Germ cells are highly specialized cells that form gametes (sperm and eggs), and they are the only cells within an organism that contribute genes to offspring. Due to the fact that the genetic information contained within germ cells is passed from generation to generation, the germ line is often thought of as immortal. Studies have revealed that germ cells are remarkably similar to pluripotent embryonic stem cells (ESCs). For example, there is a significant overlap in the gene expression profile between ESCs and primordial germ cells (PGCs), the founders of the germ cell lineage. In addition, pluripotent embryonic germ (EG) cell lines have been derived from mammalian PGCs. Secondly, a subset of testicular germ cell tumors, known as non-seminomas, often contain differentiated cells representative of all three germ layers, a definitive test of pluripotency. Lastly, recent results have demonstrated the ability of spermatogonial stem cells (SSCs) to de-differentiate into pluripotent ES-like cells, underscoring a unique relationship between the germ line and pluripotent cells that are present during the earliest stages of embryonic development. Here, we will review the factors that regulate the self-renewal and maintenance of male germline stem cells (GSCs) and discuss how these factors may allow us to manipulate the germ line to create pluripotent cells that could serve as a critical tool in cell replacement therapies and regenerative medicine.
SPECIFICATION OF PRIMORDIAL GERM CELLS AND FORMATION OF THE GONAD
In the mouse, germ cells are specified in the proximal epiblast as a consequence of Bmp4 signaling from the adjacent extraembryonic ectoderm (1). The primordial germ cells (PGCs) are first identified in the mouse at E6.25 days-post-coitum (dpc) as a cluster of cells expressing the early marker Blimp-1. At E7.5 dpc, PGCs have migrated into the proximal epiblast, where they express the germ cell specific marker gene Stella (Dppa3) and tissue non-specific alkaline phosphatase. PGCs increase in number from ∼40 at 7.2 dpc to 25 000 at E13.5 dpc. During this time, the PGCs migrate through the developing hindgut and mesentery to reach the urogenital ridge (UGR) at E10.5 dpc, where they ultimately give rise to the male or female germ cells (2). At this time, epigenetic changes begin to occur in PGCs, such as erasure of parental genomic imprints on both paternal and maternal alleles.
At E13.5 dpc, male PGCs enter into mitotic arrest and become pro-spermatogonia also known as gonocytes (3) and are organized into the seminiferous cords. Approximately 2 days post-partum, the gonocytes migrate to the basement membrane of the seminiferous tubules, re-enter the cell cycle and initiate self-renewing divisions as spermatogonial stem cells (SSCs). Although germ cells are destined to generate highly differentiated gametes, evidence suggests that at several stages during development germ cells, such as PGCs, are pluripotent and can give rise to multiple tissues (discussed below). Once male germ cells have initiated self-renewing divisions as SSCs, however, it appears that these cells are much more restricted in their differentiation potential, giving rise only to mature spermatozoa.
A NICHE FOR MALE GERMLINE STEM CELLS
Many adult stem cells lose the potential for continued self-renewal when removed from their normal microenvironment, or stem cell niche, suggesting an essential role for the niche in controlling stem cell behavior (4). Niches are customized to support stem cell maintenance and survival, as well as to provide the appropriate balance of stem and progenitor cells available for tissue maintenance and repair. Niches are able to exert such effects through a combination of both inter- and intra-cellular mechanisms, including support cells that secrete short-range signals that specify self-renewal, cell–cell and cell–extracellular matrix adhesion, as well as polarity and mechanical cues (reviewed in 5). Attachment to, orientation toward, and signaling from a support cells within the niche is a major contributing factor to regulating stem cell number, self-renewal and differentiation in tissues maintained by stem cell populations in mammals.
Drosophila melanogaster as a model for studying germline stem cell niches
Experiments in invertebrate model organisms, such as Caenorhabditis elegans and Drosophilamelanogaster (D. melanogaster), have provided a number of paradigms for how stem cell behavior is regulated by the local environment. In adult D. melanogaster testes, an average of eight GSCs lie at the apical tip in a ring closely surrounding a cluster of post-mitotic somatic cells called the hub (6). Male GSCs initiate the first self-renewing divisions during late embryogenesis and maintain production of sperm throughout life. GSCs were functionally identified as stem cells in situ using genetic marking strategies for inducible, site-specific recombination that allow generation of marked clones of wild-type cells at low frequency (7). Similar genetic strategies can be used to generate marked clones of homozygous mutant cells in an otherwise heterozygous animal to allow tests of cell autonomy and examination of the effects of mutations that cause early lethality in a whole animal (8).
Male GSCs in D. melanogaster divide with invariant asymmetry to generate one cell that remains next to the hub and retains stem cell identity and one cell, called a gonialblast, which is displaced away from the hub (6,9,10). The gonialblast and its progeny begin differentiation by undergoing four rounds of synchronous mitotic divisions, giving rise to a cluster of 16 spermatogonia. The spermatogonial divisions are comparable with the transit amplifying divisions described in blood, skin, intestinal epithelial and many other stem cell lineages. In D. melanogaster, cytokinesis is incomplete and the spermatogonia derived from a single gonialblast remain interconnected by cytoplasmic bridges. A similar strategy is apparently utilized by SSCs in mammals, as chains of interconnected spermatogonia are apparent in whole mount preparations (11).
Male GSCs in the D. melanogaster testis are flanked by a second population of stem cells: the somatic stem cells (SSCs). SSCs sit adjacent and just distal to the GSCs and extend narrow processes up past the GCSs to touch the apical hub. The SSCs also self-renew and give rise to a differentiating cell population: the somatic cyst cells. Two cyst cells enclose each gonialblast and continue to envelope its differentiating progeny throughout the process of spermatogenesis (10). The somatic cyst cells, which may be the functional equivalent of mammalian Sertoli cells, play a major role in ensuring spermatogonial differentiation and the transition from spermatogonia to spermatocyte (12–14).
The somatic hub cells at the apical tip of the testis are a critical component of the stem cell niche. Hub cells express the secreted factor Unpaired (Upd), which activates the Janus kinase (JAK)-Signal transducer and activator of transcription (STAT) pathway in adjacent stem cells to specify self-renewal (8,15). There are no obvious orthologues of Upd in vertebrate systems; however, minimal sequence homology exists between the Upd receptor, Domeless, and members of the IL-6 transmembrane receptor family, including the Leukemia inhibitory factor (LIF) receptor (16). Therefore, activation of the JAK-STAT pathway by Upd likely represents an ancestral cytokine signaling pathway (reviewed in 17).
Characterization and manipulation of mammalian SSCs
Spermatogenesis is an extremely complex process that can be separated into three phases: the proliferation of spermatogonia, meiotic divisions of spermatocytes, and the morphological changes that accompany spermiogenesis. These processes are controlled by both intrinsic determinants and extrinsic factors from the environment.
Like most stem cells, mammalian SSCs are quite rare: ∼0.03% of germ cells in the testis are SSCs (18). Although SSCs are known to reside along the basement membrane of the seminiferous tubules, there are still no unique biochemical markers or characteristics that can be used to conclusively identify SSCs in vivo. However, several markers of that appear to be restricted to the earliest spermatogonia have been identified, including the c-Ret receptor tyrosine kinase and the GDNF receptor, GFRα1 (19), the transcriptional repressor PLZF (20), Neurogenin3 (11) and GPR125 (21).
Two recent advances have revolutionized the study of germ cell-niche interactions during mammalian spermatogenesis: germ cell and somatic cell transplantation in murine testes (22–24) and the culturing of spermatogonia in vitro (25–27). In addition to providing clues to general mechanisms involved in the maintenance of stem cell populations, knowledge of the mechanisms that maintain SSC identity and self-renewal capacity will provide insight into causes of male infertility and the etiology of testicular germ cell tumors. Furthermore, the ability to isolate, culture, and transplant germline stem cells and spermatogonia provides the potential for germline gene therapy to correct heritable genetic lesions, including those that cause endocrine or metabolic abnormalities leading to infertility. Finally, the cryopreservation of sperm from adult cancer suffers could be extended to the preservation of SSCs from adult as well as prepubescent males. Both of these approaches serve as a protection against infertility and the accumulation of germline mutations due to anti-cancer treatments.
SSC function is assayed by transplanting testis cell suspensions into azoospermic recipients (22,28). The availability of a transplantation assay not only allows for the quantification of stem cells but also provides a way to compare the activity of SSCs from wild-type and mutant donors. Recent studies using transplantation assays have provided information regarding the timing of SSC proliferation during development (29), as well as identified several factors that control SSC behavior (20). The identification of factors that are expressed by SSCs has provided tools to facilitate the enrichment of SSCs by fluorescence activated cell sorting prior to transplantation and propagation in vitro (19,27,30,31). Surprisingly, SSCs of all mammalian species examined to date, including humans, can replicate in mouse seminiferous tubules following transplantation, indicating that the factors required for homing to the basement membrane of the seminiferous tubules and self-renewal of SSCs are likely conserved among mammalian species (reviewed in 32; 33,34 are 1° literature).
The availability of a long-term culture method for SSCs would allow for the manipulation of the SSC genome by gene targeting as well as transplantation as a strategy to treat infertile males. SSCs present several advantages over culture of embryonic stem (ES) or embryonic germ (EG) cells. Although all three lineages permit transmission of information through the germ line to create gain of function and loss of function transgenic mice, SSCs can be harvested from postnatal testes and only give rise to the germ cells, avoiding the potential for teratoma formation. However, recent studies have identified the presence of germ cells within adult mouse testes that have the ability to convert to pluripotent ES-like cells when cultured under defined conditions (21,35,36). If the same isolation and culture conditions could be applied to human SSCs, this would provide a potential source for patient-specific ES cells without the need for viral transduction (discussed below).
The mammalian spermatogonial stem cell niche
The stem cell niches that support male GSCs in D. melanogaster and SSCs in mammals are remarkably similar, which may be due to the highly conserved nature of spermatogenesis. Mammalian SSCs are located adjacent to the basement membrane along the periphery of the seminiferous tubules of the testis (37). Sertoli cells, somatic cells that also reside along the basement membrane in the mammalian seminiferous epithelium, are in intimate contact with the germ cells throughout all stages of spermatogenesis (19,23,24,38). Peritubular myoid cells surround the seminiferous tubules and are also in close proximity to SSCs. Therefore, both Sertoli cells and myoid cells are likely somatic components of the germline stem cell niche in the testis, supporting SSC self-renewal and/or survival. Recent studies have demonstrated that the distribution of undifferentiated spermatogonia, which likely includes SSCs, is non-random. These cells appear to preferentially localize close to the vascular network and interstitial cells, including Leydig cells, that are present between adjacent tubules (39). Therefore, these data suggest that SSCs, similar to hematopoietic stem cells and neural stem cells, may reside within or close to a vascular niche.
TERATOMAS, GERM CELLS AND PLURIPOTENT STEM CELLS
Teratomas are complex tumors containing (sometimes highly differentiated) structures derived from the three embryonic germ (EG) layers. Their primary occurrence in reproductive organs suggested a germline origin, but it was not until the 1960s that Leroy Stevens (40) first demonstrated that explantation of murine embryonic gonads to ectopic sites resulted in the formation of teratomas. This early work confirmed the suspicion that the founder cells of teratomas indeed reside in the gonad. In addition, it demonstrated for the first time the critical role of the tissue environment in the induction of these tumors. The quest for the identification of the cell that was responsible for teratoma formation ended with the discovery of embryonic carcinoma cells (EC cells) by Martin and Evans (41). EC cells are capable of multilineage differentiation, but these cells rarely contribute to the germ line in chimeras due to their aneuploid karyotype (42,43). With the experimental systems in place to generate teratomas from embryos and pluripotent stem cell lines from teratomas, it was only a small step to eliminate the tumor environment altogether and establish pluripotent ES cell lines through the direct explant-culture of blastocyst embryos (44,45).
ES cells demonstrate the ability to expand indefinitely in vitro while retaining the capacity to generate ectodermal, endodermal and mesodermal derivatives both in vitro and in vivo. The discovery of murine ES cells allowed the study of early embryonic development and mammalian cell differentiation as well as facilitated transgenic and knockout technologies. Subsequently, scientists speculated that if human ES cells could be derived with similar properties, the treatment of degenerative diseases would be revolutionized, as production of human ES cells would allow, in theory, the generation of any cell, tissue or even organ, ‘on demand’ in the laboratory. It was the groundbreaking work of Thomson et al. in 1998 (46,47) that led to the derivation of these long awaited ES cell lines, first from rhesus monkey and marmoset and subsequently from human blastocysts (48). In addition to ES cells, pluripotent stem cell lines have since been generated from other sources such as PGCs and neonatal- and adult testes. These cells share many important features with ES cells, but also differ on essential issues as discussed below.
Primordial germ cells and embryonic germ cells
EG cell lines can be derived from explanted PGCs recovered from embryos between E8.5 and E12.5 dpc by culturing the cells in a combination of basic fibroblast growth factor (bFGF), stem cell factor (SCF), and LIF (49–51). EG cells and ES cells are morphologically indistinguishable, and like ES cells, EG cells can generate derivatives of all three germ layers in vitro (52) and in vivo (50) and contribute to the germ line in chimeric mice. Interestingly, while bFGF and SCF are required for EG cell derivation, once established, the cells can be maintained under standard ES cell conditions (+LIF alone).
One significant difference between ES cells and EG cells is their imprint status. Imprinting is an intriguing type of epigenetic regulation of gene expression whereby genes are expressed in a parent-of-origin-dependent fashion, meaning that imprinted genes are expressed from either the maternal or paternal allele but almost never from both. DNA methylation plays an essential role in the regulation of imprinted gene expression. Imprinted genes are modified during gametogenesis upon sex determination at so-called differentially methylated regions. Allele methylation depends on the parental origin of the allele and affects gene expression, as imprinted methylation either leads to gene silencing or activation. Importantly, these methylation marks are maintained during cell division, and as such, imprinted epigenetic information is maintained in every single cell in an individual. In germ cells, however, imprints are erased and reset to assure that parent-appropriate imprints are transmitted to the next generation (53). The erasure of imprinted methylation marks starts when PGCs colonize the gonad, around E10.5 dpc (54). Thus, EG lines derived after E10.5 will be devoid of imprinted DNA methylation, which results in misexpression of imprinted genes. In contrast to ES cells, this can affect the developmental potential of EG cells resulting in skeletal misformation and fetal overgrowth defects in chimeric mice (55).
Pluripotent stem cells from spermatogonial stem cells
Induced pluripotency of germ cells in the testis was first reported in the early 1950–60s when Guthrie and Bresler showed that injection of metal salts induced testicular teratomas in mice (56,57). These experiments demonstrated that disruption of the testicular niche could induce the teratoma-forming potential of cells that had previously committed to gametogenesis. Recent experimental evidence underscored this notion when Kanatsu-Shinohara et al. (35) demonstrated that colonies of ES-like cells appeared spontaneously in cultures of cells from neonatal testes in growth factor conditions that allow the sustained self-renewal of SSCs. The developmental potential of these cells, designated multipotent germline stem cells (mGS cells), was very different from cells present in SSC cultures. SSCs are capable of restoring spermatogenesis upon transplantation into recipient testes in which endogenous germ cells have been ablated. In contrast, injection of mGS cells into the seminiferous tubules of recipient mice resulted in the formation of teratomas demonstrating evidence of somatic differentiation into derivatives of all three germ layers. In addition, blastocyst complementation studies using mGS cells demonstrated that these cells could form germline chimeras. This seminal study provided the intellectual and experimental basis for the idea that germ cells, when freed from the restrictions imposed by their niche, can give rise to pluripotent stem cell lines. A subsequent study reported that pluripotent stem cells could be derived from adult testes as well (36). Remarkably, the cells described in this study robustly contributed to a wide variety of tissues, as well as to the germ line, upon injection into blastocyst embryos, and yet these cells remained capable of restoring spermatogenesis when transplanted into recipient testes without the formation of teratomas. Pluripotency and spermatogenic potential are usually mutually exclusive, and since the cells originated from a mixed population of cells it is possible that the observed stem cell properties are a consequence of heterogeneity of the culture.
While both studies demonstrated that the mammalian testis is a source of pluripotent stem cells, the identity of the cell that could give rise to these cells was still unclear. Using GPR125, a G-protein coupled receptor expressed on spermatogonia, as a marker, Seandel et al. (21) demonstrated the germ cell origin of testicular pluripotent cells. GPR125 is expressed exclusively in spermatogonia and not in differentiating germ cells or somatic cells of the testis. Cultures of purified GPR125 positive cells could be maintained for prolonged periods of time without loss of SSC activity. Cells from these cultures generated multipotent adult spermatogonial-derived cells (MASCs) and could colonize and repopulate testes devoid of endogenous stem cells. This study provided the first identification of germ cells that could give rise to pluripotent stem cells and reported the proof-of-principle experiment that MASCs could be used to generate functional somatic tissues. Recent clonal marking of single mGS cells confirmed the germ cell origin of pluripotent stem cells and demonstrated again that gain of pluripotency is associated with loss of the SSC potential (58). The identification of SSCs as a source of pluripotent stem cell lines opens the door to important explorations into the molecular mechanism regulating stem cell conversion and pluripotency.
REGULATORS OF PLURIPOTENCY
Induction of the stem cell pluripotent state is likely the result of a dynamic interplay between extracellular (growth) factors and intracellular transcriptional regulators. A triad of three transcription factors play a critical role in the maintenance of the stem cell pluripotent state: Oct4, Nanog and Sox2 (59). These transcription factors form an auto-regulatory loop and stimulate each other’s expression (60). Oct4, Sox2 and Nanog jointly occupy promoter regions in their target genes, many of which are important regulators of early differentiation (60). The polycomb repressive complex 2 (PRC2) subunit Suz12 is also recruited to these promoter elements, suggesting that the Oct4, Sox2, Nanog transcriptional complex maintains pluripotency, in part by repressing genes that mediate stem cell differentiation through recruitment of the polycomb complex (61,62).
During germ cell development, the expression of Oct4, Sox2 and Nanog gradually wanes, and at birth these factors are present at almost undetectable levels. With the downregulation of Nanog, Sox2 and Oct4 expression, the frequency at which germ cells convert back to a pluripotent state also decreases, yet pluripotent stem cells derived from spermatogonia express high levels of these transcription factors (21,35,63). Since two of these factors (Oct4 and Sox2) have recently been shown to be critical mediators of the induction of pluripotency in somatic cells, it is tempting to hypothesize that conversion of germ cells to a pluripotent state is the result of culture-induced reactivation of the pluripotency transcription factors. Indeed cultures of SSCs express Oct4 at higher levels than freshly isolated cells (Y.-Y. Chou and N.G., unpublished data), suggesting that the in vitro culture conditions are sufficient to upregulate Oct4 expression. Alternatively, the in vitro culture of SSCs could select for high expressors of Oct4. As mGS cell cultures express relatively high levels of Sox2 RNA but not Sox2 protein, regulation of RNA translation may also play an important role in the conversion of germ cells to pluripotent ES-like cells (35,63). Exploration of the molecular mechanism of germ cell conversion to a pluripotent state may reveal important insights into the role of transcriptional regulators and the cellular environment in the induction of pluripotency.
GERM CELLS AND REGENERATIVE MEDICINE
Human ES cells hold the promise of patient-specific therapies for many degenerative diseases and provide new tools with which to study human development and disease progression. The excitement surrounding the utilization of human ES cells, however, is accompanied by controversy over the ethical issues surrounding the origin and methods of obtaining these cells. The recent discovery that pluripotent stem cell lines can be generated by the simple introduction of three to four genetic factors, including Oct4 and Sox2, into somatic cells, brought two revolutionary changes to the stem cell field (64–70). First, induced pluripotent stem cells (iPS cells) enable the generation of novel human pluripotent stem cell lines without the ethical- and budgetary constraints associated with human ES cell work. Secondly, the possibility to generate iPS cells from patients with complex genetic disorders allows the generation of novel developmental models to study diseases that currently cannot be recapitulated in animal models. In addition, iPS-based disease models can be used to identify and test new drugs (71). The ability to generate patient-specific iPS cells also largely eliminates concerns of immune rejection in transplantation therapies, opening new therapeutic avenues in regenerative medicine (72). However, the current methods of iPS cell derivation are not amenable to the patient clinic. iPS cell derivation requires the introduction of ectopic genetic elements into the somatic genome using viral vectors, which potentially introduce mutations at the integration site. In addition, two of the factors, Klf4 and c-Myc, are themselves proto-oncogenes, further increasing the risk that iPS cell-derivatives may form tumors upon transplantation.
Germ cell-derived pluripotent stem cells represent the best of both worlds, allowing the generation of patient-specific stem cells without the ethical hurdles associated with human ES cells and without direct manipulation of the stem cell genome. In addition, studying the conversion of germ cells into pluripotent stem cell lines may provide valuable insight into the role of the cellular microenvironment on the induction of the pluripotent stem cell state. Extracellular signals are known to control stem cell self-renewal and differentiation decisions, and the extracellular environment plays a critical role in the conversion of germ cells into pluripotent stem cells. Growth factor or small molecule enhancers of the germ cell conversion to ES-like cells may likewise augment the generation of iPS cells and may eliminate the need for one or more genetic factors.
N.G. is supported by grants from the NIH and the Harvard Stem Cell Institute. D.L.J. is supported by the Ellison Medical Foundation, the American Federation for Aging Research, the ACS and the NIH.
The authors would like to thank members of their labs for comments on the manuscript. We apologize to those colleagues whose data could not be referenced directly due to space constraints.
Conflict of Interest statement. None declared.