Introduction
The culture of ovarian follicles is a powerful approach for investigating folliculogenesis and oogenesis in a tightly controlled environment. It has also important clinical implications for the preservation of reproductive potential in women scheduled for chemotherapy or radiotherapy because of malignant diseases. Moreover, it can be used as an in vitro toxicity assay for drugs and chemicals. Therefore, in vitro culture of ovarian follicles has received increasing attention in recent years.
Folliculogenesis is a complex process that requires integration of autocrine, paracrine and endocrine factors together with tightly regulated interactions between granulosa cells and oocytes. To achieve a developmentally competent oocyte, it is very important that the cytoplasm of the oocyte matures at the same time as meiosis resumes. This includes the accumulation of proteins and energy substrates, organelle reallocation and changes in the structure of chromatin, etc. But in ex vivo culture of ovarian follicles, the oocytes often resumed meiosis precociously before cytoplasmic maturation. As a result, the developmental competence of oocyte, e.g. the ability to be fertilized and the ability to develop to blastocysts, will be negatively affected (Appeltant et al., Reference Appeltant, Somfai, Maes, VAN Soom and Kikuchi2016; Zhang T et al., Reference Li, Cao, Li, Li, Guo and Wang2017a).
It has been well reported that the synchronization of cytoplasmic maturation and resumption of meiosis in oocytes is regulated by CNP. CNP is encoded by the natriuretic peptide precursor C (Nppc) gene expressed mainly in mural granulosa cells. It stimulates natriuretic peptide receptor B (NPRB) on the membranes of cumulus cells to produce cyclic guanosine monophosphate (cGMP). cGMP of cumulus origin diffuses into oocytes to suppress phosphodiesterase 3 (PDE3) activity, leading to elevation of cyclic adenosine 3′,5′-monophosphate (cAMP) in oocytes. cAMP binds to protein kinase A (PKA), which in turn activates WEE 1 homolog 2 (WEE1B) and myelin transcription factor 1 (MYT1) kinase and inhibits cell division cycle 25 (CDC25) and cyclin-dependent kinase 1 (CDK1). Therefore, cAMP-dependent activation of PKA results in CDK1 inhibition leading to meiotic arrest in oocytes (Hsueh et al., Reference Hsueh, Kawamura, Cheng and Fauser2015; Yang et al., Reference Yang, Wei, Li, Ge, Zhao and Ma2016; Machaty et al., Reference Machaty, Miller and Zhang2017). Based on this knowledge, CNP has been supplemented in the culture system of ovarian follicles in mouse (Zhang MJ et al., Reference Zhang, Su, Sugiura, Xia and Eppig2010; Wei et al., Reference Wei, Zhou, Yuan, Miao, Zhao and Ma2017), porcine (Hiradate et al., Reference Hiradate, Hoshino, Taemura and Sato2013), bovine (Franciosi et al., Reference Franciosi, Coticchio, Lodde, Tessaro, Modina, Fadini, Dal Canto, Renzini, Albertini and Luciano2014), cat (Zhong YG et al., 2015), goat (Zhang JH et al., Reference Zhang, Wei, Cai, Zhao and Ma2015) and sheep (Zhong T et al., Reference Machaty, Miller and Zhang2018), etc.
In an earlier study, we supplemented CNP in combination with FSH during the whole period of in vitro growth (IVG) of murine preantral follicles. Using this strategy, oocyte meiotic arrest was efficiently maintained. In addition, follicle development and fertilization competency of oocyte were greatly improved (Li et al., Reference Li, Cao, Li, Li, Guo and Wang2021). To investigate the mechanism of the effect of CNP supplementation on ovarian follicle development in in vitro culture, cGMP concentrations in media were assessed in this study.
Materials and methods
Chemicals
The purchasing information for chemicals are as follows: DMEM/F12 from HyClone Company (USA); recombinant FSH from Merck Serono Company (Italy); human chorionic gonadotropin (HCG) from Sansheng Pharmaceutical Company (China); insulin from Wanbang Pharmaceutical Company (China); transferrin, epidermal growth factor (EGF) and sodium selenium from Sigma Company (USA); CNP from TOCRIS Company (Britain); serum protein substitute (SPS) from SAGE Company (USA); and human tubal fluid from Merck Company (USA). cGMP ELISA test kit from Cayman Chemical (USA).
Animals and preantral follicle culture
Female Kunming mice aged 3 weeks were obtained from the Laboratory Animal Center of Shanxi Medical University. Preantral follicles with multilayer granulosa cells (130–180 µm in diameter) were isolated from ovaries with a tuberculin syringe needle. Each follicle was cultured in a 30-µl of droplet under 37°C in a humidified atmosphere of 5% CO2 in air. The basic culture medium was DMEM/F12 and was supplemented with 5 µg/ml insulin, 5 µg/ml transferrin, 5 ng/ml sodium selenium, 10% SPS, and 25 mIU/ml FSH. CNP was added at the concentration of 100 nM/ml in the experimental group and was absent in the control group. The medium was changed for a half volume every other day. Follicle development was assessed on the day of medium exchange.
For follicle ovulation, 1.5 IU/ml HCG and 5 µg/ml EGF were added into the medium and incubated for 17 h to induce ovulation when the follicle size reached 350–400 µm and part of the follicle bulged out, which indicated that the follicle was ready to ovulate.
For oocyte maturity assessment, the granulosa cells in cumulus–oocyte complexes (COC) ovulated from antral follicles were stripped off by incubating the COC with 80 IU/ml hyaluronidase for 30 s and repeated aspiration with a capillary glass pipette. The preovulatory follicles not ovulated were pierced with a tuberculin syringe needle to retrieve the COC. The denuded oocytes were classified into stages of germinal vesicle (GV), germinal vesicle breakdown (GVBD) and metaphase II (MII).
cGMP assay
cGMP concentrations in the samples were tested with a cGMP ELISA kit (Cayman Chemical, USA) in accordance with the manufacturer’s instructions. Wells in plates were coated with mouse monoclonal anti-rabbit antibody. Next 100 µl of ELISA buffer was added to non-specific binding (NSB) wells and 50 µl of ELISA buffer to maximum binding (B0) wells. In total, 5 µl of medium was sampled from the culture droplet on days 2, 4 and 6 of culture (the day of follicle isolation and start of follicle culture as day 0), respectively, and diluted into 45 µl of ELISA buffer as sample according to the preliminary assay’s result. Here, 50 µl of sequentially diluted cGMP ELISA standard was added to the standard 1 to standard 8 (S1–S8) wells in duplicate and 50 µl of sample to sample wells in triplicate. Next, 50 µl of cGMP Ache tracer was added to each well except the total activity (TA) and blank wells; 50 µl of rabbit antiserum was added into the B0 wells and standard/sample wells. The plate was covered with plastic film and incubated for 18 h at room temperature. Then the wells were emptied and rinsed with wash buffer. Next, 200 µl of Ellman’s Reagent was added to each well; 5 µl of tracer was added to TA wells for ruling out organic impurities in the buffer or other technique problems. The plate was covered with plastic film and incubated for 90 min. Finally, the plate was read at wavelength 412 nm. For calculation of readings, the absorbance reading from NSB wells and B0 wells were averaged, respectively. Corrected maximum binding was calculated as B0 average subtracted by NSB average. Corrected absorbance of standard and sample in each well was calculated as the corresponding reading subtracted by NSB average, respectively. %B/B0 was calculated as rate of the corrected absorbance of a particular sample or standard well to that of the corrected maximum binding. %B/B0 for S1–S8 versus cGMP concentration using linear (Y) and log (X) axes was plotted and a 4-parameter logistic fit was performed. %B/B0 for each sample was calculated and the concentration of each sample was determined using the equation obtained from the standard curve plot. The values with %B/B0 within 20–80% were adopted.
Statistical analysis
Statistical analysis was performed using the software GraphPad Prism 6.0 (La Jolla, CA, USA). Data were analysed using chi-squared test for the development of ovarian follicles and with a t-test for cGMP concentrations. A level of P < 0.05 was considered significant.
Results
Most dominant follicles reached the criteria for ovulation induction on day 6. The follicles not reaching antral stage on day 6 of culture have the least potential to develop into antral follicles. So, the last sampling of medium for cGMP assay was set on day 6 of in vitro culture.
In the experimental group, the follicles grew in three dimensions, no granulosa cells migrated from within follicles. In the control group, the follicles flattened, granulosa cells migrated from within follicles to a different degree.
The maturity of oocytes after ovulation induction is summarized in Table 1. In 53 oocytes cultured in the control group, 29 were at the germinal vesicle breakdown (GVBD) stage and 24 reached MII. In contrast, of 61 oocytes cultured in the experimental group, 22 were in the GVBD stage and 39 reached MII. This demonstrated that oocyte maturity in the experimental group was superior to that in the control group (P < 0.05).
Table 1. The effect of CNP on oocyte maturity
P = 0.0458.
The concentration of cGMP is summarized in Table 2. Values between the experimental group and the control group on the same day were not significantly different. In comparisons between different batches (D2 vs D4, D4 vs D6, D2 vs D6) within each group, although the cGMP concentration increased successively from D2 onwards, only the pair between D2 and D6 demonstrated significant difference (P < 0.05).
Table 2. The concentration of cGMP (pM/ml) in groups
P-value was 0.0292 and 0.0367 in the comparison between D2 and D6 in the groups without CNP and with CNP, respectively. All the other comparisons within group and between groups showed no significant difference (P > 0.05).
Discussion
The importance of cytoplasmic maturation was confirmed by a test reported by Cheng et al. (Reference Cheng, Fan, Wen, Tong, Zhu, Lei, Sun and Chen2003). When the cytoplast at GV stage was used as the recipient and the karyoplast at metaphase in meiosis I (MI) or MII stage as donor, the constructed oocyte extruded a polar body after electrofusion and culture. Both the cytoplasm and the polar body had a metaphase spindle in the MI–GV pair, whereas only a clutch of condensed chromatin was observed in the cytoplasm and polar body of the MII–GV pair. When MI cytoplast was used as the recipient and GV or MII karyoplast as donor, the reconstructed oocyte also extruded a polar body. Each had one spindle and a group of metaphase chromosomes in the cytoplasm and polar body, respectively. When the MII cytoplast was used as recipient and GV or MI karyoplast as donor, the reconstructed oocytes were activated, became parthenogenetic embryos and even developed to hatching blastocysts after electrofusion. Immunoblotting showed that mitogen-activated protein kinase activity was high in MI and MII cytoplasts but not detected in the GV cytoplast. The results demonstrated that the cytoplasmic environment determines the behaviour of asynchronous donors. Therefore, it was accepted that to improve the developmental competency of ovarian follicle in in vitro culture, attention should be paid to prevent oocytes from undergoing precocious meiosis resumption.
Initially, ovarian follicles were cultured two-dimensionally. It was a common phenomenon that granulosa cells migrated out from within the follicle and spread on the dish surface. This resulted in loss of cell–cell communication between oocyte and granulosa cells leading to arrested follicle growth, ovulation suppression and impaired oocyte meiotic competence (Green and Shikanov, Reference Green and Shikanov2016; West et al., Reference West, Shea and Woodruff2007; Eppig and O’Brien, Reference Eppig and O’Brien1996). So, it was very important to maintain the three-dimensional architecture of follicles during in vitro culture of the ovarian follicle.
To do this, V-shaped microwell plates (Telfer and Zelinski, Reference Telfer and Zelinski2013), inverted hanging droplets (Wycheley et al., Reference Wycheley, Downey, Kane and Hynes2004; Choi et al., Reference Choi, Agarwal and He2013), alginate encapsulation (Kreegar et al., Reference Kreegar, Deck, Woodruff and Shea2006; Mainigi et al., Reference Mainigi, Ord and Schultz2011; Vanacker and Amorim, Reference Vanacker and Amorim2017), collagen encapsulation (Mochida et al., Reference Mochida, Akatani-Hasegawa, Saka, Ogino, Hosoda, Wada, Sawai and Shibahara2013), fibrin-alginate interpenetrating encapsulation (Shikanov et al., Reference Shikanov, Xu, Woodruff and Shea2009; Jin et al., Reference Jin, Lei, Shikanov, Shea and Woodruff2010) and encapsulation with alginate plus amino acids (Brito et al., Reference Brito, Lima, Xu, Shea, Woodruff and Figuerrado2014) have been used. Among the above-mentioned methods, encapsulation with alginate was applied most widely. It has been reported that compared with oocytes grown in vivo, spindle formation and chromosome alignment of oocytes with alginate method was abnormal and the developmental competence was compromised (Mainigi et al., Reference Mainigi, Ord and Schultz2011).In addition, the granulosa and theca cells around oocytes would be damaged by enzyme as alginate lyase was required to degrade alginate for releasing follicles (Kim et al., Reference Kim, Lee, Youm, Kim, Lee, Suh and Kim2018), which might affect oocyte development in the later period after encapsulation removal.
In our earlier experiment on ovarian follicle culture, we found that supplementation of CNP during the first 48 h of preantral follicle culture could sustain the three-dimensional structure of ovarian follicles and avoid the ovulation of naked immature oocytes. Then, we supplemented CNP during the whole IVG period of preantral follicles. We found that, in addition to maintaining the three-dimensional architecture, oocyte meiotic arrest was efficiently maintained. As a result, follicle development and fertilization competency of oocytes were greatly improved (Li et al., Reference Li, Cao, Li, Li, Guo and Wang2021).
The present study was designed to investigate how CNP supplementation might improve the developmental competency of ovarian follicle. As demonstrated in Table 1, more oocytes reached MII in experimental group compared with the control group. As demonstrated in Table 2, both experimental group and the control group showed accumulation of cGMP as the culture progressed. cGMP concentration in the experimental group did not increase significantly compared with the control group. The latter is beyond our anticipation that supplemented CNP might stimulate the secretion of cGMP by cumulus cells and the level of cGMP in the medium might increase.
As follicles grew three-dimensionally in the experimental group and flattened in the control group, supplementation of CNP might benefit oocyte development through maintaining the architecture. Between oocyte and cumulus granulosa cells, there are many transzonal projections (TZP). In addition, between granulosa cells there are gap junctions. Gap junctions play essential roles in the communication between granulosa cells. Transport of nutrient and signalling molecules such as cAMP is mediated by TZPs from granulosa cells to the oocyte. Therefore, the communication through TZP and gap junction have fundamental roles in oocyte meiosis control and oocyte maturation (Eppig et al., Reference Eppig, Pendola, Wigglesworth and Pendola2005; Gilchrist et al., Reference Gilchrist, Luciano, Richani, Zeng, Wang, Devos, Sugimura, Smitz, Richard and Thompson2016). If granulosa cells spread onto the dish surface during ovarian follicle culture, the communication between granulosa cells and between granulosa cells and oocyte will be broken. As a result, the development of oocyte will be adversely affected. Romero et al. (Reference Romero, Sanchez, Lolicato, Van Ranst and Smitz2016) reported that CNP-treated COC showed a higher (P < 0.05) density of TZP. Soto-Heras et al. (Reference Soto-Heras, Peramzo and Thompson2019) reported that supplementation of cAMP modulators in the in vitro culture of cattle ovarian follicle resulted in maintenance of the density of TZPs, leading to enhanced oocyte developmental competence. Franciosi et al. (Reference Franciosi, Coticchio, Lodde, Tessaro, Modina, Fadini, Dal Canto, Renzini, Albertini and Luciano2014) reported that the percentage of COCs with functionally open gap junctions 6–8 h after CNP supplementation in vitro was significantly higher than that in the control COCs as measured by Lucifer yellow injection assay. With more TZP and functionally open gap junctions, the communication between the oocyte and surrounding granulosa cells might be improved.
In addition to the mechanism mentioned above, CNP could also stimulate ovarian follicle growth and increase ovarian follicle viability both in vivo and in vitro. In mice, Nppc and Npr2 begin expression in early preantral follicles and increase during early to late preantral follicle development. Treatment of cultured ovarian explants from infantile mice with CNP promoted the development of primary and early secondary follicles to the late secondary stage. In vivo studies indicated that in infantile mice, daily injection of CNP for 4 d promoted ovarian growth and the follicles ovulated after ovulation induction. In prepubertal mice, CNP treatment alone also promoted early antral follicle growth to the preovulatory stage, resulting in efficient ovulation using gonadotropin. Mature oocytes retrieved after CNP treatment could be fertilized in vitro and developed into blastocysts. After embryo transfer, viable offspring were delivered (Sato et al., Reference Sato, Cheng, Kawamura, Takae and Hsueh2012; Xi et al., Reference Xi, Wang, Fazlani, Yao, Yang, Hao, An and Tian2019). It has been reported that CNP increased the expression of paracrine or autocrine factors such as Wingless-type mouse mammary tumour integration site family 2b (Wnt2b), Wnt5a, cytochrome P450 11a1 (Cyp11a1) and repressed the expression of oestrogen metabolic enzymes cytochrome P450 1a1 (Cyp1a1), which contributed to follicle growth or granulosa cell viability (Xi et al., Reference Xi, Wang, Fazlani, Yao, Yang, Hao, An and Tian2019). This beneficial effect of CNP is more prominent in small follicles compared with large follicles in in vitro culture (Zhang YH et al., Reference Zhang, Wang, Liu, Yang, Wang, Zhang, Guo, Wang and Xia2017b). In addition, CNP could stimulate follicle growth and decrease reactive oxygen species (Sato et al., Reference Sato, Cheng, Kawamura, Takae and Hsueh2012; Tiwari M et al., Reference Kreegar, Deck, Woodruff and Shea2015; Xi et al., Reference Xi, Wang, Fazlani, Yao, Yang, Hao, An and Tian2019). In the present study, more MII oocytes developed in the experimental group compared with the control group; this might be partly explained by the mechanism mentioned above in this paragraph.
Several chemicals in addition to CNP such as forskolin, 6-dimethylaminopurine (6-DMAP) have also been reported as substances that have the potency to improve the nuclear–cytoplasmic synchronization of oocytes after IVM. But it has been reported that some of these chemicals may have a prolonged inhibitory effect due to a longer half-life or have detrimental effects by inducing cellular apoptosis (Bouhelan et al., Reference Bouhelan, Bockaert, Mermet-Bouvier, Guillon and Homburger1987; Samake and Smith, Reference Samake and Smith1997; Simli et al., Reference Simli, Pellerano, Pigullo, Tavosanis, Ottaggio, de Saint-Georges and Bonatti1997; Alexander et al., Reference Alexander, Coppola, Di Berardino, Rho, St John, Betts and King2006; Follin-Arbelet et al., Reference Follin-Arbelet, Misund, Naderi, Ugland, Sundan and Blomhoff2015). In contrast, CNP is a natural substance in vivo, and could orchestrate the meiotic progress in cooperation with involved growth factors and hormones. More importantly, the inhibitory effect of CNP-NPR2 signalling is reversible and will be inactivated soon after the luteinizing hormone surge. This makes CNP a promising agent to improve the developmental competency of ovarian follicles cultured in vitro.
Author contributions
Wang Huaixiu and Guo Xingping designed the experiment, Wang Huaixiu wrote the manuscript. Guo Xingping, Li Ang and Cao Haixia performed the experiment. Wang Zhulin carried out the statistical analysis. All authors read and approved the final manuscript.
Competing interests
No competing interest was declared.
Funding
None
Ethics approval
The study protocol was approved by the ethics committee of Shanxi Provincial Hospital.
