Abstract
Sertoli cells (SCs) are supporting cells in the mammalian testis that proliferate throughout fetal and postnatal development but exit the cell cycle and differentiate at puberty. In our previous study, we isolated a population of highly proliferative Sertoli-like cells (SLCs) from the region of the adult mouse testis containing the rete testis and adjacent seminiferous tubules. Here RNA-seq of the adult SLC culture as well as qPCR analysis and immunofluorescence of the adult and immature (6 dpp) SLC cultures were performed that allowed us to identify SLC-specific genes, including Pax8, Cdh1, and Krt8. Using these, we found that SLCs are mostly localized in the rete testis epithelium; however, some contribution of transitional zones of seminiferous tubules could not be excluded. The main feature of SLCs indicating their relationship to SCs is DMRT1 expression. More than 40% of both adult and immature SLCs expressed DMRT1 at different levels in culture. Only rare DMRT1+ cells were detected in the adult rete testis, whereas more than 40% of cells were positively stained for DMRT1 in the immature rete testis. One more SC protein, AMH, was found in some rete cells of the immature testis. It was also demonstrated that SLCs expressed such SC genes as Nr5a1, Dhh, Gdnf, and Kitl and interacted with germ cells in 3D co-culture with immature testicular cells. All these similarities between SLCs and rete cells on one the hand and SCs on the other, suggest that rete cells could share a common origin with SCs.
Sertoli cells (SCs) are specialized supporting cells for developing germ cells and localized within seminiferous tubules of the mammalian testis. SCs actively proliferate throughout fetal and postnatal development but cease proliferation and differentiate at puberty (Sharpe et al. 2003). However, many studies have revealed that adult SCs could partly revert to an immature state under different stimuli (Tarulli et al. 2012). Seasonal breeding Djungarian hamsters resume SC proliferation after experimental changes in gonadotropin levels (Tarulli et al. 2006), and the same process occurs in human testes after gonadotropin suppression, although to a lesser extent (Tarulli et al. 2013). Monkey SCs begin to re-express KRT18, a marker of immature cells, in the testis and culture after experimental cryptorchidism and heat treatment (Zhang et al. 2004, 2006). The proliferation of postpubertal rodent and human SCs was demonstrated in culture (Ahmed et al. 2009, Nicholls et al. 2012) and after transplantation (Mital et al. 2014). Our previous study demonstrated that some SCs isolated from adult mouse seminiferous tubules indeed entered the cell cycle in vitro, but they were not able to proliferate more than once or twice. In contrast, cells isolated from the testis region containing rete testis and adjacent seminiferous tubules actively proliferated in culture producing large colonies (Kulibin & Malolina 2016). These cells, hereafter termed Sertoli-like cells (SLCs), expressed SC genes such as Wt1, Sox9, Gata4, and Vim, and more importantly, some of them also exhibited the expression of DMRT1.
DMRT1 is expressed exclusively in SCs, gonocytes and spermatogonia in male mammals (Raymond et al. 2000, Lei et al. 2007). DMRT1 is a transcription factor that shares the DM domain, a DNA-binding motif, with proteins controlling sex determination and sex differentiation in many metazoans (Zarkower 2013). DMRT1 is not required for primary male sex determination in mammals (Capel 2017), but it is essential for SC differentiation after birth (Raymond et al. 2000) and maintains SC identity throughout postnatal life protecting them from reprogramming into female granulosa cells (Matson et al. 2011). The fact that SLCs can express DMRT1 suggests their relationship to SCs.
We hypothesized that SLCs were localized at the border between seminiferous tubules and the rete testis, a system of tubules and cavities that transports sperm from seminiferous tubules to efferent ducts (Kulibin & Malolina 2016). The most prominent candidates for SLCs were SCs with modified morphology from transitional zones, the terminal segments of seminiferous tubules adjacent to the rete testis (Dym 1974, Nykänen 1979). Those SCs were reported to be capable of proliferation in adult testes of Syrian hamsters and rats (Aiyama et al. 2015, Figueiredo et al. 2016). However, strict localization of SLCs was unclear.
Here we addressed this question, found SLC-specific genes and provided evidence that mouse SLCs mostly reside in the rete testis epithelium. However, there is the possibility that transitional zones also contribute to some SLCs. Further examination of adult and immature rete testis epithelia and SLCs isolated from them revealed many similarities between SLCs and rete cells on one hand and SCs on the other. The most intriguing is DMRT1 expression in a substantial part of both adult and immature SLCs and in the rete testis epithelium.
Materials and methods
Animals
C57Bl/6J mice, B10.GFP mice expressing GFP under the β-actin promoter, and ICR mice were obtained from ‘Stolbovaya’ breeding center of the Scientific Centre of Biomedical Technologies (Russia). B10.GFP mice were maintained on a C57BL/6J background. Animals were housed in accordance with the European Convention for the Protection of Vertebrate Animals, and all experiments were approved by the Animal Care and Use Committee of Koltzov Institute of Developmental Biology RAS.
Cell isolation and culture
For SLC culture, adult (8–12 weeks) and immature (6 days postpartum (dpp)) testis regions containing the rete testis and adjacent seminiferous tubules were dissected, decapsulated, and digested at 37°С in two steps with collagenase type IV (4 mg/mL, 15 min, Sigma) and 0.125% trypsin with 1 mM EDTA (10 min, Thermo Fisher Sci) for adult tissue, and in three steps with collagenase (15 min), and the mixture of collagenase and hyaluronidase I-S (2 mg/mL; Sigma) repeated twice (25 and 15 min) for pup testes. DNase I (0.04%; Sigma) was added at all digestion steps. Tissue was extensively washed between digestions. Cells were plated at 37°C in Matrigel (Corning)-coated culture plates at 2 × 105 and 5 × 104 cells/cm2 for adult and pup tissue, respectively. The culture medium was DMEM/F12+GlutaMAX (Thermo Fisher Sci) supplemented with sodium pyruvate, insulin-transferrin-selenium (Thermo Fisher Sci), penicillin/streptomycin, and 1% fetal bovine serum (Thermo Fisher Sci) for adult cultures or 10% Knockout Serum Replacement (KSR, Thermo Fisher Sci) for pup cultures. A combination of 10 mM Y-27632 (Abcam), 0.5 mM A-83-01 (Sigma), and 3 mM CHIR99021 (Sigma) (YAC) was added to some adult cultures. After 24 h culture, germ cells were removed from the adult cultures by repeated washing. The medium was changed every 3 days. Cultures of cells from seminiferous tubules (ST cultures) were obtained from testis regions without the rete, and isolated and cultured as described earlier for SLCs of the corresponding age.
For co-culture of SLCs with neonatal testicular cells, SLCs isolated from adult B10.GFP mice were grown in culture with YAC for 9 days, then mixed (at a ratio of 1:10) with testicular cell suspension obtained from ICR pups (4–6-days-old) by 0.125% trypsin digestion (10 min at 37°C), embedded (2 × 105 cells/sample) into a collagen matrix (1.6 mg/mL, Corning) as described previously (Kulibin & Malolina 2016), and cultured onto floating Nuclepore membranes (Sigma, WHA110406) in the same medium as that for pup SLC cultures. GDNF (10 mg/mL, Sigma) was added to the medium for first 3–4 days until tubular structures were formed. The medium was changed every week. Samples were fixed after 14 days in culture.
RNA-sequence analysis
RNA-sequence data collection and analysis were carried out by the First Oncology Research and Advisory Center (Moscow, Russia). Total RNA was extracted from three independent adult SLC and adult ST culture samples (grown for 5 days in the culture medium without YAC) using RNeasy Mini Kit (Qiagen). RNA sequencing was performed on an Illumina HiSeq3000 System, which generated from 20 to 30 million paired-end reads of 50 bp in length for each sample. To map the reads, STAR software was used (Dobin et al. 2013). DESeq2 was used for the data normalization and the quantification of differential expression between SLC and SC culture samples (Love et al. 2014). Genes were considered to be differentially expressed if the Q-value (false discovery rate) was below 0.05.
Laser capture microdissection (LCM)
For LCM SLC and ST cultures, both adult and immature, were used. All cells were grown in Matrigel-coated culture dishes with PEN foil on the bottom (WillCo Wells). Cultured cells were live stained with SYBR green I (1:5000) for 3 min and underwent LCM on a Leica LMD 7000 system as previously described (Podgorny 2013). For procedure details, see Supplementary Fig. 1 (see section on supplementary data given at the end of this article).
Quantitative RT-PCR (qRT-PCR)
To confirm RNA-seq data, qRT-PCR analysis was performed from the same RNA samples. For LCM samples, total RNA was extracted using an RNeasy Micro Kit (Qiagen). Three biological replicates were performed for each group. cDNA was synthesized using an MMLV RT kit (Evrogen, Russia), and real-time PCR was performed in triplicates using SYBR green qPCRmix-HS with ROX (Evrogen) on a StepOnePlus Real-Time PCR System (Applied Biosystems). PCR amplification conditions were as follows: 45 cycles of 95°C for 15 s, 60°C for 30 s. Primer sequences (Supplementary Table 1) were obtained from PrimerBank (Spandidos et al. 2010), and primers were ordered from Evrogen. Hprt was used as a reference gene to calculate ΔCt values. Fold-change of gene expression was calculated using the 2−ΔΔCt method (Livak & Schmittgen 2001). Data were presented as mean ± s.e.m. Statistical significance was determined by a nonparametric Mann–Whitney U test.
Immunofluorescence
Cultured cells were fixed in 4% paraformaldehyde for 10 min. Testes and 3D co-culture samples were fixed in 10% neutral buffered formalin for 24 h at 4°C, dehydrated, embedded in paraffin, and 4 μm sections were cut. Next, immunofluorescence staining was performed. For staining procedure and antibodies used, see Supplementary Table 2. Negative control images for key antibodies are shown in Supplementary Fig. 2.
Culture samples were imaged on a Leica DMI6000 microscope. For triple staining with one anti-mouse (against DMRT1) and two anti-rabbit antibodies one of which labels a nuclear antigen (WT1 or SOX9) and the other cytoplasmic (KRT8 or ACTA2), samples were first stained for a nuclear antigen and photographed with storing reference points and imaged positions in the LAS AF software (Leica), and then re-stained with two other antibodies and imaged again using the saved positions. Subtraction of nuclear staining with anti-rabbit antibodies from the second set of images was performed in the CellProfiler software (Carpenter et al. 2006). Sections were imaged on a Leica TCS SP5 confocal microscope. Image processing, including deconvolution and image stitching, was performed in the LAS AF software.
Cell counts
For cultured cells, at least three biological replicates were performed for each experiment, and at least 40 fields of view were imaged on a 10× objective from each sample. For cell counts in sections, three 6 dpp male mice were analyzed. Three to five equally spaced testis sections with the rete testis were obtained from one testis of each animal and imaged. The percent of cells was calculated using CellProfiler software. Data were presented as mean ± s.e.m. Statistical significance was determined by a nonparametric Mann–Whitney U test.
Results
Transcriptome analysis of SLCs
To determine SLC transcriptional markers and to further investigate the nature of these cells, we examined the transcriptome of adult SLCs using RNA-seq. To do this, we evaluated differential expression between the sub-confluent SLC culture containing 64.8 ± 2.2% SLCs (WT1+ cells in colonies), 1.7 ± 0.4% SCs (WT1+ single cells) and 33.5 ± 1.9% peritubular myoid cells (WT1− cells), and the ST culture containing 4.6 ± 0.7% SCs and 95.4 ± 0.7% peritubular myoid cells. Neither cells from SLC culture nor from ST culture express Hsd3b6 and Hsd17b3 specific for adult Leydig cells (Supplementary Fig. 3A). No cells with lipid droplets characteristic for Leydig cells were observed in SLC culture (Supplementary Fig. 3B). The methods of SLC isolation and culturing were the same as in our previous study (Kulibin & Malolina 2016) except some minor modifications.
According to RNA-seq, the expression of 3156 genes was significantly (Q < 0.05) altered in the SLC culture with 1527 genes upregulated and 1629 genes downregulated (Supplementary Table 3). Due to the substantial number of peritubular myoid cell transcripts obscuring the expression patterns of SLCs, we considered only the top 100 genes upregulated in the SLC culture. For further analysis, we selected some genes reported to be expressed within the urogenital tract and/or be the markers of stem and progenitor cells (Table 1). Krt18 that was not in top 100 genes was also chosen due to its known expression in immature SCs (Sharpe et al. 2003). The differential expression of these genes was validated by qRT-PCR analysis (Supplementary Fig. 3C).
Selected genes upregulated in the adult mouse SLC culture compared to the culture of cells from seminiferous tubules.
| Gene symbol | Gene name | log2 Fold change | Q level | References |
|---|---|---|---|---|
| Pax8 | Paired box 8 | 7.0853 | 2.34E-93 | Wistuba et al. (2007), Sharma et al. (2015) |
| Upk3b | Uroplakin 3B | 5.4633 | 2.24E-135 | Kuriyama et al. (2017) |
| Ndp | Norrie disease (pseudoglioma) (human) | 4.5793 | 1.71E-29 | Mendive et al. (2006), Hoshii et al. (2007), Hsu et al. (2014) |
| Podxl | Podocalyxin-like | 3.9109 | 1.52E-18 | Doyonnas et al. (2005), Lee et al. (2009), Moscoso et al. (2016), Dumont-Lagacé et al. (2017) |
| Elf3 | E74-like factor 3 | 3.8531 | 1.43E-19 | Grassmeyer et al. (2017) |
| Tacstd2 | Tumor-associated calcium signal transducer 2 | 3.5742 | 2.17E-16 | Goldstein et al. (2008), McDougall et al. (2015) |
| Arl4c | ADP-ribosylation factor-like 4C | 3.3599 | 8.89E-56 | Matsumoto et al. (2014) |
| Ptprz1 | Protein tyrosine phosphatase, receptor type Z, polypeptide 1 | 3.2278 | 1.90E-50 | Soh et al. (2007), Dumont-Lagacé et al. (2017), Michelotti et al. (2016) |
| Gdf15 | Growth differentiation factor 15 | 3.2117 | 9.28E-52 | Noorali et al. (2007), Matsumoto et al. (2014) |
| Ppargc1a | Peroxisome proliferative activated receptor, gamma, coactivator 1 alpha | 2.9480 | 2.28E-10 | Bouma et al. (2010) |
| Cdh1 | Cadherin 1 | 2.9236 | 1.95E-11 | Nagasawa et al. (2018) |
| Krt8 | Keratin 8 | 2.3252 | 9.73E-07 | Dinges et al. (1991), Appert et al. (1998), Mandon et al. (2015) |
| Krt18 | Keratin 18 | 1.0661 | 0.042289457 | Dinges et al. (1991), Appert et al. (1998), Sharpe et al. (2003) |
It was previously demonstrated that primary SCs were unstable and became more fibroblast-like within several days in culture (Buganim et al. 2012). Similar changes seemed to take place in the SLC culture as seen by the appearance of ACTA2, a mesenchymal marker, in SLC colonies from day 4 of the culture (Kulibin & Malolina 2016). Here and after, to improve culture conditions for adult SLCs, we added to the medium Rho-associated kinase inhibitor (Y-27632), type 1 TGFb receptor inhibitor (A-83-01), and glycogen synthase kinase-3 inhibitor (CHIR99021), abbreviated as YAC, that are used for stable culturing of stem and progenitor cells and inhibit epithelial-mesenchymal transition (Efe & Ding 2011, Katsuda et al. 2017). We found that YAC prevented the initiation of ACTA2 expression in SLCs (Supplementary Fig. 4), changed their morphology to more epithelial-like (Fig. 1A–D) and dramatically increased the number of SLCs expressing DMRT1 at high and low levels (Fig. 2A and B).
Morphological appearance of SLC colonies. Phase-contrast (A, C and E) and bright-field (B, D and F) images of SLC colonies from adult cultures without YAC (A and B) and with YAC (C and D), and from immature culture (E and F). Cells were stained with hematoxylin in (B, D and F). Arrows point to SLC colonies. Scale bars: 200 µm (A, C); 50 µm (B, D and F); 100 µm (E).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
Morphological appearance of SLC colonies. Phase-contrast (A, C and E) and bright-field (B, D and F) images of SLC colonies from adult cultures without YAC (A and B) and with YAC (C and D), and from immature culture (E and F). Cells were stained with hematoxylin in (B, D and F). Arrows point to SLC colonies. Scale bars: 200 µm (A, C); 50 µm (B, D and F); 100 µm (E).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
Morphological appearance of SLC colonies. Phase-contrast (A, C and E) and bright-field (B, D and F) images of SLC colonies from adult cultures without YAC (A and B) and with YAC (C and D), and from immature culture (E and F). Cells were stained with hematoxylin in (B, D and F). Arrows point to SLC colonies. Scale bars: 200 µm (A, C); 50 µm (B, D and F); 100 µm (E).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
Characterization of adult and immature SLCs. (A and C) Representative colonies of adult (A) and immature (C) SLCs positively stained for WT1 (red) and having heterogeneous DMRT1 expression (green). (A’–A’’’) Higher magnifications of boxed cells with different intensities of a DMRT1 signal, (A’) shows an adult SC outside the colony. (1, 2) Higher magnifications of boxed immature SLCs (1) and SCs (2). Nuclei were counterstained with DAPI (blue). (B) The percent of SLCs with different levels of DMRT1 in the adult cultures with and without YAC, and in the immature culture at day 9 and day 3, respectively (mean ± s.e.m., 3–6 biological replicates). *P < 0.001. (D and E) Expression of SLC markers revealed by RNA-seq analysis in adult (D) and immature (E) SLC colonies obtained by LCM (mean ± s.e.m., three biological replicates). *P < 0.05. Scale bars: 250 µm (A), 200 µm (C), 40 µm (A’–A’’’, 1, 2).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
Characterization of adult and immature SLCs. (A and C) Representative colonies of adult (A) and immature (C) SLCs positively stained for WT1 (red) and having heterogeneous DMRT1 expression (green). (A’–A’’’) Higher magnifications of boxed cells with different intensities of a DMRT1 signal, (A’) shows an adult SC outside the colony. (1, 2) Higher magnifications of boxed immature SLCs (1) and SCs (2). Nuclei were counterstained with DAPI (blue). (B) The percent of SLCs with different levels of DMRT1 in the adult cultures with and without YAC, and in the immature culture at day 9 and day 3, respectively (mean ± s.e.m., 3–6 biological replicates). *P < 0.001. (D and E) Expression of SLC markers revealed by RNA-seq analysis in adult (D) and immature (E) SLC colonies obtained by LCM (mean ± s.e.m., three biological replicates). *P < 0.05. Scale bars: 250 µm (A), 200 µm (C), 40 µm (A’–A’’’, 1, 2).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
Characterization of adult and immature SLCs. (A and C) Representative colonies of adult (A) and immature (C) SLCs positively stained for WT1 (red) and having heterogeneous DMRT1 expression (green). (A’–A’’’) Higher magnifications of boxed cells with different intensities of a DMRT1 signal, (A’) shows an adult SC outside the colony. (1, 2) Higher magnifications of boxed immature SLCs (1) and SCs (2). Nuclei were counterstained with DAPI (blue). (B) The percent of SLCs with different levels of DMRT1 in the adult cultures with and without YAC, and in the immature culture at day 9 and day 3, respectively (mean ± s.e.m., 3–6 biological replicates). *P < 0.001. (D and E) Expression of SLC markers revealed by RNA-seq analysis in adult (D) and immature (E) SLC colonies obtained by LCM (mean ± s.e.m., three biological replicates). *P < 0.05. Scale bars: 250 µm (A), 200 µm (C), 40 µm (A’–A’’’, 1, 2).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
Dense epithelial-like SLC colonies were present in the cultures established from the rete testis region since birth (Fig. 1E and F, 2C). For cells from immature testes, we used serum-free KSR medium that allowed to culture SLCs without the overgrowth of fibroblast-like cells. The proportion of DMRT1+ SLCs in the immature culture was similar to the adult culture grown in the medium with YAC (Fig. 2B and C).
To confirm the identity of adult SLCs grown in the medium with YAC and immature (6 dpp) SLCs, SLC colonies were isolated using LCM procedure and qRT-PCR analysis was performed indicating that all genes selected from RNA-seq data were upregulated when compared with cultured adult and immature SCs respectively (Fig. 2D and E). The only exception was Tacstd2 whose expression was similar in immature SLCs and SCs.
Adult and immature SLCs express PAX8, CDH1, and KRT8
The expression of three SLC markers, namely the most highly expressed gene Pax8, Cdh1, and Krt8, was examined in the adult and immature (6 dpp) SLC cultures by immunofluorescence. PAX8 expression was detected in all SLC colonies identified by SOX9 or WT1 staining in the adult confluent culture grown for 9 days (Fig. 3A). The proportion of PAX8+ cells relative to the total SLC number was more than 90% in the adult culture (Fig. 3B). Because we were not able to accurately define boundaries of 6 dpp SLC colonies without DMRT1 staining, PAX8+ immature SLCs were not counted. However, most cells in the SLC colonies were positively stained for PAX8 in the confluent immature culture grown for 3 days (Fig. 3C), and there were no colonies without PAX8+ cells.
Immunofluorescence detection of some genes revealed by RNA-seq in the adult and immature SLC cultures. (A and C) SLCs (SOX9+ colonies, red) were labeled with a PAX8 antibody (green) in the adult (A) and immature cultures (C). (B) The percent of PAX8+ SLCs in the adult culture (mean ± s.e.m., three biological replicates). (D and E) SLCs (WT1+ colonies, red) were positively stained for CDH1 (green) in the adult (D) and immature cultures (E). (F and G) SLCs (WT1+ colonies, red) were labeled with a KRT8 antibody (yellow) in the adult (F) and immature cultures (G); some of them co-expressed DMRT1 (green, indicated by arrows). Arrowheads point to adult SCs. Scale bars: 200 µm (A, C, D, E, F and G).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
Immunofluorescence detection of some genes revealed by RNA-seq in the adult and immature SLC cultures. (A and C) SLCs (SOX9+ colonies, red) were labeled with a PAX8 antibody (green) in the adult (A) and immature cultures (C). (B) The percent of PAX8+ SLCs in the adult culture (mean ± s.e.m., three biological replicates). (D and E) SLCs (WT1+ colonies, red) were positively stained for CDH1 (green) in the adult (D) and immature cultures (E). (F and G) SLCs (WT1+ colonies, red) were labeled with a KRT8 antibody (yellow) in the adult (F) and immature cultures (G); some of them co-expressed DMRT1 (green, indicated by arrows). Arrowheads point to adult SCs. Scale bars: 200 µm (A, C, D, E, F and G).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
Immunofluorescence detection of some genes revealed by RNA-seq in the adult and immature SLC cultures. (A and C) SLCs (SOX9+ colonies, red) were labeled with a PAX8 antibody (green) in the adult (A) and immature cultures (C). (B) The percent of PAX8+ SLCs in the adult culture (mean ± s.e.m., three biological replicates). (D and E) SLCs (WT1+ colonies, red) were positively stained for CDH1 (green) in the adult (D) and immature cultures (E). (F and G) SLCs (WT1+ colonies, red) were labeled with a KRT8 antibody (yellow) in the adult (F) and immature cultures (G); some of them co-expressed DMRT1 (green, indicated by arrows). Arrowheads point to adult SCs. Scale bars: 200 µm (A, C, D, E, F and G).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
Both adult and immature SLC colonies from the confluent cultures had heterogeneous CDH1 expression. The staining of adult cells was faint but clear (Fig. 3D), whereas immature colonies often exhibited high-level expression of CDH1 (Fig. 3E). KRT8+ SLCs were rarely observed in the adult confluent culture, but they were more often detected in the sub-confluent culture grown for 5 days; some of them co-expressed DMRT1 (Fig. 3F). KRT8+ SLCs with and without DMRT1 expression were present in the immature SLC cultures (Fig. 3G). PAX8, KRT8, and CDH1-positive cells were never observed in ST cultures, except a few faintly stained KRT8+ SCs in the immature culture.
Because CDH1 and KRT8 expression was low in the adult confluent and sub-confluent cultures we additionally performed the immunofluorescent staining of SLCs at 1 day of culture. SLCs were recognized by WT1 or SOX9 staining and typically produced clusters of several cells, whereas SCs, which were also positive for WT1 and SOX9, were rare, sparsely dispersed and excluded from the analysis by their characteristic nuclear morphology. Most SLCs highly expressed CDH1 and KRT8 (Fig. 4A and C). Triple immunofluorescence revealed different SLC phenotypes, including CDH1+KRT8+, CDH1+KRT8−, CDH1−KRT8−, but never CDH1−KRT8+ cells (Fig. 4A). The proportion of DMRT1+ SLCs relative to the total SLC number was low (9.3 ± 0.7%) and more than 80% of DMRT1+ SLCs co-expressed KRT8 (Fig. 4B and C).
Immunofluorescence examination of adult SLCs at 1 day of culture. (A) Clusters of SLCs, small WT1+ cells (red), were labeled with KRT8 (yellow) and CDH1 (green) antibodies. Asterisks indicate CDH1−KRT8− SLCs. (B) Some KRT8+ SLCs co-expressed DMRT1 (green, indicated by dots). Compare to a SC (arrowhead, inset) that was KRT8 negative and had characteristic nuclear morphology. (C) The percent of CDH1+ and KRT8+ cells among SLCs, and the percent of KRT8+ cells among DMRT1+ SLCs (mean ± s.e.m., three biological replicates). (D) Staining of the culture with an ace-TUB antibody; a WT1+KRT8− cell with a high ace-TUB signal (green) is indicated by an arrow. Nuclei in (A, B and D) were counterstained with DAPI (blue). Scale bars: 50 µm (A, B and D); 25 µm (inset in B).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
Immunofluorescence examination of adult SLCs at 1 day of culture. (A) Clusters of SLCs, small WT1+ cells (red), were labeled with KRT8 (yellow) and CDH1 (green) antibodies. Asterisks indicate CDH1−KRT8− SLCs. (B) Some KRT8+ SLCs co-expressed DMRT1 (green, indicated by dots). Compare to a SC (arrowhead, inset) that was KRT8 negative and had characteristic nuclear morphology. (C) The percent of CDH1+ and KRT8+ cells among SLCs, and the percent of KRT8+ cells among DMRT1+ SLCs (mean ± s.e.m., three biological replicates). (D) Staining of the culture with an ace-TUB antibody; a WT1+KRT8− cell with a high ace-TUB signal (green) is indicated by an arrow. Nuclei in (A, B and D) were counterstained with DAPI (blue). Scale bars: 50 µm (A, B and D); 25 µm (inset in B).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
Immunofluorescence examination of adult SLCs at 1 day of culture. (A) Clusters of SLCs, small WT1+ cells (red), were labeled with KRT8 (yellow) and CDH1 (green) antibodies. Asterisks indicate CDH1−KRT8− SLCs. (B) Some KRT8+ SLCs co-expressed DMRT1 (green, indicated by dots). Compare to a SC (arrowhead, inset) that was KRT8 negative and had characteristic nuclear morphology. (C) The percent of CDH1+ and KRT8+ cells among SLCs, and the percent of KRT8+ cells among DMRT1+ SLCs (mean ± s.e.m., three biological replicates). (D) Staining of the culture with an ace-TUB antibody; a WT1+KRT8− cell with a high ace-TUB signal (green) is indicated by an arrow. Nuclei in (A, B and D) were counterstained with DAPI (blue). Scale bars: 50 µm (A, B and D); 25 µm (inset in B).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
We also performed immunofluorescent staining for acetylated alpha-tubulin (ace-TUB), which has been reported to be a marker of SCs from the transitional zones (Nagasawa et al. 2018). At 1 day of culture, most SLCs were not stained for ace-TUB, or the signal was low. However, some WT1+KRT8− cells displayed bright ace-TUB staining (Fig. 4D, an arrow). Later, all cells in culture, including SLCs and peritubular myoid cells, began to stain positive for ace-TUB, so we were unable to determine whether SCs from the transitional zones contributed to SLC colonies.
The rete testis is the only testicular region co-expressing PAX8, CDH1, and KRT8
To localize SLCs in the testis, we examined the region of the adult mouse testis containing the rete testis and adjacent STs. SOX9 and WT1 were expressed in both SCs and epithelial cells of the rete testis, although at substantially lower levels in the latter (Supplementary Fig. 5A and B). These findings were in conflict with our previous reports (Kulibin & Malolina 2016, Malolina & Kulibin 2017) showing no SOX9 or WT1 expression in the adult rete. Potential reasons for this include the sensitivity of the antibodies used and the change from frozen to paraffin sections in the present study. The same reasons explain the fact that here we were able to detect low DMRT1 expression in the SCs of transitional zones (Supplementary Fig. 5C), which was in contrast to our previous study.
Double staining for PAX8 and SOX9 showed that PAX8 was expressed in rete cells but not in SCs from ST and transitional zones (Fig. 5A). The same results were obtained for KRT8 and CDH1 (Fig. 5B). Note that some rete cells were positive for CDH1 but negative for KRT8 (Fig. 5B’), which corresponded to the data from the adult culture described earlier. Ace-TUB detected in the cytoplasm of SCs from transitional zones was never co-localized with KRT8 (Fig. 5C), just as it was in vitro.
SLC markers revealed by RNA-seq analysis were detected in the epithelium of the adult rete testis by immunofluorescent staining. (A) SOX9+ (red) epithelial cells of the rete testis (RT) were labeled with a PAX8 antibody (green). ST, seminiferous tubules; TZ, SCs from transitional zones. (B) A stitched image of the rete testis epithelium positively stained for KRT8 (green) and CDH1 (red). (B’) shows a higher magnification of a boxed area with CDH1+KRT8+ cells (dots) and CDH1+KRT8− cells (arrows). (C) SCs from transitional zones were brightly stained for ace-TUB (green) but were KRT8 (red) negative. Nuclei were counterstained with DRAQ5 (blue). Scale bars: 50 µm (A, B, C, D and E); 25 µm (B’).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
SLC markers revealed by RNA-seq analysis were detected in the epithelium of the adult rete testis by immunofluorescent staining. (A) SOX9+ (red) epithelial cells of the rete testis (RT) were labeled with a PAX8 antibody (green). ST, seminiferous tubules; TZ, SCs from transitional zones. (B) A stitched image of the rete testis epithelium positively stained for KRT8 (green) and CDH1 (red). (B’) shows a higher magnification of a boxed area with CDH1+KRT8+ cells (dots) and CDH1+KRT8− cells (arrows). (C) SCs from transitional zones were brightly stained for ace-TUB (green) but were KRT8 (red) negative. Nuclei were counterstained with DRAQ5 (blue). Scale bars: 50 µm (A, B, C, D and E); 25 µm (B’).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
SLC markers revealed by RNA-seq analysis were detected in the epithelium of the adult rete testis by immunofluorescent staining. (A) SOX9+ (red) epithelial cells of the rete testis (RT) were labeled with a PAX8 antibody (green). ST, seminiferous tubules; TZ, SCs from transitional zones. (B) A stitched image of the rete testis epithelium positively stained for KRT8 (green) and CDH1 (red). (B’) shows a higher magnification of a boxed area with CDH1+KRT8+ cells (dots) and CDH1+KRT8− cells (arrows). (C) SCs from transitional zones were brightly stained for ace-TUB (green) but were KRT8 (red) negative. Nuclei were counterstained with DRAQ5 (blue). Scale bars: 50 µm (A, B, C, D and E); 25 µm (B’).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
Similar to the staining pattern observed in adult testes, PAX8 (Fig. 6A) and CDH1 (Fig. 6B and B’) were highly expressed in the immature (6 dpp) rete epithelium and were not detected in SCs of STs. Some immature rete cells were positively stained for KRT8, although the signal was also observed in sporadic cells from STs (Fig. 6B and B’). Broad immunofluorescence signal for ace-TUB detected in both STs and the rete testis made it impossible to localize transitional zones at this age (Fig. 6C).
SLC markers were detected in the immature rete testis by immunofluorescent staining. (A) SOX9+ (red) epithelium of the rete testis (RT) was positively stained for PAX8 (green). An arrowhead points to a seminiferous tubule with SOX9+PAX8− SCs. (B) Cells of the rete testis expressed CDH1 (green) and KRT8 (red). (B’) shows a higher magnification of a boxed area, arrows point to a KRT8 signal in seminiferous tubules (ST). (C) An ace-TUB signal (green) was detected in the rete testis marked by CDH1 staining (yellow) and in seminiferous tubules; a WT1 antibody (red) labeled both rete cells and SCs. Nuclei were counterstained with DRAQ5 (blue). Images in (A and B) were stitched together from individual images. Scale bars: 50 µm (A, B’ and C); 100 µm (B).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
SLC markers were detected in the immature rete testis by immunofluorescent staining. (A) SOX9+ (red) epithelium of the rete testis (RT) was positively stained for PAX8 (green). An arrowhead points to a seminiferous tubule with SOX9+PAX8− SCs. (B) Cells of the rete testis expressed CDH1 (green) and KRT8 (red). (B’) shows a higher magnification of a boxed area, arrows point to a KRT8 signal in seminiferous tubules (ST). (C) An ace-TUB signal (green) was detected in the rete testis marked by CDH1 staining (yellow) and in seminiferous tubules; a WT1 antibody (red) labeled both rete cells and SCs. Nuclei were counterstained with DRAQ5 (blue). Images in (A and B) were stitched together from individual images. Scale bars: 50 µm (A, B’ and C); 100 µm (B).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
SLC markers were detected in the immature rete testis by immunofluorescent staining. (A) SOX9+ (red) epithelium of the rete testis (RT) was positively stained for PAX8 (green). An arrowhead points to a seminiferous tubule with SOX9+PAX8− SCs. (B) Cells of the rete testis expressed CDH1 (green) and KRT8 (red). (B’) shows a higher magnification of a boxed area, arrows point to a KRT8 signal in seminiferous tubules (ST). (C) An ace-TUB signal (green) was detected in the rete testis marked by CDH1 staining (yellow) and in seminiferous tubules; a WT1 antibody (red) labeled both rete cells and SCs. Nuclei were counterstained with DRAQ5 (blue). Images in (A and B) were stitched together from individual images. Scale bars: 50 µm (A, B’ and C); 100 µm (B).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
All these results suggest that SLCs mostly reside in the rete testis, whereas the contribution of SCs from transitional zones remains unclear. It is also unknown whether all rete cells or only part of them are able to form SLC colonies.
Rete testis cells and cultured SLCs express genes characteristic of SCs
The main feature of cultured SLCs indicating their relationship to SCs is DMRT1 expression. So we examined DMRT1 expression in the rete testis by immunofluorescence and found that 42.6 ± 3.2 % rete cells (identified by co-expression of SOX9 and CDH1) were positive for DMRT1 in the immature testis (Fig. 7A), which corresponded to the number of DMRT1+ SLCs in immature culture. Some DMRT1+ rete cells expressed KRT8 (Fig. 7B), as did SLCs in culture. On the contrary, only rare adult rete cells were positively stained for DMRT1; they were KRT8 positive (Fig. 7C, C’ and C’’) or negative (Fig. 7D and D’). That correlated with the low number of DMRT1+ SLCs at day 1 of adult culture.
Cells in the adult and immature rete testis epithelium expressed DMRT1. (A) A stitched image of the immature rete testis (RT) marked by co-expression of CDH1 (red) and SOX9 (blue) with many DMRT1+ (green) rete cells. ST, seminiferous tubules. Asterisks indicate CDH1+DMRT1+ germ cells in the rete epithelium that were SOX9-negative. (B) Some DMRT1+ (green) rete cells expressed KRT8 (red) in the immature testis (indicated by arrows). (C and D) Rare rete cells expressed DMRT1 (green) in the adult testis. TZ, transitional zones. (C’, C’’ and D’) show higher magnifications of boxed areas with DMRT1+ rete cells positive (C’, C’’, asterisks) or negative (D’, an arrowhead) for KRT8 (red). Nuclei were counterstained with DRAQ5 (blue). Scale bars: 50 µm (A, C and D); 25 µm (B, inset).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
Cells in the adult and immature rete testis epithelium expressed DMRT1. (A) A stitched image of the immature rete testis (RT) marked by co-expression of CDH1 (red) and SOX9 (blue) with many DMRT1+ (green) rete cells. ST, seminiferous tubules. Asterisks indicate CDH1+DMRT1+ germ cells in the rete epithelium that were SOX9-negative. (B) Some DMRT1+ (green) rete cells expressed KRT8 (red) in the immature testis (indicated by arrows). (C and D) Rare rete cells expressed DMRT1 (green) in the adult testis. TZ, transitional zones. (C’, C’’ and D’) show higher magnifications of boxed areas with DMRT1+ rete cells positive (C’, C’’, asterisks) or negative (D’, an arrowhead) for KRT8 (red). Nuclei were counterstained with DRAQ5 (blue). Scale bars: 50 µm (A, C and D); 25 µm (B, inset).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
Cells in the adult and immature rete testis epithelium expressed DMRT1. (A) A stitched image of the immature rete testis (RT) marked by co-expression of CDH1 (red) and SOX9 (blue) with many DMRT1+ (green) rete cells. ST, seminiferous tubules. Asterisks indicate CDH1+DMRT1+ germ cells in the rete epithelium that were SOX9-negative. (B) Some DMRT1+ (green) rete cells expressed KRT8 (red) in the immature testis (indicated by arrows). (C and D) Rare rete cells expressed DMRT1 (green) in the adult testis. TZ, transitional zones. (C’, C’’ and D’) show higher magnifications of boxed areas with DMRT1+ rete cells positive (C’, C’’, asterisks) or negative (D’, an arrowhead) for KRT8 (red). Nuclei were counterstained with DRAQ5 (blue). Scale bars: 50 µm (A, C and D); 25 µm (B, inset).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
AMH is a SC-specific protein highly expressed in embryonic and neonatal testes but disappearing when the testis matures (Sharpe et al. 2003). We found that AMH was also present in some cells of the immature rete testis identified by co-expression of SOX9 and CDH1 (Fig. 8A) or SOX9 and PAX8 (Fig. 8B). Triple immunofluorescence for AMH, DMRT1, and a rete marker CDH1 revealed AMH+DMRT1+ rete cells preferentially localized near seminiferous tubule entries (Fig. 8C) and AMH−DMRT1+ rete cells broadly distributed in the rete testis except in the region adjacent to efferent ducts (Fig. 8C).
Cells in the immature rete testis and SLCs in culture expressed SC genes. (A and A’) AMH+ (yellow) cells in the rete testis (RT) positively stained for CDH1 (green) are indicated by asterisks, SOX9 (red) labels SCs and rete cells. Arrows point to AMH+ cells not expressing CDH1 (i.e., SCs). (B and B’) AMH+ (yellow) cells in the rete positively stained for PAX8 (green) are indicated by arrowheads. ST, seminiferous tubules. (A’ and B’) Higher magnifications of boxed areas represent maximum projections of serial confocal optical sections; orthogonal projections of areas denoted by yellow lines are shown in the right and bottom panels. (C) A stitched image of the immature (6 dpp) rete testis stained for AMH (blue), CDH1 (red), and DMRT1 (green). Dots point to AMH+ rete cells. A dotted line outlines part of the rete without DMRT1+ and AMH+ cells. ED, efferent duct. Nuclei were counterstained with DRAQ5 (blue). (D and E) Expression of the selected genes in the adult (D) and immature (E) SLC colonies obtained by LCM. Dashed lines indicate the gene expression levels in SCs. The data are presented as the mean ± s.e.m. from three biological replicates. *P < 0.05. Scale bars: 100 µm (A, B and C); 20 µm (A’ and B’).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
Cells in the immature rete testis and SLCs in culture expressed SC genes. (A and A’) AMH+ (yellow) cells in the rete testis (RT) positively stained for CDH1 (green) are indicated by asterisks, SOX9 (red) labels SCs and rete cells. Arrows point to AMH+ cells not expressing CDH1 (i.e., SCs). (B and B’) AMH+ (yellow) cells in the rete positively stained for PAX8 (green) are indicated by arrowheads. ST, seminiferous tubules. (A’ and B’) Higher magnifications of boxed areas represent maximum projections of serial confocal optical sections; orthogonal projections of areas denoted by yellow lines are shown in the right and bottom panels. (C) A stitched image of the immature (6 dpp) rete testis stained for AMH (blue), CDH1 (red), and DMRT1 (green). Dots point to AMH+ rete cells. A dotted line outlines part of the rete without DMRT1+ and AMH+ cells. ED, efferent duct. Nuclei were counterstained with DRAQ5 (blue). (D and E) Expression of the selected genes in the adult (D) and immature (E) SLC colonies obtained by LCM. Dashed lines indicate the gene expression levels in SCs. The data are presented as the mean ± s.e.m. from three biological replicates. *P < 0.05. Scale bars: 100 µm (A, B and C); 20 µm (A’ and B’).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
Cells in the immature rete testis and SLCs in culture expressed SC genes. (A and A’) AMH+ (yellow) cells in the rete testis (RT) positively stained for CDH1 (green) are indicated by asterisks, SOX9 (red) labels SCs and rete cells. Arrows point to AMH+ cells not expressing CDH1 (i.e., SCs). (B and B’) AMH+ (yellow) cells in the rete positively stained for PAX8 (green) are indicated by arrowheads. ST, seminiferous tubules. (A’ and B’) Higher magnifications of boxed areas represent maximum projections of serial confocal optical sections; orthogonal projections of areas denoted by yellow lines are shown in the right and bottom panels. (C) A stitched image of the immature (6 dpp) rete testis stained for AMH (blue), CDH1 (red), and DMRT1 (green). Dots point to AMH+ rete cells. A dotted line outlines part of the rete without DMRT1+ and AMH+ cells. ED, efferent duct. Nuclei were counterstained with DRAQ5 (blue). (D and E) Expression of the selected genes in the adult (D) and immature (E) SLC colonies obtained by LCM. Dashed lines indicate the gene expression levels in SCs. The data are presented as the mean ± s.e.m. from three biological replicates. *P < 0.05. Scale bars: 100 µm (A, B and C); 20 µm (A’ and B’).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
We also selected several other genes important for SC function and evaluated their expression in the adult and immature SLC colonies by qRT-PCR using the same LCM samples that were employed for measuring the expression of SLC markers. Decreased Dmrt1 levels were found to be the feature of both adult and immature SLCs (Fig. 8D,E), which consistent with heterogeneous staining for DMRT1 in the SLC cultures. Wt1, Sox9, Kitl, and Trf were expressed in SLCs at levels similar to SCs or even higher. Gdnf levels were not significantly changed between adult SLCs and SCs (Fig. 8D), although these were slightly decreased in immature SLCs (Fig. 8E). Adult SLCs had decreased expression of Nr5a1, Dhh, Shbg, and especially Inha, whereas immature SLCs expressed Nr5a1 and Dhh at even higher levels than SCs of the corresponding age. Shbg and Inha levels were decreased in immature SLCs but not as much as in the adult cells (Fig. 8D and E). Both adult and immature SLCs had elevated expression of Nr0b1 (Dax1), an orphan nuclear receptor involved in sex determination and gonadal development (Fig. 8D and E).
SLCs interact with germ cells in co-culture with neonatal testicular cells
DMRT1+CDH1+SOX9− cells identified as spermatogonia were detected in the epithelium of the immature rete testis (Fig. 7A). Germ cells disappeared from the rete during testis maturation along with DMRT1 loss. These findings suggest the ability of rete cells to support germ cell function and the importance of DMRT1 for this process.
To test if cultured SLCs are able to interact with germ cells, we mixed SLCs obtained from adult GFP mice and grown in culture for 9 days and testicular cell suspension isolated from ICR neonatal mice and established 3D co-culture. GDNF was added to the cells for first 3 days to increase germ cell viability. Immunofluorescent staining demonstrated GFP+ SLCs expressing SOX9 and DMRT1 (Fig. 9A) and forming, together with neonatal cells, tubular structures surrounded by peritubular myoid cells labeled by ACTA2 (Fig. 9B). Some SLCs were found to be in direct contact with germ cells expressing DDX4 (Fig. 9C) or markers of meiosis initiation and progression, STRA8 (Fig. 9D) and SCP3 (Fig. 9E), respectively. The most differentiated germ cells appeared to progress until the meiotic prophase (Fig. 9E’). Staining for cleaved caspase-3, an apoptosis marker, confirmed the viability of germ cells. Apoptotic cells were few and located at the center of 3D samples (Fig. 9F), whereas germ cells were in the periphery (Fig. 9G).
Adult SLCs contacted with germ cells in the 3D co-culture with immature testicular cells. (A and B) The most cells from the SLC culture identified by GFP staining (green) were SLCs (SOX9+ cells, red, A) with different levels of DMRT1 expression (yellow), and not peritubular myoid cells (ACTA2+ cells, red, B). Asterisks indicate SLCs. A dot points to a GFP+SOX9− cell. (C, D and E) Some germ cells (arrowheads) positively stained for DDX4 (red, C), a pre-meiotic marker STRA8 (red, D), and a meiotic marker SCP3 (red, E) were closely associated with GFP+ cells. Dotted lines in (B and E) outline tubule-like structures. (C, D and E) represent maximum projections of serial confocal optical sections; orthogonal projections of areas denoted by yellow lines are shown in the right and bottom panels. (E’) represents a higher magnification of one of the optical sections in (E, arrow) and shows a spermatocyte in metaphase with an adjacent GFP+ cell. (F and G) Single apoptotic cells labeled by anti-cCASP3 antibody (red, F) and germ cells labeled by anti-DDX4 antibody (red, G) were located at different sites, at the center and in the periphery of 3D-culture samples respectively. An inset in (F) shows a higher magnification of a boxed area. Nuclei were counterstained with DRAQ5 (blue). Scale bars: 20 µm (A, B, C, D and E); 50 µm (F and G); 5 µm (E’); 25 µm (inset).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
Adult SLCs contacted with germ cells in the 3D co-culture with immature testicular cells. (A and B) The most cells from the SLC culture identified by GFP staining (green) were SLCs (SOX9+ cells, red, A) with different levels of DMRT1 expression (yellow), and not peritubular myoid cells (ACTA2+ cells, red, B). Asterisks indicate SLCs. A dot points to a GFP+SOX9− cell. (C, D and E) Some germ cells (arrowheads) positively stained for DDX4 (red, C), a pre-meiotic marker STRA8 (red, D), and a meiotic marker SCP3 (red, E) were closely associated with GFP+ cells. Dotted lines in (B and E) outline tubule-like structures. (C, D and E) represent maximum projections of serial confocal optical sections; orthogonal projections of areas denoted by yellow lines are shown in the right and bottom panels. (E’) represents a higher magnification of one of the optical sections in (E, arrow) and shows a spermatocyte in metaphase with an adjacent GFP+ cell. (F and G) Single apoptotic cells labeled by anti-cCASP3 antibody (red, F) and germ cells labeled by anti-DDX4 antibody (red, G) were located at different sites, at the center and in the periphery of 3D-culture samples respectively. An inset in (F) shows a higher magnification of a boxed area. Nuclei were counterstained with DRAQ5 (blue). Scale bars: 20 µm (A, B, C, D and E); 50 µm (F and G); 5 µm (E’); 25 µm (inset).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
Adult SLCs contacted with germ cells in the 3D co-culture with immature testicular cells. (A and B) The most cells from the SLC culture identified by GFP staining (green) were SLCs (SOX9+ cells, red, A) with different levels of DMRT1 expression (yellow), and not peritubular myoid cells (ACTA2+ cells, red, B). Asterisks indicate SLCs. A dot points to a GFP+SOX9− cell. (C, D and E) Some germ cells (arrowheads) positively stained for DDX4 (red, C), a pre-meiotic marker STRA8 (red, D), and a meiotic marker SCP3 (red, E) were closely associated with GFP+ cells. Dotted lines in (B and E) outline tubule-like structures. (C, D and E) represent maximum projections of serial confocal optical sections; orthogonal projections of areas denoted by yellow lines are shown in the right and bottom panels. (E’) represents a higher magnification of one of the optical sections in (E, arrow) and shows a spermatocyte in metaphase with an adjacent GFP+ cell. (F and G) Single apoptotic cells labeled by anti-cCASP3 antibody (red, F) and germ cells labeled by anti-DDX4 antibody (red, G) were located at different sites, at the center and in the periphery of 3D-culture samples respectively. An inset in (F) shows a higher magnification of a boxed area. Nuclei were counterstained with DRAQ5 (blue). Scale bars: 20 µm (A, B, C, D and E); 50 µm (F and G); 5 µm (E’); 25 µm (inset).
Citation: Reproduction 158, 5; 10.1530/REP-19-0183
Discussion
Previously, we reported that highly proliferative cells resembling SCs (SLCs) could be isolated from the region of the adult mouse testis containing the rete testis and adjacent ST (Kulibin & Malolina 2016). In the present study, we performed RNA-seq analysis of the adult SLC culture and identified SLC-specific genes. We also improved SLC culture conditions by supplementation of YAC, a combination of small molecules, and demonstrated the presence of SLC colonies in the culture from immature testes. Both adult SLC colonies grown with YAC and immature SLC colonies expressed marker genes selected from RNA-seq data. Their feature was the high numbers of cells positive for SC protein DMRT1.
Three SLC genes selected from RNA-seq data, Pax8, Cdh1, and Krt8, were further investigated. PAX8 is a transcription factor belonging to the PAX protein family. PAX8, along with another member of the PAX family, PAX2, are the earliest specific markers of the urogenital system and demonstrate some functional redundancy during pro- and mesonephros development (Sharma et al. 2015). However, according to our RNA-seq data, only Pax8 is expressed in SLCs. PAX8 expression was reported in the human rete testis (Ozcan et al. 2011). In the mouse, PAX8 was demonstrated in the efferent ducts and the epididymis; its deficiency leads to their absence or severe defects (Wistuba et al. 2007). Here, we found that PAX8 was expressed in all epithelial cells of the adult rete testis but not in other testicular cells. CDH1 is a component of adherens junctions prominently expressed in epithelial tissues (Schneider & Kolligs 2015). The previous study (Nagasawa et al. 2018) and our current findings demonstrated that the only somatic cells expressing CDH1 in the mouse testis were epithelial cells of the rete testis. The third marker examined, KRT8, in pair with KRT18, forms intermediate filaments in many simple epithelia (Owens & Lane 2003). Its expression was reported previously in the human rete testis (Dinges et al. 1991). KRT8 immunoreactivity was also observed in fetal and neonatal SCs in the mouse (Appert et al. 1998). Here, we showed the presence of KRT8 in the adult rete testis but, in contrast to PAX8 and CDH1, it was expressed in many but not all rete cells. We were also able to find a few rete cells expressing DMRT1. These findings indicate the heterogeneity of rete cells. This feature is even more prominent in the immature testis where the number of DMRT1+ rete cells is dramatically increased and some of them express another SC protein AMH, that is consistent with the previous study (Rebourcet et al. 2014).
As the rete testis is the only testicular region where PAX8, CDH1, KRT8, and DMRT1 are co-expressed we conclude that SLCs are mostly localized in the rete testis epithelium. Like the rete cells, cultured SLCs are heterogeneous. Especially intriguing is their heterogeneity for DMRT1 expression. We showed previously that some adult SLCs were positive for DMRT1 (Kulibin & Malolina 2016). Here we found that many more DMRT1+ SLCs are present in immature culture that correlates with a higher number of DMRT1+ rete cells in the immature testis. The proportion of DMRT1+ adult SLCs can be increased to the immature value by YAC, a combination of three small molecules. Small molecules are used for stable culturing of embryonic and tissue-specific stem cells and for facilitating somatic cell reprogramming (Efe & Ding 2011). They can have a general effect on cellular metabolism or modulate specific signaling pathways. YAC was reported to stimulate proliferation of mouse hepatocytes in vitro, improve their survival, and induce the expression of progenitor markers (Katsuda et al. 2017). It seems that the effects of YAC on SLCs are similar. We speculate that YAC induces DMRT1 expression in adult SLCs. Another explanation is the preferential proliferation of DMRT1+ SLCs initially presenting in culture. However, it is less possible as the proliferative rates of DMRT1− and DMRT1+ SLCs did not differ, at least at days 5 and 9 of culture (data not shown).
The issue that have remained unclear is the contribution of SCs from transitional zones to SLCs. There is growing evidence indicating that these SCs have unique characteristics. They can proliferate in adult testes of Syrian hamsters and rats (Aiyama et al. 2015, Figueiredo et al. 2016). Some of them do not express maturation markers GATA4 and AR (Figueiredo et al. 2016). According to our data SCs from mouse transitional zones have decreased levels of DMRT1. They do not proliferate in the adult testis (Kulibin & Malolina 2016) but exit the cell cycle later than other SCs when the testis matures (Malolina & Kulibin 2017). Here we demonstrated that SCs from transitional zones are not positive for SLC markers PAX8, CDH1, and KRT8 and so could not contribute to most of SLCs. However, we do not exclude the possibility that they can be among PAX8− cells present in a few number in SLC colonies or initiate the expression of SLC genes.
qPCR analysis of the adult and immature SLC colonies showed that, besides Dmrt1, they expressed some other SC genes, including Nr5a1 and Dhh, at the levels comparable to SCs. SLCs appeared to be able to support germ cells in a similar way as SCs. They expressed Gdnf and Kitl encoding growth factors that regulate the maintenance and survival of germ cells (Oatley & Brinster 2012). In the 3D co-culture with neonatal testicular cells, SLCs closely interacted with germ cells. All these similarities between SLCs and SCs suggest that rete cells could share a common origin with SCs. The fact that many rete cells in the immature testis are positive for SC proteins DMRT1 and AMH confirms this hypothesis.
Immature SLCs and adult SLCs cultured with YAC are similar in DMRT1 expression but, according to qPCR analysis, immature SLCs more closely resemble SCs of the corresponding age that could reflect progressive diversification of rete cells and SCs during postnatal development. Another feature of both adult and immature SLCs demonstrated by qPCR analysis is the elevated expression of Nr0b1 (Dax1). Nr0b1 deficiency leads to infertility in male mice because of the rete testis obstruction by aberrantly located proliferating SCs (Jeffs et al. 2001). NR0B1 was reported to repress Amh in the fetal testis and immature SC cultures (Tremblay & Viger 2001, Bowles et al. 2018). Based on these findings, we speculate that NR0B1 could play an important role in the specification of rete cells by inhibiting SC genes such as Amh. Another transcription factor that could be involved is PAX8.
Further studies on the embryonic and postnatal testis are required to elucidate the origin of rete cells and the mechanisms of their specification.
Supplementary data
This is linked to the online version of the paper at https://doi.org/10.1530/REP-19-0183.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
The work was supported in part by the program of Presidium of the Russian Academy of Sciences for 2018–2020 ‘Basic research for development of biomedical technologies’ (0088-2019-0013) and the Russian Foundation for Basic Research (16-34-60119). The study on the 3D co-culture of SLCs with neonatal testicular cells was funded by the Russian Science Foundation (17-74-10076). The research was done using equipment of the Core Centrum of Institute of Developmental Biology RAS.
Author Contribution Statement
E A M and A Y K designed research and analyzed data; E A M performed immunofluorescence and LCM procedure; A Y K performed cell culturing, PCR analysis, and cell counting; E A M wrote the manuscript with input from A Y K.
References
Ahmed EA, Barten-van Rijbroek AD, Kal HB, Sadri-Ardekani H, Mizrak SC, van Pelt AM & de Rooij DG 2009 Proliferative activity in vitro and DNA repair indicate that adult mouse and human Sertoli cells are not terminally differentiated, quiescent cells. Biology of Reproduction 1084–1091. (https://doi.org/10.1095/biolreprod.108.071662)
Aiyama Y, Tsunekawa N, Kishi K, Kawasumi M, Suzuki H, Kanai-Azuma M, Kurohmaru M & Kanai Y 2015 A niche for GFRα1-positive spermatogonia in the terminal segments of the seminiferous tubules in hamster testes. Stem Cells 2811–2824. (https://doi.org/10.1002/stem.2065)
Appert A, Fridmacher V, Locquet O & Magre S 1998 Patterns of keratins 8, 18 and 19 during gonadal differentiation in the mouse: sex- and time-dependent expression of keratin 19. Differentiation; Research in Biological Diversity 273–284. (https://doi.org/10.1046/j.1432-0436.1998.6350273.x)
Bouma GJ, Hudson QJ, Washburn LL & Eicher EM 2010 New candidate genes identified for controlling mouse gonadal sex determination and the early stages of granulosa and Sertoli cell differentiation. Biology of Reproduction 380–389. (https://doi.org/10.1095/biolreprod.109.079822)
Bowles J, Feng CW, Ineson J, Miles K, Spiller CM, Harley VR, Sinclair AH & Koopman P 2018 Retinoic acid antagonizes testis development in mice. Cell Reports 1330–1341. (https://doi.org/10.1016/j.celrep.2018.06.111)
Buganim Y, Itskovich E, Hu YC, Cheng AW, Ganz K, Sarkar S, Fu D, Welstead GG, Page DC & Jaenisch R 2012 Direct reprogramming of fibroblasts into embryonic Sertoli-like cells by defined factors. Cell Stem Cell 373–386. (https://doi.org/10.1016/j.stem.2012.07.019)
Capel B 2017 Vertebrate sex determination: evolutionary plasticity of a fundamental switch. Nature Reviews Genetics 675–689. (https://doi.org/10.1038/nrg.2017.60)
Carpenter AE, Jones TR, Lamprecht MR, Clarke C, Kang IH, Friman O, Guertin DA, Chang JH, Lindquist RA & Moffat J et al.2006 CellProfiler: image analysis software for identifying and quantifying cell phenotypes. Genome Biology R100. (https://doi.org/10.1186/gb-2006-7-10-r100)
Dinges HP, Zatloukal K, Schmid C, Mair S & Wirnsberger G 1991 Co-expression of cytokeratin and vimentin filaments in rete testis and epididymis. An immunohistochemical study. Virchows Archiv 119–127. (https://doi.org/10.1007/BF01600287)
Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, Batut P, Chaisson M & Gingeras R 2013 STAR: ultrafast universal RNA-seq aligner. Bioinformatics 15–21. (https://doi.org/10.1093/bioinformatics/bts635)
Doyonnas R, Nielsen JS, Chelliah S, Drew E, Hara T, Miyajima A & McNagny KM 2005 Podocalyxin is a CD34-related marker of murine hematopoietic stem cells and embryonic erythroid cells. Blood 4170–4178. (https://doi.org/10.1182/blood-2004-10-4077)
Dumont-Lagacé M, Gerbe H, Daouda T, Laverdure JP, Brochu S, Lemieux S, Gagnon É & Perreault C 2017 Detection of quiescent radioresistant epithelial progenitors in the adult thymus. Frontiers in Immunology 1717. (https://doi.org/10.3389/fimmu.2017.01717)
Dym M 1974 The fine structure of monkey Sertoli cells in the transitional zone at the junction of the seminiferous tubules with the tubuli recti. American Journal of Anatomy 1–25. (https://doi.org/10.1002/aja.1001400102)
Efe JA & Ding S 2011 The evolving biology of small molecules: controlling cell fate and identity. Philosophical Transactions of the Royal Society of London: Series B, Biological Sciences 2208–2221. (https://doi.org/10.1098/rstb.2011.0006)
Figueiredo AF, França LR, Hess RA & Costa GM 2016 Sertoli cells are capable of proliferation into adulthood in the transition region between the seminiferous tubules and the rete testis in Wistar rats. Cell Cycle 2486–2496. (https://doi.org/10.1080/15384101.2016.1207835)
Goldstein AS, Lawson DA, Cheng D, Sun W, Garraway IP & Witte ON 2008 Trop2 identifies a subpopulation of murine and human prostate basal cells with stem cell characteristics. PNAS 20882–20887. (https://doi.org/10.1073/pnas.0811411106)
Grassmeyer J, Mukherjee M, deRiso J, Hettinger C, Bailey M, Sinha S, Visvader JE, Zhao H, Fogarty E & Surendran K 2017 Elf5 is a principal cell lineage specific transcription factor in the kidney that contributes to Aqp2 and Avpr2 gene expression. Developmental Biology 77–89. (https://doi.org/10.1016/j.ydbio.2017.02.007)
Hoshii T, Takeo T, Nakagata N, Takeya M, Araki K & Yamamura K 2007 LGR4 regulates the postnatal development and integrity of male reproductive tracts in mice. Biology of Reproduction 303–313. (https://doi.org/10.1095/biolreprod.106.054619)
Hsu PJ, Wu FJ, Kudo M, Hsiao CL, Hsueh AJ & Luo CW 2014 A naturally occurring Lgr4 splice variant encodes a soluble antagonist useful for demonstrating the gonadal roles of Lgr4 in mammals. PLoS ONE e106804. (https://doi.org/10.1371/journal.pone.0106804)
Jeffs B, Meeks JJ, Ito M, Martinson FA, Matzuk MM, Jameson JL & Russell LD 2001 Blockage of the rete testis and efferent ductules by ectopic Sertoli and Leydig cells causes infertility in Dax1-deficient male mice. Endocrinology 4486–4495. (https://doi.org/10.1210/endo.142.10.8447)
Katsuda T, Kawamata M, Hagiwara K, Takahashi RU, Yamamoto Y, Camargo FD & Ochiya T 2017 Conversion of terminally committed hepatocytes to culturable bipotent progenitor cells with regenerative capacity. Cell Stem Cell 41–55. (https://doi.org/10.1016/j.stem.2016.10.007)
Kulibin AY & Malolina EA 2016 Only a small population of adult Sertoli cells actively proliferates in culture. Reproduction 271–281. (https://doi.org/10.1530/REP-16-0013)
Kuriyama S, Tamiya Y & Tanaka M 2017 Spatiotemporal expression of UPK3B and its promoter activity during embryogenesis and spermatogenesis. Histochemistry and Cell Biology 17–26. (https://doi.org/10.1007/s00418-016-1486-8)
Lee RH, Seo MJ, Pulin AA, Gregory CA, Ylostalo J & Prockop DJ 2009 The CD34-like protein PODXL and alpha6-integrin (CD49f) identify early progenitor MSCs with increased clonogenicity and migration to infarcted heart in mice. Blood 816–826. (https://doi.org/10.1182/blood-2007-12-128702)
Lei N, Hornbaker KI, Rice DA, Karpova T, Agbor VA & Heckert LL 2007 Sex-specific differences in mouse DMRT1 expression are both cell type- and stage-dependent during gonad development. Biology of Reproduction 466–475. (https://doi.org/10.1095/biolreprod.106.058784)
Livak KJ & Schmittgen TD 2001 Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method. Methods 402–408. (https://doi.org/10.1006/meth.2001.1262)
Love MI, Huber W & Anders S 2014 Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biology 550. (https://doi.org/10.1186/s13059-014-0550-8)
Malolina EA & Kulibin AY 2017 Rete testis and the adjacent seminiferous tubules during postembryonic development in mice. Russian Journal of Developmental Biology 385–392. (https://doi.org/10.1134/S1062360417060029)
Mandon M, Hermo L & Cyr DG 2015 Isolated rat epididymal basal cells share common properties with adult stem cells. Biology of Reproduction 115. (https://doi.org/10.1095/biolreprod.115.133967)
Matson CK, Murphy MW, Sarver AL, Griswold MD, Bardwell VJ & Zarkower D 2011 DMRT1 prevents female reprogramming in the postnatal mammalian testis. Nature 101–104. (https://doi.org/10.1038/nature10239)
Matsumoto S, Fujii S, Sato A, Ibuka S, Kagawa Y, Ishii M & Kikuchi A 2014 A combination of Wnt and growth factor signaling induces Arl4c expression to form epithelial tubular structures. EMBO Journal 702–718. (https://doi.org/10.1002/embj.201386942)
McDougall AR, Tolcos M, Hooper SB, Cole TJ & Wallace MJ 2015 Trop2: from development to disease. Developmental Dynamics 99–109. (https://doi.org/10.1002/dvdy.24242)
Mendive F, Laurent P, Van Schoore G, Skarnes W, Pochet R & Vassart G 2006 Defective postnatal development of the male reproductive tract in LGR4 knockout mice. Developmental Biology 421–434. (https://doi.org/10.1016/j.ydbio.2005.11.043)
Michelotti GA, Tucker A, Swiderska-Syn M, Machado MV, Choi SS, Kruger L, Soderblom E, Thompson JW, Mayer-Salman M & Himburg HA et al.2016 Pleiotrophin regulates the ductular reaction by controlling the migration of cells in liver progenitor niches. Gut 683–692. (https://doi.org/10.1136/gutjnl-2014-308176)
Mital P, Kaur G, Bowlin B, Paniagua NJ, Korbutt GS & Dufour JM 2014 Nondividing, postpubertal rat Sertoli cells resumed proliferation after transplantation. Biology of Reproduction 13. (https://doi.org/10.1095/biolreprod.113.110197)
Moscoso I, Tejados N, Barreiro O, Sepúlveda P, Izarra A, Calvo E, Dorronsoro A, Salcedo JM, Sádaba R & Díez-Juan A et al.2016 Podocalyxin-like protein 1 is a relevant marker for human c-kit(pos) cardiac stem cells. Journal of Tissue Engineering and Regenerative Medicine 580–590. (https://doi.org/10.1002/term.1795)
Nagasawa K, Imura-Kishi K, Uchida A, Hiramatsu R, Kurohmaru M & Kanai Y 2018 Regionally distinct patterns of STAT3 phosphorylation in the seminiferous epithelia of mouse testes. Molecular Reproduction and Development 262–270. (https://doi.org/10.1002/mrd.22962)
Nicholls PK, Stanton PG, Chen JL, Olcorn JS, Haverfield JT, Qian H, Walton KL, Gregorevic P & Harrison CA 2012 Activin signaling regulates Sertoli cell differentiation and function. Endocrinology 6065–6077. (https://doi.org/10.1210/en.2012-1821)
Noorali S, Kurita T, Woolcock B, de Algara TR, Lo M, Paralkar V, Hoodless P & Vielkind J 2007 Dynamics of expression of growth differentiation factor 15 in normal and PIN development in the mouse. Differentiation; Research in Biological Diversity 325–336. (https://doi.org/10.1111/j.1432-0436.2006.00142.x)
Nykänen M 1979 Fine structure of the transitional zone of the rat seminiferous tubule. Cell and Tissue Research 441–454. (https://doi.org/10.1007/bf00234189)
Oatley JM & Brinster RL 2012 The germline stem cell niche unit in mammalian testes. Physiological Reviews 577–595. (https://doi.org/10.1152/physrev.00025.2011)
Owens DW & Lane EB 2003 The quest for the function of simple epithelial keratins. BioEssays 748–758. (https://doi.org/10.1002/bies.10316)
Ozcan A, Shen SS, Hamilton C, Anjana K, Coffey D, Krishnan B & Truong LD 2011 PAX 8 expression in non-neoplastic tissues, primary tumors, and metastatic tumors: a comprehensive immunohistochemical study. Modern Pathology 751–764. (https://doi.org/10.1038/modpathol.2011.3)
Podgorny OV 2013 Live cell isolation by laser microdissection with gravity transfer. Journal of Biomedical Optics 55002. (https://doi.org/10.1117/1.JBO.18.5.055002)
Raymond CS, Murphy MW, O'Sullivan MG, Bardwell VJ & Zarkower D 2000 Dmrt1, a gene related to worm and fly sexual regulators, is required for mammalian testis differentiation. Genes and Development 2587–2595. (https://doi.org/10.1101/gad.834100)
Rebourcet D, O'Shaughnessy PJ, Pitetti JL, Monteiro A, O'Hara L, Milne L, Tsai YT, Cruickshanks L, Riethmacher D & Guillou F et al.2014 Sertoli cells control peritubular myoid cell fate and support adult Leydig cell development in the prepubertal testis. Development 2139–2149. (https://doi.org/10.1242/dev.107029)
Schneider MR & Kolligs FT 2015 E-cadherin's role in development, tissue homeostasis and disease: insights from mouse models. BioEssays 294–304. (https://doi.org/10.1002/bies.201400141)
Sharma R, Sanchez-Ferras O & Bouchard M 2015 Pax genes in renal development, disease and regeneration. Seminars in Cell and Developmental Biology 97–106. (https://doi.org/10.1016/j.semcdb.2015.09.016)
Sharpe RM, McKinnell C, Kivlin C & Fisher JS 2003 Proliferation and functional maturation of Sertoli cells, and their relevance to disorders of testis function in adulthood. Reproduction 769–784. (https://doi.org/10.1530/rep.0.1250769)
Soh BS, Song CM, Vallier L, Li P, Choong C, Yeo BH, Lim EH, Pedersen RA, Yang HH & Rao M et al.2007 Pleiotrophin enhances clonal growth and long-term expansion of human embryonic stem cells. Stem Cells 3029–3037. (https://doi.org/10.1634/stemcells.2007-0372)
Spandidos A, Wang X, Wang H & Seed B 2010 PrimerBank: a resource of human and mouse PCR primer pairs for gene expression detection and quantification. Nucleic Acids Research D792–D799. (https://doi.org/10.1093/nar/gkp1005)
Tarulli GA, Stanton PG, Lerchl A & Meachem SJ 2006 Adult Sertoli cells are not terminally differentiated in the Djungarian hamster: effect of FSH on proliferation and junction protein organization. Biology of Reproduction 798–806. (https://doi.org/10.1095/biolreprod.105.050450)
Tarulli GA, Stanton PG & Meachem SJ 2012 Is the adult Sertoli cell terminally differentiated? Biology of Reproduction 13, 1–13,11. (https://doi.org/10.1095/biolreprod.111.095091)
Tarulli GA, Stanton PG, Loveland KL, Rajpert-De Meyts E, McLachlan RI & Meachem SJ 2013 A survey of Sertoli cell differentiation in men after gonadotropin suppression and in testicular cancer. Spermatogenesis e24014. (https://doi.org/10.4161/spmg.24014)
Tremblay JJ & Viger RS 2001 Nuclear receptor Dax-1 represses the transcriptional cooperation between GATA-4 and SF-1 in Sertoli cells. Biology of Reproduction 1191–1199. (https://doi.org/10.1095/biolreprod64.4.1191)
Wistuba J, Mittag J, Luetjens CM, Cooper TG, Yeung CH, Nieschlag E & Bauer K 2007 Male congenital hypothyroid Pax8-/- mice are infertile despite adequate treatment with thyroid hormone. Journal of Endocrinology 99–109. (https://doi.org/10.1677/JOE-06-0054)
Zarkower D 2013 DMRT genes in vertebrate gametogenesis. Current Topics in Developmental Biology 327–356. (https://doi.org/10.1016/B978-0-12-416024-8.00012-X)
Zhang ZH, Hu ZY, Song XX, Xiao LJ, Zou RJ, Han CS & Liu YX 2004 Disrupted expression of intermediate filaments in the testis of rhesus monkey after experimental cryptorchidism. International Journal of Andrology 234–239. (https://doi.org/10.1111/j.1365-2605.2004.00477.x)
Zhang XS, Zhang ZH, Jin X, Wei P, Hu XQ, Chen M, Lu CL, Lue YH, Hu ZY & Sinha Hikim AP et al.2006 Dedifferentiation of adult monkey Sertoli cells through activation of extracellularly regulated kinase 1/2 induced by heat treatment. Endocrinology 1237–1245. (https://doi.org/10.1210/en.2005-0981)
