The size of the resting follicle pool determines the ovarian reserve of an individual and correlates with graft sustainment after restoration of fertility by ovarian tissue cryopreservation and transplantation. Assessment of ovarian reserve following ovarian tissue transplantation is challenging when using established markers and indicates high variability between patients . Besides age of the donor  and ongoing recruitment of follicles into the maturation pathway, it is challenged by external influences such as previous transportation , cryopreservation procedure and grafting conditions . Xenografting to ovarectomized versus intact animals has previously been considered to be beneficial for graft survival and growth due to elevated gonadotropin levels following interruption of the inhibitory pituitary feedback loop [25, 16]. Gonadotrophin treatment has been variously studied in its effect on cytoplasmic maturation and oocyte developmental competence in different species including pig , mouse  and the monkey model . There are anecdotal reports on a possible effect of FSH on cyctoplasmic maturation in human oocytes , but the question of FSH effects on the resting follicle pool in human ovarian tissues has rarely been addressed systematically. Oktay has described FSH influence on follicular growth at earlier stages than previously suspected, although he confirmed independence of primordial follicle growth inititation from gonadotrophins . The value of FSH stimulation for obtaining pregnancy in patients with autologous transplantation of ovarian tissue is still a matter of debate .
Enhanced follicular depletion through FSH effects during ovarian stimulation has anecdotally been reported in the human , but controlled experimental studies on the effect of FSH on the resting follicle pool are still missing. According to our results in the present study, initial loss of primordial and primary follicles after grafting of fresh or frozen-thawed tissues is rescued at least in parts by FSH treatment starting within the 2nd week of transplantation. In freshly but not in frozen-thawed grafted tissues, a positive effect of FSH treatment on primordial follicle number was also observed if the start of treatment was postponed to 3 weeks post transplantation. As we have previously documented in the marmoset , and others in different species [14, 33, 34], cryopreservation applying a slow-freeze protocol significantly reduces normal follicular morphology in the resting follicle pool. In the present paper, we have also assessed the dramatic initial follicular loss following xenografting of frozen-thawed tissue. It is assumed that tissue damages occurring during cryopreservation and the initial avascular period after grafting are masked by a more receptive host milieu in FSH treated tissues. This is also supported by higher numbers of normal follicular morphology following FSH treatment in frozen-thawed as well as fresh tissues at all different points in time. A positive effect of FSH on follicle survival in frozen-thawed ovarian tissue applying an in vitro co-culture system had previously been described for other non-human primates . In human ovarian tissue that was cryopreserved and xenografted to the murine back muscle, earlier beginning of vessel formation had been demonstrated following FSH stimulation , thus enhancing the supply with nutritional and differentiating factors. Interestingly, FSH receptor expression had been described for tumor blood vessels in various organs , and for ovarian surface epithelium , suggesting that its effect on follicular morphology is rather an indirect one and not transmitted through granulosa cells as the classical target of FSH action.
When started after a prolonged grafting period of frozen tissues, FSH negatively influences primordial follicle numbers that are already low and most likely at this point reflect manifestation of preceding tissue damages due to cryopreservation. Since primary follicle numbers are increased in these tissues, the reduction in the primordial follicle pool may additionally mirror the beginning of follicular recruitment becoming apparent in already low primordial follicle numbers. However in fresh tissues, baseline values of follicle numbers are higher also after a prolonged grafting period, and FSH treatment following early or at a later point in time after transplantation prevents primordial follicle loss. This is in agreement with viability studies on fresh and frozen ovarian tissue resulting in significantly reduced viability when applying a slow-freezing protocol .
Proportional analysis of primary and primordial follicle number can indicate initiation of follicular growth, and for xenografted human ovarian tissue the influence of cryopreservation  and timing of FSH stimulation  on primordial follicle activation has been speculated upon. However, in this study we have not applied any staining techniques as a definitive marker of proliferation, and higher percentages of primary follicles over all un-advanced follicles in FSH-treated freshly-grafted tissues cannot be clearly assigned to a shift from the primordial to the primary follicle compartment. From a clinical point of view this may be of outstanding importance for the efficacy of ovarian tissue grafting for fertility restoration as the majority of stored human tissues today have most likely been treated according to a slow freezing protocol, and lower chances of pregnancy from these tissues compared with fresh or vitrified tissues are disputed . We have previously discussed the impact of cryopreservation per se on follicle pool dynamics and primary follicle activation in adult and prepubertal ovarian tissues . This could rather indicate absence of inhibitory signals deriving from the stromal cells due to cryodamages than an increase in activating factors . As the present study has not included in depth analysis of factors potentially involved in follicular growth and activation, involvement remains speculative. We can only assume possible secondary effects, e.g. via recruitment of antral follicles, as these were too small in number to be analyzed statistically between FSH-treated and untreated tissues. In view of an increasing clinical application of ovarian tissue grafting based on previously cryopreserved human tissues, these results may further stimulate research on intraovarian factors involved in the regulation of the resting follicle pool and positively influence clinical outcomes.