Abnormal expression of TLRs may play a role in lower embryo quality of women with polycystic ovary syndrome

ABSTRACT

Toll-like receptors (TLRs) localize in mammalian ovary, including granulosa cells, cumulus cells, and theca cells. Previous studies demonstrated that TLRs may be important for the cumulus-oocyte complex expansion and fertilization. There is no evidence to indicate that the deletion of TLRs will induce infertility; however, the abnormal expression of TLRs may decrease oocyte quality and fertility rate. In the present study, we investigated the effects of polycystic ovary syndrome (PCOS) on the expression of TLRs in cumulus cells by using western-blot and quantitative real-time PCR (qRT-PCR) analyses. We found that the expression of TLR4 and 9 in cumulus cells was influenced significantly by PCOS. We also observed that overweight/obesity changed the expression of TLR2 and 5 in cumulus cells of PCOS subjects. In addition, we found that the rate of available embryos of women with PCOS was slightly lower. These results indicate that the abnormal expression of TLRs in cumulus may be a reason for the lower embryo quality of women with PCOS.

Abbreviations: ART: assisted reproductive technology BMI: body mass index COC: cumulus-cell-oocyte complex PCOS: polycystic ovary syndrome qRT-PCR: quantitative real-time PCR TLRs: Toll-like receptors

KEYWORDS:

• Cumulus
• PCOS
• TLR

Introduction

Toll is a transmembrane receptor protein, which was originally identified as a crucial component of the pathway determining the dorsal-ventral axis during Drosophila embryogenesis [Hashimoto et al. 1988]. In Drosophila, Toll plays a key role in actively defending against the invading of fungi and/or bacteria [Bulet et al. 1999]. The homologues of Toll were also identified in mammalian cells, designated as Toll-like receptors (TLRs), the role of which in innate immunity had been demonstrated [Medzhitov et al. 1997; Rock et al. 1998]. Up to now, 12 members of the TLR family were identified [Chaudhary et al. 1998; Takeuchi et al. 1999; Takeuchi and Akira 2001]. TLRs localize in many tissues and cells, such as cardiac myocytes, dermal endothelial cells, and intestinal endothelial cells [Cario et al. 2000; Faure et al. 2000; Frantz et al. 1999; Mokuno et al. 2000]. Also, TLRs were detected in cumulus cells, granulosa cells, and theca cells [Liu et al. 2008]. For example, the expression of TLR5 was detected in human ovary, peripheral blood, and prostate [Chaudhary et al. 1998]. In mouse, the expression of TLR2, 4, 8, and 9 was detected in cumulus cells and granulosa cells. Shimada et al. [2006] indicated that the TLR pathway has a function in the cumulus-cell-oocyte complex (COC). TLR2 and TLR4 can regulate the expression of Ptgs2, Il6, and Tnfa by their ligands [Shimada et al. 2006]. IL-6, which regulates many genes expressed in cumulus cells and stimulates COCs expansion, is produced by COCs of ovulated follicles [Liu et al. 2009]. TLR2/4 in cumulus cells was regulated by sperm in a hyaluroan fragment-dependent manner. The chemokines secreted from TLR2/4-stimulated COCs could induce sperm capacitation and enhance fertilization [Shimada et al. 2008]. The results in the Drosophila ovary demonstrated that a member of TLRs, 18-wheeler, was crucial for cell migration during folliculogenesis [Kleve et al. 2006]. Furthermore, ovulation is an inflammatory-like process in which COC release from follicle involving tissue remodeling similar to wound repair [Espey 1980]. Polycystic ovary syndrome (PCOS) is one of the common diseases for reproductive age women. Oocyte quality is seriously compromised in PCOS women. However, it is still obscure whether TLRs cause the lower oocyte quality in PCOS women. In the present study, we investigated the expression of TLRs (including TLR 2, 4, 5, 6, 7, 9, and 10) in cumulus cells of PCOS women. Cumulus cells from women without PCOS were used as a control. A previous study reported that TLRs are crucial in adipose tissue inflammation of chronic diseases including diabetes, overweight/obesity, and cardiovascular [Lucas and Maes 2013]. We also observed the effects of overweight/obesity on the expression of TLRs in cumulus cells for PCOS patients because PCOS women are generally compromised by overweight/obesity. In addition, we found that the expression of TLRs in cumulus cells was significantly influenced by PCOS.

Results

Characteristics of women

The characteristics of women included in the study are presented in Table 1. The average age of the PCOS women and control group were similar. The mean body mass index (BMI) of PCOS women with overweight/obesity was higher than PCOS women without overweight/obesity and control (p<0.001), but there was no significant difference for the average BMI between PCOS women without overweight/obesity and control. The average basic level of FSH, progesterone, and E2 was similar between the control and PCOS group. But the basic level of LH and testosterone in blood was higher in the PCOS group than that in the control (p<0.01, Table 1). In the subgroup of PCOS, the level of LH and testosterone was not affected by overweight/obesity (Table 1). There were no other known diseases which may influence folliculogenesis besides PCOS for all the included women.

Table 1. Characteristics of PCOS group and control.

	Control (n=30)	PCOS (n=38)	PCOS with overweight/obesity (n=17)	PCOS without overweight/obesity (n=21)
Average BMI	21.13 ± 1.93^a	24.54 ± 3.25^a	27.49 ± 1.87^b	22.03 ± 1.63^a
Average age (year)	28.76 ± 4.38^a	28.05 ± 3.42^a	28.59 ± 3.50^a	27.62 ± 3.37^a
FSH (mIU/ml)	7.52 ± 2.92^a	6.38 ± 1.78^a	5.93 ± 0.96^a	6.72 ± 2.17^a
E2 (pg/ml)	44.30 ± 17.22^a	38.32 ± 16.10^a	35.60 ± 11.76^a	40.46 ± 18.89^a
P (ng/ml)	0.59 ± 0.25^a	0.58 ± 0.30^a	0.57 ± 0.27^a	0.60 ± 0.23 ^a
LH (mIU/ml)	4.65 ± 1.94^a	10.13 ± 4.27^b	9.57 ± 3.87^b	10.55 ± 4.59^b
Testoterone (ng/ml)	0.22 ± 0.12^a	0.43 ± 0.16^b	0.41 ± 0.15^b	0.45 ± 0.17^b

PCOS: polycystic ovary syndrome; BMI: body mass index; FSH: follicle stimulating hormone; E2: estradiol; P: progesterone; LH: luteinizing hormone; the same letter represents that there is no significant difference; the different letter represents p value is <0.05 or <0.01.

Clinical outcome of assisted reproductive technology (ART)

We analyzed the clinical outcome of ART and the statistical data is presented in Table 2. The mean number of retrieved oocytes in the PCOS group was 14.1% higher than that in the control, but the difference was not significant (p=0.706). The mean number of retrieved oocytes in the PCOS group with overweight/obesity was significantly higher than that in the control p=0.037). Although the fertility rate of the PCOS group without and/or with overweight/obesity was higher than that in the control (p<0.05), the available embryo rate of the PCOS group with and/or without overweight/obesity was slightly lower than that in the control (p>0.05).

The expression of TLR4 and 9 was affected by PCOS in cumulus cells

In the present study, we detected the expression of TLRs in cumulus cells from PCOS patients. The average mRNA levels of TRL2, 5, 6, 7, and 10 in cumulus cells were similar between the control and PCOS group (Figure 1A). However, the average mRNA expression of TRL4 (p=0.018) and 9 (p=0.029) in cumulus cells of the PCOS group were significantly higher compared to the control (Figure 1A). Thereafter, we investigated the protein levels of TRL4 and 9 in cumulus cells. The average protein levels of TRL4 and 9 in the PCOS group (n=38) were significantly different with the control (n=28, p<0.05, Figure 1B,C).

Figure 1. The expression of TRL4 and TRL9. The mRNA and protein expressions of TRL4 and TRL9 were evaluated by qRT-PCR and western-blotting. (A) mRNA expression for control (n=30) and PCOS (n=38); (B, C) protein expression control (n=28) and PCOS (n=38). *p<0.05, **p<0.01.

Overweight/obesity may induce changes of the expression of TLR2 and 5 in cumulus cells of PCOS subjects

When we divided the patients with PCOS into two subgroups a) with overweight/obesity (O-PCOS) and b) without overweight/obesity (N-PCOS), the difference of the mRNA expression levels of TRL2 and 5 in OP (n=18) was significantly changed compared to NOP (n=20, p=0.032 and 0.029, respectively, Figure 2A). The western-blot results also showed that the average protein levels of TRL2 and 5 were significantly different between O-PCOS (n=18) and N-PCOS (n=20, Figure 2B,C).

Figure 2. The expression of TRL2 and TRL5. The mRNA and protein expressions of TRL2 and TRL5 were evaluated by qRT-PCR and western-blotting. (A) mRNA expression for control (n=30), obese-PCOS (n=17), and non-obese-PCOS (n=16); (B, C) protein expression for obese-PCOS (O-PCOS, n=15) and non-obese-PCOS (N-PCOS, n=14). *p<0.05, **p<0.01.

Discussion

TLR4 and 9 are present in COCs including cumulus cell and granulosa cell, and the expression of TLR4 and 9 in COC occurs in a time-dependent manner [Shimada et al. 2006]. In hen ovarian granulosa cells, the expression of TLR4 is dependent on the stage of follicle maturation [Woods et al. 2009]. Previous studies also showed that lipopolysaccharides could perturb bovine folliculogenesis and meiotic competence via the TLR4 pathway [Bromfield and Sheldon 2011; 2013; Shimizu et al. 2012]. These studies indicate that TLR4 and 9 may play a key role in folliculogenesis. In this present study, we found that TLR4/9 was expressed in cumulus cells and the expression of TLR4/9 in the PCOS group was significantly higher than the control.

TLRs are crucial factors in the innate immune surveillance system [Lucas and Maes 2013]. Hyaluronan fragments in the COC mix can bind and activate TLR4 [Jiang et al. 2005; Termeer et al. 2002]. When COC was co-cultured with sperm, TLR4 was activated in a hyaluronan fragment-dependent manner [Shimada et al. 2008]. Once the neutralizing antibodies of TLR2 and TLR4 were added to the culture medium, the activation of the NFkappaB pathway was suppressed [Shimada et al. 2008]. Furthermore, TLR2/4 can induce sperm protein tyrosine phosphorylation [Shimada et al. 2008]. These results indicate that TLR4 and TLR2 are vital for folliculogenesis and fertilization. In this present study, we found that the fertility rate of the PCOS group was significantly higher than that in the control. The increase of the fertility rate of the PCOS group may be caused by the increase of the expression of TLR4 and TLR9 in cumulus cells of the PCOS group including the obese and non-obese PCOS women. In addition, we found that the average BMI of PCOS subjects was higher than that of the control. Thus we investigated whether the changed expression of TLR4/9 was caused by overweight/obesity. We divided the PCOS women into two subgroups: with overweight/obesity and without overweight/obesity. Our data showed that the expression of TLR4/9 was similar between the subgroups of PCOS (Figure 2A). This indicates that the changed expression of TLR4/9 was not induced by overweight/obesity. However, we found that the basic level of LH and testosterone was higher in the PCOS group than that in the control. The base level of LH and testosterone in the PCOS group was not affected by overweight/obesity. Previous studies have shown that the expression of TLRs was regulated by LH in different stages of folliculogenesis in granulosa cells of hens [Jiang et al. 2005] and testosterone could affect the expression of TLRs in hepatic cells [Al-Quraishy et al. 2014]. Therefore, we conclude that the changed expression of TLR4/9 may be induced by the abnormal basic level of LH and testosterone.

However, we found that the expression of TLR2/5 increased in overweight/obese PCOS subjects compared with PCOS subjects without overweight/obesity. A previous study has shown that there is an increase in the concentration of TLR5 in blood of obese people [Kim et al. 2014]. Using an in vitro adipocyte model shows that tissue metabolism and inflammation in obese humans may be induced by fagellin-activated TLR5 signaling [Pekkala et al. 2015]. In mice, overweight/obesity also activates the TLR signaling cascades [Kim et al. 2012]. These studies indicate that the folliculogenesis may be influenced by maternal overweight/obesity. Because it is difficult to collect enough materials from overweight/obese women without PCOS in the clinic more studies are needed to clarify the relationship between the abnormal expression of TLR2/5 and women with PCOS with overweight/obesity.

A previous study indicated that TLR2/4 is not only vital for COC expansion, but also plays a key role during fertilization [Shimada et al. 2008]. In an in vitro fertilization assay, TLR2/4 activates cumulus cells to release cytokines and chemokines which can enhance fertilization [Jiang et al. 2005]. TLR2/4 may also respond to factors in addition to sperm that enter the reproductive tract when infected [Girling and Hedger 2007]. Therefore, the expression changes of TLRs may be a reason for the lower available embryo rate of PCOS women. The detailed mechanisms of how TLRs decreases available embryo rate are still obscure.

Conclusion

In summary, we detected the expression of TLRs in cumulus cells of PCOS women, and found that the protein and mRNA levels of TLRs was changed in cumulus cells when PCOS subjects were compared to the control. Therefore, we concluded that TLRs may play a role in the lower available embryo rate of women with PCOS.

Materials and methods

Cumulus cells collection

The present study was supported by the Ethics Committee of Zhengzhou University. All the people included in the present study had signed the consent and permissions. We had obtained the consent from participants to publish the data. A total of 38 PCOS women and 30 women without PCOS received artificial assisted reproduction for the first time in the Reproductive Medicine Center of People’s Hospital of Zhengzhou University. Notably, 17 of the 38 PCOS women are overweight/obesity. At 36-37 h post-human chorionic gonadotropin (hCG), COCs were retrieved by transvaginal ultrasound-guided. Cumulus cells of PCOS and non-PCOS patients were collected and stored at -80ºC until used.

RNA purification and qRT-PCR

Total RNA of cumulus cells was extracted by using EZ-10 Spin Column Total RNA Isolation Kit (Sangon Biotech, Beijing, China) according to the manufacturer’s instructions. The first strand cDNA was synthesized by utilizing FastQuant RT Kit (Tiangen, Beijing, China). A Roche LightCycler480 (Roche, Switzerland) was used for qRT-PCR. Triple samples were analyzed for each gene and β-actin was used as a control. The expression level was evaluated by 2^−ΔΔCt [Ge et al. 2013]. All the primers are shown in Table 3.

Table 2. Clinical outcome of assisted reproductive technology.

	Mean no. of retrieved oocyte	Fertility rate (%)	Available embryo rate (%)
Control (n=30)	12.8 ± 6.8 (385/30)^a	62.6 (241/385)^a	59.8 (144/241)^a
PCOS (n=38)	14.6 ± 7.0 (556/38)^a	69.1 (384/556)^b	54.4 (209/384)^a
PCOS without overweight/obesity (n=17)	17.2 ± 6.6 (293/17)^b	70.3 (206/293)^b	53.4 (110/206)^a
PCOS with overweight/obesity (n=21)	12.5 ± 6.7 (263/21)^a	67.7 (178/263)^ab	55.6 (99/178)^a

PCOS: polycystic ovary syndrome; the same letter represents that there is no significant difference; the different letter represents p value is <0.05.

Table 3. Primers used in qRT-PCR.

Genes	Forward	Reverse
TRL2	5’ ATCCTCCAATCAGGCTTCTCT-3’	5’-GGACAGGTCAAGGCTTTTTACA-3’
TRL4	5’- AGGGCATGCCTGTGCTGAGTTTGAAT-3’	5’-TATGCAGCCAGCAAGAAGCATCAGGT-3’
TRL5	5’- TCCCTGAACTCACGAGTCTTT-3’	5’-GGTTGTCAAGTCCGTAAAATGC-3’
TRL6	5’-TGAATGCAAAAACCCTTCACCT-3’	5’- CCAAGTCGTTTCTATGTGGTTGA-3’
TRL7	5’-TCGTGGACTGCACAGACAAG-3’	5’-GGTATGTGGTTAATGGTGAGGGT-3’
TRL9	5’-CTGCCACATGACCATCGAG3’	5’-GGACAGGGATATGAGGGATTTGG-3’
TRL10	5’-GATGGTTGGATGGTCAGATTC-3’	5’- AAGCCCACATTTACGCCTATC-3’
β-actin	5’-CTACAATGAGCTGCGTGTGG-3’	5’-TAGCTCTTCTCCAGGGAGGA-3’

Western-blotting

COCs were added in 1% hyaluronidase solution and cumulus cells were separated from oocytes by mechanically pipetting for less than 1 min. Then cumulus cells were transferred to 15 mL centrifuge tubes and centrifuged for 10 min at 1,500 rpm. Cells were washed 3 times using culture medium (DMEM/F12, Sigma-Aldrich, USA) containing 10% fetal bovine serum and then cultured in cell culture dishes. The dishes with cumulus cells were incubated at 37ºC for 48 h with 5% CO₂ and Genipin (Alexis, USA). Cultured cumulus cells were washed by PBS and incubated at 4ºC. Washed cumulus cells were lysed by RIPA (Beyotime, China) with PMSF (1 mM) on ice for 30 min and crushed by ultrasonication. Then it was centrifuged at 12,000 rpm, 4ºC for 10 min using a refrigerated centrifuge (Sigma, Germany). The supernatant was transferred to a new EP tube and stored at -20ºC until use. Total protein concentration was analyzed using a BCA kit (DingGuo, Beijing). The protein expression was analyzed by SDA-PAGE (sodium dodecyl sulfate polyacrylamide gel electropheresis). The anti-TLRs antibodies (abcam, USA) were used as 1:300. The protein of β-actin was used as a control. Relative quantity expression of protein was evaluated by Quantity One software.

Statistical analysis

Data are represented as mean ± SD. The significance between groups was compared by independent-samples T test and One-way ANOVA. The difference of fertility rate and available embryo rate was tested using Chi-square test. A probability level of p<0.05 was considered significant.

Declaration of interest

This work was supported by Scientific and Technological Foundation of Henan Province (142102310077). There is no conflict for all the authors.

Acknowledgments

We appreciate all the members who gave us so much help during this study in the Reproductive Medicine Center of People’s Hospital of Zhengzhou University.

Additional information

Notes on contributors

Bao-Xia Gu

Designed the study and analyzed the data: B-XG, C-LZ; Collected the materials and performed qR-PCR and western-blot: XW, B-LY, H-BG, H-LZ, S-DZ; Wrote the manuscript:C-LZ .