25.2.1 In Vitro Gametogenesis in Mice

PGCs are a scarce and transient entity, present only at the embryonic period thereby precluding the detailed biochemical analysis requiring a large number of cells. Induction of PGCs from pluripotent cells, such as mouse ESCs (mESCs) or iPSCs (miPSCs), therefore, have been awaited to overcome such hurdles. Many of such attempts have been done since early 2000, but none of them have successfully demonstrated the contribution to functional gametes or healthy offspring, likely due to the poor induction efficiency and/or the lack of appropriate markers to isolate and characterize the in vitro derivatives [13–16]. Notably, these studies utilized markers that have poor PGC specificity or are expressed only at a late stage, thereby failing to recapitulate the early PGC development and thus the following developmental trajectories leading to mature gametes. To reconstitute the perplexed and protracted germ cell developmental processes in vitro, faithful reconstitution of such processes in a stepwise manner is of paramount importance. By using epiblast cells harboring Blimp1-mVenus (BV), Stella-Ecfp (SC) reporters to visualize the formation of early PGCs, Ohinata et al. successfully induced the BVSC (+) PGC-like state from pre-gastrulating epiblast cells in a dish by adding BMP4 and other growth factors (SCF, EGF, LIF, BMP8b) (Fig. 25.2) [8]. These cells closely resembled PGCs in vivo in both gene expression and epigenome pattern. Upon transplantation into the seminiferous tubules of c-Kit mutant (W/Wv) mice depleted of endogenous germ cells, they differentiated into functional sperm and successfully contributed to the healthy offspring after intracytoplasmic sperm injection (ICSI). In the following study, Hayashi et al. successfully induced mouse primordial germ cell-like cells (mPGCLCs) in vitro from BVSC (+) naïve mESCs or miPSCs maintained in the presence of 2i (CHIR99021, PD0325901) + LIF [17]. In this study, they first cultivated naïve mESCs or miPSCs for 2 days in the presence of Activin-A and bFGF to induce epiblast-like cells (EpiLCs), which closely resemble germline competent E 5.75 pre-gastrulating epiblast. EpiLCs were further induced into mPGCLCs by BMP4-based cytokine cocktails as above used by Ohinata et al. Notably, mPGCLCs exhibit highly similar transcriptome and epigenome profiles with in vivo PGCs at ~E9.0 and contribute to the functional spermatogenesis upon transplantation into W/Wv mice. Again, healthy pups were generated through intracytoplasmic sperm injection (ICSI) of the mPGCLC-derived sperms and transferring the resultant zygotes to foster mothers.

In a subsequent study, Hayashi et al. similarly induced female mESCs into mPGCLCs in vitro (Fig. 25.2) [18]. By culturing with somatic cells obtained from embryonic ovaries (“reconstituted ovary” method), mPGCLCs further differentiated into the oocytes that were in prophase of the first meiotic division. Upon transplantation into ovaries of immune-deficient mice, mPGCLC-derived oocytes matured into fully grown oocytes, which, through in vitro maturation (IVM) and in vitro fertilization (IVF), became zygotes and contributed to fertile offspring [18]. Together, these studies for the first time provided convincing evidence that the mouse germline can be reconstituted in vitro by using pluripotent stem cells.

Notably, all of the above attempts of mouse in vitro gametogenesis relied on host niche environment for terminal maturation into oocytes or spermatozoa through the transplantation of mPGCLCs into recipient mice (Fig. 25.2) [8, 17, 18]. More recently, however, several groups have reported the successful reconstitution of the latter half of oogenesis in vitro. In these studies, authors divided the culture into three steps – in vitro differentiation (IVD), in vitro growth (IVG), and IVM according to the in vivo oogenesis pathway. By careful validation of each step using the reconstituted ovaries, authors successfully reconstituted the entire process of oogenesis in vitro and generated healthy offspring by IVF of the in vitro reconstituted eggs [19–21]. These studies provide a valuable platform for analyzing the mechanisms of oogenesis in vitro and will serve as a guide for future in vitro gametogenesis in humans.

In male testes, the presence of spermatogonial stem cells with self-renewal capabilities allow for production of terminally differentiated spermatozoa for a prolonged period of time [21]. In mice, long-term culture of cells with robust spermatogonial potential, named germline stem cells (GSCs), can be established from both neonatal and adult testes [22, 23]. Recently, Ishikura et al. successfully induced GSC-like cells (GSCLCs) from male mPGCLCs by culturing with gonadal somatic cells obtained from E12.5 male embryos (Fig. 25.2) [24]. Upon transplantation, GSCLCs were capable of colonizing in the adult testis and generated functional sperm, features characteristic of GSCs. Previously, Ogawa and colleagues successfully induced sperm from GSCs through organ culture of neonatal testes into which GSCs were transplanted before culture [25]. Combining these methods together, reconstitution of the entire process of spermatogenesis in vitro will be technically feasible and achieved in the near future.

25.3 The Development of Germline in Humans and Nonhuman Primates

Precise understanding of how human germline develops in vivo is the prerequisite information for human in vitro gametogenesis. However, until very recently, human germ cell development has been poorly understood due to ethical and technical constraints in analyzing a scarce population which is only present in embryos and fetuses. To date, our knowledge of human germ cell development has been mostly attributed to the morphologic description conducted in the early to mid-twentieth century.

One of such earliest descriptions dates back to nearly a century ago, when Fuss identified the cluster of cells with unique morphologic features suggestive of human PGCs (hPGCs) at the posterior aspect of the yolk sac at ~4 weeks post conception (wpc) of human embryos [26, 27]. Subsequent series of histological and ultrastructural evaluation provided further evidence that PGCs emigrate from the yolk sac endoderm and further migrate through the hindgut endoderm and dorsal mesentery until they reach the gonads at ~5 wpc [28–30]. Recently, several groups reported transcriptome and immunohistochemical profiles of hPGCs within the gonads [31–34]. Gonadal hPGCs show expression of germ cell markers (e.g., BLIMP1, TFAP2C, NANOS3, DND1, DDX4, DAZL) and pluripotency-associated genes (e.g., POU5F1, NANOG, SALL4, LIN28A, PRDM14) similar to mPGCs. Notable divergence, however, does exist between hPGCs and mPGCs. SOX2, a core transcription factor for pluripotency and critical for germ cell development in mPGCs, is not expressed in hPGCs, which instead shows strong and persistent expression of SOX17, previously known as an endodermal marker [32, 35]. SOX17, on the other hand, is only transiently expressed in early mPGCs but not in gonadal mPGCs [36]. Epigenetic features of hPGCs in particular genome-wide methylation has also been investigated over the last few years, which showed that global demethylation occurs in hPGCs at the migratory and early gonadal stage with their genome-wide 5-methylcytosine (5mC) levels reaching ~5% at 9 wpc, similar to mPGCs [31–33].

Although the above recent studies provide some insight into the biologic properties of gonadal hPGCs in vivo from 5 wpc and onward, how hPGCs develop and are specified in earlier phases is still a mystery due to our inaccessibility of such specimens. Of note, there are striking differences in the anatomical structures in pre-gastrulating embryos between mice and primates that include humans [37]. For example, EXE, a source of BMP4, which also plays an essential role in mPGC specification, is not present in early postimplantation primate embryos. On the other hand, the nascent amnion which emerges in early pre-gastrulating primate embryos dorsal to the epiblastic disc is not present at the pre-gastrulating mouse embryos (Figs. 25.1 and 25.3). Therefore, direct extrapolation of the signaling principles in the formation of mice pre-gastrulating embryos to those of primates is not possible. To overcome this hurdle, Sasaki et al. recently used cynomolgus monkey embryos as a surrogate model, which exhibit very similar early developmental processes to human embryos [38, 39]. By series of immunohistochemical evaluation in cynomolgus embryos at various stages, they found that the origin of cynomolgus PGCs (cyPGCs) are traced back to clusters of ~10 cells at the dorsal aspect of the nascent amnion in the pre-gastrulating embryos at E11, which thereafter move toward the posterior end and further into the yolk sac (visceral) endoderm by E17 (Fig. 25.3) [38]. These early cyPGCs express germ cell markers (BLIMP1, TFAP2C, and SOX17) and pluripotency-associated markers (POU5F1, NANOG, and PRDM14) but lack SOX2, similar to gonadal hPGCs. Interestingly, cyPGCs before E17 do not express DDX4 or DAZL, suggesting that these markers serve as a late (gonadal) PGC marker in both mice and primates. In situ hybridization on histologic sections further demonstrated the specific localization of BMP4, a key determinant of mPGC specification, and their downstream target genes, ID2 and MSX2 in the nascent amnion, suggesting that cyPGCs may be specified within the nascent amnion through the autocrine action of BMP4 signaling [38]. This surprising finding is at odds with the mode of mPGC specification, where mPGCs are originated from the posterior epiblast by receiving BMP4 signaling from EXE (Fig.25.1) and highlights the mechanistic divergence of PGC specification among species [7, 8]. Nonetheless, the study is based on static observation and therefore awaits further validation, such as the tracing of cyPGCs using a reporter system ex vivo. This will prove to be challenging even with the current developmental engineering technologies.

25.3.1 Induction of Human PGC-Like Cells in Humans

Reconstitution of the human germline in vitro using hESCs or hiPSCs has a tremendous impact on both basic research and clinical perspectives. Since human embryos before 4 wpc, which is when specification of hPGCs occurs, are inaccessible due to ethical reasons, human PGC-like cells (hPGCLCs) induced in vitro from pluripotent stem cells provide the indispensable tool in understanding the molecular mechanisms of hPGC specification. Additionally, the reconstitution of the human germline in vitro will also enable the molecular dissection of the critical events during the subsequent germline development including epigenetic reprogramming, genomic imprinting, X-chromosome reactivation, or meiosis through gain- or loss-of-function analysis which, in turn, provides critical insight into causes of infertility or birth defects as a consequence of abnormal germ cell development [2]. Moreover, creation of fertile eggs or sperms in a dish provides new opportunities for infertile couples possessing defective gametes. Despite such significance, human in vitro gametogenesis has several hurdles to be overcome, making it more challenging than its mouse counterpart. First, our knowledge of human germline in vivo, in particular, that of before 4 wpc is very limited, which makes it difficult to assure that hPGCLCs recapitulate hPGC development in vivo. Second, species specific differences exist in the property of ESCs/iPSCs between mice and humans. hESCs/hiPSCs exhibit “primed” pluripotency, resembling peri-gastrulating epiblasts in vivo, while mESCs/miPSCs show “naïve” pluripotency recapitulating the features characteristic of inner cell mass (ICM) [37]. Although mESCs/miPSCs are known to be germline competent, it has yet to be known whether hESCs/hiPSCs have such competency. Third, an appropriate assay system to test the functionality of germ cells is lacking in humans. Although with ethical and legal constraints, fertilization using hPGCLC-derived gametes and observing the successive embryo development may provide clues that such germ cells have a potential to create the embryonic and extraembryonic tissues, a hallmark of germline [2].

Despite such technical and ethical constraints, many laboratories have attempted the induction of human germline from pluripotent stem cells over the last two decades. These studies used various methods including random differentiation [40, 41], directed differentiation using BMP4 [42–44] or retinoic acids [45], or overexpression of DAZ gene family [43]. These studies however suffer from either low induction efficiency due to the failure to recapitulate hPGC development in vivo or inappropriately gauge the induction using markers that is not specific to or only detectable at the late stage of human germ cells (e.g., DDX4, DAZL, or meiotic markers).

More recently, Sasaki et al. employed a stepwise approach to reconstitute the first step of human PGC development (Fig. 25.4) [39]. To recapitulate early PGC development, they generated dual reporter hiPSCs harboring tdTomato or EGFP reporter genes under the promoter of BLIMP1 and AP2γ (also known as TFAP2C), respectively [BLIMP1-tdTomato (BT); AP2γ-EGFP (AG)], the earliest known markers in both mPGCs and cyPGCs [10, 38, 46] as well as in gonadal hPGCs in vivo [32, 47]. BTAG hiPSCs were maintained in feeder-free, defined conditions through multiple passages with stable karyotypes and a gene expression pattern consistent with “primed pluripotency.” Authors demonstrated that BTAG hiPSCs first differentiated into incipient mesoderm-like cells (iMeLCs) by WNT agonist, CHIR99021, and Activin-A could be further induced robustly into hPGCLCs by culturing in the presence of BMP4, SCF, EGF, and LIF, the similar growth factors used for mPGCLC induction [17, 39]. These cells strongly express both BT and AG fluorescence proteins with efficiency reaching 60% and exhibit a gene expression pattern similar to that of cyPGCs and hPGCs in vivo. Notably, however, in contrast to gonadal PGCs, hPGCLCs lack late germ cell markers such as DDX4 or DAZL that are only expressed after they migrate into the gonads. Further comparison of hPGCLCs with early cyPGCs (obtained from E13–E21 embryos) versus gonadal cyPGCs confirmed that hPGCLCs do in fact resemble early cyPGCs more than gonadal cyPGCs they have yet to undergo overt epigenetic reprogramming such as global demethylation [38, 39]. Sasaki et al. also identified highly specific cell surface markers, ITGA6 and EpCAM, which, when used in combination, enables isolation of hPGCLCs with nearly 99% accuracy from hPGCLC-induction culture using various hiPSCs, which do not harbor fluorescence reporter genes [39].

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

No related posts.

Stay updated, free articles. Join our Telegram channel

Join