected by gout. The prevalence of gout and hyperuricemia is about 1 to 2% and 10 to 25% in males, respectively, in Japan (7) and other countries (8, 9). In men over the age of 65 years, gout prevalence is approaching 7% (9). The prevalence of gout and hyperuricemia is now increasing, and therefore, many more individuals will carry increased risks of related disorders (2), such as hypertension (10, 11), cerebrovascular diseases (12, 13), and kidney diseases (14, 15).

We have previously identified a urate transporter gene designated URAT1/SLC22A12, which shows kidney-specific expression by the candidate gene approach using a human genome database (16). Another urate transporter gene, glucose transporter 9 (GLUT9/SLC2A9), has also been identified by genome-wide association studies (GWAS) of SUA (1719). So far, we have demonstrated that loss-of-function mutations in two urate transporter genes, URAT1 and GLUT9, cause renal hypouricemia type 1 [Mendelian Inheritance in Man (MIM) 220150] (16) and type 2 (MIM612076) (20), respectively. These findings, together with their renal expression patterns, also showed that URAT1 and GLUT9 physiologically mediate renal urate reabsorption in humans (16, 20).

The adenosine 5′-triphosphate (ATP)–binding cassette (ABC), subfamily G, member 2 gene ABCG2/BCRP locates in a gout-susceptibility locus (MIM 138900) on chromosome 4q (21), which was demonstrated by a genome-wide linkage study of gout (21). Recent GWAS of SUA and gout also identified several genes, including GLUT9 and ABCG2 (22, 23). ABCG2 encodes a multispecific transporter that is expressed on the apical membrane in several tissues, including intestine, liver (24), and kidney (25). Besides its transport of nucleotide analogs (2628) that are structurally similar to urate, we have reported that ABCG2 is an exporter that has polymorphic reduced functionality or nonfunctional variants (29). These findings suggest that ABCG2 could be a urate secretion transporter gene (30, 31) and thus be a promising candidate gene for gout. Here, we show that ABCG2 is a high-capacity urate secretion transporter and that its common variants increase SUA. On the basis of these results, and to identify pathogenic nonfunctional mutations of ABCG2, we further performed a function-based genetic analysis (fig. S1) of the ABCG2 gene in gout patients in a Japanese population.


High-capacity urate transport via ABCG2

Using membrane vesicles prepared from ABCG2-expressing human embryonic kidney 293 (HEK293) cells (32), we examined the inhibitory effect of urate on ABCG2-mediated transport of its typical substrate estrone-3-sulfate (ES) labeled with 3H (33). Although urate required a higher concentration than did unlabeled ES to inhibit [3H]ES transport via ABCG2, the potency of urate was similar to that of the previously reported substrate, 3′-azido-3′-deoxythymidine (AZT) (26) (Fig. 1A). To test whether urate is a substrate of ABCG2, transport assays were performed with isotope-labeled [14C]urate. ATP-dependent transport of urate was detected in ABCG2-expressing vesicles but not in control vesicles (Fig. 1B). Because urate had a lower potency in inhibiting ES transport, it was assumed to be a high-capacity (low-affinity) substrate of ABCG2 compared with typical ABCG2 substrates—for example, sulfate conjugates such as ES, 4-methylumbelliferone sulfate, and E3040 sulfate—transported with low capacity [Michaelis constant (Km) values of ~20 μM] (33). Kinetic analysis revealed that ABCG2 mediated the high-capacity transport of urate, maintaining its function even under high-urate conditions (Fig. 1C). The calculated parameters of ABCG2-mediated transport of urate were a Km of 8.24 ± 1.44 mM and a Vmax (maximum velocity) of 6.96 ± 0.89 nmol/min per milligram of protein (Fig. 1C). The calculated Km value exceeded the highest concentration in the experimental condition. This is due to the low-solubility limitation of urate, a property related to the monosodium urate crystal formations in gout patients. Our data indicate that ABCG2 could play a physiological role as a high-capacity urate exporter.

Fig. 1

High-capacity urate transporter function of ABCG2. (A) Urate inhibits ABCG2-mediated transport. Vesicles prepared from HEK293 cells expressing ABCG2 were incubated with 500 nM 3H-labeled ES plus the indicated inhibitors or unlabeled ES with or without ATP. The amount of 3H-labeled ES was measured after 1 min. (B) ABCG2-mediated urate transport. ATP-dependent transport of 14C-labeled urate was detected in ABCG2-expressing vesicles but not in control vesicles after indicated periods. (C) ABCG2 transports urate with high capacity. Concentration dependence of ABCG2-mediated transport of 14C-labeled urate was detected with 5-min incubation. All results are expressed as means ± SD.

Nonsynonymous ABCG2 mutations in hyperuricemia patients

To find candidate variants for the causal ABCG2 mutations in gout, we performed mutation analysis of all coding regions and intron-exon boundaries of the ABCG2 gene (table S1 and fig. S2) in 90 Japanese patients with hyperuricemia. The following six nonsynonymous mutations were found: V12M, Q126X, Q141K, G268R, S441N, and F506SfsX4 (Table 1). The first three mutations are single-nucleotide polymorphisms (SNPs). Maekawa et al. reported that allele frequencies for these SNPs, which are quite common in the Japanese population, were 31.9% for Q141K, 19.2% for V12M, and 2.8% for Q126X, respectively (Table 1) (34). Using Hardy-Weinberg equilibrium and these data of a Japanese population reported by Maekawa et al. (34), we calculated estimates of the frequencies of Japanese individuals with these minor alleles to be 53.6% for Q141K, 34.7% for V12M, and 5.5% for Q126X. Figure 2A shows the location of these variants in a topological diagram of human ABCG2 (35) in the intracellular N-terminal region. The results of our sequence analysis of ABCG2 are shown in fig. S3.

Table 1

Nonsynonymous mutations in ABCG2 found by mutation analysis of 90 hyperuricemia patients. NCBI, National Center for Biotechnology Information; N.D., not detected.

View this table:
Fig. 2

ABCG2 mutations and their impairment of transporter function. (A) The topological model of ABCG2 and six nonsynonymous mutation sites (magenta) found in hyperuricemic patients. #, N-linked glycosylation site (N596); *, cysteine residues for disulfide bonds (C592, C603, and C608). (B) Urate transport analysis of mutated ABCG2. Vesicles prepared from HEK293 cells expressing the wild-type or variants of ABCG2 were incubated with 14C-labeled urate with or without ATP. The amount of 14C-labeled urate was measured after 5 min. Results are expressed as means ± SD. (C to E) QTL analysis of Q141K and SUA concentrations was performed in 739 Japanese individuals (C), including 245 male subjects (D) and 494 female subjects (E). “C/C,” “C/A,” and “A/A” indicate wild-type subjects, heterozygous mutation carriers, and homozygous mutation carriers of Q141K, respectively. Results are expressed as means ± SE.

Loss of function of urate transport in ABCG2 mutants

To clarify the effect on ABCG2 function, the urate transport capacity of the six variant proteins was examined and compared with that of wild-type ABCG2. ATP-dependent transport of urate was reduced by approximately half (46.7%) in Q141K and was nearly eliminated in Q126X, G268R, S441N, and F506SfsX4 mutants (Fig. 2B). Western blot analysis showed that ABCG2 protein expression levels in the Q141K variant decreased by half (47.2%), whereas Q126X resulted in no protein on membrane vesicles (fig. S4). The decreased activity of Q141K is probably a result of decreased amounts of ABCG2 protein, consistent with our previous study on ES transport (29). The V12M variant did not show any changes in urate transport or in protein amounts relative to wild-type ABCG2. Our data show that a decrease in ABCG2 protein expression levels is directly proportional to a decrease in the urate transport activity.

Common variant of ABCG2 increases SUA concentrations

Quantitative trait locus (QTL) analysis of SUA was performed with the high-frequency dysfunctional variant Q141K in ABCG2 in a random sample of 739 Japanese individuals. The analysis revealed that SUA significantly increased as the number of minor alleles of Q141K increased (P = 6.60 × 10−5), and when adjusted for sex, the corrected P value is 2.02 × 10−6 (Fig. 2C). A significant increase in SUA was also observed in both male (P = 0.0144) (Fig. 2D) and female subjects (P = 0.0137) (Fig. 2E). Different from SUA, Q141K had no significant association with other clinical parameters such as age, body mass index, or sex (table S2). These findings indicate that ABCG2 controls SUA in vivo and that there could be great interindividual differences in this function because of its polymorphic nature.

Additional genotyping and association analysis of gout

Through additional genotyping of ABCG2 SNPs for 228 Japanese men with hyperuricemia (including 161 men with gout), Q126X homozygous (n = 2) and heterozygous (n = 24) mutations were identified (Table 2). Two patients with Q126X homozygous mutations showed very high SUA (>10 mg/dl) before they were treated for hyperuricemia. For association studies, 871 Japanese men (SUA ≤ 7.0 mg/dl) were also genotyped as controls. The association study showed that Q126X increased the risk of hyperuricemia [odds ratio (OR), 3.61; 95% confidence interval (CI), 2.14–6.08; P = 2.91 × 10−7]. Among the 161 patients with gout, Q126X homozygous (n = 1) and heterozygous (n = 21) mutations were found, which revealed that Q126X dramatically increased gout risk (OR, 4.25; 95% CI, 2.44–7.38; P = 3.04 × 10−8). The partially functional SNP Q141K also increased gout risk (OR, 2.23; 95% CI, 1.75–2.87; P = 5.54 × 10−11). The call rate, or the ability of the SNP to be reliably decoded, for V12M, Q126X, and Q141K was 98.8%, 100%, and 99.2%, respectively. P values for Hardy-Weinberg equilibrium of V12M, Q126X, and Q141K were 0.08, 0.72, and 0.01, respectively. P values that suggested mistyping were not obtained.

Table 2

Association analysis of ABCG2 variants in gout patients. MAF, minor allele frequency.

View this table:

Haplotype frequency analysis revealed that there is no simultaneous presence of the minor alleles of Q126X and Q141K in one haplotype (Table 3). The haplotype with Q126X is present in up to 13.5% of gout patients, and it markedly increases gout risk (OR, 5.97; 95% CI, 3.39–10.51; P = 4.10 × 10−12) compared with nonrisk haplotypes. Q141K is assigned to another independent risk haplotype. Our data also show enrichment of the Q126X minor allele in gout or hyperuricemia patients relative to normouricemic subjects (SUA ≤ 7.0 mg/dl). Thus, the Q126X mutation of the ABCG2 gene is identified as a major cause of primary gout. Together, these findings suggest that nonfunctional variants of ABCG2, such as Q126X, essentially block urate excretion and cause gout. Our findings showed that V12M is exclusively assigned to a nonrisk haplotype (Table 3) and that the V12M variant does not exhibit altered urate transport activity (Fig. 2B). These findings may help explain why V12M decreases gout risk (OR, 0.68; 95% CI, 0.49–0.94; P = 0.02) (Table 2).

Table 3

Haplotype frequency analysis of V12M, Q126X, and Q141K. OR is obtained by comparing with the nonrisk haplotypes GCC and ACC. Risk alleles for Q126X and Q141K are underlined.

View this table:

Because Q126X and Q141K are assigned to different risk haplotypes, nonfunctional and half-functional haplotype, respectively, their genotype combinations are divided into four groups on the basis of the estimated ABCG2 transport functions, that is, full function, ¾ function, ½ function, and ≤¼ function (Table 4). Gout risk of “¾ ABCG2 transport function” was increased with an OR of 3.02 (95% CI, 1.96–4.65; P = 2.29 × 10−7) and that of ½ function was increased with an OR of 4.34 (95% CI, 2.61–7.24; P = 2.23 × 10−9). A remarkable increase in gout risk was observed in genotype combinations of ≤¼ function (OR, 25.8; 95% CI, 10.3–64.6; P = 3.39 × 10−21) and up to 10.1% of gout patients had these genotypes. In contrast, only 0.9% of Japanese males (SUA ≤ 7.0 mg/dl) have the same genotype combinations (Fig. 3). In addition, genotype combinations of full function are detected in 50.8% of the normouricemic subjects but only in 21.4% of gout patients who may have other gout risks. These findings suggested that combinations of nonfunctional and partially functional variants are important for the development of gout, thereby providing evidence for a common disease caused by common nonfunctional variants and their combinations.

Table 4

Association analysis of ABCG2 genotype combination in gout patients. OR is obtained by comparing with nonrisk genotype combination C/C(Q126X) and C/C(Q141K). Risk alleles for Q126X and Q141K are underlined.

View this table:
Fig. 3

Relation between ABCG2 transport dysfunction and gout. The genotype combinations of Q126X and Q141K are divided into several groups based on estimated ABCG2 transport functions. The Q126X homozygous and heterozygous mutations were identified in up to 13.5% of total gout patients (n = 161). Up to 10.1% of total gout patients have genotype combinations resulting in ≤25% function, whereas the asymptomatic carriers of these genotype combinations, who would have possible risk of gout, were only 0.9% of the normal population (n = 865).


Here, we identified nonfunctional mutations of ABCG2, which encodes a high-capacity urate secretion transporter, in more than 10% of Japanese gout patients. We also demonstrated that nonfunctional ABCG2 mutation Q126X is assigned to the identical haplotype, which increases gout risk, conferring an OR of 5.97. Importantly, we found that some genotype combinations remarkably decrease ABCG2 function (≤25% of control). These dysfunctional genotype combinations markedly increase gout risk, conferring an OR of 25.8. Our findings indicate that dysfunctional genotype combinations of ABCG2 are major causes of gout.

ABCG2 protein—a high-capacity transporter for urate excretion

ABCG2, also known as breast cancer resistance protein (BCRP), belongs to the ABC transporter superfamily. Together with P-glycoprotein (ABCB1) (36, 37) and multidrug resistance–associated proteins (ABCCs) (3840), ABCG2 is a well-known multispecific transporter that is expressed on the plasma membrane where it mediates cellular extrusion of various compounds in an ATP-dependent manner. Because of its polymorphic nature in humans (29, 34, 41), clinical studies on the pharmacokinetics, efficacy, and toxicity of the substrate xenobiotics of ABCG2, such as rosuvastatin (42) and gefitinib (43), have been reported. Besides these clinically used drugs, several compounds are good substrates of ABCG2; sulfate conjugates of endogenous and exogenous chemicals are high-affinity substrates of ABCG2 (33), and the ABCG2-mediated excretion of dietary carcinogens may reduce cancer susceptibility (44). Besides its ability to export porphyrins (45), regulation of SUA through urate secretion may be an essential physiological role of ABCG2 in humans.

Generally, SUA concentrations in humans are higher than those in most other mammals because humans lack the uric acid–degrading enzyme hepatic uricase (46, 47). Higher SUA in humans is reported to be associated with a longer life span relative to those of other mammals because uric acid is biologically active as an antioxidant (48). On the other hand, uric acid induces inflammation and oxidative stress (2), suggesting that uric acid in humans can have both beneficial and harmful characteristics. Because uric acid metabolism is considerably different in humans and mice, the study of patients with loss-of-function mutations in the ABCG2 gene can provide advantages over abcg2 gene-deficient mouse models and help to analyze the physiological roles of ABCG2-mediated urate transport.

Two-thirds of the uric acid in the human body is normally excreted through the kidney (30), whereas one-third gains entrance to the gut where it undergoes uricolysis (decomposition of uric acid) (30, 31). In the human kidney, urate is bidirectionally reabsorbed and secreted via urate transporters. Transporters for renal urate reabsorption, URAT1 and GLUT9, have been identified by us previously (16, 20).

Expression of urate exporter ABCG2 detected on the apical side of the proximal tubular cells in human kidneys, and on the apical sides of enterocytes and hepatocytes in the human intestine and liver, has been reported (24, 25). These findings could suggest a role for ABCG2 in not only renal urate excretion but also gut urate excretion via intestinal and biliary secretion in humans (fig. S5).

ABCG2 gene—major causative gene for gout

Most genetic findings on gout have been derived from rare Mendelian disorders (21). For example, rare familial diseases with abnormal uric acid metabolism include hypoxanthine guanine phosphoribosyltransferase deficiency, including Lesch-Nyhan syndrome (MIM 300322) (49) and Kelley-Seegmiller syndrome (MIM 300323) (50), phosphoribosylpyrophosphate synthetase superactivity (MIM 300661) (51, 52), and familial juvenile hyperuricemic nephropathy (MIM 162000) (53, 54). The approach reported here (fig. S1) revealed that nonfunctional variants of ABCG2, such as Q126X, essentially block urate excretion and cause gout. In contrast to the rare Mendelian disorders described above, the disease haplotype carrying the ABCG2 Q126X mutation (OR, 5.97; 95% CI, 3.39–10.51; P = 4.10 × 10−12) is present in as many as 13.5% of primary gout patients in our population, suggesting that Q126X is a major causative mutation for gout. We also found that 10.1% of gout patients have combinations of mutations in ABCG2 that decreased the ABCG2 transport function (≤25% of control), which remarkably increase gout risk (OR, 25.8; 95% CI, 10.3–64.6; P = 3.39 × 10−21). Thus, the more severe the ABCG2 transport dysfunction, the more likely the individual is to develop gout.

Chen et al. previously identified a genetic locus harboring a gout-susceptibility gene on chromosome 4q through a genome-wide linkage study of 21 multiplex pedigrees with gout from an aboriginal tribe in Taiwan (21). Because the Pacific Austronesian population, including Taiwanese aborigines, has a remarkably high prevalence of gout and hyperuricemia, there may be a founder effect across the Pacific region (21). Because the ABCG2 gene is located in the same locus on chromosome 4q, these findings imply that ABCG2 could be a major causative gene not only for Japanese gout patients but also for those in other Pacific regions.

Dehghan et al. recently reported the association of several genetic loci, including those of GLUT9 and ABCG2, with SUA in a GWAS (22). They also found that a missense SNP (Q141K) was associated with self-reported gout in Caucasians (OR, 1.74; 95% CI, 1.51–1.99) (22), which is consistent with our finding from clinically diagnosed gout patients from a Japanese population. A meta-analysis of GWAS of European descent also showed that the nine loci, including GLUT9 and ABCG2, influence SUA (23), which confirms the findings of Dehghan et al. (22). In this study, we identified several nonfunctional ABCG2 mutations, including Q126X, in Japanese gout patients, which shows stronger effects on gout development than Q141K did in a previous study (OR < 2.0) (22). There is a possibility that these nonfunctional mutations, including Q126X, are specific to a Japanese or Asian population, or alternatively, there may be nonfunctional mutations in the ABCG2 gene that cause gout in Caucasians and other groups. Further experiments in non-Japanese gout patients are required to clarify the presence of such nonfunctional mutations. Additionally, the urate efflux via ABCG2 expressed on oocytes has also been recently reported (55), consistent with our findings using the ABCG2-expressing vesicle system. We and Woodward et al. (55) independently found an ability of ABCG2 to transport urate and characterized the effects of a partially functional variant Q141K using different methods. Functional studies using a vesicle system also enabled us to show the ATP dependence and urate concentration dependence of ABCG2-mediated urate transport. In this study, we also showed the relation between the protein expression levels and the ABCG2-mediated urate transport function (fig. S4), as demonstrated in our previous study on ABCG2-mediated ES transport (29). Collectively, our comprehensive analysis shows that ABCG2 is a high-capacity urate exporter and that nonfunctional mutations of ABCG2 cause gout.

Recently, Nejentsev et al. showed that rare variants of the IFIH1 gene, including a nonsense mutation, protect against type 1 diabetes (OR, 0.51-0.74) by resequencing study (56). They demonstrated that such resequencing studies can pinpoint disease-causing genes in genomic regions initially identified by GWAS, and they also pointed out the importance of future functional analysis. Our findings also show that function-based genetic analysis is useful to pinpoint truly causal variants, especially nonfunctional variants, and to clarify the molecular pathogenesis of common diseases such as gout.

Materials and Methods

Genetic analysis

A function-based genetic analysis of gout and the ABCG2 gene was performed as shown in fig. S1 and Supplementary Material (32). All procedures were carried out in accordance with the standards of the institutional ethical committees involved in this project and according to the Declaration of Helsinki. After written consent had been given by each participant, genomic DNA was extracted from blood samples (57). Mutation analysis of all coding regions and intron-exon boundaries of the ABCG2 gene (58) was performed for 90 Japanese hyperuricemia patients. Primers for mutation analysis of ABCG2 were designed as shown in table S1 according to the genomic structure of the human ABCG2 gene (see fig. S2). For QTL analysis of SUA concentrations, genotyping of Q141K in 739 Japanese individuals was performed. For association studies, 228 Japanese male hyperuricemia cases (including 161 gout cases) as well as 871 Japanese male controls (SUA ≤ 7.0 mg/dl) were additionally genotyped. All gout patients were clinically diagnosed as having primary gout. Individuals whose SUA concentration had been more than 8.0 mg/dl were selected as hyperuricemia cases.

Functional analysis

Wild-type ABCG2 complementary DNA was inserted into the Nhe I and Apa I sites of pcDNA3.1(+) vector plasmid (Invitrogen), with a myc-tag sequence attached at the 5′ end. To prepare membrane vesicles, HEK293 cells were transiently transfected with an expression vector for ABCG2 or an empty vector by FuGENE6 (Roche Diagnostics) according to the manufacturer’s instructions. Forty-eight hours later, cells were harvested and the membrane vesicles were isolated with a standard method described previously (29). The study of [3H]ES (500 nM) and [14C]urate (28 μM) uptake was performed as reported previously (59). Transport experiments with high concentrations of urate were performed under alkaline conditions, and kinetic parameters were calculated by nonlinear regression analysis of the ABCG2-mediated transport of urate. Using the site-directed mutagenesis technique, we constructed mutants of ABCG2 (V12M, Q126X, Q141K, G268R, S441N, and F506SfsX4), which were used for urate transport analysis, on the expression vector for ABCG2. Western blot analysis of the membrane vesicles (20 μg) with 800-fold–diluted antibody against myc-tag (Roche Diagnostics) and 200-fold–diluted antibody against Na+/K+-ATPase α (Santa Cruz Biotechnology) was performed as described (60).

Statistical analysis

For all calculations of statistical analysis, the software R was used (32). The differences in the clinical covariates between the genotypes of the SNPs of ABCG2 were compared with Mann-Whitney and Kruskall-Wallis tests. Regression analysis was used to obtain corrected P values. The χ2 test and Fisher’s exact test were used to compare the difference in genotype frequencies and allele frequencies between the case and control samples. Haplotype estimation was performed with the EM algorithm (61).


[3H]ES was purchased from PerkinElmer Life Science. [14C]Uric acid was purchased from American Radiolabeled Chemicals. Unlabeled ES, AZT, and ATP were purchased from Sigma Chemicals. Unlabeled uric acid was purchased from Nacalai Tesque. All other chemicals used were commercially available and of reagent grade.

Supplementary Material

Materials and Methods

Fig. S1. Flowchart for molecular-function-based clinicogenetic analysis of gout with ABCG2 polymorphic variants.

Fig. S2. Genomic structure and mutation sites of the human ABCG2 gene.

Fig. S3. Results of sequence analysis of ABCG2 gene.

Fig. S4. Western blot analysis of wild-type and mutated ABCG2.

Fig. S5. Proposed model of renal and gut urate excretion.

Table S1. Primer pairs used for mutation analysis of the human ABCG2 gene.

Table S2. Association of Q141K in ABCG2 with clinical parameters.



    References and Notes

    1. Supplementary Materials and Methods are presented as Supplementary Material on Science Translational Medicine Online.

    2. Acknowledgments: We thank all of the patients and healthy volunteers involved in this study; M. Nudejima, K. Nakanishi, T. Tamatsukuri, A. Kudo, T. Iwamoto, M. Miyazawa, J. Inoue, M. Watanabe, M. Wakai, Y. Utsumi, Y. Kitamura, Y. Kawamura, R. Kinoshita, and A. Fujii for genetic analysis; K. Matsumura and S. Matsuyama for technical support; T. Nakazono, H. Itosu, T. Oda, and Y. Kikuchi for patient analysis; M. Yamashiro, K. Takahashi, H. Nakashima, K. Ishii, A. Enomoto, T. Shimizu, T. Itoh, H. Sato, M. Emi, S. Suzuki, Y. Kobayashi, and J. Fukuda for helpful discussion; and D. E. Nadziejka for technical editing.Funding: Ministry of Defense of Japan (H.M., K.K., Y.S., and N.S.); Ministry of Education, Science, and Culture of Japan (Scientific Research on Priority Areas, Transportsome; T.T., K. Ito, Y. Kanai, and H.S.); Japan Society for the Promotion of Science (K. Ichida); Nakabayashi Trust for ALS Research; Kawano Masanori Memorial Foundation for Promotion of Pediatrics; Takeda Science Foundation (H.M.); and Gout Research Foundation of Japan (H.M. and K. Ichida).Author contributions: H.M., T.T., K. Ichida, and T.N. designed the experiment; T.T., Y.I., K. Ito, Y. Kanai, and H.S. performed functional analysis; H.M., K. Ichida, J.N., H.D., S.W., M.F., Y.M., T.H., and N.S. performed patient analysis; H.M., K. Ichida, Y. Kusanagi, T.C., S.T., Y.T., Y.O., H.I., K. Niwa, K.K., and N.S. carried out genetic analysis; H.M., T.N., A.N., S.N., Y.S., and N.H. performed QTL analysis; H.M., K. Ichida, A.N., K.S., R.O., M.N., K. Nishio, A.H., K.W., Y.A., T.H., and N.H. collected samples of patients and normal controls and analyzed data of samples; T.N. performed statistical analysis; H.M. wrote the paper; H.M., T.T., A.N., and N.S. revised the paper.Competing interests: The authors have a patent pending based on the work reported in this paper.Accession numbers: The amino acid sequence of ABCG2 shown in Fig. 2A was obtained from GenBank (accession code NM_004827) . This sequence is identical to that from the ABCG2 clone used in this study. OMIM accession code for gout-susceptibility locus GOUT1 (GOUT SUSCEPTIBILITY 1) on chromosome 4q is MIM 138900.
    View Abstract

    Stay Connected to Science Translational Medicine

    Navigate This Article