NOTCH signaling in Sertoli cells regulates gonocyte fate

Abstract

Gonocytes (or prospermatogonia) are the precursors to spermatogonial stem cells (SSCs), which provide the foundation for spermatogenesis through their ability to both self-renew and generate daughter cells. Despite their relative importance, the regulatory mechanisms that govern gonocyte maintenance and transition to SSCs are poorly understood. Recently, we reported that constitutive activation of NOTCH1 signaling in Sertoli cells causes gonocyte exit from quiescence—the first suggestion of the potential role of this signaling pathway in the testis.

This Extra View will review what is known about NOTCH signaling, particularly in Sertoli cells and germ cells in the testes, by providing a background on germ cell biology and a summary of our recently published data on NOTCH1 signaling in Sertoli cells. We also describe additional data showing that aberrant proliferation and differentiation of gonocytes in response to constitutive activation of NOTCH1 signaling in Sertoli cells involves de novo expression of cell cycle proteins and a marked upregulation of the KIT receptor. These data further suggest that NOTCH signaling orchestrates a dynamic balance between maintenance and differentiation of gonocytes in the perinatal testis.

Keywords: :

• gonocyte
• Sertoli cell
• NOTCH signaling
• CCND1
• testis development
• mitotic arrest

Introduction

A significant percentage of young men are infertile owing to a severe or complete lack of germ cells, and for the majority of these men, the underlying causes of the germ cell deficit remain unknown.^1-3 In addition, the most common cancers affecting young men are testicular cancers of germ cell origin.⁴ Despite recent progress, many aspects of mammalian germ cell proliferation and differentiation and the relationship between germ cells and their somatic niche are not understood. Sertoli cells provide for the germ cell microenvironment within the developing and adult testis by delivering growth factors and other signaling molecules that coordinate the precise development of germ cells throughout life. One such signal, previously identified in the testis of Caenorhabditis elegans, is the protein LAG-2, which is produced by the distal tip cell (a somatic cell equivalent to the vertebrate Sertoli cell), which forms the germline stem cell niche in C. elegans. LAG-2 associates with the GLP-1/NOTCH receptor at the surface of germ cells, and this signal maintains the germ cells in an undifferentiated mitotic state.⁵ Lack of interaction between the distal tip cell and germ cells induces germ cell differentiation, meiosis, and gametogenesis. Therefore, in C. elegans, NOTCH signaling directly controls germline stem cell development and maintenance. In Drosophila, male germline stem cell maintenance depends not on NOTCH signaling, but rather on Unpaired (Upd) signals delivered by a group of somatic cells, called the hub, that form the niche.⁶ However, it has recently become apparent that in Drosophila, NOTCH signaling is crucial for niche cell specification.^7,8 Because evolutionarily conserved intracellular signaling mechanisms (such as NOTCH) are often employed to regulate germline niche formation and germ cell development, NOTCH signaling might also be employed in mammalian Sertoli cells to specify the fate of germ cells during embryonic development and after birth.

Here, we review what is currently known about germ cell development and the NOTCH signaling pathway in the mammalian testis, particularly in Sertoli cells. Then, we summarize results previously obtained by our group on the constitutive activation of NOTCH1 signaling in Sertoli cells and present additional data on the mechanism underlying the germ cell dysregulation we observed in our model. We close with an overview of our current understanding of the mechanisms regulating prenatal germ cell development.

Gonocytes Set the Stage for Spermatogenesis

In mammals, gonocytes (or prospermatogonia) are the precursors to spermatogonial stem cells (SSCs), which provide the foundation for spermatogenesis by their ability to both self-renew and generate daughter cells, which differentiate through meiosis to generate spermatozoa. Defects in proper gonocyte/SSC maintenance, in spermatogonia differentiation, or in the progression of meiosis lead to Sertoli cell-only syndrome or maturation arrest. These disorders are the leading causes of azoospermia, which occurs in 1% of all men³ and in 10–15% of infertile men.^1,2

During the embryonic development of male mice, primordial germ cells (PGCs) migrate to the genital ridges and by approximately embryonic day (E)12.5 are sequestered within the newly formed testis cords.⁹ Between E12.5 and E14.5, PGCs differentiate into gonocytes (also referred to as T1 prospermatogonia) and undergo mitotic arrest (quiescence) in an unsynchronized manner, exiting the cell cycle at the G₁/G₀ transition.^10-12 In mice, only after birth and between postnatal day (P)3 and P10 do the germ cells re-enter the cell cycle and resume mitosis, whereby they either develop into SSCs by P6 or enter prophase I of meiosis at puberty by P7 to P8.^9,10,13 Humans share the same developmental program, except that the timing is considerably extended, lasting the first 12 y of life.^9,14 During the first 2–3 y in boys, gonocytes transition into reserve and active adult dark and adult pale spermatogonia. By 5 y of age, these adult dark and adult pale spermatogonia begin to differentiate into type B spermatogonia, which by the 10th year represent only 10% of all spermatogonia.^9,14 Despite recent progress, a considerable gap in knowledge remains in our understanding of the regulatory mechanisms that govern the entry into and exit from gonocyte quiescence, the transition of gonocytes into SSCs, and the self-renewal vs. differentiation fate of SSCs. By elucidating these fundamental developmental mechanisms, it will be possible to determine the testicular causes of azoospermia and to design novel effective treatments for men with infertility.

The NOTCH Signaling Pathway

The NOTCH signaling pathway is best understood as a ligand- and receptor-based signaling pathway between adjacent cells, whereby both the ligands (JAG1, JAG2, DLL1, DLL3, and DLL4 in mammals) and receptors (NOTCH1 to NOTCH4 in mammals) are single-pass transmembrane proteins anchored to their respective cell surfaces. Upon ligand binding to a receptor, the intracellular domain of the receptor (NOTCH intracellular domain [NICD]) is released from the plasma membrane and translocates into the nucleus, where together with mastermind-like protein (MAML), NICD converts recombining binding protein suppressor of hairless (RBPJ) from a transcriptional repressor to a transcriptional activator. This activity triggers the expression of NOTCH target genes, the most notable of which are the hairy/enhancer-of-split (HES) and HES-related with YRPW motif (HEY) families of transcription factors.^15,16 In general, HES and HEY proteins inhibit the expression of other genes by forming complexes with co-repressors.^17,18

Modulation of NOTCH Signaling Through Receptor Modifications

NOTCH receptor activity can be regulated at numerous levels, including at glycosylation and fucosylation of the extracellular domain.^19,20 The NOTCH extracellular domain contains many tandem epidermal growth factor-like repeats, which can be modified with O-linked fucose, glucose, or N-acetylglucosamine, which, in turn, modulates the affinity of the NOTCH receptors for their ligands.¹⁹ The addition of O-fucose to epidermal growth factor-like repeats is catalyzed mainly by Protein O-fucosyltransferase 1 (POFUT1) in mammals,²¹ and Lunatic fringe (LFNG), which belongs to a family of β1–3 N-acetyltransferases, can modify these O-fucose moieties.^22-24 Pofut1 global null and Lfng global null mice display an embryonic lethal phenotype similar to mice lacking components of the NOTCH pathway,^25-28 although presentation of this phenotype may be due to indirect effects.

Expression of NOTCH Signaling Components in Sertoli Cells and Germ Cells

The expression of NOTCH signaling components in neonatal and adult rodent and human testes was first reported in 2001.²⁹ Through co-culture of 6-d-old BALB/c mouse Sertoli cells and germ cells and subsequent immunocytochemical analyses, Dirami et al. demonstrated that 3 NOTCH receptors (NOTCH1, NOTCH2, and NOTCH3) and 2 NOTCH ligands (JAG1 and DLL1) were expressed on the cell surface of spermatogonia.²⁹ Sertoli cells were shown to express both ligands but only 1 of the 3 NOTCH receptors tested, NOTCH2.²⁹ Concurrently, the studies of Hayashi et al. demonstrated NOTCH1 and JAG2 immunostaining of rat and human spermatocytes as well as round and elongated spermatids in paraffin-embedded testes cross-sections.³⁰ Interestingly, the authors’ examination of NOTCH1 and JAG2 testis expression in 11 patients with maturation arrest revealed a complete lack of NOTCH2 expression, despite continued expression of JAG2.³⁰ In 2003, Mori et al. further characterized immunostaining of NOTCH1 to NOTCH4 in frozen sections of adult mouse testes and included the activated intracellular domains of the receptors in their analyses.³¹ Their immunostaining demonstrated membrane-localized NOTCH1, NOTCH2, and NOTCH4 in spermatogonia, spermatocytes, and spermatids.³¹ They also found nuclear-localized intracellular domain of NOTCH3 (NICD3) in spermatogonia and nuclear-localized NICD1 and NICD4 in spermatocytes and spermatids, indicating that the respective receptors had been activated in these cells.³¹ Additionally, Sahin et al. demonstrated NOTCH2 immunostaining in spermatogonia and spermatocytes and NOTCH1 immunostaining in elongated spermatids.³² Although the results of these studies generally disagree with regard to the particular receptor or ligand expressed in each cell type in the seminiferous epithelium, cumulatively, these studies strongly indicate that Sertoli cells and germ cells can elicit and respond to NOTCH signaling during spermatogenesis, and that the NOTCH signaling system may play a role in male germ cell differentiation and survival. However, these studies lack null mouse models of NOTCH signaling components linking NOTCH signaling in these cells with significant physiologic effects.

Reproductive Phenotypes of Mouse Models of NOTCH Signaling

Notch3^−/−, Notch4^−/−, Hes5^−/−, Hey1^−/−, and Heyl^−/− global null mice are viable and fertile with no apparent lack of reproductive function.^15,33-36 Oppositely, Notch1^−/−, Notch2^−/−, Jag1^−/−, Jag2^−/−, Dll1^−/−, Dll3^−/−, Dll4^-/+, Hes1^−/−, Hes7^−/−, and Hey2^−/− mice cannot be studied for male reproductive function, because they all exhibit either complete embryonic or complete perinatal/postnatal lethal phenotype.^37-47 Thus, using the Cre/loxP system⁴⁸ and examining cell-specific null mouse models are of paramount importance for understanding the function and role of NOTCH signaling during spermatogenesis. Germ cell- and/or Sertoli cell-specific null mice of Notch1 that display no apparent phenotype have been generated,^49,50 but cell-specific deletion and testing of the remaining receptors, ligands, and/or Hes/Hey target genes that otherwise bear a lethal phenotype in global null mice have not been reported to date.

Since NOTCH receptor, ligand and target gene paralogs may have redundant functions when deleted in a cell-specific fashion, recent studies of Sertoli and germ cell-specific disruption of the NOTCH pathway have aimed to target modulators of the NOTCH receptor that are considered less redundant than individual receptors, ligands, or Hes/Hey target genes. One such modulator that has been tested and independently reported by 2 groups is POFUT1.^49,50 Results from both studies demonstrated that the NOTCH modifications brought about by POFUT1 are dispensable for testis development and spermatogenesis.^49,50 However, previous studies on POFUT1 demonstrate that the protein is not required for stable cell surface expression of NOTCH receptor in mammals;⁵¹ that an unrelated ER glucosidases can substitute for POFUT1 in promoting NOTCH folding and function;⁵² and that mammalian NOTCH receptors are capable of signaling in the absence of POFUT1 and O-fucose.⁵² Therefore, the finding that POFUT1 in Sertoli cells is dispensable for testis development and spermatogenesis does not necessarily preclude the possibility that NOTCH signaling can still function, either in entirety or to a limited or different extent, in these cells in modulating or helping to maintain proper regulation of spermatogenesis.

Because expression of NOTCH signaling components in a cell does not demonstrate that the pathway is truly activated, our recent study used a transgenic mouse model that expressed green fluorescent protein (GFP) in cells that actively received a NOTCH signal.⁵³ We observed that the canonical NOTCH signaling pathway is active exclusively in Sertoli cells in the perinatal testis. To circumvent redundancy and compensation in loss-of-function models, we then examined a Sertoli cell-specific NOTCH gain-of-function model in which NICD1 was constitutively expressed following Cre-mediated recombination in Sertoli cells (formerly Amh-cre;Rosa^NICD/+, hereafter referred to as AMH-NICD1).⁵³

AMH-NICD1 Mouse Model

It is well established that the maintenance and regulation of male germ cells rely on cues from the somatic environment, but the pathways that regulate gonocyte entry into and exit from quiescence, the transition of gonocytes into SSCs, and the self-renewal vs. differentiation fate of SSCs remain largely unknown. We demonstrated that increased activation of NOTCH signaling in Sertoli cells from approximately E13.5 onward led to drastic changes in gonocyte fate, including exit from quiescence, premature differentiation, and migration. These changes led to gonocyte apoptosis and to a Sertoli cell-only phenotype by P2. To glean additional insight into the molecular signature of the gonocytes in our model and to clarify the role for NOTCH signaling in Sertoli cells, we performed gene expression analysis of whole gonads isolated from E14.5 and E17.5 fetuses.⁵³ We found that several transcripts were differentially regulated in a manner consistent with NOTCH signaling activation in Sertoli cells and with premature differentiation of germ cells. For instance, RBPJ target genes such as Hey1 and Heyl were upregulated in Sertoli cells, as expected; in germ cells, Nanos2, a gene that suppresses meiosis,⁵⁴ was downregulated, and genes that are expressed at the beginning of or during meiosis, such as Stra8 and Rec8,^55,56 were all upregulated (GSE37073). Furthermore, in accord with the observation that germ cell differentiation had taken place, a significant decrease in Pou5f1 (a marker for undifferentiated spermatogonia)⁵⁷ and a significant increase in Sohlh2 (a marker for differentiating spermatogonia)⁵⁸ were also found in these mutant germ cells. Our molecular analyses demonstrated that these effects correlated with a downregulation of Sertoli cell-specific factors that are essential to maintaining the undifferentiated state of germ cells. These factors include glial cell-derived neurotrophic factor (GDNF), which is required for fetal germ cell proliferation and survival^59,60 and for spermatogonial stem cell self-renewal,^61,62 and CYP26B1, an enzyme that inhibits germ cell progression toward meiosis.⁶³ Conversely, complete inhibition of NOTCH signaling in Sertoli cells in vitro by the NOTCH inhibitor DAPT resulted in the opposite phenotype, with dose-dependent increases in Gdnf and Cyp26b1 expression. These data demonstrated that NOTCH signaling in perinatal Sertoli cells maintains gonocyte quiescence by establishing the proper balance between germ cell proliferation and differentiation.

Molecular Signature of Germ Cells in Wild-Type and AMH-NICD1 Mouse Testes

In male wild-type mice at E13.5, most PGCs have entered the bipotential gonads. By E15.5, they have all reached mitotic arrest and become gonocytes. Mitotic arrest coincides with downregulation of expression of CCND3 and CCNB1, the major cyclins found in PGCs.^64,65 The gonocytes remain mitotically quiescent until shortly after birth, when cyclins are re-expressed and the cell cycle resumes. Germ cells then actively proliferate as they differentiate and express different types of cyclins depending on the stage of maturation.⁶⁶ Previous research demonstrated that CCND1 is expressed in spermatogonia only; that CCND2 and CCND3 are expressed throughout the premeiotic and meiotic stages; and that CCNB3 is uniquely expressed in leptotene and zygotene spermatocytes.⁶⁶ Therefore, the particular expression pattern of cyclins in the mutant germ cells may represent the cells’ level of differentiation.

According to our microarray and quantitative real-time polymerase chain reaction data analysis, at E14.5, Ccnd1 transcripts were significantly upregulated in AMH-NICD1 gonads, at levels 3 times those of the littermate controls, suggesting that the germ cells had actively proliferated; indeed, the mutant testes had significantly more germ cells than did the controls.⁵³ We further examined protein expression of CCND1, as well as CCNB3 and CCND3, to glean additional insight into the stage of differentiation of the aberrant gonocytes. Through immunohistochemistry, we found that CCND1 and CCND3, but not CCNB3, were markedly upregulated in mutant gonocytes at E15.5. (Fig. 1); in stark contrast, the control gonads showed no staining above background levels for any of the cyclins tested at this age (Fig. 1). Therefore, the aberrant gonocytes presented signs of differentiation toward spermatogonia but were not advanced enough to proceed through the first meiotic division.

Figure 1. Aberrant germ cell proliferation and differentiation in AMH-NICD1 testes involves CCND1 and CCND3, but not CCNB3. (A–R) Immunofluorescence images of control (Rosa^NICD/+) (A–C, G–I, and M–O) and AMH-NICD1 (Amh-cre;Rosa^NICD/+) (D–F, J–L, P–R) testis sections at E15.5, showing CCNB3 (A–F; green), CCND1 (G–L; green), CCND3 (M–R; green), germ cell nuclear antigen (GCNA; A–R; red), and DAPI (A–R; blue) staining. Scale bars represent 100 µm. Higher magnification inset in upper right of each image. Immunostainings are representative of at least 6 separate gonad cross-sections, from at least 3 separate fetuses.

In mammals, the KIT membrane receptor at the surface of germ cells and the receptor’s ligand (KITL), produced by Sertoli cells, are of crucial importance for normal germ cell development in the embryo and after birth.⁶⁷ Additionally, KIT is a well-known marker for PGCs. Together with KITL, KIT mediates PGC proliferation and survival as well as PGC migration toward the genital ridges.⁶⁸ In wild-type mice, after the germ cells have reached the male gonads and are quiescent, KIT expression is downregulated. KIT is then re-expressed in a fraction of gonocytes after birth and is thought to mediate their migration toward the basement membrane.^69-71 While spermatogonial stem cells and other undifferentiated spermatogonia do not express KIT, this molecule is expressed in all differentiating spermatogonia in the first wave of spermatogenesis and in the adult.⁷² Our microarray data set (GSE37073) revealed that Kit transcripts were upregulated in AMH-NICD1 to levels 1.63 times those of the controls at E14.5 (p = 2.45 × 10⁻²). To determine whether KIT protein levels also were upregulated, we stained gonads at E15.5 with anti-KIT antibody. As shown in Figure 2, KIT was indeed markedly upregulated in a sub-population of gonocytes in the mutant. Because KIT expression was upregulated concomitant with a decrease in Pou5f1 and an increase in Sohlh2, Stra8, Sycp3, and Rec8,⁵³ our data overall establish that the aberrant gonocytes had acquired properties of differentiating spermatogonia.

Figure 2. Germ cells in AMH-NICD1 testes express KIT at high levels in a subpopulation of germ cells. (A–F) Immunofluorescence images of control (Rosa^NICD/+) (A–C) and AMH-NICD1 (Amh-cre;Rosa^NICD/+) (D–F) testis sections at E15.5, showing KIT (green), germ cell nuclear antigen (GCNA; red), and DAPI (blue) staining. Scale bars represent 100 µm. Areas outlined in panels (B and E) are magnified 4 times in panels (C and F), rotated 90° clockwise and counter-clockwise, respectively. Immunostainings are representative of at least 6 separate gonad cross-sections, from at least 3 separate fetuses.

Phenotype of Sertoli cells in Wild-Type and AMH-NICD1 Testes

The gonadal microenvironment is crucial for the control of germ cell differentiation. After they have arrived in the male gonad, the germ cells are nurtured by the Sertoli cells, which form the major component of the germ cell microenvironment. Through the timely delivery of growth and differentiation factors, Sertoli cells coordinate the precise development of the germ cells throughout life. One such growth factor is GDNF, which is already expressed at high levels by precursors of Sertoli cells at E11.5,⁵⁹ and which binds to a receptor complex formed by the receptors RET and GFRA1 in fetal and adult germ cells.^60,73,74 Miles et al. recently investigated the putative function of GDNF signaling in the fetal gonad and demonstrated that Gndf is exclusively expressed in the male gonad between E12.5 and E15.5,⁶⁰ when gonocytes have entered the gonads but still proliferate until E15.5. These authors demonstrated that RET signaling is required for gonocyte proliferation and maintenance, and that Ret^−/− germ cells underwent apoptosis.

As we previously reported, transcript levels for Gdnf were significantly downregulated in the AMH-NICD1 mutant testes at E14.5 and E17.5 and in cultures of Sertoli cells isolated from mutant testes at P10.⁵³ In addition, γ-secretase inhibitor DAPT-induced downregulation of NOTCH signaling in Sertoli cells in vitro upregulated Gdnf expression levels. These findings therefore strongly suggest that NOTCH signaling is a negative regulator of GDNF production that counteracts the positive effects of FGF2⁷⁵ (Fig. 3). NOTCH signaling specifically downregulates the levels of expression of GDNF in the perinatal testis and may contribute to the entry of gonocytes into mitotic arrest (M-prospermatogonia into T1-prospermatogonia). Since the levels of GDNF increase after birth and influence gonocyte migration (T2-prospermatogonia)⁷⁶ and spermatogonial stem cell self-renewal,^61,62 GDNF expression in the perinatal testis must be cyclic and may be inversely related to the cyclicity of NOTCH activity.⁴⁹ NOTCH signaling may maintain gonocytes in a quiescent state by keeping GDNF at basal levels to ensure gonocyte survival; however, a lack of GDNF through NOTCH signaling overactivation induces germ cell apoptosis.

Figure 3. Model depicting the possible role of NOTCH signaling in Sertoli cells before birth. Previous studies have shown that FGF2 induces GDNF production, and that SOX9/SF1 induces CYP26B1 expression by Sertoli cells. Our data suggest NOTCH1 signaling is a negative regulator of GDNF and CYP26B1, which balances the effects of FGF2 and SOX9/SF1. This normally results in gonocyte maintenance in the quiescent, undifferentiated state. However, overactivation of NOTCH1 signaling further suppresses the expression of GDNF and CYP26B1, which induces direct differentiation into KIT-expressing proliferating spermatogonia.

We also demonstrated that overactivation of NOTCH signaling in vivo downregulated CYP26B1, an enzyme crucial for maintaining germ cells in an undifferentiated state.^77,78 Abrogation of NOTCH activation by DAPT in Sertoli cells in vitro resulted in an increase in Cyp26b1 expression, indicating that NOTCH signaling downregulates the levels of this enzyme. Normally, CYP26B1 in Sertoli and Leydig cells regulates the concentration of retinoic acid by catabolizing all-trans-retinoic acid into inactive oxidized metabolites.^79-81 Since retinoic acid controls germ cell progression toward meiosis, the presence of CYP26B1 in the male embryo and perinatal testis prevents meiotic entry.⁸² Therefore, decrease of CYP26B1 in the AMH-NICD1 gonads likely promotes both premature differentiation of gonocytes into spermatogonia and germ cell progression toward meiosis, causing the cells to express Kit, Stra8, and Sycp3, as reported in our previous investigation.⁵³ Until the recent study by Kashimada et al., little was known about the regulation of CYP26B1 expression in normal gonads.⁸³ It now appears that SOX9 and SF1 significantly activate Cyp26b1 expression in vivo in male mice,⁸³ which represses meiosis until puberty. Therefore, NOTCH appears to promote differentiation toward meiosis by negatively regulating Cyp26b1. However, the effects of SOX9 and SF1 presumably outweigh the effects of NOTCH in fetal Sertoli cells (Fig. 3). Sex reversal was not observed in the AMH-NICD1 model, because the levels of Sox9,^84,85 Dax1,⁸⁶ and Foxl2⁸⁷ expression in Sertoli cells were not significantly altered in our microarray (GSE37073).

Conclusion

The NOTCH signaling pathway is crucial for tissue patterning and homeostasis and plays an important role in regulating C. elegans spermatogenesis and the Drosophila male germ cell niche.^5,7,8 In mice, this pathway’s role in the development of the ovaries and oocytes has begun to be clarified;^88,89 however, its role in controlling the fate of gonocytes and the development of male germ cells after birth and into adulthood is still not clear. We have now demonstrated that NOTCH signaling in Sertoli cells is a crucial modulator of germ cell development, and that when overactivated, NOTCH signaling inhibits quiescence and pushes germ cells toward differentiation and meiosis (Fig. 3). While further research is required to decipher the molecular pathways underlying these effects, as well as the relationship between NOTCH and other signaling pathways in Sertoli cells and germ cells, it appears that the absence of NOTCH activation ensures germ cell proliferation, whereas NOTCH overactivation promotes germ cell differentiation toward meiosis and, ultimately, apoptosis. However, it is not known whether germ cell apoptosis is linked to inadequate levels of GDNF and CYP26B1 or to an inadequate response of the microenvironment to prematurely differentiating germ cells. In addition, data from our in vitro experiment suggest that loss of NOTCH signaling may induce a germ cell differentiation block by increasing the levels of GDNF produced by Sertoli cells. This event implies that dysregulation of this pathway plays a role in the maintenance and growth of testicular germ cell tumors, as previously suggested by others.⁹⁰

Materials and Methods

Generation of conditional AMH-NICD1 mice

Heterozygous Tg(AMH-cre)1Flor (Amh-cre)⁹¹ female mice were mated to homozygous floxed Gt(Rosa)26Sor^{tm1(Notch1)Dam}/J (Rosa^NICD)⁹² male mice to generate F₁ Amh-cre;Rosa^NICD/+ overexpressors and Rosa^NICD/+ controls as previously described.⁵³

Histology and immunohistochemistry

Immunohistochemistry of Rosa^NICD and Amh-cre;Rosa^NICD testes was performed as previously described.⁵³ All multi-channel fluorescent images shown in the figures and used for analysis were tiled and stitched images of entire gonad cross-sections, acquired with a Nikon A1 laser scanning confocal microscope (Nikon, Tokyo, Japan) equipped with the NIS-Elements acquisition and analysis software platform (version 4.1), with equivalent exposure levels between control and mutant sections at each staining. The primary antibodies used and their dilutions are as follows: CCNB3 (Santa Cruz Biotechnology; Cat. No. SC-131482; 1:1000), CCND1 (Abcam; Cat. No. AB6152; 1:1000), CCND3 (Abcam; Cat. No. AB28283; 1:1000), germ cell nuclear antigen (GCNA; a generous gift of Dr George Enders;⁹³ 1:30), and KIT (R&D Systems, Minneapolis, MN; Cat. No. AF1356; 1:40). Each staining was performed with at least 6 separate testes from at least 3 separate fetuses.

Abbreviations:
SSCs	=	spermatogonial stem cells
PGCs	=	primordial germ cells
E	=	embryonic day
P	=	postnatal day
NICD	=	NOTCH intracellular domain

Acknowledgments

This work was supported by NIH R01 HD044543 and R21 HD068989 to MCH and in part through Cancer Center Support Grant CA16672. We are also grateful to Dr Walter N Hittelman and the Center for Targeted Therapy at The University of Texas MD Anderson Cancer Center for generously providing access to the Nikon confocal microscope.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.