Background

Gametes derived from pluripotent stem cells may provide potential reproductive options to individuals who are rendered infertile due to injuries, exposure to toxicants or immune-suppressive treatments, in cases with gonadal insufficiency due to premature ovarian failure or azoospermia, reproductive aging and idiopathic cases of poor gametes quality and IVF failure. These artificial gametes derived from stem cells may also serve as an invaluable model system to study both genetic and epigenetic programming of germ cells development in vivo and also help obtain better insights into causes for idiopathic cases of infertility. Premature ovarian failure (POF) is a heterogeneous disorder that occurs at the frequency of less than 1% in women less than 40 years of age. Besides genetic basis and autoimmune etiologies, POF is caused by surgical removal of ovaries for conditions such as severe endometriosis, cancer and also as a side effect of oncotherapy for various non-gynecological malignancies. Similarly, besides a genetic basis, azoospermia in men occurs as a side effect of oncotherapy or infections. The option to preserve fertility prior to oncotherapy by way of cryopreservation of gametes or embryos is not yet widely available in several countries and also not useful to young pre-pubertal cancer patients due to non-availability of gametes. Women willingly go through 6–7 failed IVF cycles with a hope to become pregnant. However, assisted reproductive technologies of IVF and ICSI fail to benefit 30% of couples diagnosed with unexplained infertility and in cases where patients are entirely devoid of viable gametes. Donor gametes or adoption are available options however, the urge to have one’s own biological child supersedes any other desire in life. Recent advances in the field of reproductive medicine are focused on exploiting pluripotent stem cells to differentiate into gametes with a hope to deal with infertility.

First human pluripotent embryonic stem (hES) cell lines were reported more than 15 years ago [1] but their induction into gametes remains highly inefficient till date. A recent 2014 Views and Reviews section in Fertility and Sterility was dedicated to stem cells, their differentiation into germ cells and the related efforts towards translation. To summarize it is still a long way before realizing clinical potential of stem cells to make gametes for reproductive medicine [2]. We encourage the readers to refer these publications for latest update in the field [3–7]. Our review provides an altogether a different perspective to overcome existing hurdles to obtain gametes from stem cells. We put forth our case in favor of VSELs as an alternative source of pluripotent stem cells to obtain gametes.

Pluripotent stem cells differentiation into gametes – recent advances

A careful review of published literature shows that a group from Japan, including Prof. Hayashi and Prof. Saitou has achieved major progress in the field of generating gametes from mouse pluripotent stem cells (mES/iPS cells). In 2011 they published in Cell that it is possible to obtain live pups from sperm derived from pluripotent stem cells (ES or iPS cells) [8]. In 2012 they published in Science that following a similar strategy, offspring are obtained from oocytes derived from ES or iPS cells [9]. In 2013, they have published their detailed protocols in Nature Protocols describing the method to generate eggs starting with mouse ES cells and iPS cells [10]. Basic reasoning that led to this remarkable success was that it is important to recapitulate in vitro what happens in vivo during early embryo development. Two main strategies that have been used in the past to induce germ cells from pluripotent stem cells (PSCs) include (i) spontaneous differentiation of PSCs to make embryoid bodies (EBs), isolate cells expressing germ cell markers for further manipulation and (ii) to use mouse epiblast stem cell lines to obtain germ cells. Both these approaches, although provide proof of concept that it may be possible to differentiate PSCs into germ cells, remain highly inefficient. Primordial germ cells (PGCs) are available in very few numbers and are relatively quiescent and thus the embryonic germ cell lines derived from them [11] have shown limited long-term proliferation potential [12]. Thus Hayashi’s group carried out experiments to first differentiate PSCs into epiblast-like cells and then induced them into PGC-like cells (PGCLCs). They demonstrate that once PGCLCs are obtained, it is possible to transplant them into testis/ovary to enable their further differentiation into sperm or oocytes respectively resulting in offspring. It is important to note that in both the publications, Hyashi et al. [8, 9] have reported existence of genetic anomalies in the offspring. When PSCs were induced to undergo spermatogenesis, some of the offspring underwent premature deaths because of tumors around the neck region. Similarly reduced number of pups were obtained from PSCs (3.9%) compared to those obtained by transplanting E12.5 PGCs (12.7%) or 3 weeks oocytes derived pups (17.3%). Almost half of the PSCs-derived oocytes failed to extrude second polar body resulting in 3PN zygotes. This is not surprising since extended cultures of ES/iPS cells are bound to result in the acquisition of genetic and epigenetic alterations during in vitro culture and parallel studies in humans remain a distant dream [2, 13]. Besides them, few other groups have also reported that PGCs have the ability to undergo gametogenesis when transplanted in adult tissues. Chuma et al. [14] transplanted PGCs in testis and obtained mature sperm whereas Matoba et al. [15] reported that PGCs isolated from E12.5 male fetus under the kidney capsule yield spermatids. Both these groups reported birth of healthy offspring. Similarly Matoba et al. [15] and Hashimoto et al. [16] reported that PGCs isolated from female fetus when transplanted under the ovarian bursa or kidney capsule result in functional eggs. It is intriguing to note that offspring born when starting with PGCs are normal compared to when starting with ES/iPS cells . Hayashi et al. [17] reviewed recent advances towards obtaining human gametes to treat infertility. They highlighted the existing hurdles in the existing differentiation protocols and discuss alternative use of germline stem cells (SSCs or OSCs) as a source to produce synthetic gametes (Figure 1). It may also be possible to obtain germ cells by transdifferentiation of somatic cells e.g. bone marrow and mesenchymal cells. Efforts are also ongoing to mature the primordial follicles in ovarian cortical tissue which are cryopreserved prior to cancer therapy.

Figure 1
figure 1

Left yellow panel depicts event that occur naturally. Right purple panel represents human efforts to make synthetic gametes. Fertilization of gametes results in a blastocyst with inner cell mass (ICM) which comprises of pluripotent cells (grown in vitro as ES cells) and further develops into a epiblast-stage embryo where specification into somatic cells and primordial germ cells (PGCs) occurs. PGCs are pluripotent, express nuclear OCT-4, differentiate into gonocytes in testes and primordial follicles in ovaries (please refer to the main text for greater details) and persist in adult gonads as pluripotent, nuclear OCT-4 positive VSELs. Thus in addition to SSCs and OSCs in testes and ovaries [42], VSELs also exist [48] as reviewed recently. VSELs self-renew and give rise to progenitors (SSCs in testis and OSCs in ovary) which undergo clonal expansion, meiosis and further differentiation into gametes. Solid blue arrows represent asymmetric cell division of VSELs [48]. Differentiation of ES and iPS cells into synthetic gametes is a distant dream as they do not efficiently differentiate into PGCs. VSELs and OSCs spontaneously differentiate into oocyte-like structures in vitro[43, 63, 74–76, 78, 79] as they are indeed PGCs that survive into adulthood. Limited success has been achieved using bone marrow [27–29], fetal skin [30] and mesenchymal cells [19]. However, on transplantation –these cells are able to colonize but differentiation remains inefficient. Recent success was reported by Hermann et al. [20] who obtained functional sperm after autologus SSCs transplantation in non-human primates which after IVF also resulted in the formation of blastocysts. However lot more work needs to be done before it can reach the clinic and for more reading in this area readers may refer to recent reviews [21, 22]. Tilly’s group has made significant contributions to the field of ovarian stem cells (OSCs) since their first landmark paper challenging the basic dogma that females are born with fixed number of eggs [23]. OSCs are localized in the ovary surface epithelium and can be isolated from the ovarian cortex, expanded in culture and later transplantation in adult mice - they differentiate into functional eggs and result in offspring [63]. We also noted that the 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 human OSE cultures along with characteristic expression of stem/germ cell/oocyte markers [46]. Time lapse imaging of develo** oocyte –like cells with distinctly moving cytoplasmic extensions have also been reported by Bukovsky’s group [43, 79].

The striking fact is the spontaneous nature of this kind of differentiation of VSELs into oocyte-like structures. No additional growth factors are added to the medium to induce differentiation of oocyte-like structures. It appears that the VSELs in the OSE scra**s are pre-programmed to differentiate into oocytes. This is indeed facilitated by the epithelial cells which form a bed of fibroblasts and were present in close association with the differentiating stem cells. Similarly the isolated OSCs also undergo spontaneous differentiation into oocytes in culture [25, 43, 80]. Parte et al. [81] have shown that ovarian cortical tissue slices besides being a source of primordial follicles are also an excellent source of stem cells that spontaneously differentiate into oocyte-like structures after 3 weeks culture. Evidently the reason for this spontaneous differentiation of VSELs into oocyte-like structures is that the VSELs closely resemble PGCs (Table 1 and Figure 1).

Conclusions

It may be possible to obtain human gametes provided efficient and directed differentiation of ES or iPS cells into PGCs is achieved. But this may not be mandatory since emerging literature suggests that PGCs persist as a sub-population of VSELs along with SSCs in testis and OSCs in ovary. Similar to the PGCs, VSELs are quiescent in nature, do not expand in culture like ES or iPS cells and throughout life serve as a backup pool and give rise to SSCs/OSCs which undergo clonal expansion, meiosis and further differentiation to produce haploid gametes. Ovarian VSELs respond to FSH via FSHR3 and spontaneously differentiate into oocyte-like structures in vitro during OSE culture. Similar in vitro culture studies are ongoing in our lab using testicular VSELs. More studies are required to further substantiate the potential of VSELs and their ability to differentiate into gametes. We propose that rather than the existing concept of in vitro differentiation of stem cells into oocytes and sperm for assisted reproduction, it would be ideal to manipulate VSELs that survive oncotherapy in vivo to achieve restoration of gonadal function (since they exist in menopausal/ POF ovary and also in azoospermic human testis).

In contrast to genetically affected offspring born from ES/iPS derived gametes, healthy offspring born starting with OSCs and the oocytes formed after in vitro spontaneous differentiation of ovarian stem cells show normal ploidy status. This is evidently because of the similar epigenetic status of PGCs and VSELs which is possibly difficult to be replicated in vitro while differentiating ES/iPS cells into PGCs (although some success has been achieved as described above). Scientific community needs to slow down, re-think and make efforts to exploit clinical potential of pluripotent stem cells (VSELs) and progenitors (SSCs and OSCs) which exist in the adult gonads as an alternate option to ES/iPS cells!

Key messages

  •  Current status of making gametes from pluripotent stem cells (ES and iPS) to help infertile couples is highly inefficient and still remains a distant dream

  •  Major obstacle in the field is apparently to establish protocols to obtain primordial germ cells (PGCs) from the pluripotent stem cells (ES and iPS) in vitro. PGCs are pre-programmed and hence easily and spontaneously differentiate into gametes

  •  Published literature is reviewed suggesting that this challenge of making gametes can be easily overcome since PGCs indeed survive in adult human ovaries and testes as very small embryonic-like stem cells (VSELs)

  •  VSELs are pluripotent stem cells (surviving PGCs) which exist as a sub-population localized in the adult ovary surface epithelium and in the basement membrane of seminiferous tubules in the testes. They are present in normal adult and aged testes and ovaries (including POF and menopausal ovaries). Moreover VSELs survive oncotherapy because of their quiescent nature.

  •  Three weeks culture (simple culture medium with no added growth factors) of ovary surface epithelial cells enriched with VSELs and ovary stem cells (OSCs) spontaneously differentiate into oocyte-like structures - because the gonadal VSELs (PGCs) and OSCs (arise from the VSELs) are pre-programmed to develop into gametes

  •  We propose that rather than manipulating gonadal VSELs (PGCs) in vitro, a better approach will be to manipulate them in vivo to give rise to functional gametes. This approach will give rise to autologus gametes, with no associated ethical/regulatory constraints and epigenetic/genetic issues may not exist by avoiding in vitro culture.

Authors’ information

DB is working on pluripotent stem cells for almost 11 years. IH is leading IVF expert and well understands the need of synthetic gametes by infertile couples. HP is a PhD student at NIRRH and RB works at Hinduja Hospital.