Given the clinical significance of POF in women, this study was designed to investigate the efficacy of using intra-ovarian transplantation of bone marrow-derived c-Kit+ cells in restoring ovarian function in rats experiencing experimentally-induced POF. To test this hypothesis, a combination of histopathological, IHC, gene expression, hormonal, and fertility assessments were used. Taken together, the results are in support of our hypothesis; to our knowledge, we are first to report positive therapeutic effects of enriched populations of c-Kit+ bone marrow-derived cells on the rejuvenation of ovarian tissue in a rat model of POF.
As a first step, we set out to generate a reliable rat model of POF. This was achieved through daily administrations of VCD, a chemical that is known to induce premature ovarian insufficiency in animal models [30, 31]. In a study by Mayer and colleagues, various doses of VCD were used to induce the menopausal model in mice (80–320 mg/kg/day for 15 days), where 160 mg/kg/day for 15 days was reported as an optimal dose for the induction of ovarian failure [29, 32]. In the present study, we used this latter dose to achieve the depletion of follicles in rats. Our findings confirm that VCD injections can induce ovarian fibrosis and cause atresia of follicles at all stages of their development in rats. It has been suggested that this of VCD action occurs through its effect on the KIT-KITLG and apoptotic signaling pathways, leading to ovarian deficiency [33, 32]. Our results are consistent with previous studies that reported VCD as being capable of reducing the number of ovarian follicles, especially in their earlier and later stages [32, 34, 30].
As a next step, we used MACS to isolate bone marrow-derived which were then confirmed to be highly enriched for c-Kit+ cells. These cells along with c-Kit− cells and saline were used for intra-ovarian injections in different groups of VCD-treated rats.
In the present study, the administration of c-Kit+ cells, but not c-Kit− cells or saline, restored ovarian competence compared with the untreated VCD rats. In our rat model of menopause, restoration of ovarian competence was defined as an increased relative number of primary, secondary, and antral follicles. Our results also showed that bone marrow-derived cells, including c-Kit− but especially c-Kit+, have the potential to slow down aberrant remodeling in VCD-treated ovarian tissue. Similarly, although intra-ovarian injection of c-Kit− cells improved the VCD-induced angiogenesis deficiency, transplantation of c-Kit+ cells restored the vascularization levels to a greater degree, which were ultimately comparable to the control values.
In line with these observations, our results confirmed that the expression of Angpt2 and KDR genes were altered in VCD groups; Angpt2 expression was likely elevated in response to the VCD-induced inflammatory and degenerative changes occurring within the ovarian tissues [35, 36], but it is not clear why KDR expression had decreased. It may be reasonable to hypothesize that the severe inflammation had promoted angiogenic signaling through the up-regulation of the pro-angiogenic factor Angpt2, which then led to a compensatory down-regulation of its relevant receptor (KDR) gene. In either case, levels of Angpt2 and KDR expression returned to their normal range by 8 wk post-transplantation, which would correlate with the reduction of inflammation responses by this time. We also noted that VCD injections contributed to the weight loss recorded during the first 5 days, which was then followed by a gradual pattern of weight gain that continued until the end of the experiment. We assume this to be caused by the effects of VCD being triggered during drug administration, which would later have been metabolized via renal and hepatic routes. Consistent with our study, bone marrow-derived c-kit/Sca-1 cells have been shown to restore oocyte numbers in mice sterilized by chemotherapy, further demonstrating their restorative potential. In their study, Johnson and co-workers highlighted the up-regulation of genes associated with germline markers such as Dazl, Fragilis, Stella, Nobox, etc [37]. In support for the regenerative role of bone marrow c-kit cells, we showed that the reduction of fibrosis and induction of angiogenesis could be an effective strategy in the restoration of follicles.
The number of offspring born from the VCD & c-Kit+ rats was significantly greater than those in the VCD & c-Kit− group and was in the range of control rats. In support of this finding, Takehara and colleagues used adipose-derived stem cells, bone marrow mesenchymal stem cells, and tail fibroblast cells in their treatment of rats with cyclophosphamide-induced POF. According to their results, rats treated with bone marrow stem cells and tail fibroblasts exhibited significantly greater offspring numbers compared with POF rats left untreated [18]. Similarly, Leia and colleagues reported that skin-derived stem cells have the potential to increase the number of offspring born to mice experiencing busulfan-induced menopause [19].
Finally, we monitored the serum levels of FSH in rats on the first day after receiving their final dose of VCD, as well as 8 weeks after cell transplantation. Neither pre- nor post-transplantation FSH data differed among the groups, which was rather unexpected given that VCD treated rats showed a significant reduction of antral follicles, the only follicle which produces enough estradiol to regulate FSH through a feedback system. The reasons for this lack of change are unclear, but could result from the low number of animals (three) tested per time point, which may not have been sufficient to allow detection of changes in a pulsatile hormone such as FSH. It could also be due to the short experimental period of our study, which lasted only 8 weeks, since it was previously reported that hormonal alterations reach significant values at 240 days post-induction of POF [30]. In another study, however, VCD injection in mice resulted in the elevation of FSH hormone only 37 days after induction, with significant differences compared with the control [29]. Therefor, according to our hormonal and histological findings, it is conceivable that hormonal examination would require a greater number of samples per time point, in addition to longer intervals than the time allotted in the present study. Future studies will be necessary to determine whether FSH should serve as a viable biomarker for this disease model and/or its therapies, since our results suggest that it should not be used to measure the restoration of ovarian tissue following c-Kit+ cells transplantation.