DMRTC2, PAX7, BRACHYURY/T and TERT Are Implicated in Male Germ Cell Development Following Curative Hormone Treatment for Cryptorchidism-Induced Infertility

Defective mini-puberty results in insufficient testosterone secretion that impairs the differentiation of gonocytes into dark-type (Ad) spermatogonia. The differentiation of gonocytes into Ad spermatogonia can be induced by administration of the gonadotropin-releasing hormone agonist, GnRHa (Buserelin, INN)). Nothing is known about the mechanism that underlies successful GnRHa treatment in the germ cells. Using RNA-sequencing of testicular biopsies, we recently examined RNA profiles of testes with and without GnRHa treatment. Here, we focused on the expression patterns of known gene markers for gonocytes and spermatogonia, and found that DMRTC2, PAX7, BRACHYURY/T, and TERT were associated with defective mini-puberty and were responsive to GnRHa. These results indicate novel testosterone-dependent genes and provide valuable insight into the transcriptional response to both defective mini-puberty and curative GnRHa treatment, which prevents infertility in man with one or both undescended (cryptorchid) testes.


Introduction
During mini-puberty, which occurs between 30 and 90 days of postnatal life in male infants, the substantial increase in gonadotropin releasing hormone (GnRH) secretion induces gonadotropin and testosterone production [1][2][3]. As a consequence, transformation of gonocytes into adult dark (Ad) spermatogonia takes place. Ad spermatogonia have a characteristic nuclear feature that distinguishes them from the other germ cells (e.g., fetal, transient, and pale-type (Ap) spermatogonia) (Figures 1 and 2). Ad spermatogonia appear at three months of age and remain present in the testes for the rest of an individual's life. Therefore, the transformation of gonocytes into Ad spermatogonia, either directly or through intermediate stages, is not simply another step in a succession Cryptorchid boys, of a median age of eight years, who were treated with a gonadotropinreleasing hormone agonist, showed post-puberty improved sperm concentrations, when compared to an untreated control group [10]. The treatment resulted in increased luteinizing hormone (LH) levels and regeneration of atrophic juvenile Leydig cells, and increased numbers of germ and Leydig cells [11,12]. Worth noting, long term follow-up in a high infertility risk group of cryptorchid boys, treated before the age of six, showed normal sperm concentrations in 86% of cases [13]. This result strongly contrasts with those of a 'surgery only' group, in which not a single patient had a normal semen analysis and 20% suffered from azoospermia [13].
Though expression patterns may differ, development in both humans and mice appear to involve a similar set of genes, which can also be used as markers to distinguish gonocytes and spermatogonia (reviewed in [14,15]). Because the transition process is similar between both species, animal models are commonly used, since the testicular tissues necessary to study this process in humans are difficult to obtain. ALPP/PLAP, EPS8, KIT/c-KIT, NANOG, POU5F1 and TFAP2C/AP2γ are gonocyte markers, while ALPP, NANOG, POU5F1, and TFAP2C encode transcription factors known to be important for pluripotency. The differentiation of gonocytes into spermatogonia is associated with upregulation of certain genes, including MAGEA4, DDX4 and TSPY1 [16,17], while POU5F1, ALPP, TFAP2C, NANOG and KIT are downregulated [17][18][19][20][21]. Markers for self-renewal (ETV5, FOXO1, GFRA1, ID4, RET, SALL4, UTF1, CHD1L, and TAF4B) or differentiation (DMRT1, ZBTB16/PLZF, FGF9, FGFR3, Cryptorchid boys, of a median age of eight years, who were treated with a gonadotropin-releasing hormone agonist, showed post-puberty improved sperm concentrations, when compared to an untreated control group [10]. The treatment resulted in increased luteinizing hormone (LH) levels and regeneration of atrophic juvenile Leydig cells, and increased numbers of germ and Leydig cells [11,12]. Worth noting, long term follow-up in a high infertility risk group of cryptorchid boys, treated before the age of six, showed normal sperm concentrations in 86% of cases [13]. This result strongly contrasts with those of a 'surgery only' group, in which not a single patient had a normal semen analysis and 20% suffered from azoospermia [13].
In this study, we investigated the molecular events underlying human male germ cell development, focusing on the testosterone-dependent transition from gonocytes to Ad spermatogonia as well as the molecular impact of early GnRHa (Buserelin INN) treatment. Utilizing testicular gene expression profiles from testes with insufficient testosterone secretion, before and after GnRHa administration, and testes with completed mini-puberty, we identified the DMRTC2, PAX7, BRACHYURY/T, and TERT genes to be associated with defective mini-puberty and responsive to GnRHa.

Study Population and Biopsy Sample Collection
We selected 15 patients with isolated cryptorchidism, based on histological results, and divided them into 2 groups. Seven belonged to the Ad− (lacking Ad spermatogonia) and 8 to the Ad+ (presenting Ad spermatogonia) group. Data from Ad− bilateral cryptorchid boys treated with GnRHa (Buserelin) following the first orchidopexy (surgery) (4 patients) were retrieved from an ongoing randomized study. Initial biopsies revealed no Ad spermatogonia, indicating defective mini-puberty (Ad− group). The second testis was managed by orchidopexy and biopsied 6 months after the initial surgery. Thus, results from 19 biopsies were compared. Patients were age and ethnicity matched. RNA sequencing data from manually selected germ cell marker genes from our two previous studies [22,23] were analyzed.
A cryptorchid testis is defined as a testis localized outside of the scrotum and incapable of being brought into a stable scrotal position. All undescended testes in this study were located in the inguinal region. Testicular biopsies were taken at the time of orchidopexy. This sample was then subdivided, with one fragment fixed in glutaraldehyde for histological processing, while the other one was immediately immersed in RNAlater (ThermoFisher Scientific, Waltham, MA, USA) and stored at -25 • C until further processing (for RNA extraction and RNA-sequencing).

Histological Analyses
Biopsies were fixed in 3% glutaraldehyde in phosphate-buffered saline (PBS, pH 7.4) and then embedded in Epon resin. Semi-thin sections (1 µm) were cut using a Reichert Om-U3 ultramicrotome (Reichert AG, Vienna, Austria). Sections were mounted on glass slides, stained with 1% toluidine blue, and examined under a Zeiss Axioskop light microscope (Carl Zeiss Microscopy Gmbh, Jena, Germany) with an integrated photo-camera. Biopsies were histologically examined by two of the authors (F.H. and D.D.), each with expertise in the interpretation of semi-thin sections of prepubertal testes.
During histological analyses, at least 100 tubular cross sections per biopsy were evaluated, with regard to their number and absence of Ad spermatogonia. In the prepubertal testes, Ad spermatogonia were identified according to the criteria first published by Seguchi and Hadziselimovic [24]. This type of germ cell has a typical halo in the nucleus, termed the rarefication zone, and cytoplasm with a darker aspect in comparison to Ap or fetal spermatogonia.

RNA Preparation, Sequencing, Data Analyses, and RNA Expression Levels
The workflow from RNA isolation, through to purification, library preparation, sequencing, data analyses, and expression level analyses, has been described previously [22,23].

Data and Differential Gene Expression Analyses
Determination of differentially expressed genes, statistical analyses and model design were described previously [22,23]. Only genes with at least one read per million, in at least two samples, were included. p values and fold-changes were calculated for the treatment factor and differentially expressed genes were defined as those displaying a false discovery rate (FDR) of less than 0.05 and an absolute change in expression of at least two-fold. Raw data files are available at the Database of Genotypes and Phenotypes (dbGaP) with the accession number phs001275.v1.p1.

Protein Interaction Network
Two gonocyte marker genes, 19 spermatogonial marker genes and 8 putative spermatogonia genes, all of which are differentially expressed between the two groups (Ad− and Ad+), were used as input to obtain the protein-protein interaction network using STRING version 10.0 [25].

Ethics Statement
Investigations were carried out in accordance with the Declaration of Helsinki of 1975, revised in 2008. All aspects of this study were approved by the Institutional Review Board and the Independent Ethics Committee of Vilnius University. Approval was also provided for research involving the use of material (data records or biopsy specimens) that had been collected for non-research purposes (Vilnius Regional Biomedical Research Ethics Committee, No. 158200-580-PPI-17, 11 June 2013).

Results
Here, we focused on selected marker genes for gonocytes and Ad spermatogonia (Table 1). Table 1. Differential expression of gonocyte and spermatogonial marker genes involved in self-renewal and differentiation of spermatogonial stem cells (SSCs). Absolute fold changes (logFC) and false discovery rates (FDR) of differentially expressed genes in the Ad− vs. Ad+ group, and in the GnRHa (Buserelin, INN) treated vs. untreated group are indicated. Absolute fold changes <2 are highlighted in red; n.d.: not determined, n.s.: not significant. Gonocytes are defined as small cells originating from the primordial germ cells and localized predominately in the center of the tubule and small typical mitochondria. They give rise to the fetal spermatogonia, which are the largest germ cells in prepubertal testis (Figure 2). This type of germ cells represents a population of so called undifferentiated spermatogonia. In the group of undifferentiated spermatogonia we included also all transient forms of the germ cells, which evolve into A spermatogonia [24]. Gonocytes are defined as small cells originating from the primordial germ cells and localized predominately in the center of the tubule and small typical mitochondria. They give rise to the fetal spermatogonia, which are the largest germ cells in prepubertal testis (Figure 2). This type of germ cells represents a population of so called undifferentiated spermatogonia. In the group of undifferentiated spermatogonia we included also all transient forms of the germ cells, which evolve into A spermatogonia [24].

Decreased Expression of Gonocyte and Spermatogonial Marker Genes in Testes with Altered Mini-Puberty
Of the six gonocyte markers selected, POU5F1 and TFAP2C showed lower expression levels in Ad− testes with testosterone deficiency (Table 1), and 19 of 34 spermatogonial marker genes showed

Gonocyte and Spermatogonial Marker Genes Respond to GnRHa Treatment
Out of the six gonocyte marker genes tested, EPS8 and KIT showed decreased RNA expression after GnRHa treatment (Table 1). Neither gene was differentially expressed between testes, with or without Ad spermatogonia. ALPP, NANOG, POU5F1, and TFAP2C expression levels were similar between the treated and untreated testes, and GnRHa treatment did not lead to upregulation of any gonocyte markers.
Downregulation was observed in nine of 34 spermatogonial marker genes (ADGRA3, CHD1L, DMRT1, ETV5, FOXO1, ID4, TAF4B, THY1, UCHL1), and the expression levels of ETV5, ID4 and UCHL1 were lower than in testes with Ad spermatogonia, indicating that GnRHa treatment further decreased the expression of these genes (Table 1 and blue nodes in Figure 3). The expression of 20 spermatogonial markers showed no significant response to GnRHa treatment (Table 1). Five spermatogonial genes (PAX7, POU2F2, BRACHYURY/T, TERT, TSPAN8) responded with an increase in RNA expression. TSPAN8 was not differentially expressed between Ad− and Ad+ testes. PAX7, BRACHYURY/T, and TERT were both less expressed in Ad− testes and upregulated following GnRHa treatment (Table 1 and red nodes in Figure 3). These genes showed the strongest treatment effect. spermatogonial markers showed no significant response to GnRHa treatment (Table 1). Five spermatogonial genes (PAX7, POU2F2, BRACHYURY/T, TERT, TSPAN8) responded with an increase in RNA expression. TSPAN8 was not differentially expressed between Ad− and Ad+ testes. PAX7, BRACHYURY/T, and TERT were both less expressed in Ad− testes and upregulated following GnRHa treatment (Table 1 and red nodes in Figure 3). These genes showed the strongest treatment effect.  Only a few of the marker genes downregulated in Ad-testes were then upregulated after GnRHa treatment. Therefore, we searched for additional genes involved in testosterone-dependent gonocyte-to-Ad spermatogonia transition (Supplementary Tables S1 and S2). We identified eight additional candidates matching these criteria: DMRTC2, EGR2, NRG1, NRG3, RBMY1B, RBMY1E, RBMY1J, and TSPY4 (Table 1). A positive effect of GnRHa on EGR2, NRG1, POU2F2, RBMY1B, RBMY1E and RBMY1J gene expression has been previously reported [22,23].
We next interpreted the 29 markers (including putative markers, and all of which are differentially expressed between Ad− and Ad+) in the context of physical protein-protein interactions and functional interactions, by integrating our data with information available in the literature (STRING interaction network; http://string-db.org [25]). Markers responding positively to GnRHa (red nodes in Figure 3) are mostly not connected to the non-responsive (grey nodes) key germ cell markers such as FGF9, NANOS2, SOHLH1 and SOHLH2, suggesting that, at the protein level, GnRHa stimulates and activates alternative pathways in germ cells. This is consistent with our previous observations of alternate GnRHa responsive genes in the hypothalamus-pituitary-gonadal (HPG) axis [23]. Especially the markers, PAX7, BRACHYURY/T, EGR2, NRG1 and NRG3, seem to represent an alternative pathway, activated by GnRHa and involved in gonocyte-to-Ad spermatogonia transition. It should be noted that although not visualized by STRING, BRACHYURY/T expression was reported to be partially influenced by POU5F1 [29], and to be a downstream effector of GDNF/ETV5 signaling to promote self-renewal [30].

Luteinizing Hormone and Testosterone Deprivation Decreases Gonocyte and Spermatogonial Marker Gene Expression
POU5F1 encodes a transcription factor that plays a key role in both embryonic development and stem cell pluripotency. In human fetal gonads, POU5F1 regulation differs in male and female germ cells. While POU5F1 expression is gradually downregulated during gonocyte differentiation, in human males, in females it is downregulated much earlier in fetal life, when the oocytes enter the first meiotic prophase [31]. TFAP2C/AP2γ expression is also gradually reduced during gonocyte differentiation [32]. The reduced levels of POU5F1 and TFAP2C RNA, observed in Ad− testes compared to Ad+ testes, lead to the assumption that they play an important role in LH and in the testosterone-dependent gonocyte-to-Ad spermatogonia transition.
ETV5 encodes a glial cell-derived neurotrophic factor (GDNF)-inducible transcription factor, regulating several genes known to be important for stem cell self-renewal, including BCL6B, LHX1, CXCR4, BRACHYURY/T, and RET [30,33]. The fact that ETV5-null mice are infertile demonstrates the importance of ETV5 for spermatogenesis [34]. Moreover, Wu and coworkers showed that transplantation of SSCs in vivo following Brachyury/T silencing significantly reduces the number of donor cell-derived colonies formed in recipient mouse testes, suggesting that BRACHYURY/T functions as a part of GDNF/ETV5 signaling to promote self-renewal [30]. BRACHYURY/T is a classical mesodermal marker, which is regulated by WNT and BMP signaling and expressed in early mouse and human primordial germ cells [35,36]. Ad− testes showed reduced expression levels for GDNF signaling factors-ETV5, CXCR4, BRACHYURY/T, and RET-which indicates disturbed GDNF-dependent self-renewal in the germ cells.
The pluripotency transcription factor, SALL4, was found to be localized to primordial germ cells and most gonocytes in the prenatal and early postnatal testes, as well as in undifferentiated spermatogonia of marmoset, human and mouse pubertal and adult testes [37,38]. SALL4 regulates the expression of genes required for either self-renewal or differentiation, along with ZBTB16 and DMRT1 [39].
NANOS2 and NANOS3 are RNA binding proteins from the NANOS family. In mice, a Nanos3 deficiency results in the loss of primordial germ cells during migration and leads to sterility in male and female mice [40]. In contrast, a Nanos2 loss results in decreased germ cell numbers and causes infertility in only male mice [40]. Murine NANOS2 activates a male-specific genetic program in female germ cells by inhibiting meiosis, which suppresses the female pathway and induces male-type differentiation [41]. Furthermore, Nanos2 expression is directly dependent on retinoic acid (RA) for its downregulation and fibroblast growth factor 9 (FGF9) for its upregulation [42]. While Nanos2 expression is restricted to prenatal germ cells and a small number of spermatogonia in adult mice [40,43], NANOS2 was reported to be more widely expressed in adult humans, including in spermatocytes and round spermatids [44]. Therefore, Kusz and colleagues suggested that NANOS2 is not a suitable spermatogonial marker in adult men, although our results indicate that NANOS2 could be a potential marker for spermatogonia in young boys. SOHLH1 and SOHLH2 together with DMRT1 and DMRT6/DMRTB1 ensure that meiosis starts only when spermatogonia have reached the appropriate differentiation step [45,46]. In summary, Nanos2, Sohlh1 and Sohlh2 all play substantial roles in male germline development in mice, and our results strongly support the notion that they fulfill the same role during mini-puberty in humans.

Genes with Augmented Levels after GnRH Treatment
Testosterone treatment of mouse satellite cells was shown to increase Pax7 expression [47], supporting the present observation of decreased PAX7 expression in Ad− testes with low testosterone levels as well as increased PAX7 expression in the testes of boys with increased testosterone levels, following GnRHa treatment.
Lim and colleagues observed a predominant expression of the transcription factor POU2F2 in Ad spermatogonia [48]. Although it has yet not been confirmed by other studies in humans whether POU2F2 is indeed a specific marker for Ad spermatogonia, our RNA expression results after GnRHa treatment do support this finding.
High telomerase expression levels were found to be a hallmark of undifferentiated spermatogonia using telomerase reverse transcriptase (TERT) reporter mice [49]. It was also shown that telomere dysfunction caused undifferentiated spermatogonia depletion, which disrupted male germ cell development. While high telomerase expression has yet not been demonstrated as a hallmark for human spermatogonia, our differential expression data strongly support this possibility. Atrophic testes had lower TERT expression levels relative to normal testes, leading to the suggestion that telomerase plays a role in maintaining germ cell proliferation [50]. Furthermore, the TERT mRNA expression level was shown to be effective in both classifying spermatogenesis disorders in patients, and in predicting successful sperm recovery in azoospermia patients [51,52]. Androgens were reported to enhance TERT expression in human primary hematopoietic cells [53]. Calado and colleagues also demonstrated that the aromatase, CYP19, known to convert testosterone into estradiol, is necessary for the testosterone-dependent increase in TERT expression. This supports both, our finding of reduced TERT expression in lower testosterone Ad-testes, but also the observed increased TERT expression in Ad-testes with increased testosterone levels after treatment. Additionally, GnRHa treated testes showed significantly increased CYP19A1 gene expression levels (absolute fold change logFC GnRHa + 2.51) [23].
Neither testosterone, nor LH dependent gene expression of BRACHYURY/T, has yet been reported. However, BRACHYURY/T was shown to bind to the promotor of the androgen receptor (AR) and regulate AR expression in prostate cancer cells [54]. Increased BRACHYURY/T levels after GnRHa treatment point towards its role in Ad spermatogonia formation. This is a new observation which indicates this gene to be an important marker of Ad spermatogonia.
EGR2 and EGR3 are transcription factors used as spermatogonial markers, and EGR3 immunoreaction was reported in A single or paired germ cells in mice [55]. Furthermore, in mice EGR3 expression is stimulated by RA and downregulated by KIT Ligand (KITLG) [56].
The neuregulins, NRG1 and NRG3, are essential for the proliferation of spermatogonia and the initiation of meiosis [57]. NRG1 and KITLG were also reported to activate alternative pathways downstream of RA signaling in the germline, known to be essential for spermatogonial differentiation [58]. Notably, after GnRHa treatment NRG1 was upregulated, while KITLG was downregulated (logFC GnRHa − 1.10).
While mouse RBMY is expressed only in spermatogonia and early spermatocytes [59] and its mRNA is not detected in meiotic or post-meiotic cells [60], human RBMY is expressed in spermatogonia, spermatocytes, and round spermatids [61,62]. Therefore, it is notable that testes lacking Ad spermatogonia showed significantly reduced RBMY RNA levels that increased strongly after GnRHa treatment.
We also found a treatment related increase in DMRTC2/DMRT7 and TSPY4 gene expression. DMRT7 mutant mice show meiotic arrest at the pachytene stage [63,64], and DMRT7 protein is present in germ cells, localized to the male XY body during meiosis, and essential for male but not female fertility in mice [64]. While murine DMRT7 was predominantly expressed in mid-to late-pachytene spermatocytes and not detected in other germ cells including spermatogonia [64], our results point to a role of DMRTC2/DMRT7 in the early stage of human male germ cell development. Although the function of TSPY4 is unknown, from sequence similarity it is assumed to be involved in sperm differentiation and proliferation (UniProtKB/Swiss-Prot, TSPY4_HUMAN, P0CV99). Interestingly, TSPY was reported to bind to androgen receptor and AR variants, and thereby increase the transactivation of the AR/AR-V7 target genes [65]. Whether testosterone-dependent TSPY4 does also bind AR is unknown.

Combining Classical Physiological Information and Cutting-Edge Genomics Data into a Complete Picture
Here, we report that testes with defective mini-puberty, with lower testosterone levels, and lack of Ad spermatogonia had significant lower RNA levels for selected gonocyte and spermatogonial marker genes (21 genes) relative to testes with Ad spermatogonia. We suggest that these differentially expressed genes reflect molecular functions involved in the gonocyte-to-Ad spermatogonia transition in humans during mini-puberty. Furthermore, we propose that higher expression levels of these 21 genes in testes presenting Ad spermatogonia are the result of testosterone-dependent expression, since the lack of testosterone increase during mini-puberty causes developmental arrest. The finding of four gonocyte markers and 15 spermatogonial marker genes that are not differentially expressed argues against a dilution effect and supports the importance of these findings.
PAX7, BRACHYURY/T, and TERT responded positively to GnRHa treatment, and were markers with reduced expression in Ad− testes. We suggest that the genes DMRTC2, EGR2, NRG1, NRG3, RBMY1B, RBMY1E, RBMY1J and TSPY4 represent potential new markers for spermatogonia in infant testes, and that they may have key functions in the gonocyte-to-Ad spermatogonia transition. Additionally, it seems likely that they are testosterone-responsive genes, given that GnRHa treated boys were reported to show increased testosterone and LH levels.
The absence of an apparent GnRHa stimulation for 11 out of 16 genes remains unexplained, but one possibility could be that they are epigenetically downregulated and therefore unresponsive to GnRHa treatment at the mRNA level. Nonetheless, the absence of GnRHa-responsive key players strengthens and supports the need for alternative pathways, for which we suggest the stimulation of the transcription factors, EGR2, DMRTC2, PAX7 and BRACHYURY/T, the growth factor like proteins, NRG1 and NRG3, and the RNA binding and Y chromosome encoded genes, RBMY1B, RBMY1E, RBMY1J and TSPY4.
The balance between self-renewal and differentiation depends not only on the described intrinsic factors, but also on extrinsic factors, including GDNF, RA, WNT and testosterone signaling. Spermatogonial cell development in mice depends on testosterone-dependent secretion of GDNF by peritubular myoid cells [71]. GDNF expression was significantly increased after GnRHa treatment (logFC GnRHa + 1.47), suggesting expression induced by GnRHa. LH-dependent testosterone secretion was reported to regulate SSC self-renewal by suppressing Wnt5a expression in mouse Sertolli cells [72]. GnRHa treated testes showed reduced WNT5A expression (logFC GnRHa − 0.86), suggesting a similar testosterone-dependent regulation of SSC self-renewal, by Wnt5a suppression in humans.

Conclusions
Our differential gene expression profiling of gonocyte and spermatogonial markers, particularly DMRTC2, PAX7, BRACHYURY/T, and TERT, highlights their importance for the development of Ad spermatogonia with specific functionalities in self-renewal and differentiation, and following GnRHa curative treatment. We suggest that GnRHa induced testosterone and a LH increase reconstitute self-renewal properties of the Ad spermatogonial stem cells, and induce RA-responsive genes, such as NRG1, NRG3 and PAX7, to help prepare them for commitment to differentiation.
Supplementary Materials: The following are available online at www.mdpi.com/2073-4425/8/10/267/s1, Table S1: Differentially expressed genes involved in the self-renewal of spermatogonial stem cells in HIR/LIR and after GnRHa treatment, Table S2: Differentially expressed genes involved in the differentiation of spermatogonial stem cells in HIR/LIR and after GnRHa treatment.