Introduction
Fertility preservation is an important subject in the field of reproductive health. Cryopreservation of ovarian tissue including primordial follicles (PFs) is a strategy for fertility preservation among women suffering from cancer (1). Reimplantation of frozen-thawed ovarian tissue has been associated with malignancies (2), which is preventable via in vitro maturation of PFs. Such a mechanism can be developed by some factors activating PFs (3).
The purpose for culture of ovarian cortical tissue is the transition of PFs at resting phase into the primary and secondary follicular phase (4, 5). However, follicle development can not be supported and there is a need for nutrients or growth factors. Several intra-and extra-ovarian factors essential for activation of PFs and development of follicles in organ cultures have been investigated (6).
Developing antral follicle is associated with follicle stimulating hormone (FSH) within the mammalian ovary; however, the mechanism underlying preantral growth has not yet been identified. In genetically hypogonadotropic mice, hypophysectomy or pharmacological suppression of FSH leads to antral follicles depletion and prevents developing follicles through the late preantral phases. Likewise, in the mice with mutant FSH receptor gene, there was no improvement in follicle development over the multilayered preantral phase. Regarding human subjects, the mutations of the FSH receptor gene have been linked to the lack of mature antral follicles, estrogen, and amenorrhea (7).
It has widely been shown that FSH can induce in vivo development of preantral follicles and also stimulate culture of in vitro ovarian tissue. In vivo administration of FSH promoted survival of preantral follicles. FSH administration into cultures of preantral follicles resulted in the same outcomes (8). Recent research has shown that FSH binding sites are available in preantral follicles in rodents and sheep ovaries. Adult mouse ovarian surface epithelium can express FSH receptors locating on the pluripotent stem cells and ovarian germ stem cells. These stem cells are activated by pregnant mare serum gonadotropin administration leading to augmented neo-oogenesis as well as PFs accumulation (9). Parte et al. showed induction of pluripotent stem cells located in ovarian surface epithelium (OSE) by FSH in cultured human and marmoset ovarian cortical tissue (6). FSH therapy caused vast proliferation of OSE and an increase in pluripotent stem cells and germ cells (7).
Basic fibroblast growth factor (bFGF) is a crucial causative agent for PF transition, which can be expressed in the ooplasm of primordial and primary follicles, OSE, vascular smooth muscle cells, and corpus luteum (10). In goats, development and survival of PF is enhanced by the combination of bFGF and FSH (8). bFGF has been shown to be involved in follicle recruitment and development via stimulation of granulosa cell proliferation (10). Quennell et al. (11) indicated that bFGF transcripts are available in PFs and there is a reduction in their level by growing follicles. According to Tang et al. (12), the effectiveness of FSH on PF growth and survival was increased by bFGF through culturing bovine ovarian tissue for a long period. Interestingly, FSH receptors and bFGF can be expressed in OSE (10) located on ovarian stem cells (13). Further, it has been shown that bFGF is an effective factor in the culture medium needed for maintaining the pluripotency in human embryonic stem cells (14).
Neuronal growth factors, including neurotrophins (NTs), are associated with early folliculogenesis (14). So far, five neurotrophins have been found (15). Two isoforms have been identified for neurotrophin 3 (NT3) (16). NGFs (10), containing NT3 (17), induce activation of the primordial follicle. NT3 is possibly associated with the primordial-to-primary transition in humans. It also acts via synergistic interactions with other NTs and other factors.
Accordingly, in this study, an attempt was made to investigate FSH, bFGF, and NT3 impacts on PF development focusing on ovarian stem cells within the ovarian cortical tissues in human samples. Since previous reports indicated that days 2-3 are suitable for spontaneous activation of PF in vitro, the third day as the time point was selected in this study (18). qRT-PCR analysis was used for different stem cells, oocytes, and granulosa cell specific markers for studying the impact of FSH, bFGF, and NT3 on ovarian stem cells as well as PF development in vitro. Moreover, this study was performed to investigate whether cells extracted from the ovarian cortex can be considered a promising source in obtaining oocytes in vitro.
Methods
Ovarian tissue preparation: The use of human ovarian tissue in this research was confirmed by the Ethics Committee of Iran University of Medical Sciences (Code: 92-03-117-24106). Ovarian tissue was collected from a perimenopausal woman under 40 without underlying diseases and cancer who had undergone total abdominal hysterectomy. The Obstetrics and Gynecology unit at Imam Khomeini Hospital of Tehran, Iran provided perimenopausal ovarian sample after obtaining the donor’s consent in compliance with ethical rules.
Ovarian cortical tissue culture
Cell isolation: A human ovary was transported to the laboratory in phosphate buffered salin (PBS) including 100 IU/ml of penicillin (3%) as well as 100 ug/ml of streptomycin. Afterward, ovarian cortical tissue was dissected into 1 mm3 piece and was placed in a 25 cm2 cell culture flask, and incubated with alpha-MEM (Invitrogen, USA) supplemented with sodium pyruvate (2 mM/L), L-glutamine (2 mM/L), 1% insulin, transferrin, and selenium (Sigma-Aldrich, USA), 1% penicillin (100 IU/ml)/streptomycin (100 μg/ml) (Invitrogen, USA) and 10% fetal bovine serum (Gibco, UK) and the medium was changed every 2 days. After 5 days, cells migrated from explants and attached to the cell culture flask surface and were subcultured after reaching 80% confluency; moreover, cells at passage 3 were used for subsequent experiments. The cells isolated from the explant had a variety of shapes. After three passages, the other cells were removed and most of the remaining cells had a fibroblastic morphology (Figure 1).
Flow cytometry: Cell surface markers expression in the isolated cells was evaluated using flow cytometry for CD105, CD44, and CD90, as MSCs positive markers and CD34, CD45 as negative markers. Cells were fixed using 4% V/V paraformaldehyde (Sigma Aldrich, USA). Then, the cells were incubated with primary antibodies against CD105 (1:500, Cat. N: E01423-1631, eBioscience, USA), CD73 (1:50, Cat. N: 344009, BioLegend, USA), CD90 (1:300, Cat. N: 328113, BioLegend, USA), CD34 (1:100, Cat. N: IQP-144R, IQ Products, Netherlands), and CD45 (1:50, Cat. N: 560978, BioLegend, USA) according to the manufacturer’s instructions. After removing primary antibodies, PBS was used to wash the cells and incubation was done using FITC-conjugated secondary antibody (1:500/dark environment), PE-conjugated secondary antibody (1:300/dark environment), PerCP-conjugated secondary antibody (1:200/dark environment) and marker expression was analyzed using BD FACSCalibur™ flow cytometer (BD Biosciences, USA).
Treatments: The isolated cells were divided into eight groups as follows: control group included cells without any treatment, FSH group included cells treated with 0.5 IU/ml human urinary FSH (Utrofol, Kuanart Pharmaceuticals, India), bFGFG group included cells treated with 100 ng/ml basic FGF (R&D Systems Inc, USA), NT3 group included cells treated with 100 ng/ml NT3 (R&D Systems Inc, USA), and other groups included cells treated with one of the combination of mentioned growth factors including FSH+NT3, FSH+ bFGF, NT3+bFGF, and FSH+NT3+bFGF. Each group included 3 same members (10000 cells). They were kept at 37°C in 5% CO2 followed by monitoring by inverted microscope using Hoffman modulation contrast (Eclipse TE2000-S, Nikon, Japan). Cortical tissue sections were sampled on day three and RNA extraction was done after processing.
Quantitative reverse transcription PCR (qRT-PCR) analysis: Real-time PCR was used for evaluation of transcripts rates for pluripotent stem cells, including OCT4-A and Nanog, early germ cells or oocyte specific markers of c-KIT and VASA as well as for reflecting PF transition to primary follicles (Oocyte specific GDF-9 and Lhx).
RNA extraction kit was used for total RNA extraction. RNX-Plus (Sinaclon, Iran) was resolved in RNase-free water (50 µl) based on the manufacturer's instruction. NanoDrop (Thermo scientific, USA) was applied to analyze RNA for determining its concentration as well as purity. Denaturing gradient gel electrophoresis (DGGE) was employed for measuring RNA integrity. RNA was treated using DNase I (EN0521, Sigma-Aldrich, Germany) for reducing DNA contamination. The process of converting RNA to cDNA was done through cDNA Synthesis Kit (PrimeScript RT Master Mix, Takara Bio INC Company, Japan). Synthesis of cDNAs was performed by RNA (1 μg), oligo dT (0.5 µl), and random hexamer (0.5 μl). The analyses were performed according to the recommended instruction by the producer. SYBR Green-based RT-PCR was done using the synthesized cDNA (1 µg) through 2X qPCR kit (RR820L, Tli RNaseH Plus, Takara Bio INC Company, Japan). Table 1 shows the applied primer pairs. Thermocycling was done based on the following parameters: initial denaturation (95°C/30 s), 40 cycles at 95°C within 5 s followed by annealing and elongation (60°C/30 s). β-actin as a loading control was used for normalization of data. ΔΔCt formula was employed for determining relative alteration in gene expression in the control group (19).
Statistical analysis: All data were expressed as mean±standard error (SEM). Oneway ANOVA was applied to clarify significant differences between groups. When a significant effect was found, the Tukey post-hoc test was performed. All analyses were performed using SPSS v 23 (IBM, USA). The statistical significance level was set at p≤0.5.
Results
Flow cytometry analysis: The isolated mesenchymal stem cells from ovarian tissue with fibroblast like morphology expressed mesenchymal lineage specific markers including CD73, CD90, and CD105 (88.9%, 88.8%, and 81.0%, respectively) and did not express CD34 and CD45 according to flow cytometry results (Figure 2).
Expression of pluripotent stem cells markers: Levels of transcripts for pluripotent stem cells (OCT4-A, Nanog) were analyzed. Results were represented as fold change values over untreated control values expressed as 1 after normalizing with the housekeeping gene. Nanog transcripts showed increased expression treated with FSH, bFGF or NT3. Treatment with FSH+NT3 or NT3+bFGF had a synergistic effect on the expression of Nanog. OCT4-A transcripts did not show increased expression when treated with FSH, bFGF or NT3 separately. Treatment with FSH+NT3 or bFGF+FSH had an impressive effect on the expression of OCT4-A (Table 2).
Expression of germ cell markers: Levels of transcripts for germ cell markers (c-KIT, VASA) were analyzed. Results were represented as fold change values over untreated control values expressed as 1 after normalizing with the housekeeping gene. c-KIT transcripts showed increased expression treated with FSH, bFGF or NT3 separately. Treatment with two or three agents together had no synergistic effect on expression of this marker. VASA transcripts showed increased expression when only treated with FSH+NT3 (Table 3).
Expression of transition specific markers: Levels of transcripts for transition specific markers (GDF-9 and Lhx-8) were analyzed. Results were represented as fold change values over untreated control values expressed as 1 after normalizing with housekeeping gene. Lhx transcript showed increased expression when only treated with bFGF. GDF transcript showed increased expression when only treated with FSH+NT3 (Table 4).
Discussion
Ovarian cortex culture for obtaining preantral follicles has been an appropriate technique to preserve fertility. Survival and activation of PFs can be promoted by many factors and hormones (20). Primordial to primary follicle transition has been shown to be regulated by many growth factors.
There are several factors associated with activation of PFs, like FGF-2 (21), NTs, and members of the TGF-β family, including activin and growth differentiation factor 9 (GDF-9) (22-24). Accordingly, in this study, human ovarian cortex culture was used for evaluating the effectiveness of FSH, bFGF, and NT3 in follicle growth.
Various studies have shown that PI3K pathway plays a key role in oocyte development (25). As a result, activation of this pathway by growth factors may help stem cells differentiation into oocytes in vitro; in fact, they are important in activation of PFs as well as transition of resting PF to primary follicles via the elevated expression of GDF-9 and Lhx-8. Various studies have shown that primary oocytes express GDF-9. Also, some studies demonstrated that GDF-9 has important role in early follicular growth and in oocyte to granulosa cell interaction (26). Activation of precursor follicles in vivo is a slow process in which GDF-9 and BMP-15 play role and also it involves the PI3K pathway (27, 28). In non-growing oocytes, the Lhx transcription factor blocks the activity of the PI3K pathway and keeps the oocyte in a quiescent state (29). Indirect interaction of Lhx-8 with PI3K pathway has also been proven in some studies (30). In addition, there was an enhanced expression of pluripotent stem cells, like OCT4-A and Nanog and germ cells specific markers, such as c-KIT and VASA. Our findings indicated the direct effect of FSH, bFGF, and NT3 on ovarian stem cells and also PF development.
The growth of germ cells needs several phases including up-and down-regulating the expression of specific genes (31). According to recent investigations, primordial germ cells and oocyte like cells are produced by embryonic, differentiated pluripotent and adult stem cells in vitro (32). Regarding mice, pigs, and humans, somatic stem cells could be differentiated to oocyte like cells by culturing under the influence of growth factors (33), estrogenic stimuli (32), follicular fluid, and gonadotropins (34).
PF transition from primary to secondary follicles occurs through 8 to 20 days following the culture of ovarian cortical tissue (6). Parte et al. showed that stem cells within the ovary are capable of differentiation to oocyte like cells (6). Isolated ovarian stem cells can express mRNAs for OCT4 as transcription factors involving in the maintenance of cell pluripotency. OCT4 has also been reported as a regulator of Nanog expression (35). Non-coding RNA transcription can be activated by OCT4 and Nanog and they increase stem cells self-renewal as well as inhibit differentiation (36). According to our results, the expression level of OCT4 in FSH+NT3 group and bFGF+NT3 group showed a significant increase while in other groups, no increase was observed. FSH, by binding to its receptor via the PI3K pathway, can activate AKT and β-catenin and ultimately increase OCT expression (37). The results of our study showed that treatment of cells with FSH or NT3 alone could not lead to a significant increase in OCT4 expression. As a result, FSH and NT may have a synergistic effect on OCT expression. Treatment of cells with bFGF+NT3 also resulted in a significant increase in the expression of pluripotency markers, although this increase was lower than the one in the group treated with FSH+NT3.
On the other hand, the expression of Nanog in all groups showed a significant increase and the highest expression was observed in the FSH+NT3 group. As our results show, treatment of cells with any of the growth factors of FSH, bFGF, and NT3 increases the expression of Nanog, while simultaneous treatment of cells with bFGF+NT3 or FSH+NT3 increases the expression of Nanog significantly producing a synergistic effect. Since the expression of Nanog in the two groups of FSH+NT3 and bFGF+NT3 is significantly higher than the group of FSH+bFGF and also the expression of Nanog is higher in NT3 group compared to the two groups of bFGF and FSH, it is suggested that NT3 has a greater effect on Nanog expression than bFGF and FSH.
Furthermore, treatment of cells with FSH+NT3 significantly increases the expression of VASA. VASA has been shown to be expressed in the post-migratory primordial germ cells until the post-meiotic stage of oocytes (38). Based on Dyce et al.’s results, VASA expression and differentiation of stem cells, obtained from skin, to OLCs is promoted by follicular fluid (39). VASA is a marker for germ cell differentiation among several species (40). Considering that OCT4 marker is also a marker of germ cells and its expression is significantly increased due to the treatment of cells with FSH+NT3, it can be concluded that the treatment of ovarian cortical cells with FSH+NT3 can be a good option for achieving germ cells in vitro. Unexpectedly, the expression of c-KIT as another germ cell marker, unlike the other two germ cell markers, did not increase in the FSH+NT3 group. Treatment of cells with FSH, NT3, and bFGF significantly increases the expression of this marker while FSH+bFGF, NT3+bFGF, and FSH+NT3+bFGF did not increase the expression of this marker compared to the control group. These results may suggest that the combination of growth factors is not a suitable method to induce the expression of germ cell markers in ovarian cortical stem cells, but other methods of cell treatment, including periodic treatment with these growth factors should be considered in next studies.
The expression of GDF-9 is specific for oocytes and is a marker for the identification of oocytes derived from stem cells (41).
Our study showed that these markers increased the influence of FSH+NT3 treatment. Silva et al. reported that five days of culture using FSH increased oocyte diameter in primary follicles (42). It has been shown that the expression of FSH receptor is developed progressively through the primordial-to-primary follicle transition (43) and also recent studies have reported the existence of FSH receptors in primary follicles (44), which indicates the essential role of FSH in oocyte development. Moreover, Roy and Albee (45) stated the crucial effect of FSH for somatic cells to early granulosa cell differentiation through initial PF growth. Our study shows that FSH alone does not increase the expression of GDF-9, but the combination of FSH with NT3 can significantly increase the expression of GDF-9. On the other hand, our results show that Lhx expression is significantly increased only in the FSH group and has no expression in the FSH+NT3 group.
In our study, FSH alone could increase transition markers including Lhx, pluripotent stem cells marker (Nanog), and early germ cell marker (c-KIT). c-Kit, a member of tyrosine kinase receptors, together with its ligand, KIT, plays considerable role in female reproductive system (46). Consistent with the results of Parte et al., Nanog expression was indirectly representative of existence of pluripotent stem cells within the surface epithelium of ovary (47) and increase in the expression of Lhx-8 indicated PFs transition into primary follicles, which is consistent with other relevant studies (48).
The stimulation of pluripotent stem cells as well as early germ cell specific markers via bFGF is not unusual, as bFGF receptors have shown to be available in OSE (10). Based on Kezele et al.’s (49) report, bFGF can stimulate several PFs to primary and secondary follicles among rat samples. Garor et al. (50) indicated the enhanced rate of PF in the bFGF-treated cortical tissue culture of the ovary.
In our study, bFGF alone could increase transition markers including Lhx, pluripotent stem cells marker (Nanog), and early germ cells marker (c-KIT). A combination of FSH and bFGF had no such effect.
FSH and bFGF could induce an elevation in OSE proliferation and also an increase in the expression of the pluripotent markers, including OCT4-A and Nanog in perimenopausal human ovarian sample devoid of PF (6). Matos et al. (8) reported that the combination of FSH+bFGF can increase PF growth and survival in goats and bFGF-neutralizing antibody could stop the change (51). Tang et al. showed that the combination of FSH and bFGF significantly increased the rates of primary and normal follicles and also decreased the apoptotic cell count in comparison with FSH alone (12).
However, Derrar et al. (52) demonstrated that bFGF+FSH, insulin-like growth factor 1 (IGF-I), epidermal growth factor (EGF), and transforming growth factor β (TGF-β) were unable to change PF activation or morphology of primary or secondary follicles in short-term cultures of the ovarian cortex in the cow. In our study, combination of FSH and bFGF did not have synergistic effect.
NT3 is associated with the progression of rat PFs to developing follicles in culture as well as upregulating the expression of 18 genes, such as the genes involved in the plasma membrane and cytoskeleton synthesis (48). NT3 is possibly linked to the stimulation of steroid hormone formation in the antral follicle (53). By culturing hamster follicles plus NT3, a two fold increase was observed in E2 release (53). Moreover, adding NT3 into the culture medium of bovine theca cells led to elevated cell proliferation (54). According to Dissen et al.’s study, by eliminating neurotrophin genes in knockout mice, follicle production and development are influenced (55). Nilsson et al. indicated that NT3 signaling probably occurs by paracrine/autocrine mechanism in PFs.
In our study, NT3 alone could increase pluripotent stem cells maker (Nanog) and and early germ cells marker (c-KIT). NT3 concomitant with FSH or bFGF could increase pluripotent stem cells markers (Nanog and OCT4-A).
NT3 with FSH or bFGF in normal circumstances probably facilitates PF to primary follicle transition. NT3 with FSH or bFGF is capable of stimulating ovarian stem cells and it may be effective in stem cells accumulation through developed cultures. It is worth studying whether the long-term culture of perimenopausal tissue of ovary can cause PF accumulation from the stem cells with NT3, FSH, or bFGF therapy. The results can provide the context for next investigations on the effect of other hormones and growth factors on activation of PFs in vitro. The culture of ovarian cortical tissue seems to be a useful technique for studying ovarian stem cell biology and also PF transition.
Conclusion
The obtained findings can be used for both oncotherapy and infertility treatment. Our results are helpful for in vitro production of human oocytes from ovarian stem cells. The function of produced oocytes in vitro is considered one of the main issues for investigators. Obviously, more studies are needed to reach the final result and achieve functional oocytes in vitro. In the future, more research should be done to clarify the molecular pathways involved in the process studied in this paper. It is also necessary to conduct further studies on the effect of other growth factors with different concentrations.
Acknowledgement
The authors would like to thank Dr. Mansoureh Soleimani for her sincere support and guidance.
Conflict of Interest
There is no conflict of interest.
Funding: This research was supported by a grant (No92-03-117-24106) from Iran University of Medical Sciences, Tehran, Iran.