Association between leptin and leptin receptor gene polymorphisms and breast cancer risk in premenopausal and postmenopausal Mexican women

Purpose: This study aims to evaluate the association between leptin (-2548 G>A) and leptin receptor (p.K109R, p.Q223R, p.K656N) polymorphisms with breast cancer (BC) risk in premenopausal and postmenopausal Mexican women. Material and methods: 274 women were included, 76 with BC in premenopause, 66 with BC in postmenopause, 108 without BC in premenopause, and 22 without BC in postmenopause. Four polymorphisms were genotyped by PCR, High Resolution Melting, and DNA direct sequencing. χ, logistic regression analysis was performed with SPSS and XLSTAT software. Results: There was statistically significant risk of BC in postmenopausal women with LEPR p.Q223R+p.R223R (OR: 8.4, 95% CI: 2.7-25.6, P <0.001), and in premenopausal women with LEP-2548G>A+ -2548A>A genotype (OR: 1.7, 95% CI: 1.0-3.0, P = 0.04). Discussion and Conclusions: This is the first study from Mexico that includes four polymorphisms on LEP and LEPR genes. Our data indicate that both LEPR p.Q223R in postmenopausal women and LEP-2548G>A in premenopausal women have a strong association with BC, which suggests that these polymorphisms play a role in the development of BC in Mexican women. We detected coincidences and differences with respect to other populations. This highlights the importance of conducting genetic epidemiologic studies to identify the risk of gene polymorphisms and BC for each population.


INTRODUCTION
Breast cancer (BC) is a heterogeneous and complex disease due to an interaction between environmental factors and genetic susceptibility.The major risk factors for developing BC include being of female sex, receiving hormone replacement therapy during menopause, having received ionizing radiation, being of an early age at first menstruation, having children late or none at all, being of older age, having a family history of BC, and being obese (1).Obesity is a well-established risk factor for postmenopausal BC (2) The adipocyte dysfunction in obese individuals stimulates insulin resistance, ineffectual lipolysis, metabolic disease, and the release of mitogenic factors/ proinflammatory cytokines (3).
Leptin, a cytokine of the adipocytes encoded by the leptin (LEP) gene, interacts with the leptin receptor (LEPR), which is present in breast and other tissues (4).LEP and LEPR are overexpressed in BC tissue (5,6).LEP 2548G>A polymorphism seems to influence leptin mRNA expression in breast cancer cell lines affecting signaling capacity and receptor functionality, increasing the LEPR-LEP binding capacity (7,8).This induces a cascade of events in the tumor microenvironment, causing the proliferation, angiogenesis, and invasion of the cell (9,10).QR and RR genotypes of the LEPR p.Q223R polymorphism have been associated with modest risk for BC in nonobese premenopausal women (11), whereas QR and QR+QQ genotypes of the LEPR p.Q223R polymorphism with BC in overweight postmenopausal women (12).
The RR genotype of the LEPR p.K109R polymorphism has been associated with large tumors in obese premenopausal women (13).The LEP -2548G>A polymorphism has been associated with modest risk of BC in obese postmenopausal women (14).In a previous study, no statistical differences were found between BC and LEP 2548G>A in obese or non-obese Mexican women (15).

Patient population
The study included 274 women (mean age of 42.5±10.4years), 142 with BC (mean age of 47.3±9.1 years), and 132 healthy controls (mean age of 37.3±9.2years); 184 were premenopausal (mean age of 36.4±6.8 years) and 90 were postmenopausal (mean age of 53.2±5.9 years).The stage of the patients was III according to the breastcancer.org.All of the unrelated subjects were born in Mexico, with at least three generations of Mexican parents.All women resided in metropolitan areas.The protocol was approved by the Ethics Committee of the General Hospital of Mexico.Informed consent was obtained from all subjects.Eating habits, reproductive and menstrual histories, exogenous hormone use, familial history of cancer, physical activity, alcohol and smoking histories, weight in kilograms (Kg), height in meters (m), and body mass index (BMI = Kg/m 2 ) were determined in all subjects.

Statistical analyses
The agreement of genotype frequencies with Hardy-Weinberg equilibrium (HWE) expectations and allelic frequencies were analyzed with the chi-square test (χ 2 ).Binary logistic regression was used to determine OR (odds ratio), 95% confidence interval (CI), and significant differences (P value) of the genotypic frequencies among BC patients and healthy controls, between premenopausal versus postmenopausal women, and among BMI >30 versus BMI< 30.Postmenopausal status was considered by absence of menses for at least 1 year or according to the average of the Mexican population if the record was not obtained (23).For analysis, because patients with the homozygous minor allele or genotype were uncommon in our sample, these patients were included with their respective heterozygotes.This is consistent with analysis techniques reported elsewhere (24,25).P value <0.05 was considered statistically significant.These analyses were carried out with SPSS (v.22.0) and XLSTAT (v.2015) software.

Distribution of genotype and allele frequencies of LEP and LEPR gene polymorphisms
HRM curve and sequencing analysis of the four SNPs in the 274 Mexican women showed heterozygous and homozygous curves, suggesting the presence of allelic variation (Figure 1).HRM variation detection sensitivity was 100% concordant with 274 variations, as confirmed through DNA direct sequencing analysis.
The genotype distribution and allele frequencies of the LEP (-2548 G>A) and LEPR p.K109R, p.Q223R, p.K656N polymorphisms in all patients with BC, and in the healthy controls, are shown in Table 1.The genotype frequencies were in HWE in both patients and healthy controls for LEP (-2548 G>A) (P = 0.32, P = 0.31), LEPR p.Q223R, (P = 0.22, P = 0.40) and LEPR p.K656N (P = 0.75, P = 0.75); there was not Hardy-Weinberg equilibrium for LEPR p.K109R (P = 0.035, P = 0.042).Analyses to determine the association of LEP and LEPR polymorphisms with overall BC risk indicated statistical differences (p<0.05) with low OR in patients with BC, and in carriers of LEPR 109KR+RR and LEPR 223QR+RR genotypes.No statistical differences were observed for LEP (-2548G>A) and for LEPR p.K656N polymorphisms (Table 1).

Association between LEP and LEPR polymorphism with BC risk according menopausal status
Analyses of stratification of the premenopausal and postmenopausal patients to detect BC risk showed statistical differences for the LEPR 223QR+RR genotype, with a higher BC risk in postmenopausal women (OR: 8.4, 95% CI: 2.7-25.6,P < 0.001, whereas the LEP-2548 GA+AA genotype indicated statistical differences with a higher risk of BC in premenopausal women (OR: 1.7, 95% CI: 1.0-3.0,P = 0.04).No risk was observed with the remaining LEP and LEPR gene polymorphisms and BC (Table 2).

DISCUSSION
In this study, LEPR p.Q223R (QR+RR) established a statistical difference with a high OR for BC in postmenopausal women, and LEP -2548G>A (GA+AA) had a statistical difference with a higher OR in premenopausal women with BC.These data show the involvement of LEP and LEPR gene polymorphisms in BC risk in the Mexican population.
Contrary to a previous study in Mexican patients (15), our study detected a higher risk of BC in postmenopausal women with a BMI>30, suggesting obesity contributes significantly to the risk of BC in postmenopausal women, but not in premenopausal women (Table 3).This could be due to a bias in the sample selection in the previous Mexican study.
In other populations, the LEPR p.K109R polymorphism has been associated with BC risk in obese premenopausal women (13); however, a meta-analysis suggested that the LEP p.K109R polymorphism did not affect risk of BC (26).LEPR p.Q223R and LEP -2548G>A polymorphisms have been associated with BC in obese postmenopausal women (18,19), whereas LEPR p.Q223R and p.R223R polymorphisms are associated with BC in non-obese premenopausal women (14).Conversely, in our Mexican sample, LEPR p.Q223R was associated with BC in postmenopausal women; this demonstrates the importance of conducting these studies in several populations.
Five out of ten studies reported an association between the LEPR p.Q223R polymorphism with BC risk.In one of them, p.Q223R and p.R223R were associated with an increased BC risk and with shorter overall survival (16).Other studies reported an association between p.R223R with luminal A  subtype of BC (19), and tumor grade of BC (26), with poorly differentiated BC (20).In the present study, LEPR p.Q223R + p.R223R were associated with BC in obese postmenopausal women.Conversely, four studies showed no risk of BC with LEPR p.Q223R/p.R223R (14,17,27,28), emphasizing the importance of the relationship between genetic backgrounds with environmental/epigenetic factors.
On the other hand, the LEP (-2548G>A) polymorphism has been analyzed in ten studies; in two of them, the A allele was associated with BC (16,18,29).Seven studies showed no association between BC and the LEP -2548G>A polymorphism (9, 17-19, 27, 28).In our population, we found statistical differences in the LEP (-2548G>A) (GA+AA) polymorphism and BC in premenopausal women.
Breast cancer is a complex and multifactorial disease, and gen-gene and gene-environmental interactions may contribute significantly to its occurrence (30).Many non-infectious diseases are polygenic and involve polymorphisms that predispose people to obesity, glucose intolerance, insulin resistance, and a higher risk of developing BC (14).Major risk factors for BC and linkage disequilibrium of LEP and LEPR polymorphisms in each population play a role in BC risk (12,17,18,27,28).In our study, the combination of LEP -2548, LEPR p.K109R, and LEPR p.Q223R polymorphisms have a higher risk for BC with OR of 9.0, 95% CI 2.48-32-5, P = 0.001) (Table 4).This demonstrates the cumulative effect of the gene-gene interaction and the complexity of determining the BC risk (31,32).
The postmenopausal state involves hormonal changes such as decreases in estrogen.The importance of the LEP and LEPR gene polymorphisms in BC could be explained by the changes in the expression levels of leptin, with the effect on leptin receptors possibly at the transcriptional level (20,33).These changes stimulate the production of estrogen through the increase of aromatase expression in the cells of BC (34).Further studies of BC risk in several populations are necessary due to the different genetic background of each population.
In conclusion, we detected an association between LEP (-2548G>A) and BC in premenopausal women and LEPR p.Q223R with BC in obese postmenopausal women, which indicates that these polymorphisms play a role in the development of BC in Mexican women.One limitation of the present study is perhaps the sample size, even though there were statistical differences probably a larger sample could be a better indicator.We detected coincidences and differences with respect to other populations.

Figure 1 .
Figure 1.Melting curve analysis (A and C) and electropherograms of a partial sequence of the (B) LEP (-2548G>A) and (D) LEPR (p.Q223R) gene polymorphisms in the BC patients and healthy controls are showed.Wild-type homozygous, heterozygous and mutant homozygous genotypes are represented by colors different in the melting curves.The black arrows show the SNPs sites in the electropherograms.Abbreviations: R, Arginine; Q, glutamine.

Table 1
Genotypic and allelic frequency of LEP and LEPR polymorphisms and their associations with risks of breast cancer (BC) in patients with BC and healthy controls Abbreviations: P, patients with breast cancer; HC, healthy controls; OR, Odds ratio; CI, confidence interval, adjusted for age; n, number of patients; *P, statistical significance;

Table 3 .
Breast cancer risk according at menopausal status and BMI status

Table 2 .
Genotypic frequency distribution of LEP and LEPR Polymorphisms and BC risk according at menopausal status