Introduction

Human skin color is highly heritable and one of the most variable phenotypic traits that can vary dramatically within and across ethnic populations [1]. It is known that the human skin color is predominantly determined by pigments include melanin, hemoglobin (red), hemosiderin (brown), carotene (yellow), and bilirubin (yellow) [2]. Among those, the amount, type, and distribution of melanin play key roles in determining human skin pigmentation. Studies indicate that the human skin pigmentation in global populations is highly associated with latitude, and fundamentally, the distribution of ultraviolet (UV) radiation [3, 4]. Moreover, the researchers believe that geographic variation in skin pigmentation was influenced by the concerted action of different types of natural selection, including climate, lifestyle, diet, metabolism [1]. However, the genetic causes and environmental selective pressures underlying this range of skin color variation have remained largely unknown.

With the rapid development of genetics and genomics, researchers have gradually realized that the human skin color diversity is due to the natural positive selection of those genes that impact on human pigmentation, especially in the melanosome biogenesis or the melanin biosynthetic pathways [5, 6]. Recently, a large number of genome-wide association studies (GWAS) for pigmentation have been established and identified that some single nucleotide polymorphisms (SNPs) on TYR, IRF4, TYRP1, OCA2, SLC45A2, MC1R and KITLG genes are significantly associated with human skin color [7,8,9,10]. The solute carrier family 45, member 2 (SLC45A2) gene encodes the membrane associated transporter protein (MATP). The SLC45A2 protein expresses in melanocyte cell lines and mediates melanin synthesis by tyrosinase trafficking and proton transportation to melanosomes [11]. SLC45A2 mutations cause oculocutaneous albinism type IV (OCA4) and polymorphisms of SLC45A2 gene are associated with dark skin, hair, and eye pigmentation [12, 13]. In addition, the TYR gene encodes tyrosinase, a multifunctional enzyme that plays a major role in melanin biosynthesis in melanocytes [14]. TYR is commonly known as the albino locus since the homozygous or compound heterozygous mutations of this gene result in oculocutaneous albinism type 1 (OCA1), an autosomal recessive genetic disorder characterized by hypopigmented hair, skin and eyes [15].

However, the genetic causes and environmental selective pressures underlying this range of phenotypic variation have remained largely unknown. Therefore, to investigate whether the six polymorphisms in the two pigmentation related-genes SLC45A2 (rs11568737, rs28777 and rs183671) and TYR gene (rs1042602, rs1393350 and rs1126809) are associated with the geographic environmental variables, we selected randomly a total of 795 healthy individuals from eight ethnic groups in nine provinces in China, while collected the geographic environmental variables (altitude, longitude, latitude, air pressure, sunshine hours, and annual average temperature). The results of this study will improve our understanding of the impact of environmental variables in genetic differentiation and maintenance of genetic variation.

Results

A total of 795 samples including eight ethnic groups from nine provinces in China (Tibet-Tibetan accounted for 13.2%, Inner Mongolia-Ewenki 12.6%, Hainan-Han 6.2%, Ningxia-Hui 12.6%, Hainan-Li 12.5%, Inner Mongolia-Mongolian 12.6%, Guizhou-Miao 11.2%, Xinjiang-Uighur 13.3%, and Shaanxi-Han 5.9%) were collected to study the relationship between skin pigmentation-related gene variants and environmental variables. We also collected the detailed geographical environment information of different ethnic regions (Fig. 1), including altitude (m), longitude (°), latitude (°), atmosphere pressure (kPa), sunshine duration (hours), and year-round average temperature (°C), as shown in Table 1.

Fig. 1
figure 1

Distribution of allele frequencies of rs1042602 (A/C), rs28777 (A/C) and rs183671 (G/T) among different ethnic groups in China

Full size image
Table 1 Detailed geographical environment information of different ethnic regions
Full size table

The six SNPs on the two skin pigmentation-related gene SLC45A2 (rs11568737, rs28777 and rs183671) and TYR (rs1042602, rs1393350 and rs1126809) were successfully genotyped from 795 samples (call rate > 95%). The basic information (SNP-ID, chromosome number, position, alleles and gene name) and polymerase chain reaction (PCR) primer sequence (1st-PCRP, 2nd-PCRP and unique base extension primer sequence (UEP-SEQ) of the six SNPs was showed in Table 2. The minor allele frequency (MAF), genotype frequency and Hardy-Weinberg equilibrium (HWE)-P value of each SNPs are shown in Supplementary Table 1, Tables 2, and 3, respectively. The results showed that except for rs1393350 in TYR was not in accordance with the HWE in Uighur (P < 0.01), other five SNPs were in accordance with the HWE in the nine groups (P > 0.01).

Table 2 The basic information and primer sequence of SNPs
Full size table
Table 3 Differences in genotype distributions of SNPs among different ethnic groups
Full size table

In addition, we used the Chi-square test to evaluate the difference of genotype frequency distribution of the five SNPs among eight ethnic groups, as shown in Table 3. The results indicated that the genotype frequency distribution of rs28777 and rs183671 (SLC45A2) and rs1042602 (TYR) were significantly different between the Xinjiang-Uighur and other ethnic groups (P < 0.05). The allele frequency distribution of these three significantly different SNPs was shown in Fig. 1.

Simultaneously, we analyzed the relationship between SNP genetic diversity and environmental variables using Spearman correlation analysis (Table 4). It was found that the allele frequencies of rs28777-A (r = − 0.090, P = 0.011), rs183671-G (r = − 0.105, P = 0.003), rs1042602-A (r = − 0.108, P = 0.002), rs1126809-A (r = − 0.151, P < 0.001) were negatively correlated with the longitude. However, the positive correlation between the alleles frequencies of rs183671-G (r = 0.151), rs1042602-A (r = 0.157) and rs1126809-A (r = 0.138) and the latitude were extremely significant (P < 0.001). The alleles frequencies of rs183671-G (r = 0.116, P = 0.001), rs1042602-A (r = 0.105, P = 0.003) and rs1126809-A (r = 0.070, P = 0.048) were found to be significantly positively correlated with the sunshine hours. However, the alleles frequencies of rs183671-G (r = − 0.076, P = 0.033), rs1042602-A (r = − 0.079, P = 0.027) and rs1126809-A (r = − 0.076, P = 0.031) were significantly negatively correlated with the annual average temperature. The correlations between the allele frequencies of other SNPs and environmental variables were not significant. These findings indicate that environmental factors have selective pressure on these SNPs.

Table 4 The association between polymorphisms and geographic environmental variables
Full size table

Discussion

To investigate whether the two pigmentation related genes (SLC45A2 and TYR) polymorphisms are associated with the geographic environmental variables (altitude, longitude, latitude, and air pressure, sunshine hours, and annual average temperature), we selected randomly selected 795 healthy individuals from eight ethnic groups in nine provinces in China. The results of this study found that the genotype frequency distribution of rs28777 and rs183671 in SLC45A2 and rs1042602 in TYR were significantly different between the Xinjiang-Uighur and other ethnic groups (P < 0.05). Simultaneously, the rs28777, rs183671, rs1042602, rs1126809 polymorphisms were found to be correlated with the geographic environmental variables (longitude, latitude, sunshine hours or annual average temperature).

SLC45A2 (as also AIM1 or MATP) encodes a transporter protein that mediates melanin synthesis, which is expressed in a high percentage of melanoma cell lines. It has been reported that some SLC45A2 mutations cause OCA4 and polymorphisms of this gene were found to be significantly associated with human skin, hair, and eye pigmentation, and its mutation frequency varies significantly among the global population. Yuko Abe et al. found that rs11568737 in SLC45A2 (T500P) was significantly associated with melanin index [16]. A multi-stage GWAS of natural hair color in European ancestry found that rs28777 (SLC45A2) was associated with skin color and tanning ability [17]. A large Australian population-based case control study reveal that rs28777 exhibited the strongest crude association with risk of cutaneous malignant melanoma [18]. The study found that rs183671 (SLC45A2) was in strong linkage disequilibrium (LD) with rs16891982 (F374L) in CEU. A previous GWAS declared that the frequency of the rs183671 derived allele increased from Southern to Northern Europe, and this SNP was associated with skin pigmentation, and that each copy of the derived allele lightens the skin by 1.2 M index units [19]. Moreover, a previous GWAS demonstrated that the SNP rs183671 can explain skin color variation in three European studies RS, BTNS, and TwinsUK [20].

TYR is located at human chromosome 11q14.3, and encodes tyrosinase, which regulates the biosynthesis of melanin. Previous study demonstrated that mutations in TYR can cause OCA1 [15]. The non-synonymous polymorphism rs1042602 (Ser192Tyr) in TYR derived allele has specifically high frequency in Europe, and rs1042602 was significantly associated with eye color, freckles and lighter skin pigmentation [21,22,23,24]. It has been reported that rs1393350 was also associated with human hair, eye and skin color and tanning ability [23, 25,26,27]. A GWAS of melanoma conducted by the GenoMEL consortium identifies the locus rs1393350 associated with melanoma risk [28]. The rs1126809 variant is located in exon 4 of TYR gene and encodes a tyrosinase enzyme with an arginine-to-glutamine substitution at codon 402 (R402Q), and is a strong linkage with rs1393350 [29, 30]. The mutation of rs1126809 (A-G) causes the TYR enzyme to be thermosensitive, thus less active [31]. The rs1126809 has previously been used as a marker for skin pigmentation and also influence brown eye color formation [23, 30]. Previous GWAS indicated that the allele A of rs1042602 (TYR) was highly associated with lighter skin color in a South Asian descent population [32]. It has reported that the allele A of rs1042602 was over-represented in the Indo­Europeans population [33]. The two polymorphisms (rs1042602 and rs1126809) in TYR appear at high frequency in Europeans and are largely absent in African populations [34].

This study indicated that the genotype distribution of rs28777 and rs18367 in Xinjiang-Uighur was significantly different from other ethnic groups. Moreover, the allele frequencies of rs28777-A, rs183671-G, rs1042602-A, rs1126809-A were negatively correlated with the longitude; rs183671, rs1042602 and rs1126809 allele frequencies were positively associated with the latitude and the sunshine hours, while were negatively correlated with the annual average temperature in Chinese population. At present, there are few research reports on the association between genetic polymorphism and environmental factors. In 2010, Ji et al. [35] found that the disease-predisposition polymorphisms of the melatonin receptors were associated with sunshine duration in the global human populations. These results indicated that environmental factors had selective pressure on these loci, and their changes were related to environmental variables, that is, differences in selection caused by differences in environmental factors play an important role in genetic differentiation.

However, this study has some limitations that cannot be ignored. First, the sample size is small and the statistical power is relatively low. Second, this study is the first to explore the correlation between the allele frequencies of these six SNPs and geographical environmental factors. Third, we only selected six SNP loci on two genes to explore their correlation with geographical environmental factors. Finally, the effect of these genetic variations on human skin color diversity is not involved in this study. Therefore, we will further collect a larger sample and choose more SNPs and design functional experiments to explore the impact of environmental factors on genetic mutations.

In summary, this study results indicate that rs28777, rs183671 (SLC45A2) and 1,042,602 (TYR) polymorphisms were different among different populations. More importantly, our results confirm the idea that environmental factors have been an important selective pressure upon pigmentation related gene polymorphisms (rs28777, rs183671, rs1042602 and rs1126809). Further association and functional studies need to confirm our results in a large sample and explore the influence of geographical environment factors on the skin pigmentation-related genes polymorphisms and the mechanism of action.

Materials and methods

Study design

This study randomly selected a total of 795 healthy individuals from eight ethnic groups in nine provinces in China, including 105 Tibetan individuals, 100 Ewenki individuals, 49 Hainan Han individuals, 100 Hui individuals, 99 Li individuals, 100 Mongolian individuals, 89 Miao individuals, 106 Uighur individuals, and 47 Shaanxi-Han individuals. The basic situation of each population was shown in Table 1. The climate data (sunshine hours and annual average temperature) are quoted from China’s surface climate data in 2019. The information of altitude, longitude, latitude, and air pressure was collected through online query. Individuals who have a history of skin pigmentation-related diseases (albinism or melanoma), history of serious illness, mental illness, pregnancy were excluded from the study.

DNA extraction

The peripheral venous blood sample (5 mL) from each subjects were taken from fasting in the morning using the Ethylene diamine tetraacetic acid (EDTA) tube, and stored at − 20 °C refrigerator for further experiment. The GoldMag-Mini Whole Blood Genomic DNA Purification Kit (GoldMag. Co. Ltd., Xi’an, China) was used to extract genomic DNA, including blood lysis, adding GoldMag® gold magnetic particles to bind DNA, magnetic separation, washing, elution, magnetic separation to obtain DNA. In order to determine the concentration and purity of the extracted DNA, we use a spectrophotometer (Nanodrop 2000, Thermo Fisher Scientific, Waltham, MA, USA). If the ratio of OD260/OD280 ratios is about 1.8, the extracted DNA is qualified.

SNP selection and genotyping

We randomly selected the six SNPs (rs11568737, rs28777 and rs183671 in the SLC45A2 gene and rs1042602, rs1393350 and rs1126809 in the TYR gene) based on previously published genes related to pigmentation [18, 21, 28, 36,37,38,39,40]. The online software Agena Bioscience Assay Design Suite Version 2.0 (https://agenacx.com/online-tools/) was used to design the primers sequence (Table 2). The Agena MassARRAY platform (Agena Bioscience, San Diego, CA, USA) was used to genotype the six SNPs from 795 samples, according to the manufacturer’s instructions, including DNA sample preparation; polymerase chain reaction (PCR) amplification (95 °C pre-denaturation 2 min; 45 cycles (95 °C denaturation 30s, 56 °C annealing 30s, 72 °C extension 60s); 72 °C extension 5 min; 4 °C storage); shrimp alkaline phosphatase purification; Unique base extension primer (UEP) reaction; resin purification; spotting and mass spectrometry. Genotyping results data management and analysis using the Agena Bioscience TYPER software (version 4.0).

Statistical analysis

We used the Microsoft Excel (Microsoft Corp., Redmond, WA, USA) and Statistical Package for the Social Sciences (SPSS) version 25 (SPSS, Chicago, IL) to perform statistical analysis. The Chi-square test was used to evaluate whether each SNP was consistent with Hardy-Weinberg Equilibrium (HWE), and compare whether there are significant differences in the frequency distribution of genotypes among different ethnic groups. The relationship between SNP genetic diversity and environmental variables was analyzed using Spearman correlation analysis. All statistical analyses were two sided and the P < 0.05 was considered as statistically significant.