Association of Luteinizing hormone and LH receptor gene polymorphism with susceptibility of Polycystic ovary syndrome

ABSTRACT

Altered folliculogenesis and reproductive anomalies in polycystic ovary syndrome (PCOS) suggest that variations of genes involved in folliculogenesis might influence etiopathogenesis of this syndrome. The objective of this study was to assess the association of LHβ (rs1056917) and lutropin receptor (LHR) (rs61996318) polymorphism with polycystic ovarian syndrome and to interrelate the levels of luteinizing hormone (LH) with severity of clinical manifestations of PCOS. Three hundred women of reproductive age were enrolled in this retrospective case-control study. Rotterdam Criteria was used to diagnose PCOS patients. Nucleotide mutations of LH and LHR gene was analyzed using polymerase chain reaction-restriction fragment length polymorphism. High LH levels were found in 88% of PCOS patients. LHβ TC and CC genotypes were significantly associated with PCOS risk (OR [odds ratio] 13.95, CI [confidence interval] 6.30–30.86, p < 0.0001 and OR 3.31, CI 1.30–8.41, p = 0.01). The frequency of the C allele was 0.31 in PCOS and 0.02 in controls (OR 18.80, CI 8.54–41.37, p < 0.0001). LHR CA and AA genotype conferred a significant risk in development of PCOS (OR 5.07, CI 2.50–10.31, p < 0.0001). The frequency of the A allele was 0.51 in PCOS and 0.03 in controls (OR 26.62, CI 13.99–50.65, p < 0.0001). The results show an association between polymorphism of LHβ, LHR and PCOS, indicating that variants of these genes may affect the metabolic pathways involved in this syndrome. Majority of the affected women were found to have elevated LH levels. This study sheds new light in the diagnosis, treatment and management of PCOS syndrome.

Abbreviations: AUC: area under curve; BMI: body mass index; C: cholesterol; CI: confidence interval; DBP: diastolic blood pressure; DHEAS: dehydroepiandrosterone sulfate; FG: Ferriman–Gallway; FSH: follicle stimulating hormone; GHQ: general health questionnaire; HA: hyperandrogenism; HDL-C: high-density lipoprotein cholesterol; HOMA-IR: homeostatic model assessment for insulin resistance; HWR: hip waist ratio; LDL-C: low-density lipoprotein cholesterol; LH: luteinizing hormone; LH: luteinizing hormone; LHR: lutropin receptor; O: oligomenorrhea; OR: odds ratio; PCO: polycystic ovaries; PCO: polycystic ovary; PCOS: polycystic ovary syndrome; PCR: polymerase chain reaction; ROC: receiver operating curve; SBP: systolic blood pressure; SE: standard error of coefficient; SNP: single nucleotide polymorphism; TG: triglycerides; TSH: thyroid stimulating hormone; VD: vitamin D

KEYWORDS:

• Luteinizing hormone
• PCOS
• luteinizing hormone receptor
• polymorphism
• PCR-RFLP

Introduction

Polycystic ovary syndrome (PCOS) is a multifactorial disorder where genes individually or gene–gene interactions have been reported to influence predisposition to PCOS development (Gilbert et al. 2018). With varied clinical manifestations, unknown etiology, complex pathophysiology and poor diagnosis, it has been considered as a major scientific issue (Palomba et al. 2015). PCOS is primarily a disorder of hyperandrogenism, anovulation and polycystic ovaries (Soni et al. 2018). Hyperandrogenism may arrest folliculogenesis that in turn leads to polyfollicular morphology, disturbing menstrual cycle causing anovulatory infertility (Yao et al. 2017).

Luteinizing hormone (LH) is a 29-kD heterodimeric glycoprotein, consists of well conserved α chain and a unique β-subunit that confers biological specificity for the hormone receptor interaction in the target tissues (Jianga et al. 2014). The genes for the LH subunits are located on chromosome 19q13.32 (Cahoreau et al. 2015). It is produced and released by the anterior pituitary and is vital for follicular growth stimulation, oocyte maturation, ovulation and estrogen production in females (Young et al. 2003). LH binds to and activates a receptor known as lutropin receptor (LHR). LHR gene is located on chromosome 2p21. LHR is expressed in multiple cells of ovary, including theca, luteal, interstitial and granulosa cells, testis and uterus (Choia and Smitz 2014).

A high level of circulating LH is an important biochemical feature of PCOS (Baldani et al. 2012; Zhang et al. 2017). LH hypersecretion appears as a result of increased amplitude and frequency of pulsatile LH release (Tsutsumi and Webster 2015). Increased LH levels, in turn, favor increased androgen production by ovarian theca cells (Baptiste et al. 2010). Microheterogeneity and genetic variants of LHβ and LHR might affect bioactivity of LH and LHR causing polycystic ovary syndrome (Tapanainen et al. 1999). Most of the pharmacological interventions are generalized for metabolic aberrations in PCOS and thus treatment remains less effective (Badawy and Elnashar 2011). Therefore, the clinical management needs to consider individualized symptomatology with knowledge of genetic variations, limiting the exposure of patients to drugs ineffective for them. The identification of single nucleotide polymorphisms (SNPs) in PCOS susceptible genes will prove useful in defining the individual’s risk profile to PCOS and drug response. Therefore, studying polymorphic association of LH and LHR with PCOS may contribute toward a better understanding of the disease. This will further help to better understand if genotypic variations would have important predictive power for tailored treatment response by allowing selection of susceptible and non-susceptible patients.

The purpose of the study was to determine the frequency of LHβ rs1056917 variant and rs61996318 LHR gene variant in patients with PCOS and controls. We also tried to clarify the influence of these polymorphisms on the clinical and biochemical profile of PCOS. In addition, we assessed levels of LH in PCOS women. This is the first integrated study conducted to establish an association of LHβ and LHR polymorphism among North Indian PCOS women.

Results

Clinical and biochemical characteristics of participants

The basal demographic, anthropometric, clinical and biochemical parameters of PCOS women and controls are summarized in Table 1. PCOS patients have significantly larger values of LH, LH/follicle stimulating hormone (FSH) ratio, weight, body mass index (BMI), HDL-C, LDL-C, TSH, testosterone, fasting glucose, insulin, homeostatic model assessment for insulin resistance (HOMA-IR), DHEA-S, hirsutism, GHQ, VD, depression compared with controls. There were no significant differences between the control and PCOS group in height, age, hip waist ratio, total cholesterol, triglycerides, prolactin and estradiol. Eighty-eight percent of PCOS women had raised LH/FSH ratio. Oligomenorrhea and amenorrhea were present in 77% and 15% of PCOS women, respectively. Sixty-four percent of PCOS women had presence of polycystic ovaries by ultrasound (supplementary figure 1). Multiple logistic regression analysis of various clinical and anthropometric variables predicting the likelihood of PCOS was performed (Table 2). A multiple logistic full model was constructed by including all significantly related factors BMI, LH/FSH, LH, T, DHEA, depression, and GHQ. In the second step, the stepwise multiple logistic reduced model analysis was performed by excluding one least significant factor from the full model. The prediction power of the model was considered as a function of the R-squared value.

Table 1. Anthropometric, clinical, and biochemical characteristics of study population.

Variable	All (300)	PCOS (150)	Control (150)	p	95% CI
Age (year)	27.56 ± 6.76	26.47 ± 6.99	27.55 ± 6.21	0.30	−5.58 to −2.57
Height (m)	1.62 ± 0.06	1.63 ± 0.06	1.62 ± 0.05	0.1179	−0.002 to 0.02
BMI (kg/m^2)	30.93 ± 97.81	34.90 ± 8.34	26.96 ± 5.54	0.0001	−9.5488 to −6.33
HWR (cm)	1.10 ± 0.34	1.13 ± 0.48	1.08 ± 0.03	0.20	−0.1273 to 0.02
Oligomenorrhea (%)	–	116 (77.33)	0	–	–
Amenorrhea (%)	–	23 (15.33)	0	–	–
PCO (%)	–	64 (42.66)	0	–	–
DBP (mmHg)	80.68 ± 6.89	79.38 ± 6.44	73.38 ± 5.44	0.0001	4.64–7.35
SBP (mmHg)	120.57 ± 11.86	123.21 ± 11.09	117.94 ± 12.05	0.0001	2.63–7.90
FG	7.64 ± 6.12	9.01 ± 8.42	6.36 ± 1.73	0.0002	1.26–4.03
Total C (mg/dl)	188 ± 25.27	187 ± 25.98	189 ± 24.56	0.49	−7.74 to 3.74
TG (mg/dl)	157.88 ± 76.36	165.08 ± 101.92	151.03 ± 34.69	0.11	−3.24 to 31.34
HDL-C (mg/dl)	63.09 ± 48.53	54.61 ± 44.78	71.58 ± 50.74	0.002	−27.84 to −6.09
LDL-C (mg/dl)	83.81 ± 23.21	95.19 ± 24.25	72.44 ± 15.24	0.0001	18.14–27.35
Prolactin (ng/ml)	10.77 ± 4.09	10.70 ± 3.99	10.71 ± 3.99	0.06	−0.90 to 0.90
TSH (µIU/ml)	2.69 ± 1.39	2.96 ± 1.49	2.41 ± 1.22	0.0005	0.24–0.85
Fasting glucose (mg/dl)	71.07 ± 13.48	74.82 ± 14.50	67.32 ± 11.23	0.0001	4.55–10.44
Fasting insulin (µIU/ml)	11.34 ± 8.26	13.61 ± 10.24	9.07 ± 4.64	0.0001	2.73–6.34
HOMA-IR (mol × µU/l^2)	1.40 ± 1.01	1.67 ± 1.26	1.14 ± 0.56	0.0001	0.30–0.75
Testosterone (ng/dl)	61.69 ± 27.05	80.39 ± 26.94	43 ± 6.26	0.0001	32.94–41.83
Estradiol (pg/ml)	55.59 ± 29.05	55.01 ± 23.99	56.18 ± 34.11	0.73	−7.87 to 5.53
Progesterone (ng/ml)	1.77 ± 4.71	2.65 ± 6.92	0.89 ± 2.51	0.003	0.57–2.94
DHEA-S (ng/ml)	1469 ± 985	1876 ± 1101	1063 ± 870	0.0001	587.52–1038.48
LH (µIU/ml)	7.44 ± 6.08	11.52 ± 6.43	3.42 ± 0.44	0.0001	7.06–9.13
FSH (µIU/ml)	4.97 ± 1.85	4.47 ± 2.23	4.99 ± 0.44	0.005	−0.88 to −0.15
LH/FSH ratio	1.92 ± 0.48	2.73 ± 0.58	1.21 ± 0.39	0.0001	1.40–1.63
GHQ	9.64 ± 5.71	11.13 ± 5.87	8.15 ± 5.15	0.0001	1.72–4.23
VD (ng/ml)	21.55 ± 12.9	17.6 ± 7.3	25.5 ± 11.2	0.0001	−10.04 to −5.75
Depression	22.48 ± 13.78	27.08 ± 14.65	17.89 ± 11.14	0.0001	6.23–12.14
Infertility	–	60 (40)	0	–	–

Values are described as mean ± SD and 95% confidence interval. Independent Student’s t test was used to compute p value. BMI: body mass index; PCOS: polycystic ovary syndrome; HWR: hip waist ratio; SBP: systolic blood pressure; DBP: diastolic blood pressure; PCO: polycystic ovaries; GG: Ferriman–Gallway score; C: cholesterol; TSH: thyroid stimulating hormone; TG: triglycerides; LH: luteinizing hormone; FSH: follicle-stimulating hormone; HOMA-IR: homeostatic model assessment for insulin resistance; HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol; VD: vitamin D; DHEAS: dehydroepiandrosterone sulfate.

Table 2. Multiple logistic regression analysis of various clinical and anthropometric variables predicting the likelihood of PCOS.

Variables	Coefficient	SE	t	R^2	p
Model 1	12.47	5.50	2.26	0.71	0.026
Model 2	13.24	5.39	2.45	0.29	0.016
Model 3	9.91	3.44	2.87	0.33	0.005
Model 4	11.62	2.42	5.19	0.41	0.002
Model 5	16.14	1.57	10.24	0.54	0.001
Model 6	14.19	1.08	13.07	0.61	0.000

SE: Standard error of coefficient; Model 1 (full model) including BMI, LH/FSH, LH, T, DHEA, depression and GHQ; Model 2 including BMI, LH/FSH, LH, T, DHEA, depression; Model 3 including BMI, T, DHEA, depression; Model 4 including BMI, T, depression; Model 5 including BMI, T; Model 6 including BMI.

BMI: body mass index; LH/FSH: luteinizing hormone/follicle stimulating hormone; DHEAS: dehydroepiandrosterone sulfate, GHQ: general health questionnaire; p value of <0.05 is significant.

Genotypic distribution of LHβ and LHR polymorphisms

Table 3 evaluated the distribution of LHβ and LHR variants genotypic and allelic frequencies between two groups. The result illustrates the number (n) and percentage (%) of relative genotypes, allele frequencies and percentage, odds ratio (OR), 95% confidence interval (CI), and p values. The LHβ gene was analyzed for rs1056917 in exon 3 causing the substitution of serine for glycine at codon 1502. The TT genotype distribution prevailed in 44% PCOS patients and 95.3% controls. The allelic variant of LHβ was found in 56% of PCOS subjects and 4.6% of controls (Table 3). A total of 50% (75) of PCOS women had a heterozygous TC genotype and 6% (9) had CC homozygous genotype distribution. The CC genotype was not detected in controls. The frequency of T (69% in PCOS and 97% in control) and C allele (31% in PCOS and 2% in control) differs significantly among two groups (Table 3). The minor allele frequency allele was 31% of PCOS and 2.33% in controls. Patients carrying TC genotype exhibit a significant higher risk of PCOS than those carrying TT genotype (OR [95% CI] 20 [8.96–46.54], p < 0.001).

Table 3. LHβ exon 3 variant and LHR rs61996318 (C>A) variant genotype distribution in PCOS patients and control.

	PCOS (150)		Control (150)
LHβ	n	%	n	%	OR	CI	p
TT	66	44	143	95.3	–	–	–
TC	75	50	7	4.6	13.95	6.30–30.86	<0.0001
CC	9	6	0	0	3.31	1.30–8.41	0.01
TC + CC	84	56	7	4.6	16.27	7.38–35.89	<0.0001
T	207	69	293	97.6
C	93	31	7	2.33	18.80	8.54–41.37	<0.0001
LHR
CC	43	28.6	139	92.6	–	–	–
CA	63	42	11	7.3	5.07	2.50–10.31	<0.0001
AA	44	29.3	0	0	100.7	6.15–1637.57	0.001
AA + CA	107	71.3	11	7.3	14.56	7.6284–27.81	<0.0001
C	149	49.7	289	96.4
A	151	50.3	11	3.6	26.62	13.99–50.65	<0.0001

Data are presented as number (percentage). Chi square test; OR odds ratio, and p value was computed. PCOS: polycystic ovary syndrome; LH: luteinizing hormone; LHR: luteinizing hormone receptor. G wild type, A mutant allele. p Value of <0.05 is considered as significant.

The LHR was analyzed for rs61996318 polymorphism which is a C to A change at position 546 in exon. The PCOS women had significantly higher frequency of mutant allele ‘A’ (50.3%) compared to controls (3.6%). The homozygous CC genotype was present more frequently in controls (92%) than in cases (43%). The number of CA heterozygotes was greater in PCOS (42%) than in controls (7.3%). The homozygous AA genotype was prevalent in 29% of PCOS. The minor allele (A) frequency 50.3% in PCOS cases was 3.6% in controls. The risk of PCOS was significantly higher among patients carrying CA genotype than those carrying CC genotype (OR 5.07, CI 2.50–10.31, p < 0.0001). The result of chromatogram for LHβ and LHR was shown in Figure 1(A,B), respectively. The distribution of the polymorphisms was confirmed by Hardy–Weinberg equilibrium test.

Figure 1. Chromatogram showing sequence variation of LHβ and LHR gene SNP. The chromatogram graph illustrating wild-type and genetic variant pattern is visualized by Finch TV. Each colored peak represents a base pair. Arrow indicates the position of mutation.(A) LHβ rs1056917 polymorphism-position 294 represents nucleotide substitution where T is changed to C, replacing GGT to GGC. Upper left part represents wild-type and right part represents the mutated sequence. (B) LHR rs61996318 polymorphism at position 546 represents nucleotide substitution; C is changed to A.

Impact of gene polymorphisms on endo-metabolic characteristics among PCOS patients

In PCOS patients, we found that LHβ mutant T/C or C/C polymorphism was significantly associated with increased insulin, HOMA-IR, PCO, oligomenorrhea, and infertility when compared with GG genotype (Table 4). The LH/FSH ratio was significantly higher in mutant homozygous and heterozygous variants when compared with wild-type variants (p 0.0001). LHR CA/AA mutation was found to be strongly associated with high LH/FSH ratio. LHR mutated genotype showed nonsignificant association with BMI, testosterone, insulin, HOMA-IR, hirsutism and vitamin D as compared to wild-type genotype.

Table 4. Association of various parameters with LHβ and LHR SNPs genotype in PCOS patients.

Variables	LHβ TT 66	LHβ TC (75)	LHβ CC (9)	p Value
BMI (kg/m^2)	30.93 ± 7.81	31.02 ± 8.01	32.90 ± 8.34	0.035
LH/FSH	1.72 ± 0.99	1.74 ± 1.01	2.21 ± 1.06	0.0001
Testosterone	75.75 ± 26.18	75.81 ± 26.27	76.67 ± 27.98	0.51
FG	17.36 ± 4.11	16.65 ± 4.13	17.93 ± 4.67	0.136
Insulin	12.05 ± 9.37	16.71 ± 11.33	18.76 ± 13.36	0.005
HOMA-IR	1.73 ± 0.85	2.51 ± 2.03	2.56 ± 2.2	0.004
VD	9.82 ± 7.17	7.18 ± 5.98	8.54 ± 6.74	0.065
Oligomenorrhea	45 (68.18%)	62 (82.66%)	9 (100%)	0.000
PCO	0	57 (76%)	7 (77.77%)	0.000
Infertility	11 (16.66%)	10 (13.33%)	9 (100%)	0.000
O + PCO	0	32 (42%)	7 (77%)	0.09
O + HA	16 (24%)	33 (44%)	8 (88%)	0.02
PCO + HA	0	16 (21%)	7 (77%)	0.41
PCO + HA + O	0	15 (20%)	8 (88%)	0.73
	LHR CC (43)	LHR CA (63)	LHR AA (44)	p
BMI (kg/m^2)	29.24 ± 7.01	31.02 ± 7.19	31.74 ± 7.78	0.06
LH/FSH	1.61 ± 0.82	2.45 ± 0.46	2.53 ± 0.54	0.0001
Testosterone	75.56 ± 27.87	80.11 ± 26.15	79.39 ± 25.17	0.21
FG	71.32 ± 14.11	73.88 ± 16.17	71.21 ± 14.11	0.94
Insulin	15.87 ± 9.34	14.04 ± 9.56	14.24 ± 9.88	0.14
HOMA-IR	2.01 ± 1.57	2.52 ± 2.28	2.45 ± 2.08	0.17
VD	7.78 ± 2.22	7.44 ± 2.24	8.01 ± 2.59	0.18
Oligomenorrhea	24 (20%)	29 (46%)	16 (36%)	0.33
PCO	17 (39%)	22 (34%)	12 (27%)	0.23
Infertility	9 (20%)	8 (12%)	13 (29%)	0.33
O + PCO	11 (16%)	24 (32%)	6 (66%)	0.04
O + HA	13 (19%)	36 (48%)	8 (88%)	0.04
PCO + HA	3 (4%)	16 (21%)	4 (44%)	0.27
PCO + HA + O	7 (10%)	12 (16%)	4 (44%)	0.21

ANOVA test; data are described in mean ± SD. p ≤ 0.05 was considered statistically significant. BMI: body mass index; LH: luteinizing hormone; FSH: follicle stimulating hormone; FG: Ferriman–Gallway; HOMA-IR: homeostatic model assessment for insulin resistance; VD: vitamin D; PCO: polycystic ovary; O: oligomenorrhea; HA: hyperandrogenism.

LH accuracy as a diagnostic criterion for PCOS

For the assessment of the use of LH as a diagnostic test for PCOS, receiver operating curve (ROC) analysis was performed. Strength of using LH as a positive predictor for diagnosis PCOS was measured using ROC analysis (Figure 2). Area under curve, p value, specificity and sensitivity were computed for variables including LH, testosterone, BMI, PCO, and HOMA-IR. ROC near upper left corner suggested that LH showed highest specificity and sensitivity with p value <0.001 and AUC value of 1. This makes LH a better diagnostic predicator for PCOS. We also estimated strength of linear association of LH with endo-metabolic different variables of PCOS viz. hyperandrogenism, PCO, BMI, HOMA-IR, VD. The result demonstrated positive correlation with all variables except vitamin D. Table 5 explains data for ROC and correlation analysis of LH with certain hallmark features of PCOS.

Table 5. ROC analysis and correlation coefficient values of various parameters with LH in PCOS patients.

Tests	AUC	p Value	r	95% CI	p
LH	1	<0.001	–	–	–
BMI	0.93	<0.001	0.69*	0.57–0.78	0.03
HA	0.99	<0.001	0.51*	0.34–0.64	0.001
PCO	0.64	0.004	0.197*	0.00–0.37	0.05
VD	0.77	0.002	−0.26	−0.436 to−0.06	0.009
IR	0.91	<0.001	0.21*	0.017–0.395	0.03
Infertility	–	–	0.67*	0.55–0.76	0.001

ROC: receiver operating curve; LH: luteinizing hormone; BMI: body mass index; HA: hyperandrogenism; PCO: polycystic ovary; VD: vitamin D; AUC: area under curve; r: correlation coefficient; CI: confidence interval. p Value of <0.05 is considered as statistically significant. *indicates positive correlation.

Figure 2. ROC analysis of LH as diagnostic test for predicting PCOS. Roc curve was plotted using specificity versus sensitivity. The closer the curve follows the top left-hand border, the more accurate is the test. The area under the curve is a measure of test accuracy. The slope of the tangent line gives the likelihood ratio (LR) for the test value. The AUC for LH is 1, making it better diagnostic determinant among all. ROC: receiver operating curve, p value based on z statistics. p Value <0.05 was considered as significant. Curve toward upper left corner indicates better performance.

Discussion

LH plays critical roles in folliculogenesis, ovulation and androgen production (Filicori 1999). Variations in LH and its receptor are associated with the risk of development of various reproductive disorders including PCOS. We have evaluated the influence of LH (rs1056917) and LHR (rs61996318) genetic polymorphism as a potential risk factor for development of PCOS. To the best of our knowledge, there has been no study conducted investigating the association of these SNPs with PCOS in Indian population. This study observed a significant difference in genotype frequencies of both the polymorphisms between two groups.

Association of LH levels with PCOS phenotype

Elevated levels of LH and LH/FSH ratio are found in PCOS patients. Few studies published earlier showed that increased ovarian volume and follicle number are positively correlated to LH (Marisol and Vicente 2015; Rackow et al. 2018). A study in 1995 established an association of increased LH concentrations with severe menstruation dysfunction and infertility rate (Kumar and Sait 2011). A recent study demonstrated the role of LH in stimulating anti-mullerian hormone production in ovarian cells of PCOS women which contributes to the arrest of folliculogenesis and subsequent development of the disorder (Pellatt et al. 2007). Elevated LH and LH/FSH prevalence varies from 35% to 77% in PCOS patients (Banaszewska et al. 2003; Kadhim 2017; Lal et al. 2017). This study also evaluated the levels of LH in PCOS patients and found that 88% of PCOS patients had elevated LH levels. In correlation testing, LH showed strong correlation with BMI (r = 0.69), hyperandrogenism (r = 0.51) and infertility (r = 0.67). No correlation of LH was found between IR (r = 0.21) and PCO (0.19). In ROC analysis too, AUC for PCO was 0.64, illustrating low accuracy. This reveals that LH was not a better predictor of PCO in this study.

Association of LHβ polymorphism with PCOS

To date, there are no data exploring the role of LHβ rs1056917 polymorphism in PCOS women available. However, a few earlier studies reported the association of three LHβ polymorphisms with PCOS viz. G1502A, Trp8Arg and Ile15Thr. Liu et al., in 2016, studied prevalence of both LHβ and LHR polymorphisms in Chinese PCOS women and reported the association of LH G1052A polymorphism only with PCOS susceptibility (Liu et al. 2012). A report by El‐Shal et al. (2016) investigated LH G1502A polymorphism in Egyptian PCOS women and found a significant positive association with increased risk of PCOS. Also, LHβ GA genotype was found to be less prevalent in the lean PCOS group when compared with obese patients (OR 5.6, 95% CI 1.30–24.56, p 0.01). Some published literature speculated the role of LH polymorphism in PCOS-related diseases. Ramanujam et al. (1999) proposed that microheterogeneity of LHβ rs1056917 SNP affects gonadal function and related to menstrual cycle irregularities in Chinese women. Prevalence of insLQ was reported to be higher in patients with endometriosis and infertility (Schmitz et al. 2015). The present study reported that 56% of PCOS women were found to carry LHβ rs1056917 variation and the mutant phenotypes showed statistically significant difference between PCOS women and controls.

Association of LHR polymorphism with PCOS

To date, only one report assessed the impact of LHR rs61996318 SNP in PCOS and showed a negative association with the risk of PCOS development (Liu et al. 2012). Our results were in contrast to LHR rs6168318 polymorphism suggesting that this variation may be conserved in the studied Chinese population. Supplementary Table 1 speculates the impact of some other SNPs present in LHR gene (rs12470652, G935A, ins18LQ, rs2293275) on PCOS (Suganuma et al. 1995; Rajkhowa et al. 1995; Elter et al. 1999; Valkenburg et al. 2009; Thathapudi et al. 2015; Robeva et al. 2018). Robeva et al. (2018) examined the association of LHR rs12470652 and rs2293275 in Bulgarian PCOS women and found that the rs2293275 variation modulates PCOS characteristics especially obesity. Thathapudi et al. (2015) demonstrated a significant correlation between LHR rs2293275 GG genotype with BMI, LH/FSH ratio and IR in south Indian PCOS women. Two studies reported lack of association between LHR polymorphism and PCOS (Valkenburg et al. 2009; Liu et al. 2012). Our study demonstrated a significant positive association of LHR rs6168318 variation between PCOS and controls. The present study also found strong association of LHβ and LHR variations with high LH/FSH ratio, IR, insulin levels, oligomenorrhea, PCO and infertility. Genetic polymorphisms contribute to a better understanding of etiology and pathophysiology of this syndrome. The results showed important associations between variant and the PCOS characteristics suggesting the need of more LHR polymorphism studies in different PCOS subgroups. LH and LHR are the most important genes affecting steroidogenesis with metabolic and transport pathways of sex steroids (Cui et al. 2011). The prevalence and importance of high LH concentrations in PCOS seem to be underestimated, and its importance within the diagnosis of PCOS and future research should be reestablished.

The present study provides better insight into the possible association of LHβ and LHR genetics in the pathophysiological mechanism of PCOS. We found that both LHβ and LHR mutations might contribute to the pathways involved in the development of PCOS. We also reported that elevated levels of LH and LH/FSH ratio are found in majority of PCOS patients. LH hypersecretion had a vital effect on ovulation, subfertility/infertility and pregnancy complications in women with PCOS. Thus, exploring the environmental and genetic factors would be of clinical importance. Therefore, LH and LHR polymorphism may be useful as a molecular marker for early detection of ‘at-risk’ women for developing PCOS. The genotype of an individual could potentially fulfill the role for targeted therapeutic decision-making. This necessitates further research in different populations in future possibly helping to guide personalized care.

Materials and methods

Subjects

This study is a retrospective case-control study. One hundred and fifty PCOS patients and 150 controls ranging between 16 and 45 years of age were enrolled in this study from Post Graduate Institute of Medical Sciences, India. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional ethical committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all individual participants included in the study.

Inclusion criteria

Rotterdam Criteria was used to diagnose PCOS (Lizneva et al. 2016). Any two out of three criteria were needed for the diagnosis: (a) anovulation or oligoanovulation (when menstrual cycles ≥45 days). (b) Polycystic ovaries on sonography (>12 follicles measuring 2–9 mm in one or both ovaries). (c) Clinical or biochemical hyperandrogenism. Age-matched healthy females who had regular menstruation with no clinical appearance of hirsutism, obesity, thyroid, infertility or any other chronic illness were considered as controls. The study was approved by the institutional ethical committee and written informed consent was obtained from participants.

Exclusion criteria

Subjects with thyroid disorder, obesity, diabetes, hyperprolactinemia, Cushing’s syndrome, adrenal hyperplasia, acromegaly, ovarian tumors, infertility and who were taking regular oral contraceptive pills for the last 6 months were excluded from the study.

Clinical and biochemical analysis

The general medical information was recorded including age, weight, height, hip and waist circumference, menstruation, history of fertility. Personal medical history (obstetric history, skin problems and metabolic diseases) and family history of PCOS was also recorded, if any. Details pertaining to psychosocial aspects like depression, stress, anxiety and quality of life (GHQ) were obtained using standard questionnaires (Beck Depression Inventory, GHQ-12) (Montazeri et al. 2003; Englbrecht et al. 2017). BMI was estimated using the mathematical formula: weight (kg)/height (m²). Blood pressure was recorded using the electronic OMRON machine. Two readings were taken at an interval of 10 min and the mean of the two values was taken as blood pressure. Degree of hirsutism was assessed using modified Ferriman–Gallwey score (Aswini and Jayapalan 2017).

Biochemical analysis was done on a Stratec open analyzer (SR 300) using kits supplied by Beckman coulter. Blood samples were collected on day 2/3 of menstrual cycle and hormonal measurement was performed at the Obstetrics and Gynecological Department, PGIMS, Rohtak, Haryana Serum LH (Reference Value: 1.7–13.3 mIU/ml), FSH (Reference Value: 4.5–11.0 mIU/ml), total testosterone (Reference Value: 9.8–82.1 ng/dl), serum triglycerides (mg/dl), cholesterol (mg/dl), LDL (mg/dl), HDL cholesterol (mg/dl) and VLDL (mg/dl) were measured. Fasting blood glucose was measured by the glucose oxidase method. Fasting serum insulin was measured by ECLIA method (immunoanalyzer cobas e411). IR was calculated by Homeostasis model assessments of IR using the formula fasting insulin (µU/l) × fasting glucose (nmol/l)/22.5. IR was diagnosed when HOMA-IR is more than 2.5 (Wongwananuruk et al. 2012). Additionally, serum prolactin (Reference Value: 4.1–28.9 ng/ml), estradiol (Reference Value: 40.7–220.4 pg/ml), TSH (µmol/l), T3 (pg/ml), T4 (ng/dl) and 17-hydroxyprogesterone (Reference Value: ≤0.87 ng/ml) were also measured to exclude other causes of menstrual disorders. A transvaginal ultrasound examination was also performed day 2/3 of menstrual cycle by using a scanner with a 5-MHz probe.

Single nucleotide polymorphisms identification and selection

LHR SNP is located in intronic region causing missense mutation with significant effect on mRNA stability, level of gene expression, formation of an alternative protein isoform (Pache et al. 1993). LHβ mutation aborts the ability of LH to bind to LHR, resulting in the abnormal steroidogenesis and subsequent infertility (Chang et al. 2000). Because of their potential association with progression of PCOS, we selected these SNPs for this study.

Genotyping and sequencing

Genomic DNA was extracted from blood by using kit from Genetix Biotech Asia Pvt. Ltd. DNA amplification was carried out using the polymerase chain reaction (PCR) with specific primers: LHβ 5′ forward (AGTCTGAGACCTGTGGGGTCAGCTT), LH 3′ reverse (GGAGGATCCGGGTGTCAGGGCTCCA) and LHR 5′ forward (TGATGGTGGTGGTGATGATG) and LHR 3′ reverse (GGTTTCTAGCCA GCCAGTTG) (Liu et al. 2012). A volume of 50 μl reaction mixtures was used containing 25 μl of master mix (GoTaq® Promega green master mix), 2 μl, each primer (10 pmol/ml), 2 μl of template DNA, 0.5 μl Taq polymerase and 18.5 μl distilled water. PCR cycle was carried out at an initial denaturation at 94°C for 1 min, 94°C for 5 min, annealing at 65°C for 40 s, extension at 72°C for 1 min, followed by a final extension at 72°C for 5 min. Samples were subjected to 30 cycles. The PCR cycle setup was same for LHR except that annealing temperature was kept at 64°C. The fragment of amplification was 395 bp for LHβ and 379 bp for LHR examined by 2% agarose gel. PCR amplified products were subjected to restriction digestion by restriction fragment length polymorphism using fast digest enzymes Eco0109I and Rsa1 (Fermentas) for LHβ and LHR, respectively. Restriction digestion was performed at 37°C for 15 min. Digested products were visualized on a 3% agarose gel electrophoresis. Mutations were verified by using an ABI PRISM 3100 automatic genetic analyzer. The obtained sequence results were interpreted by using FinchTV software. LHβ mutation is from T to C. Genotypes are represented as TT and CC for homozygous type and TC for heterozygous. Digestion with Eco0109I produced two bands of 279 and 126 bp in CC normal genotype and a single band of 395 bp in AA homozygous mutant genotype and three bands of 395, 269 and 126 bp in heterozygous CA genotype. Rsa1 digestion produced two bands of 170 and 210 bp in normal TT genotype, a single band of 350 bp in C allele homozygotes mutant and 3 bands of 379, 210 and 170 bp in heterozygotes TC.

Statistical analysis

Statistical analysis was performed using Medcalc statistical software. The results were expressed as mean ± SD. Continuous variables were compared using Student’s t test and categorical variables were compared using chi square test. ANOVA test was used to compare significance between three variables. ROC analysis was used to measure strength of association of LH with various variables. Area under curve and p values were computed using z statistics. Strength of LH association with various parameters was analyzed using Pearson’s coefficient of correlation (r). Multiple logistic regression was used to find association of various parameters with PCOS. The genotypic and allelic frequencies variants were analyzed using χ², OR and 95% CI. p Value of <0.05 was considered as a value of statistically significant association.

Author’s contributions

Designed the experiments: ASD; performed the experiments and wrote the manuscript: RD; analyzed the data: SN.

Acknowledgments

The authors are thankful to PGIMS for providing access to labs and patient clinic and to all the participants.

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplementary material

Supplemental data for this article can be accessed here.

Additional information

Funding

This study was not funded by any agency or grant.