Abstract
In this study, we systematically compared the morphological, functional and molecular characteristics of granulosa cells and oocytes obtained by a three-dimensional in vitro model of ovine ovarian follicular growth with those of follicles recovered in vivo. Preantral follicles of 200 µm diameter were recovered and cultured up to 950 µm over a 20-day period. Compared with in vivo follicles, the in vitro culture conditions maintained follicle survival, with no difference in the rate of atresia. However, the in vitro conditions induced a slight decrease in oocyte growth rate, delayed antrum formation and increased granulosa cell proliferation rate, accompanied by an increase and decrease in CCND2 and CDKN1A mRNA expression respectively. These changes were associated with advanced granulosa cell differentiation in early antral follicles larger than 400 µm diameter, regardless of the presence or absence of FSH, as indicated by an increase in estradiol secretion, together with decreased AMH secretion and expression, as well as increased expression of GJA1, CYP19A1, ESR1, ESR2, FSHR, INHA, INHBA, INHBB and FST. There was a decrease in the expression of oocyte-specific molecular markers GJA4, KIT, ZP3, WEE2 and BMP15 in vitro compared to that in vivo. Moreover, a higher percentage of the oocytes recovered from cultured follicles 550 to 950 µm in diameter was able to reach the metaphase II meiosis stage. Overall, this in vitro model of ovarian follicle development is characterized by accelerated follicular maturation, associated with improved developmental competence of the oocyte, compared to follicles recovered in vivo.
Introduction
Fundamental studies of ovarian dynamics are crucial for understanding the underlying causes of infertility and to improve assisted reproductive technologies (ART) in humans and domestic animal species in order to overcome infertility problems and optimize fertility preservation by cryopreservation and in vitro follicle development. Ovarian follicle development, and particularly the environment in which development occurs, significantly influences oocyte quality, and consequently, the success of pregnancy. Extensive research on follicle development has been conducted, mainly at the late antral and preovulatory stages, from those recovered in vivo as well as cultured in vitro granulosa cells. However, a better understanding of the early steps of ovarian folliculogenesis, in particular, during the preantral to early antral transition, is still needed (Palma et al. 2012, McGee & Raj 2015). Research on preantral follicles is justified by the fact that they constitute a large proportion of the total follicular population in the ovarian cortex (Gougeon 1996, Scaramuzzi et al. 2011), and they survive cryopreservation in sheep, goats and humans (Cecconi et al. 2004, Figueiredo et al. 2011, Vanacker et al. 2013). Moreover, immediately prior to antrum formation, preantral follicles are the only follicles able to develop properly and individually in vitro (in contrast to isolated primordial and primary follicles) until the antral stage in domestic animal species and humans (Telfer & Zelinski 2013).
In vitro culture of isolated preantral follicles has been successfully developed in ruminants (Newton et al. 1999, Gutierrez et al. 2000, Silva et al. 2015) and humans (Telfer et al. 2008, Xu et al. 2009) by several research groups. Various in vitro follicle culture methodologies have been developed, including two-dimensional (2D) systems with attached follicles and three-dimensional (3D) systems with or without gel-based encapsulation. The 2D system developed successfully in the mouse (Cortvrindt et al. 1996) does not work for human follicles or those of large domestic animals, mainly due to the longer culture period required, the greater diameter expansion and the disruption of follicle architecture leading to oocyte death. In vitro cultures of individual preantral follicles without attachment, which maintain the 3D morphology of the follicles and support substantial follicle growth and development, were accompanied by a progressive decrease in viability and weakening of the follicular architecture (Newton et al. 1999, Gutierrez et al. 2000). The follicle integrity can be improved by including follicles in a fibrin–alginate–matrigel matrix, which preserves the optimal oocyte–somatic cell connections; however, the compressive force needs to be adapted in response to follicle expansion (Picton et al. 2008, Shea et al. 2014).
Studies that have focused on the in vitro development of preantral follicles in large non-rodent species have evaluated the quality of the follicle and its enclosed oocyte in culture by studying their survival and growth capacity, associated with the secretion of steroids or inhibins (sheep (Cecconi et al. 2004), goat (Chaves et al. 2012, Silva et al. 2015) and human (Xu et al. 2009)) or anti-Müllerian hormone (primate (Xu et al. 2016)) and/or oocyte competence to development (buffalo (Gupta et al. 2008), sheep (Barboni et al. 2011) and goat (Magalhaes et al. 2011)). The transcriptome of small follicles developed in vivo has been established for sheep (Bonnet et al. 2011) and goat (Magalhaes-Padilha et al. 2013). However, gene expression studies for follicles developed in vitro are limited, as only a small number of genes are analyzed at the end of the culture, usually including GDF9/BMP15 and their receptors (sheep (Kona et al. 2016) and goat (Lima et al. 2012)), BCL2/BAX (sheep (Praveen Chakravarthi et al. 2015)), CYP19A1 (sheep (Lakshminarayana et al. 2014)), EGF/EGFR (goat (Silva et al. 2013)), INSR/FSHR (goat (Chaves et al. 2012)) and the activin system genes (goat (da Silva et al. 2015)). Only one study compared the in vivo and in vitro expression of 14 genes from small antral follicles of the rhesus macaque, which was performed using a microarray (Xu et al. 2013). To our knowledge, there have been no systematic studies of granulosa cell proliferation, differentiation and viability, with oocyte growth and maturation, and their association with the expression of molecular markers of somatic follicular cells and oocytes. Moreover, a systematic comparison of these characteristics between follicles developed in vitro and those at the same stage developed in vivo would assist in providing a comprehensive overview of the events associated with in vitro follicle development.
In the present work, we studied the in vitro development of ovine preantral follicles (160–240 µm) with the following aims: (1) to simultaneously establish the kinetics of morphological and cellular changes in cultured preantral follicles and the growth of their enclosed oocytes; (2) to analyze the impact of in vitro follicle culture in terms of hormone secretion and gene expression on the oocyte and follicular cells, and compare these data to those obtained in vivo and (3) to compare the maturation state of the follicular cells and oocytes at each stage of in vitro follicle development.
Materials and methods
Unless otherwise stated, all culture media and chemicals used in the present study were purchased from Sigma-Aldrich Chemical.
Collection of ovaries and isolation and in vitro culture of preantral follicles
Ovaries (n = 20 for each culture experiment) were recovered from prepubertal sheep (6–8 months of age) at a local abattoir, which provided Charolais, Bleu du Maine and Suffolk ewe lambs as main breeds. Ovaries were washed with sterile 0.9% NaCl supplemented with 25 µg/mL gentamicin, and then transported to the laboratory within 2 h of collection in tubes containing the HEPES buffered tissue culture medium 199 supplemented with 0.4 mg/mL bovine serum albumin (BSA) and 25 µg/mL gentamicin (TCM199+). The ovaries were cut into thin slices using a sterile surgical blade. These slices were incubated at 37°C for 1 h in a cell dissociation PBS-based solution containing 0.1% collagenase IA and 0.01% DNase I, and then in a stop solution of PBS containing 2% BSA. After rinsing in warmed TCM199+ medium, follicles were mechanically isolated from the cortical ovary slices by micro-dissection under a stereomicroscope using two 30 gauge needles fitted to 1 mL syringe barrels, after which the follicles were stored in petri dishes containing TCM199+ medium. On day 0, the initial diameter of all follicles was measured on perpendicular axes with a stereomicroscope equipped with a calibrated ocular micrometer. Only preantral follicles between 160 and 240 µm in diameter, with no apparent damage to the basal membrane, no visible signs of degeneration (darkness of the oocyte and follicular cells) and no antral cavity were selected for culture (Fig. 1). For comparison with those developed in vivo, additional follicles of 160–950 µm in diameter were either individually analyzed for cellular characterization or pooled according to their size and frozen at −80°C in 20 µL lysis buffer (from the ARCTURUS PicoPure RNA isolation kit, Kbib202) for gene expression analysis.
The culture medium was prepared the day prior to culture (or medium renewal) with bicarbonate buffered Eagle minimum essential medium (MEM; alpha modification) supplemented with 2 mM glutamine, 2 mM hypoxanthine, 50 µg/mL ascorbic acid and ITS+ pre-mix (6.25 µg/mL insulin, 6.25 µg/mL transferrin, 6.25 ng/mL selenium, 1.25 mg/mL BSA and 5.35 µg/mL linoleic acid; Becton Dickinson, Le Pont de Claix, France). The control medium (αMEM+) was otherwise supplemented with 100 ng/mL of purified ovine FSH (oFSH), obtained from Dr Yves Combarnous (Nouzilly, France; batch no. CY1771-II, oFSH activity was 28-fold higher than the activity of NIH oFSH-S1). We have based our choice of this active but submaximal FSH concentration on previous data obtained from our laboratory concerning cultures of granulosa cells (Pierre et al. 2004) and cultures of ovine ovarian cortex to sustain the development of small follicles (Bertoldo et al. 2016). For each culture, petri dishes containing 10 droplets of 100 µL of culture medium overlaid with mineral oil were pre-equilibrated overnight at 38.5°C in 5% CO2 in air under 95% relative humidity. Isolated follicles were washed twice in TCM199+, measured, and then randomly allocated into two treatment groups with different culture media. Follicles were individually placed into each of the 100 µL droplets and incubated for up to 20 days. This culture period was found to be optimal for follicular growth and antrum formation. Petri dishes were removed from the incubator for morphological evaluation at day 1 (D1), D6, D13 and D20, taking care to manipulate follicles on a heating plate (37°C) for the least amount of time. The morphology of follicles was assessed using three criteria: (1) their ‘integrity’ (no breakdown of the basal lamina and extrusion of the oocyte); (2) their increase in diameter and (3) formation of an antral cavity, defined as a visible, translucent area within the follicular cell mass. Follicles were visualized and photographed using an inverted microscope at 20×, 10× or 4× magnification to assess their growth and survival throughout the culture period. At D6 and D13, half the culture medium in each droplet was replaced with fresh pre-incubated medium, and the medium removed from each droplet was individually frozen at −20°C for AMH and estradiol (E2) determination.
Morphological analysis of follicles developed in vivo and in vitro
Nine independent experiments were conducted for the morphological evaluation of intact follicles (with structural integrity) developed in vivo (n = 111) and those developed in vitro in the control (n = 157) and FSH-treated (n = 188) cultures. After each follicle had been measured, cells were dissociated after brief exposure to an accutase solution at 37°C. Oocytes and follicular cell suspensions were visualized under an inverted microscope at 40× and 10× magnification respectively. The photographs were analyzed to determine the oocyte diameter and cell number.
For each follicle, the follicular diameter, the external and internal oocyte diameters, the number of follicular cells and the presence of an antrum were evaluated. The zona pellucida thickness was calculated as the difference between the external and internal radius of the oocyte. Data for the in vitro developed follicles were first analyzed grouped by the day of culture. Then for in vitro and in vivo comparisons, follicles between 160 and 850 µm were grouped into 12 follicular diameter classes of 50 µm intervals.
The antrum volume VA was calculated by the equation: VA = VF − ((NCVC) + VO), where VF is the follicle volume, VO is the oocyte volume, NCVC is the follicular cell volume, VC is the volume of one follicular cell and NC is the number of follicular cells. The volume of one follicular cell was estimated from follicles at the preantral stage collected in vivo, using the equation: VC = (VF − VO)/NC. Second-order polynomial regressions were fitted to the data sets to describe the increase in antrum volume with follicular size for the different developmental conditions (in vivo and in vitro control or FSH-treated culture).
Follicle quality (healthy or atretic) was assessed by the presence of mitosis and pycnosis in the granulosa cells of intact follicles developed in vivo (n = 29) and in vitro in the control (n = 35) and FSH-treated (n = 30) culture conditions over the 20-day period. The follicles were dissected and classified according to two follicular diameter classes, either 550 to <750 µm or 750–1000 µm. Immediately, each follicle was slit open and a smear of granulosa cells was prepared on histological slides, which were then fixed in a solution containing methanol, formaldehyde and acetic acid (80:15:5 v:v:v), and subsequently stained with Feulgen. The quality of each follicle was assessed by microscopic examination of the smears, using classical histological criteria (Supplementary Fig. 1, see section on supplementary data given at the end of this article) where follicles were judged as either healthy (frequent mitosis and no pycnosis in granulosa cells), slightly atretic (mitosis, some pycnotic bodies in granulosa cells) or atretic (no mitosis, numerous pycnotic bodies in granulosa cells), as previously described (Besnard et al. 1996).
In vitro follicular secretion of anti-Müllerian hormone and estradiol
The secretion of AMH and E2 by the cultured follicles was assessed from four replicate cultures performed in the absence or presence of FSH (100 ng/mL). For hormonal determination, the culture media (50 µL) was recovered at D6, D13 and D20 only from follicles with maintained structural integrity, which had formed an antral cavity and had reached a diameter of at least 550 µm at the end of the culture period (D20). The concentrations of AMH and E2 in the culture medium were determined for at least 20 individual follicles per treatment at each of the selected time points.
The AMH concentration in the culture media was determined using the AMH Gen II ELISA kit (Beckman Coulter, Villepinte, France), which had previously been validated for the analysis of ovine samples (Estienne et al. 2015). The AMH concentrations were determined in the 50 µL aliquots of culture medium, diluted to 1/5 for samples recovered at D6 and 1/10 and 1/100 at D13 and D20 respectively. The limit of detection of the assay for samples in this study was found to be 0.02 ng/mL (1.4 pg/well). The intra- and inter-assay coefficients of variations (s.e.m./mean) were 2.9% and 4% respectively.
The E2 concentrations in the culture media were determined using the E2-EASIA immunoassay kit (DIAsource, Louvain-la-Neuve, Belgium) from the 50 µL aliquots of culture medium, diluted to 1/10 for samples recovered at D6 and 1/50 and 1/200 at D13 and D20 respectively. The limit of detection of the assay was found to be 2.8 pg/mL (0.4 pg/well). The intra-assay and inter-assay coefficients of variation (s.e.m./mean) were 6.2% and 8.2% respectively.
Hormonal secretion was expressed as pg per follicle (in the 100 µL culture drop). As half of the medium had been replaced every 6–7 days of culture, the remaining amount of each hormone detected at the time of previous medium renewal was subtracted from the total amount measured for the studied culture period, and the data were expressed as ‘de novo secretion’. The de novo secretion was further expressed relative to the follicular cell number, estimated from the size of the corresponding follicle. The number of follicular cells had been determined from counting and averaging the number of follicles in each of the 50 µm-interval size classes. The hormone concentration was then expressed as pg per 104 follicular cells.
In vitro maturation of oocytes from follicles developed in vivo and in vitro
Seven independent experiments were conducted for the oocyte maturation evaluation, involving 456 intact follicles developed in vivo (n = 250) and those developed in vitro in the control (n = 101) and FSH-treated (n = 105) culture conditions. The follicles collected in vivo were separated into four groups depending on follicular size: 400 to <550 µm, 550 to <1000 µm, 1–2 mm and >2 mm. For in vitro experiments, at the end of the culture period (D20), follicles with maintained structural integrity were divided into four groups depending on treatment, firstly for whether they were cultured with or without FSH (100 ng/mL), and then for their follicular size (400 to <550 µm or 550 to <1000 µm). After the follicles were opened, the oocyte-cumulus complexes (COCs) were carefully retrieved, washed in TCM199+ and examined under a stereomicroscope. Only oocytes that were completely surrounded by cumulus cells were tested for in vitro maturation (IVM). Groups of COCs (≤10 per droplet) were subjected to IVM evaluation, after they had been placed into 100 µL droplets of TCM199 supplemented with 10 ng/mL EGF and 100 µM cysteamine, overlaid with mineral oil, and then incubated for 24 h at 38.5°C under 5% CO2 in humidified air, as previously described. The positive in vivo controls included COCs collected from follicles of 2–5 mm diameter that had been transported to the laboratory within 2 h in TCM199+ at 35°C and subjected to the contemporary IVM experiment. The negative controls consisted of COCs that had been recovered at D20 from the cultured follicles, or in vivo, which were then directly (without IVM) analyzed for their oocyte nuclear status. Oocytes were denuded, fixed in 4% paraformaldehyde in PBS, incubated with Hoechst 33342 (1 µg/mL) for chromatin labeling, and then placed on slides and mounted with Mowiol. Oocytes were observed with a microscope under ultraviolet illumination and photographed at 40× magnification. The meiotic stages were identified from the intact germinal vesicle stage until metaphase II (MII) (Supplementary Fig. 2).
RNA extraction and cDNA synthesis
A total of 1758 follicles were used for the evaluation of gene expression. Follicles from 80 to 850 µm developed in vivo (n = 1389) were collected from ovarian cortex strips, as described in the isolation procedure. In parallel, follicles developed in vitro under the control conditions (n = 369) were recovered at D6, D13 and D20 from 10 independent culture experiments. Follicles were pooled according to their size, preserved in 20 µL extraction buffer (from ARCTURUS PicoPure RNA isolation kit, Kbib202) and stored at −80°C until use. For preparation of the 88 samples, follicles were allocated to one of five developmental stages (Supplementary Table 1) including small preantral (SPA; 80–160 µm) and medium preantral (MPA; >160 to 240 µm) follicles, corresponding only to follicles developed in vivo; and large preantral (LPA; >240 to 300 µm), early antral (EA; >300 to 550 µm) and antral (A; >550 to 850 µm) follicles, corresponding to follicles developed in vivo and in vitro. Follicles of the same stage were pooled to obtain a minimum of 500 ng of extracted RNA per sample.
Total RNA was extracted using the PicoPure RNA isolation kit, according to the manufacturer’s instructions, including on-column DNase treatment. Total RNA was quantified using a NanoDrop ND-1000 spectrophotometer (Nyxor Biotech, Paris, France). The quality and integrity of the RNA samples were assessed using an Agilent 2100 bioanalyzer (Agilent Technologies) with an RNA 6000 Nano LabChip kit, analyzed using the RNA integrity number (RIN) algorithm. Only samples that showed a RIN greater than or equal to 7 were selected. For each sample, 500 ng of total RNA was reverse transcribed into cDNA using the iScript cDNA synthesis kit (Bio-Rad Laboratories) with a blend of oligo (dT) and random hexamers to provide complete RNA sequence representation. The cDNAs were diluted five-fold in 10 mM Tris–HCl (pH 8.0) and 0.1 mM EDTA (TE), and then stored at −20°C.
Quantitative PCR with the BioMark HD system from Fluidigm
Specific primers were designed for 40 target genes and five reference genes (ACTB, RPL19, SDHA, TMED4 and YWHAZ). Gene names and primer sequences are presented in Supplementary Table 2.
The expression of the 40 genes, known to be involved in preantral to antral follicular transition, in addition to the five control genes, was determined using the 96.96 Dynamic Array integrated fluidic circuits (IFCs) and the BioMark HD system from Fluidigm, as previously described (Bonnet et al. 2013). Briefly, one specific target amplification (STA) with 14 cycles was performed on the 88 cDNA samples, derived from 43 samples of in vivo developed follicles, 45 samples of in vitro follicles (Supplementary Table 1) and two calibrator samples (cDNA derived from a pool of follicles, obtained by mixing equal amounts of RNA representing each of the five follicular stages with a pool of cDNA from each of the 88 follicles-derived cDNA samples). STA was performed with a pre-amplification primer mix that contained primers for the 40 target genes and five reference genes. The resulting STA cDNA samples were treated with exonuclease I, and then diluted fivefold in TE buffer and stored at −20°C. Following priming, each inlet of the dynamic array IFCs was loaded with the 45 primer pre-mix and a control primer pre-mix (RNAseP) in one section, and with the pre-mixed sample and the 88 STA cDNA samples, 1 STA calibrator, 1 internal control (ADNg) and the STA cDNA pool of the 88 cDNA samples in fivefold serial dilutions (1/3) (to determine the PCR amplification efficiency), in the other section. The chip was transferred to the BioMark HD, and qPCR was performed using the previously described thermal protocol (Bonnet et al. 2013) followed by a melting phase. All quality controls were performed consistently with the Genome and Transcriptome platform (Toulouse, France) and the manufacturer’s recommendations. Data were analyzed using Fluidigm Digital PCR Analysis software (version 4) using the linear (derivative) baseline correction method and the auto (global) cycle threshold (Ct) method.
After determination of the Ct for each cDNA sample and PCR efficiency for each gene, the calculated relative expression (R = (ERefCt(Ref)/EtargetCt(target))) was normalized for the geometric average expression of the three most stable internal control genes, chosen from the five control genes (ACTB, RPL19, SDHA, TMED4 and YWHAZ) after they had been tested using the geNorm (version 3.4) algorithm. The three genes with the best qPCR efficiency (>1.9) and average expression stability value (M; M < 0.4) were SDHA (expressed at a low level in follicles), ACTB and RPL19 genes (highly expressed in follicles).
For the 10 oocyte-specific genes (oocyte specificity previously validated by Bonnet and coworkers (Bonnet et al. 2013)), we used the expression ratio of the gene: R = E(−Ct)/N, where N is the number of equivalent oocytes in the sample, to avoid the impact of reference gene expression in the somatic follicular cells.
Statistical analysis
GraphPad Prism 6 (GraphPad Software) was used for statistical analyses, and a probability level of 5% (P < 0.05) indicated significant differences. After the Bartlett test had been performed to confirm the homogeneity of variances, the effect of culture time or follicle growth on the morphological parameters and hormonal secretions were analyzed using one-way ANOVA and repeated measures one-way ANOVA respectively, followed by the Tukey multiple comparisons test. Where the variances were found to be heterogeneous, the Kruskal–Wallis test (or Friedman test for repeated-measures ANOVA) was performed, followed by the Dunn multiple comparisons post hoc test. The effects of different follicular developmental conditions and culture conditions (for hormonal secretions), or follicle diameter (for morphological parameters), were analyzed using two-way ANOVA followed by the Bonferroni multiple comparisons test. Chi-square analysis and Fishers test was used to assess the proportion of follicles that maintained their structural integrity and had formed an antral cavity, those with differences in follicular atresia and differences in the meiotic progression of the oocytes according to their follicular size, origin (in vivo vs in vitro) and treatment (control or FSH condition).
The effect of the different follicle development conditions (in vitro vs in vivo) and stages (preantral to antral stages) on gene expression were studied with a hierarchical analysis method using Cluster 3.0 software, which was applied to 40 genes and 79 biological samples, separated according to their development stage (LPA, EA and A). The data were log2-transformed, centered (mean) and normalized. Hierarchical clustering was performed using the Euclidian algorithm for dissimilarity with average linkage, and the results were visualized using TreeView software.
A one-way ANOVA was performed for the gene expression data, after the Bartlett test had confirmed the homogeneity of variances, followed by the Bonferroni multiple comparisons test. Kruskal–Wallis test was performed with the Dunn multiple comparisons post hoc test when the data were not sampled from Gaussian distributions.
Results
Morphological changes of follicles during culture
To analyze their growth and morphological changes during culture, follicles were classified into five stages according to their size (Fig. 1). The first stage, named small preantral stage (SPA, 80 to <160 µm), only corresponded to follicles obtained in vivo. At the beginning of culture, all the follicles were at the medium preantral stage (MPA, 160–240 µm). During the growth period from day 6 to 13 of culture, they reached the large preantral (LPA, >240 to 300 µm) and early antral (EA, >300 to 550 µm) stages. Thereafter, until the end of the culture period at day 20, follicles reached the antral stage (A, >550 to 850 µm).
Between day 1 and 20 of culture under control conditions, the follicle and oocyte diameters increased by 2.5-fold and 1.3-fold respectively (P < 0.01; Fig. 2A and C) and the number of follicular cells increased by 20.5-fold (P < 0.01; Fig. 2B). The thickness of the oocyte zona pellucida increased progressively until day 13 of culture (P < 0.01) and reached a plateau between day 13 and 20 (Fig. 2D). Similar changes were observed when follicles were cultured in the presence of FSH. Some of the cultured follicles did not survive until the end of the 20-day culture period, mainly due to the rupture of the basal membrane and opening of the follicular structure. Follicular structural integrity was lost in 39% and 27% of the follicles cultured in the absence or presence of FSH respectively, which was not significantly different between the culture conditions (Supplementary Fig. 3). Follicles that maintained their structural integrity were viable, as shown by their labeling with calcein-AM (Supplementary Fig. 4). Subsequent analyses were only performed on follicles that had maintained their structural integrity.
Follicular growth in vivo and in vitro
The morphological features of follicles with maintained structural integrity developed in vitro (n = 340) in the absence (control medium) or presence of FSH, were compared to follicles of the same size collected in vivo (n = 111).
During the culture period, and in parallel with increasing follicle diameter, the oocytes grew significantly (P < 0.05), reaching a mean external diameter of 140 µm in follicles larger than 600 µm after either in vivo or in vitro development. For in vitro development, the change in oocyte size was similar between follicles cultured with or without FSH. However, the diameter of the oocytes recovered from follicles 300 to 600 µm in diameter after culture was lower than that of similarly sized follicles retrieved in vivo (P < 0.05), suggesting that the oocyte growth rate was slower in vitro (Fig. 3A).
During in vivo development of the follicle from 200 to 800 µm in diameter, the number of follicular cells increased by 18-fold, whereas in vitro development was associated with 32-fold and 37-fold increases in cell number in control and FSH-treated culture conditions respectively (P < 0.05) (Fig. 3B). For follicles between 180 and 550 µm in diameter, the number of cells recovered after culture in control or FSH-treated conditions was comparable to those obtained in vivo. However, beyond 550 µm, cell numbers were higher in the cultured follicles (P < 0.01), indicating that the cell proliferation rate was higher in vitro.
The progressive antrum formation was related to the increase in follicular diameter. All follicles larger than 600 µm had a visible antral cavity after in vivo or in vitro development (Fig. 3C). However, the percentage of follicles for which an antral cavity could be detected was lower in the follicles between 300 and 550 µm in diameter developed in vitro compared to those developed in vivo (P < 0.01 and P < 0.05 for in vivo vs in vitro control or FSH-treated culture conditions respectively), with no significant difference observed between the in vitro control and FSH-treated culture conditions. Moreover, from estimation of the antrum volumes, the regression curves established between the follicular diameter and the antrum volume differed between the follicles developed in vivo and in vitro (P < 0.001), indicating that the antrum growth rate was lower in vitro (Fig. 3D). Again, no significant difference was observed for antrum growth between control and FSH-treated in vitro conditions.
Atresia rate in vivo and in vitro
Histological analysis of granulosa cells showed that atresia was present in less than 5–15% of follicles with a diameter between 550 and <750 µm, developed in vivo or in vitro (with and without FSH), and the atresia rate (percentage of atretic follicles) increased to more than 50% in follicles 750–1000 µm in diameter (Supplementary Table 3). For both size classes, there was no difference between the atresia rate of follicles developed in vivo or in vitro, regardless of the culture conditions.
Follicular secretion of anti-Müllerian hormone and estradiol during culture
The functional changes of growing follicles was assessed by studying their secretion of AMH and E2 during the culture period. Only follicles that had maintained their structural integrity, formed an antral cavity and reached a minimal diameter of 550 µm on day 20 of culture were analyzed.
The secretion of AMH by growing follicles in the culture medium increased between day 6 and 20 of culture for both the control (P < 0.001) and FSH-treated (P < 0.001) conditions, with a 3.3-fold and 3.2-fold increase observed respectively (Fig. 4A). However, when expressed relative to the number of follicular cells, AMH secretion per 104 cells decreased progressively within the culture period for both the control (3.8-fold decrease, P < 0.001) and FSH-treated (2.2-fold decrease, P < 0.001) conditions (Fig. 4B). In the presence of FSH, a 1.8-fold drop in AMH secretion per 104 follicular cells was observed on D6 (P < 0.001); however, AMH secretion after this point was similar for the control and FSH-treated conditions.
The E2 secretion per follicle increased by 12-fold (P < 0.001) and 14-fold (P < 0.001) between day 6 and 20 of culture in the control and FSH-treated conditions respectively (Fig. 4C). On D13 and D20 of culture, the E2 secretion per follicle was significantly higher (P < 0.05 and P < 0.01 respectively) for the FSH-treated follicles. When expressed based on the number of follicular cells, E2 secretion per 104 cells was decreased slightly between day 13 and 20 of culture in the control culture (P < 0.05), but no significant effect was observed for the FSH-treated conditions. Secretion of E2 per 104 cells was enhanced by FSH at D13 and D20 compared to that in the control condition (both P < 0.05, Fig. 4D).
Oocyte competence to resume meiosis after culture
The functional maturation of oocytes enclosed in cultured follicles was assessed by studying their ability to resume meiosis in vitro, when removed from follicles and incubated for an additional 24 h (Table 1). At the end of the culture period, regardless of follicle treatment (with or without FSH), all oocytes recovered from cultured follicles (n = 10) had an intact germinal vesicle (GV). After in vitro maturation (IVM), there was no difference in oocyte maturation ability between the control and FSH-treated follicles. Of the oocytes retrieved from cultured follicles larger than 550 µm (n = 146), 53% resumed meiosis and 36% matured to the MII stage, whereas 85% of those recovered from cultured follicles with a diameter less than 550 µm (n = 59) remained immature (P < 0.0001). For follicles under 550 µm developed in vitro, there was no significant difference with those developed in vivo, with only 15% of the oocytes able to resume meiosis. In contrast, for follicles larger than 550 µm, a better maturation stage (P = 0.0002) was obtained for oocytes derived from cultured follicles, with a lower percentage of oocytes in the GVBD/MI stage (18% vs 38%) but a greater percentage in the MII stage (36% vs 9%) for in vitro vs in vivo origin respectively (Table 1). In our IVM conditions, 85% of the oocytes recovered from 1 to 2 mm follicles isolated from fresh ovarian cortex fragments (n = 58) were able to resume meiosis, only 35% of which matured to the MII stage and 72% of oocytes recovered from follicles up to 2 mm in diameter (n = 100) matured to MII (positive controls).
Meiotic competence of oocytes derived from follicles of different sizes obtained in vivo or after in vitro culture in control medium or with FSH.
Meiotic stages reached (%) | ||||||
---|---|---|---|---|---|---|
Follicular diameter | Follicle origin | No. of oocytes | GV | GVBD/MI | MII | Significance |
400 to <550 µm | In vivo | 27 | 85.2 | 14.8 | 0.0 | a |
In vitro control | 31 | 93.5 | 6.5 | 0.0 | a | |
In vitro FSH | 28 | 75.0 | 17.9 | 7.1 | a,b | |
550–950 µm | In vivo | 56 | 53.6 | 37.5 | 8.9 | b |
In vitro control | 71 | 49.3 | 16.9 | 33.8 | c | |
In vitro FSH | 75 | 44.0 | 18.7 | 37.3 | c | |
1–2 mm | In vivo | 58 | 15.5 | 50.0 | 34.5 | d |
>2 mm | In vivo | 100 | 12.0 | 16.0 | 72.0 | e |
Cumulus cell-enclosed oocytes were cultured for 24 h (IVM) and then classified in 3 different maturation stages: germinal vesicle (GV), germinal vesicle breakdown (GVBD) or metaphase I (MI), metaphase II (MII).
Different letters indicate significant differences between groups of follicles considering the three meiotic stages reached (P < 0.05).
Differential gene expression between follicles developed in vivo and in vitro
To better understand the transcriptional changes during follicular growth in vivo compared to those induced by the in vitro culture conditions, we performed qPCR on follicles sampled from the preantral to antral stages collected in vivo or follicles cultured in control in vitro conditions for 6, 13 or 20 days. We focused on 40 genes known to be expressed by follicles at these developmental stages, which were classified according to their expression levels in vivo or in vitro (Supplementary Table 4) and their change in expression (Supplementary Table 5) at different follicular stages.
Based on the expression of these 40 target genes, a non-supervised hierarchical clustering and heat map was generated (Fig. 5). As expected, this analysis distinctly separated the follicles sampled into two clusters based on their gene expression profiles for those developed in vivo vs in vitro. Furthermore, the large preantral and early antral follicles were clustered in the same region (correlation node = 0.6), whereas the antral follicles were in a distinct region (correlation node = 0.3).
The cluster analysis also identified two major clusters in the gene axis. Of the 40 genes analyzed, downregulated expression of 17 genes (correlation node = 0.05) was found in follicles cultured in vitro, and 20 genes were upregulated (correlation node = 0.5) compared to their expression in follicles at the same stage collected in vivo. The cluster of downregulated genes included receptors (AMHR2, KIT, BMPR1B and INSR), ligands (KITL, AMH, BMP6, BMP15 and LIF) and genes involved in cell proliferation, meiosis and apoptosis (CDKN1A, GADD45A, BCL2, WEE2 and MYC). Another gene cluster comprised three genes whose expression increased with follicle development similarly in vivo and in vitro, including the oocyte-specific gene SPO11 and two follicular somatic receptor genes, LIFR and IGF1R. The cluster of upregulated genes mainly included those that encode nuclear receptors (PR, ESR1, ESR2 and AR), membrane receptors (BMPR1A, BMPR2, FSHR and LHR), members of the inhibin/follistatin system (INHA, INHBA, INHBB and FST) and genes involved in cellular proliferation, communication and apoptosis (CCND2, GJA1, CCDC80, BAX and TP53) and steroidogenesis (CYP19A1, CYP17A1 and CYP11A1).
Gene expression in the somatic cells and the oocyte of follicles developed in vivo and in vitro
Figure 6 shows the results for 12 representative genes expressed in somatic cells of the follicles cultured in vitro compared to follicles obtained in vivo at different developmental stages. These genes were classified into three groups. The first group represented genes whose expression was highly increased during follicular growth in vitro compared to those developed in vivo, and included CCND2, GJA1, BAX, CYP19A1, ESR2, FSHR, INHA, INHBA and INHBB. In vivo, the expression of FSHR, INHBA and INHBB was stable, whereas the expression of CCND2, CYP19A1 and INHA slightly increased during follicular growth. The second group represented genes whose expression was decreased during follicular growth in vitro compared to in vivo, including CDKN1A and BCL2. CDKN1A expression was highly downregulated in vivo (7.6-fold) between the small preantral and antral stages and drastically reduced (3-fold) in the large preantral follicles cultured in vitro compared to similar follicles obtained in vivo (P < 0.01), which remained stable thereafter. BCL2 expression was also progressively downregulated (2-fold) in vivo between the small preantral and antral stages, with a similar pattern observed in vitro, in which BCL2 expression was reduced fourfold during the culture period compared to follicles obtained in vivo. The third group included only the AMH gene, whose expression was highly increased during follicular growth in vivo compared to in vitro. Moreover, antral follicles developed in vitro showed a 50% reduction in the level of AMH mRNA compared to in vivo follicles. For follicles developed in vitro, the initial increase in AMH expression from the large preantral to early antral follicle stage (EA/LPA = 1.6) was attenuated thereafter (A/EA = 0.9).
Figure 7 shows the results of the qPCR analysis for six representative genes that were specifically expressed in oocytes, the expression of which increased at the different follicular developmental stages in vivo. For five of these genes, KIT, BMP15, GJA4, ZP3 and WEE2, the expression was lower in follicles developed in vitro compared to those developed in vivo. The expression of KIT, BMP15 and WEE2 increased during in vitro culture, whereas GJA4 and ZP3 expression remained low and stable. In contrast, SPO11 expression was found to be similar in follicles developed in vitro and in vivo, with increased expression observed from the preantral to the antral stages in vivo and in vitro.
Discussion
This study reports the developmental changes accompanying the growth of ovine ovarian follicles between 160 and 950 µm in diameter during a 20-day culture period. We have established the growth and secretory features and the transcriptomic profiles of the cultured follicles, in addition to the ability of their enclosed oocytes to mature when compared to follicles developed in vivo.
The animal model chosen was the prepubertal ewe lamb, between 6 and 8 months of age, characterized by low serum FSH levels and ovaries containing high numbers of growing follicles (Rawlings & Churchill 1990). This model can be of particular interest for the human species. Indeed, fertility preservation by cryopreservation in human mainly concerns children and adolescents, patients who are at high risk of malignant contamination of their ovarian tissue. Anderson and coworkers showed a reduced growth of isolated follicles for young patients compared to adults, indicating true intrafollicular differences in addition to potential differences in their local environment. There are likely maturational processes occurring in the ovary through childhood and adolescence, which may involve the loss of abnormal follicles and increasing follicle developmental competence (Anderson et al. 2014).
This is the first detailed study to accurately describe the morphological modifications of follicles induced by in vitro culture conditions. Given that the 3D relationship (spatial positioning) between the different cell types of the follicle in culture may influence the somatic cell proliferation and differentiation, as well as oocyte developmental competence (Woodruff & Shea 2007), we must pay special attention to the choice of culture conditions. Our study is based on a simplified preantral follicle culture system that allows the follicles to partly attach to the bottom of the plastic well without spreading (Newton et al. 1999, Picton et al. 2008). This system is a compromise between an adhesive 2D culture system, which may lead to distorted follicular morphology and frequent follicle disruptions, and a compressive 3D culture using complex hydrogels, which is better able to support the dynamic spatiotemporal regulation of follicle development (Shea et al. 2014).
In sheep, a follicle is known to progress from the preantral (210 µm) to the small antral (620 µm) stage in 27 days in vivo (Cahill & Mauleon 1980). However, in our in vitro conditions, a similar increase in size took only 20 days and was not affected by the presence of FSH. The shorter time required for follicle development was accompanied by several morphological differences between the in vivo and in vitro conditions. Firstly, the oocyte growth rate was decreased in vitro, as the maximal external diameter of 140 µm was reached in follicles with a diameter of 375 µm in vivo (in agreement with previous observations (Cahill & Mauleon 1980)), but only in follicles of 625 µm in vitro. Secondly, the number of somatic cells per follicle increased about 18-fold from the preantral to the antral stage in vivo, as shown previously (Cahill & Mauleon 1980, Lundy et al. 1999), whereas for follicles developed in vitro, there was an additional increase in proliferation rate in the small antral follicles larger than 550 µm of diameter. Thirdly, antrum formation was delayed and slowed in vitro. The components of the follicular fluid that fills the antrum and is involved in its expansion in vivo are known to be derived from both follicular cells and serum (Rodgers & Irving-Rodgers 2010). During in vitro follicle development, some of these components likely originate from the culture medium; however, the absence of functional follicular vascularization may have prevented the movement of fluid. In conclusion, follicular morphogenesis was affected by the in vitro culture conditions, evidenced by a decrease in oocyte growth rate, an increase in somatic cell proliferation and a delay in antrum formation.
Early follicular growth in vitro and in vivo was further investigated by comparing the expression of a panel of 40 genes. The proliferation rate of granulosa cells is known to be regulated by the balance of positive and negative cell cycle regulators (Robker & Richards 1998). In vitro follicular growth was accompanied by an increase in the expression of CCND2 with a simultaneous decrease in CDKN1A expression. Interestingly, the expression of these genes in follicles developed in vitro was higher and lower respectively, than follicles developed in vivo, suggesting that these differences may be, at least partly, responsible for the higher rate of granulosa cell proliferation observed in vitro. Moreover, in vitro, the mRNA levels of BCL2 and BAX, which are anti- and pro-apoptotic factors respectively, were drastically decreased (4-fold) and increased (2-fold) respectively, suggesting that the increased BAX/BCL2 ratio that exists during follicular growth in vivo ((Tilly et al. 1997) and present results) is enhanced in vitro. Nevertheless, the increased rate of follicular atresia was similar during the growth of small antral follicles in vivo and in vitro.
Furthermore, the culture conditions induced a higher GJA1 (encoding the connexin 43 protein) expression compared with in vivo conditions, likely ensuring improved coordination and synchronization of the processes involved in granulosa cell proliferation and differentiation previously observed in vivo (Nuttinck et al. 2000). The differentiation state of granulosa cells was assessed by determining the expression of several well-known markers of follicular maturation. In the present study, the expression of CYP19A1, FSHR, ESR2, INHA, INHBA, INHBB and FST was very low during early folliculogenesis in vivo, as expected (Scaramuzzi et al. 2011) but was increased by 3- to 5-fold during the culture period, finally reaching 5- to 16-fold higher levels of expression in antral follicles developed in vitro than those developed in vivo. These observations suggest that the in vitro culture conditions induced early and fast acquisition of a mature stage of follicle development. Moreover, elevated E2 concentrations were observed in the culture medium throughout the 20-day culture period, with the conditions including FSH stimulation leading to the highest E2 follicular production. This increase in E2 secretion by follicles was associated with an increase in the number of granulosa cells per follicle, resulting in a global steady state of E2 secretion per cell. It is suggested that follicular production of E2 in vitro may be limited by the weak development or a lack of theca cells required to provide androgens as a substrate for the aromatase enzyme present in granulosa cells. In any case, such an early increase in the in vitro expression of markers of granulosa cell differentiation in large preantral or/and early antral follicles suggests accelerated follicular maturation as a result of in vitro culture, leading to prematurely estrogen-active follicles compared with those developed in vivo.
In the sheep, in vivo intrafollicular expression of AMH reaches its maximum level in early antral follicles of about 1 mm diameter and gradually declines in the subsequent stages of follicle development (Monniaux et al. 2013). In the present study, AMH expression was indeed highly upregulated (>5 fold) in vivo from the small preantral to the small antral stage; however, AMH mRNA was twofold lower at the small antral stage in vitro than that in vivo, accompanied by decreased AMH secretion per granulosa cell during culture. These observations are consistent with those reported by Xu and coworkers (Xu et al. 2016) who showed a decreased rate of AMH secretion during the final stages of antral follicular growth in vitro. This decrease may alleviate the inhibitory action of AMH on follicular differentiation in vitro, described elsewhere (Campbell et al. 2012).
The increased expression of the oocyte-specific genes ZP3, GJA4 (encoding the connexin 37 protein), KIT and BMP15 between the preantral and the small antral stage of follicular growth in vivo is in agreement with previous observations (Roller et al. 1989, Wright et al. 2001, Hutt et al. 2006, Feary et al. 2007, Bonnet et al. 2011). However, the expression of these genes was drastically downregulated during culture. This suggests that communication between the oocyte and its surrounding follicular cells is probably impaired in vitro. The lower expression of GJA4 and KIT may explain the decreased oocyte growth rate observed in vitro. Indeed, mutations in either of these genes has been shown to affect the completion of oocyte growth (Carabatsos et al. 2000, Reynaud et al. 2001). Moreover, the altered BMP15 expression in the oocyte observed in vitro may contribute to the advanced follicle maturation in vitro compared to in vivo developmental conditions (Moore & Shimasaki 2005).
Oocytes reached nuclear maturation, despite the alterations in oocyte gene expression in vitro, leading to a higher degree of meiotic competence compared to oocytes from follicles of the same size developed in vivo (Cognié et al. 1998). Indeed, in our culture conditions, half of the oocytes obtained from follicles of 550–950 µm in diameter were able to resume meiosis, with fourfold more oocytes reaching the MII stage after in vitro follicle development compared to in vivo development. In sharp contrast, this competence to resume meiosis was only reached in follicles larger than 1 mm in vivo. Interestingly, although the expression of WEE2, a gene involved in the maintenance of meiotic arrest in oocytes (Solc et al. 2010), increased from the preantral to antral stages in vivo (Bonnet et al. 2015) and in vitro (present results), its expression was lower in vitro. This could correspond to the maintenance of meiotic arrest of follicle-enclosed oocytes in vivo, and the earlier acquisition of oocyte competence to resume meiosis in vitro. Indeed, downregulation of WEE2 expression resulting from RNAi injection in vitro or transgenic overexpression of RNAi in vivo, has been reported to result in meiotic resumption in mouse (Han et al. 2005) and monkey oocytes (Hanna et al. 2010). In contrast, the expression of SPO11, a gene preferentially expressed in oocytes from small antrum follicles (Bonnet et al. 2015) and involved in meiotic DNA repair and recombination (Lam & Keeney 2015), was similarly increased in vivo and in vitro. Therefore, our culture conditions appeared to maintain meiotic arrest of oocyte-enclosed follicles, but facilitated the acquisition of meiotic competence, even preparing the oocyte to perform meiosis. Overall, further investigations are needed to obtain a better understanding of the functional state of the oocyte enclosed in the follicle in vitro.
The results of the present study demonstrate that ovine preantral follicles developed in vitro can reach the small antral follicle stage, with some characteristics of fully mature follicles. This in vitro development requires insulin supplementation but can be achieved without FSH. The main effect of FSH observed in our in vitro study was an increase in steroidogenesis accompanied by a decrease in AMH secretion by granulosa cells. For ovine follicles developed in vivo, FSH is neither essential for follicles to grow up to 2 mm diameter nor for the proliferation of granulosa cells (McNeilly et al. 1986, Scaramuzzi et al. 2011). However, these small follicles have FSH receptors (Tisdall et al. 1995) and are therefore sensitive to this hormone. Moreover, FSHR expression, found to be stably expressed in vivo from the preantral to the small antral stage, was progressively increased by up to 2.8-fold in vitro, supporting previous observations in goat (Silva et al. 2013) and suggesting better sensitivity to FSH in culture than those developed in vivo.
In our culture conditions, the expression of thecal markers like LHR, CYP17A1 and CYP11A1 suggests the presence of some theca cells in variable numbers, likely due to the technique of follicle dissection. Interestingly, the expression of these genes was upregulated in vitro compared to that observed in vivo, corresponding to the earlier differentiation status of the follicle induced by the in vitro culture.
In conclusion, compared to in vivo developmental conditions, the in vitro culture conditions utilized in this study induced decreased rates of oocyte growth, a delay in antrum formation, an increase in the rate of follicular cell proliferation and advanced follicular cell differentiation, reflected by higher expression of gene markers of follicular maturation and an earlier ability of some oocytes to resume meiosis after in vitro maturation. These changes, summarized in Fig. 8, likely result from the privileged dialog existing between the oocyte and its surrounding follicular cells, which coordinate the different phases of follicle development (Eppig et al. 2002, Buratini & Price 2011, Monniaux 2016), this dialog being affected by the in vitro culture system.
Supplementary data
This is linked to the online version of the paper at http://dx.doi.org/10.1530/REP-16-0627.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This work was supported by the Région Centre (CRYOVAIRE, grant number 320000268).
Acknowledgements
The authors thank Dr Yves Combarnous (INRA, Tours) for providing the ovine FSH. They thank J P Dubois and A Arnoult of the CIRE team and the staff of the slaughtering house of Vendôme for their help in ovarian collection. They also thank our colleagues, Cécile Donnadieu and Alain Roulet, at the genomic platform (GeT-PlaGe, Genotoul, Toulouse) for their contribution to the Fluidigm experiment.
References
Anderson RA, McLaughlin M, Wallace WH, Albertini DF & Telfer EE 2014 The immature human ovary shows loss of abnormal follicles and increasing follicle developmental competence through childhood and adolescence. Human Reproduction 29 97–106. (doi:10.1093/humrep/det388)
Barboni B, Russo V, Cecconi S, Curini V, Colosimo A, Garofalo ML, Capacchietti G, Di Giacinto O & Mattioli M 2011 In vitro grown sheep preantral follicles yield oocytes with normal nuclear-epigenetic maturation. PLoS ONE 6 e27550. (doi:10.1371/journal.pone.0027550)
Bertoldo MJ, Bernard J, Duffard N, Tsikis G, Alves S, Calais L, Uzbekova S, Monniaux D, Mermillod P & Locatelli Y 2016 Inhibitors of c-Jun phosphorylation impede ovine primordial follicle activation. Molecular Human Reproduction 22 338–349. (doi:10.1093/molehr/gaw012)
Besnard N, Pisselet C, Monniaux D, Locatelli A, Benne F, Gasser F, Hatey F & Monget P 1996 Expression of messenger ribonucleic acids of insulin-like growth factor binding protein-2, -4, and -5 in the ovine ovary: localization and changes during growth and atresia of antral follicles. Biology of Reproduction 55 1356–1367. (doi:10.1095/biolreprod55.6.1356)
Bonnet A, Bevilacqua C, Benne F, Bodin L, Cotinot C, Liaubet L, Sancristobal M, Sarry J, Terenina E & Martin P et al. 2011 Transcriptome profiling of sheep granulosa cells and oocytes during early follicular development obtained by laser capture microdissection. BMC Genomics 12 e417. (doi:10.1186/1471-2164-12-417)
Bonnet A, Cabau C, Bouchez O, Sarry J, Marsaud N, Foissac S, Woloszyn F, Mulsant P & Mandon-Pepin B 2013 An overview of gene expression dynamics during early ovarian folliculogenesis: specificity of follicular compartments and bi-directional dialog. BMC Genomics 14 e904. (doi:10.1186/1471-2164-14-904)
Bonnet A, Servin B, Mulsant P & Mandon-Pepin B 2015 Spatio-temporal gene expression profiling during in vivo early ovarian folliculogenesis: integrated transcriptomic study and molecular signature of early follicular growth. PLoS ONE 10 e0141482. (doi:10.1371/journal.pone.0141482)
Buratini J & Price CA 2011 Follicular somatic cell factors and follicle development. Reproduction, Fertility, and Development 23 32–39. (doi:10.1071/RD10224)
Cahill LP & Mauleon P 1980 Influences of season, cycle and breed on follicular growth rates in sheep. Journal of Reproduction and Fertility 58 321–328. (doi:10.1530/jrf.0.0580321)
Campbell BK, Clinton M & Webb R 2012 The role of anti-Mullerian hormone (AMH) during follicle development in a monovulatory species (sheep). Endocrinology 153 4533–4543. (doi:10.1210/en.2012-1158)
Carabatsos MJ, Sellitto C, Goodenough DA & Albertini DF 2000 Oocyte-granulosa cell heterologous gap junctions are required for the coordination of nuclear and cytoplasmic meiotic competence. Developmental Biology 226 167–179. (doi:10.1006/dbio.2000.9863)
Cecconi S, Capacchietti G, Russo V, Berardinelli P, Mattioli M & Barboni B 2004 In vitro growth of preantral follicles isolated from cryopreserved ovine ovarian tissue. Biology of Reproduction 70 12–17. (doi:10.1095/biolreprod.103.016774)
Chaves RN, Duarte AB, Rodrigues GQ, Celestino JJ, Silva GM, Lopes CA, Almeida AP, Donato MA, Peixoto CA & Moura AA et al. 2012 The effects of insulin and follicle-simulating hormone (FSH) during in vitro development of ovarian goat preantral follicles and the relative mRNA expression for insulin and FSH receptors and cytochrome P450 aromatase in cultured follicles. Biology of Reproduction 87 69. (doi:10.1095/biolreprod.112.099010)
Cognié Y, Benoit F, Poulin N, Khatir H & Driancourt MA 1998 Effect of follicle size and of the FecB Booroola gene on oocyte function in sheep. Journal of Reproduction and Fertility 112 379–386. (doi:10.1530/jrf.0.1120379)
Cortvrindt R, Smitz J & Van Steirteghem AC 1996 In-vitro maturation, fertilization and embryo development of immature oocytes from early preantral follicles from prepuberal mice in a simplified culture system. Human Reproduction 11 2656–2666. (doi:10.1093/oxfordjournals.humrep.a019188)
da Silva CM, Castro SV, Faustino LR, Rodrigues GQ, Brito IR, Rossetto R, Saraiva MV, Campello CC, Lobo CH & Souza CE et al. 2015 Activin-A promotes the development of goat isolated secondary follicles in vitro. Zygote 23 41–52. (doi:10.1017/S0967199413000294)
Eppig JJ, Wigglesworth K & Pendola FL 2002 The mammalian oocyte orchestrates the rate of ovarian follicular development. PNAS 99 2890–2894. (doi:10.1073/pnas.052658699)
Estienne A, Pierre A, di Clemente N, Picard JY, Jarrier P, Mansanet C, Monniaux D & Fabre S 2015 Anti-Mullerian hormone regulation by the bone morphogenetic proteins in the sheep ovary: deciphering a direct regulatory pathway. Endocrinology 156 301–313. (doi:10.1210/en.2014-1551)
Feary ES, Juengel JL, Smith P, French MC, O’Connell AR, Lawrence SB, Galloway SM, Davis GH & McNatty KP 2007 Patterns of expression of messenger RNAs encoding GDF9, BMP15, TGFBR1, BMPR1B, and BMPR2 during follicular development and characterization of ovarian follicular populations in ewes carrying the Woodlands FecX2W mutation. Biology of Reproduction 77 990–998. (doi:10.1095/biolreprod.107.062752)
Figueiredo JR, Rodrigues AP, Silva JR & Santos RR 2011 Cryopreservation and in vitro culture of caprine preantral follicles. Reproduction, Fertility and Development 23 40–47. (doi:10.1071/RD10227)
Gougeon A 1996 Regulation of ovarian follicular development in primates: facts and hypotheses. Endocrine Reviews 17 121–155. (doi:10.1210/edrv-17-2-121)
Gupta PS, Ramesh HS, Manjunatha BM, Nandi S & Ravindra JP 2008 Production of buffalo embryos using oocytes from in vitro grown preantral follicles. Zygote 16 57–63. (doi:10.1017/S096719940700442X)
Gutierrez CG, Ralph JH, Telfer EE, Wilmut I & Webb R 2000 Growth and antrum formation of bovine preantral follicles in long-term culture in vitro. Biology of Reproduction 62 1322–1328. (doi:10.1095/biolreprod62.5.1322)
Han SJ, Chen R, Paronetto MP & Conti M 2005 Wee1B is an oocyte-specific kinase involved in the control of meiotic arrest in the mouse. Current Biology 15 1670–1676. (doi:10.1016/j.cub.2005.07.056)
Hanna CB, Yao S, Patta MC, Jensen JT & Wu X 2010 WEE2 is an oocyte-specific meiosis inhibitor in rhesus macaque monkeys. Biology of Reproduction 82 1190–1197. (doi:10.1095/biolreprod.109.081984)
Hutt KJ, McLaughlin EA & Holland MK 2006 Kit ligand and c-Kit have diverse roles during mammalian oogenesis and folliculogenesis. Molecular Human Reproduction 12 61–69. (doi:10.1093/molehr/gal010)
Kona SS, Praveen Chakravarthi V, Siva Kumar AV, Srividya D, Padmaja K & Rao VH 2016 Quantitative expression patterns of GDF9 and BMP15 genes in sheep ovarian follicles grown in vivo or cultured in vitro. Theriogenology 85 315–322. (doi:10.1016/j.theriogenology.2015.09.022)
Lakshminarayana BNV, Praveen Chakravarthi V, Brahmaiah KV & Rao VH 2014 Quantification of P450 aromatase gene expression in cultured and in vivo grown ovarian follicles in sheep. Small Ruminant Research 117 66–72. (doi:10.1016/j.smallrumres.2013.11.016)
Lam I & Keeney S 2015 Mechanism and regulation of meiotic recombination initiation. Cold Spring Harbor Perspectives in Medicine 7 a016634. (doi:10.1101/cshperspect.a016634)
Lima IM, Brito IR, Rossetto R, Duarte AB, Rodrigues GQ, Saraiva MV, Costa JJ, Donato MA, Peixoto CA & Silva JR et al. 2012 BMPRIB and BMPRII mRNA expression levels in goat ovarian follicles and the in vitro effects of BMP-15 on preantral follicle development. Cell and Tissue Research 348 225–238. (doi:10.1007/s00441-012-1361-4)
Lundy T, Smith P, O’Connell A, Hudson NL & McNatty KP 1999 Populations of granulosa cells in small follicles of the sheep ovary. Journal of Reproduction and Fertility 115 251–262. (doi:10.1530/jrf.0.1150251)
Magalhaes DM, Duarte AB, Araujo VR, Brito IR, Soares TG, Lima IM, Lopes CA, Campello CC, Rodrigues AP & Figueiredo JR 2011 In vitro production of a caprine embryo from a preantral follicle cultured in media supplemented with growth hormone. Theriogenology 75 182–188. (doi:10.1016/j.theriogenology.2010.08.004)
Magalhaes-Padilha DM, Geisler-Lee J, Wischral A, Gastal MO, Fonseca GR, Eloy YR, Geisler M, Figueiredo JR & Gastal EL 2013 Gene expression during early folliculogenesis in goats using microarray analysis. Biology of Reproduction 89 1–12. (doi:10.1095/biolreprod.112.106096)
McGee EA & Raj RS 2015 Regulators of ovarian preantral follicle development. Seminars in Reproductive Medicine 33 179–184. (doi:10.1055/s-0035-1552584)
McNeilly AS, Jonassen JA & Fraser HM 1986 Suppression of follicular development after chronic LHRH immunoneutralization in the ewe. Journal of Reproduction and Fertility 76 481–490. (doi:10.1530/jrf.0.0760481)
Monniaux D 2016 Driving folliculogenesis by the oocyte-somatic cell dialog: lessons from genetic models. Theriogenology 86 41–53. (doi:10.1016/j.theriogenology.2016.04.017)
Monniaux D, Drouilhet L, Rico C, Estienne A, Jarrier P, Touzé JL, Sapa J, Phocas F, Dupont J & Dalbies-Tran R et al. 2013 Regulation of anti-Mullerian hormone production in domestic animals. Reproduction, Fertility and Development 25 1–16. (doi:10.1071/RD12270)
Moore RK & Shimasaki S 2005 Molecular biology and physiological role of the oocyte factor, BMP-15. Molecular and Cellular Endocrinology 234 67–73. (doi:10.1016/j.mce.2004.10.012)
Newton H, Picton H & Gosden RG 1999 In vitro growth of oocyte-granulosa cell complexes isolated from cryopreserved ovine tissue. Journal of Reproduction and Fertility 115 141–150. (doi:10.1530/jrf.0.1150141)
Nuttinck F, Peynot N, Humblot P, Massip A, Dessy F & Flechon JE 2000 Comparative immunohistochemical distribution of connexin 37 and connexin 43 throughout folliculogenesis in the bovine ovary. Molecular Reproduction and Development 57 60–66. (doi:10.1002/1098-2795(200009)57:1<60::AID-MRD9>3.0.CO;2-6)
Palma GA, Arganaraz ME, Barrera AD, Rodler D, Mutto AA & Sinowatz F 2012 Biology and biotechnology of follicle development. Scientific World Journal 2012 938138. (doi:10.1100/2012/938138)
Picton HM, Harris SE, Muruvi W & Chambers EL 2008 The in vitro growth and maturation of follicles. Reproduction 136 703–715. (doi:10.1530/REP-08-0290)
Pierre A, Pisselet C, Dupont J, Mandon-Pepin B, Monniaux D, Monget P & Fabre S 2004 Molecular basis of bone morphogenetic protein-4 inhibitory action on progesterone secretion by ovine granulosa cells. Journal of Molecular Endocrinology 33 805–817. (doi:10.1677/jme.1.01545)
Praveen Chakravarthi V, Kona SS, Siva Kumar AV, Bhaskar M & Rao VH 2015 Quantitative expression of antiapoptotic and proapoptotic genes in sheep ovarian follicles grown in vivo or cultured in vitro. Theriogenology 83 590–595. (doi:10.1016/j.theriogenology.2014.10.024)
Rawlings NC & Churchill IJ 1990 Effect of naloxone on gonadotrophin secretion at various stages of development in the ewe lamb. Journal of Reproduction and Fertility 89 503–509. (doi:10.1530/jrf.0.0890503)
Reynaud K, Cortvrindt R, Smitz J, Bernex F, Panthier JJ & Driancourt MA 2001 Alterations in ovarian function of mice with reduced amounts of KIT receptor. Reproduction 121 229–237. (doi:10.1530/rep.0.1210229)
Robker RL & Richards JS 1998 Hormone-induced proliferation and differentiation of granulosa cells: a coordinated balance of the cell cycle regulators cyclin D2 and p27Kip1. Molecular Endocrinology 12 924–940. (doi:10.1210/mend.12.7.0138)
Rodgers RJ & Irving-Rodgers HF 2010 Formation of the ovarian follicular antrum and follicular fluid. Biology of Reproduction 82 1021–1029. (doi:10.1095/biolreprod.109.082941)
Roller RJ, Kinloch RA, Hiraoka BY, Li SS & Wassarman PM 1989 Gene expression during mammalian oogenesis and early embryogenesis: quantification of three messenger RNAs abundant in fully grown mouse oocytes. Development 106 251–261.
Scaramuzzi RJ, Baird DT, Campbell BK, Driancourt MA, Dupont J, Fortune JE, Gilchrist RB, Martin GB, McNatty KP & McNeilly AS et al. 2011 Regulation of folliculogenesis and the determination of ovulation rate in ruminants. Reproduction, Fertility, and Development 23 444–467. (doi:10.1071/RD09161)
Shea LD, Woodruff TK & Shikanov A 2014 Bioengineering the ovarian follicle microenvironment. Annual Review of Biomedical Engineering 16 29–52. (doi:10.1146/annurev-bioeng-071813-105131)
Silva CM, Castro SV, Faustino LR, Rodrigues GQ, Brito IR, Rossetto R, Saraiva MV, Campello CC, Lobo CH & Souza CE et al. 2013 The effects of epidermal growth factor (EGF) on the in vitro development of isolated goat secondary follicles and the relative mRNA expression of EGF, EGF-R, FSH-R and P450 aromatase in cultured follicles. Research in Veterinary Science 94 453–461. (doi:10.1016/j.rvsc.2012.12.002)
Silva GM, Rossetto R, Chaves RN, Duarte AB, Araujo VR, Feltrin C, Bernuci MP, Anselmo-Franci JA, Xu M & Woodruff TK et al. 2015 In vitro development of secondary follicles from pre-pubertal and adult goats cultured in two-dimensional or three-dimensional systems. Zygote 23 475–484. (doi:10.1017/S0967199414000070)
Solc P, Schultz RM & Motlik J 2010 Prophase I arrest and progression to metaphase I in mouse oocytes: comparison of resumption of meiosis and recovery from G2-arrest in somatic cells. Molecular Human Reproduction 16 654–664. (doi:10.1093/molehr/gaq034)
Telfer EE & Zelinski MB 2013 Ovarian follicle culture: advances and challenges for human and nonhuman primates. Fertility and Sterility 99 1523–1533. (doi:10.1016/j.fertnstert.2013.03.043)
Telfer EE, McLaughlin M, Ding C & Thong KJ 2008 A two-step serum-free culture system supports development of human oocytes from primordial follicles in the presence of activin. Human Reproduction 23 1151–1158. (doi:10.1093/humrep/den070)
Tilly JL, Tilly KI & Perez GI 1997 The genes of cell death and cellular susceptibility to apoptosis in the ovary: a hypothesis. Cell Death and Differentiation 4 180–187. (doi:10.1038/sj.cdd.4400238)
Tisdall DJ, Watanabe K, Hudson NL, Smith P & McNatty KP 1995 FSH receptor gene expression during ovarian follicle development in sheep. Journal of Molecular Endocrinology 15 273–281. (doi:10.1677/jme.0.0150273)
Vanacker J, Luyckx V, Amorim C, Dolmans MM, Van Langendonckt A, Donnez J & Camboni A 2013 Should we isolate human preantral follicles before or after cryopreservation of ovarian tissue? Fertility and Sterility 99 1363–1368. (doi:10.1016/j.fertnstert.2012.12.016)
Woodruff TK & Shea LD 2007 The role of the extracellular matrix in ovarian follicle development. Reproductive Sciences 14 6–10. (doi:10.1177/1933719107309818)
Wright CS, Becker DL, Lin JS, Warner AE & Hardy K 2001 Stage-specific and differential expression of gap junctions in the mouse ovary: connexin-specific roles in follicular regulation. Reproduction 121 77–88. (doi:10.1530/rep.0.1210077)
Xu M, Barrett SL, West-Farrell E, Kondapalli LA, Kiesewetter SE, Shea LD & Woodruff TK 2009 In vitro grown human ovarian follicles from cancer patients support oocyte growth. Human Reproduction 24 2531–2540. (doi:10.1093/humrep/dep228)
Xu J, Xu M, Bernuci MP, Fisher TE, Shea LD, Woodruff TK, Zelinski MB & Stouffer RL 2013 Primate follicular development and oocyte maturation in vitro. Advances in Experimental Medecine and Biology 761 43–67. (doi:10.1007/978-1-4614-8214-7_5)
Xu J, Xu F, Letaw JH, Park BS, Searles RP & Ferguson BM 2016 Anti-Mullerian hormone is produced heterogeneously in primate preantral follicles and is a potential biomarker for follicle growth and oocyte maturation in vitro. Journal of Assisted Reproduction and Genetics 33 1665–1675. (doi:10.1007/s10815-016-0804-3)