Allele Frequencies of the Single Nucleotide Polymorphisms Related to the Body Burden of Heavy Metals in the Korean Population and Their Ethnic Differences.

This study was performed to select single nucleotide polymorphisms (SNPs) related to the body burden of heavy metals in Koreans, to provide Korean allele frequencies of selected SNPs, and to assess the difference in allele frequencies with other ethnicities. The candidate-gene approach method and genome-wide association screening were used to select SNPs related to the body burden of heavy metals. Genotyping analysis of the final 192 SNPs selected was performed on 1,483 subjects using the VeraCode Goldengate assay. Allele frequencies differences and genetic differentiations between the Korean population and Chinese (CHB), Japanese (JPT), Caucasian (CEU), and African (YIR) populations were tested by Fisher's exact test and fixation index (F ST), respectively. The Korean population was genetically similar to the CHB and JPT populations (F ST < 0.05, for all SNPs in both populations). However, a significant difference in the allele frequencies between the Korean and CEU and YIR populations were observed in 99 SNPs (60.7%) and 120 SNPs (73.6%), respectively. Ten (6.1%) and 26 (16.0%) SNPs had genetic differentiation (F ST > 0.05) among the Korean-CEU and Korean-YIR comparisons, respectively. The SNP with the largest F ST value between the Korean and African populations was cystathionine-β-synthase rs234709 (F ST: KOR-YIR, 0.309; KOR-CEU, 0.064). Our study suggests that interethnic differences exist in SNPs associated with heavy metals of Koreans, and it should be considered in future studies that address ethnic differences in heavy-metal concentrations in the body and genetic susceptibility to the body burden of heavy metals.

INTRODUCTION

It is well known that heavy metals induce adverse health effects in humans, including kidney damage, bone loss, neurological disorders, developmental abnormalities, vascular diseases, and cancer (1,2). Even the general population that does not have occupational exposure is chronically exposed to a low concentration of heavy metals because heavy metals are widely distributed in the environment (1,3). Heavy-metal concentration in the body is affected by various factors such as age, sex, smoking, diet, and nutritional status, and the environmental exposure level is a critical factor in determining the body burden of heavy metal (1,3,4). However, heavy metals go through the processes of absorption, distribution, metabolism, and excretion, in which a number of genetic factors are involved directly or indirectly. Therefore, in addition to environmental factors, genetic factors and their interactions may also play important roles in determining heavy-metal concentrations in the body (5). Previous studies reported that single nucleotide polymorphisms (SNPs) of a gene involved in iron metabolism were associated with not only the iron level but also with the lead and cadmium levels (6,7). Furthermore, in a twin study, the blood cadmium concentration was more strongly affected by genetics than by environmental factors (8). Therefore, genetic predisposition can play an important role in the body burden of heavy metals.

The blood cadmium and mercury levels in the general Korean population are approximately 2~4 times higher than the levels in the American population (9). Although consuming grains and shellfish was predicted to be a major factor in the heavy-metal high exposure levels of Korean populations (10), the general Korean population’s estimated total dietary intake of cadmium was not high compared to that of other nations and was considerably lower (about 30%) than the provisional tolerable weekly intake (11). This mismatch between external exposure and internal concentration indicates that there is the possibility that Koreans have a genetic predisposition associated with high absorption, low excretion, and high accumulation rates of heavy metals. Therefore, the goal of this study was to select SNPs related to the body burden of heavy metals, such as lead, mercury, cadmium, and arsenic, provide Korean allele frequencies of selected SNPs, and assess the difference in allele frequencies with other ethnicities.

MATERIALS AND METHODS

Study subjects

This study was based on a cohort established by the Korean Research Project on Integrated Exposure Assessment to Hazardous Materials for Food Safety (KRIEFS). The characteristics of this KRIEFS cohort and the method used to select the study subjects were described in detail in previous studies (12). Out of the 2,118 adults who enrolled in a KRIEFS cohort, 1,558 consented to participating in the genetic study. Among them, 71 subjects were excluded for the following reasons: incomplete data on heavy-metal exposure (n = 48) and insufficient blood sample (n = 23). Ultimately, 1,487 subjects were selected as study subjects. This study was approved by the Institutional Review Board of Dankook University Hospital, Republic of Korea (IRB No. 2013-03-008), and informed consent was obtained from all individual participants included in the study.

Selection of SNPs-related body burden of heavy metals in the Korean population and genotyping analysis

The candidate-gene approach method and genome-wide association screening using an exome chip were performed to select SNPs related to the body burden of heavy metals in the Korean population.

Candidate-gene approach

The genes involved in absorption, distribution, metabolism, and excretion of heavy metals were selected as candidate genes through a literature review, and databases search, such as Catalog of Published GWAS (13) and HuGE Navigator (14). SNPs located in the transcription regulatory region (promoter region or start codon) and the coding region (splice site, exon, or stop codon) of the selected candidate genes were selected as candidate SNPs using the Functional Element SNPs Database II (15). We searched the International HapMap Project database (HapMap Data Rel 27, population CHB and JPT/R-square cutoff 0.9, minor allele frequency cutoff 0.05) for the haplotype tagging SNP of each candidate gene and selected the candidate SNPs from this source.

Genome-wide association screening

After randomly selecting 500 people from the study subjects, genome-wide association screening was conducted using a Human Exome chipv1.2 (Illumina, San Diego, USA) in which 244,770 SNPs could be simultaneously analyzed. There were 783 SNPs not in Hardy-Weinberg equilibrium (HWE) (p < 0.001), and 309 SNPs had call rates of less than 95%. The average call rate of all samples was greater than 99.9%, with a minimum value of 99.4%. As a result of conducting a blind replication test on 20 randomly selected samples, the error rate of all samples was less than 0.05%, and the average concordance rate was 99.96%. For the SNPs located on autosomal chromosomes that satisfied the call rate (> 95%) and were in HWE (p > 0.001), the association with the marker of heavy-metal body burden (blood lead, blood cadmium, blood mercury, urinary cadmium and total arsenic) was evaluated by multiple regression analysis using the program PLINK, and 81 significant SNPs (p < 1.0 × 10−4) were selected.

Genotyping analysis

Ultimately, 192 SNPs were selected based on the candidate-gene approach method and genome-wide association screening. Genotyping analysis was performed on the selected 192 SNPs using the VeraCode Goldengate assay (Illumina, San Diego, CA, USA). An analysis was performed on 1,483 subjects who passed the DNA quality control (QC). The average call rate of the samples was 99.41%, and the average call rate of the SNPs was 99.38%. From 15 of the 192 total SNPs that were not in HWE, six SNPs with call rates less than 95% and two samples with call rates less than 95% were excluded from the final analysis. As a result of conducting a blind replication test on 19 randomly selected samples, high reproducibility was confirmed with an average concordance rate of 99.5%.

SNP frequencies in other ethnic populations

The frequencies of the selected SNPs in other ethnic populations were investigated using the Database of Single Nucleotide Polymorphisms (dbSNP build 142) and International Hap-Map DB (HapMap Data Rel #27 Phases I, II, and III). In this study, the gene frequencies in the Korean population were compared to those in four ethnic populations: Han Chinese individuals from Beijing, China (CHB), Japanese individuals from Tokyo, Japan (JTP), Caucasian individuals from Utah, USA of Northern and Western European ancestry from the Centre de’Etude du Polymorphism Humain-collection (CEU), and African Yoruba individuals in Ibadan, Nigeria (YRI).

Statistical analysis

HWE and allele frequency, as determined by the program PLINK, were used to analyze the data for 192 SNPs in the Korean individuals in this study. Based on the minor allele in the Korean population, the allele frequencies in each ethnic group were calculated. For the 163 SNPs that passed SNP QC, the difference in SNP frequencies between the Korean populations and other ethnic groups was compared using Fisher’s exact test. For each of the SNPs, we used Bonferroni correction for multiple tests and set the statistical significance threshold to p-value < 3.1 × 10−4 (0.05/163 SNPs = 3.1 × 10−4). Genetic differentiation among four ethnicities was measured by the Fixation index (FST), which describes the degree of population differentiation based on genetic polymorphisms (16). FST among a pairwise comparison between different ethnic groups was schematized with a Manhattan plot. FST at 0.05 to 0.15 was interpreted as moderate genetic differentiation, 0.15 to 0.25 was high genetic differentiation, and above 0.25 was very high genetic differentiation.

RESULTS

The study was conducted on 1,487 Korean subjects to calculate the allele frequencies of SNPs involved in the body burden of heavy metals, and their demographic characteristics and the level of heavy metals in subjects are presented in Table 1. The mean age of study subjects was 45.5 ± 14.5 years, 56.8% of all subjects was females. The geometric means of blood lead, mercury, cadmium levels in all subjects were 2.21 μg/dL, 4.05 μg/L and 1.06 μg/L, respectively. The geometric mean concentrations of cadmium and total arsenic in urine were 1.06, 102.7 μg/g creatinine, respectively.

Table 1

General characteristics of study subjects

		N (%)
Total subjects		1,487
Gender	Males	643 (43.2)
	Females	844 (56.8)
Age, mean ± std.		45.5 ± 14.5
Age groups	−29	255 (17.2)
	30~39	266 (17.9)
	40~49	341 (22.9)
	50~59	334 (22.5)
	60+	291 (19.6)
Smoking history	Never smokers	966 (65.0)
	Ex-smokers	243 (16.3)
	Current smokers	278 (18.7)
Alcohol use	Non-drinkers	362 (24.3)
	Drinkers	1125 (75.7)
Heavy metal levels*
Blood lead, unit: μg/dL		2.21 (2.17, 2.26)
Blood mercury, unit: μg/L		4.05 (3.91, 4.19)
Blood cadmium, unit: μg/L		1.06 (1.03, 1.09)
Urinary cadmium, unit: μg/g creatinine		1.09 (1.05, 1.13)
Urinary total arsenic, unit: μg/g creatinine		102.7 (98.03, 107.60)

Presented as geometric mean and 95% confidence intervals.

Table 2 shows the annotation information, minor allele frequency and selection rationale for the 192 selected SNPs. For the 163 SNPs that passed SNP QC, the allele frequency of minor (variant) alleles in the Korean population and the allele frequencies in CHB, JPT, CEU, and YIR were compared by pairwise comparison; the results are presented in Supplemental Table 1. Six SNPs (3.7%) showed a statistically significant difference in allele frequency between the Korean and CHB populations, and eight SNPs (4.9%) differed between the Korean and JPT populations. However, there was no genetic differentiation among populations because FST was less than 0.05 in all SNPs. In the allele frequency comparison between the Korean and CEU populations, significant differences were found in 99 SNPs (60.7%), and FST was above 0.05 in 10 SNPs (6.1%). In comparison between the Korean and YIR populations, 120 SNPs (73.6%) showed a significant difference in the allele frequency, and FST was above 0.05 in 26 SNPs (16.0%). Therefore, the biggest genetic divergence was observed between the Korean and YIR populations (Fig. 1).

Table 2

Information about the 192 SNPs and allele frequencies tested in this study

rs ID	Chr.	Gene	Location	Minor allele	MAF	Selection rationale	Related heavy metals
rs1948368	1	S1PR1/OLFM3	Intergenic	A	0.003	Exome chip based	Cd
rs714282	1	GPR177	Intron	A	0.419	Exome chip based	Cd
rs3736930	1	ATP6V1G3	Complex	T	0.057	Candidate gene approached	Cd
rs2666839	1	CENPF	Coding	T	0.163	Exome chip based	Cd
rs34545462	1	SLC2A7	Coding	T	0.050	Exome chip based	Hg
rs11265263	1	DUSP23/CRP	Intergenic	A	0.170	Exome chip based	Cd
rs13306731	1	SOAT1	Coding	G	0.380	Candidate gene approached	Cd, Hg
rs11118075	1	RRP15	Coding	C	0.070	Exome chip based	Hg
rs11805194	1	NUP133	Coding	C	0.140	Exome chip based	Cd
rs2479409	1	BSND/PCSK9	Intergenic	A	0.366	Exome chip based	Cd
rs35351292	1	LAPTM5	Coding	A	0.065	Exome chip based	Cd
rs41268474	1	C1orf68	Coding	A	0.068	Exome chip based	Pb
rs1284852	1	FLVCR1/VASH2	Intergenic	G	0.446	Candidate gene approached	Cd
rs58275168	1	SLC35F3	Intron	A	0.282	Exome chip based	Cd
rs1476413	1	MTHFR	Intron	A	0.176	Candidate gene approached	As
rs4845625	1	IL6R	Intron	T	0.443	Exome chip based	Pb
rs267733	1	ANXA9	Coding	G	0.077	Exome chip based	Pb
rs2698530	2	PELI1/HSPC159	Intergenic	A	0.350	Candidate gene approached	Cd, Pb
rs1457451	2	LOC729348/LOC100131818	Intergenic	A	0.172	Candidate gene approached	Cd
rs4664325	2	RBMS1	Intron	G	0.315	Exome chip based	Cd
rs12623234	2	MRPS9/GPR45	Intergenic	G	0.476	Exome chip based	Cd
rs1130609	2	RRM2	UTR	G	0.338	Candidate gene approached	Pb
rs2165738	2	NCOA1/ITSN2	Intergenic	G	0.387	Exome chip based	Hg
rs61197218	2	LOC100128572/IQCA1	Intergenic	A	0.271	Exome chip based	Hg
rs2287059	2	NOL10	Coding	T	0.114	Exome chip based	Hg
rs10455	2	CYBRD1	UTR	A	0.331	Candidate gene approached	Pb
rs3747673	3	TNK2	Coding	T	0.111	Exome chip based	Cd
rs2293232	3	MUC4	Coding	T	0.219	Exome chip based	Cd
rs3817672	3	TFRC	Coding	A	0.175	Candidate gene approached	Cd
rs72953098	3	C3orf30	UTR	G	0.067	Exome chip based	Hg
rs7640978	3	CMTM6	Intron	T	0.057	Exome chip based	Cd
rs832038	3	GABRR3	Intron	G	0.452	Candidate gene approached	Pb, Cd
rs6799969	3	RAD18/OXTR	Intergenic	G	0.358	Exome chip based	Cd
rs1799852	3	TF	Coding	T	0.218	Candidate gene approached	Cd, Pb
rs3804141	3	TFRC	Intron	A	0.212	Candidate gene approached	Cd
rs2718812	3	TOPBP1/TF	Intergenic	A	0.490	Candidate gene approached	Cd
rs1830084	3	TF/SRPRB	Intergenic	A	0.472	Candidate gene approached	Cd, Pb
rs75123867	3	CCDC50	Coding	T	0.048	Exome chip based	Cd
rs3811647	3	TF	Intron	A	0.419	Candidate gene approached	Cd
rs1561072	3	SOX2OT/ATP11B	Intergenic	C	0.180	Exome chip based	Hg
rs2276790	3	MFI2	Coding	T	0.061	Candidate gene approached	Cd
rs1049296	3	TF	Coding	T	0.266	Candidate gene approached	Cd
rs34193982	4	NEIL3	Coding	G	0.118	Exome chip based	Hg
rs74511500	4	FAT1	Coding	A	0.091	Exome chip based	Hg
rs11556167	4	PET112L	Coding	A	0.059	Exome chip based	Cd
rs4073	4	RASSF6/IL8	Intergenic	A	0.367	Candidate gene approached	As
rs2725264	4	ABCG2	Intron	G	0.219	Candidate gene approached	Hg
rs17208187	5	TMCO6	Coding	G	0.258	Exome chip based	Hg
rs7579	5	SEPP1	UTR	A	0.329	Candidate gene approached	Hg
rs3822751	5	GLRX	Intron	C	0.294	Candidate gene approached	As
rs2052550	5	ARSB	Intron	G	0.452	Candidate gene approached	Cd, Pb
rs3877899	5	SEPP1	Coding	-	0.000	Candidate gene approached	Hg
rs13188386	5	GHR/LOC100129630	Intergenic	-	0.000	Candidate gene approached	Cd, Pb
rs2354124	5	MRPL36/LOC728613	Intergenic	G	0.255	Exome chip based	Cd
rs1130435	5	FABP6	Complex	T	0.456	Exome chip based	Cd
rs3749779	5	SLC25A2	Coding	G	0.095	Exome chip based	Hg
rs1801394	5	MTRR	Complex	G	0.283	Candidate gene approached	Cd
rs3765467	6	GLP1R	Coding	T	0.252	Exome chip based	Hg
rs2301227	6	HLA-DPA1	Intron	C	0.073	Exome chip based	Cd, Hg
rs3129953	6	C6orf10/BTNL2	Intergenic	T	0.083	Exome chip based	Cd
rs76100089	6	LOC729792	Coding	T	0.203	Exome chip based	Hg
rs1800629	6	TNF/LTA	Intergenic	A	0.068	Candidate gene approached	Cd
rs17270561	6	SLC17A1	Intron	A	0.145	Candidate gene approached	Pb, Cd
rs13194984	6	BTN1A1/BTN2A1	Intergenic	T	0.007	Candidate gene approached	Cd, Pb
rs17342717	6	SLC17A1	Intron	T	0.008	Candidate gene approached	Cd, Pb
rs2071593	6	ATP6V1G2	UTR	T	0.084	Candidate gene approached	Hg
rs3957356	6	GSTA1/GSTA5	Intergenic	T	0.156	Candidate gene approached	Hg
rs932316	6	SCGN/LRRC16A	Intergenic	C	0.136	Candidate gene approached	Cd, Pb
rs12216125	6	HIST1H1A/TRIM38	Intergenic	T	0.122	Candidate gene approached	Cd, Hg
rs1799945	6	HFE	Complex	G	0.048	Candidate gene approached	Cd, Pb
rs9357283	6	DNAH8	Coding	A	0.314	Candidate gene approached	Cd
rs4516970	6	WTAP/SOD2	Intergenic	-	0.000	Candidate gene approached	Cd, Pb
rs2274089	6	LRRC16A	Intron	A	0.031	Candidate gene approached	Cd, Pb
rs1183201	6	SLC17A1	Intron	A	0.143	Candidate gene approached	Hg
rs17883901	6	GCLC/KLHL31	Intergenic	T	0.115	Candidate gene approached	Hg
rs2858881	6	HLA-DQB1/HLA-DQA2	Intergenic	G	0.048	Exome chip based	Hg
rs3736781	6	BTN1A1	Coding	G	0.314	Candidate gene approached	Hg
rs2142672	6	MYLIP/GMPR	Intergenic	C	0.264	Exome chip based	Pb
rs972275	6	LOC728666/RSPO3	Intergenic	G	0.458	Candidate gene approached	Cd, Pb
rs35868297	7	GALNTL5	Coding	C	0.196	Exome chip based	Cd
rs194524	7	STEAP2	Complex	A	0.213	Candidate gene approached	Pb
rs2718021	7	SEPT7/EEPD1	Intergenic	T	0.480	Exome chip based	Cd
rs13225097	7	LOC100288724/GIMAP4	Intergenic	G	0.188	Exome chip based	Cd
rs4722266	7	STK31	Complex	A	0.260	Exome chip based	Pb
rs13306698	7	PON1	Coding	G	0.086	Candidate gene approached	Cd
rs29880	7	INHBA/C7orf10	Intergenic	G	0.144	Candidate gene approached	Cd, Pb
rs662	7	PON1	Coding	A	0.355	Candidate gene approached	Pb
rs6971925	7	DGKB	Intron	T	0.078	Exome chip based	Cd
rs1106634	8	ATP6V1B2	Intron	A	0.211	Candidate gene approached	Hg
rs8191664	8	NEIL2	Complex	T	0.193	Exome chip based	Cd
rs11544484	8	TOP1MT	Coding	A	0.063	Exome chip based	Hg
rs4732748	8	ESCO2	Coding	T	0.200	Exome chip based	Cd, Hg
rs74846385	8	C8orf86	Coding	C	0.106	Exome chip based	Cd
rs17058207	8	SCARA5	Coding	G	0.320	Candidate gene approached	Pb, Cd
rs4872511	8	PPP3CC/SORBS3	Intergenic	T	0.084	Exome chip based	Pb
rs1800435	9	ALAD	Coding	C	0.073	Candidate gene approached	Pb
rs10818708	9	OR1N1	Coding	G	0.099	Exome chip based	Cd
rs3740393	10	AS3MT	Intron	C	0.253	Candidate gene approached	As
rs743572	10	CYP17A1	UTR	G	0.496	Candidate gene approached	As
rs1046778	10	AS3MT	UTR	C	0.385	Candidate gene approached	As
rs10749138	10	NRAP	Coding	T	0.419	Exome chip based	Hg
rs4462262	10	IPMK/ZWINT	Intergenic	T	0.078	Exome chip based	Hg
rs717620	10	ABCC2	UTR	A	0.222	Candidate gene approached	Hg
rs11191439	10	AS3MT	Coding	C	0.014	Candidate gene approached	As
rs10748835	10	AS3MT	Intron	A	0.491	Candidate gene approached	As
rs156697	10	GSTO2	Coding	C	0.259	Candidate gene approached	Cd
rs11191453	10	AS3MT	Intron	C	0.250	Candidate gene approached	As
rs7085104	10	C10orf32/AS3MT	Intergenic	G	0.435	Candidate gene approached	As
rs2297235	10	GSTO2	UTR	G	0.149	Candidate gene approached	As
rs4925	10	GSTO1	Coding	A	0.150	Candidate gene approached	As
rs2273697	10	ABCC2	Coding	A	0.080	Candidate gene approached	Cd
rs3740066	10	ABCC2	Coding	A	0.245	Candidate gene approached	Hg
rs3740390	10	AS3MT	Intron	A	0.250	Candidate gene approached	As
rs10891692	11	FAM55A	Coding	C	0.382	Exome chip based	Cd
rs1695	11	GSTP1	Coding	G	0.176	Candidate gene approached	Cd, Hg
rs4149182	11	SLC22A8	Intron	C	0.316	Candidate gene approached	Hg
rs11568496	11	SLC22A8	Coding	-	0.000	Candidate gene approached	Hg
rs45566039	11	SLC22A8	Coding	-	0.000	Candidate gene approached	Hg
rs77030286	11	SNHG1/SNORD28	Intergenic	-	0.000	Candidate gene approached	Hg
rs10047462	11	KIAA0999	Intron	G	0.499	Candidate gene approached	Cd, Pb
rs12362209	11	CCDC83	Coding	G	0.082	Exome chip based	Hg
rs236918	11	PCSK7	Intron	C	0.444	Candidate gene approached	Cd, Hg
rs4752805	11	PTPRJ	Intron	G	0.211	Exome chip based	Cd
rs4149170	11	SLC22A6	UTR	A	0.278	Candidate gene approached	Hg
rs1965	12	LOC341378/CKAP4	Intergenic	G	0.345	Candidate gene approached	Hg
rs12229654	12	LOC100131138/CUX2	Intergenic	G	0.139	Exome chip based	Pb
rs11111245	12	NAV3/SYT1	Intergenic	C	0.080	Exome chip based	Cd
rs2291075	12	SLCO1B1	Coding	T	0.422	Candidate gene approached	As
rs7975232	12	VDR	Intron	A	0.249	Candidate gene approached	Pb
rs2464196	12	HNF1A	Coding	C	0.454	Candidate gene approached	Pb
rs11066280	12	LOC100287871	Intron	A	0.178	Exome chip based	Pb
rs4304840	12	CLEC4D	Coding	G	0.160	Exome chip based	Hg
rs885389	12	GPR133	Intron	G	0.423	Exome chip based	Pb
rs1564370	12	SLCO1B1	Intron	C	0.259	Candidate gene approached	As
rs10842971	12	PZP	Coding	T	0.063	Exome chip based	Hg
rs17124715	12	LARP4	Complex	C	0.079	Exome chip based	Cd, Hg
rs757343	12	VDR	Intron	A	0.190	Candidate gene approached	Pb
rs1800802	12	ERP27/MGP	Intergenic	C	0.340	Candidate gene approached	Pb
rs671	12	ALDH2	Coding	A	0.158	Exome chip based	Pb
rs1544410	12	VDR	Intron	A	0.051	Candidate gene approached	Pb
rs60683621	12	OR6C70	Coding	G	0.489	Exome chip based	Hg
rs17278868	13	LATS2/SAP18	Intergenic	C	0.366	Exome chip based	Hg
rs636437	13	RFC3/NBEA	Intergenic	G	0.132	Exome chip based	Cd, Hg
rs973968	14	FLJ43390/KCNH5	Intergenic	G	0.059	Candidate gene approached	Cd
rs12879346	14	SLC7A8	UTR	T	0.486	Candidate gene approached	Hg
rs12588118	14	SLC7A8	Intron	G	0.096	Candidate gene approached	Hg
rs34691153	14	SLC7A8	Coding	-	0.000	Candidate gene approached	Hg
rs1130650	14	NP	Coding	T	0.227	Candidate gene approached	As
rs8005905	14	HSP90AA1	Coding	T	0.223	Candidate gene approached	Hg
rs2234636	14	SLC39A2	Coding	C	0.424	Candidate gene approached	As
rs11549465	14	HIF1A	Coding	T	0.053	Candidate gene approached	Cd, Hg
rs4984390	15	MCTP2	Intron	A	0.318	Exome chip based	Hg
rs55799438	15	C15orf56	Coding	G	0.047	Exome chip based	Cd
rs13180	15	IREB2	Coding	T	0.465	Candidate gene approached	Cd
rs11643815	16	MT4	Coding	A	0.004	Candidate gene approached	Hg
rs28366003	16	MT2A	UTR	G	0.127	Candidate gene approached	Cd
rs9936741	16	MT1M	UTR	C	0.069	Candidate gene approached	Hg
rs12919719	16	CDH1	Intron	G	0.164	Candidate gene approached	As
rs11076161	16	MT1A	Intron	A	0.292	Candidate gene approached	Cd
rs4148356	16	ABCC1	Coding	A	0.069	Candidate gene approached	Pb
rs35529209	16	ABCC1	Coding	-	0.000	Candidate gene approached	Hg
rs41395947	16	ABCC1	Coding	-	0.000	Candidate gene approached	Hg
rs33916661	16	SLC7A5/CA5A	Intergenic	G	0.119	Candidate gene approached	Hg
rs11075290	16	ABCC1	Intron	T	0.379	Candidate gene approached	Hg
rs10636	16	MT2A	UTR	C	0.266	Candidate gene approached	Cd
rs3785879	17	LOC100130148/MAPT	Intergenic	A	0.388	Candidate gene approached	Hg
rs78388447	17	EFCAB3	Complex	G	0.102	Exome chip based	Cd
rs242557	17	MAPT/LOC100130148	Intergenic	G	0.471	Exome chip based	Cd
rs542939	17	ABHD15	Coding	T	0.070	Exome chip based	Cd
rs7216284	17	GGT6	Coding	A	0.146	Candidate gene approached	Cd
rs312893	17	SEPT9	Intron	T	0.163	Exome chip based	Cd
rs3744807	17	PYCR1	UTR	T	0.048	Exome chip based	Hg
rs2660917	18	SOCS6/CBLN2	Intergenic	C	0.057	Candidate gene approached	Cd
rs2276199	18	PSTPIP2	Coding	G	0.439	Exome chip based	Pb
rs11555891	19	IRGC	Coding	A	0.132	Exome chip based	Hg
rs3745262	19	RAVER1	Coding	C	0.080	Exome chip based	Cd
rs10427027	19	PRDX2	Intron	C	0.077	Candidate gene approached	As
rs1644731	19	RDH8	Coding	A	0.439	Exome chip based	Cd
rs4452075	19	ZNF527	Coding	G	0.315	Exome chip based	Hg
rs1043673	19	NLRP2	Coding	A	0.225	Candidate gene approached	Cd
rs3761144	20	GSS/MYH7B	Intergenic	C	0.463	Candidate gene approached	Hg
rs1056720	20	CDC25B	Complex	T	0.331	Candidate gene approached	Cd
rs2762934	20	CYP24A1	UTR	A	0.114	Exome chip based	Cd
rs4925386	20	LAMA5	Intron	T	0.225	Exome chip based	Cd
rs62200482	20	FERMT1	Coding	A	0.071	Exome chip based	Cd
rs6126559	20	VSTM2L	Intron	A	0.472	Exome chip based	Pb
rs4920037	21	CBS	Intron	A	0.026	Candidate gene approached	As
rs234709	21	CBS	Intron	T	0.091	Candidate gene approached	As
rs855791	22	TMPRSS6	Coding	C	0.106	Candidate gene approached	Cd, Pb
rs987710	22	PRAMEL/VPREB1	Intergenic	G	0.310	Candidate gene approached	Cd, Pb
rs4820268	22	TMPRSS6	Coding	G	0.490	Candidate gene approached	Cd, Pb
rs2430212	X	KLHL13	Intron	C	0.299	Candidate gene approached	Cd, Pb

Chr.: chromosome, MAF: minor allele frequency, UTR: untranslated region.

Fig. 1

Genetic differentiation between Korean and other ethnic populations. A: Korean versus Chinese (CHB). B: Korean versus Japanese (JPT). C: Korean versus Caucasian (CEU). D: Korean versus African (YIR).

Table 3 shows that 31 SNPs had FST above 0.05 at least once in a pairwise comparison between ethnic groups. The SNP with the largest FST value between the Korean and CEU populations was rs636437, which is located in the intergenic region between replication factor C subunit 3 (RFC3) and neurobeachin (NBEA) (FST: KOR-CEU, 0.255; KOR-YIR, 0.209). The SNP with the largest FST value between the Korean and African populations was cystathionine-β-synthase (CBS) rs234709 (FST: KOR-YIR, 0.309; KOR-CEU, 0.064). The three SNPs had FST above 0.05 both in pairwise comparison between the Korean and CEU populations and between the Korean and YIR populations [vitamin D receptor (VDR) rs1544410 (FST: KOR-CEU, 0.136; KOR-YIR, 0.061), inositol polyphosphate multikinase/ZW10 interacting kinetochore protein (IPMK/ZWINT) rs4462262 (FST: KOR-CEU, 0.082; KOR-YIR, 0.207), and mitochondrial topoisomerase I (TOP1MT) rs11544484 (FST: KOR-CEU, 0.051; KOR-YIR, 0.190)].

Table 3

Allele frequencies and fixation index (FST) among different ethnics for selected 31 SNPs

SNP ID	Gene symbol	Chr.	Referent/variant allele*	Variant allele* frequency					KOR versus CHB		KOR versus JPT		KOR versus CEU		KOR versus YIR
				KOR	CHB	JPT	CEU	YIR	P†	FST	P†	FST	P†	FST	P†	FST
rs2479409	BSND/PCSK9	1	T/C	0.37	0.32	0.39	0.65	0.79	0.115	0.0008	0.518	0.0002	8.1 × 10−17	0.0225	5.9 × 10−46	0.0620
rs10455	CYBRD1	2	G/A	0.33	0.33	0.40	0.73	0.96	1.000	0.0001	0.047	0.0014	1.0 × 10−32	0.0467	4.2 × 10−105	0.1343
rs1130609	RRM2	2	A/G	0.34	0.37	0.35	0.74	0.98	0.573	0.0001	0.819	0.0001	1.5 × 10−19	0.0283	1.1 × 10−52	0.0695
rs2698530	PELI1/HSPC159	2	T/C	0.35	0.37	0.36	0.72	0.90	0.467	0.0002	0.718	0.0001	8.7 × 10−28	0.0388	8.9 ×10−79	0.1035
rs61197218	LOC100128572/IQCA1	2	T/G	0.27	0.32	0.28	0.04	0.86	0.114	0.0008	0.931	0.0000	2.1 × 10−15	0.0148	6.8 × 10−56	0.0860
rs1561072	SOX2OT/ATP11B	3	G/A	0.18	0.19	0.15	0.10	0.78	0.565	0.0003	0.368	0.0004	0.001	0.0033	1.6 × 10−97	0.1643
rs1830084	TF/SRPRB	3	G/A	0.47	0.58	0.50	0.65	0.91	3.9 × 10−4	0.0039	0.447	0.0002	4.0 × 10−7	0.0080	9.2 × 10−53	0.0626
rs3817672	TFRC	3	T/C	0.18	0.15	0.19	0.60	0.14	0.451	0.0002	0.651	0.0001	3.7 × 10−42	0.0736	0.166	0.0006
rs7640978	CMTM6	3	T/C	0.06	0.05	0.05	0.10	0.31	0.785	0.0003	0.764	0.0004	0.012	0.0026	5.4 × 10−36	0.0736
rs2725264	ABCG2	4	T/A	0.22	0.23	0.19	0.05	0.92	0.760	0.0000	0.357	0.0002	3.6 × 10−11	0.0109	1.7 × 10−132	0.2004
rs4073	RASSF6/IL8	4	T/C	0.37	0.41	0.33	0.39	0.86	0.283	0.0004	0.297	0.0004	0.513	0.0001	7.3 × 10−40	0.0546
rs2142672	MYLIP/GMPR	6	T/C	0.26	0.29	0.20	0.69	0.26	0.392	0.0003	0.033	0.0015	3.7 × 10−38	0.0587	0.835	0.0001
rs2858881	HLA-DQB1/HLA-DQA2	6	T/C	0.05	0.05	0.12	0.01	0.24	0.767	0.0000	3.8 × 10−5	0.0063	0.003	0.0019	4.2 × 10−26	0.0516
rs11544484	TOP1MT	8	G/A	0.06	0.08	0.05	0.30	0.53	0.198	0.0009	0.773	0.0005	6.4 × 10−25	0.0513	1.4 × 10−86	0.1900
rs10818708	OR1N1	9	T/C	0.10	0.13	0.09	0.58	0.15	0.141	0.0009	0.727	0.0002	1.1 × 10−61	0.1300	0.015	0.0021
rs156697	GSTO2	10	C/A	0.26	0.27	0.29	0.39	0.83	0.719	0.0001	0.307	0.0005	8.9 × 10−5	0.0054	8.8 × 10−85	0.1262
rs4462262	IPMK/ZWINT	10	A/G	0.08	0.05	0.03	0.42	0.61	0.190	0.0009	0.002	0.0029	7.8 × 10−39	0.0820	6.4 × 10−98	0.2073
rs4752805	PTPRJ	11	A/G	0.21	0.28	0.19	0.16	0.98	0.112	0.0008	0.791	0.0000	0.187	0.0005	8.5 × 10−75	0.1245
rs11111245	NAV3/SYT1	12	G/T	0.08	0.09	0.09	0.00	0.46	0.487	0.0000	0.612	0.0000	5.5 × 10−5	0.0032	2.2 × 10−57	0.1166
rs1544410	VDR	12	G/A	0.05	0.04	0.11	0.44	0.27	0.383	0.0012	0.001	0.0046	5.7 × 10−56	0.1364	4.7 × 10−30	0.0611
rs2464196	HNF1A	12	T/A	0.45	0.52	0.38	0.70	0.90	0.031	0.0015	0.037	0.0014	6.6 × 10−13	0.0160	8.9 × 10−54	0.0652
rs4304840	CLEC4D	12	A/G	0.16	0.15	0.12	0.22	0.62	0.730	0.0000	0.072	0.0008	0.032	0.0013	1.1 × 10−61	0.1070
rs636437	RFC3/NBEA	13	C/T	0.13	0.17	0.14	0.90	0.76	0.097	0.0010	0.608	0.0003	2.4 × 10−133	0.2552	5.4 × 10−113	0.2089
rs973968	FLJ43390/KCNH5	14	A/G	0.06	0.04	0.08	0.17	0.27	0.345	0.0000	0.313	0.0000	4.2 × 10−8	0.0119	1.3 × 10−26	0.0508
rs55799438	C15orf56	15	G/A	0.05	0.06	0.02	0.41	0.05	0.298	0.0008	0.062	0.0016	8.6 × 10−41	0.1078	1.000	0.0005
rs312893	SEPT9	17	A/G	0.16	0.21	0.19	0.00	0.63	0.062	0.0012	0.226	0.0005	1.6 × 10−15	0.0129	1.8 × 10−63	0.1099
rs2660917	SOCS6/CBLN2	18	C/A	0.06	0.10	0.05	0.25	0.30	0.015	0.0023	0.764	0.0003	6.5 × 10−19	0.0376	6.4 × 10−33	0.0667
rs10427027	PRDX2	19	G/A	0.08	0.07	0.08	0.10	0.56	0.553	0.0002	0.699	0.0001	0.304	0.0005	2.6 × 10−85	0.1802
rs234709	CBS	21	G/A	0.09	0.12	0.15	0.44	0.93	0.159	0.0004	0.012	0.0020	2.8 × 10−30	0.0637	1.3 × 10−135	0.3093
rs4920037	CBS	21	C/T	0.03	0.01	0.03	0.23	0.16	0.315	0.0004	0.666	0.0000	2.5 × 10−27	0.0673	1.4 × 10−18	0.0379
rs2430212	KLHL13	X	A/G	0.30	0.36	0.39	0.24	0.91	0.279	0.0005	0.105	0.0010	0.186	0.0007	5.3 × 10−72	0.1028

Chr.: chromosome, KOR: Koreans in this study, CHB: Han Chinese in Beijing, China, JTP: Japanese in Tokyo, Japan, CEU: Utah residents with Northern and Western European ancestry from the CEPH collection, YRI: Yoruba in Ibadan, Nigeria.

Variant allele defined as the minor allele in the Korean population.

P value calculated by Fisher’s exact test.

DISCUSSION

Our interethnic comparison study for SNPs related to the body burden of heavy metals revealed that Koreans were genetically very similar to other East Asians, including Chinese and Japanese individuals but considerably different from Caucasian and African individuals. This result was consistent with the ethnic differences in previous studies on SNPs associated with asthma (17), pharmacogenesis (18), and autoimmunity (19), although direct comparison is impossible because the studied SNPs differed. The ethnic differences in SNPs are affected by genetic drift, migration, and natural selection, and verifying these differences will help us better understand the ethnic variations in disease susceptibility and phenotypes as well as complex genetic-environment interactions (20).

There are several studies reported that the body concentration of heavy metals differs across ethnicity (21,22). The U.S. National Health and Nutrition Examination Survey (NHANES) report shows that the body concentration of heavy metals in Asians was higher than in all other ethnic populations, especially for cadmium, mercury, and arsenic (23). Blood cadmium, mercury and the urinary total arsenic levels in our cohort subjects were about two, five and ten times greater than those in the U.S. population, respectively (23). Until now, it mainly focused on the ethnic differences in environmental factors including dietary habit to explain for this variation. However, our study is the first to verify the ethnic divergence in SNPs that may be related to heavy metal body burden in Koreans.

In this study, CBS rs234709 showed the highest FST value compared between Korean and African individuals (FST = 0.309), and moderate genetic differentiation was observed for both CBS rs234709 and rs4920037 in the comparison between Korean and Caucasian individuals. CBS gene was-selected as a candidate gene because of the association with arsenic metabolism (24). CBS enzyme catalyzes the synthesis of cystathionine from homocysteine. A decrease in CBS activity is associated with the increases in homocysteine concentration in the body. Elevated homocysteine can deplete S-adenosylmethionine which is a methyl donor. Therefore, a modulation in CBS activity by genetic variation might affect methylation capacity in human (24–26). Recently, the evidence for this mechanism has been reported that CBS rs234709 or rs4920037 variant allele were associated with an increased in monomethylarsonous acid (a less-methylated form of arsenic metabolites), while with a decrease in dimethylarsinic acid (a more-methylated form) (25,26). That is, interethnic genetic variations in enzymes involved in arsenic metabolism can affect interethnic differences in methylation capacity, which results in ethnic differences in urine arsenic methylated metabolite compositions (26,27).

In this study, there was a genetic variation between Korean and CEU populations in Transferrin receptor 1 (TFRC) rs3817672 (FST = 0.0736), which is involved in iron absorption, and VDR rs1544410 (FST = 0.1364), which is involved in calcium absorption. Because heavy metals such as cadmium and lead are not metabolized in the body, interactions with various essential minerals during absorption and excretion processes can act as an important factor that affects body burden. Deficiency of essential metals such as iron, calcium, and zinc in the body increases absorption of heavy metals such as cadmium and lead (4). Genetic factors associated with iron homeostasis were identified by several GWAS studies (28), and the association between SNPs associated with iron homeostasis and urine cadmium concentration in non-smoking women was reported (7).

Comparison between Korean and CEU populations and between Korean and YIR populations revealed intergenic SNPs, including RFC3/NBEA rs636437 and IPMK/ZWINT rs4462262, with FST values that indicated moderate genetic differentiation. No studies on these two SNPs and body burden of heavy metals have been conducted to date, and the functions of these SNPs have not been identified. Only the association of IPMK/ZWINT rs4462262 with diabetes retinopathy was reported by a Taiwanese GWAS study (29).

To our knowledge, this is the first report on ethnic differences in SNPs associated with the body burden of heavy metals. In this study, we presented the Koreans allele frequencies of SNPs highly associated with the body burden of heavy metals, which were selected using a candidate-gene approach and GWAS in Korean individuals, and compared the allele frequencies with those of Caucasian, African, and other ethnic Asian populations. Compared with other ethnic Asian populations such as Chinese and Japanese people, Korean individuals were not genetically different (FST < 0.05). However, compared to the Caucasian and African populations, significant differences in allele frequencies were confirmed in more than 60% of the SNPs analyzed in this study, and high genetic divergence (FST > 0.05) was observed in ten (6.1%) and 26 (16.0%) SNPs, respectively. Because there have not been many studies on the genetic effects of the body burden of heavy metals to date, ethnic differences in SNPs associated with heavy metals confirmed in this study should be considered in future studies that address ethnic differences in heavy-metal concentrations in the body and genetic susceptibility to the body burden of heavy metals.