Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Single-nucleotide polymorphisms link gout with health-related lifestyle factors in Korean cohorts

Abstract

Gout—a very painful inflammatory arthritis caused by the deposition of monosodium urate crystals in the joints—is influenced by several factors. We identified the association of single- nucleotide polymorphisms (SNPs) that link gout with health-related lifestyle factors using genomic data from the Korean Genome and Epidemiology Study. We conducted a genome-wide association study (GWAS) on 18,927 samples of 438 Korean patients with gout and 18,489 controls for the discovery stage. For the replication stage, another batch containing samples of 326 patients with gout and 2,737 controls were analyzed. Lastly, a meta-analysis was performed using these two cohorts. We analyzed the effects of health-related lifestyle factors, including eating habits, physical activity, drinking behavior, and smoking behavior, on gout. After identifying the association between GWAS-derived SNPs and health-related lifestyle factors, we confirmed the interaction between the polygenic risk score (PRS) and health-related lifestyle factors. We identified 15 SNPs related to gout, among which rs1481012 of ABCG2 located on chromosome 4 has been newly discovered (P = 2.46e-11). On examining the interaction between SNPs and health-related lifestyles, rs3109823—located in ABCG2—was found to be associated with smoking status. In addition, rs11936395—located in SLC2A9—was significantly associated with the average momentum of exercise per session, whereas rs11066325 located in PTPN11, showed a significant association with the number of exercise sessions per week, smoking status, drinking status, and amount of soju drink per session. rs9421589—located in FAM35A—was significantly associated with the duration of smoking. In addition, we verified that the association between PRS and duration of smoking affects gout. Thus, in this study, we identified novel SNPs that link gout with health-related lifestyle factors in the Korean population.

Citation: Jeon HK, Yoo HY (2023) Single-nucleotide polymorphisms link gout with health-related lifestyle factors in Korean cohorts. PLoS ONE 18(12): e0295038. https://doi.org/10.1371/journal.pone.0295038

Editor: Ashish Verma, Boston University Chobanian & Avedisian School of Medicine, UNITED STATES

Received: October 21, 2022; Accepted: November 14, 2023; Published: December 7, 2023

Copyright: © 2023 Jeon, Yoo. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: The data underlying the results of our study were obtained from the Korean Genome and Epidemiology Study (KoGES), National Institute of Health. Legal restriction does not permit the authors to publicly share the dataset. All interested researchers can request the dataset from the Korean Center for Disease Control (https://nih.go.kr/eng/main/main.do) through due process.

Funding: This study was supported by a National Research Foundation of Korea grant funded by Korea government (MSIT) (grant no. NRF-2019R1F1A1062965 and no. NRF-2022R1F1A1068307). The funder did not participate in the study's design, data collection and analysis, publication decision, or manuscript preparation. There was no additional external funding received for this study.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Gout is a very painful inflammatory arthritis caused by the deposition of monosodium urate crystals in the joints due to abnormal purine metabolism and underexcretion of serum urate, causing inflammation and pain in the joints and the surrounding tissues [1]. It is influenced by a combination of genetic variations and environmental factors, and genetic factors being the key reason for renal excretion of serum urate. The change in serum urate concentration can be primarily (40–80%) attributed to genetic factors [2]. One of the genetic factors related to uric acid excretion by the kidneys is single nucleotide polymorphisms (SNPs). A SNP is a germline substitution of a single nucleotide occurring at the same location in a gene for different individuals. These genetic mutations have a varied prevalence depending on race and group [3].

With the advancement of genome-wide association study (GWAS), several genes associated with gout have been reported [4]. According to previous GWAS, the representative genes related to gout include ABCG2 (encodes the strongest renal apical urate secretion transporter), SLC2A9 (encodes renal basolateral GLUT9, the only urate reabsorption transporter that transports urate from proximal tubule epithelial cells to blood stream), SLC17A1 (encodes NPT1, an apical renal urate secretion transporter), SLC17A3 (encodes NPT4, an apical renal urate secretion transporter), SLC22A11 (encodes OAT4, an apical renal urate reabsorption transporter), SLC22A12 (encodes URAT1, the strongest apical renal urate/nicotinate exchanging reabsorption transporter), and PKD2 (encodes polycystin-2, an integral membrane protein having characteristics of a calcium-permeant cation channel) [4–10]. Effects of the interaction between PKD2 and ABCG2 on gout has been studied in Chinese and Japanese male populations, respectively [5]. In addition, the genetic association between gout and rs2230054 of CXCR2 in Chinese Han men has been recently studied [8]. In Korea, the association between electric genes and the interaction between genes in Korean gout was investigated through the Korean Association Resource (KARE) cohort [11]. Sull et al. [12] also investigated the impact of SLC2A9 mutations on serum urate levels in Koreans. Additionally, a significant association between rs2231142 or rs2054576 of ABCG2 and Korean gout was revealed [6, 13]. Consequently, prior studies on gout have had varied results based on race or study group.

In Korea, the number of patients with gout is increasing rapidly, and it is prevalent not only among those in their 50s and 60s but also among those in their 20s and 30s owing to the increase in meat intake and sedentary lifestyles, which leads to obesity [14]. The average serum urate concentration of the entire population is increasing in Korea, attributed to multifaceted factors such as lifestyle and dietary habits [15].

Epigenetics asserts that differences in the environment can modify gene expression. It also shows that expression of genes varies depending on the environment, diet, and behavioral habits [16]. Moreover, the gene–environment correlation does not exist independently, and changes in one affect the characteristics of the other through a feedback loop [17, 18]. Therefore, it is necessary to promote good health by enhancing environmental factors owing to the role the interaction of genetic and environmental factors play in human diseases. In addition, although gout is a disease affected by both genetic variations and environmental factors, previous studies on genetic variations related to gout in Korean patients with gout focused on only the genetic aspect without considering environmental factors. Therefore, detailed research considering both environmental and genetic factors in Korean gout is urgently needed.

In this study, we conducted a GWAS analysis on patients with gout in Korea based on the Health Examinees study (HEXA) cohort and KARE cohort of the Korean Genome and Epidemiology Study (KoGES). Furthermore, the polygenic risk score (PRS) was calculated as a weighted sum of the risk alleles of the SNPs to assess the risk level of the SNPs identified in the GWAS analysis [19]. PRS has been used extensively to predict the genetic predisposition toward diseases, since it was established that the genetic risk for schizophrenia is a predictor of bipolar disorder [20]. Ultimately, this study seeks to identify the SNPs related to gout in Koreans, analyze their interactions with health-related lifestyle factors, and determine whether these interactions affect gout. In addition, this study calculated the PRS to analyze the genetic risk of gout in Koreans and assessed the interaction between PRS and health-related lifestyle factors.

Material & methods

Data collection

We used the epidemiologic and genomic data from the first HEXA-based survey conducted from 2004 to 2013 from the KoGES, as well as the genomic data from the first KARE-based survey conducted from 2001 to 2002, from the KoGES, conducted by the Korea Centers for Disease Control and Prevention. The Korean Biobank Array (also called the K-Chip or KoreanChip) was used to process the HEXA and KARE genomic data.

Study participants

This study included data from subjects in the age group of 40 to 69 years, recruited by the KoGES. Among the 173,208 people who were included in the HEXA-based survey, the epidemiologic and genomic data of 58,700 people, including their genetic information, were used. Similarly, the genomic data of 5,493 people out of the 10,031 people studied in the KARE-based study were used.

In the discovery stage, a total of 18,927 subjects were analyzed, including the 438 self-reported subjects with gout in the HEXA cohort. It was assumed that gout is diagnosed by rheumatologists according to the American College of Rheumatology diagnostic criteria [21].

Subjects who met any of the following criteria were excluded: “No response” (n = 7,624) and “Other diseases” (n = 32,149) such as chronic, metabolic, cardiovascular, cerebral, and respiratory diseases, detected through disease history investigation. In the replication stage, a total of 3063 subjects were analyzed, including the 326 subjects with gout in the KARE cohort (S1 Fig). This study was conducted according to the guidelines of the Declaration of Helsinki and approved by the institutional review boards of Chung-Ang University (approval no.: 1041078-202005-HRBR-137-01) and Korean center for Disease Control and Prevention (KBN-2020-080). All participants provided their written informed consent voluntarily.

General and clinical characteristics

The data in this study were investigated using self-reported questionnaires and clinical examinations. The questionnaire was developed by including items from a semi-quantitative food frequency questionnaire, socio-demographic status, lifestyle (i.e., diet, smoking, drinking, and physical activity), and disease history for the KoGES [22].

Parameters like gender, age, marital status, occupation, education, monthly household income, medical history (hypertension/diabetes), age during gout diagnosis, height, weight, body mass index (BMI), systolic blood pressure, diastolic blood pressure, pulse, fasting blood sugar, blood urea nitrogen, creatinine, uric acid, total cholesterol (TC), low density lipoprotein cholesterol (LDL-C), high density lipoprotein cholesterol (HDL-C), triglycerides (TG), aspartate transaminase (AST), alanine transaminase (ALT), and high sensitivity C-reactive protein (hs-CRP) were analyzed. Bio-specimens included fasting blood samples that were collected in a serum separator tube and two ethylenediaminetetraacetic acid (EDTA) tubes, and a 10 ml midstream urine sample. For long-term storage, both serum and plasma were prepared and aliquoted in 6–10 vials (300–500 μl per vial), and 80–100 μg samples of blood DNA were also prepared [22].

Health-related lifestyle factors

Health-related lifestyle factors are environmental factors such as eating habits, physical activity, drinking behavior, and smoking behavior. Eating habits included the number of regular meals per day and the average frequency of food intake over the past year. Physical activity included the exercise status, number of exercise sessions per week, and average momentum of exercise per session. Drinking behavior included drinking status, duration of drinking, average frequency of alcoholic beverages (soju/beer) consumed over the past year, and quantity of alcohol consumed (average of the amount soju/beer drink per session: one bottle of soju = 6.5 cups, 50 cc per cup/one bottle of beer = 2.5 cups, 220 cc per cup). Smoking behavior included smoking status, duration of smoking, and number of cigarettes smoked per day.

Genetic data refinement and extraction

Data purification and SNP detection along with the analysis of the interaction between SNPs and health-related lifestyle factors were performed using the Plink (v.1.9, http://pngu.mgh.harvard.edu/purecell/plink/) genome analysis program. Functional Mapping and Annotation of Genome-Wide Association Studies was used to interpret the genomic data results and express them through Manhattan plot and Q–Q plot [23]. The SNPs for meta-analysis results were visualized using PhenoGram [24]. Locuszoom version 1.1 (http://csg.sph.umich.edu/locuszoom) was utilized to confirm the association between SNPs of ABCG2 on chromosome 4 and to create regional plots [25].

Genomic DNA samples were genotyped using the Korea Biobank Array (K-chip). The K-chip is a new array, a Korean-customized dielectric chip developed by the Korea National Institute of Health in 2015. The K-chip consists of approximately 830,000 markers, including more than 247,000 rare-frequency and functional variants estimated through the sequencing data of more than 2,500 Koreans [26]. In this study, genotyping was conducted to determine the genotypes of 64,193 people and investigate the genotypes for 6,176,658 SNPs. In addition, SNP quality control (QC) was performed to purify genomic data to analyze 18,927 of the 58,700 subjects from the HEXA study included in the K-chip. Finally, 5,442,390 SNPs were used for actual analysis after QC by applying minor frequency (MAF) <0.05, SNP call rate <0.95, and Hardy–Weinberg equilibrium (HWE) P-value < 1E-6.

Discovery stage: HEXA

In this study, the genomic data from the HEXA study included in the K-chip were analyzed to perform GWAS. Similar to that in a previous study [11], age, gender, and BMI were used as covariates and GWAS was performed using logistic regression analysis. We analyzed 18,927 out of 58,700 subjects in the HEXA study, and 15 lead SNPs significant for gout were detected. Lastly, the identified SNPs (n = 15) were assigned points (0, 1, or 2) based on the number of risk alleles present; the total score was equivalent to PRS.

Replication stage: KARE

A replication study was performed on 3,063 subjects from the KARE cohort, with 326 subjects in the gout group and 2,737 subjects in the control group, to examine whether the 15 SNPs derived from the HEXA study were reproduced in this cohort. As with the discovery stage, age, gender, and BMI were adjusted as covariates. A meta-analysis was performed after the replication stage. Lastly, the PRS calculation method was the same as that in the discovery stage.

Statistical analysis

Meta-analysis was performed by logistic regression analysis using PLINK by combining the discovery stage and replication stage of GWAS. The combined subject was composed of 764 cases and 21,226 controls.

Health-related lifestyle factors with SNPs of HEXA were analyzed by univariate logistic regression using PLINK. Binary logistic regression was performed with health-related lifestyle factors and PRS as independent variables, and the presence or absence of gout as a dependent variable.

Results

General and clinical characteristics

Based on gender, the gout subjects were categorized as male (369, 84.2%) and female (69, 15.8%), with relatively higher male subjects and significant differences between gout and healthy control group (χ2 = 559.32, P<0.001). The age of gout diagnosis was in the range of 50.31±9.02 years in the gout group. The serum urate concentration was 6.67±1.84 mg/dl in the gout group and 4.51±1.18 mg/dl in the control group, with significantly higher serum urate levels in the gout group (t = -24.46, P<0.001). The TC and LDL-C levels were significantly higher in the control group than in the gout group. HDL-C level was significantly lower in the gout group than in the control group (t = 14.33, P<0.001), whereas the TG, AST, ALT, and hs-CRP levels were significantly higher in the gout group (S1 Table).

GWAS

Discovery stage. GWAS analysis, which was conducted to identify the genetic variants associated with gout in all the subjects in the discovery stage, revealed the presence of 15 SNPs. The most significant SNP was rs1481012 (P = 1.12e-14) located in ABCG2, based on the GWAS results. The second most significant SNP was rs11936395 (P = 1.37e-06) located in SLC2A9 and rs200888518 (P = 1.84e-06) located in APP. The rest of the SNPs in order of their significance are as follows: rs11066325 (P = 2.00e-06) of PTPN11, rs3798728 (P = 2.80e-06) of RP3-510L9.1, rs9532070 (P = 5.71e-06) of RP11-14O22.1, rs3109823 (P = 6.02e-06) of ABCG2, rs339405 (P = 6.50e-06) of TRIP10, rs28674878 (P = 6.72e-06) of CHST15, rs56205418 (P = 6.98e-06) of CLPS, rs17794144 (P = 7.90e-06) of CWC22, rs17013965 (P = 7.98e-06) of PPM1, rs59517147 (P = 8.00e-06) of KADAMTS9, rs9421589 (P = 9.42e-08) of FAM35A, and rs146386352 (P = 9.76e-06) of ROBO2 (S2 Table) (Figs 1 and 2).

thumbnail
Download:
Fig 1. Manhattan plot of GWAS on HEXA.

Manhattan plot showing the results of the GWAS analysis of patients with gout in the HEXA study. The X-axis of the plot shows the genomic location and the Y-axis shows the level of association and each dot represents an SNP. Significant SNPs associated with gout were identified in the genetic region associated with gout by the appearance of SNPs with a p-value of 1e-5 or less rising upward. The red line shows critical values for genome-wide statistical significance (P<1e-5).

https://doi.org/10.1371/journal.pone.0295038.g001

thumbnail
Download:
Fig 2. Q–Q plot of GWAS on HEXA.

Q–Q plot showing the results of the GWAS analysis of patients with gout in the HEXA study. The blue dots representing SNPs coincide with the red oblique line below. They then deviate from the oblique line around the middle. SNPs that do not coincide with the oblique line are considered to be associated with gout.

https://doi.org/10.1371/journal.pone.0295038.g002

Replication stage.

In this stage, it was confirmed whether the 15 SNPs derived from the HEXA study in the discovery stage were reproduced in the KARE study as well (S3 Table).

Meta-analysis results.

Table 1 shows details regarding the most significant SNP in Korean patients with gout. A meta-analysis was conducted using the 15 SNPs identified from the two cohorts. The rs1481012 SNP of the ABCG2 gene (4:89039082_A/G) located on chromosome 4 satisfying the GWAS signature p threshold (P<5e-8) was discovered (P = 2.46e-11) (Figs 3 and 4).

thumbnail
Download:
Fig 3. PhenoGram plot of gout.

PhenoGram plot of the SNPs associated with gout. The long bars represent the 9 out of 22 chromosomes in which gout-related SNPs were found. The position of the SNP on each chromosome is indicated by a line and the line is connected to colored circles; Each different colored circles represent gout-related SNPs.

https://doi.org/10.1371/journal.pone.0295038.g003

thumbnail
Download:
Fig 4. Regional plot of SNPs of ABCG2 on chromosome 4.

Regional plots were based on hg 19 version ASN (Asian population). A high r2 SNP was observed around rs1481012, which has the highest statistical significance. The purple diamond is an SNP that is the standard for regional plots, and rs1481012 showed the highest level of significance as a result of association analysis. One SNP circled in red has a high r2 and is found in the same region, indicating that the SNPs are highly correlated.

https://doi.org/10.1371/journal.pone.0295038.g004

thumbnail
Download:
Table 1. Significant SNPs related to gout established through meta-analysis.

https://doi.org/10.1371/journal.pone.0295038.t001

Differences in PRS between the two groups

The PRS for the 15 lead SNPs was significantly higher for HEXA (P = 7.56E-63), and a higher PRS has been shown to affect gout (OR = 1.48, 95% CI = 1.41–1.55; P = 7.56e-63). However, for KARE, there was no significant differences between control group and patients with gout (S4 Table).

Health-related lifestyle factors associated with gout

With respect to the number of exercise sessions per week, daily exercise was positively associated with the risk of gout (P = 0.036). For drinking behavior, past drinking and current drinking showed a 3.53-fold and 1.74-fold higher risk of gout, respectively (P = 5.23e-10, P = 2.56e-05, respectively). In particular, consuming more than 0.5 bottles of soju and beer increased the risk of gout by 0.30-fold and 0.38-fold, respectively (P = 0.000, P = 0.002, respectively).

Furthermore, current smoking was positively associated with risk of gout (P = 0.042). Subjects who had smoked for >41 years developed a 0.44-fold higher risk of gout than those who smoked for less than 5 years (P = 0.042). No significant differences in the risk of gout development were found with respect to eating habits (S5 Table).

Interactions between 15 SNPs and health-related lifestyle factors

The interactions between SNPs and health-related lifestyle factors are presented in Table 2. The minor allele of rs3109823 located in ABCG2 interacted with smoking status (OR = 0.73, 95% CI 0.54–0.98; P = 0.037). rs11936395 located in SLC2A9 interacted with the average momentum of exercise per session (OR = 0.76, 95% CI 0.59–0.97; P = 0.033). The interaction between rs11066325 located in PTPN11 and the number of exercise sessions per week, smoking status, drinking status, and amount of soju drink per session was significant. Duration of smoking was found to interact with rs9421589 located in FAM35A (OR = 1.18, 95% CI 1.01–1.39; P = 0.036).

thumbnail
Download:
Table 2. Interaction between 15 SNPs and health-related lifestyle factors.

https://doi.org/10.1371/journal.pone.0295038.t002

In KARE, the interaction between rs146386352 located in ROBO2 and smoking status was significant (OR = 2.16, 95% CI 1.42–3.28; P = 0.000). In addition, rs11936395 located in SLC2A9 was significantly associated with physical activity during the day (middle-intensity activity) as well as smoking status (P = 0.041, P = 0.008, respectively).

Interaction between PRS and health-related lifestyle factors

Table 3 shows the analysis results of the interaction of PRS and health-related lifestyle factors. Among them, the duration of smoking did interact with PRS for the risk of gout (OR = 1.06, 95% CI = 1.00–1.11; P = 0.030).

thumbnail
Download:
Table 3. Interaction between PRS and health-related lifestyle factors.

https://doi.org/10.1371/journal.pone.0295038.t003

Similarly, in KARE, the duration of smoking showed a significant association with gout risk (OR = 1.18 95% CI = 1.02–1.35; P = 0.019). Furthermore, the average number of soju drinks per year and smoking status were also significantly related with the risk of developing gout (P = 0.023, P = 0.048, respectively).

Discussion

SNPs related to gout

We identified 15 SNPs related to gout in the Korean population. rs1481012 was one of the 24 SNPs in ABCG2 screened for their association with gout in the Chinese population [27], but its presence in Koreans was detected for the first time in this study. In previous studies, rs1481012 in ABCG2 has been associated with risk of hyperuricemia, gout, coronary artery disease (CAD), B-cell non-Hodgkin lymphoma (B-NHL), and chronic lymphocytic leukemia (CLL) [28–30]. Rs2728121 in PKD2 and rs1481012 in ABCG2 affect the etiology of diseases and are associated with elevated serum urate, hyperuricemia, and gout [5], which was in agreement with our findings that rs1481012 affects gout. We also identified rs3109823 in ABCG2 and rs11936395 in SLC2A9. Several studies have established that ABCG2 and SLC2A9 are related to serum urate levels [6, 31]. Rs3109823 of ABCG2 was also detected as one of the SNPs related to serum urate levels in Koreans [32]. Moreover, rs4529048 in SLC2A9 affects serum urate levels. This is an iSNP located in the intron (DNA segment that does not contain genetic information or code for proteins) of SLC2A9, which also contains rs11936395 found in this study, and is associated with increased serum urate levels and risk of gout [33].

A study on Chinese patients revealed that the epistatic interactions between PKD2 and ABCG2 affect serum urate concentration and gout risk. This study confirmed that pairs of rs2728121 in PKD2 and rs1481012 in ABCG2 can impact the etiology of elevated serum urate, hyperuricemia, and gout [5], which supports the fact that rs1481012 discovered in this study influences gout. In addition, this study identified APP, PTPN11, RP3-510L9.1, RP11-14O22.1, TRIP10, CHST15, CLPS, CWC22, PPM1, KADAMTS9, FAM35A, and ROBO2 as genes related to gout.

The amyloid precursor protein (APP) is located on chromosome 21 and is linked to gout as a genetic risk factor for Alzheimer’s disease, which can induce disease through interaction with the mutated HPRT1 [34, 35]. Additionally, HPRT (hypoxanthine-guanine phosphoribosyltransferase) deficiency can be caused by a mutant HPRT1 [36]. In absence of HPRT enzyme, hypoxanthine and guanine cannot be converted to nucleotides and are disintegrated in the form of uric acid, leading to an increase in serum urate concentration and abnormal accumulation of urate crystals in the kidneys and joints [37]. Ultimately, mutations in HPRT1 can result in manifestation of disease through variations in the interaction between HPRT1 and APP [38]. Protein tyrosine phosphatases (PTP) are known to regulate various processes such as cell growth, differentiation, division cycle, and tumor conversion [39]. We found in this study that PTPN11 (protein tyrosine phosphatase non-receptor type 11), located on chromosome 12, interacts with three health-related lifestyle factors. In addition, mutations in PTPN11 modulate serum HDL-C and are associated with hemostatic pathways such as platelet activation, aggregation, and sensitization [40, 41], as well as major genes associated with hypertension, ischemic heart disease, leukemia, and breast cancer [42]. Meanwhile, PTPN11 encodes the Src homology 2 domain–containing protein phosphatase 2 (SHP-2), which mediates cell responses to various growth factors, hormones, and cytokines and is crucial for the migration and invasion of fibroblasts and tumor cells [42, 43]. SHP-2 enhances the survival and invasion of fibroblast-like synoviocytes (FLS) and the responsiveness to platelet-derived growth factor (PDGF) and tumor necrosis factor (TNF) in rheumatoid arthritis through the activation of focal adhesion kinase (FAK). It was shown that SHP-2 promotes the aggressiveness of FLS in rheumatoid arthritis [44]. These findings demonstrate that the PTPN11 is associated with rheumatoid arthritis and that SHP-2 contributes to the development of rheumatoid arthritis [43]. FAM35A (family with sequence similarity 35 member A), located on chromosome 10, encodes proteins with 835 amino acids and is ubiquitous, including the kidneys and other organs such as thyroid, duodenum, and small intestine [45]. Recently, it was confirmed that rs7903456 of FAM35A was associated with hyperuricemia and gout in Japanese and Chinese populations [7]. Currently, there is no GWAS analysis on FAM35A in the Korean population. In this study, rs9421589 of FAM35A was identified as a significant lead SNP in the gout group. ROBO2 on chromosome 3 encodes a transmembrane receptor for ligands and is essential for ureteric bud formation [46]. Mutations in this gene are associated with ureter reflux in the bladder, which is characterized by urine reflux from bladder to ureter or kidney [47]. Regarding this urinary reflux, mutations in the ROBO2 are thought to be a potential genetic factor in patients with hyperuricemia and gout.

Minor allele frequency (MAF), the lower frequency of allelomorphic characters (alleles) in one SNP, is widely used in genetics because it distinguishes between common and rare variants in a population. MAF can vary in any population distribution as genetic variation and distribution vary with race [48]. Therefore, the prevalence of the disease may also vary with genetic variation and distribution can vary with race. According to a GWAS analysis in Han Chinese and Solomon Islanders, SNPs rs3733591, rs3733589, and rs1014290 in SLC2A9 affect gout in Han Chinese but not in Solomon Islanders [49]. rs3733591 does not affect Polynesian or Caucasian gout patients [50]. Hence, the effect of SNPs in the same gene differ based on race, ethnicity, and region. Identification of SNPs through K-Chip provides a clearer understanding of the genetic factors of gout unique to Koreans. Further studies on other Korean population cohorts are needed to discover other genetic factors affecting gout in Koreans.

Health-related lifestyle factors related to gout

The high purine content in beer may be partly responsible for increased risk of hyperuricemia/gout [51]. The major purine component in beer is guanine, and the increase in serum urate caused by purines contained in beer is sufficient to boost the effect of alcohol elevating serum urate levels [52, 53]. Ethanol from excessive consumption of soju/beer/distilled spirit causes hyperuricemia by decreasing urate excretion and increasing its production by enhancing ATP degradation to uric acid precursors [54]. Excessive ethanol consumption also causes a swift increase in alcohol metabolism (SIAM) thus inducing a hypermetabolic state with increased oxygen consumption in the liver cells primarily mediated by Kupffer cells, the resident hepatic macrophages [55]. Ethanol was reported to induce Kupffer cells to release prostaglandin E2 (PGE2), which stimulates glucose production from endogenous hepatic glycogen and increases oxygen uptake in hepatocytes leading to pericentral hypoxia [55]. Experimental hypoxia, correlated with preeclampsia, reportedly causes hyperuricemia [56]. In addition, previous studies have indicated that excessive consumption of soju (Korean distilled spirits) or other distilled spirits has been linked with an increased risk of gout, which supports our findings [52, 57]. Therefore, refraining from alcohol consumption can prevent gout.

Notably, we found that current smokers and those who have smoked for more than 41 years had a substantial positive correlation with gout-related SNPs. This differs from the findings of some studies on the association between gout and smoking. Previous research has found that the incidence of gout was 20–30% greater in nonsmokers than in smokers [58–60]. However, these studies did not consider genetic factors, limitedly considered the sex of the subjects, and the environmental factors considered were different for each study. Furthermore, several studies have noted the harmful effects of smoking and concluded that smoking was not recommended in patients with gout [58–60]. Therefore, the findings of this study are significant, suggesting that long-term smoking may increase the risk of gout-related genetic modification.

Purines are abundant in foods such as pork and beef, meat by-products (organ meat, blood, etc.), fish, shellfish, high-fructose fruits, and sugar or fructose-rich drinks [4]. Although several epidemiological studies have indicated that purine intake is associated with the incidence of gout [53, 61], purine intake and the number of regular meals per day were not associated with the gout-related SNPs in this study As purines are water-soluble, their concentration can be reduced by discarding the water during the cooking process. Moreover, by drinking more water, uric acid excretion can be increased by increasing urine discharge [62]. Hence, even if purine-rich foods are consumed, the degree of absorption may differ depending on the water intake pattern and cooking methods [63]. Therefore, to analyze the relationship between purine intake and gout more accurately, further studies focused on water intake and detailed cooking methods should be conducted.

In physical activity, the risk of gout was higher for those exercised for 61–120 min or for more than 120 min per session compared to those who exercised for less than 30 min. These results support those of a previous study in which people who exercised had a higher prevalence of gout than those who did not [64]. Furthermore, the prevalence of hyperuricemia was the highest when vigorous exercise was performed and decreased as exercise intensity decreased with moderate and mild exercise [65]. Several studies indicated that high-intensity exercise increases uridine and hypoxanthine levels in the blood, leading to an increase in serum urate concentration [65–67]. Therefore, it is recommended that patients with gout pay special attention to their exercise intensity and that further studies are needed on exercise interventions that consider the exercise intensity of patients with gout.

Gout-related gene–lifestyle interactions and significance of PRS

The PRS of 15 SNPs related to gout were calculated, and their interactions with 8 health-related lifestyle factors were analyzed. The PRS associated with the duration of smoking, and if the smoking period was long, the genetic risk of gout increased, whereas if the smoking period was short, the genetic risk of gout was relatively low.

The PRS for the 15 gout-related lead SNPs was significantly higher in the gout group than in the control group (12.68±2.26 points in the gout group and 10.82±2.23 points in the control group in the HEXA cohort). Therefore, it is useful for predicting the genetic risk and its impact and for investigating gene–environment interactions. A single SNP that is associated with the outbreak of a disease has a low influence on diseases, whereas PRS assigns scores according to risk alleles for multiple SNPs and has a higher influence [68]. However, no gene–lifestyle interaction studies on gout have been reported in Korea. This study is significant as this study identified SNPs, for the first time, linking gout with health-related lifestyle factors.

Conclusions

In conclusion, this study found, for the first time, that rs1481012 of ABCG2 was associated with the risk of gout in Koreans. We also showed that genetic factors interact with health-related lifestyle factors such as lifestyle and have a significant impact on gout. In particular, the duration of smoking and amount of drinks consumed were shown to increase gout risk. These results suggest treatment and intervention require consideration of both genetic and lifestyle factors.

Supporting information

S1 Fig. Case/control selection process.

https://doi.org/10.1371/journal.pone.0295038.s001

(TIF)

S1 Table. General and clinical characteristics.

https://doi.org/10.1371/journal.pone.0295038.s002

(DOCX)

S2 Table. The significant 15 SNPs after discovery stage.

https://doi.org/10.1371/journal.pone.0295038.s003

(DOCX)

S3 Table. Reproduced SNPs in KARE cohort after replication.

https://doi.org/10.1371/journal.pone.0295038.s004

(DOCX)

S4 Table. Differences in PRS between the two groups.

https://doi.org/10.1371/journal.pone.0295038.s005

(DOCX)

S5 Table. Health-related lifestyle factors associated with gout.

https://doi.org/10.1371/journal.pone.0295038.s006

(DOCX)

Acknowledgments

This study was conducted with bioresources from National Biobank of Korea, the Korea Disease Control and Prevention Agency, Republic of Korea (KBN-2020-080).

References

  1. 1. Abhishek A, Roddy E, Doherty M. Gout–a guide for the general and acute physicians. Clin Med. 2017; 17: 54–59. pmid:28148582
  2. 2. Johnson RJ, Titte S, Cade JR, Rideout BA, Oliver WJ. Uric acid, evolution and primitive cultures. Semin Nephrol. 2005; 25: 3–8. pmid:15660328
  3. 3. Robinson PC.Gout–An update of aetiology, genetics, co-morbidities and management. Maturitas. 2018; 118: 67–73. pmid:30415758
  4. 4. MacFarlane LA, Kim SC. Gout: A review of nonmodifiable and modifiable risk factors. Rheum Dis Clin North Am. 2018; 40: 581–604. pmid:25437279
  5. 5. Dong Z, Zhou J, Jiang S, Li Y, Zhao D, Yang C, et al. Epistatic interaction between PKD2 and ABCG2 influences the pathogenesis of hyperuricemia and gout. Hereditas. 2020; 157: 1- pmid:32000861
  6. 6. Kim YS, Kim Y, Park G, Kim SK, Choe JY, Park BL, et al. Genetic analysis of ABCG2 and SLC2A9 gene polymorphisms in gouty arthritis in a Korean population. Korean J Intern Med. 2015; 30: 913–920. pmid:26552468
  7. 7. Nakayama A, Nakaoka H, Yamamoto K, Sakiyama M, Shaukat A, Toyoda Y, et al. GWAS of clinically defined gout and subtypes identifies multiple susceptibility loci that include urate transporter genes. Ann Rheum Dis. 2017; 76: 869–877. pmid:27899376
  8. 8. Wang Y, Wu Y, Xing Q, Chu N, Shen L, Yu X, et al. Genetic association of polymorphism rs2230054 in CXCR2 gene with gout in Chinese Han male population. Cent Eur J Immunol. 2020; 45: 80–85. pmid:32425684
  9. 9. Mandal AK, Mercado A, Foster A, Zandi-Nejad K, Mount DB. Uricosuric targets of tranilast. Pharmacol Res Perspect. 2017; 5: e00291. pmid:28357121
  10. 10. Mandal AK, Mount DB. The molecular physiology of uric acid homeostasis. Annu Rev Physiol. 2015: 77: 323–345. pmid:25422986
  11. 11. Shin J, Kim Y, Kong M, Lee C. Genetic architecture for susceptibility to gout in the KARE cohort study. J Hum Genet. 2012; 57: 379–384. pmid:22513714
  12. 12. Sull JW, Park EJ, Lee M, Jee SH. Effects of SLC2A9 variants on uric acid levels in a Korean population. Rheumatol Int. 2013; 33: 19–23. pmid:22212408
  13. 13. Son CN. Genome-wide association study for hyperuricemia in a community-based Korean cohort. Doctoral dissertation. [Seoul]: Hanyang University; 2016 Feb. Korean
  14. 14. Zhang Q, Gong H, Lin C, Liu Q, Baima Y, Wang Y, et al. The prevalence of gout and hyperuricemia in middle-aged and elderly people in Tibet Autonomous Region, China: A preliminary study. Med (Baltim). 2020; 99: e18542. pmid:31914028
  15. 15. Kim Y, Kang J, Kim GT. Prevalence of hyperuricemia and its associated factors in the general Korean population: an analysis of a population-based nationally representative sample. Clin Rheumatol. 2018; 37: 2529–2538. pmid:29790110
  16. 16. Butler F, Alghubayshi A, Roman Y. The epidemiology and genetics of hyperuricemia and gout across major racial groups: A literature review and population genetics secondary database analysis. J Pers Med. 2021; 11: 231. pmid:33810064
  17. 17. Simon C, Hessen DO. Selection as a domain-general evolutionary process. Behav Processes. 2019; 161: 3–16. pmid:29278778
  18. 18. Chi HM, Zhai QF, Zhang M, Ming S, Donghong Cao W, et al. App/Ps1 gene-environment noise interaction aggravates Ad-like neuropathology in hippocampus via activation of the Vdac1 positive feedback loop. Curr Alzheimer Res. 2021; 18: 14–24. pmid:33761858
  19. 19. Wray NR, Goddard ME, Visscher PM. Prediction of individual genetic risk to disease from genome-wide association studies. Genome Res. 2007; 17: 1520–1528. pmid:17785532
  20. 20. Purcell SM, Wray NR, Stone JL, Visscher PM, O’Donovan MC, Sullivan PF, et al. Common polygenic variation contributes to risk of schizophrenia and bipolar disorder. Nature. 2009; 460: 748–752. pmid:19571811
  21. 21. Neogi T, Jansen TLTA, Dalbeth N, Fransen J, Schumacher HR, Berendsen D, et al. 2015 gout classification criteria: An American College of Rheumatology/European League Against Rheumatism collaborative initiative. Ann Rheum Dis. 2015; 67: 2557–2568. pmid:26359487
  22. 22. Kim YJ, Han BG, KoGES group. Cohort Profile: The Korean Genome and Epidemiology Study (KoGES) Consortium. Int J Epidemiol. 2017; 46: e20. pmid:27085081
  23. 23. Watanabe K, Taskesen E, Van Bochoven A, Posthuma D. Functional mapping and annotation of genetic associations with FUMA. Nat Commun. 2017; 8: 1826. pmid:29184056
  24. 24. Wolfe D, Dudek S, Ritchie M, Pendergrass S. Visualizing genomic information across chromosomes with PhenoGram. BioData Min. 2013; 6: 18. pmid:24131735
  25. 25. Pruim RJ, Welch RP, Sanna S, Teslovich TM, Chines PS, Gliedt TP, et al. LocusZoom: Regional visualization of genome-wide association scan results. Bioinformatics. 2010; 26: 2336–2337. pmid:20634204
  26. 26. Moon SH, Kim YJ, Han SH, Hwang MY, Shin DM, Park MY, et al. The Korea Biobank Array: Design and identification of coding variants associated with blood biochemical traits. Sci Rep. 2019; 9: 1–11. pmid:30718733
  27. 27. Yu KH, Chang PY, Chang SC, Wu-Chou YH, Wu LA, Chen DP, et al. A comprehensive analysis of the association of common variants of ABCG2 with gout. Sci Rep. 2017; 7: 9988. pmid:28855613
  28. 28. Dong Z, Zhou J, Jiang S, Li Y, Zhao D, Yang C, et al. Effects of multiple genetic loci on the pathogenesis from serum urate to gout. Sci Rep. 2017; 7: 1–12. pmid:28252667
  29. 29. Kononov S, Mal G, Azarova I, Klyosova E, Bykanova M, Churnosov M, et al. (2021). Pharmacogenetic loci for rosuvastatin are associated with intima-media thickness change and coronary artery disease risk. Pharmacogenomics, 23: 15–34. pmid:34905955
  30. 30. Campa D, Butterbach K, Slager SL, Skibola CF, De Sanjosé S, Benavente Y, et al. A comprehensive study of polymorphisms in the ABCB1, ABCC2, ABCG2, NR1I2 genes and lymphoma risk. Int J Cancer. 2012; 131: 803–812. pmid:21918980
  31. 31. Jung SJ. SLC2A9 gene polymorphism in Korean patients with gouty arthritis. Doctoral dissertation. [Seoul]: Yonsei University; 2014 Feb. Korean
  32. 32. Lee J, Lee Y, Park B, Won S, Han JS, Heo NJ. Genome-wide association analysis identifies multiple loci associated with kidney disease-related traits in Korean populations. PLoS One. 2018; 13: e0194044. pmid:29558500
  33. 33. Gee HY, Cachau R, Choi JM, Park D, Jee SH, Ryu S, et al. Genetic predisposition to hypouricemia on whole-exome sequencing analysis and its utilities in primary screening purposes. bioRxiv. 2018; 459727.
  34. 34. Nguyen KV. Epigenetic regulation in amyloid precursor protein and the Lesch-Nyhan syndrome. Biochem Biophys Res Commun. 2014; 446: 1091–1095. pmid:24680827
  35. 35. Lee SH, Park GW. Clinical aspects of genetic testing for dementia. J Korean Geriatr Soc. 2008; 12: 5–10
  36. 36. Torres RJ, Puig JG. Hypoxanthine-guanine phosophoribosyltransferase (HPRT) deficiency: Lesch-Nyhan syndrome. Orphanet J Rare Dis. 2007; 2: 1–10. pmid:18067674
  37. 37. Nguyen KV. β-Amyloid precursor protein (APP) and the human diseases. AIMS Neuroscience. 2019; 6: 273–281. pmid:32341983
  38. 38. Nguyen KV, Nyhan WL. Quantification of various APP-mRNA isoforms and epistasis in Lesch-Nyhan disease. Neurosci Lett. 2017; 643: 52–58. pmid:28192196
  39. 39. Tiganis T. Protein tyrosine phosphatases: dephosphorylating the epidermal growth factor receptor. IUBMB life. 2002; 53: 3–14. pmid:12018405
  40. 40. Liu M, Jin HS, Park S. Protein and fat intake interacts with the haplotype of PTPN11_rs11066325, RPH3A_rs886477, and OAS3_rs2072134 to modulate serum HDL concentrations in middle-aged people. Clin Nutr. 2020; 39: 942–949. pmid:31006500
  41. 41. Li X, Meng X, He Y, Spiliopoulou A, Timofeeva M, Wei WQ, et al. Genetically determined serum urate levels and cardiovascular and other diseases in UK Biobank cohort: A phenome-wide mendelian randomization study. PLoS Med. 2019; 16: e1002937. pmid:31626644
  42. 42. Chan G, Kalaitzidis D, Neel BG. The tyrosine phosphatase Shp2 (PTPN11) in cancer. Cancer Metastasis Rev. 2008; 27: 179–192. pmid:18286234
  43. 43. Yu DH, Qu CK, Henegariu O, Lu X, Feng GS. Protein-tyrosine phosphatase Shp-2 regulates cell spreading, migration, and focal adhesion. Biol Chem. 1998; 273: 21125–21131. pmid:9694867
  44. 44. Stanford SM, Maestre MF, Campbell AM, Bartok B, Kiosses WB, Boyle DL, et al. PTPome profile of rheumatoid arthritis fibroblast-like synoviocytes: A novel role for SHP-2 as a modulator of invasion and survival. Arthritis Rheum. 2013; 65: 1171–1180.
  45. 45. Yan F, Sun P, Zhao H, Zhao C, Zhang N, Dai Y. Genetic association and functional analysis of rs7903456 in FAM35A gene and hyperuricemia: A population based study. Sci Rep. 2018; 8: 1–6. pmid:29942023
  46. 46. Fillion ML, Watt CL, & Gupta IR. Vesicoureteric reflux and reflux nephropathy: From mouse models to childhood disease. Pediatr Nephrol. 2014; 29: 757–766. pmid:24500705
  47. 47. Darlow JM, Dobson MG, Green AJ, Puri P, Barton DE. Investigation of DNA variants specific to ROBO2 Isoform ‘a’ in Irish vesicoureteric reflux patients reveals marked CpG island variation. Sci Rep. 2020; 10: 1–9. pmid:32041992
  48. 48. Sidore C, Busonero F, Maschio A, Porcu E, Naitza S, Zoledziewska M, et al. Genome sequencing elucidates Sardinian genetic architecture and augments association analyses for lipid and blood inflammatory markers. Nat Genet. 2015; 47: 1272–1281. pmid:26366554
  49. 49. Tu HP, Chen CJ, Tovosia S, Ko AM, Lee CH, Ou TT, et al. Associations of a non-synonymous variant in SLC2A9 with gouty arthritis and uric acid levels in Han Chinese subjects and Solomon Islanders. Ann Rheum Dis. 2010; 69: 887–890. pmid:19723617
  50. 50. Hollis-Moffatt JE, Gow PJ, Harrison AA, Highton J, Jones PB, Stamp LK, et al. The SLC2A9 nonsynonymous Arg265His variant and gout: Evidence for a population-specific effect on severity. Arthritis Res Ther. 2011; 13: R85. pmid:21658257
  51. 51. Jun JB, Na YI, Kim HJ, Kim S, Park Y, Kang J. Measurement of purine contents in Korean alcoholic beverages. J Korean Rheum Assoc. 2010; 17: 368–375.
  52. 52. Choi HK, Atkinson K, Karlson EW, Willett W, Curhan G. Alcohol intake and risk of incident gout in men: A prospective study. Lancet. 2004; 363: 1277–1281. pmid:15094272
  53. 53. Fukuuchi T, Yasuda M, Inazawa K, Ota T, Yamaoka N, Mawatari K, et al. A simple HPLC method for determining the purine content of beer and beer-like alcoholic beverages. Anal Sci. 2013; 29: 511–517. pmid:23665623
  54. 54. Puig JG, Fox IH. Ethanol-induced activation of adenine nucleotide turnover. Evidence for a role of acetate. J Clin Invest. 1984; 74: 936–941. pmid:6470146
  55. 55. Arteel GE, Raleigh JA, Bradford BU, Thurman RG. Acute alcohol produces hypoxia directly in rat liver tissue in vivo: role of Kupffer cells. Am J Physiol. 1996; 271: G494–G500. pmid:8843775
  56. 56. Hassen FS, Malik T, Dejenie TA. Evaluation of serum uric acid and liver function tests among pregnant women with and without preeclampsia at the University of Gondar Comprehensive Specialized Hospital, Northweat Ethiopia.PLos One. 2022; 17: e0272165. eCollection 2022.
  57. 57. Lee CH, Sung NY, Lee J, Bae SC. Factors associated with gout in South Koreans: analysis using the National Health Insurance Corporation and the National Health Screening Exam databases. Clin Rheumatol. 2013; 32: 829–837. pmid:23397143
  58. 58. Wang W, Krishnan E. Cigarette smoking is associated with a reduction in the risk of incident gout: Results from the Framingham Heart Study original cohort. Rheumatology (Oxford). 2015; 54: 91–95. pmid:25086327
  59. 59. Gee Teng GG, Pan A, Yuan JM, Koh WP. Cigarette smoking and the risk of incident gout in a prospective cohort study. Arthritis Care Res. 2016; 68: 1135–1142. pmid:26714165
  60. 60. Polidori MC, Mecocci P, Stahl W, Sies H. Cigarette smoking cessation increases plasma levels of several antioxidant micronutrients and improves resistance towards oxidative challenge. Br J Nutr. 2003; 90: 147–150. pmid:12844386
  61. 61. Villegas R, Xiang YB, Elasy T, Xu WH, Cai H, Cai Q, et al. Purine-rich foods, protein intake, and the prevalence of hyperuricemia: The Shanghai Men’s Health Study. Nutr Metab Cardiovasc Dis. 2012; 22: 409–416. pmid:21277179
  62. 62. Ellington A; 2007. Reduction of purine content in commonly consumed meat products through rinsing and cooking. Doctoral dissertation. Athens Ga
  63. 63. Kaneko K, Aoyagi Y, Fukuuchi T, Inazawa K, Yamaoka N. Total purine and purine base content of common foodstuffs for facilitating nutritional therapy for gout and hyperuricemia. Biol Pharm Bull. 2014; 37: 709–721. pmid:24553148
  64. 64. Kwon GH, Chun BY, Kwon YH. The prevalence of gout and its risk factors in healthy workers. J Kor Phys Ther. 2016; 28: 259–263.
  65. 65. Westing YH, Ekblom B, Sjödin B. The metabolic relation between hypoxanthine and uric acid in man following maximal short‐distance running. Acta Physiol Scand. 1989; 137: 341–345. pmid:2596329
  66. 66. Dudzinska W, Lubkowska A, Dolegowska B, Safranow K. Blood uridine concentration may be an indicator of the degradation of pyrimidine nucleotides during physical exercise with increasing intensity. J Physiol Biochem. 2010; 66: 189–196. pmid:20533099
  67. 67. Dudzinska W, Lubkowska A, Dolegowska B, Suska M, Janiak M. Uridine-an indicator of post-exercise uric acid concentration and blood pressure. Physiol Res. 2015; 64: 467–477. pmid:25470512
  68. 68. Jung KJ, Kim S, Yun M, Jeon C, Jee SH. Review on genetic risk score and cancer prediction models. J Health Info Stat. 2014; 39: 1–14