Genetic polymorphisms in very important pharmacogenomic variants in the Zhuang ethnic group of Southwestern China: A cohort study in the Zhuang population.

Pharmacogenomics, the study of the role of genetics in drug response, has recently become a focal point of research. Previous studies showed that genes associated with drug detoxification vary among different populations. However, pharmacogenomic information of the Zhuang ethnic group is scarce. The aim of the present study was to screen members of the Zhuang ethnicity in southwestern China for genotype frequencies of very important pharmacogenomic (VIP) variants and to determine the differences between the Zhuang ethnicity and other human populations.We genotyped 80 variants of VIP genes in 100 unrelated healthy Zhuang adults from the Yunnan province of China. Next, we analyzed the genotyping data with Structure and F-statistics (Fst).We compared our data with those of other populations using the HapMap data set, and observed that the frequency distribution of Zhuang population in Yunnan closely resembles that of JPT. Furthermore, population structure and Fst analysis showed that the Zhuang population is closely related to the Shaanxi Han population with respect to genetic background.Our study supplements existing information on Zhuang population pharmacogenomics and provides an extensive overview for developing personalized medicine.

Introduction

Drug response and reaction vary among individuals. A nationwide study conducted in Spain from 2001 to 2006 showed that 3.5 million people were hospitalized with adverse drug reactions (ADRs), and >5% of these patients eventually died.[ Pharmacogenomics focuses on the inheritance of individual variations in drug response, and eventually provides guidance to precision medical treatment.[ Since the term pharmacogenomics appeared in the literature in 1997, the number of articles identifying genetic variations is rapidly increasing.[ In 2005, a database called Pharmacogenetics and Pharmacogenomics Knowledge Base (PharmGKB) was established for sharing genotype, phenotype, or other data on genetic variation among researchers.[ Currently, PharmGKB is an easily accessible versatile knowledge database, which contains information on gene variant annotations, drug-centered pathways, and very important pharmacogenes (VIPs).

Evidence shows that genetic variant characteristics vary with populations or ethnicities.[ For example, CYP2C9, a member of the CYP450 superfamily, is an enzyme related to metabolism of many drugs such as diclofenac and warfarin. Reports show that the allele frequency of the CYP2C9∗2 allele is 15% among Caucasians, 1% to 3.6% among African Americans, but 0% among Asians. Thus, one of the major tasks in population pharmacogenetics and pharmacogenomics is to determine the frequencies of polymorphisms in drug detoxification genes among different ethnicities.[

China is the most populated country in the world. In addition to the Han people who make up 96% of the country's total population, there are 55 ethnic minority groups in China. Previously, we reported VIP variants in several Chinese ethnic groups, including the Deng, Han, Li, Lohoba, Kyrgyz, Miao, Mongol, Sherpa, Tajik, Uygur, and Tibetan population.[ According to the data of the sixth nationwide population census, the Zhuang, a minority with the largest population in China, has > 16 million people. Most Zhuang people live in Guangxi, Yunnan, Guangdong, Guizhou, and Hunan provinces. However, information on the VIP variants of the Zhuang ethnic group is limited. Therefore, identification of pharmacogenomic variants of Zhuang may extend our understanding of VIP gene variants among different populations.

In this study, we selected and genotyped 80 VIP variants in 100 Zhuang ethnic volunteers from the Yunnan province of China. Next, we compared the frequency differences between the selected Zhuang cohort and 11 major HapMap populations. Finally, Fst values were calculated to infer the population structure. Our results will supplement the existing VIP variant data of the Zhuang ethnic group, and may extend our understanding of ethnic diversity and pharmacogenomics.

Methods

Study subjects

We randomly recruited 100 unrelated healthy Zhuang adults (50 males and females each in the age range of 25–40 years) from the Yunnan province of Northwestern China and confirmed their ethnicity from lineage and birth place information. Written informed consent was obtained from all subjects. The study was performed in accordance with the Declaration of Helsinki and approved by the Human Research Committee of the Northwest University for Approval of Research Involving Human Subjects.

Variant screening and genotyping

We searched the PharmGKB database (https://www.pharmgkb.org/) and selected 80 genetic variants according to available data on frequency, functionality, and linkage based on published research. Genomic DNA was extracted from blood samples using Gold Mag-Mini whole blood genomic DNA purification kit (Gold Mag Ltd., Xi’an, China) according to the manufacturer's protocol. Optical density at 260 nm (OD260) was detected by spectrometry (DU530 UV/VIS spectrophotometer, Beckman Instruments, Fullerton, CA) to estimate DNA concentration. Multiplexed SNP MassEXTEND arrays were designed using the Sequenom MassARRAY Assay Design 3.0 software (San Diego, California).[ Genotyping of SNPs was conducted by Sequenom MassARRAY RS1000 (San Diego, California) according to manufacturer's instructions. Sequenom Typer 4.0 software was used for data collection and analysis as described previously.[

HapMap genotype data

We downloaded the genotype data of eleven populations from the International HapMap Project website (HapMap_release127) at http://hapmap.ncbi.nlm.nih.gov/biomart/martview/e4f42d4d0acde5ea6c35312381c1e461. The full names of the 11 populations are as follows:African ancestry in Southwest USA (ASW), Utah, USA residents with Northern and Western European ancestry from the CEPH collection (CEU), Han Chinese in Beijing, China (CHB), Chinese in metropolitan Denver, CO (CHD), Gujarati Indians in Houston, TX (GIH), Japanese in Tokyo, Japan (JPT), Luhya in Webuye, Kenya (LWK), Mexican ancestry in Los Angeles, CA (MEX), Maasai in Kinyawa, Kenya (MKK), Toscani in Italy (TSI), and Yoruba in Ibadan, Nigeria (YRI).

Statistical analysis

We used Excel and SPSS 19.0 statistical packages (SPSS, Chicago, IL) to conduct Hardy–Weinberg equilibrium (HWE) analysis and the χ2 tests. Validation of the frequency of each variant in the Zhuang people was tested by assessing the departure from HWE using an exact test. In this study, all the P values were calculated 2 sided, and the criterion of statistical significance was P < .05, or P < .000625 (.05/80) after Bonferroni's multiple adjustment.[ After χ2 tests, we selected 2 SNPs which showed more difference between Zhuang and the 11 other populations to perform a global allele frequency analysis. The allele data were downloaded from ALFRED (https://alfred.med.yale.edu/).

Structure analysis

To analyze the pairwise genetic distance among populations, pairwise Fst values were calculated using Arlequin v3.5. Population structure was investigated using the Bayesian clustering algorithm STRUCTURE ver. 2.3.1 (Pritchard Lab, Stanford University, http://pritchardlab.stanford.edu/structure.html).[ The data of the Deng, Han, Li, Lohoba, Kyrgyz, Miao, Mongol, Sherpa, Tajik, Uygur, and Tibetan populations were obtained from our previous studies. Analyses were performed using the ancestry model with correlated allele frequencies in runs from K = 6 to K = 10 (K is the number of genetically distinct clusters). The model choice criterion implemented in structure to detect the true K is an estimate of the posterior probability of the data for a given K, Pr(X|K). This value is obtained by first computing the log likelihood of the data at each step of the MCMC, called “Ln P(D).” To infer the number of clusters, ΔK was calculated using the method of Evanno.[. Graphs of STRUCTURE results were generated using Excel.

Population tree

A population tree was constructed using the Fst data and MEGA7 to infer the evolutionary relationship between the 12 populations.[ The evolutionary history was subsequently inferred using the neighbor-joining method.[

Results

We successfully genotyped 80 VIP variants selected from PharmGKB VIP in 100 members of the Zhuang population. The basic information of 80 selected variants is listed in Table 1, including those associated with related genes, families, phases, primary locus, alleles, alternative amino acids, and genotype frequencies of 100 Zhuang people. All the variants tested met the HWE.

Table 1

Basic information of selected SNP in the Zhuang people.

We first compared the allele frequency differences among the Zhuang ethnic group and the 11 groups selected from the International HapMap project database. In the ASW population, 22 of the selected VIP variants showed differences with Zhuang. The results of the other groups are as follows: CEU, 27; CHB, 11; CHD, 28, GIH, 29; JPT, 8; LWK, 26; MEX, 20; MKK, 24; TSI, 19; YRI, 34 (Table 2). To adjust for multiple comparisons, the level of statistical significance was reduced to 0.000625 (0.05/80), using which, the number of variants obtained with existing significant differences were as follows: ASW, 17; CEU, 21; CHB, 2; CHD, 26; GIH, 27; JPT, 3; LWK, 22; MEX, 7; MKK, 22; TSI, 15; YRI, 27. Obviously, the frequency distribution of the Zhuang population in Yunnan was similar to that of CHB, followed by JPT and MEX. In these different loci, rs7294 and rs689466 (located in VKORC1 and PTGS2, respectively) were significantly different in the Zhuang population compared to in other populations.

Table 2

Genotype frequency differences between Zhuang and 11 populations after multiple adjustment.

Among the 80 variants listed in Table 1, 67 variants could be classified as specific families. Based on the classification, the number of differing loci changed slightly as follows (after adjustment): ASW, 12; CEU, 15; CHB, 2; CHD, 23; GIH, 23; JPT, 3; LWK, 16; MEX, 4; MKK, 16; TSI, 9; YRI, 21 (Table 3).

Table 3

Significant variants in Zhuang compared to the other 11 populations after classification.

To further verify the ubiquitous differences between different groups, we downloaded the data pertaining to rs7294 and rs689466 from ALFRED (https://alfred.med.yale.edu/) and performed a global frequency analysis using the new data.

Figure 1A shows the global allele frequencies of rs7294. We observed that the A allele frequencies in the East Asian populations were lower than those in most of the other populations of the world. The frequencies of the Zhuang people (0.145) were relatively closer to those of the East Asian populations, especially the southern Chinese populations such as Yi (0.15) and Lahu (0.15). In addition, the Zhuang people showed frequency similar to those of other East Asian populations for rs689466 (Fig. 1B).

Figure 1

(A) Global allele frequency of rs7294. NA is short for North America; O is for Oceanic; S is for Siberia; SA is for South America. (B) Global allele frequency of rs689466.

Pairwise FST values were used to estimate and assess the magnitude of differentiation among geographic populations (0 indicating no divergence, 1 indicating complete separation). As shown in Table 4, the Fst values of the Zhuang and CHD population were the smallest (FST = 0.00884), followed by those of CHB (FST = 0.01701), and JPT (FST = 0.02057), indicating that the allele frequencies of the Zhuang and these 3 populations are similar. In addition, the highest divergence was observed for the MKK (FST = 0.21627) population.

Table 4

Pairwise Fst distances among the 12 populations.

Combining these results with our previous data on 11 Chinese ethnicities and 11 populations from the International HapMap project, a genetic structure was derived using STRUCTURE 2.3.1 (Fig. 2A). Individuals were divided into K clusters to display the genetic components. At K = 6, population structure was almost in accordance with the major geographic regions, and populations could be divided into 6 subgroups (subgroup 1: Zhuang, Shaanxi Han, CHB, CHD, and JPT; subgroup 2: ASW, LWK, MKK, and YRI; subgroup 3: Kyrgyz, Tajik, Mongol, and Uygur; subgroup 4: CEU, GIH, MEX, and TSI; subgroup 5: Deng, Lohoba, and Sherpa; subgroup 6: Miao, Li, and Tibetan) according to the clusters in each population. In addition, we inferred that Zhuang is most closely related to Shaanxi Han, followed by 3 other East Asian populations (CHD, CHB, and JPT) (Fig. 2A).

Figure 2

(A) Structure analysis of 23 populations. Each individual is represented by a vertical line which was partitioned into colored segments. K is the number of estimated clusters. ASW = African ancestry in Southwest USA; CEU = Utah, USA residents with Northern and Western European ancestry from the CEPH collection; CHB = Han Chinese in Beijing; CHD = Chinese in metropolitan Denver; GIH = Gujarati Indians in Houston; JPT = Japanese in Tokyo; LWK = Luhya in Webuye, Kenya; MEX = Mexican ancestry in Los Angeles, California, USA; MKK = Maasai in Kinyawa, Kenya; TSI = Toscani in Italy; YRI = Yoruba in Ibadan, Nigeria. Among them, CHB, CHD And JPT come from Asia; CEU, GIH, MEX and ASW come from America; TSI come from Europe; LWK, MKK and YRI come from Africa. (B) Evolutionary relationships of populations. The evolutionary history was inferred using the neighbor-joining method. The optimal tree with the sum of branch length = 0.26960109 is shown. The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. Evolutionary analyses were conducted in MEGA7.

Since individuals from the same population show similar ancestry proportions, statistical evaluation of the genetic relationships among populations is important. Therefore, we constructed a phylogenetic tree using the neighbor-joining method (Fig. 2B). Results indicate that ASW are located near the root of the tree, and JPT shares the closest evolutionary relationship with Zhuang.

Discussion

As early as 2002, a World Health Organization (WHO) report showed that more than half of the medicines were prescribed, dispensed, or sold inappropriately, while 50% of the patients did not consume them correctly. For safety issues, approximately 150 prescription drugs have been removed from the market since 1960, and certain new expensive drugs do not appear in the market.[ This is largely because of the individual differences in toxicity and response caused by pharmacogenomic variants. For providing more information on VIP variants of different ethnicities, we selected and genotyped 80 VIP variants in 100 Zhuang people from the Yunnan province of China, and conducted a series of statistical analyses. According to Fst values and STRUCTURE analysis, we speculated that Zhuang is most closely related to the Shaanxi Han population.

Comparison of genotype frequency distribution obtained from χ2 tests showed that rs7294, a SNP located in VKORC1 on chromosome 16, differs significantly among populations. VKORC1, a gene encoding vitamin K epoxide reductase complex 1, is the target of warfarin,[ a widely used anticoagulant prescribed for chronic atrial fibrillation, mechanical valves, pulmonary embolism, and dilated cardiomyopathy.[ To achieve the same anticoagulant effect (normally defined by the international normalized ratio, INR), individuals harboring polymorphisms in dosage-related genes require lower or higher warfarin doses than people carrying the wild type genes.[ For example, an investigation involving 279 patients of European ancestry on warfarin medication indicated that individuals with the TT genotype of rs7294 required 53% higher dose than individuals with the CC genotype.[ Similarly, reports show that in Chinese populations, patients with allele T of rs7294 require higher plasma concentration to achieve similar INR.[ Our previous study showed that the allele frequency of rs7294-T in Uygur from northwestern China and the Han ethnic group of Shaanxi province is lower than that in most of the other populations in the world. The same trend was observed in Zhuang people, suggesting that most Zhuang individuals may require lower dosage of warfarin.[

In addition to rs7294, the genotype frequency of rs689466 in the Zhuang population also shows notable differences with other populations. Rs689466 or COX-2 -1195G>A, is a SNP located in COX-2/PTGS2 (chromosome 1q25.2-3). COX-2 (cyclooxygenase-2), also known as prostaglandin-endoperoxide synthase 2, is composed of 10 exons and is a key enzyme that converts arachidonic acid to prostaglandins, which is involved in several important biological processes such as inflammation, immune function, cell proliferation, and angiogenesis.[ Evidence show that COX-2 is over-expressed in tumor tissue, whereas it is rarely detected in normal tissue.[ According to a meta-analysis involving 50,672 subjects in 2015, the A allele of rs689466 was associated with higher risk of cancer; in addition, a subgroup analysis by ethnicity showed it to be associated with high cancer risk in Asians.[ In contrast, the G allele is associated with high blood pressure in Asian individuals such as Koreans and Japanese.[ Consistent with the results obtained with the East Asian population, our data showed that the rs689466 A allele frequency of Zhuang (0.52) is relatively lower than those of other populations, especially European and African populations, indicating that Zhuang may have lower risk of cancer. Nevertheless, owing to their similarity with the Tibetan people, the G allele frequency of Zhuang is relatively higher, highlighting that Zhuang people should be cautious about the risk of developing hypertension, especially that caused by side effects of non-steroidal anti-inflammatory drugs (NSAIDs).[

Populations are usually defined based on culture or geography but not genetic relationships. Therefore, analysis of population structure may assist in investigating human evolutionary history.[ Comparison with 11 Chinese ethnicities showed that the genetic background of Zhuang closely resembles that of the Shaanxi Han population. In addition, we observed that some populations such as Miao have partial membership in multiple clusters, which may be caused by continuous gradations in allele frequencies across regions or admixture of neighboring groups.[

The Clinical Pharmacogenetics Implementation Consortium (CPIC) is a shared project of PharmGKB and the Pharmacogenomics Research Network (PGRN),[ which was established in 2009 for reducing the barrier in translating genetic laboratory test results into actionable prescribing decisions for affected drugs. Currently, CPIC guidelines are being developed and updated using established methods, including a rigorous review and grading of the relevant scientific literature.[ Our study may provide information for developing CPIC guidelines.

Conclusions

We identified the features of 80 VIP variants of Zhuang people from southwestern China, and observed that the genetic background of the Zhuang population of Yunnan is closest to that of the Shaanxi Han population. This study supplements the existing knowledge regarding different pharmacogenomic variants and may provide guidance for developing personalized medicine in future. However, our study has certain limitations. Our sample size was relatively small, and further studies with larger groups are necessary to verify the accuracy of our study.

Acknowledgment

We are grateful to the patients and control subjects for their participation in this study. We also thank the clinicians and hospital staff who obtained the blood samples and collected data for this study.

Author contributions

Conceptualization: Peng Chen, Bin Li, Tianbo Jin.

Data curation: Chenghao Guo.

Formal analysis: Jing Li, Tianbo Jin.

Investigation: Jing Li, Mengdan Yan, Fanglin Niu.

Project administration: Mengdan Yan.

Supervision: Tianbo Jin.

Writing – original draft: Chenghao Guo.

Writing – review & editing: Chenghao Guo.