Glutamine and norepinephrine in follicular fluid synergistically enhance the antioxidant capacity of human granulosa cells and the outcome of IVF-ET

An increasing number of studies demonstrate that changes in neurotransmitters metabolic levels in follicular fluid are directly related to oocyte maturation, fertilization, the quality of embryo and pregnancy rates. However, the relationship between the intra-follicular neurotransmitters and the function of granulosa cells (GCs), and the outcome of in vitro fertilization-embryo transfer (IVF-ET) is not clear. Human follicular fluid and cumulus GCs were harvested from large follicles obtained from patients undergoing IVF. Neurotransmitters and steroid hormones in follicular fluid were measured through liquid chromatography-tandem mass spectrometry (LC–MS/MS) and high-performance liquid chromatography-mass spectrometry (HPLC–MS/MS). Based on the content of glutamine (Gln) in follicular fluid, the samples were divided into two groups: high Gln level group and low Gln level group. The expression of proliferation-, steroidogenesis- and antioxidant-related genes in GCs was detected by qRT-PCR. In vitro, KGN cells were used to further verify the effects of Gln and NE on GCs function. Primary and secondary outcomes were the number of mature and retrieved oocytes, and the ratio of high-quality embryos, respectively. Gln (46.75 ± 7.74 μg/mL) and norepinephrine (NE, 0.20 ± 0.07 μg/mL) were abundant neurotransmitters in follicular fluid, and exhibited a significantly positive correlation (R = 0.5869, P < 0.005). In high Gln level group, the expression of proliferation, steroidogenesis and antioxidant-related genes in GCs were higher than those in low Gln level group, and the contents of estriol and E2 in follicular fluid were more abundant. Moreover, the concentrations of Gln and NE in follicular fluid showed significantly positive correlation with IDH1 expression in GCs (R = 0.3822, R = 0.4009, P < 0.05). Importantly, a significantly positive correlation was observed between IDH1 expression in GCs and the ratio of higher-quality/cleaved embryos (R = 0.4480, P < 0.05). In vitro studies further demonstrated that Gln and NE played synergistically function in improving GCs proliferation and E2 production by upregulating IDH1 expression. These data demonstrate that Gln and NE in follicular fluid might play significant positive roles in GCs function, and may be potential predictors for selecting optimal quality oocytes and evaluating the quality of embryonic development.


Background
Subfertility, also commonly referred to infertility, is a reproductive system disease characterized by failure to establish a clinical pregnancy after normal unprotected sexual intercourse for 12 months or more, or impairment of the reproductive capacity of an individual or with his or her partner [1,2]. Subfertility has been recognized as a worldwide public health problem by the World Health Organization (WHO), and approximately half of these cases lack an de nitive explanation for pregnancy failure [3]. It is estimated that in western countries, one in six couples has subfertility, which is recognized as the cause of psychological distress in both men and women [4]. Assisted reproductive technology (ART) including in vitro fertilization (IVF) and intracytoplasmic sperm injection (ICSI) is a promising option with suitable interventions for the clinical management of subfertility [5]. IVF offers a possibility for achieving pregnancy, but the success rate is still suboptimal [6,7].
A healthy intrafollicular environment supports the production of developing oocytes and affects oocyte quality by coordinating the cumulus GCs and the follicular uid [8]. Follicular uid originates from both plasma and secretion by cumulus GCs. As an important bio uid, follicular uid provides a special microenvironment for follicular development and impacts oocyte quality, implantation and early embryo development. Follicular uid is rich in proteins, including sex hormones, nutrient substances, cytokines, growth factors and neurotransmitters, which is critical for oocyte development [9]. Cumulus granulosa cells represent the predominant somatic cell type of follicle development and are involved in steroidogenesis and folliculogenesis. Furthermore, they connect with oocytes via direct gap junctions to support the developing oocyte [10][11][12]. The process of follicular development involves local biochemical exchanges and substantial modi cations in cellular metabolism.
Previous studies have demonstrated that NE in the follicular uid derived from the sympathetic neurons of the peripheral nervous system in the mammalian ovary takes part in the regulation of ovarian function [13,14]. Follicular cells such as granulosa and theca express β-adrenergic receptor in respond to the β-adrenergic agonist, leading to a physiological follicular development, ovulation and the synthesis and production of steroid hormones [15]. In addition, NE and its metabolites could induce the generation of reactive oxygen species (ROS) in human GCs [16]. Once the level of NE concentration exceeds the physiological concentration, it may lead to ROS-related pathophysiology events including oxidative stress and cell death [17]. Thus, the content of NE in the follicular microenvironment could be a key physiology or physiological indicator for the signaling in steroidogenic Leydig cells and the process of ovulation.
Amino acids serve as substrates for the synthesis of proteins, not only enhancing embryonic development but also affecting blastocyst formation in mice. A study demonstrated that Glycine (Gly), Glutamine (Gln), alanine (Ala), Glutamate (Glu) and Proline (Pro) were the most abundant amino acids in follicular uid [18]. The disturbance of follicular amino acid metabolism is related to women with polycystic ovary syndrome (PCOS), which may be the reason for the poor pregnancy outcome in obese patients and increased risk of abortion in PCOS patients [19]. In addition, they are not only components of proteins, but also regulate neurotransmitter homeostasis, such as a metabolite shuttle known as the glutamate/GABA-glutamine cycle describes the release of neurotransmitter glutamate or GABA from neurons and subsequent uptake into astrocytes, play import roles in nervous system [20,21]. However, whether the neurotransmitters metabolic levels provided by follicular uid in uences the function and hormone synthesis of cumulus granulosa cells need to be further investigated.
Oxidative stress plays an important role in reproductive processes, including follicular development, oocyte maturation, ovulation and fertilization, as well as embryo implantation and embryo development [22]. Excessive generation of ROS, if not e ciently counterbalanced by antioxidative enzymes, may cause damage to oocytes and granulosa cells, leading to poor oocyte quality [11]. In addition, increased levels of endogenous ROS and reduced antioxidant genes expression diminish cellular antioxidant defense capability, resulting in oxidative damage to lipids, proteins and DNA in ovary [23]. In our previous study, psychological stress-induced oxidative stress decreased the expression of the antioxidant gene IDH1 to disturb and compromise the function of granulosa cells, accelerating ovarian senescence [24]. Antioxidant treatment with melatonin can provide reproductive medicine with a way to improve IVF outcomes [25]. Additionally, antioxidant supplementation in IVF culture medium could increase fertilization rates and subsequent healthy embryo development [26].
However, little is known about the impact of neurotransmitters in follicular uid on oxidative capacity of granulosa cells.
During the process of follicle growth to ovulation, which is mediated by the preovulation production of gonadotropins, metabolic pro les change dynamically in follicular uid. Thus, a better understanding of the metabolic levels in follicular uid is important for exploring the metabolic cooperativity between oocytes and cumulus granulosa cells. In recent years, metabolomics, as a new approach, has been used to explore various molecular mechanisms involved in tissues and organs. Among the various metabolomics techniques, LC-MS/MS is widely used as a conventional detection method for analyzing small molecules, which could provide a quantitative evaluation with minimal handling [27].
In this study, we identi ed the neurotransmitters in follicular uid from IVF patients by high sensitivity, convenient and fast LC-MS/MS technology and evaluated the impact of neurotransmitter Gln and NE in follicular uid on the function and antioxidative capacity of cumulus granulosa cells, which could impact the clinical outcome of IVF patients. Additional in vitro cell experiments were performed to explore the effects of Gln and NE on the proliferation and steroidogenesis of granulosa cells.

Study design
This is an observational study of infertile women before and during IVF treatment at the International Peace Maternity and Child Health Hospital (IPMCH), Shanghai, China. The Institutional Ethics Committee of the IPMCH approved the acquisition protocol. Follicular uid and cumulus granulosa cells on the day of oocyte retrieval were obtained from 27 patients undergoing IVF at the reproductive center. The subjects were recruited with male factor infertility or tubal factor infertility. Subjects with endometriosis and PCOS were excluded. The age of the patients ranged from 25 to 40 years, and the size of the follicles ranged from 19 to 24 mm.

Collection of GCs and follicular uid
The size of the follicle was estimated by ultrasound at the time of oocyte retrieval. Follicular aspirates contain mural GCs and oocytes surrounded by cumulus GCs. After the oocytes were removed by the embryologist, the remaining material was centrifuged at 3000 g for 15 mins to isolate the GCs. The uid from the rst aspirated follicle was used to measure its hormone and neurotransmitter metabolic levels.
Quanti cation of neurotransmitters in follicular uid by liquid chromatography-tandem mass spectrometry (LC-MS/MS) Pretreated human follicular uid was prepared for targeted metabolomics experiments. For this purpose, 12-point standard calibration curves in the respective range with reference material were measured, and the limit of detection (LOD) and limit of quanti cation (LOQ) for each neurotransmitter under investigation were calculated. LC-MS/MS was conducted by an ultra-performance liquid chromatography method using a Waters ACQUITY UPLC BEH C18 column (2.1 mm×100 mm, 1.7 µm, Waters, USA). The Waters acquisition UPLC was coupled to an API 4000 triple quadrupole mass spectrometer (AB SCIEX, USA). The mobile phase A buffer was 10% methanol in water (containing 0.1% formic acid), and the mobile phase B buffer was 50% methanol in water (containing 0.1% formic acid). The LC gradient elution was at a ow rate of 0.3 mL/min at 0-8.0 min and 0.4 ml/min at 8.0-17.5 min, then 10%-30% B at 0-6.5 min; 30%-100% B at 6.5-7 min; 100% B at 7-14 min; and 100%-10% B at 14-17.5 min. The injection volume

RNA Extraction and qRT-PCR
Granulosa cells were collected in TRIzol reagent (Invitrogen, Carlsbad, CA, USA), and total RNA was extracted according to the manufacturer's instructions. A total of 1 mg RNA was converted to cDNA with a Takara kit (Applied Biosystems Foster City, CA, USA; Takara, Shiga, Japan). The genes of interest were ampli ed with a 7900HT fast real-time PCR system (Applied Biosystems) and SYBR Green Real-time PCR Master Mix (Applied Biosystems/Takara). The PCR primers were designed according to the cDNA sequences in the NCBI database. The primer sequences used are shown in Table 1. The cycling conditions used for the PCR analysis were as follows: 95 °C for 5 s, 60 °C for 30 s, and 72 °C for 30 s (40 cycles). 18S was used as an internal control. The 2 -DDCt method was employed to determine the relative mRNA expression level.

Hormone metabolic levels in the follicular uid
The concentrations of twenty kinds of hormones in the follicular uid were detected by high-performance liquid chromatography-mass spectrometry (HPLC-MS/MS). An Agilent 1200 series high-performance liquid chromatography (HPLC) instrument (Agilent, USA) was utilized. A PAL autosampler (CTC, Swiss) and a Gemini-NX-C18 column (2.0 mm×50 mm, 3 μm, Waters, USA) were used. The ion source was an API-4000 quadrupole electrostatic eld orbit trap high-resolution mass spectrometer (Applied Biosystems, USA). The scanning mode was multiple reaction monitoring (MRM) (Agilent-1200 LC system coupled to an API400 mass spectrometer).
Testosterone concentration in the follicular uid Follicular uid samples were drawn into prechilled tubes containing sodium heparin (143 USP/10 ml). The samples were vortexed and centrifuged immediately at 4 °C for 20 min at 3000 g and then stored at -80 °C until they were assayed. The chemiluminescence method was used to detect the level of testosterone in the follicular uid.

KGN cell culture and treatment
The human ovarian granulosa cell line KGN was kindly gifted by Dr. Zuwei Yang (IPMCH, School of Medicine, Shanghai Jiao Tong University, Shanghai, China). KGN was initially cultured in Dulbecco minimal essential medium (DMEM)-F12 medium (Gibco, Grand Island, NY, USA) supplemented with 10% fetal calf serum and antibiotics (100 IU/mL penicillin, 100 μg/ml streptomycin) in a 5% CO 2 atmosphere at 37 °C. After deprivation of glutamine for 24 h, the cells were treated with the corresponding conditioned medium for 48 h.

Enzyme-linked immunosorbent assay (ELISA)
Cell culture supernatant samples were collected and centrifuged for 20 min at 2500 rpm to acquire the supernatant at -4 °C. The supernatant was stored at -80 °C until further use.

Western blotting
Proteins were extracted from cultured cells with RIPA lysis buffer (Yeasen, Shanghai, China) supplemented with protease inhibitor cocktail (Yeasen, Shanghai, China) and phosphatase inhibitor (Yeasen, Shanghai, China) on ice. Cell lysates were cleared with centrifugation at 13000 rpm for 10 min. The protein concentration was measured via a BCA kit (Thermo Scienti c, Rockford, IL, USA). The protein was separated on a 10% SDS polyacrylamide gel (Yamei, Shanghai, China), transferred to PVDF membranes (Millipore, Billerica, MA, USA), and then blocked in a blocking solution at room temperature for 1 h. The following primary antibodies were used: rabbit antibodies against β-tubulin (1:1000, CST, MA, USA) and anti-PCNA antibody (1:1000, CST, MA, USA). After incubation with primary antibodies at 4 °C overnight, the membranes were incubated with horseradish peroxidase-conjugated secondary antibodies (1:2500, CST, MA, USA) for 1 hour at room temperature. The bands were detected by electrochemiluminescence (ECL, Millipore, Billerica, MA, USA). The relative intensity of the target proteins was normalized to β-tubulin using ImageJ software (NIH, Bethesda, MD, USA).

EdU Staining
EdU (RiboBio, Guangdong, China) was added to the KGN cell culture medium at a concentration of 10 µM. KGN cells were cultured for 24 hours and stained with Alexa 594 and DAPI. The detailed and complete steps were performed as recommended by the manufacturer of the EdU detection kits (RiboBio, Guangzhou, China) and observed under a microscope (Leica, WetzlaR, Germany). The number of EdU-positive cells per 100 cells was evaluated.

Statistical analysis
All statistical analyses were performed using Prism software, and a p value of <0.05 was considered statistically signi cant. Means and standard error were used as descriptive statistics. T-tests were used to compare different variables between the low-level and high-level glutamine groups. Linear correlation analysis was used to compare the expression of the IDH1 gene and steroidogenesis in luteinized granulosa cells. In the current study, when a power analysis was performed with 80% power and an a value of 0.05, the number of samples in each group needed to be 13 to con rm statistical signi cance.

Results
Glutamine is positively correlated with norepinephrine content in the follicular uid of IVF patients Dynamic metabolic changes within the follicular uid are important for cumulus expansion, which represents the consequence of oocyte cumulus complex (OCC) metabolic cooperation. In this study, we rst measured the content of neurotransmitters in follicular uid by LC-MS/MS. The results regarding the concentrations of Ach, Gln, Glu, His, NE, Tyr, Trp, Kyn in human follicular uid are shown in Table 2.
The other fteen neurotransmitters were not detected in most of the follicular uid samples. Correlation analysis showed that Gln was positively correlated with the NE content in follicular uid (R=0.5869, P<0.005, Figure 1A). However, there was no relationship between the contents of Gln and Glu in follicular uid (R=0.2374, P>0.05, Figure 1B).
Gln serves as a substrate for the synthesis of proteins in follicular uid, which directly promotes oocyte nuclear maturation and enhances embryo development. However, the effect of Gln and NE in follicular uid on the function of cumulus GCs is still unclear. To further investigate the function of Gln and NE in follicular uid, 27 samples were divided into two groups according to their concentration of Gln, the high Gln level group (>45mg/mL) and low Gln level group (<45mg/mL).
Effect of intra-follicular glutamine/norepinephrine on the function of granulosa cells The function of GCs involved in a proliferative capacity, receptor expression and steroid hormone synthesis in cumulus GCs was determined by qRT-PCR. Compared with the low Gln level group, those with high Gln level had a signi cant 2-fold increase in proliferating cell nuclear antigen (PCNA) and cyclin D2 (CCND2) mRNA levels in the GCs (P<0.001 and P<0.01, Figure 2A-B). However, there were no statistically signi cant differences in luteinizing hormone receptor (LHR) or follicle stimulating hormone receptor (FSHR) mRNA levels between the two groups (P>0.05, Figure 2C-D). Notably, the mRNA levels of the steroid hormone synthesis-related genes steroid acute regulatory protein (STAR), cytochrome P450 (CPY11A, CYP19A) and hydroxysteroid dehydrogenase-3beta (HSD-3b) signi cantly decreased in the low Gln level group (P<0.05, Figure 2E-H). However, there was no signi cant difference in hydroxysteroid dehydrogenase-17beta (HSD-17b) mRNA levels between the two groups (P>0.05, Figure 2I). These data demonstrated that the increased Gln and NE in the follicular uid could promote the proliferation and steroidogenesis of GCs.
Glutamine/norepinephrine promote estrogen synthesis in the follicular uid of IVF patients To obtain an overview of hormone metabolic characteristics in follicular uid between high and low Gln level groups, targeted metabolic pro ling was conducted using HPLC-MS/MS. Nine kinds of hormones were detected in the two groups, including estrone, estriol, 17b-estradiol, progesterone, methyl testosterone, (D)-norgestrel, 19-nortestosterone, epitestosterone and medroxy progesterone. In addition, the concentration of testosterone in follicular uid was measured by a chemiluminescence immunoassay. The results demonstrated that the concentrations of estrone and 17b-Estradiol were signi cantly higher in the high Gln level group than in the low Gln level group (P<0.05 and P<0.001, Figure  3A-B). There were no statistically signi cant differences in the other hormone contents between the two groups. These results revealed that Gln and NE in follicular uid played important roles in promoting estrogen synthesis and the production.
Glutamine/norepinephrine in the follicular uid increase the expression of antioxidant genes in granulosa cells Oxidative metabolism is the major oxygen consumer in the follicle, which could be affected by amino acid metabolism. Antioxidants are indicated as factors that can maintain the balance between ROS reproduction and clearance. Then, we detected the expression levels of the antioxidant genes in GCs via qRT-PCR between the two groups. In the group of high Gln level, antioxidant gene IDH1 and glutathione peroxidase 1 (GPX1) increased signi cantly in GCs compared with that in group of low Gln level (P<0.001 and P<0.0001, Figure 4A-B). Moreover, the expression of nuclear factor erythroid 2-related factor 2 (NRF2) and its endogenous inhibitor, kelch-like epichlorohydrin-associated protein-1 (KEAP1), signi cantly increased in the high Gln level group (P<0.01, Figure 4C-D). Notably, the expression level of IDH1 mRNA in GCs showed a positive correlation with the Gln concentration in follicular uid (R=0.3822, P<0.05, Figure  4E). Furthermore, there was a positive correlation between the NE concentration in follicular uid and IDH1 expression in GCs (R=0.4009, P<0.05, Figure 4F). To further clarify whether this difference in the expression of antioxidative genes was caused by individual differences, we summarized the clinical data of the IVF patients as Table 3 and found that there were no statistically signi cant differences in age (P=0.1601), body mass index (BMI, P=0.5927), serum Anti-Mullerian Hormone (AMH, P=0.6033) or thyrotropin (TSH, P=0.8698) between this two groups. These results demonstrated that the contents of Gln and NE in follicular uid might affect the antioxidant capacity of GCs.

IDH1 expression in granulosa cells is related with the quality of embryo development of IVF patients.
To determine whether follicular development and oocyte growth were different between the high Gln level and low Gln level groups, we compared several clinical indices in the two groups, including the number of mature follicles (diameters ≥14 mm) on the human chorionic gonadotrophin (HCG) injection day, the number of retrieved oocytes, the fertilization rate, and the number and percentage of higher-quality embryos. There were no signi cant differences between the two groups in the above parameters (P>0.05, Figure 5A). Pearson correlation analysis was used to identify the relationship between the Gln/NE concentration in the follicular uid and the ratio of higher-quality/cleaved embryos, and IDH1 expression in granulosa cells and the ratio of higher-quality/cleaved embryos. Results showed that there were no signi cant correlation between the Gln/NE concentration in the follicular uid and the ratio of higherquality/cleaved embryos (R=0.1158 and R=0.0134, P>0.05, Figure 5B-C). However, a signi cantly positive correlation was observed between IDH1 gene expression in GCs and the ratio of higher-quality/cleaved embryos (R=0.4766, P<0.05, Figure 5D). These results demonstrated that the IDH1 expression in cumulus GCs is directly related with the quality of oocyte and embryo development in IVF patients.

Glutamine/norepinephrine improve the function of granulosa cells in vitro
Previous studies have reported that Gln is required for oocyte maturation and embryo development [28] and adrenergic activity through NE participates in the control of follicular development and steroidal secretion [29]. To study the roles of Gln and NE in granulosa cells, the proliferation of cultured KGN cells in vitro was evaluated via EdU staining and western blotting. Fluorescent staining showed that the percentage of EdU-positive cells was dramatically reduced in the Gln deprivation group. The addition of NE partially increased the ratio of EdU-positive KGN cells. Importantly, Gln supplementation increased signi cantly the EdU-positive number (P<0.0001, Figure 6A-B). Moreover, the concentrations of E2 and Prog in the cell culture supernatant were measured by ELISA. The results showed that the concentration of E2 was reduced in the Gln deprivation group, and adding Gln to the medium signi cantly increased E2 production in KGN cells, which could be further improved by NE supplementation (P<0.01, Figure 6C). Although Gln deprivation reduced the level of Prog in the cell supernatant, Gln and NE supplementation had no obvious effect on Prog production in KGN cells ((P<0.05, Figure 6D). The expression of PCNA protein in KGN cells exhibited similar trends to the morphological results in the different treatment groups. Notably, Gln deprivation obviously reduced IDH1 protein expression in GCs, which could be recovered by NE and Gln supplementation (P<0.0001, P<0.01 and P<0.05, Figure 6F-G). These results revealed that the neurotransmitters Gln and NE could synergistically enhance the cell proliferation of granulosa cells and increase the E2 production.

Discussion
Follicular uid is an important external environment for the oocyte growth and follicular development. Ideally, follicular uid, a super uous, abundant and easily available biological sample in IVF cycles, would be the optimum source of noninvasive prediction of oocyte quality [30]. Currently, investigations of follicular uid biomarkers for oocyte quality mainly focus on measuring steroid hormones secreted by granulosa cells, such as estradiol and progesterone. Some studies have demonstrated that high levels of Prog and E2 in follicular uid are positively related to the quality of oocyte and could be indicators for assessing oocyte mature and development. However, other studies have reported no correlation between hormones in the follicular uid and oocyte quality or that they may even be correlated with a lower quality [31]. Steroid hormone production mainly relies on the function of granulosa cells, which are dynamically regulated by small molecule metabolites in follicular uid. Therefore, exploring the relationship between metabolites and granulosa cell function may be conducive to search reliable biomarkers for assessing oocyte quality and predicting the outcome of IVF patients.
A metabolic approach is a powerful tool to analyze the complexities of follicular uid. Previous study reported the application of gas chromatography-mass spectrometry (GC-MS) to detect polybrominated diphenyl ethers (PBDEs) in follicular uid and found that the presence of PBDEs could be associated with failed embryo implantation [32]. In the current study, a more simple, fast and e cient LC-MS/MS technology was successfully applied for determination of simultaneous neurotransmitter in human follicular uid. Consistent with other studies, we found that many different neurotransmitter metabolites exist in human follicular uid, including Ach, Gln, Glu, His, NE, Tyr, Trp and Kyn, as shown in Table 2. Among them, high levels of Gln and NE have attracted our attention; there was a positive relationship between the contents of Gln and NE in follicular uid. NE is released from ovarian nerve terminals wherein the ovarian compartment is directly innervated by sympathetic nerves [29]. Whether ovarian sympathetic nerves control the concentration of Gln in follicular uid needs to be further studied.
Currently, extensive scienti c attention has been given to an important metabolite Gln, which is one of the most abundant amino acid in the human body, such as skeletal muscles, plasma [33] and follicular uid [34]. It has anti-in ammatory and antioxidant functions as well as effects that modulate the heatshock protein response during stress; moreover, as a precursor of glutamate, Gln participates in multiple neurotransmitter metabolic pathways [33,35,36]. In addition, peripheral neurotransmitter NE participates in the control of steroidal secretion from the ovary and follicular development [29]. To study the effect of intra-follicular Gln and NE on cumulus granulosa cell function, we compared the expression of proliferation-and steroid hormone synthesis-related genes between the low Gln level and high Gln level groups. The results showed that Gln and NE could promote granulosa cell proliferation and steroid hormone synthase gene expression. To con rm this relationship, hormone synthetic metabolites were also quantitatively analyzed by LC-MS/MS technology. Among the many hormones, only the concentrations of estriol and 17 b-estradiol were signi cantly decreased in the low Gln level group. These ndings indicated that the content of Gln and NE in follicular uid plays important roles in maintaining the proliferation of cumulus GCs and E2 production.
An altered follicular microenvironment may directly damage follicle recruitment and growth. Previous studies show that metabolites in the follicular uid are direct or indirect regulators of oxidative stress and the function of granulosa cells [37]. A research found that Gln was used as an additive substance to enhance antioxidant capacity in the breast muscle of heat-stressed broilers [38]. Moreover, Gln improves the body's resistance to stress [39] and restores the ovarian function of PCOS rats [40]. In addition, Gln, as the main energy supply substance for mitochondria to form ATP, is an important source of energy for oocyte maturation and the development of embryo [34]. Therefore, Gln has become an essential component in embryo culture medium. A study reported that human ovarian tissue was cryopreserved using freezing medium supplemented with cryoprotectants and antioxidants (L-glutamine and taurine) and exhibited good preservation of ovarian vascular integrity and functionality of granulosa cells post thawing [41]. Furthermore, the glutamine supplementation could effectively alleviate in ammatory reaction and oxidative stress in PCOS rats [40] and protect Chinese hamsters from radiation damage [42]. Thus, the relationships among the levels of Gln in follicular uid, cumulus GC function, and oocyte development are worthy of further study.
Excessive ROS damages DNA, reduces cell proliferation, and induces mitochondrial-mediated apoptosis in GCs, then affecting oocyte development [43,44]. Studies have demonstrated that there is a correlation between the apoptosis of GCs and poor oocyte quality and IVF outcomes in patients with PCOS [11,45].
Many environmental stressors may cause inferior oocyte competence and alter oocyte development and growth by affecting GC function [44,46,47]. The bidirectional interplay between cells and microenvironment affects the expression of genes and ultimately alter cellular behaviors by dynamic reciprocity [48]. In the current study, we found that Gln and NE not only upregulated IDH1 gene expression but also increased GPX1 expression in cumulus GCs. Notably, the concentration of Gln and NE in the follicular uid were positively related to IDH1 gene expression in cumulus GCs, which could be a predictor of high-quality embryos. To date, a clear correspondence between speci c metabolites in follicular uid and measurable embryo-related variables has not been established. Taking the level of Gln and NE in follicular uid and antioxidant gene expression in cumulus granulosa cells into account will provide more accurate information to help embryologists choose optimal quality oocytes.
There are several limitations in current study. First, the number of clinical samples was insu cient. Although the number of samples in the two groups met the statistical requirements, the limited number of samples in uenced the subsequent analysis of reproductive outcomes of IVF patients. The second limitation was that only antioxidant genes expressions were assayed in cumulus GCs, however, other oxidative stress makers including ROS, MDA, GSH did not detected due to lack of enough samples. Finally, only KGN cells were used to con rm the roles of Gln and NE in maintaining granulosa cell function in vitro experiments. Whether appropriate supplementation of Gln and/or NE in animal models with ovarian dysfunction could improve fertility outcomes needs to be explored in future studies.

Conclusions
In summary, our research demonstrates that the concentrations of Gln and NE in follicular uid in uences proliferation and steroidogenesis in human cumulus GCs. Moreover, intrafollicular Gln and NE could upregulate the expression of the antioxidative gene IDH1 in cumulus GCs, which could be a good predictor for evaluating the quality of embryonic development. Declarations Ethics approval and consent to participate Informed consent was obtained from all the couples. In line with the Chinese legislation on clinical studies, the study protocol was approved by our hospital committee with competency for research.

Consent for publication
All authors have agreed to publish this article.

Availability of data and materials
All the raw datasets generated during the current study are available from the corresponding author on reasonable request. All datasets analysed during this study are included in this published article as