Culture of Rodent Spermatogonial Stem Cells, Male Germline Stem Cells of the Postnatal Animal

Abstract

Spermatogonial stem cells (SSCs), postnatal male germline stem cells, are the foundation of spermatogenesis, during which an enormous number of spermatozoa is produced daily by the testis throughout life of the male. SSCs are unique among stem cells in the adult body because they are the only cells that undergo self-renewal and transmit genes to subsequent generations. In addition, SSCs provide an excellent and powerful model to study stem cell biology because of the availability of a functional assay that unequivocally identifies the stem cell. Development of an in vitro culture system that allows an unlimited supply of SSCs is a crucial technique to manipulate genes of the SSC to generate valuable transgenic animals, to study the self-renewal mechanism, and to develop new therapeutic strategies for infertility. In this chapter, we describe a detailed protocol for the culture of mouse and rat SSCs. A key factor for successful development of the SSC culture system was identification of in vitro growth factor requirements for the stem cell using a defined serum-free medium. Because transplantation assays using immunodeficient mice demonstrated that extrinsic factors for self-renewal of SSCs appear to be conserved among many mammalian species, culture techniques for SSCs of other species, including farm animals and humans, are likely to be developed in the coming 5–10 years.

I. Introduction

Germ cells are specialized cells that pass the genetic information of an individual to the next generation. Production of functional germ cells is essential for continuation of the germline of the species. Spermatogenesis, the process of male germ cell production, takes place in the seminiferous tubules of the postnatal testis and is a highly productive system in the body. In the mammalian testis, more than 20 million sperms per gram of tissue are produced daily (Amann, 1986). The high productivity relies on spermatogonial stem cells (SSCs). Like other types of stem cells in adult tissues, SSCs self-renew and produce daughter cells that commit to diVerentiate throughout life of the male (Meistrich and van Beek, 1993). Furthermore, in mammals, SSCs are unique among stem cells in the adult body, because they are the only cells that undergo self-renewal and transmit genes to subsequent generations.

Stem cells are defined by their biological function; therefore, unequivocal identification of a stem cell requires a functional assay (Weissman et al., 2001). A functional transplantation assay for SSCs was developed in mice a decade ago and made it possible to study the biological function of SSCs, including self-renewal and diVerentiation (Brinster and Avarbock, 1994; Brinster and Zimmermann, 1994). In the transplantation assay, donor cells are harvested from the testes of fertile mice and are microinjected into seminiferous tubules of recipient mice. A small subpopulation of transplanted testis cells, the SSCs, colonize the basement membrane of the seminiferous tubules and begin to proliferate. Individual donor SSCs eventually form spermatogenic colonies in the seminiferous tubules of the recipients. When testicular cells from transgenic mice that express the Escherichia coli LacZ gene, which encodes a β-galactosidase, or the green fluorescence protein (GFP) gene are transplanted, donor colonies can be identified and counted readily because of the transgene expression in the donor cells (Fig. 1). About 2 months after transplantation of mouse SSCs, diVerentiated germ cells in a colony have progressed to spermatozoa in the seminiferous tubules of the recipient animal. The reconstituted spermatogenic colonies continuously produce spermatozoa throughout the remaining life of the recipient males. Several lines of evidence indicate that each donor-derived spermatogenic colony arises from a single SSC (Dobrinski et al., 1999b; Kanatsu-Shinohara et al., 2006b; Zhang et al., 2003), and the colonization eYciency of transplanted SSCs is estimated to be 5–12% (Nagano, 2003; Ogawa et al., 2003). Following transplantation, the recipient males can become fertile and produce progeny with donor cell haplotype, demonstrating normal function of the spermatozoa originating from transplanted germ cells (Brinster and Avarbock, 1994). Thus, it is clear that the spermatogenic colony-forming cells are SSCs, and the spermatogonial transplantation technique is a functional assay that allows quantitative evaluation of SSCs from a variety of sources.

Fig. 1.

Colony formation of donor-derived spermatogenesis in infertile recipient mouse testes. Testicular germ cells were isolated from transgenic mice that express reporter genes (LacZ or GFP) and injected into testes of mice treated with busulfan. Two months after transplantation, donor-derived spermatogenesis is reconstituted. Left: testis transplanted with LacZ expressing SSCs from transgenic mouse line B6.129S-Gt(ROSA)26Sor/J (ROSA, Jackson Laboratory). The testis was stained with 5-bromo-4-choloro-3-indolyl β-D- galactoside (X-gal). Right: testis transplanted with GFP-expressing SSCs from transgenic mouse line C57BL/6-TgNACTB-EGFP)1Osb/J (Jackson Laboratory). Each blue stretch or green stretch of cells in the testes represents a colony of spermatogenesis that arises from a single SSC.

Existence of a definitive functional assay to unequivocally identify SSCs provides an ideal experimental system to study stem cell biology. Using the functional assay, SSCs and the surrounding microenvironment, or the stem cell niche, in the seminiferous tubules have been studied (Brinster, 2002). Furthermore, by means of genetic modification of SSCs isolated from testes followed by transplantation, it has been shown that an SSC is a valuable vehicle to generate genetically modified animals (Hamra et al., 2002; Nagano et al., 2001a).

Because of the enormous potential of SSCs in basic research and applied science, including agriculture and medicine, development of an in vitro culture system for stem cells is extremely important. An early study demonstrated that SSCs could survive on STO (SIM mouse embryo-derived thioguanine and ouabain resistant) mouse embryonic fibroblast feeder layers for several months in culture (Nagano et al., 1998). Recently, several methods to culture rodent SSCs for long periods have been reported (Hamra et al., 2005; Kanatsu-Shinohara et al., 2003; Kubota et al., 2004b; Ryu et al., 2005). To develop a long-term culture system for SSCs, one of the most crucial objectives is identification of extrinsic factors essential to promote self-renewal and expansion of SSCs in vitro. Previous studies using transgenic mice with gain-of-function and loss-of-function of glial cell line-derived neurotrophic factor (GDNF) indicated that this ligand is a key growth factor to control survival and proliferation of undiVerentiated spermatogonia and perhaps SSCs in vivo (Meng et al., 2000). Using a serum-free culture system, we clearly demonstrated that GDNF is indeed the primary growth factor for SSC self-renewal (Kubota et al., 2004b). In the presence of GDNF, SSCs formed tightly packed clumps of cells and continuously proliferated. Clump-forming germ cells kept on expanding for more than 6 months in the defined serum-free medium supplemented with GDNF, and reconstituted long-term spermatogenesis following transplantation into recipient testes (Fig. 2). Furthermore, GDNF was found to be the crucial extrinsic factor for rat SSC self-renewal and proliferation (Ryu et al., 2005). With slight modifications of the culture condition for mouse SSCs, rat SSCs can be expanded in vitro, and progeny derived from cultured rat SSCs could be generated. In this chapter, we describe a detailed method to culture mouse and rat SSCs. Since this is a technical and methodological chapter, neither the developmental biology of male germ cells nor the physiology of spermatogenesis will be reviewed. Excellent publications have described these processes in detail (de Rooij, 1998; Nagy et al., 2003; Russell et al., 1990; Zhao and Garbers, 2002).

Fig. 2.

Histological cross sections of seminiferous tubules of recipient W⁵⁴/W^r testes transplanted with cultured SSCs. Histological sections of testes transplanted with ROSA SSCs that were cultured for 6 months (left, stained with X-gal, counter stain; Nuclear Fast Red) and 7.5 months (right, stain; Hematoxylin-Eosin).

II. Rationale

A. Basic Concept of SSC Culture

Development of new culture conditions for animal cells has been largely empirical. At present no unifying culture technique applicable to various tissue-specific stem cells has been developed. Despite early hopeful expectations by stem cell biologists, recent studies have indicated that characteristics of tissue-specific stem cells do not appear to be conserved (Kubota et al., 2003). However, because the general process of spermatogenesis is believed to be conserved among many species, the self-renewal machinery of SSCs in diVerent species might be similar. In fact, xenotransplantation experiments demonstrated that SSCs from all mammals examined, including rats, rabbits, dogs, pigs, cattle, horses, baboons, and humans, colonized and proliferated or were maintained in the seminiferous tubules of immunodeficient mice (Clouthier et al., 1996; Dobrinski et al., 1999a, 2000; Nagano et al., 2001b, 2002; Oatley et al., 2004). These results suggest that critical exogenous factors to promote self-renewal of SSCs are conserved among various species. Therefore, we employed a systematic approach to develop a SSC culture system using the mouse as a model because once a defined culture condition for mouse SSCs was developed, it would form the basis for other species.

Stem cells generally divide rarely in normal physiological conditions (Meistrich and van Beek, 1993; Morrison et al., 1997; Potten and Morris, 1988). The micro-environment surrounding stem cells, which is called the stem cell niche, controls the timing of proliferation and diVerentiation of the stem cells (Spradling et al., 2001). To maintain stem cells in vitro, reconstitution of the stem cell niche would be ideal, but likely not completely attainable. However, if one can maintain self-renewing stem cells in vitro, the culture condition probably provides essential signals that promote self-renewal found endogenously in the stem cell niche in vivo. Clearly, any knowledge about stem cell niche factors provides valuable information for the development of an in vitro culture system for maintaining stem cells.

To develop a defined culture condition for mouse SSCs, we chose a culture system that consists of a germ cell population enriched for stem cells, serum-free hormonally defined culture medium, and mitotically inactivated STO feeder cells (Kubota et al., 2004a). Originally this culture system was developed for hepatic progenitors, hepatoblasts, in the rat (Kubota and Reid, 2000). A previous study clearly demonstrated that the culture system is useful to investigate stem/progenitor cells and the surrounding microenvironment, because the condition minimized unknown components in the culture (Kubota and Reid, 2000). Such an in vitro system will allow controlled and detailed investigation of factors involved in cell fate decisions. We first optimized the basic culture condition to allow mouse SSCs to survive for a short period by modifying the serum-free medium for hepatoblasts. The modified culture condition was able to maintain mouse SSCs without loss of the stem cell activity assessed by the transplantation assay for at least 1 week (Kubota et al., 2004a). The availability of a functional assay made possible the evaluation of various conditions for survival of SSCs. Subsequently, we sought factors that increase stem cell number in the culture. After extensive screening utilizing the serum-free culture technique and the transplantation assay, we determined extrinsic factors essential for self-renewal proliferation of SSCs in vitro (Kubota et al., 2004b). As described earlier, the culture system consists of three components, enrichment of SSCs, serum-free medium, and feeder cells. Because each element was important for the systematic development of the mouse SSC culture system, we describe the three components in the following sections.

B. Components of SSC Culture System

1. Spermatogonial Stem Cells

The number of SSCs in the testis is very low, presumably as few as 1 in 3000– 4000 cells in the adult mouse testis (Tegelenbosch and de Rooij, 1993). Therefore, unfractionated testis cell suspensions are not ideal as a starting cell population for SSC culture. The great majority of unfractionated testicular cells in adult testes are diVerentiated or diVerentiating germ cells. These germ cells have limited proliferative activity; therefore, they die and disappear gradually in culture although the cell number is high at the beginning. In addition, since diVerentiating germ cells are nonadherent cells, they can be removed easily by changing the medium, while SSCs and spermatogonia generally adhere to the stromal feeder cells or extracellular matrices (Shinohara et al., 1999). Although the diVerentiated germ cells showed limited proliferative activity in culture, there are two reasons to remove them from starting cell populations. First, the stem cells are rare in the testis cell suspension; therefore, it is almost impossible to identify the stem cells in culture. Because microscopic observation of cell divisions in culture is a valuable indication of stem cell proliferation, a large number of diVerentiated cells are an impediment to accurate assessment of stem cell behavior. Second, it is thought that diVerentiating germ cells may provide growth-inhibitory signals to undiVerentiated spermatogonia, including stem cells (Bootsma and Davids, 1988; de Rooij et al., 1985; Meistrich and van Beek, 1993). Existence of such inhibitory signals or negative feedback systems is believed to occur in other stem cell systems as well (LoeZer and Potten, 1997). Although the identity of these signals is not known, it is desirable to eliminate any possibility of a detrimental eVect on SSCs.

Somatic cells, such as Sertoli cells, Leydig cells, myoid cells, and fibroblasts, also exist in the testis, although they are not major cell populations in adult testes. It is believed that terminally diVerentiated cells do not proliferate; therefore, most somatic cells will disappear in culture during subculturing. However, because they could produce a variety of hormones, growth factors, or extracellular matrices, it is important to remove as many somatic cells as possible to minimize the complexity of the culture conditions. In addition, fibroblastic cells generally can proliferate in regular culture medium containing animal serum; even if they are isolated from adult tissues (see below). Based on our experience, SSCs from testes of immature animals are easier to culture, but removal of testicular somatic cells from these cell suspensions is even more critical. Rapid overgrowth of fibroblastic cells can be a major problem in primary cultures, and fibroblastic cells from young animals have a high proliferative ability. In the adult, the number of testicular fibroblastic cells is fewer. However, they can gradually become dominant if serum is added to the culture medium. Serum-supplemented culture conditions selectively expand fibroblasts, because serum contains a variety of trophic factors for fibro-blasts (Sato et al., 1960). Use of a germ cell population enriched for SSCs is crucial to diminish the detrimental eVects of fibroblastic cells and other somatic cells on SSC survival and proliferation.

2. Serum-Free Defined Medium

In general, addition of serum facilitates survival and proliferation of animal cells in vitro; however, several crucial drawbacks exist. First, serum contains complex materials, which have been as yet poorly defined or characterized. In addition, there is considerable batch variation depending on the physiological conditions, sex, and age of donors. Second, serum contains inhibitors of certain tissue-specific cells (Barnes and Sato, 1980b; Enat et al., 1984). Third, serum enriches growth factors for mesenchymal cells, such as platelet-derived growth factor or fibroblast growth factors; therefore, mesenchymal cells, particularly fibroblasts, selectively overgrow in serum-supplemented medium (Sato et al., 1960). These fibroblasts produce factors to inhibit proliferation of other cell types.

Serum-free hormonally defined medium was developed by Gordon Sato's group in the 1970s (Barnes and Sato, 1980b; Bottenstein et al., 1979; Hayashi and Sato, 1976). This series of studies revealed that one of the major functions of serum is to provide hormones or growth factors which stimulate replication of cells. In serum-free media supplemented with specific hormones or growth factors, many mammalian cells were able to be maintained without loss of the cell lineage-specific and developmental stage-specific characteristics (Barnes and Sato, 1980a,b; Bottenstein et al., 1979). Although there was no report of long-term cultures to maintain functional germ cells in such serum-free hormonally defined medium, early studies culturing testicular cells in serum-free medium clearly demonstrated that using defined medium is a powerful approach to study the physiology of testicular somatic cells, such as Sertoli cells or Leydig cells (Mather, 1980; Mather et al., 1981). Serum-free culture is a prerequisite to identify hormones or growth factors essential for self-renewal of SSCs.

3. Feeder Cells

Generally, when a small number of cells are placed in a culture dish, they do not grow well. Conditioned media or feeder cells commonly have been used to culture cells at low cell densities (Ham, 1963). While conditioned media support cell growth by soluble factors, feeder cells are able to stimulate cells cocultured not only by soluble factors but also by insoluble signals through direct cell–cell contact or via extracellular matrices. Originally, nonmultiplying irradiated feeder cells were used to supply conditioning factors for colony formation from single HeLa cells (Puck and Marcus, 1955). The most commonly used coculture system with a feeder layer technique was first developed for human epidermal keratinocyte culture (Rheinwald and Green, 1975). The original culture system consisted of irradiated 3T3 cells as mesenchymal feeders with serum-supplemented medium, and it supported colony formation and serial cultivation of human keratinocytes (Rheinwald and Green, 1975). A coculture system for clonal growth of rat hepatic progenitors using STO feeder cells and a defined serum-free medium (Kubota and Reid, 2000) is an evolved form of the human keratinocyte culture.

Although SSCs can be enriched by several methods (Kubota et al., 2003, 2004a; Shinohara et al., 1999, 2000), it is still laborious to obtain a large number of stem cells. In addition, the slow proliferation rate is a common characteristic of tissue stem cells. Therefore, a coculture system using feeder cells is reasonable for cultivation of stem cells. Early studies with hematopoietic stem cell culture have shown that coculture using stromal monolayers derived from hematopoietic tissues was able to support self-renewal of hematopoietic stem/progenitor cells for several months (Dexter et al., 1977). It is worthwhile to point out that primary mouse embryonic fibroblast (MEF) feeders or STO feeder cells have generally been used for culture of germline-derived pluripotent stem cells, such as embryonic carcinoma (EC) stem cells, embryonic stem (ES) cells, or embryonic germ (EG) cells (Evans and Kaufman, 1981; Martin, 1981; Martin and Evans, 1975; Matsui et al., 1992; Resnick et al., 1992).

One drawback to using feeder cells is that they produce factors that are not defined. Although our culture system uses feeder cells, they were prepared from a well-established mouse cell line, STO cells, instead of primary MEFs. Recent studies have demonstrated that fibroblasts from diVerent anatomic locations are significantly diVerent, and substantial heterogeneity of embryonic fibroblasts exists (Chang et al., 2002). Using feeders of an established cell lines minimizes the variability of the unknown contribution of the feeder cells to the culture conditions.

III. Methods

A. Protocol for Mouse SSC Culture

1. Preparation of Testis Cell Suspension and Enrichment of SSCs

Identification of surface antigens on SSCs is necessary for immunoselection. Using fluorescence-activated cell sorting (FACS) in conjunction with the transplantation assay, the surface phenotype of mouse SSC was determined to be Thy-1⁺ αV-integrin ^–/dim α6-integrin⁺ c-kit^– and major histocompatibility complex class I (MHC-I)^– (Kubota et al., 2003). Thy-1 is a surface marker expressed on SSCs in neonatal, pup, and adult testes of the mouse. Following identification of the surface antigen for SSC by FACS, a method to enrich mouse SSCs using microbeads conjugated with a Thy-1 antibody was developed (Kubota et al., 2004a). The method, magnetic-activated cell sorting (MACS), using Thy-1 antibody cannot completely purify the Thy-1⁺ aV-integrin ^–/dim a6-integrin⁺ c-kit^– MHC-I^– cell population, but the technique requires less time and less tissues to obtain an SSC-enriched cell population. The SSC content of the cell population (MACS Thy-1 cells) depends on the age of animal. Usually, testis cells from younger mice result in cell populations with a greater percentage of SSCs than do cells from adult testes, because fewer Thy-1⁺ somatic cells are present in younger testes. Gonocytes (prespermatogonia) can be enriched by MACS from neonatal testes; however, they are quiescent in the testis and appeared to take more time to start proliferating at the beginning of the SSC culture than do MACS Thy-1 cells from pup testes (unpublished observation). Based on our experience, MACS Thy-1 cells from pup testis at 5–8 days postpartum (dpp; day of birth is 0 dpp) are the best population for SSC culture in the mouse. The following procedure is for cell preparation of MACS Thy-1⁺ cells from 4 to 8 pup testes (2–4 pups) at the age of 5–8 dpp. The procedure consists of three steps: digestion of pup testes; Percoll fractionation; and MACS separation. The procedure to enrich Thy-1⁺ cells by MACS was performed according to the manufacture's protocol. Details for preparing each reagent in the following procedures are described in the Materials section.

Testis cell preparation

1. Remove testes from pups with fine forceps using sterile procedures and collect the testes in a 35-mm petri dish in 3 ml of Hank's balanced salt solution (HBSS).

2. Transfer the testes to a second dish of HBSS and remove tunica albuginea under a dissecting microsope.

3. Using a p200 pipette, transfer the testis tissue without tunica to a 15 ml conical centrifuge tube containing 0.5 ml of 7 mg/ml DNase I solution and 4.5 ml of 0.25% Trypsin-EDTA.

4. Pipette up and down with p1000 pipette to disperse seminiferous tubules.

5. Incubate the tissues at 37 °C for 5 min.

6. After pipetting with p1000 pipette several times, incubate the tube at 37 °C for an additional 3 min. At this point, the cell suspension will be viscous.

7. Add 0.7 ml of fetal bovine serum (FBS) (one tenth volume) to stop enzymatic digestion, also add 0.5 ml of 7 mg/ml DNase to digest genomic DNA from dead cells.

8. Pipette well to make a single cell suspension. If cells remain clumped, add another 0.5 ml of 7 mg/ml DNase and pipette again.

9. Filter the cell suspension through a 40-μm pore nylon cell-strainer (BD Biosciences 352340) followed by washing the cell-strainer with HBSS.

10. Centrifuge the cell suspension at 600 × g for 7 min at 4 °C.

11. Remove supernatant and resuspend cells in 10 ml of PBS-S (Dulbecco's PBS supplemented with 1% FBS, 10 mM HEPES, 1 mg/ml glucose, 1 mM pyruvate, 50 units/ml penicillin, and 50 μg/ml of streptomycin).

12. Count cells. Typical testicular cell number obtained by this method is 1 × 10⁶ to 1.3 × 10⁶ cells/testis at 5 dpp and 1.7 × 10⁶ to 2 × 10⁶ cells/testis at 8 dpp.

Percoll fractionation

1. Slowly overlay 5 ml of cell suspension on 2 ml of 30% Percoll solution in a 15 ml conical centrifuge tube. Do not put more than 2 × 10⁷ cells in the 15 ml tube.

2. Centrifuge at 600 × g for 7 min at 4 °C without using the centrifuge brake.

3. Carefully remove the cells and debris at the interface between the HBSS and the 30% Percoll solution. Then, remove all aqueous phases containing HBSS and 30% Percoll solution. Leave the pellet at the bottom of the tube.

4. Resuspend the pellet of cells in 2 ml of PBS-S and transfer the cell suspension into a 5 ml polypropylene tube (BD Biosciences 352063). Count cell number. The range of cell recovery is 40–70%.

5. Centrifuge cell suspension at 600 × g for 7 min at 4 °C.

6. Resuspend the pellet in 90 μl of PBS-S and pipette well to make a single cell suspension.

MACS separation

1. Add 10 ml of magnetic microbeads conjugated with anti-Thy-1 antibody (Miltenyi Biotec 130–049–101, Auburn, CA) into the 90 μl of the cell suspension and mix well. Ten microliter of Thy-1 microbeads is for one separation column. The typicalrangeof the cell numberfor 10 μl of Thy-1 microbeads is 3 × 10⁶ to 6 × 10⁶ cells. Use 20 μl of Thy-1 microbeads when cell number is more than 6 × 10⁶ cells.

2. Incubate the cell suspension containing Thy-1 microbeads for 20 min at 4 °C. Mix gently by tapping every 10 min. Longer incubation (for 30–40 min) occasionally increases the recovery rate of Thy-1⁺ cells.

3. Add 2 ml of PBS-S to the tube to dilute Thy-1 microbeads and centrifuge at 600 × g for 7 min at 4 °C. Remove the supernatant completely and resuspend in 1 ml of PBS-S.

4. Place a separation column (MS Column; Miltenyi Biotec 130-042-201) in the magnetic field of the mini MACS Separation Unit (Miltenyi Biotec 130-142-102) and rinse with 0.5 ml of PBS-S.

5. Apply the cell suspension to the column. After the cell suspension has passed through the column and the column reservoir is empty, wash the column with 0.5 ml of PBS-S three times.

6. Remove the column from the MACS Separation Unit and elute the magnetically retained cells slowly into a 5 ml polypropylene tube (BD Biosciences 352063) with 1 ml of serum-free culture medium (see below) using the plunger supplied with the column.

7. Centrifuge the tube containing the cells at 600 × g for 7 min at 4 °C and resuspend the cell pellet with 1 ml of mouse serum-free medium (SFM) for rinsing. Repeat this step once.

8. After the final rinsing step, resuspend cells in 0.5 ml of SFM and count the cell number. The recovery of Thy-1⁺ cells from one MS Column is 1.6 × 10 to 2.2 × 10⁵ cells.

2. Serum-Free Medium

The composition of SFM for mouse SSCs is shown in Table I. To prepare mouse SFM, bovine serum albumin (BSA) powder and antibiotics are added to MEMα medium and stored at 4 °C overnight, because it takes several hours to dissolve BSA completely. The following day, stock solutions of transferrin, free fatty acid (FFA) mixture (Chessebeuf and Padieu, 1984, see Table II), selenium (Na₂SeO₃), freshly prepared 2-mercaptoethanol (2-ME), insulin, HEPES, and putrescine are added to the MEMα containing BSA and antibiotics. Preparation of stock solutions for each supplemental component is described in the Materials section. SFM is sterilized by filtration using a 0.2 μm membrane filter and stored at 4 °C.

Table I.

Components in Mouse Serum-Free Medium (SFM) and Modified Mouse SFM

	Manufacturer	Catalog number	Mouse SFMa	Modified mouse SFMb
MEMα	Gibco/Invitrogen	12561	Basal media	Basal media
Penicillin	Gibco/Invitrogen	15140	50 units/ml	50 units/ml
Streptomycin	Gibco/Invitrogen	15140	50 μg/ml	50 μg/ml
BSA	MP Biomedicals/ICN	See Notec	0.2%	0.6%
Transferrin	Sigma	T 1283	10 μg/ml	100 μg/ml
FFA mixture	Sigma	See Table II	7.6 μeq/L	15.2 μeq/L
Na2SeO3	Aldrich/Sigma	481815	3 × 10–8 M	6 × 10–8 M
l-glutamine	Gibco/Invitrogen	25030	2 mM	2 mM
2-ME	Sigma	M 7522	50 μM	100 μM
Insulin	Sigma	I 5500	5 μg/ml	25 μg/ml
HEPES	Sigma	H 0887	10 mM	10 mM
Putrescine	Sigma	P 5780	60 μM	120 μM

^aThe concentration of each component in serum-free medium for mouse SSC culture is indicated.

^bModified mouse SFM is made by increasing several components of mouse SFM. Serum-free medium for rat SSC culture is modified mouse SFM plus 10% DDW (10 ml of water is added to 100 ml of modified mouse SFM). Thus, the final concentrations of each supplemental component in rat SFM is about 90% of that in modified mouse SFM.

^cNote: The source and lot of BSA are important. BSA of MP Biochemicals (formerly ICN), catalog number 810661 (lot number 2943C) and 194774 (lot number R14550), supported long-term in vitro proliferation of mouse SSCs. BSA from Sigma, catalog number A3803 (lot numbers 064K0720, 025K1497, and 124K0729), could be used for mouse SSC culture. For rat SSC culture, BSA of MP Biochemicals, catalog number 810661 (lot number 2943C and 4561H), was effective.

Table II.

Preparation of Free Fatty Acid Mixture

Free fatty acid (FFA)	Manufacturer	Catalog number	Stock solutiona	100 meq/l FFA mixturebb
Linolenic acid	Sigma	L 2376	1 M	5.6 μl (5.6 mM)
Oleic acid	Sigma	O 1008	1 M	13.4 μl (13.4 mM)
Palmitoleic acid	Sigma	P 9417	1 M	2.8 μl (2.8 mM)
Linoleic acid	Sigma	L 1012	1 M	35.6 μl (35.6 mM)
Palmitic acid	Sigma	P 0500	1 M	31.0 μl (31.0 mM)
Stearic acid	Sigma	S 4751	151 mM	76.9 μl (11.6 mM)
Absolute ethanol				834.7 μl
Final volume				1000 μl (100 meq/l)

^aEthanol is used to make stock solutions of each FFA. One molar stock solutions of linolenic acid, oleic acid, palmitoleic acid, and linoleic acid are liquid at room temperature. One molar palmitic acid and 151 mM stearic acid are solid. These two FFA stock solution need to be heated to 45–50 °C to dissolve.

^bSeventy-six microliter of 100 meq/l FFA mixture is added to 1000 ml of serum-free medium. This makes 7.6 μeq/l FFA mixture at the final concentration in the medium. Numbers in parentheses indicate final concentration in serum-free medium.

3. Preparation of Feeder Cells

STO cell s (SNL76/7 cell s) (McMah on and Bra dley, 1990) were obtained from Dr. A. Bradley (The Wellcome Trust Sanger Institute, London). STO cells are routinely grown in 10 cm dishes in Dulbecco's modified Eagle's medium (DMEM) supplemented with 7% FBS (serum-supplemented medium for STO cells, SSM/STO). Although the culture methods are based on previous reports using newborn calf serum for the growth medium (Robertson, 1987), STO cells in the medium with FBS appeared to grow faster. Cultures of STO cells should be passaged promptly when they reach confluency because it is possible to accumulate noncontact-inhibited cells in the population. In addition, continuous passage of STO cells may also select clones that cannot support SSC culture. Therefore, it is advisable to expand STO cells at early passages and cryopreserve immediately in several low-passage aliquots.

1. Add mitomycin C at a final concentration of 10 μg/ml in culture medium for confluent 10 cm dishes of STO cells. Incubate the dishes for 3–4 h at 37 °C. Avoid more than 5 h exposure to mitomycin C.

2. Remove the medium containing mitomycin C from the STO cells and wash plates three times with 10 ml of HBSS.

3. Digest the cells with trypsin-EDTA solution for 3 min at 37 °C and mix with SSM/STO to stop the trypsin activity. Collect digested cells and centrifuge at 600 × g for 7 min at 4 °C. Resuspend the cells in SSM/STO at 12 × 10⁶ cells/ml. Add to the cell suspension an equal volume of 2× freeze medium [20% dimethyl sulfoxide (DMSO) in SSM/STO] slowly at 4 °C. The final cell concentration will be 6 × 10⁶ cells/ml in 10% DMSO in SSM/STO.

4. Aliquot 1 ml of the STO cell suspension into each cryotube and store at –70 °C using a freezing container (Nalgene 5100–0001).

5. For preparation of monolayers of STO cell feeders, thaw one vial of mitomycin C-treated STO cells at 37 °C and put the cell suspension into a 50 ml conical centrifugation tube.

6. Add slowly 9 ml of SSM/STO on ice to dilute the freeze medium and mix gently. Centrifuge at 600 × g for 7 min at 4 °C.

7. Resuspend the cell pellet with SSM/STO to make a cell suspension at a cell concentration of 2 × 10⁵ cells/ml. The cell viability is usually more than 90%.

8. Seed the cells on to gelatinized plates at a concentration of 5 × 10⁴ cells/cm². Typically put 1 ml of the cell suspension (2 × 10⁵ cells/ml) into one well of a 12-well plate. Gelatinized plates are prepared by precoating plates with 0.1% gelatin for 1 h at 37 °C. STO monolayers can be created with fresh mitomycin-C treated STO cells as well.

9. Use the STO feeder cells within 4 days. If older STO feeder cells (5–7 days after seeding) are used for SSC culture, change the medium on the fourth day.

4. Culture of Mouse SSCs

While the culture of SSCs is dependent on the presence of appropriate growth factors, other elements can influence cell proliferation. After placing freshly isolated MACS Thy-1 cells in the primary culture, initiation of germ cell clump formation is an important indication of likely successful continuous culture of SSC (Fig. 3A) (Kubota et al., 2004b). Each culture shows some diVerence in the number of germ cell clumps and the growth speed. Probably the age of the testes, purity of Thy-1⁺ cells, and damage caused by cell preparation are factors influencing initial and continuous clump formation. One of the most influential factors observed is the mouse strain used. While SSCs from DBA/2 mice proliferate easily in the presence of GDNF alone, stem cells from other mouse strains such as C57BL/6 or 129/SvCP require soluble GDNF family receptor alpha 1 (GFRα1) and basic fibroblast growth factor (bFGF or FGF2) for continuous proliferation in vitro (Kubota et al., 2004b). The growth advantage of DBA/2-derived SSCs was reported earlier (Kanatsu-Shinohara et al., 2003). Cellular responses to GDNF are mediated by a multicomponent receptor complex consisting of RET receptor tyrosine kinase and a glycosil phosphatidylinositol-anchored ligand-binding subunit, GFRα1 (Sariola and Saarma, 2003). Because RET stimulation by soluble GFRα1 potentiates downstream signaling (Paratcha et al., 2001), addition of soluble GFRα1 may play an important role for in vitro proliferation of SSCs in most mouse strains. bFGF has been shown to be a potent growth factor for in vitro proliferation of primordial germ cells (Matsui et al., 1992; Resnick et al., 1992). Therefore, bFGF may also provide a critical stimulus to support SSC replication in culture. Moreover, recent studies suggest that bFGF may play an important role for SSC proliferation in human testes (Goriely et al., 2003).

Fig. 3.

Microscopic observation of cultured SSCs. (A) development of germ cell clumps from 129/ SvCP Thy-1⁺ pup testis cells isolated by MACS. MACS Thy-1 cells were placed on STO feeder cells in serum-free defined medium supplemented with GDNF. Single Thy-1⁺ cells at 5 h in culture (left). Initiation of cell clump formation at 2 days (middle). Growth of germ cell clumps at 5 days (right). (B) Comparison of feeder cells for their ability to support germ-cell clump formation and growth. C57BL/6-derived SSCs were placed on STO, MEF, or MSC-1 feeders in the presence of GDNF, GFRα1, and bFGF, and cultured for 9 days.

From our experience, SSCs from DBA/2 appear to proliferate faster in the SFM supplemented with GDNF, GFRα1, and bFGF than in GDNF alone. In the presence of three factors, SSCs expand in culture easily from DBA/2 mice, with more diYculty from C57BL/6 mice and the most diYculty from 129/SvCP mice.

Primary culture of Thy-1⁺ germ cells

1. Remove culture medium from STO monolayer cultures and rinse the plates with HBSS twice to wash out residual medium containing serum.

2. Place 5 × 10⁴ to 10 × 10⁴ MACS Thy-1 cells in wells of 12-well plates containing STO monolayers. Two to four days after seeding of STO feeder cells in gelatin-coated wells is optimal for SSC primary culture.

3. Add recombinant human GDNF, rat GFRα1, and human bFGF at a final concentration of 20 ng/ml, 150 ng/ml, and 1 ng/ml, respectively.

4. Maintain cells at 37 °C in a humidified 5% CO₂ atmosphere. Change the medium every other day. By 48 h in culture, Thy-1⁺ germ cells form small clumps with tight intercellular contacts (Fig. 3A). They continuously proliferate and form large clumps (Fig. 3A).

Recently, several studies reported SSC culture using feeder cells. Kanatsu-Shinohara et al. used MEF for long-term culture of SSCs derived from DBA/ 2 gonocytes (Kanatsu-Shinohara et al., 2003). Hamra et al. reported that feeder cells from a mouse Sertoli cell line (MSC-1) to be superior to STO feeder cells, although germ cells were cultured on the feeder cells for only a short period (Hamra et al., 2004, 2005). Both culture conditions contain FBS. MSC-1 was established from transgenic mice carrying a fusion gene composed of human Müllerian inhibitory substance transcriptional regulatory sequences linked to the SV40 T-antigen gene (Peschon et al., 1992). Using our culture system, MEF feeders supported initiation of germ cell clumps; however, the growth was not as fast as STO cell feeders (Fig. 3B). MSC-1 feeder cells maintained germ cell survival and proliferation poorly (Fig. 3B). By 9 days after initial plating, most germ cells on MSC-1 feeder layers died.

The first subculture is performed 6–9 days after initial culture of MACS Thy-1 cells on STO feeders. The second and subsequent subcultures also are performed at a similar interval (6–9 days). Timing of subculture must be determined subjectively on the basis of the number of proliferating germ cells and testicular fibroblasts.

1. Remove culture medium and add 0.5 ml of trypsin-EDTA solution per well of a 12-well plate to digest the germ cell clumps.

2. Incubate cultures for 3–5 min at 37 °C and add 0.1 ml of FBS to stop digestion. Pipette digested cells gently with p1000 pipette several times. In some instances, genomic DNA from dead cells may cause cell aggregation and clumping, especially when many germ cells are cultured for more than 1 week. In that case, add 0.1 ml of 7 mg/ml DNase solution with the FBS.

3. Collect digested cells from a well of the 12-well plate in a 15 ml conical centrifugation tube and dilute the cell suspension with 2 ml of mouse SFM. Centrifuge at 600 × g for 7 min at 4 °C.

4. Resuspend the cell pellet with 2 ml of mouse SFM and centrifuge at 600 × g for 7 min at 4 °C. Cells are resuspended in 3 ml of SFM and plated onto fresh STO cell feeders. The split ratio of the first subculture is 1:2.

5. Sometimes testicular fibroblastic cells outgrow clump-forming germ cells. Because clump-forming germ cells are attached on STO feeders weakly, they can be removed from the feeders by pipetting gently with p1000 pipette (Ryu et al., 2005). Following gentle pipetting on the surface of feeder cells, collect the culture medium containing the detached germ cell clumps and centrifuge at 600 × g for 7 min at 4 °C.

6. Resuspend the pellet in mouse SFM and pipette gently several times to break up cell clumps. Place cells onto fresh STO feeder cells in the same size well as the original culture (1:1).

Usually, 1 month after initial plating, clump-forming germ cells constantly proliferate on the feeder in mouse SFM supplemented with GDNF, GFRα1, and bFGF (Fig. 4A, phase contrast). Once cultures reach that stage, subculture can be conducted every 4–7 days at the split ratio of 1:2–1:4.

Fig. 4.

Phenotypic characteristics of cultured SSCs. (A) SSCs from C57BL/6 pup testes continuously proliferate on STO feeder cells in serum-free defined medium supplemented with GDNF, GFRα1, and bFGF. A phase contrast image of SSCs cultured for 1 month (left) and immunocytochemistry of RET (middle) and PLZF (right). Clump-forming germ cells were stained with antibodies against RET or PLZF. (B) Cell surface molecules on clump-forming germ cells were analyzed by flow cytometry. Cultured SSCs express GFRα1, EpCAM, and E-cadherin. Filled histograms represent stained cells with antibodies indicated. Open histograms indicate unstained cells.

5. Characteristics of Cultured Mouse SSCs

The clump-forming germ cells express germ cell markers, such as germ cell nuclear antigen 1 (GCNA1) or mouse vasa homologue (MVH) (Enders and May, 1994; Fujiwara et al., 1994; Kubota et al., 2004b; Oatley et al., 2006). More importantly, they show very similar characteristics with freshly isolated SSCs. Previous studies using transplantation assays and FACS demonstrated that the phenotype of SSCs in pup testes is αV-integrin ^–/dim α6-integrin⁺ Thy-1^lo/+ (Kubota et al., 2004a). Flow cytometric analysis showed that continuously cultured germ cells are also αV-integrin ^–/dim α6-integrin⁺ Thy-1^lo/+, and the surface phenotype of cultured germ cells was constant during a 3-month culture period (Kubota et al., 2004b). Transplantation assays indicated that the stem cell activity of clump-forming germ cells during a 3-month culture period and fresh αV-integrin ^–/dim α6-integrin⁺ Thy-1^lo/+ cells isolated by MACS was similar (Kubota et al., 2004b). Usually, the stem cell activity is represented by donor-derived colony number generated in the recipient testes per 10⁵ transplanted cells. In our studies, the values of stem cell activity of freshly isolated cells and cultured cells were both ~500 colonies per 10⁵ αV-integrin ^–/dim α6-integrin⁺ Thy-1^lo/6 cells transplanted (Kubota et al., 2004b). Since the colonization eYciency is ~5%, 1 in 10 clump-forming germ cells was an SSC. These results indicate that the culture conditions eYciently support proliferation of αV-integrin ^–/dim α6-integrin⁺ Thy1^lo/+ cells. All clump-forming germ cells expressed RET tyrosine kinase, which is the signal transducer of GDNF (Fig. 4A) (Kubota et al., 2004b). Although soluble GFRα1 was added to the culture medium, cultured germ cells expressed GFRα1 on the cell surface (Fig. 4B) (Kubota et al., 2004b). In addition to RET and GFRα1 expression, clump-forming germ cells are positive for several spermatogonia markers including promyelocytic leukaemia zinc finger (PLZF) protein, epithelial cell adhesion molecule (EpCAM, CD326), and E-cadherin (Fig. 4A and B) (Buaas et al., 2004; Costoya et al., 2004; Tokuda et al., 2007; van der Wee et al., 2001).

The cultured SSCs shared several characteristics of undiVerentiated ES cells, such as POU5F1 (previously known as Oct-3/4) expression and alkaline phosphatase activity (Kubota et al., 2004b). However, significant diVerences exist. For example, although FBS supported ES cells in culture, the constituents of FBS are detrimental to SSC proliferation (Kubota et al., 2004b). When SSCs were cultured in medium containing only 0.1% FBS, proliferation of SSCs was dramatically decreased compared with that of SSCs in SFM (Kubota et al., 2004b). Cultured SSCs do not generate tumors when transplanted to immunocompromised mice (Kubota et al., 2004b), whereas ES cells produce teratocarcinomas when injected into mice (Evans and Kaufman, 1981; Martin, 1981). In addition, SSCs are negative for Nanog, which is highly expressed in undiVerentiated ES cells (Chambers et al., 2003; Mitsui et al., 2003; Oatley et al., 2006).

B. Protocol for Rat SSC Culture

1. Enrichment of Rat SSCs

SSCs from rat pup testis can be enriched by immunoselection using an EpCAM antibody (Ryu et al., 2004). In the rat pup testis (8 dpp), ~8% of testis cells are EpCAM⁺. The transplantation assay indicated that almost all SSCs express EpCAM (Ryu et al., 2004). EpCAM is a homophilic cell–cell adhesion molecule and has been used for identification and purification of primordial germ cells or spermatogonia in rodents (Anderson et al., 1999; Moore et al., 2002; van der Wee et al., 2001). Although rat SSCs express Thy-1 antigen, a subpopulation of the testicular somatic cells also express this surface molecule; therefore, MACS using Thy-1 antibody did not enrich rat SSCs eYciently. To obtain an SSC-enriched cell population for culture, we developed a protocol for isolation of EpCAM⁺ cells using MACS with anti-EpCAM antibody (MACS EpCAM cells). The protocol is for 4–8 testes (2–4 pups) from 8–12 dpp rat pups.

Testis cell preparation and MACS separation

1. Prepare testes for digestion as described in the mouse section.

2. Put testes into 10 ml of 1 mg/ml collagenase solution and incubate 5–8 min. Tap the bottom of the tube gently every 2–3 min.

3. Centrifuge at 600 × g for 1 min at 4 °C to collect loose seminiferous tubules.

4. Remove supernatant and add 10 ml of HBSS to loosen tubules by inverting the tube. Repeat steps 3 and 4 for rinsing.

5. Remove supernatant and add 1.5 ml of 7 mg/ml DNase and 6 ml of trypsin-EDTA.

6. Digest the tissues as described in the mouse protocol.

7. Add 0.9 ml of FBS and 1 ml of 7 mg/ml DNase and pipette well using a p1000 pipette.

8. Filter the cells through a 40 mm pore nylon cell strainer and spin down cells at 600 × g for 7 min at 4 °C.

9. Resuspend cells in 14 ml of MEMa containing 1% FBS and count the cell number. Typical cell recovery is 10 × 10⁶ to 13 × 10⁶ cells/testis at 9–10 dpp.

10. Employ Percoll fractionation as described in the mouse protocol.

11. For MACS separation, resuspend 50 × 10⁶ cells of Percoll-fractionated cells in 5 ml of PBS-S.

12. Add 0.3–0.5 μg of anti-rat EpCAM antibody (Clone: GZ1) per 10⁶ cells for labeling with primary antibody and incubate for 20 min on ice.

13. After rinsing the cells twice with 10 ml of PBS-S, resuspend cells in 0.4 ml of PBS-S. Add 0.1 ml of goat anti-mouse IgG microbeads (Miltenyi Biotec, 130–048–402) to label magnetically EpCAM⁺cells. Incubate cells for 20 min on ice.

14. Add 10 ml of PBS-S and centrifuge at 600 × g for 7 min at 4 °C.

15. Remove the supernatant and resuspend in 2 ml of PBS-S.

16. Separate EpCAM ⁺ cells using two MS column according to the manufacture's protocol. For detail, see the mouse protocol described earlier.

Approximately 4–6% of cells applied to MS columns are recovered. For example, 2 × 10⁶ to 3 × 10⁶ EpCAM⁺ cells are obtained from 50 × 10⁶ cells (see step 11).

2. Culture of Rat SSCs

The basic culture condition for rat SSC culture consists of SFM, STO feeder cells, and a growth factor combination that are the same as for mouse SSC culture. Although components of SFM for rat SSCs are identical to those of mouse SFM, the concentrations of several components in the SFM for rat SSCs were increased. The SFM that contains increased concentrations of several supplements was designated modified mouse SFM. The concentration of each supplement in the modified mouse SFM is indicated in Table I. In addition, reduction of medium osmolarity by the addition of water resulted in an increase in clump-forming germ cells (Ryu et al., 2005). The modified mouse SFM diluted with 10% (vol/vol) distilled water was designated rat SFM.

To prepare rat SFM, BSA powder, transferrin, and antibiotics are added to MEMα, and the medium is stored at 4 °C overnight. The next day, stock solutions of FFA, Na₂SeO₃, freshly prepared 2-ME, insulin, HEPES, and putrescine are added to the MEMα containing BSA and transferrin followed by sterilization by filtration through a membrane filter of 0.2 μm pore size.

Because the concentration of several nutrients is increased in the rat SFM, the medium supports somatic cell growth better than mouse SFM. As a result, we observed during cultures a gradual increase in the number of testicular somatic cells contaminating the original MACS EpCAM cell population. Because testicular somatic cells interfered with SSC maintenance and replication (Kubota et al., 2004a), their number was decreased when necessary by removing germ cell clumps at the time of subculture using gentle pipetting of medium across the surface of the feeder layer and recovery of the detached clumps with the culture medium. The collected clumps are digested with a 1:25 dilution of trypsin/EDTA (final 0.01% trypsin/40 nM EDTA) for 1 min or longer if necessary. Enzymatic digestion is stopped by FBS. Following gentle pipetting, clump-forming germ cells become small clumps, but not single cells. After rinsing the germ cells twice, place them on fresh STO feeder layers. During the first 2 months after initial plating of MACS EpCAM cells, the split ratio for subculture should be 1:1 or 1:1.5 at an interval of every 7–10 days. When testicular fibroblasts outgrow germ cells in culture, germ cell clumps can be collected by gentle pipetting, even though the sizes of clumps are small. In that case, enzymatic digestion is not necessary, because trypsin treatment damages clump-forming cells and results in a decrease in cell recovery. Once clump-forming germ cells constantly proliferate in culture, subculture can be conducted at the split ratio of 1:2. Low atmospheric oxygen (5%) was used for long-term culture of rat SSCs (Ryu et al., 2005). Low atmospheric oxygen has shown a beneficial eVect on long-term proliferation of several types of mammalian cells in vitro (Ezashi et al., 2005; Parrinello et al., 2003).

3. Characteristics of Cultured Rat SSCs

The surface phenotype of freshly isolated rat SSCs in pup testis is EpCAM⁺ Thy-1^lo β3-integrin^– (Ryu et al., 2004). The surface antigen profile of rat SSCs cultured for more than 10 months was basically similar to that of fresh SSCs (Ryu et al., 2005). However, the surface expression of these antigens increased slightly in cultured cells. POU5F1 expression and alkaline phosphatase activity are present in cultured rat SSCs, indicating that mouse and rat SSCs share these characteristics found in ES cells or PGCs. Rat SSCs also express RET and GFRα1, which are the receptors for GDNF (Ryu et al., 2005).

IV. Materials

A. Reagents for Cell Preparation

DNase: DNase I (Sigma DN25) solution is prepared at 7 mg/ml in HBSS and sterilized by filtration.

Collagenase: Collagenase type IV (Sigma C5138) solution is prepared at 1 mg/ml in HBSS and sterilized by filtration.

Trypsin-EDTA: 0.25% trypsin and 1 mM EDTA (Gibco/Invitrogen 25200). Antibiotics: 10,000 units/ml penicillin and 10,000 μg/ml of streptomycin (Gibco/ Invitrogen 15140). FBS: FBS (Hyclone) is treated at 56 °C for 30 min. Aliquots are stored at –20 °C.

Pyruvate: Stock solution is prepared at 100 mM sodium pyruvate (Sigma P2256) in distilled deionized water (DDW), sterilized by filtration using 0.2 μm membrane filter, and stored at 4 °C.

PBS-S: Dulbecco's PBS supplemented with 1% FBS, 10 mM HEPES (Sigma H0887), 1 mg/ml glucose (Sigma G6152), 1 mM pyruvate, 50 units/ml penicillin, and 50 μg/ml streptomycin sterilized by filtration using 0.2 μm membrane filter, and stored at 4 °C.

Percoll solution: 30% (vol/vol) Percoll (Sigma P4937) is prepared in Dulbecco's PBS containing 1% FBS, 50 units/ml penicillin, and 50 mg/ml streptomycin. Percoll solution is sterilized by filtration using a 0.2 μm membrane filter and stored at 4 °C.

B. Stock Solution of Reagents for SFM and Cell Culture

Transferrin: Stock solution is prepared at 10 mg/ml in Dulbecco's PBS and stored at –20 °C.

FFA mixture: Detailed information is described in Table II. Make a small volume aliquot (e.g., 50–100 μl/0.5 ml tube) of FFA mixture. For storage of stock solutions of each FFA or FFA mixture aliquot, it is advised to flush vials or tubes with nitrogen before closing to prevent oxidation of FFAs. The tubes are sealed with parafilm and stored at –20 °C.

Na₂SeO₃: Stock solution is prepared at 3 × 10 ^–4 M in DDW and stored at –20 °C.

2-ME: 100 mM solution is prepared freshly each time medium is made. 100 mM solution is prepared by adding 7 μl of 14.4 M 2-ME (Sigma M7522) in 1 ml of MEMα medium.

Insulin: Stock solution is prepared at 10 mg/ml in 10 mM HCl and stored at –20 °C.

Putrescine: Stock solution is prepared at 100 mM in Dulbecco's PBS and stored at –20 °C.

GDNF: Human recombinant GDNF (R&D Systems, 212-GD-010) stock solution is prepared at 20 μg/ml in Dulbecco's PBS containing 0.1% BSA and stored at –70 °C.

GFRα1: Rat recombinant GFRα1/Fc Chimera (R&D Systems, 560-GR-100) stock solution is prepared at 100 μg/ml in Dulbecco's PBS containing 0.1% BSA and stored at –70 °C.

bFGF: Human recombinant bFGF (BD Biosciences, 354060) stock solution is prepared at 10 μg/ml in Dulbecco's PBS containing 0.1% BSA and stored at –70 °C.

C. Reagents for STO Feeder Layers

SSM/STO (serum-supplemented medium for STO cells): The medium for STO cell culture is Dulbecco's modified Eagle's medium (DMEM, Gibco/Invitrogen 11965) supplemented with 7% FBS, 100 μM 2-ME, 10 mM HEPES, 2 mM glutamine, 50 units/ml penicillin, and 50 μg/ml streptomycin. SSM/STO is sterilized by filtration using a 0.2 μm membrane filter and stored at 4 °C.

Mytomycin C: Stock mitomycin C (Sigma M4287) solution is prepared at 200 μg/ml in Dulbecco's PBS and stored at –70 °C.

Freezing medium (2×): 20% dimethyl sulfoxide (DMSO, Sigma D2650) in SSM/STO.

Gelatin: 0.1% gelatin (porcine skin type A, Sigma G2500) in DDW is autoclaved for sterilization and stored at room temperature.

V. Discussion

An exciting practical aspect of the development of culture systems for SSCs is that the technique establishes a foundation for sophisticated genetic manipulation, including targeted modification, of the species from which the stem cells were isolated (Kanatsu-Shinohara et al., 2006a). The conservation of GDNF signaling in mouse and rat SSCs as the essential pathway to stimulate in vitro self-renewal and the previously demonstrated ability of SSC of many species to maintain and proliferate in mouse seminiferous tubules suggest that a similar culture system can be developed to obtain continuous proliferation of SSCs of many mammalian species, including humans. In addition, because a large number of SSCs can be generated in culture, they represent a powerful resource for gene analysis to elucidate the mechanisms governing self-renewal and diVerentiation of the stem cells (Oatley et al., 2006). Moreover, continuous in vitro proliferation of SSCs of any species lays the foundation for the development of systems to support germ cell diVerentiation in vitro. Modulating culture conditions that support diVerentiation processes of male germ cells resulting in production of functional gametes in vitro will create a valuable model for studying the molecular and cellular biology of male germ cell diVerentiation. Such an in vitro experimental system may allow development of new therapeutic strategies for infertility (Brinster, 2007; Kubota and Brinster, 2006). The progressive development of culture systems for SSCs of other mammalian species including humans is imminent and will provide the basis for a wide range of studies on the biology of the stem cell, in vitro diVerentiation of germ cells, and modification of germlines.