Association between genetic polymorphisms of leptin receptor and preeclampsia in Chinese women

Abstract Objective Leptin signaling plays an important role in regulating metabolism and reproduction. In the present study, we investigated the relationship between polymorphisms of leptin receptor (LEPR) gene A223G and A668G and preeclampsia (PE) and evaluated influences of genotypes on clinical, metabolic, and oxidative stress indices in Chinese women. Methods This is a case-control study including 322 patients with PE and 1295 healthy pregnant women. The two polymorphisms were genotyped by polymerase chain reaction-restriction fragments length polymorphism method. Clinical and biochemical parameters were analyzed. Results The frequencies of the AA + AG genotypes (28.6% vs. 36.1%) and A allele (14.9% vs. 19.8%) of LEPR A223G polymorphism, and those of the AA + AG genotypes (17.7% vs. 24.6%) and A allele (9.0% vs. 12.9%) of LEPR A668G polymorphism were significantly lower in the PE group than those in the control group. The 223A and 668A alleles were protective factors against PE in the regression model, which included age and delivery body mass index as covariates (OR = 0.684, 95% CI: 0.506–0.926, p = .014; OR = 0.650, 95% CI: 0.456–0.927, p = .017, respectively). When the 668GG/223GG combined genotype served as the reference category, the 668A/223A combined allele had further enhanced the protective effect on PE (OR = 0.558, 95% CI: 0.374–0.833, p = .004). Patients possessing the LEPR 223A allele had higher total antioxidant capacity and lower oxidative stress index (p < .05), while those with the LEPR 668A allele had higher high-density lipoprotein cholesterol levels (p = .045) compared with those carrying the corresponding GG genotype. Conclusions The 223A and 668A alleles of LEPR polymorphisms are genetic protective factors for PE in Chinese women. The two alleles may exert a beneficial effect on oxidative stress and lipid metabolism in patients.


Introduction
Preeclampsia (PE) is a severe disease that occurs during pregnancy characterized by new onset hypertension after 20 weeks of gestation, involving one or more organs and systems [1]. PE affects approximately 2-8% of pregnant women worldwide and causes over 70,000 maternal deaths and 500,000 fetal deaths every year [1,2]. The etiology and pathogenesis of PE remains unknown, but mounting evidence hints that the onset and development of PE may be related to angiogenesis disorder, increased oxidative stress, inflammation, liver dysfunction, and genetic factors [2][3][4][5][6].
Leptin, a peptide hormone comprising 167 amino acids, is mainly secreted in white adipose tissue and placenta, and functions by binding to leptin receptor (LEPR) [7,8]. Leptin was initially identified as a hormone that regulates food intake and body weight; however, recent studies revealed that it also plays an important role in modulating vascular function, immune system, inflammatory response, and reproductive process [8][9][10]. Leptin can affect blastocyst formation, implantation, and placentation [7,8], induce gonadotrophin production in trophoblast cells, modulate proliferation, protein synthesis, invasion, and apoptosis in placental cells, and regulate lipid mobilization during pregnancy [7,8,11,12].
Leptin signaling and oxidative stress plays a significant role in the pathogenesis of PE [4,7,8,25]. Circulating oxidative stress and leptin levels are elevated in patients with PE [4,[25][26][27][28]. The relationship between LEPR A223G and A668G SNPs and PE has been reported in different populations; however, the results are inconsistent and have been obtained from limited sample size. The A allele of A223G SNP was protective factor for severe PE in Chinese women (PE, n ¼ 207; controls, n ¼ 252) [27], whereas this SNP was not associated with the risk of severe PE in Hungarian women (PE, n ¼ 124; controls, n ¼ 107) [29]. The G allele of A668G SNP was associated with an increased risk of PE in Sudanese (PE, n ¼ 122; controls, n ¼ 122) or Hungarian women [29,30]. However, no significant difference in the allele and genotype frequencies of A668G SNP was observed between the PE/pregnancy induced hypertension (n ¼ 61) and control (n ¼ 40) groups in Sri Lankan population [31]. So far, no related reports have yet revealed whether A668G SNP is associated with the risk of PE in Chinese population. Further, the association between the two SNPs and oxidative stress is relatively unknown. Therefore, we examined the association of the aforementioned two genetic variations with PE and assessed influences of genotypes on clinical, metabolic, and oxidative stress parameters using a relatively large sample size of Chinese women.

Study participants
This case-control study comprised 322 patients with PE and 1295 healthy pregnant women enrolled between 2006 and 2021. In the present study, the genotype and allele frequencies of LEPR SNPs are main variables and the sample sizes are practicable and reasonable based on a report by B-Rao [32]. All participants were recruited at the Department of Obstetrics of West China Second University Hospital in Chengdu, and provided written informed consents. This study was approved by the Institutional Review Board of the West China Second University Hospital of Sichuan University (2015-016 to Xinghui Liu; 2020-036 to Ping Fan).
PE was defined according to the American College of Obstetricians and Gynecologists Practice Bulletin [1]. A detailed description of the diagnostic criteria can be found in our previous published article [33]. The control group included women without any complications during pregnancy recruited from the same hospital and during the same period.
The exclusion criteria included patients with diabetes mellitus, twin/multiple pregnancies, systemic inflammatory disorders, cardiac, renal, and hepatic disease. Women that underwent in vitro fertilization and had premature delivery were also excluded from the control group.
The clinical variables of all participants, including maternal age, pre-pregnancy body mass index (BMI, kg/m 2 ), delivery BMI, systolic blood pressure (SBP), diastolic blood pressure (DBP), and gestational age, as well as birth height, weight, and ponderal index (PI, g/cm 3 ) of infants of these participants were measured and estimated. Blood sample from participants was obtained in the third trimester of pregnancy or before delivery after at least 8 h of fasting, transferred on ice, and centrifuged at 1500 Â g and 4 C for 15 min within 2 h. Serum and plasma aliquots were stored at À80 C. Blood cells were stored at 4 C before DNA extraction.

Genotype analysis
Genomic DNA was isolated from leukocytes in the peripheral blood of participants. Genotypes of LEPR A223G and A668G were determined by PCR restriction fragment length polymorphism. The following PCR primers were designed using Primer-BLAST: forward 5 0 -AACAACTTTCCACTGTTGCTTTC-3 0 and reverse 5 0 -AGAATTTACTGTTGAAACAAATGGC-3 0 for A223G SNP; forward 5 0 -TGAATGTCTTGTGCCTGTG-3 0 and reverse 5 0 -CCCAGTACTACATCTACCATC-3 0 for A668G SNP. PCR was performed using the following cycling conditions: pre-denaturation at 95 C for 3 min, followed by 35 cycles of 45 s at 95 C, 45 s at 55 C (A223G)/56 C (A668G), 45 s at 72 C, and a single 7 min extension step at 72 C. Thereafter, 6 lL of A223G (100 bp) or A668G (257 bp) PCR product was digested with 2.5 U of HaeIII or 4 U of BstEII-HF (New England Biolabs, Inc., Ipswich, MA) in a 10 lL of reaction volume for 1-3 h at 37 C, respectively. Digestion resulted in 76-and 24-bp fragments for the LEPR 223G allele and a non-digested 100-bp fragment for the 223A allele; 170-and 87-bp fragments for the LEPR 668G allele and a non-digested 257-bp fragment for 668A allele. The digested fragments were analyzed by electrophoresis on a 3.0% agarose gel and visualized by staining with Genecolour fluorescent dye (Beijing Genebio Biotech Co., Ltd., Beijing, China). For genotyping quality control, more than 30% of DNA samples were randomly genotyped again by a different operator to ensure identical results.

Statistical analysis
Data are presented as mean ± standard deviations (SDs) or absolute and relative frequencies (%). Variables with symmetric distribution were compared by Student's t-test or one-way ANOVA. Variables with asymmetric distribution were evaluated by Mann-Whitney's U-test or the Kruskal-Wallis H test. Assessing genotype distributions for deviation from the Hardy-Weinberg equilibrium and comparing differences of genotype or allelic frequencies between controls and patients were determined by Chi-square (v 2 ) test and Fisher's exact test. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated by v 2 test and logistic regression model to derive the risk of PE associated with gene variants. Differences in parameters between two groups/subgroups after correcting for differences in age and delivery BMI were estimated using analysis of covariance. The false discovery rate (FDR) method was used to adjust for multiple testing [35]. Statistical significance was set at p < .05. All statistical analyses were performed using the Statistical Program for Social Sciences (SPSS) software version 21.0 (IBM SPSS Statistic, IBM Corporation, Armonk, NY). A power calculation based on sample size and genotype frequencies of LEPR A223G polymorphism was performed by Power Analysis and Sample Size (PASS) software version 15.0.5 (NCSS, LLC, Kaysville, UT). The false positive report probability (FPRP) values of LEPR A223G and A668G SNPs were calculated based on the statistical power values, the p values of v 2 analyses, and the prior probabilities [36].

Results
Clinical, metabolic, and oxidative stress parameters of study participants As shown in Table 1, pre-pregnancy BMI, delivery BMI, SBP, and DBP were significantly higher, while age, gestational age, as well as birth height, weight, and PI of infants were significantly lower in the PE group than those in the control group. Serum TG, TC, non-HDL-C, apoB, TOS, and MDA levels; TG/HDL-C ratio, AI, apoB/apoA1 ratio, and OSI were higher, while HDL-C and apoA1 were lower in patients with PE compared with the control women after adjusting differences in age and delivery BMI.

Distribution of LEPR A223G and A668G genotypes and alleles
The genotypic distribution of LEPR A223G and A668G was aligned with the Hardy-Weinberg equilibrium in both groups (all p > .05).
The different genetic models of LEPR A223G and A668G SNPs are shown in Table 2. The frequencies of the AA þ AG genotype (28.6% vs. 36.1%) and A allele (14.9% vs. 19.8%) of LEPR A223G SNP were significantly lower in patients with PE than in the control women (OR ¼ 0.709, 95% CI: 0.543-0.926, p FDR ¼ .011 for the dominant model; OR ¼ 0.709, 95% CI: 0.559-0.899, p FDR ¼ .008 for the allele model). The frequencies of the AA þ AG genotype (17.7% vs. 24.6%) and A allele (9.0% vs. 12.9%) of LEPR A668G SNP were also lower in the PE group than in the control group (OR ¼ 0.658, 95% CI: 0.481-0.900, p FDR ¼ .016 for the dominant model; OR ¼ 0.671, 95% CI ¼ 0.500-0.899, p FDR ¼ .007 for the allele model). The genotype (AA þ AG) of LEPR A223G SNP (OR ¼ 0.684, 95% CI: 0.506-0.926, p ¼ .014) and that (AA þ AG) of LEPR A668G SNP (OR ¼ 0.650, 95% CI: 0.456-0.927, p ¼ .017) in dominant models remained a significant predictor for PE in prognostic models that included age and delivery BMI as covariates. When the prior probability was set to 0.1, the FPRP values were lower than 0.2 in genotype, dominant, and allele models for A223G SNP and in dominant and allele models for A668G SNP, suggesting that our results are relatively reliable in these models ( Table 2).
The combined genotypes of LEPR A223G and A668G SNPs are shown in Table 3. We combined the LEPR 223AA and 668AA homozygotes into heterozygous subgroups as their sample sizes were too small. The AG þ AA/AG þ AA frequency of the LEPR A668G and A223G SNPs was lower in patients with PE relative to the control women (13.0% vs. 21.2%, p ¼ .010). The multinomial logistic regression model, which included age and delivery BMI as covariates, demonstrated that the AG þ AA/AG þ AA combined genotype of the A668G/A223G SNPs was a protective factor for PE when the 668GG/223GG combined genotype was employed as the reference category (OR ¼ 0.558, 95% CI: 0.374-0.833, p ¼ .004).
Effect of LEPR A223G and A668G genetic variants on the clinical, metabolic, and oxidative stress parameters As shown in Table 4, patients with the A allele (AG þ AA genotype) of the LEPR A223G polymorphism had significantly higher DBP, LDL-C, apoB, and TAC levels (p < .05), but lower OSI (p ¼ .025), tended to have increased HDL-C levels (p ¼ .068), and had babies with increased neonatal PI (p ¼ .066) compared with those in patients with GG genotype and their babies. Healthy women with the A allele had babies who were taller at birth than babies of women with the GG genotype (p ¼ .035).
Patients with the A allele (AG þ AA) genotype of LEPR A668G polymorphism had higher HDL-C levels (p ¼ .045), but decreased TG/HDL-C ratio (p ¼ .052), tended to have babies with increased neonatal PI (p ¼ .062) and decreased neonatal birth height (p ¼ .066) relative to those in patients with the GG genotype and their babies (Table 5).

Discussion
In the present study, we found that the A alleles of LEPR A223G and A668G polymorphisms were genetic protective factors for PE in Chinese women. We also revealed that patients possessing the AG þ AA genotype of LEPR A223G polymorphism had significantly higher TAC and lower OSI levels, while those with the AG þ AA genotype of LEPR A668G polymorphism had Values are presented as mean ± SD. a All comparisons of parameters were corrected for differences in age and delivery BMI between the two groups, except for the parameters of age and BMI.
significantly higher HDL-C levels and relatively low TG/HDL-C ratio than in those carrying the corresponding GG genotype. Such findings imply that the decreased risk of PE in carriers with the 223A and 668A alleles may be linked to improved oxidative stress and lipid metabolism. Patients carrying the LEPR 668A allele and 223A allele were further found to have an enhanced the protective effect on PE compared with those carrying the GG/GG genotype after adjusting for age and delivery BMI. This result suggests that the combination of the two alleles may have a synergistic effect on PE. Multiple factors, including dysangiogenesis, immunological imbalances, metabolic disturbances, oxidative stress, inflammatory response, and liver dysfunction, contribute to the pathogenesis of PE [2][3][4][5]26,37,38]. Leptin, which is involved in glycolipid metabolism, angiogenesis, blood pressure regulation, reproduction, and inflammatory response, exerts its physiological and pathological effects by binding with  LEPRs in the hypothalamus and peripherally in tissues [8,9,11,39]. Serum leptin levels are increased during pregnancy, with a peak at 28-32 weeks of gestation in healthy pregnant women; however, its levels continue to rise in the third trimester until delivery in patients with PE [7]. Both PE and obesity are associated with elevated levels of circulating leptin [26]. Abnormal levels of leptin in PE can lead to dysregulation of amino acid transport, lipid metabolism, and nitric oxide production [40,41], promoting the occurrence and development of PE. Leptin exerts its function by binding to LEPR. Genetic variations of LEPR may affect the signal transduction and function of leptin. Therefore, studying SNPs in LEPR gene is a meaningful method to determine the susceptibility and pathogenesis of PE.
The A223G variant of LEPR gene caused by a point mutation at nucleotide position 223 in exon 4, resulting in lysine being changed to arginine at position 109 of the LEPR protein [42]. The G allele of A223G SNP has been reported to be associated with a lower risk of PCOS in Bahraini women [18]. On the contrary, the AA genotype of this SNP was deduced to be a protective factor for PCOS in Chinese women [19]. Moreover, the A allele of A223G SNP has been reported to be a protective factor for severe PE in Chinese population [27], whereas this SNP was not associated with the risk of PE in Hungarian women [29]. Consistent with the findings of Guan et al. [27], in the present study, the A allele of A223G SNP was associated with decreased risk of PE in Chinese population. Furthermore, patients with the AG þ AA genotype of this SNP were found to have significantly higher DBP, LDL-C, apoB, and TAC levels, lower OSI, relatively high HDL-C levels, and had babies with higher neonatal PI relative to those in patients with the GG genotype and their babies. Such findings suggest that this genetic variation may be related to the regulation of blood pressure and lipid metabolism and may enhance the body's antioxidant capacity, improve the state of oxidative stress, and exert a beneficial effect on fetal development.
The A668G variant of LEPR gene is caused by a mutation at position 668 in exon 6, leading to a change from glutamine to arginine at position 223 of its protein [14,17]. This mutation occurs in the Values are presented as mean ± SD. All comparisons of parameters were corrected for differences in age and delivery BMI between the two subgroups, except for the parameters of age and BMI.
. a p < .05: compared with the GG genotype subgroup in the control group. b p < .05: compared with the GG genotype subgroup in the preeclampsia group. extracellular domain of the LEPR protein, alters the three-dimensional conformation, directly influences its binding with leptin, and affects downstream LEPRinduced signaling events [14,18]. Quinton et al. [14] have reported lower serum levels of leptin in postmenopausal Caucasian women with GG genotype than those in carriers of the A allele. The G allele of A668G SNP has been shown to be associated with increased risk of T2DM [15][16][17], PCOS in Bahraini women [18], essential hypertension in a Northern Han Chinese population [22], and PE in Sudanese and Hungarian population but not in Sri Lankan women [29][30][31]. The A allele significantly decreases the risk of nonalcoholic fatty liver disease and coronary atherosclerosis [21], but increases the risk of essential hypertension [23] in the Chinese population. Different from the abovementioned research findings in PE [29][30][31], in the present study, the A allele was first found to be a protective factor of PE in Chinese women. Patients with the A allele of this SNP had significantly higher HDL-C levels, suggesting that the variation may be related to improved lipid metabolism. Notably, the allelic frequencies for LEPR polymorphisms differ among ethnic groups [16,17]. The G allele of A223G SNP is more common in Chinese population (82.7%) [20] than in Iranian (45.5%) [43], Danish (43.8%) [44], and Spanish (24.7%) [45] populations. The G allele frequency of A668G SNP is higher in Chinese population (87.7%) [46] than in Iranian (29.9%) [47], Spanish (43.3%) [45], and Sudanese (26.8%) [30] populations. The G allele frequencies of A223G and A668G SNPs in our study are 81.2% and 87.9%, respectively, and are comparable with several other studies on Chinese populations. Additionally, environmental factors, such as diet and obesity, may also affect LEPR expression [14,26,48,49]. In some cases, positive associations observed in smaller cohorts could not be confirmed in subsequent studies of larger cohorts [50]. Therefore, inconsistent results concerning relationships between LEPR polymorphisms and disease may be caused by different ethnic background and sample size.
This study had some limitations. First, owing to the low frequency of homozygosity of minor alleles, LEPR 223AA and 668AA, an analysis of these genotypes as subgroups was difficult to perform. Larger sample sizes of patients and controls are required to access dose-dependent genotype characteristics. Second, we did not measure leptin levels in serum of participants. Values are presented as mean ± SD. All comparisons of parameters were corrected for differences in age and delivery BMI between the two subgroups, except for the parameters of age and BMI. Ã Controls (GG ¼ 823, AG þ AA ¼ 227 þ 9); PE (GG ¼ 198, AG þ AA ¼ 42 þ 0). ÃÃ Controls (GG ¼ 628, AG þ AA ¼ 168 þ 10); PE (GG ¼ 123, AG þ AA ¼ 25 þ 0). a p ¼ .045: compared with the GG genotype subgroup in the preeclampsia group.
A further study to detect leptin levels in patients with different genotypes can help illustrate the relationship between genetic mutation and disease pathogenesis. Third, we did not measure oxidative parameters in some subjects due to inadequate sample volume or samples with hemolysis or bilirubin, which might influence the power of these parameters or result in the absence of statistical significance.
In conclusion, the present study revealed that the 223A and 668A alleles of LEPR polymorphisms are genetic protective factors for PE in Chinese women. The two alleles may exert a beneficial effect on oxidative stress and lipid metabolism in patients. Furthermore, compared with women carrying the 223GG/668GG combined genotype, those carrying 223A and 668A combined allele showed a further enhanced protective impact on PE. Altogether, our findings suggest that LEPR genetic variations may influence the susceptibility of carriers to the development of PE.