Association of Angiotensin-Converting Enzyme gene polymorphism in Pakistani women with the atypical steroidogenesis in Polycystic ovarian syndrome: A case-control study.

KN designed the study, designed the questionnaire, collected data from participants, applied statistics and drafted this manuscript. SR contributed to study design and helped in finalizing the manuscript. NML participated in questionnaire designing, statistical analysis and finalizing the manuscript. SS and AEZ helped in sample collection from respective OPDS and reviewed the final draft. SKS participated in drafting the manuscript. All authors critically reviewed the manuscript and approved the final version.

All the authors consented for the paper publication.

Ethical Review Boad of Citi Lab and Research Centre approved this work under Ref # 26-17/ERB/CLRC/ 27th. All the required material for project was evaluated by committee including questionnaire and consent form.

This work was partially funded by the University of The Punjab, Lahore, Pakistan.

A total of 252 subjects were recruited from the gynaecology outpatient clinic, Jinnah Hospital, Lady Willingdon Hospital, CRC labs and Institute of Microbiology & Molecular Genetics, University of the Punjab Lahore. Ethical approval for the study was taken from CRC Labs (Ethical Review Board Ref # 26-17/ERB/CLRC/ 27th; dated 28-07-2016). The study subjects included were 162 PCOS patients and 90 healthy women. All participants provided written, informed consent with information about the research project. Written informed consent was taken from the parents or guardians of the participants under 18 years of age. Rotterdam diagnostic criteria 2003 ( ESHRE & Group, 2004 ) was used to select the PCOS cases. Healthy subjects with the normal menstrual cycle, ovaries without any cyst and having no signs of hyperandrogenism were included as controls. The age of participants of the study ranged between 16 and 40 years. Women on hormonal medication and infectious diseases, hypothyroidism, androgen-secreting cancers, adrenal hyperplasia, insulin resistance or hyperglycemia, advanced stage (III or IV) endometriosis or suffering from premature ovarian failure were excluded from the study. All participants filled a survey-based questionnaire. It included the information of sociodemographic characters, age, height, weight, body mass index (BMI), waist circumference (WC), fasting glucose level (above 100 mg/dL), menstrual cycle and its irregularity, hirsutism, acne, weight gain, skin discolouration, mood swings, patient’s history of metabolic disorders as diabetes, hypertension, cardiovascular diseases (CVDs), vitamin and calcium deficiency (Vit/Ca deficient), use of fertility medicines and oral contraceptives and family history of PCOS, diabetes, hypertension, cardiovascular diseases and Vit/Ca deficiency. ELISA kits (Accesss 2 beckman coulter) were used to measure Leutinizing hormones (LH) and follicle-stimulating hormones (FSH). The ratio of 1:1 for LH: FSH was considered normal. A volume of 5 ml of peripheral blood was collected on the 2nd or 3rd day ( PCOS guidelines, 2012 ) of the menstrual cycle. It is equally aliquoted into EDTA and serum vials and stored at −20℃ for genomic studies and serum analysis. In nested PCR, triple primers were used to observe the ACE polymorphism. A reaction mixture of 20 μl contained forward primer 5′ TCCCATCCTTTCTCCCATTTCT’3, 0.6 μl, reverse primer CATGCCCATAACAGGTCTTCAT’3 0.6 μl and internal nested primer 5′ GGTTTCACCGTTTTAGCCG’3 ( Nawaz and Hasnain, 2009 ) 0.6 μl, master mix 10 μl, nuclease-free water 6.2 μl and DNA template 2 μl. PCR program was developed for 35 cycles with denaturation at 94 °C for 30sec, annealing at 58 °C for the 30sec, and extension at 72 °C for 30sec in a Bio-Rad T100 thermal cycler. Bands of amplified products were visualized on 2% agarose gel under UV illuminator. The PCR results were evaluated by observing the bands as 210 bp fragment for DD, 498 and 264 bp fragments for II, and 498, 264, 210 bp fragments for ID genotype. Statistical Package for Social Sciences (SPSS, IBM statistics) version 22, MEDCALC online were used for statistical analysis. Data were evaluated by calculating mean, standard deviation, percentages, odd ratio, independent T-test and Chi-square values. Alleles and genotypes were following Hardy-Weinberg equilibrium (HWE), calculated using an online calculator ( http://apps.biocompute.org.uk/hwe-mr-calc.html ) . The level of significance was <0.05. For correlation analysis, the Pearson correlation was applied.

The anthropometric variables, comorbidities, clinical characteristics in PCOS and non-PCOS groups are shown in Table 1 . PCOS women were older than controls with mean ages of 25.21 ± 4.570 years and non-PCOS 22.9663 ± 4.239 years. Means of the weight of PCOS patients were tremendously high 68.45 ± 14.18 as compared to controls 53.84 ± 8.54. There was a difference between the mean values of BMI of patients 27.694 ± 5.770 and controls 20.17 ± 6.315. In Fig. 1 , the box plot's median values represent the inclination of age and BMI among the two groups. WC of PCOS was also slightly higher than the healthy women. These differences in the mean ± standard deviation values were significant statistically between experimental and control groups regarding age, weight, BMI and WC except height and fasting glucose level. However, diabetic patients had a high fasting glucose level. Table 1 Anthropometric and clinical data of PCOS and non-PCOS group. Variables PCOS (n = 162) Controls (n = 92) P-value Anthropometric Characters Age 25.21 ± 4.570 22.9663 ± 4.23849 0.0001* Height 1.6687 ± 0.544 1.59 ± 0.064 0.1685 Weight 68.45 ± 14.18 53.84 ± 8.54 0.0001* BMI 27.694 ± 5.770 20.17 ± 6.315 0.0001* WC 0.82 ± 0.084 0.8 ± 0.021 0.0258* Fasting Glucose level 100 ± 24.07 96.70 ± 11.59 0.2019 Diagnostic criteria characteristics Ultrasound scan Confirmed 108(66.6%) 6(0%) 0.0001* Irregular periods 161 (99.3%) 3(3.3%) 0.0001* Facial hairs 121(75.2%) 4(4.3%) 0.00* Biochemicals tests Elevated LH: FSH 102 (63%) 33(35.9%) 0.0001* Comorbidities Hypertensive 64(39.5%) 4(4.3%) 0.0001* Diabetes 7(4.32%) 2(2.17%) 0.3738 CVDs 1(0.61%) 0(0%) 0.4538 Vit/Ca deficient 12(7.4%) 0(0%) 0.0076* Family history of other diseases Hypertension 88(54.32%) 11(11.9%) 0.0001* Diabetes 86(53%) 25(27.1%) 0.0001* CVDs 16(9.8%) 3(3.2%) 0.0541 Vit/Ca deficiency 1(0.61%) 0(0%) 0.4538 PCOS 41(25.3%) 0(0%) 0.0001* Other clinical conditions Rapid Gaining weight 136(85.5%) 5(5.4%) < 0.0001* Skin coloration 76(47.5%) 4(4.3%) < 0.0001* Acne 104(64.6%) 12(13%) < 0.0001* hungry/fatigue 132(81.4%) 0 (0%) < 0.0001* Mood swings 130(80.2%) 16(17.4%) < 0.0001* Infertility 91(56.13%) 5(5.4%) < 0.0001* Surgery history 26(16.1%) 3(3.2%) 0.0019* Taking contraceptives 26(16.0%) 0(0%) < 0.0001* * indicate the level of significance P ≤ 0.05. Fig. 1 The box plot (A) indicates maximum, minimum, median values of age, along with lower and higher quartile among PCOS and non PCOS women. (B) This represents the maximum, minimum, median, and lower and higher quartile values for BMI among PCOS and non PCOS women. Anthropometric and clinical data of PCOS and non-PCOS group. * indicate the level of significance P ≤ 0.05. The box plot (A) indicates maximum, minimum, median values of age, along with lower and higher quartile among PCOS and non PCOS women. (B) This represents the maximum, minimum, median, and lower and higher quartile values for BMI among PCOS and non PCOS women. Among the clinical features of criteria for assessing the PCOS, ovaries with multiple cysts, irregular menstrual cycle and hirsutism were significantly high in PCOS vs non PCOS women (P-value < 0.05) ( Table 1 ). Frequencies of comorbidities as hypertension, diabetes, CVDs and Vit/Ca deficiency among PCOS women were higher than controls. More than half of PCOS women had family histories of hypertension and diabetes. About one-quarter of the study population of patients told the presence of the PCOS in their families. These differences in the frequencies of hypertension and Vit/Ca deficiencies and family histories of hypertension, diabetes and Vit/Ca deficiencies were significant statistically among the two groups. As for other PCOS outcomes, rapid weight gain, skin discolouration, acne, fatigue, mood swings, infertility, and surgery history were common in PCOS patients with P-value  0.05) ( Table 1 ). Three different genotypes were observed on agarose gel; 210 bp fragment for DD; 498 and 264 bp fragments for II; and 498, 264, 210 bp fragments for ID genotype as shown in Fig. 2 . Fig. 2 Bands of ACE polymorphic alleles, lane 1 contain ladder of 100 bp, lane 2 with 3 bands of 498 bp, 264 bp 210 bp indicate the ID genotype, lane 3 and 5 with one band of 210 bp indicate the DD genotype, Lane 4 with 2 bands of 498 bp, 264 bp indicate II genotype. Bands of ACE polymorphic alleles, lane 1 contain ladder of 100 bp, lane 2 with 3 bands of 498 bp, 264 bp 210 bp indicate the ID genotype, lane 3 and 5 with one band of 210 bp indicate the DD genotype, Lane 4 with 2 bands of 498 bp, 264 bp indicate II genotype. The allelic and genotypic frequencies in patients and controls, along with the ACE polymorphism percentages, is shown in Table 2 , by using different penetrance models. The P-value (0.005) for the additive model showed that polymorphism at ACE was significantly associated with atypical steroidogenesis in PCOS women. The additive model shows a high allelic frequency for allele I in controls, whereas allele D has the highest frequency in patients. The codominant model reports the highest genotypic frequency for heterozygous genotype (ID) in controls vs patients; II and DD genotypic frequency is higher in patients. Furthermore, the genotypic frequencies for patients and controls are nearly similar for patients and controls in dominant and recessive models. Table 2 Different penetrance models showing genotypic and allelotypic frequencies for ACE gene polymorphism. SNP Patients % Control% OR (95% CI) p-value Codominant II 58 (36.02) 40 (0.44) 1.62 (0.96, 2.71) 0.064 DD 91 (0.57) 35 (0.38) ID 12 (0.07) 16 (0.16) Additive I 122 (0.39) 88 (0.53) 0.57 (0.39, 0.84) 0.0050* D 188 (0.61) 78 (0.47) Dominant DD 90 (0.56) 40 (0.44) 1.62 (0.96, 2.71) 0.0878 II + ID 71 (0.44) 51 (0.56) Recessive II 58 (0.56) 35 (0.38) 0.90 (0.53, 1.53) 0.7859 DD + ID 103 (0.64) 56 (0.62) Frequencies are represented as number and percentage; p-Values were calculated using Chi-square test; p-values < 0.05 were considered statistically significant; n, Number of subjects; OR, Odds ratio; CI, Confidence interval. Different penetrance models showing genotypic and allelotypic frequencies for ACE gene polymorphism. Frequencies are represented as number and percentage; p-Values were calculated using Chi-square test; p-values < 0.05 were considered statistically significant; n, Number of subjects; OR, Odds ratio; CI, Confidence interval. A significant association was found between three genotypes of ACE and Leutinizing hormone: Follicle-stimulating hormone (LH: FSH) ratio in the PCOS subjects (P ≤ 0.05), using ANOVA analysis. It was observed that both patients and controls with DD genotypes had a higher LH: FSH ratio, contrary to those having and ID and II genotypes. In controls, the distribution of genotypes was not associated with LH: FSH ratio (P = 0.5) ( Table 3 ). Also, the genotypes did not show a significant association with anthropometric characteristics, comorbidities, clinical conditions and other related parameters of PCOS (P-value > 0.05). Pearson correlation between LH: FSH ratio and PCOS patients' clinical parameters were determined in both patients and the control group. A positive correlation was revealed between patients with diabetes and their LH: FSH ratio (r = 0.199*, P-value 0.047). In controls, infertility showed a negative correlation with LH: FSH ratio (r = -0.0321, P-value 0.023). No significant correlation was found in ACE genotypic distribution regarding age, BMI, waistline, hypertension, family history of hypertension, diabetes and PCOS, irregular menstrual cycle, hirsutism, gaining weight, acne, and found ( Table 4 ). Table 3 Association of ACE polymorphism with LH: FSH ratio. PCOS Non-PCOS Genotype N Mean ± S. D P- value N Mean ± S. D P-value LH: FSH II 39 0.28 ± 0.46 0.050* 32 0.76 ± 0.83 0.5 DD 59 0.53 ± 0.50 20 1.143 ± 0.95 ID 4 0.50 ± 0.58 14 0.80 ± 0.92 Total 102 0.43 ± 0.5 66 0.88 ± 0.88 * indicate the level of significance P ≤ 0.05. Table 4 Pearson correlation between LH: FSH ratio and other parameters of PCOS. Series no. Parameters PCOS Non-PCOS Correlation ‘r’ P-value Correlation ‘r’ P-value LH: FSH ratio 1.00 0.00 1.00 0.00 1 Age −0.186 0.066 0.059 0.687 2 BMI −0.035 0.738 0.099 0.564 3 Waist line −0.074 0.476 0.057 0.693 4 Hypertension −0.067 0.508 0.022 0.884 5 Diabetes 0.199* 0.047 0.007 0.960 6 Family history of Hypertension −0.097 0.335 0.016 0.912 7 Family history of Diabetes 0.136 0.178 −0.040 0.784 8 Family history of PCOS 0.002 0.986 −0.045 0.776 9 Irregular periods −0.035 0.733 −0.054 0.710 10 Hirsutism 0.168 0.098 0.011 0.942 11 Gaining weight −0.025 0.810 −0.084 0.560 12 Acne 0.060 0.559 −0.277 0.052 13 Infertility −0.059 0.576 -0.0321* 0.023 14 Genotyping ACE −0.172 0.091 −0.104 0.482 Cannot be computed because atleast one of variables is constant. Correlation is significant at the level of 0.05* (2-tailed). Correlation is significant at the level of o.oq** (2-tailed). Association of ACE polymorphism with LH: FSH ratio. * indicate the level of significance P ≤ 0.05. Pearson correlation between LH: FSH ratio and other parameters of PCOS. Cannot be computed because atleast one of variables is constant. Correlation is significant at the level of 0.05* (2-tailed). Correlation is significant at the level of o.oq** (2-tailed).

This study showed the means of body weight, BMI and WC were significantly higher in PCOS compared to healthy women. This exhibits the role of obesity in the development of clinical features of PCOS. The research revealed that the excess of adipose tissue downregulates the adipokines necessary for insulin sensitivity in PCOS. A case-control study from Iran revealed high BMI and WC in PCOS diagnosed women ( Behboudi-Gandevani et al., 2017 ). Another research from India was in accordance with our study, also obtained 1.74 times high BMI in PCOS than non-PCOS ( Begum et al., 2017 ). In our results, PCOS patients with hypertension, vitamin and calcium deficiencies were significantly high in number. Also, the family histories of hypertension, diabetes and PCOS were common among patients. In previous studies, hypertension was presented as a PCOS risk factor and highly prevalent in women with PCOS ( Ollila et al., 2018 ). In Moldova, about half of the PCOS patients had a family history of mothers suffering from PCOS ( Voloceai, 2017 ). Deficiencies of vitamins and calcium may be triggering hormonal imbalances in the body. The beneficial effects of the co-supplementation of vitamins and calcium on hormonal profile in PCOS women show their role in the pathogenesis of the disorder. According to prior studies, women with PCOS were more likely to have obesity, hypertension, diabetes mellitus (T2DM) and cardiovascular disorders ( Fauser, 2018 , Fernandez et al., 2018 ). Hyperandrogenic conditions like acne, skin discolouration, extreme hunger, fatigue, mood swings and psychological issues were common among PCOS women. A large amount of relevant literature revealed that women with PCOS were more prone to have these clinical conditions due to elevated androgen ( Dashti et al., 2019 , Lopes et al., 2018 , Stapinska-Syniec et al., 2018 ). As an indication of the association of PCOS with the absence of pregnancy, more than half of the cases were facing infertility. Balen and Morley reported that approximately 80% of women who were suffering from anovulatory infertility had PCOS ( Balen et al., 2016 , Fauser, 2018 , Fernandez et al., 2018 ). The genotypic analysis revealed the high prevalence of allele D and genotype DD in the PCOS group. In controls, allele I and II genotype were observed more often than DD. High percentages of DD genotype in patients were reported in PCOS patients with hyperandrogenism ( Cintra et al., 2018 , Deepika et al., 2013 ). By further analysis, we found the differences in genotypic distributions and allelic frequencies were significant. Although no significant association of ACE I/D polymorphism with PCOS phenotypic characteristics was found, yet the LH: FSH ratio in the PCOS varied among the three genotypes in cases significantly (P ≤ 0.05). This ratio was found significantly higher in DD genotype than other genotypes in both patients and control. It indicates that elevated ACE levels in the plasma due to I/D polymorphism may be responsible for hyperandrogenism in PCOS patients. Ang II, the ACE product, inhibits the progesterone secretions that give rise to the secretion of LH through a negative feedback mechanism ( Che et al., 2009b ). An increase in LH and reduced or normal secretions of FSH stimulate androgen production in the ovary ( Kumar et al., 2016 ). It is evident from an observational study in which androgen levels decrease after treatment with the angiotensin-converting enzyme inhibitor for four weeks ( Connolly et al., 2018 ). This implicates the fact that polymorphism may cause a surge of LH, leading to PCO ( Che et al., 2009a , Hashemi et al., 2017 ). These findings suggest that I/D polymorphism leads to abnormal RAS activity and is associated with the onset of metabolic disturbances, glycemic dysregulation, PCOS and other clinical consequences of PCOS such as hypertension and insulin resistance ( Cintra et al., 2018 ). Another interesting finding of our research was the positive correlation of diabetes in PCOS women with their abnormal LH: FSH ratio. The high prevalence of Diabetic Mellitus type 2 in PCOS women may be due to the direct relation of insulin resistance with the LH: FSH ratio ( Smaism et al., 2016 ). A follow-up study over 1919 person-years observed PCOS as a significant risk factor of Diabetic Mellitus type 2 in women of reproductive age ( Kakoly et al., 2018 ). Pakistan has scarce data published on PCOS and its phenotypic and genotypic pathologies, while no work yet has been reported regarding the association of ACE polymorphisms with PCOS. The significance of our research is presenting the association of ACE gene polymorphism with hormonal parameters of PCOS in Pakistani women. This represents the growing interest in the importance of gene polymorphisms in the pathogenesis of PCOS. However, a short period of research limits the sample size of the study. Also, the participation of first-degree relatives of the subjects could enhance the significance of this study. New strategies are needed to screen all subjects for genetic alterations of PCOS pathogenesis. This will help prevent the disease, transition to self-management, and improve women's health in Pakistan.

In conclusion, I/D polymorphism was significantly associated with atypical LH: FSH ratio in PCOS patients of the Pakistani population. Therefore, the presence of I/D polymorphism suspects to affect the steroidogenesis process, which in turn can trigger the development of PCOS in women of reproductive age. It is suggested that ACE I/D polymorphism may play a role in the pathogenesis of the disease, while it is not the main etiological factor of PCOS.

Polycystic Ovarian Syndrome (PCOS) has become a common problem among women of reproductive age in Pakistan and affects 8–12% of the world's population ( Donà et al., 2017 ). It is a complicated endocrinopathy with several adverse effects on women health ( Abinaya et al., 2019 ). Prominent diagnostic features are polycystic ovaries, hyperandrogenism, high luteinizing hormone (LH) levels, insulin resistance (IR), menstrual disorder and hirsutism ( Lone et al., 2020 , Munawar Lone et al., 2020 , Singh et al., 2019 ). Both environmental and genetic risk factors tremendously contribute to the development of broad-spectrum metabolic defects of PCOS ( Merkin et al., 2016 ). In previous research, genes involved in ovarian and adrenal steroidogenesis, insulin action, energy homeostasis, and chronic inflammation were mainly focused on PCOS candidate genes ( Dadachanji et al., 2018 , Lone et al., 2020 ). Renin-Angiotensin System (RAS) plays a crucial role in body homeostasis, oocyte maturation, ovulation, the formation of corpus luteum, and steroidogenesis. It is expressed in various tissues, including the ovary ( Bayram et al., 2011 ). ACE, also called peptidyl dipeptidase A or kininase II, acts as a main rate-limiting component of the RAS system. The ACE gene encodes it on chromosome 17. The ACE product, Angiotensin II (Ang II) contributes to the many physiologic processes such as steroidogenesis ( Bayram et al., 2011 , Bernstein et al., 2013 ). Besides, its elevated concentrations are suspected to be involved in the pathogenesis of obesity, inflammation, oxidative stress, insulin resistance and PCOS ( Ramalingam et al., 2017 ). However, a few studies have observed the role of ACE in hormonal imbalance. The presence of the ACE gene with insertion-deletion polymorphism (I/D polymorphism), based on the insertion of 287 bp sequence, is associated with elevated ACE levels in plasma. Subsequently, it upregulates the RAS and leads to PCOS development by disrupting the normal steroidogenesis ( Palumbo et al., 2016 ). Thus, it is of interest for researchers to evaluate the role of ACE in the pathophysiology of PCOS. Many studies revealed that high concentrations of ACE responsible for insulin resistance (IR) by producing reactive oxygen species (ROS) in tissues ( Manucha et al., 2015 , Ramalingam et al., 2017 ). IR provokes hyperandrogenism that is a significant manifestation of PCOS phenotypes. I/D polymorphism in the ACE gene induces low-grade inflammation due to ROS. IR and hyperandrogenism are the metabolic risk factors of the deadly quartet (obesity, diabetes mellitus, etc.) and associated with PCOS pathophysiology ( Barrea et al., 2018 ). Previously we studied the associations of pathological causes of PCOS with different genes ( Lone et al., 2020 , Munawar Lone et al., 2020 ). Here, we aimed to investigate the association of I/D polymorphism of the ACE gene with the risk of polycystic ovary syndrome by disrupting the steroidogenesis process.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

The analysed data for the current study will be available from the corresponding author.