• Spermatogonial stem cells;
  • spermatogenesis;
  • early culture effects;
  • in vitro meiosis


  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. References

ABSTRACT: Isolation and culture of spermatogonial stem cells (SSCs) has become an approach to study the milieu and the factors controlling their expansion and differentiation. Traditional conventional cell culture does not mimic the complex situation in the seminiferous epithelium providing a basal, intraepithelial, and adluminal compartment to the developing male germ cells. SSCs are located in specific stem cell niches whose features and functional parameters are thus far poorly understood. It was the aim of this study to isolate SSCs and to explore their expansion and differentiation potential in a novel three-dimensional Soft-Agar-Culture-System (SACS). This system provides three-dimensional structural support and multiple options for manipulations through the addition of factors, cells, or other changes. The system has revolutionized research on blood stem cells by providing a tool for clonal analysis of expanding and differentiating blood cell lineages. In our studies, SSCs are enriched using Gfrα-1 as a specific surface marker and magnetic-activated cell sorting as a separation approach. At termination of the culture, we determined the type and number of germ cells obtained after the first 24 hours of culture. We also determined cell types and numbers in expanding cell clones of differentiating germ cells during the subsequent 15 days of culture. We analyzed a supportive effect of somatic cell lineages added to the solid part of the culture system. We conclude that our enrichment and culture approach is highly useful for exploration of SSC expansion and have found indications that the system supports differentiation up to the level of postmeiotic germ cells.

Analysis of male germ cell proliferation and differentiation under various in vitro conditions has increasingly come into focus. Thereby, substances influencing maturation processes have been intensively investigated. These studies aimed to specifically analyze spermatogonial stem cell (SSC) physiology in different approaches using feeder layer (Nagano et al, 2003), serum-free (Creemers et al, 2002; Kubota et al, 2004b; Kanatsu-Shinohara et al, 2005a), or feeder-free culture conditions (Kanatsu-Shinohara et al, 2005a).

The physiologic conditions needed to maintain and differentiate cultured SSCs were previously analyzed in conventional culture systems by addition of testicular cells (Lee et al, 1997; Nagano et al, 2003) and/or certain factors (eg, leukemia inhibitory factor [LIF], de Miguel et al, 1996; Kanatsu-Shinohara et al, 2003; glia cell line—derived neurotrophic factor [GDNF], Meng et al, 2000; Kanatsu-Shinohara et al, 2003; Kubota et al, 2004b; basic fibroblast growth factor, Kanatsu-Shinohara et al, 2003; Kubota et al, 2004b; or stem cell factor [SCF], Allard et al, 1996; Blanchard et al, 1998; de Rooij et al, 1998) that had been identified for spermatogonial propagation. All of these factors have been proposed to be crucial for premeiotic germ cell development.

The conditions allowing male germ cells to enter meiosis are unknown. However, entry into meiosis relies on the integrity of the testicular microenvironment, as it is easily achieved in organ culture (Schlatt et al, 1999) but rarely observed in cell culture. Therefore, we and others assume that testicular somatic cells create unique physical and paracrine support for the developing germ cells, allowing them to enter meiosis (Hofmann et al, 1992; Lee et al, 1997; Nagano et al, 2003). In vivo, the seminiferous tubule offers three compartments for germ cells. 1) The basal compartment, offering physical contacts with the basement membrane, peritubular cells, Sertoli cells, and other premeiotic germ cells. Here, the germ cell receives paracrine and endocrine signals from the interstitium. 2) The intraepithelial compartment, offering only contact with the Sertoli cells and other meiotic and postmeiotic germ cells. 3) The adluminal compartment, allowing contact with Sertoli cells and postmeiotic germ cells, as well as signal molecules from the luminal fluid. The stem cell niches are part of the basal compartment, which offers the most versatile compartment within the seminiferous tubules. Stem cell niches could be established through specific extracellular matrix—specific contacts or specific signaling cascades and will provide specific physical support and environmental features allowing recognition and settlement of SSCs. They also might provide crucial factors needed for maintenance of pluripotent abilities of SSCs (Spradling et al, 2001).

In general, mammalian SSC culture experiments have been performed in conventional “two-dimensional” cell culture approaches using culture dishes or flasks (eg, Dirami et al, 1999; Feng et al, 2002; Hasthorpe, 2003; Nagano et al, 2003; Kanatsu-Shinohara et al, 2004a, 2005a,b). The physical support for SSCs in a conventional culture is different from the natural niche environment zof the seminiferous epithelium, and it remains conjectural whether stem cell niches can be reestablished in a monolayer culture of Sertoli cells. Hence, a three-dimensional culture approach might offer more appropriate opportunities for cell growth.

The Soft-Agar-Culture-System (SACS), a three-dimensional cell culture approach, was first established to characterize clonal expansion of bone marrow cells and to identify factors involved in the regulation of their proliferation and differentiation (Lin et al, 1975; Quaroni et al, 1979; Huleihel et al, 1993; Horowitz et al, 2000). Applied to testicular stem cells, it might also provide an improved structural environment for clonal expansion of germ cells. Here, we are testing this hypothesis to explore whether SACS can be used as an innovative methodology for analysis of germ cell development. Previously published studies demonstrated the importance of a three-dimensional structure for the differentiation of mouse and human testicular cells and the support of in vitro spermatogenesis (Lee et al, 2006, 2007).

To isolate spermatogonia from testicular tissues, particularly from immature animals, several approaches are available, such as gravity sedimentation to separate cells of different size in percoll (Koh et al, 2004) or with the STAPUT technique (Dirami et al, 1999), fluorescence-activated cell sorting (Shinohara et al, 2000; Fujita et al, 2005; Guan et al, 2006), or magnetic-activated cell sorting (MACS; von Schönfeldt et al, 1999; Buageaw et al, 2005). The MACS system is fast and causes minimal stress to the spermatogonial cells during isolation and enrichment. One of the most crucial steps to enrich SSCs is the availability of highly specific markers. Signaling pathway proteins or receptors exclusively expressed on the surface of spermatogonia can be specifically utilized for cell separation by MACS (von Schönfeldt et al, 1999; Buageaw et al, 2005). To isolate the population of undifferentiated spermatogonia in mice, marker proteins such as GDNF family receptor-alpha-1 (Gfrα-1; Meng et al, 2000; von Schönfeldt et al, 2004), Cd-9 (tetraspanin transmembrane protein; Kanatsu-Shinohara et al, 2004b), and Thy-1 (glycosyl phosphatidylinositol—anchored surface antigen; Kubota et al, 2004a; Oatley et al, 2007) have been suggested to show prevalence for this cell type. MAC-sorted cells have previously been cultured using standard procedures. These studies showed the possibility of maintaining proliferating SSCs in vitro for up to 6 months (Kubota et al, 2004b). However, the in vitro production of meiotic and postmeiotic germ cells, which would indicate an optimal culture condition not only for SSCs, but also for survival and differentiation of their progeny, turned out to be extremely difficult. Thus far, no culture system was able to maintain the viability of differentiating spermatogonia and to support the meiotic and postmeiotic spermatogenic progress. In this study, we aimed to characterize a novel three-dimensional culture system and determine the survival, expansion, and differentiation of germ cells.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. References


Testes were obtained from juvenile (10 days post partum [dpp]) and mature (30 dpp) CD-1 mice from our institutional colony. All animal experimental procedures were performed in accordance with the German federal law on the handling of experimental animals (animal license no. A87/05).

Testicular Cell Isolation

Testicular cells were isolated on day 10 pp from CD-1 mice (≥20 animals per isolation). Testes were removed from the scrotum and decapsulated. The tissue was minced with fine scissors and transferred into culture medium (Dulbecco modified Eagle medium DMEM/HAM F12; Gibco, Gaithersburg, Maryland) containing collagenase type 1A (1 mg/mL; Sigma Chemical Co, St Louis, Missouri) and DNase (0.5 mg/mL; Sigma). Digestion was performed at 37°C for 10 minutes in a shaking water bath operated at 110 cycles per minute. After this step, we obtained a fraction of tubular fragments and single cells, which were separated by sedimentation at unit gravity. To obtain an enriched fraction of interstitial cells, no DNase was added to the collagenase, the supernatant was removed after 10 minutes, and the cell fraction was washed and stored in ice-cold DMEM/HAM F12.

To obtain a fraction of tubular cells (designated unsorted fraction) consisting mainly of Sertoli cells and germ cells, the fragments of seminiferous tubules obtained after the first digestion step were washed once in DMEM and further digested in a mixture of collagenase type I (1 mg/mL; Sigma), DNase (0.5 mg/mL; Sigma), and hyaluronidase (0.5 mg/mL; Sigma; Wistuba et al, 2002). The single-cell suspension (Table 1) was washed successively with medium and phosphate-buffered saline (PBS) containing 2 mM EDTA (Sigma) and 0.5% fetal calf serum (Gibco; Figure 1A). Efficiency of the digestion, cell number, and concentration were established microscopically using a Thoma chamber (Hecht, Sondheim, Germany).

Table 1. . List of cell types of different cell suspensions used for SACSa
Cell FractionsCell Types
  1. a The separation of the unsorted cell fraction with MACS resulted in 2 different tubular cell suspensions, the depleted and the enriched fraction. The percentage of Gfrα-1—positive spermatogonia is decreased in the depleted fraction, whereas the enriched fraction contained more Gfrα-1—positive spermatogonial cells. The testicular cell fraction is a mixture of tubular and interstitial cells. The different cell types of 10 dpp murine testis are also shown in Figure 2A through C (Cd-9 expression) and Figure 2G through I (Gfrα-1 expression).

Unsorted fraction (MACS)Tubular cells (undifferentiated spermatogonia up to leptotene spermatocyes and Sertoli cells)
Enriched fraction (MACS)Tubular cells (undifferentiated spermatogonia [enriched: 42%-54%] up to leptotene spermatocytes and Sertoli cells)
Depleted fraction (MACS)Tubular cells (undifferentiated spermatogonia [depleted: 11%-19%] up to leptotene spermatocytes and Sertoli cells)
Testicular cell fractionTubular and interstitial cells (undifferentiated spermatogonia up to leptotene spermatocytes, peritubular cells, macrophages, Sertoli cells, and Leydig cells)

Figure 1. . Scheme of the Soft-Agar-Culture-System (SACS) in combination with vacuum filtration, mRNA/total RNA analysis, and fixation for immunohistochemistry. The scheme contains 3 parts with the time course of the experiment. The cell separation and the magnetic-activated cell sorting (MACS) is described in A, the Soft-Agar-Culture in B, and the analysis by vacuum filtration, μMACS/RNA isolation and staining of paraffin embedded SACS sections in C.

Download figure to PowerPoint

Magnetic Labeling and Separation of Cells

Aliquots of cell suspensions from the unsorted fraction with a concentration of around 7.5 × 107 cells/mL (Table 2) were subjected to indirect labeling using a previously described protocol based on primary polyclonal antibodies against Gfrα-1, Cd-9, and Thy-1, and secondary or tertiary anti-rabbit or anti-biotin antibodies carrying ferromagnetic particles (von Schönfeldt et al, 1999). In brief, the cells were incubated with a polyclonal rabbit anti—Gfrα-1 immunoglobulin G (IgG) antibody (H-70, diluted 1:20; Santa Cruz Biotechnology, Santa Cruz, California), polyclonal rabbit anti—Cd-9 IgG antibody (H-110, diluted 1:20; Santa Cruz Biotechnology), or monoclonal mouse anti—Thy-1 IgG antibody (HIS51, diluted 1:20; Santa Cruz Biotechnology) for 15 minutes at 6°C–10°C. Afterwards, cells were washed with PBS (supplemented with EDTA and fetal calf serum as described above), labeled with goat anti-rabbit IgG MicroBeads (dilution 1:5; Miltenyi, Bergisch Gladbach, Germany) or anti-biotin MicroBeads (dilution 1:5; Miltenyi), and washed again. Before the use of anti-biotin MicroBeads, the cells were incubated with anti-rabbit IgG biotin conjugate (B-8895; Sigma) or anti-mouse IgG biotin conjugate (B-0529; Sigma) for 15 minutes. A separation column (MS separation column; Miltenyi) was placed in a strong magnetic field and flushed with 500 μL degassed buffer. The Gfrα-1-labeled cell suspension (Table 1) was resuspended in degassed buffer and poured into the column reservoir. Gfrα-1-positive cells were retained in the magnetic field within the matrix of the column, whereas nonlabeled cells passed through and were collected and designated as a depleted cell fraction (Table 1; Figure 1A). This depleted fraction was added in coculture experiments as somatic cell fraction to support spermatogonia in the SACS. To deplete unlabeled cells from the magnetic fraction, the column was rinsed 3 times with 500 μL degassed buffer. Cells eluting in these washing steps were added to the depleted fraction. In order to retrieve the enriched fraction (Table 1), the column was removed from the magnet, and 500 μL degassed buffer was added to the reservoir. The cells were flushed out of the column using a plunger. Cell size and nuclear size and shape were evaluated and documented under phase-contrast microscopy and were used as criteria to determine the homogeneity/heterogeneity of the unfixed fresh cell suspensions. Viability of MAC-sorted cells was microscopically evaluated using Trypan blue staining.

Table 2. . Primer sequences and sizes for different murine spermatogenic stages used for analysis of total and mRNA expressionsa
Spermatogenic Cell Stage and MarkerPrimerProduct Size
  1. a Spermatogonial markers (undifferentiated and differentiated): Oct3/4, C-kit, Gfrα-1, Cd-9, and α-6-integrin; meiotic germ cell stages: Prohibitin, Scp-3, and Srf-1; postmeiotic germ cell stages: Ldh, Protamine-2, and Sp-10; Sertoli cells: Abp; peritubular cells: α-smooth muscle; positive control: β-actin.

Spermatogonia (undifferentiated and differentiated)  
    Oct3/4Forward 5′-AGAAGGAGCTAGAACAGTTTGC-3′416 bp
    KitForward 5′-GCATCACCATCAAAAACGTG-3′331 bp
    Cd-9Forward 5′-ATGGCTTTGAGTGTTTCCCGCT-3′374 bp
    Gfrα-1Forward 5′-GGCCTACTCGGGACTGATTGG-3′462 bp
    α-6-integrinForward 5′-AGGAGTCGCGGGATATCTTT-3′502 bp
    DazlForward 5′-TTCAGGCATATCCTCCTTATC-3′262 bp
    ProhibitinForward 5′-GTGGCGTACAGGACATTGTG-3′306 bp
    Scp-3Forward 5′-ACAACAAGAGGAAATACAGAA-3′618 bp
    Srf-1Forward 5′-TCCATTCAGCACCTTCAACA-3′303 bp
Spermatids (postmeiotic)  
    LdhForward 5′-GCACGGCAGTCTTTTCCTTAGC-5′585 bp
    Protamine-2Forward 5′-GGCCACCACCACCACAGACACAGGCG-3′188 bp
    Sp-10Forward 5′-GGAGCACCACCAGGTCAG-3′734 bp
Stertoli cells  
    AbpForward 5′-GGAGAAGAGAGACTCTGTGG-3′900 bp
Peritubular cells  
    α-Smooth muscleForward 5′-CGATAGAACACGGCATCATC-3′524 bp
Positive control  
    β-actinForward 5′-AGAGGGAAATCGTGCGTGAC-3′463 bp

Flow Cytometry

The efficiency of spermatogonial enrichment was quantitatively assessed by flow cytometry. Forward and sideward scatter were used to gate cell populations. This gating by size and granularity allowed the exclusion of FITC-positive cell debris. Fractions from the unsorted, enriched, and depleted cell suspensions (Table 1) were stained with FITC-conjugated anti-rabbit antibody (F-0382; Sigma) for 30 minutes at 6°C–8°C. To determine the proportion of FITC-positive cells, the cells were analyzed on a Beckman Coulter flow cytometer FC500 (Krefeld, Germany) equipped with a 15-mW argon-ion laser at an excitation wavelength of 488 nm. The green signals of FITC plotted on a log scale of each cell fraction were collected using a 520-band pass filter (505–545 nm). A marker was set in the FITC histogram as the cutoff between background signals and positive staining, which was determined by comparison with the control sample. A minimum of 105 cells was analyzed in each run (according to von Schönfeldt et al, 1999).


Tissue samples were fixed in Bouin solution for up to 12 hours before being transferred into 70% ethanol, and were routinely embedded in paraffin using an automated processor. Cultured cells (within the agar phase) were fixed in 4% paraformalde-hyde (PFA) for 24 hours at 6°C–8°C before transfer into 30% (24 hours) and 50% (24 hours) ethanol and embedding as described before. Tissue and cultured cells were cut into sections of 5–7 μm and were immunohistochemically stained for Cd9, Gfrα-1 (spermatogonial markers), cAMP reponse element modulator (Crem; postmeitotic spermatids), and 5-bromodesoxyuridine (BrdU; proliferation marker; Figure 1C). Polyclonal primary antibodies against the peptide ETQEDAQKILQEAEKLNYKDKKLN (common to all 3 human BOULE isoforms) were used for detection of murine Boule protein (provided by R.A. Reijo Pera, San Francisco, California). Briefly, sections were deparaffinized in paraclear and rehydrated in a graded series of ethanol. For antigen retrieval, sections were heated in a microwave oven in Glycin/HCl buffer (50 mM, pH 3.5) for 12 minutes at 80°C. Endogenous peroxidase activity was quenched by treatment with hydrogen peroxide (3% for 5 minutes), followed by blocking of nonspecific antibodybinding with 5% normal horse serum supplemented with bovine serum albumin (BSA; 0.1%) for 20 minutes at room temperature. All antibodies were diluted in Tris buffered saline (TBS)/BSA (0.1%). The slides were incubated with a primary antibody (rabbit anti-Cd9 antibody [H-110; 1:50], rabbit anti—Gfrα-1 [H-70; 1:50], rabbit anti—Crem-1 [X-12; 1:50]; Santa Cruz Biotechnology; rabbit anti-BrdU [Bu20a; 1:30]; Sigma-Aldrich, Taufkirchen, Germany) and a polyclonal primary antibody against the peptide ETQEDAQKILQEAEKLNYKDKKLN (Boule detection: 1:300) at room temperature in a humidified chamber for 1 hour and rinsed in TBS (10 mM TBS, 150 mM NaCl, pH 7.6) for 3 × 5 minutes between each of the following incubations. Sections incubated in TBS/BSA without primary antibody served as negative control. Juvenile and adult testicular tissues were immunostained using an LSAB2 kit (DAKO Cytomation, Hamburg, Germany). Washing steps followed incubation with primary antibody and were carried out in TBS (2 × 5 minutes). Afterwards, the sections were incubated with biotinylated swine—anti-rabbit IgGs (15 minutes), washed, and covered with streptavidin-horseradish-peroxidase (HRP) solution (15 minutes), and staining was finally visualized using 3,3-diaminobenzidine tetrahydrochloride (DAB) in urea buffer for 5 to 20 minutes (Sigma-Aldrich). Positive staining appeared as a brown precipitate in the cells. Vacuum-filtrated cell colonies were stained with LSAB2 kit under identical conditions, as described before.

Staining for apoptosis was performed by the DeadEnd Colorimetric TUNEL System (Promega, Madison, Wisconsin). Sections were deparaffinized in paraclear and rehydrated in a graded series of ethanol. After washing with 0.9% NaCl and PBS for 5 minutes, all sections were fixed before and after incubation with proteinase K (20 μg/mL) for 15 minutes in 4% PFA. The recombinant terminal deoxynucleodityl transferase (rTdT) reaction was maintained for 1 hour in a humidified chamber, followed by 15 minutes of twofold SSC incubation. All sections were immunostained with streptavidin horseradishperoxidase solution after washing with PBS for 30 minutes. To visualize positive TUNEL-stained cells, DAB in urea buffer was added for 8 minutes. All sections were counterstained in hematoxylin, mounted, and analyzed by light microscopy. Additionally, the analysis of anti-BrdU was counterstained with Hoechst 33528 (Sigma) for 15 minutes. These sections were analyzed by light and fluorescence microscopy.

To analyze Gfrα-1 expression in unsorted and enriched MAC-sorted fractions, single-cell solutions were stained for Gfrα-1 with an FITC-labeled secondary antibody (Sigma) in combination with Hoechst 33528 for 30 minutes. Immunohistochemical results were documented by digital imaging using a fluorescence microscope (Axiovert 200; Zeiss, Oberkochen, Germany).

The expression of Gfrα-1 on the cell surface of undifferentiated spermatogonia was evaluated by confocal microscopy (TCS SL; Leica, Wetzlar, Germany).


The enriched fraction (Table 1) was used for culture in the gel phase of SACS. The cells were added to the gel-agar medium (0.35% [w/v]) settled on a solid-agar base (0.5% [w/v]; Figure 1B; Lin et al, 1975; Kimball et al, 1978; Hofmann et al, 1992; Huleihel et al, 1993). Depending on the experimental setup, the solid base was either empty or supplemented with cells from the depleted fraction (Table 1). To establish the final concentrations of agar and cells, 0.7% (w/v) agar and 1.0% (w/v) agar were dissolved in distilled water to prepare the gel and solid phases, respectively (Fisher Scientific, Loughborough, United Kingdom). This solution was mixed with the same volume of DMEM high glucose (Gibco, pH 7.4) to achieve a final concentration of 0.35% and 0.5% (Figure 1B). Cell suspensions were added to the DMEM prior to mixing with the agar. The agar and the cells in DMEM were mixed at 37°C, avoiding heat-induced cellular stress and premature coagulation of the agar. Culture conditions were 35°C in 5% CO2. For standard cell culture experiments, regular 24-well plates (Nunc, Wiesbaden, Germany) and 24-well plates with standard Transwell inserts (Corning, New York) were used. To investigate cell proliferation, BrdU (B-5002; Sigma) was added in a final concentration of 100 M to the cell/DMEM suspension before mixing with the agar solution.

Vacuum Filtration

The gel-agar phase containing cultured cells was separated from the solid-agar phase by pipetting and subsequent vacuum filtration using Whatman 47 filters (Whatman, Maidstone, United Kingdom) with a pore size of 0.2 μm. After filtration, cells were fixed on the filter material in 4% PFA and washed twice with PBS. After washing, cell nuclei were stained with Hoechst 33258 for 30 minutes before they were washed again. For evaluation of the cultured cells, 10 micrograph images at a fivefold magnification (Axiovert microscope, CCD camera; Zeiss) of each experiment were scored for cell numbers (Figure 1C), and the total cell number per filter was calculated. Three filters per time point were evaluated in the 24-hour approach, and 12 filters per time point in the 1- to 16-day approach.

Total RNA Extraction, cDNA Synthesis, and Reverse Transcription—Polymerase Chain Reaction of Fresh Tissue

Total RNA was extracted from immature (10 dpp) mouse testes using the EZ-RNA Reagent protocol (Biological Industries, Beit Haemek, Israel). First-strand cDNAs were synthesized from 2.5 μg total RNA with 0.5 μg random oligonucleotide primers (Roche Molecular Biochemicals, Mannheim, Germany) and 200 units of Moloney-Murine Leukemia Virus—Reverse Transcriptase (M-MLV-RT; Life Technologies Inc, Paisley, Scotland, United Kingdom) in a total volume of 20 μL Tris-HCl-MgCl reaction buffer, supplemented with dithiothreitol, dinucleotriphosphates (0.5 mM; Roche Molecular Biochemicals), and RNase inhibitor (40 units; Roche Molecular Biochemicals). The reverse transcriptase (RT) reaction was performed for 1 hour at 37°C and stopped for 10 minutes at 75°C. The volume of 20 μL was subsequently filled up to 60 μL with water. Negative controls for the reverse transcriptase reaction (RT–) were prepared in parallel using the same reaction preparations with the same samples and without M-MLV-RT. The polymerase chain reaction (PCR), performed subsequently, contained cDNA samples in final dilution of 1:15 with 2 pairs of oligonucleotide primers (Sigma) which were exon spanning (Table 2).

To assess the absence of genomic DNA contamination in RNA preparations and RT-PCR reactions, PCR was performed with negative controls of the RT reaction (RT–) and without cDNA (cDNA–). The PCR reactions were carried out on a Cycler II System Thermal Cycler (Ericomp, San Diego, California). A total of 20 μL of each PCR product was run on 2% agarose gel containing ethidium bromide and was photographed under ultraviolet light (Figure 1C).

Messenger RNA Isolation from SACS-Cultured Cells

Messenger RNA from 10 dpp murine testicular cells and from testicular cells cultured with SACS were isolated using the μMACS oneStep cDNA Kit (Miltenyi), following the manufacturer's protocol. In brief, the cells were prepared fresh or were snap frozen before mRNA isolation. The samples were thawed in Lysis/Binding buffer (Miltenyi) on ice and were lysed by mixing and additional vortexing for 5 minutes. Afterwards, the sheared lysate samples were placed in a LysateClear column (Miltenyi) and centrifuged at 26 450 × g for 3 minutes to separate the mRNA. After separation, the lysate was mixed with 50 μL Oligo(dt) MicroBeads (Miltenyi). Magnetic separation was preceded using a prepared column within a magnetic field of the thermoMACS Separator (Miltenyi). Magnetically labeled mRNA was retained in the column during washing steps to remove rRNA and DNA according to the manufacturer's protocol. To proceed with cDNA synthesis, 100 μL equilibration/wash buffer was added 2 times to the column, followed by incubation with the enzyme mix (Miltenyi) for 1 hour. During cDNA synthesis the thermoMACS separator was set to 42°C for 1 hour. Subsequently, cDNA was washed 2 times with equilibration/wash buffer within the column. To release the cDNA from the magnetic beads, 20 μL cDNA-Release solution (Miltenyi) was applied for 10 minutes at 42°C on the top of the column. Synthesized cDNA was eluted with 50 μL cDNA Elution buffer (Miltenyi). The efficiency of cDNA synthesis was verified by PCR amplification of the β-actin gene. Specific expression of different spermatogenic stage-specific markers (Table 2) was investigated by PCR amplification. PCR conditions were 2 minutes at 94°C, 35 × 50 seconds at 94°C, 50 seconds at 58°C, and 1 minute at 72°C (annealing and extension). Messenger RNA isolation was performed using the same cell numbers (Figure 1C).


Statistic evaluation was performed by Student's t test (SigmaStat3; Statcon, Witzenhausen, Germany). Mean ± SD is given in the figures as described in the legends.


  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. References

Immunohistochemical and Flow Cytometric Evaluation of Spermatogonial Markers

Immunoreactivity of anti—Gfrα-1 and anti—Cd-9 for subpopulations of SSCs was confirmed by staining of representative histologic sections in day 10 pp and day 30 pp mice (Figure 2A through L). Positive spermatogonial cells were observed at the basal membrane of the seminiferous tubules. Differentiating germ cells and Sertoli cells were negative for Gfrα-1 and Cd-9. In the juvenile testis, Gfrα-1 and Cd-9 were expressed in single spermatogonial cells (Figure 2B, C, H, and I). In the adult testis, the predominant Gfrα-1-positive cells were isolated single spermatogonia (Figure 2K and L). In contrast, Cd-9 labeling was often detected in groups and chains of spermatogonia (Figure 2E and F). Omission of the primary antibodies or the use of rabbit IgGs (data not shown) as negative control showed identical results in the absence of any specific positive staining in the juvenile and adult tissue sections (Figure 2A, D, G, and J).


Figure 2. . Expression of Cd-9 (A-F) and Gfrα-1 (G-L) in the juvenile (A-C, G-I) and the adult (D-F, J-L) mouse testis. In the adult tissue, Cd-9 is expressed in aligned (E, F) and Gfrα-1 in single spermatogonia (K, L). The staining is confined to spermatogonia (arrowheads). Sertoli and Leydig cells remain unstained. A representative control of the method specificity shows no specific staining within the tubular compartment. Scale bars = 50 μm.

Download figure to PowerPoint

To confirm these results and to demonstrate both markers to be sufficient to isolate undifferentiated spermatogonia, flow cytometric analysis was performed on unsorted, depleted, and enriched fractions (Table 1) after MACS with anti—Gfrα-1 (Figure 3A through C), anti—Cd-9 (Figure 3D through F) and, as a third marker of undifferentiated cells, anti—Thy-1 (Figure 3G through I) antibodies. All markers showed separation of cell populations of the same size and granularity pattern in the depleted and the enriched fractions (Figure 3), whereas Gfrα-1 showed the highest percentage of events in the cell population of the enriched fraction. Therefore, the staining pattern and the flow cytometric analysis indicated that Gfrα-1 is the most suitable of the 3 markers analyzed for isolation of undifferentiated spermatogonia.


Figure 3. . Images of flow cytometric analysis of Gfrα-1, Cd-9, and Thy-1 unsorted and MAC-sorted cell fraction. Three different cell fractions (unsorted [A, D, G], depleted [B, E, H], and enriched [C, F, I]) are shown for MACS using anti—Gfrα-1 (A-C), anti—Cd-9 (D-F), and anti—Thy-1 (G-I) to separated undifferentiated spermatogonia. All three antibodies detect the same cell population defined by size (forward scatter [FS log: x-axis]) and granularity (side scatter [SS log: y-axis]). The dominant fractions in the depleted, and enriched fractions are surrounded with a black line.

Download figure to PowerPoint

Cell Separation

The 2-step enzymatic digestion resulted in a single-cell suspension (Table 1), which was used for cell separation by MACS. The eluted depleted fraction contained a heterogeneous suspension of living cells similar to the unsorted cell suspension (Figure 4A and B). After separation, the enriched fraction of fresh cells was homogenous, containing clusters of cells with similar sizes and shapes and comparable nuclear-cytoplasm ratio and nuclear morphology (Figure 4C). Microscopic imaging of immunohistochemical staining of isolated cells confirmed that the enriched fraction contained a higher number of Gfrα-1-positive cells compared with the unsorted and depleted fraction (Figure 4D through F). Total RNA isolation and analysis were performed to analyze the enriched fraction after MACS with Gfrα-1 (Figure 4G). RNA analysis revealed an expression profile typical for spermatogonial cells in the enriched fraction after MACS only. The expression of Gfrα-1 on the cell surface of undifferentiated spermatogonia is proven by confocal microscopy (Figure 4H through K). Cell survival was verified by trypan blue exclusion test at around 90%.


Figure 4. . Images of fluorescence-labeled cells showing freshly and fixed isolated cells from mouse testes (10-day-old mice). A heterogeneous cell suspension containing single cells is observed in the unsorted cell fraction after digestion of cells before the separation procedure (A) and in the depleted fraction after MAC sorting with Gfrα-1 (B). A homogeneous cell suspension is observed in the enriched fraction (C). These cells tend to form clusters of variable size. Images of fluorescent cells depicting fixed spermatogonia before and after MAC sorting. The mentioned fractions are shown after immunofluorescent labeling with anti—Gfrα-1 (unsorted fraction [D], depleted fraction [E], and enriched fraction [F]; arrowheads: Gfrα-1-positive cells [FITC]). DNA staining by Hoechst: A-F. Expression analysis of different spermatogenic marker genes (murine spermatogonial stages: Oct3/4, C-kit, Gfrα-1, Cd-9, and α-6-integrin; murine meiotic stages: Prohibitin and Srf-1; murine postmeiotic stages: Ldh, Protamine-2, and Sp-10; positive control: β-actin before (G; unsorted [us] lane) and after sorting with anti—GFRα-1 (G; enriched [+] lane). Confocal microscopy images of fluorescent-labeled cells showed expression of Gfrα-1 (arrows) on the cell surface of spermatogonial cells (H-J; overlay K). Scale bar = 20 μm.

Download figure to PowerPoint

Flow Cytometric Analysis of the Separation Technique

Flow cytometric analysis allowed a quantification of enrichment efficiency. Table 3 presents a comparison of the two methods used to establish the unsorted (Table 3A and D), depleted (Table 3B and E), and enriched fractions (Table 3C and F). Recovery rates were threefold higher when MACS was performed with biotin-labeled instead of magnetically labeled secondary antibodies. The unsorted fraction from the juvenile testes contained 21%–24% Gfrα-1-positive cells. The use of magnetically labeled secondary antibodies resulted in a twofold to threefold (up to 42% positive cells) enrichment rate for Gfrα-1-positive cells derived from 10 dpp murine cells. These enriched and depleted fractions from the day 10 pp testes were used for the SACS experiments.

Table 3. . Quantitative and flow cytometric evaluation of MACS by the use of standard MicroBeads compared with anti-biotin MicroBeadsa
  1. a Absolute cell numbers/mL of all 3 MACS fractions (unsorted, enriched, and depleted) are shown in white columns. The enrichment and depletion of Gfrα-1-positive cells in the obtained fractions after isolation are shown for both MicroBeads (standard MicroBeads: unsorted [A], depleted [B], and enriched [C]; anti-biotin MicroBeads: unsorted [D], depleted [E], and enriched [F]; y-axis: events; x-axis: FL1 log for positive FITC signals). Both MicroBeads show a depletion in FITC-positive signals in the depleted fraction (B, E) and an enrichment of FITC-positive signals in the enriched fraction (C, F) compared with the unsorted ones (A, D). The absolute cell numbers/mL of Gfrα-1-positive cells of the same fractions are shown in gray columns. In the anti-biotin MicroBeads-enriched fraction, more Gfrα-1-positive cells were found compared with cell numbers enriched by standard MicroBeads. The unsorted fraction showed a similar content of cells in both MACS variations. The use of anti-biotin MicroBeads resulted in fewer Gfrα-1-positive cells in the depleted fraction compared with the use of standard MicroBeads. Results are given as mean (%) ± SD of 3 and 7 experiments.

inline image

Evaluation of Germ Cell Development With and Without Supporter Cells Using the SACS

Three experimental approaches were selected to compare the outgrowth of colonies in the gel phase of the three-dimensional culture system: 1) cells from the enriched fraction growing in the gel phase (Figure 5A, D, and G) without additional supporting cells; 2) cells from the enriched fraction growing in the gel phase (Figure 5B, E, and H) and cells from the depleted fraction added to the solid phase; and 3) cells from the enriched fraction growing in the gel phase (Figure 5C, F, and I) together with cells of the interstitial and depleted fractions.


Figure 5. . Images of clonal expansion of SACS-cultured spermatogonia after different culture periods: without supporter cells (A, day 6; D, day 11; G, day 16) and with cell support: depleted fraction (tubular cells): (B, day 6; E, day 11; H, day 16); testicular cells (tubular and interstitial cells): (C, day 6; F, day 11; I, day 16) found in the gel phase. Colonies found in somatic cell—supported SACS showed fewer connected cells at the edge of the colonies (arrowheads) or even highly condensed structures in the inner region (arrows). Scale bar = 50 μm.

Download figure to PowerPoint

Morphologic analysis of colonies in the gel phase revealed a different growth pattern over time in response to the 3 different conditions (Figure 5). Colonies in the absence of supporter cells were heavily compacted and exhibited a round shape with sharp edges (Figure 5A, D, and G). The colonies growing in the gel phase in the presence of supporter cells in the solid phase were less dense and showed single cells and small groups of cells in loose contact with the colony (Figure 5B, E, and H). Those colonies that also contained interstitial cells (Table 1) in the gel phase show similar colony structures (Figure 5C, F, and I).

Evaluation of cell numbers on filters demonstrated a positive effect of supporter cells in the solid phase on the number of cells in the gel phase as consistently throughout all time points (days 1–16); a higher number of cells was determined in these groups (Figure 6A).


Figure 6. . Evaluation of testicular cells cultured in a three-dimensional SACS for 16 days and 24 hours with (dark bars, A-C) or without (light bars, A-C) intratubular supporter cells (depleted fraction) in the solid phase or supplementation with all somatic testicular cells (D). Analysis of vacuum-filtrated cells showed different cell numbers in the approach with or without supporter cells after 16 days of culture (A) and 24 hours (B). Proliferation rate analysis shows a significant increase of proliferating cell numbers during the 12th and 20th hours in the approach with support of intratubular cells in the solid phase (C). Evaluation of TUNEL-positive cells in the approach using all testicular cells in the gel phase depicts a decrease of apoptosis at day 6 and day 11 of culture (D). Results of mRNA isolation and expression after 1 day ofculture of cells in the gel phase (E-G). In both approaches with isolated spermatogonia (without intratubular supporter cells in the solid phase [E]; with intratubular supporter cells in the solid phase [F]), strong expression of Cd-9 and α-smooth muscle is shown, whereas the approach containing all testicular cells shows a variety of markers different spermatogenic stages (G). Primer used for analysis: A/O standard marker (500-bp band is shown on both sites), B: Oct3/4; C: Gfrα-1; D: C-kit; E: α-6-integrin; F: Dazl; G: Prohibitin; H: Srf-1; I: Ldh; J: Protamine-2; K: Scp-3; L: Sp-10; M: Abp; N: α-smooth muscle; P: β-actin; Q: Cd-9. Pooled results are shown as mean ± SD of 12 (A) and 3 (B-D) experiments. *P < .05; **P < .01; ***P < .001.

Download figure to PowerPoint

A high loss of total cells but a positive effect of supporter cells in the solid phase were also determined during the first 24 hours of SACS (Figure 6B). To determine the cause for the 40%–70% loss of cells from the gel phase during the initial 24 hours of culture, we evaluated the number of apoptotic cells after 24 hours of culture. In all groups, we determined rates of 37.2%–47.7% (+10% SD) TUNEL-positive cells independent of the presence of supporter cells. The proliferation analysis showed a significant increase of BrdU-positive cells between the 12th and 20th hour of culture in the experiment with isolated spermatogonia in the gel phase supported by cells from the depleted fraction in the solid phase (Figure 6C), whereas no significant increase could be observed in the approach without supporting cells in the solid phase. Analysis of apoptotic events during the culture period of the third approach, when all testicular cells were used, showed a decrease of apoptosis (to around 20%) between day 6 and day 11 of culture (Figure 6D).

To exclude a potential migration of cells from the solid phase into the gel phase, we separated the 2 phases by a cell-impermeable membrane. The cell numbers were evaluated by counting cells in 40 paraffin sections per culture approach after 24 hours of culture. In this approach we obtained no difference in cell number proportion (without supporter cells: 101.2 ± 39.1 [SD]; with supporter cells: 357.6 ± 79.0 [SD]) compared with the approach without using a membrane, and therefore no indices for cell migration between the 2 phases.

To investigate spermatogenic development during culture in all 3 experimental settings, mRNA was isolated with the μMACS kit. The small amount of cell numbers resulted in very low levels of mRNA. Therefore, we could only analyze expression profiles on 1 day of culture (Figure 6E and F). In the 2 approaches using Gfrα-1-isolated spermatogonia (Figure 6E and F), a strong expression of Cd-9 (undifferentiated spermatogonia; Figure 6E and F, lane Q), α-smooth muscle (peritubular cells; Figure 6E and F, lane N), and β-actin (positive control; Figure 6E and F, lane P), and weak expression of prohibitin and Srf-1 (meiotic spermatocytes; Figure 6E and F, lanes G and H) could be observed after 1 day of culture. In the approach containing all testicular cells, mRNA of different spermatogenic stages was observed (Figure 3G). In this approach, the mRNA expression was detectable for spermatogonial and meiotic genes.

Characterization of the cultured cells in the gel phase by immunohistochemistry revealed the presence of Gfrα-1-positive cells in histologic sections and on filters after vacuum filtration at different time points (Figure 7A through D). To determine the degree of differentiation of germ cells in the gel phase when exposed to different culture conditions and to confirm these data obtained by mRNA analysis, we performed immunohistochemical localization of Boule (Figure 7E through J) and Crem (Figure 7K and L). Boule is considered a reliable marker for meiotic germ cells (Xu et al, 2001), and its expression in murine testis was observed earliest in the stage of late pachytene spermatocytes. Immunohistochemistry of tissue sections confirms that Boule is present after day 15 pp in the late spermatocyte stage (Figure 7N), but is not expressed at an earlier time point of development (Figure 7M).


Figure 7. . Expression Gfrα-1 (A, B, D) in SACS-cultured cells. The staining is confined to spermatogonia (DAB-positive, A, B [brown precipitate; arrowheads] and FITC-positive, D) directly in the gel phase in associated cell groups (B, D) and single cells (A) after 24 hours (A, B) and 16 days (D). An image of a control labeling omitting a primary antibody is shown in C. Expression of the Boule protein (E-I) and Crem (K) in SACS-cultured cells and in the murine testis at different developmental stages (M-R). Culture approaches without (E, 11 days of culture; and G, 16 days of culture) and with tubular cell (depleted fraction) support (F, 11 days of culture; and H, 16 days of culture) showed no positive expression of Boule after vacuum filtration, whereas stained paraffin sections of the approach with all testicular cells (tubular and interstitial cells) as supporter resulted in associated cell groups confined to meiotic cells (DAB-positive, arrowheads [I]) directly in the gel phase after 13 days of culture. Crem-positive cells are shown in this approach after at least 21 days of culture. Images of a control labeling omitting a primary antibody are shown for Boule in J and for Crem in L. In the 10 dpp old tissue (used in our experiments as starting material for SACS), Boule and Crem are not expressed in any tubular cell type (for Boule: M; for Crem: P). We observed positive Boule expression in the testis starting with day 16 pp (day 16: N). The staining is confined to meiotic cells in the stage of pachytene spermatocytes (arrows, N). An image of a control labeling omitting a primary antibody is shown (O). We observed positive Crem expression in the testis starting with day 21 pp (day 21: Q). The staining is confined to postmeiotic cells in the stage of round spermatids (open arrowheads [Q]). An image of a control labeling omitting a primary antibody is shown (R). Scale bars: 10 μm (A-L) and 20 μm (M-R).

Download figure to PowerPoint

No Boule-positive cells were observed in cell fractions after SACS without supporter cells or with cells from the depleted fraction in the solid phase (Figure 7E through H). However, when interstitial cells and cells from the depleted fraction were added to the gel phase, Boule-positive cells were detected consistently when the cultures were maintained for at least 13 days (Figure 7I).

Crem is considered a marker for postmeiotic cells at the stage of round spermatids (Delmas et al, 1993; Wistuba et al, 2002). In the third approach containing all testicular cells, positive signals for this postmeiotic spermatogenic stage of round spermatids were observed for at least 21 days of culture (Figure 7K). Immunohistochemistry of tissue sections confirms that Crem is present 3 weeks after birth in the spermatogenic stage of round spermatids (Figure 7Q) but not in cells at the time point when cell isolation was performed (Figure 7P).


  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. References

Many studies have analyzed in vitro culture effects on gonocytes or SSCs (Hasthorpe et al, 2000; Hasthorpe 2003; Izadyar et al, 2002, 2003; Kanatsu-Shinohara et al, 2003, 2004a, 2005a; Nagano et al, 2003; Kubota et al, 2004b). In general, conventional culture approaches were employed. These did not provide structural conditions that would closely resemble the natural testicular environment. Therefore, our study aimed to establish and validate a novel method for spermatogonial cell culture to improve propagation and differentiation of these undifferentiated germline cells. In contrast to conventional culture performed in dishes or flasks, the three-dimensional agar structure of SACS offers conditions mimicking some structural features of the in vivo situation. SACS was used to examine supporting and limiting effects of somatic testicular cells cocultured in the solid phase of the system. As previously published, the use of different supporter cell types revealed aspects of SSC culture (single spermatogonia; Asingle; de Rooij et al, 2000) in terms of SSC line establishment (Shinohara et al, 2000; Nagano et al, 2003; Kubota et al, 2004b; Kanatsu-Shinohara et al, 2005a) and differentiation into more developed spermatogenic stages (Tres et al, 1983; Gerton et al, 1984; Hue et al, 1998; Tesarik et al, 1998a,b; Feng et al, 2002; Sousa et al, 2002). However, a culture system that allows complete spermatogenesis to occur is still far from routine methodology.

Successful enrichment and separation of isolated testicular cells from mouse tissue using MACS has previously been described (von Schönfeldt et al, 1999; Kubota et al, 2004b; Buageaw et al, 2005; Oatley et al, 2007). Supplementation with somatic cells resulted in a stabilized and more differentiated in vitro population of germ cells. The experimental setting combining MACS separation of the testicular cell fraction and SACS allowed the use of the various fractions achieved by the MAC sorting. A successful MACS separation depends on the use of cell surface markers expressed exclusively on undifferentiated SSCs (Sofikitis et al, 2005). Therefore, Gfrα-1, Cd-9, and Thy-1 were analyzed as putative mouse SSC markers (Meng et al, 2000; Kanatsu-Shinohara et al, 2004b; von Schönfeldt et al, 2004; Oatley et al, 2007; He et al, 2007). The fact that all 3 markers detect the same cell population (same size and granularity in flow cytometric analysis of the enriched fractions) and that we localized exclusively single spermatogonia and small chains/groups of spermatogonia indicated that Cd-9 is a marker for Asingle and Aaligned spermatogonia. In contrast, Gfrα-1 appeared to be expressed exclusively in single spermatogonia, rendering out our favorite marker for SSC separation. Previous studies using MACS confirmed that Gfrα-1 is an excellent marker for SSCs as a co-enrichment of Oct-3/4, which is considered a specific marker for pluripotent cells and germline stem cells (primordial germ cell, embryonic germ cells, and embryonic stem cells), was observed in the enriched fraction (Ohbo et al, 2003; Buageaw et al, 2005). Furthermore, our results are strongly supported by a recently published study by He et al (2007), who showed the double expression of Oct-3/4 and Gfrα-1 in type A spermatogonia in 6 dpp murine testes. Taken together, these and our results indicate an expression of Gfrα-1 in SSCs before the initial differentiation and expansion into pairs and chains starts, which is also indicated by expression of Cd-9.

To explore the effect of more sensitive separation approaches, we compared indirect approaches with magnetically labeled secondary antibodies to strategies using biotin-labeled secondary antibodies and antibiotin magnetic MicroBeads. The better result in cell numbers but similar outcome in the degree of enrichment using the latter approach let us conclude that the efficiency of isolation depends on the enhancement of a rather weak cellular labeling. This indicates that even low expression of Gfrα-1 on the cell surface could be detected. This finding also confirms previous observations that sub-populations of SSCs exist which are characterized by different levels of Gfrα-1 expression (Buagaew et al, 2005).

On day 10 pp in the juvenile immature mouse testis, the SSC proportion is up to 100-fold higher compared with adult tissue (de Rooij et al, 2000; McLean et al, 2003; Aponte et al, 2005). We determined a proportion of 21%–24% Gfrα-1-positive cells in immature preparations prior to sorting. In addition, isolated spermatogonia from immature mice showed better viability (Creemers et al, 2002) and differentiation potential (Nagano et al, 2003). Therefore, the use of juvenile male germ cells seems to be beneficial for spermatogonial in vitro development.

The SACS we used consists of 2 phases of different agar concentrations forming a gel and a solid phase according to Huleihel et al, 1993. This arrangement allows addition of different supplemental factors or supporter cell lines (eg, Sertoli cells) to the solid agar phase without contaminating the gel phase containing the enriched SSCs.

Colony morphology was different when the cultured spermatogonia were grown in different SACS approaches; once established, it did not change during continued culture. During the first 24 hours of SACS spermatogonial cell number decreased independently of the presence of tubular cells and was shown to occur via apoptosis due to abundant TUNEL-positive cells. However, cell survival was enhanced when germ cells are cocultured with cells of the tubular and interstitial fraction, and this early effect was sustained throughout the culture period up to 16 days. Better survival of spermatogonia in the presence of somatic cells confirms findings from conventional in vitro experiments (Dirami et al, 1999; Izadyar et al, 2003).

During murine male germ cell development, a first wave of apoptosis occurs at day 16 pp (Zheng et al, 2006). We also observed an apoptotic wave in 16-dayold germ cells (isolated at day 10 pp and maintained for 6 days in vitro). This response could reflect the first wave of apoptosis found in vivo. However, when a somatic cell supported the germ cell, this apoptotic wave was not seen. It can be speculated that factors produced by Sertoli cells and/or Leydig cells have a positive effect on the cultured spermatogonia.

During the development of the immature testis, Sertoli cells differentiate terminally (eg, Tarulli et al, 2006). Sertoli cells produce 2 isoforms of SCF, a paracrine growth factor, which has inhibiting effects on apoptosis in early spermatogenesis (Print et al, 2000; Huleihel et al, 2004). The soluble form is predominantly expressed and important in the juvenile testis; the membrane-bound isoform is crucial for adult spermato-genesis (Blanchard et al, 1998; de Rooij et al, 1998). Considering the antiapoptotic effect of SCF, our data obtained from SACS suggest an effect on apoptotic inhibition during spermatogonial differentiation.

If optimal culture conditions exist, meiosis should be initiated and completed in vitro. Boule is a meiosis marker highly expressed in mice in late pachytene or diplotene stage spermatocytes (Xu et al, 2001). We confirmed here that Boule protein in immature mouse testis is not detected before day 16 pp. In SACS-cultured germ cells, we found Boule expression at day 13 of culture when the spermatogonia were cocultured with all other somatic testicular cells in the gel phase of the agar, but not when spermatogonia were cultured alone or with the tubular somatic fraction. As an additional marker to determine meiotic processes in vitro, we analyzed Crem expression indicating for postmeiotic/round spermatid stages. Crem is known to be expressed in round spermatids (Delmas et al, 1993; Wistuba et al, 2002), and is therefore the optimal marker to analyze meiosis completion. In the SACS approach using all testicular cells (intratubular and interstitial; Table 1), we show that Crem-positive cells appear at least at day 21 of culture. This might be an effect of testosterone as a product of Leydig cells located in the interstitium, which is considered to be a crucial factor inhibiting apoptotic events during meiosis (Print et al. 2000).

The observed mRNA expression profile supports the results obtained by immunohistochemistry. In the approach containing all testicular cells, almost all meiotic genes were expressed at the mRNA level already after 1 day of culture. This can be explained by the well-known shift between transcription and translation of genes during spermatogenesis (Kleene et al, 1984; Kleene, 1996; Iguchi et al, 2006). Although only the germ cells that were supported by all other testicular cells progress up to meiosis, the early expression of these mRNAs might be a necessary step preparing the later differentiation. Hence, in the other experiments, we did not find this expression pattern, and maybe for this reason we also failed to detect meiosis. Therefore, these results indicate that our coculture approach allowed germ cells to enter meiosis in vitro without any addition of growth factors.

The hypothesis that in vitro meiosis is even possible without a direct cell-cell contact has to be investigated further. Therefore, additional experiments combining the supporting factors (eg, LIF, GDNF, SCF, and/or hormones like testosterone) with a three-dimensional environment might result in completed spermatogenesis in vitro.


  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. References

The authors thank Prof Dr R. Reijo at the University of California, San Francisco, CA, for providing us with BOULE antibodies. J. Salzig and H. Kersebom are thanked for technical assistance, and M. Heuermann, G. Stelke, and O. Damm for animal caretaking. Finally, we are grateful to Prof M. Simoni for comments on the manuscript and Susan Nieschlag, MA, for language editing.


  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. References
  • Allard EK, Blanchard KT, Boekelheide K.. Exogenous stem cell factor (SCF) compensates for altered endogenous SCF expression in 2,5-hexanedione-induced testicular atrophy in rats. Biol Reprod. 1996;55: 185193.
  • Aponte PM, van Bragt MPA, de Rooij DG, van Pelt AMM. Spermatogonial stem cells: characteristics and experimental possibilities. APMIS. 2005;113: 727742.
  • Blanchard KT, Lee J., Boekelheide K.. Leuprolide, a gonadotropin-releasing hormone agonist, reestablishes spermatogenesis after 2,5-hexanedione-induced irreversible testicular injury in the rat, resulting in normalized stem cell factor expression. Endocrinology. 1998;139: 236244.
  • Buageaw A., Sukhwani M., Ben-Yehudah A., Ehmcke J., Rawe VY, Pholpramool C., Orwig KE, Schlatt S.. GFRalpha-1 phenoptype of spermatogonial stem cells in immature mouse testes. Biol Reprod. 2005;73: 10111016.
  • Creemers LB, den Ouden K., van Pelt AMM, de Rooij DG. Maintenance of adult mouse type A spermatogonia in vitro: influence of serum and growth factors and comparison with prepubertal spermatogonial cell culture. Reproduction. 2002;124: 791799.
  • de Miguel MP, de Boer-Brouwer M., Paniagua R., van den Hurk R., de Rooij DG, van Dissel-Emiliani FMF. Leukemia inhibitory factor and ciliary neurotrophic factor promote the survival of Sertoli cells and gonocytes in a coculture system. Endocrinology. 1996;137: 18851893.
  • de Rooij DG, Grootegoed JA. Spermatogonial stem cells. Curr Opin Cell Biol. 1998;10: 694701.
  • de Rooij DG, Russell LD. All you wanted to know about spermatogonia but were afraid to ask. J Androl. 2000;21: 776798.
  • Delmas V., van Der Hoorn F., Mellström B., Jéqou B., Sassone-Corsi P.. Induction of CREM activator proteins in spermatids: down-stream targets and implications for haploid germ cell differentiation. Mol Endocrinol. 1993;7: 15021514.
  • Dirami G., Ravindranath N., Pursel V., Dym M.. Effects of stem cell factor and granulocyte macrophage-colony stimulating factor on survival of porcine type A spermatogonia cultured in KSOM. Biol Reprod. 1999;61: 225230.
  • Feng LX, Chen Y., Dettin L., Pera Reijo RA, Herr JC, Goldberg E., Dym M.. Generation and in vitro differentiation of a spermatogonial cell line. Science. 2002;297: 392395.
  • Fujita K., Tsujimura A., Takao T., Miyagawa Y., Matsumiya K., Koga M., Takeyama M., Fujioka H., Aozasa K., Okuyama A.. Expression of inhibin α, glia cell line-derived neurotrophic factor in Sertoli cell only syndrome: relation to successful sperm retrieval by microdis-section testicular sperm extraction. Hum Reprod. 2005;20: 22892294.
  • Gerton GL, Millette CF. Generation of flagella by cultured mouse spermatids. J Cell Biol. 1984;98: 619628.
  • Guan K., Nayernia K., Maier LS, Wagner S., Dressel R., Lee JH, Nolte J., Wolf F., Li M., Engel W., Hasenfuss G.. Pluripotency of spermatogonial stem cell from adult mouse testis. Nature. 2006;440: 11991203.
  • Hasthorpe S.. Clonogenic culture of normal spermatogonia: in vitro regulation of postnatal germ cell proliferation. Biol Reprod. 2003;68: 13541360.
  • Hasthorpe S., Barbic S., Farmer PJ, Hutson JM. Growth factor and somatic cell regulation of mouse gonocyte-derived colony formation in vitro. J Reprod Fertil. 2000;119: 8591.
  • He Z., Jiang J., Hofmann MC, Dym M.. Gfrα1 silencing in mouse spermatogonial stem cells results in their differentiation via the inactivation of RET tyrosine kinase. Biol Reprod. 2007;77: 723733.
  • Hofmann MC, Narisawa S., Hess RA, Millan JL. Immortalization of germ cells and somatic testicular cells using the SV40 large T antigen. Exp Cell Res. 1992;201: 417435.
  • Horowitz D., King AG. Colorimetric determination of inhibition of hematopoietic progenitor cells in soft agar. J Immunol Methods. 2000;244: 4958.
  • Hue D., Staub C., Perrard-Sapori MH, Weiss M., Nicolle JC, Vigier M., Durand P.. Meiotic differentiation of germinal cells in three-week cultures of whole cell population from rat seminiferous tubules. Biol Reprod. 1998;59: 379387.
  • Huleihel M., Douvdevani A., Segal S., Apte RN. Different regulatory levels involved in the generation of hemopoietic cytokines (CSFs and IL-6) in fibroblasts by inflammatory products. Cytokine. 1993;5: 4756.
  • Huleihel M., Lunenfeld E.. Regulation of spermatogenesis by paracrine/autocrine testicular factors. Asian J Androl. 2004;6: 259268.
  • Iguchi N., Tobias JW, Hecht NB. Expression profiling reveals meiotic male germ cells mRNAs that are translationally up- and down-regulated. Proc Natl Acad Sci U S A. 2006;103: 77127717.
  • Izadyar F., den Ouden K., Creemers LB, Posthuma G., Parvinen M., de Rooij DG. Proliferation and differentiation of bovine type A spermatogonia during long-term culture. Biol Reprod. 2003;68: 272281.
  • Izadyar F., Spierenberg GT, Creemers LB, den Ouden K., de Rooij DG. Isolation and purification of type A spermatogonia from the bovine testis. Reproduction. 2002;124: 8594.
  • Kanatsu-Shinohara M., Inoue K., Lee J., Yoshimoto M., Ogonoki N., Miki H., Baba S., Kato T., Kazuki Y., Toyokuni S., Toyoshima M., Ohtsura N., Oshimura M., Heike T., Nakahata T., Ishino F., Ogura A., Shinohara T.. Generation of pluripotent stem cells from neonatal mouse testis. Cell. 2004a;119: 10011012.
  • Kanatsu-Shinohara M., Miki H., Inoue K., Ogonuki N., Toyokuni S., Ogura A., Shinohara T.. Long-term culture of mouse male germline stem cells under serum-or feeder-free conditions. Biol Reprod. 2005a;72: 985991.
  • Kanatsu-Shinohara M., Ogonuki N., Inoue K., Miki H., Ogura A., Toyokuni S., Shinohara T.. Long-term proliferation in culture and germline transmission of mouse male germline stem cells. Biol Reprod. 2003;69: 612616.
  • Kanatsu-Shinohara M., Ogonuki N., Iwano T., Lee J., Kazuki Y., Inoue K., Miki H., Takehashi M., Toyokuni S., Shinkai Y., Oshimura M., Ishino F., Ogura A., Shinohara T.. Genetic and epigenetic properties of mouse male germline stem cells during long-term culture. Development. 2005b;132: 41554163.
  • Kanatsu-Shinohara M., Toyokuni S., Shinohara T.. CD-9 is a surface marker on mouse and male germline stem cells. Biol Reprod. 2004b;70: 7075.
  • Kimball PM, Brattain MG, Pitts AM. A soft-agar procedure measuring growth of human colonic carcinomas. Br J Cancer. 1978;37: 10151019.
  • Kleene KC. Patterns of translational regulation in the mammalian testis. Mol Reprod Dev. 1996;43: 268281.
  • Kleene KC, Distel RJ, Hecht NB. Translational regulation and deadenylation of a protamine mRNA during spermiogenesis in the mouse. Dev Biol. 1984;105: 7179.
  • Koh KB, Komiyama M., Toyama Y., Adachi T., Mori C.. Percoll fractionation of adult mouse spermatogonia improves germ cell transplantation. Asian J Androl. 2004;6: 9398.
  • Kubota H., Avarbock MR, Brinster RL. Culture conditions and single growth factors affect fate determination of mouse spermatogonial stem cells. Biol Reprod. 2004a;71: 722731.
  • Kubota H., Avarbock MR, Brinster RL. Growth factor essential for self-renewal and expansion of mouse spermatogonial stem cells. Proc Natl Acad Sci U S A. 2004b;101: 1648916494.
  • Lee J., Richburg JH, Younkin SC, Boekelheide K.. The Fas system is a key regulator of germ cell apoptosis in the testis. Endocrinology. 1997;138: 20812088.
  • Lee JH, Gye MC, Choi KW, Hong JY, Lee YB, Park DW, Lee SJ, Min CK. In vitro differentiation of germ cells from nonobstructive azoospermic patients using three-dimensional culture in a collagen gel matrix. Fertil Steril. 2007;87: 824833.
  • Lee JH, Kim HJ, Kim H., Lee SJ, Gye MC. In vitrospermatogenesis by three-dimensional culture of rat testicular cells in collagen gel matrix. Biomaterials. 2006;27: 28452853.
  • Lin HS, Kuhn C., Kuo T.. Clonal growth of hamster free alveolar cells in soft agar. J Exp Med. 1975;142: 877886.
  • McLean DJ, Friel PJ, Johnston DS, Griswold MD. Characterization of spermatogonial stem cell maturation and differentiation in neonatal mice. Biol Reprod. 2003;69: 20852091.
  • Meng X., Lindahl M., Hyvönen ME, Parvinen M., de Rooij DG, Hess MW, Raatikainen-Ahokas A., Sainio K., Rauvala H., Lakso M., Pichel JG, Westphal H., Saarma M., Sariola H.. Regulation of cell fate decision of undifferentiated spermatogonia by GDNF. Science. 2000;287: 14891493.
  • Nagano M., Ryu BY, Brinster CJ, Avarbock MR, Brinster RL. Maintenance of mouse male germ line stem cells in vitro. Biol Reprod. 2003;68: 22072214.
  • Oatley JM, Avarbock MR, Brinster RL. Glia cell line-derived neurotrophic factor regulation of genes essential for self-renewal of mouse spermatogonial stem cells is dependent on SRC family kinase signaling. J Biol Chem. 2007;282: 2584225851.
  • Ohbo K., Yoshida S., Ohmura M., Ohneda O., Ogawa T., Tsuchiya H., Kuwana T., Kehler J., Abe K., Schöler HR, Suda T.. Identification and characterization of stem cells in prepubertal spermatogenesis in mice small star. Dev Biol. 2003;258: 209225.
  • Print CG, Loveland KL. Germ cell suicide: new insights into apoptosis during spermatogenesis. Bio Essays. 2000;22: 423430.
  • Quaroni A., Wands J., Trelstad RL, Isselbacher KJ. Epithelioid cell cultures from rat small intestine. Characterization by morphologic and immunologic criteria. J Cell Biol. 1979;80: 248265.
  • Schlatt S., Zhengwei Y., Meehan T., de Kretser DM, Loveland KL. Application of morphometric techniques to postnatal rat testes in organ culture: insights into testis growth. Cell Tissue Res. 1999;298: 335343.
  • Shinohara T., Orwig KE, Avarbock MR, Brinster RL. Spermatogonial stem cell enrichment by multiparameter selection of mouse testis cells. Proc Natl Acad Sci U S A. 2000;97: 83468351.
  • Sofikitis N., Pappas E., Kawatani A., Baltogiannis D., Loutradis D., Kanakas N., Giannakis D., Dimitriadis F., Tsoukanelis K., Georgiou I., Makrydimas G., Mio Y., Tarlatzis V., Melekos M., Miyagawa I.. Efforts to create an artificial testis: culture systems of male germ cells under biomedical conditions resembling the seminiferous tubular biochemical environment. Hum Reprod Update. 2005;11: 229259.
  • Sousa M., Cremades N., Alves C., Silva J., Barros A.. Developmental potential of human spermatogenic cells co-cultured with Sertoli cells. Hum Reprod. 2002;17: 161172.
  • Spradling A., Drummond-Barbosa D., Toshie K.. Stem cells find their niche. Nature. 2001;414: 98104.
  • Tarulli GA, Stanton PG, Lerchl A., Meachem SJ. Adult Sertoli cells are not differentiated in the Djungarian hamster: effect of FSH on proliferation and junction protein organization. Biol Reprod. 2006;74: 798806.
  • Tesarik J., Greco E., Rienzi L., Ubaldi F., Guido M., Cohen-Bacrie P., Mendoza C.. Differentiation of spermatogenic cells during in-vitro culture of testicular biopsy samples from patients with obstructive azoospermia: effect of recombinant follicle stimulating hormone. Hum Reprod. 1998a;13: 27722781.
  • Tesarik J., Guido M., Mendoza C., Greco E.. Human spermatogenesisin vitro: respective effects of follicle-stimulating hormone and testosterone on meiosis, spermiogenesis, and Sertoli cell apoptosis. J Clin Endocrinol Metab. 1998b;83: 44674473.
  • Tres LL, Kierszenbaum AL. Viability of rat spermatognenic cellin vitro is facilitated by their coculture with Sertoli cells in serum-free hormone-supplemented medium. Proc Natl Acad Sci U S A. 1983;80: 33773381.
  • vo Schönfeldt V., Krishnamurthy H., Foppiani L., Schlatt S.. Magnetic cell sorting is a fast and effective method of enriching viable spermatogonia from Djungarian hamster, mouse, and marmoset monkey testes. Biol Reprod. 1999;61: 582589.
  • vo Schönfeldt V., Wistuba J., Schlatt S.. Notch-1, c-kit and GFR alpha-1 are developmentally regulated markers for premeiotic germ cells. Cytogenet Genome Res. 2004;105: 235239.
  • Wistuba J., Schlatt S., Cantauw C., vo Schönfeldt V., Nieschlag E., Behr R.. Transplantation of wild-type spermatogonia leads to complete spermatogenesis in testes of cyclic 3′,5′-adenosine monophoshate response element modulator-deficient mice. Biol Reprod. 2002;67: 10521057.
  • Xu EY, Moore FL, Pera Reijo RA. A gene family required for human germ cell development evolved from an ancient meiotic gene conserved in metazoans. Proc Natl Acad Sci U S A. 2001;98: 74147419.
  • Zheng S., Turner TT, Lysiak JJ. Caspase 2 activity contributes to the initial wave of germ cell apoptosis during the first round of spermatogenesis. Biol Reprod. 2006;74: 10261033.
  • Supported by the German-Israeli Foundation (grant 1: 760–205.2/2002) and a grant from the medical faculty of the University of Münster (IZKF Project No. WI 2/023/07). J.W. was supported by the Deutsche Forschungsgemeinschaft WI 2723/1–1, and S.S. by a U54 grant from the National Institutes of Health.

  • These authors contributed equally to this article and share coauthorship.