VSELs may obviate cryobanking of gonadal tissue in cancer patients for fertility preservation

More than a decade ago, Professor Jonathan Tilly from Harvard University, USA challenged the existing dogma that a female is born with a fixed number of eggs at birth; rather he suggested that a continuous replenishment of follicles occurs throughout life [3]. Several groups were drawn to this field of research and evidence was generated that stem cells are housed in adult mammalian ovary surface epithelium (human, sheep, marmoset, rabbit, and mouse) as well as in women with premature ovarian failure (POF) and menopause [4‐8]. Presence of stem cells in adult ovary is still not well accepted [9]; however Woods and Tilly [10] have published protocols to isolate ovarian stem cells (OSCs) from human ovaries and equated them to spermatogonial stem cells in testis [11]. Zhang et al. [12] recently described their inability to sort human OSCs using DDX4 as a cell surface marker by compiling together independent work done in four different laboratories to which Woods and Tilly have responded [13]. Silvestris et al. [14] successfully isolated DDX4 positive OSCs and showed further that the DDX4 positive OSCs are also positive for OCT-4. We routinely isolate VSELs and OSCs by mechanical scraping of ovary surface epithelium or by enzymatic method and use flow cytometry only as a tool to further characterize the stem cells [15].

Interesting data has been generated showing that the stem cells in adult ovary can spontaneously differentiate into oocyte-like structures in vitro [7, 16, 17] and on injection in human cortical biopsies lead to primordial follicle assembly [18]. A careful examination of the OSE cells smears showed the presence of small (2–5μm), round cells which expressed pluripotent markers SSEA-4, Oct-4, Nanog, Sox-2, and c-kit [5, 6, 19]. Using a polyclonal OCT-4 antibody, Parte et al. [7] for the first time documented the presence of two distinct populations of stem cells in OSE smears including very small stem cells (earlier described by Viran-Klun’s group as well) expressing nuclear OCT-4 and cell surface SSEA-4 termed very small embryonic-like stem cells (VSELs) and slightly bigger stem cells termed ovarian germ stem cells (OGSCs) with cytoplasmic OCT-4. Based on size difference between VSELs (2–5 μm) and OGSCs (>8–10 μm), we speculate that Tilly’s group detected the OGSCs (OSCs) and missed out on the VSELs in adult ovary because of very small size and low abundance of VSELs. Protocols to isolate and characterize both (VSELs and OGSCs) the stem cell populations in adult mammalian ovaries were recently published [15, 20]. Follicle stimulating hormone (FSH) exerts direct action on the ovarian VSELs (express FSH receptors) to undergo self-renewal, clonal expansion to form germ cell nests and further oocyte differentiation [21]. We also discussed that menopause occurs despite high circulatory FSH due to a compromised niche that is unable to support stem cells differentiation into follicles [22]. Stem cells from menopausal ovary spontaneously differentiate into oocyte-like structures when the inhibitory factors are overcome in vitro [5, 7]. Similarly stem cells from aged mouse ovaries differentiate and give rise to oocytes on being transplanted into a young somatic environment [23]. Similar VSELs were earlier reported by our group in adult human testis as a sub-group among spermatogonial stem cells (SSCs) on the basis of size and nuclear versus cytoplasmic staining of OCT-4. This was established through extensive characterization by immunolocalization using 3 different OCT-4 antibodies, qRT-PCR studies, in- situ hybridization and Western analysis [24]. VSELs have also been extensively characterized in adult mouse testis [25].

To conclude this section, both ovary and testis harbor pluripotent VSELs along with the specific progenitors which include OSCs in the ovary and SSCs in the testis.

The VSELs were first reported by Ratajczak’s group [26] in various adult mouse organs including testis and they postulate that pluripotent primordial germ cells (PGCs) during their migration along the dorsal mesentery towards the gonadal ridge to form the germ cells, migrate and settle in various adult organs throughout life [27]. The work became controversial recently when a leading stem cell biologist was unable to detect VSELs in mouse bone marrow [28], but the underlying reasons for their failure were technical as discussed by Ratajczak’s group [29, 30]. We have recently confirmed and extensively characterized VSELs in human cord blood [31]. It is probably because of the very small size, low abundance and minimal cytoplasm that VSELs have remained obscure till now. When cord blood/bone marrow is subjected to Ficoll-Hypaque centrifugation – the VSELs settle down with red blood cells and have been invariably discarded unknowingly in the past [32]. In the gonads, it is relatively easier to conceptualize that a small number of PGCs survive in adult gonads as VSELs and this has also been suggested by other group [33]. Ratajczak’s group have shown that VSELs are relatively quiescent and when mice are subjected to total body irradiation, bone marrow gets depleted of hematopoietic stem cells whereas the VSELs survive and show increased uptake of BrdU [34]. Shin et al. [35] reported that quiescent state of VSELs is because of unique DNA methylation pattern of developmentally crucial imprinted-genes showing hypomethylation/erasure of imprints in paternally methylated genes and hypermethylation of imprints in maternally methylated ones. As a result VSELs express increased levels of H19and Cdkn1c and lowered levels of Igf2 and Rasgrf1accounting for their quiescence.

Available literature suggests that all renewing body organs including skin, hair follicle, gut epithelium, hematopoietic system harbor two populations of stem cells which include quiescent and actively dividing stem cells [36, 37]. Based on these published literature, we decided to study VSELs in mouse ovary and testis to gauge the effect of busulphan and cyclophosphamide treatment on them [8, 25]. Besides immuno-localization and qRT-PCR analysis to show presence of pluripotent VSELs, we were able to quantitate them by flow cytometry as <6 μm sized cells which are LIN-/CD45-/SCA-1+. Results shown in Table 1 demonstrate that VSELs exist in normal gonads, survive chemotherapy and undergo self-renewal in response to PMSG treatment.

OCT-4 positive stem cells were also detected by flow cytometry in OSE cells from sheep ovaries (VSELs 1.26 ± 0.19 % of 2–4μm and OGSCs 6.86 ± 0.5 % of 4–9μm). Parte et al. [38] showed that a large number of stem cells are shed on cell culture inserts during both marmoset and human cortical tissue culture and these stem cells undergo spontaneous differentiation into oocyte-like structures in vitro. Presence of germ cell nests, Balbiani body-like structures and cytoplasmic streaming extensively described during fetal ovary development, are indeed well recapitulated during in vitro oogenesis in adult OSE cultures along with characteristic expression of stem/germ cell/oocyte markers [13]. Similar VSELs have also been reported in testicular biopsy collected from azoospermic cancer survivors [39] and also in POF ovaries [5, 6].

Spontaneous paternity and live birth following bone marrow conditioning and transplantation has surprised everyone. Ignatove et al. [40] reviewed 8 such cases and reported two more themselves, where although the patients were azoospermic, testicular sperm helped attain biological parenthood. Dupont et al. [41] recently reported spontaneous conception after autologus BMT. A 20 years old man suffering from acute promyelocytic leukemia, even though was repeatedly diagnosed azoospermic with zero sperm count fathered healthy children spontaneously [41]. The underlying reason is not yet clear but it is likely that there may be a stem cell link for fertility restoration in men similar to that discussed by Oktay et al. [42] in women (please refer to next section).

Transplanting germ cells from cryopreserved testicular tissue still remains experimental and has not been extensively studied in men as yet. Hermann et al. [43] have reported initial success and carried out SSCs transplantation in 18 adult and 5 pre-pubertal busulphan treated Rhesus monkeys through ultrasound guided transfer via rete testis. Sperm collected from these monkeys were injected in 85 oocytes by ICSI, 81 fertilized and 4 cells to blastocyst stage embryos were obtained. 7/81 embryos were of donor origin. Struijk et al. [44] critically reviewed the progress made to restore fertility in survivors of childhood cancer by auto- transplanting SSCs and concluded that transplanted cells colonize but do not differentiate. Navaeian-Kalat et al. [45] reviewed available literature and suggested that trans-differentiation of mesenchymal cells is the only way to treat infertility. Several attempts have been reported to restore spermatogenesis in chemoablated testis by transplanting mesenchymal cells as shown in Table 2. Different doses of mesenchymal cells (and isolated from different sources) have been injected directly into the seminiferous tubules via rete testis or efferent ducts, intravenous injections or direct inter-tubular injections into the testis. It has been postulated that mesenchymal cells are ‘pluripotent’ and can trans-differentiate into gametes. But this is possibly not true [57] and our results suggest that mesenchymal cells only provide a niche (source of growth factors) to the VSELs that survive chemotherapy. Rather than injections into the rete testis or efferent ducts of the seminiferous tubules (challenging technically and not a practical solution), inter-tubular injections of Sertoli cells or mesenchymal cells as a simple OPD procedure may suffice to restore spermatogenesis – based on our studies in mice [25].

A total of 37 births have occurred till now by transplanting cryopreserved cortical tissue in cancer survivors [58]. Recently, Demeestere et al. [59] report spontaneous pregnancy and live-birth in a 27 years old survivor from cryopreserved ovarian cortical tissue which was cryopreserved as she suffered from sickle cell anemia at the age of 5 years. Four thawed ovarian fragments were grafted on the residual left ovary, six were grafted in the right peritoneal bursa, and five were grafted subcutaneously. Normal menstruation along with normal FSH and undetectable AMH levels was maintained after transplantation and the woman became pregnant spontaneously after more than 2 years of transplantation of cortical tissue. They reported that cortical transplantation led to successful recovery of ovarian function. Since other reports have shown spontaneous recovery of atretic ovary even without direct ovarian transplantation, it is almost impossible to reason out source of oocytes that gave rise to the babies - whether it is cortical tissue or the ‘regenerated’ non-functional ovary. Four years ago, Oktay and his team reported 4 spontaneous pregnancies and 3 live births after subcutaneous transplantation of cryopreserved ovarian tissue [42] in a 28 years old survivor of Hodgkin’s disease. Origin of the pregnancy was proposed to be due to improvement of function of in situ ‘menopausal’ ovary by hormones secreted by transplanted cortical tissue [60]. Oktay et al. [61] discussed a possible stem cell link to spontaneous pregnancy. Besides destroying follicles, oncotherapy also destroys the somatic niche and transplantation of cortical tissue provides ‘regenerative signals’ to intact non-functional ovary enabling oocyte formation from stem cells (source could be bone marrow) and hence pregnancy and live births.

Authors propose that it could be the surviving VSELs in non-functional ovary that become functional and undergo neo-oogenesis and primordial follicle assembly and these follicles then undergo normal maturation and ovulation resulting in spontaneous pregnancy and live births. These views are strongly supported by a recent study by our group on chemoablated mice ovaries wherein Sriraman et al. [8] found the VSELs survive chemotherapy, are located in the OSE and are functional since they self-renew in response to PMSG treatment (Table 1). Culture of intact chemoablated ovary in presence of FSH resulted in appearance of proliferating germ cell clusters in situ that incorporated BrdU and expressed MVH and STRA8. When enzymatically isolated OSE in vitro were cultured, germ cell nests that co-expressed PCNA and OCT-4 were observed. Prolonged culture showed differentiated GDF-9 and MVH expressing oocyte-like structures in vitro. The study underwent a very strict review and we were advised to publish 4 images of germ cell nests (in the Supplement) to convince the readers as it is very significant finding that postnatal ovary even after chemoablation retains stem cells which can form germ cell nests. These results are in contrast to those published by Lei and Spradling [62] who earlier failed to detect any evidence for presence of stem cells in adult ovary and we had argued that absence of evidence does not necessarily imply evidence for absence [63]. It is interesting to point out that (i) Lee et al. [64] showed that although bone marrow transplantation could rescue long-term fertility in chemoablated mice- but all the offspring were of recipient germline (ii) Niikura et al. [23] reported that stem cells in aged ovary can form follicles when exposed to a young niche of the host (iii) Edessy et al. [65] transplanted autologus mesenchymal cells isolated from bone marrow into the POF ovary resulted in resumption of menstruation and secretory changes in the endometrium and (iv) Zhu et al. [66] reported beneficial effect of transplanting human cord blood mesenchymal cells in rat chemoablated ovary. Table 3 shows various studies demonstrating beneficial effect of transplanting mesenchymal cells in chemoablated ovary associated with live births.

Just as in men, spontaneous pregnancy and birth of babies has been reported after BMT in women. Veitia et al. [76] observed spontaneous menstruation followed by pregnancy in a woman after 8 years of BMT, although she was expected to be infertile after suffering from Fanconi anemia and aggressive BMT conditioning regimen. Genetic analysis of the mother (patient), daughter and the donor revealed genetic relationship between the mother and daughter. As a result they ruled out that transplanted bone marrow was a source of oocytes. Similarly Nabhan et al. [77] also confirmed genetic relationship between the transplanted mother and daughter and excluded the possibility of germ cells transmission from the donor. We postulate that the transplanted bone marrow only helps to improve the niche (is not a source of germ cells) and endogenous VSELs differentiate into oocytes resulting in spontaneous birth of babies.