Both variants of A1CF and BAZ1B genes are associated with gout susceptibility: a replication study and meta-analysis in a Japanese population

Gout is a common type of acute arthritis that results from elevated serum uric acid (SUA) levels. Recent genome-wide association studies (GWASs) have revealed several novel single nucleotide polymorphism (SNPs) associated with SUA levels. Of these, rs10821905 of A1CF and rs1178977 of BAZ1B showed the greatest and the second greatest significant effect size for increasing SUA level in the Japanese population, but their association with gout is not clear. We examined their association with gout using 1411 clinically-defined Japanese gout patients and 1285 controls, and meta-analyzed our previous gout GWAS data to investigate any association with gout. Replication studies revealed both SNPs to be significantly associated with gout (P = 0.0366, odds ratio [OR] with 95% confidence interval [CI]: 1.30 [1.02–1.68] for rs10821905 of A1CF, P = 6.49 × 10–3, OR with 95% CI: 1.29 [1.07–1.55] for rs1178977 of BAZ1B). Meta-analysis also revealed a significant association with gout in both SNPs (Pmeta = 3.16 × 10–4, OR with 95% CI: 1.39 [1.17–1.66] for rs10821905 of A1CF, Pmeta = 7.28 × 10–5, OR with 95% CI 1.32 [1.15–1.51] for rs1178977 of BAZ1B). This study shows the first known association between SNPs of A1CF, BAZ1B and clinically-defined gout cases in Japanese. Our results also suggest a shared physiological/pathophysiological background between several populations, including Japanese, for both SUA increase and gout susceptibility. Our findings will not only assist the elucidation of the pathophysiology of gout and hyperuricemia, but also suggest new molecular targets.


Introduction
Gout is a common disease which displays severe noninfectious acute arthritis and results from elevated serum uric acid (SUA) level, or hyperuricemia [1]. Recent genetic studies including genome-wide association studies (GWASs) have revealed several genes to be associated with SUA levels [2-7] as well as clinically-defined gout [8][9][10][11][12][13][14]. Of these, Köttgen et al. [7] reported 18 novel loci associated with SUA in European, Indian, African-American and Japanese populations. Two of them, rs10821905 of A1CF (chromosome 10q11.23) and rs1178977 of BAZ1B (chromosome 7q.11.23), showed the statistically significantly greatest and the second greatest effect size for increasing SUA level in a Japanese population [7]. While they showed a nominally significant association with gout in individuals of European Makoto Kawaguchi, Akiyoshi Nakayama contributed equally to this work. ancestry, their association was not clarified in Japanese gout cases.
This prompted us to examine the association between common variants of A1CF, BAZ1B and gout using clinically-defined Japanese gout patients through a candidate gene approach, and to meta-analyze it with our previous gout GWAS data [10].

Patients and controls
1411 male Japanese patients in total were recruited from the outpatients of the gout clinics of Ryougoku East Gate Clinic (Tokyo, Japan), Nagase Clinic (Tokyo, Japan), Wakasa Clinic (Saitama, Japan) and Tokorozawa Central Hospital (Saitama, Japan). All the subjects had been diagnosed with primary gout according to the criteria established by the American College of Rheumatology [23]. As the control group, 1,285 Japanese males without a history of gout or hyperuricemia (SUA levels > 7.0 mg/dL) were selected from participants in the Nagoya and Shizuoka area in the Japan Multi-Institutional Collaborative Cohort Study (J-MICC Study) [24,25]. The mean age and standard deviation of cases and controls were, respectively, 48.2 ± 11.7 and 53.4 ± 9.9 years, and their mean body-mass indexes were 25.2 ± 3.7 and 22.9 ± 2.9 kg/m 2 , respectively.

Genetic and statistical analyses
Genomic DNA was extracted from whole peripheral blood [26]. Genotyping of A1CF polymorphism (rs10821905) and BAZ1B polymorphism (rs1178977) was performed using a TaqMan assay (Applied Biosystems) employing a Lightcycler 480 (Roche Diagnostics) as reported in our previous study [27]. GWAS genotyping data were obtained from our previous study [10] which was performed using the Illumina HumanOmniExpress-12 v1.0 (Illumina) platform employing 945 clinically-ascertained cases and 1213 Japanese male controls.
For the calculations in the statistical analyses, we used R (version 4.0.1) [28] including a meta package [29] in a fixed effect model. The chi-squared test was used for the association and Hardy-Weinberg equilibrium analyses. A P value of < 0.05 was regarded as statistically significant. Table 1 shows the genotyping results for rs10821905 of A1CF and rs1178977 of BAZ1B in 1,411 gout cases and 1,285 controls. The genotyping call rate for these SNPs exceeded 98%. These SNPs in the control group were in Hardy-Weinberg equilibrium (P > 0.05), which suggested no mistyping. Table 1 shows that both SNPs showed a significant association with gout.

Results
As shown in Fig. 1, the meta-analysis between the present study and previous gout GWAS [10] also showed both SNPs to have significant associations with gout (P meta = 3. 16

Discussion
The present study showed, for the first time, an association between rs10821905 of A1CF as well as rs1178977 of BAZ1B and gout in a Japanese population.
Since it has been established that hyperuricemia is associated with dyslipidemia [30] such as hypertriglyceridemia [31], variants of A1CF may affect urate metabolism via the following ApoB-related mechanisms. A1CF (APOBEC1 Table 1 Association analysis between A1CF and BAZ1B and gout SNP single nucleotide polymorphism, RAF risk allele frequency, OR odds ratio, CI confidence interval a The non-risk allele referred to as allele 1 and the risk allele as 2. Allele 1 is G and allele 2 is A in both rs10821905 and rs1178977 b P values were obtained by chi-squared tests  [32]. In other words, mRNA coding ApoB-100 is converted to mRNA coding ApoB-48 by this enzyme complex containing A1CF. Variants of A1CF may, therefore, affect urate metabolism through ApoB production and/or ApoB-related insulin resistance. Moreover, A1CF is reported to be expressed in the liver, kidney and intestine [33,34], from where urate is also mainly produced and excreted in humans. Indeed, in addition to the present study, Dong et al. and Rasheed et al. have reported an association between gout and the A1CF variant in Han Chinese and in New Zealand European and Polynesian populations [35,36]. Nevertheless, further studies need to be conducted to elucidate the precise pathophysiological background, when taking into account the fact that rs10821905 is located at about 2 kbp upstream of the A1CF gene. In other words, while the present study showed significance at rs10821905 of A1CF, it is possible that this is a mere marker and that the true risk gene is present close by. This is the first report to identify an association between clinically-defined gout and BAZ1B. BAZ1B is possibly involved in urate metabolism due to transcriptional changes. BAZ1B (bromodomain adjacent to zinc finger domain 1B), also known as WSTF (Williams syndrome transcription factor), encodes a member of the bromodomain protein family. BAZ1B shows ubiquitous expression in humans and is mainly involved in the chromatin-dependent regulation of transcription, including chromatin assembly, RNA polymerase I and III gene regulation, vitamin D metabolism, and DNA repair [37]. Since Köttgen et al. [7] reported rs1178977 of BAZ1B to be in linkage disequilibrium (LD) with several SNPs of MLXIPL, it is probable that BAZ1B is a mere surrogate marker of MLXIPL, since MLXIPL/ ChREBP is involved in the glucose-6-phosphate production, an upstream pathway of de novo urate production [38,39]. However, because urate is an end metabolite of purine bodies including ATP and some nucleosides, it is also possible that changes in regulation of transcription by BAZ1B variants are associated with urate metabolism.
While deletion, including BAZ1B, is known to cause Williams syndrome [40], the mechanism between SUA or gout and BAZ1B variants remains to be elucidated. It is compatible in that the same SNP (rs1178977) is reported to have an association with triglyceride levels from a previous GWAS [41], when taking into account the association between urate and triglycerides [31].
We previously reported that ABCG2 (ATP-binding cassette subfamily G member 2) is a renal and intestinal urate exporter and that its dysfunctional variants have a significant and strong effect on susceptibility to gout/hyperuricemia [8, 42,43]. We, therefore, recalculated the results shown in Table 1 with and without these variants. As shown in Supplementary Table S1, the A1CF variant still showed a significant association with gout in the presence of the dysfunctional variants of ABCG2 but was no longer significant without those variants, while the BAZ1B variant remained significant both with and without ABCG2 dysfunction. A1CF might, therefore, have synergistic effects on Fig. 1 Meta-analysis of a rs10821905 of A1CF and b rs1178977 of BAZ1B for gout in the Japanese male population. The meta-analysis was conducted using the present study and our previous gout GWAS of Japanese male populations (Matsuo et al. [10]). Both SNPs showed a statistically significant association with gout (P meta = 3.16 × 10 -4 for rs10821905 of A1CF, P meta = 7.28 × 10 -5 for rs1178977 of BAZ1B). OR odds ratio, CI confidence interval susceptibility to gout with dysfunctional variants of ABCG2, for which further analyses are necessary.
Our findings should not only assist the elucidation of the pathophysiology of gout and hyperuricemia, but also suggests new molecular targets.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/. I, Hastie N, Balkau B, Froguel P, Esko T, Salumets A, Khaw