| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Testis |
Department of Animal Biology, School of Veterinary Medicine, University of Pennsylvania, Philadelphia, Pennsylvania 19104
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
6-integrin+ v-integrin–/dim throughout postnatal life, the most enriched population of SSCs was obtained from cryptorchid adult testes by cell-sorting techniques based on Thy-1 expression. This enriched population of SSCs was used to develop a culture system that consisted of serum-free defined medium and STO (SIM mouse embryo-derived thioguanine and ouabain resistant) feeders, which routinely maintained stem cell activity for 1 wk. Combining the culture system and the transplantation assay provided a mechanism to study the effect of single growth factors. A negative effect was demonstrated for several concentrations of basic fibroblast growth factor and leukemia inhibitory factor, whereas glial cell line-derived neurotrophic factor and stem cell factor appeared to have a positive effect on stem cell maintenance. The stem cell enrichment strategies and the culture methods described provide a reproducible and powerful assay system to establish the effect of various environmental factors on SSC survival and replication in vitro.
developmental biology, gametogenesis, growth factors, spermatogenesis, testis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Serum-free culture systems are an important approach to investigating the biological properties of mammalian cells in vitro [5, 6]. Serum contains complex undefined materials, and batch variations occur depending on many uncontrollable factors, e.g., the physiological condition or sex of donors. In addition, substances in serum are toxic for certain cell types [5, 7]. Once mammalian cells were shown to proliferate in serum-free hormonally defined medium without altering the cell type-specific characteristics [8], serum-free culture became a major resource to study cells in vitro and to identify novel growth factors or regulatory mechanisms for proliferation and differentiation. Using serum-free culture systems, it was determined that most cell types require specific growth factors and hormones to proliferate in vitro [5, 8]. In culture studies using SSCs, serum has been used at various concentrations, perhaps because embryonic stem (ES) cells have generally been maintained with high concentrations of serum. In early reports on the culture of SSCs, media contained 10% fetal bovine serum (FBS), and some SSCs survived for more than 3 mo [9]. A similar concentration of serum was present in media testing the effect of growth factors and various feeder cell types [10]. Recently, long-term survival and proliferation of SSCs were reported in a proprietary medium (Stem Pro-34 SFM; Invitrogen, Carlsbad, CA) with 1% FBS and mouse embryonic feeder cells [11]. While this medium contains serum and is not defined, the long-term proliferation of SSCs in vitro is a significant development. A major challenge still remaining is to establish a defined serum-free culture condition that supports maintenance of the stem cell and allows definitive experiments to analyze the effect of individual medium modifications on proliferation.
Cell-fate determination between self-renewal or differentiation of SSCs in the testis is precisely regulated to maintain normal spermatogenesis. Fate determination of stem cells is controlled to a large extent by the surrounding microenvironment, particularly the stem cell niche [12]. Little is known about the components of the stem cell niche. However, studies with hematopoietic stem cells suggest that feeder cells are an essential element to reconstitute stem cell niches in vitro [13]. Likewise, coculture with STO (SIM mouse embryo-derived thioguanine and ouabain resistant) cell feeders improved in vitro maintenance of SSCs compared with no feeders, although the coculture system maintained only 10–20% of stem cells for 7 days [9, 10]. This result suggested that STO cell feeders, which can support ES cells [14], might reconstitute a stem cell niche for SSCs in vitro. However, because crude cryptorchid testis cell populations were used as a stem cell source in the study, it is not clear whether STO cell feeders alone provide the beneficial effects on SSC survival or the combination of STO cells and testis cells was necessary. Alternatively, components of the testis somatic cell population may have been detrimental to SSC maintenance by stimulating differentiation [10]. To avoid ambiguity or a negative effect associated with the diverse testis somatic cell population on SSC maintenance, it is important to use highly enriched SSCs for in vitro culture studies.
Because SSCs are rare in the testis, presumably 1 in 3000–4000 cells in adult mouse testis [15], several approaches to enrich stem cells have been attempted. Experimental cryptorchid surgery resulted in approximately a 20- to 25-fold enrichment of SSCs [16]. Cell suspensions from cryptorchid mouse testis contained about one SSC in 200 cells. In addition, immunological separation using surface antigenic properties is a major approach for enrichment of SSCs [17, 18], as has been shown in other stem cell systems [19–21]. To obtain a pure or highly enriched stem cell population, it is critical to identify unique surface markers that are expressed on stem cells because the antigenic profile of stem cells establishes the basis for selective separation. Particularly, identification of surface markers that are expressed uniquely on SSCs, but not on other somatic cells or differentiated spermatogenic cells, facilitates enrichment of SSCs. It is also important to establish that expression of stem cell markers is conserved during development, indicating possible association with biological properties of the stem cells.
In a previous study, we identified Thy-1 as a positive marker expressed uniquely on SSCs [17]. Thy-1 is a glycosyl phosphatidylinositol-anchored surface antigen and is expressed on other stem cells, including hematopoietic stem cells, mesenchymal stem cells, or ES cells [20, 22, 23]. The data indicated that major histocompatibility complex class I (MHC-I)– Thy-1+ c-kit– cells isolated by flow cytometric sorting from experimental cryptorchid testis cells contained SSCs at a concentration of 1 in 15 cells and that the MHC-I– Thy-1+ c-kit– cells contained almost all the SSCs in the testis [17]. Because most of the MHC-I– Thy-1+ cells in the testis were c-kit–, Thy-1 antigen is a key molecule to enrich SSCs. However, the expression of Thy-1 on SSCs in neonate or pup testis has not been examined. Therefore, it is unclear whether SSCs express Thy-1 constitutively throughout postnatal life. Although the concentration of SSCs appears to be lower in neonatal and pup testes than in cryptorchid [24], it has not been determined whether stem cell activity of SSCs enriched by a common characteristic from neonate, pup, and adult testes are identical.
In this study, we hypothesized that an in vitro system that consisted of highly enriched populations of SSCs and defined culture conditions would facilitate identifying external factors that affect replication of SSCs and studying mechanisms of stem cell fate determination. A primary objective of the research was to develop a serum-free defined culture system for SSCs in which the effects of single growth factors could be determined.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cryptorchid and wild-type adult donor testis cells were obtained from the transgenic mouse line B6.129S7-Gtrosa26 (designated ROSA; The Jackson Laboratory, Bar Harbor, ME) that express the Escherichia coli lacZ gene in virtually all cell types, including all stages of spermatogenesis [25]. Neonate (0.5–1.5 days postpartum, dpp; day of birth is 0.5 dpp), and pup (4.5–5.5 dpp) testis cells were collected from the hemizygous transgenic mice, C57BL/6 x ROSA F1 hybrid. Experimental cryptorchid testes were produced as previously described [16]. Cell suspensions from cryptorchid adult and wild-type adult, neonate, and pup testes were prepared by enzymatic digestion [26]. In several experiments, testis cells were fractionated by using Percoll (Sigma, St. Louis, MO) to remove cellular debris and large cells. The dissociated testis cell suspension, after enzymatic digestion, was overlaid on 30% (v/v) Percoll prepared in Dulbecco PBS containing 1% FBS (Hyclone, Logan, UT) and centrifuged at 600 x g for 8 min at 4°C. Cells from the interphase and Percoll phase were collected as top fraction. Sedimented cells were used as bottom fraction.
Cell Staining and Fluorescence-Activated Cell Sorting
Dissociated testis cells were suspended (5 x 106 cells/ml) in Dulbecco PBS supplemented with 1% FBS, 10 mM HEPES (Sigma), 1 mM pyruvate (Sigma), antibiotics (50 U/ml penicillin and 50 µg/ml streptomycin; Invitrogen), and 1 mg/ml glucose (Sigma) (PBS-S). All antibodies were obtained from BD Biosciences (Franklin Lakes, NJ) unless otherwise stated. For fluorescence-activated cell sorting (FACS) analysis of cryptorchid adult testis cells, dissociated cells were incubated with anti-
2 microglobulin (2M; S19.8) and biotin-conjugated anti-Thy-1 (53-2.1) antibodies for 20 min on ice and washed twice with excess PBS-S. All staining and washing were performed with a similar protocol. Then cells were stained with Alexa Fluor 647-conjugated goat anti-mouse IgG2b antibody (Alexa Fluor 647-IgG2b; Molecular Probes, Eugene, OR) and Alexa Fluor 488-conjugated streptavidin (Alexa Fluor 488-SAv; Molecular Probes) for ß2M and Thy-1, respectively. For FACS analysis of wild-type adult testis cells, cell suspensions from the bottom fraction after Percoll centrifugation were stained with anti-
6-integrin (GoH3), anti-2M, and R-phycoerythrin (PE)-conjugated anti-Thy-1 (30H-12) antibodies. The staining pattern of mouse testis cells by two anti-Thy-1 antibodies (53-2.1 and 30H-12) showed no difference (data not shown). 6-Integrin was detected by Alexa Fluor 488-SAv after staining with biotin-conjugated rat anti-mouse IgG1/2a (G28-5) antibody. ß2M was detected by Alexa Fluor 647-IgG2b. For FACS analysis of pup and neonate testis cells, dissociated cells were stained with anti-2M, biotin-conjugated anti-Thy-1 (53-2.1), and PE-conjugated anti-v-integrin (RMV-7) antibodies, followed by Alexa Fluor 647-IgG2b and Alexa Fluor 488-SAv. Prior to FACS, 1 µg/ml propidium iodide (Sigma) was added to the cell suspensions to exclude dead cells. Cell sorting was performed on a FACStar Plus (BD BioSciences) equipped with Coherent Enterprise II laser (488 nm) and air-cooled helium neon laser (633 nm) and operated by Flow Cytometry and Cell Sorting Shared Resource at the University of Pennsylvania. Cells were sorted into 5-ml polypropylene tubes containing 4 ml of PBS supplemented with 10% FBS, 10 mM HEPES, 1 mM pyruvate, antibiotics, and 1 mg/ml glucose. An aliquot of stained cells in each experiment was not used for FACS (unsorted control). The sorted and unsorted control cells were centrifuged and resuspended in 3 ml of Ham nutrient mixture F-10 (F10) supplemented with 10% FBS, 50 µM 2-mercaptoethanol (Sigma), 10 mM HEPES, 2 mM glutamine (Invitrogen), and antibiotics. The tubes were gassed with 5% CO2 and stored overnight at 4°C [17].
Magnetic-Activated Cell Sorting
Magnetic microbeads conjugated to anti-Thy-1 antibody (30-H12; Miltenyi Biotec, Gladbach, Germany) were used for magnetic-activated cell sorting (MACS) to enrich Thy-1+ cells from testis cell suspensions. The procedure to obtain Thy-1+ cells was performed according to the manufacture's protocol with minor modification. Briefly, dissociated cells from cryptorchid adult and wild-type adult, pup, or neonate testes were fractionated by Percoll centrifugation as described earlier. The single-cell suspension (3–8 x 106 cells in 90 µl of PBS-S) from the bottom fraction of Percoll centrifugation was incubated with 10 µl of Thy-1 microbeads for 20 min at 4°C. After rinsing with PBS-S, Thy-1+ cells were selected by passing through an MS separation column (Miltenyi Biotec) that was placed in a magnetic field. After removal of the column from the magnetic field, the magnetically retained Thy-1+ cells were eluted.
Cell Culture
The basic culture system consisted of serum-free medium and mitotically inactivated STO cell feeders as described previously for hepatic progenitors with minor modification [19]. The serum-free medium for SSCs consisted of minimum essential medium-alpha (MEM; Invitrogen) or F10, to which was added 0.2% bovine serum albumin (ICN Biomedicals, Irvine, CA), 5 µg/ml insulin (Sigma), 10 µg/ml iron-saturated transferrin (Sigma), 7.6 µ=/L free fatty acids [27], 3 x 10–8 M H2SeO3 (Sigma), 50 µM 2-mercaptoethanol, 10 mM HEPES, 60 µM Putrescine (Sigma), 2 mM glutamine, and antibiotics. Free fatty acids comprised palmitic, palmitoleic, stearic, oleic, linoleic, and linolenic acids (all Sigma) in the respective millimolar proportions of 31.0:2.8:11.6:13.4:35.6:5.6 for 100 meq/L stock solution [27]. Serum-supplemented medium was prepared by adding heat-inactivated (56°C, 30 min) FBS into the serum-free medium at the concentrations indicated (0.1–10%; v/v). Growth factors used were mouse leukemia inhibitory factor (LIF; Chemicon International, Temecula, CA), human insulin-like growth factor-I (IGF-I; R&D Systems, Minneapolis, MN), human basic fibroblast growth factor (bFGF; BD Biosciences), mouse epidermal growth factor (EGF; BD Biosciences), mouse stem cell factor (SCF; R&D Systems), and human glial cell line-derived neurotrophic factor (GDNF; R&D Systems).
For cultures with ß2M– Thy-1+ cells isolated by FACS from ROSA cryptorchid adult testis cells, 2 x 104 ß2M– Thy-1+ cells were plated in two wells of a six-well plate (1 x 104 cells/9.6 cm2) with STO feeders or newborn (NB) testis cell feeders and in serum-free media or in serum-supplemented media. STO cells (STO SNL76/7 cells) were obtained from Dr. A. Bradley (Baylor College of Medicine, Houston, TX), and STO feeders were prepared as described [14]. Briefly, STO cells were maintained in Dulbecco modified Eagle medium (DMEM) supplemented with 7% FBS, 100 µM 2-mercaptoethanol, 2 mM glutamine, and antibiotics. For making feeders, STO cells were treated with 10 µg/ml of mitomicin C (Sigma) for 3–4 h and plated at a density of 5 x 105 cells per well of a six-well plate coated with 0.1% gelatin (Sigma) in the same medium. Before culture with donor testis cells, STO feeders were rinsed with Hank balanced salt solution twice. For NB testis feeders, NB testis cells (0.5– 1.5 dpp) from C57BL/6 x 129/SvCP F1 hybrid mice were prepared by enzymatic digestion. From 2 to 2.5 x 106 cells were placed in a 10-cm tissue culture dish and cultured in F10 supplemented with 10% FBS, 50 µM 2-mercaptoethanol, 10 mM HEPES, 2 mM glutamine, and antibiotics for 2 days. Mitomycin C-treated NB testis feeders were prepared as described for STO feeders. FACS-sorted ß2M– Thy-1+ cells were maintained on feeders for 8–10 days. Culture medium (2 ml/well) was changed every other day. Likewise, Thy-1 microbead-selected cells were cultured on STO feeders in the MEM-based medium in two wells of a six-well plate at a density of 1 x 104 cells per well (1 x 104 cells/9.6 cm2). When the culture period was extended to 2 wk, cells were harvested from two wells after 1 wk of culture and reseeded on fresh STO feeders in 2 wells of a six-well plate. All cultures were maintained at 37°C in a humidified 5% CO2 atmosphere.
Testis Cell Transplantation and Analysis of Recipient Testes
C57BL/6 x 129/SvCP F1 hybrid male mice were used as recipients. The mice were treated with busulfan (55 mg/kg, Sigma) at 5–7 wk of age to deplete endogenous germ cells in the testes [2, 3]. Approximately 10 µl of donor cell suspension was transplanted into the seminiferous tubules of each recipient testis through the efferent duct 4–6 wk after busulfan treatment [26]. The injection resulted in 70–80% filling of the tubules in each testis. Preparation of donor cells isolated by FACS for transplantation was described previously [17]. Thy-1+ cells obtained by MACS were resuspended in MEM-based serum-free medium and transplanted into the recipient testes. Cultured donor cell suspensions for transplantation were prepared in the same medium as used for each culture condition. The institutional Animal Care and Use Committee of the University of Pennsylvania approved all experimental procedures in accordance with the Guide for Care and Use of Laboratory Animals from the National Academy of Sciences.
Two months after transplantation, recipient testes were collected and stained with 5-bromo-4-chloro-3-indolyl ß-D-galactoside (X-gal) to visualize donor-derived spermatogenic colonies [28]. Individual blue stretches of spermatogenesis represent spermatogenic colonies generated from donor-derived single SSCs [28, 29]. Colony number was counted by using a dissection microscope. For the analysis of testes injected with testis cells isolated by FACS or MACS, colony number was normalized to 105 cells injected because the number of cells that could be recovered and injected varied. Colony number for injected cultured testis cells was normalized to 105 cells originally seeded in culture to compare with values of colony number generated by the cell population before culture.
Statistics
To determine if there were any significant differences in the number of colonies/105 cells seeded between the control group of mice and mice treated with different concentrations of growth factors, a series of mixed-effects analysis of variance (ANOVA) models were run to compare each of the treated groups to the control group of mice. Fixed effects for treatment group and testis side were included in the models, along with a random effect of mouse. This method of analysis assumes that the data are symmetrically distributed around the mean. Bonferroni adjustments were made to each of the pairwise comparisons versus the control group to control the Type I experimentwise error rate. All analyses were performed using SAS version 8.2 (SAS Institute, Cary, NC).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Surface expression of Thy-1 in wild-type adult testis cells as well as cryptorchid adult testis cells was analyzed to determine whether Thy-1 is a positive marker of SSCs throughout postnatal life of the mouse. Initially, cryptorchid testis cells were stained with ß2M, the light chain of MHC-I, and Thy-1 antibodies and analyzed by FACS (Fig. 1, A and B). Some ß2M+ cells produced autofluorescent signals in the control sample that was stained with ß2M antibody only (Fig. 1A, gate 3; G3); however, the Thy-1+ cell population was identified clearly in the ß2M– cells (Fig. 1B, G1). Three populations, ß2M– Thy-1+ (G1), ß2M– Thy-1– (G2), and ß2M+ cells (G3), were isolated by FACS and transplanted into the seminiferous tubules of busulfan-treated infertile recipient mice to determine the stem cell activity in each fraction. Two months after transplantation, spermatogenic colonies in the recipient testes stained with X-gal were counted (Fig. 1, C and D). Stem cell activity was detected almost exclusively in the ß2M– Thy-1+ cell fraction, while few spermatogenic colonies were generated from ß2M– Thy-1– cells and ß2M+ cells, indicating that ß2M– Thy-1+ cells contained most (
95%) of SSCs in the testis. The ß2M– Thy-1+ cells produced about 280 colonies of spermatogenesis per 105 cells transplanted (Fig. 1C), and stem cell concentration was enriched 15-fold (282.6/18.6) over unsorted cryptorchid testis cells.
|
Subsequently, Thy-1 expression of wild-type adult testis cells was analyzed. FACS analysis, however, did not show a distinct Thy-1+ subpopulation in the wild-type adult testis (data not shown), probably due to the very low number of Thy-1+ cells among the many differentiating germ cells, autofluorescent large cells, and cellular debris. Therefore, we used cell separation by centrifugation in Percoll to concentrate SSCs and to reduce cellular debris before antibody staining for FACS analysis. Approximately 11% of the original testis cells were sedimented in the bottom of the centrifugation tube after Percoll separation. All floating cells (top fraction), including the Percoll phase, were collected as well. The top fraction contained about 68% of total cells, indicating that about 79% of cells were recovered after Percoll centrifugation. To examine the stem cell activity of each fraction, a transplantation assay was performed. The number of spermatogenic colonies in the bottom fraction and the top fraction were 12 and 1.4 per 105 cells transplanted, respectively, while the original wild-type testis cells generated 2.6 colonies per 105 transplanted. This result indicated that the bottom fraction contained about a 5-fold enriched population of SSCs compared with the original wild-type adult testis cells. The bottom fraction was stained with antibodies against Thy-1, ß2M, and 6-integrin, a surface marker of SSCs [30]. ß2M+ cells, which were about 10% in the bottom fraction (data not shown), were gated out for analysis of Thy-1 and 6-integrin expression. In addition, side scatterhigh cells were removed by preliminary gating because of autofluorescence. In the ß2M– cells, FACS analysis identified Thy-1+ cells in the 6-integrin+ cell fraction (Fig. 2B). About 10% of the 6-integrin+ cells expressed Thy-1 (2.6%/25%; Fig. 2 legend). Three subpopulations, Thy-1+ 6-integrin+ (G1), Thy-1– 6-integrin+ (G2), and 6-integrin– (G3), were isolated by FACS followed by transplantation. After 2 mo, recipient testes were analyzed. The spermatogenic colony number in Thy-1+ 6-integrin+, Thy-1– 6-integrin+, and 6-integrin– cells, were 162, 10, and 0 per 105 cells transplanted, respectively (Fig. 2C). These results confirmed that Thy-1 antigen is expressed on SSCs in wild-type adult testis as well as cryptorchid adult testis cells. However, the concentration of SSCs in the ß2M– Thy-1+ cryptorchid testis cells was higher than that in the ß2M– Thy-1+ 6-integrin+ wild-type testis cells.
|
Thy-1 Is Expressed on Spermatogonial Stem Cells in Pup and Neonate Testes
We then investigated Thy-1 expression on SSCs in pup and neonate testes. Preliminary experiments showed that there were few ß2M+ cells in pup and neonatal testis cells. Therefore, the testis cell suspension was stained with anti-áv-integrin antibody as well as anti-2M and anti-Thy-1 antibodies, because most of ß2M+ cells expressed v-integrin in the cryptorchid testis cells (data not shown), and SSCs did not express v-integrin in the adults [17]. FACS analysis of the stained testis cells identified Thy-1+ cells in the v-integrin– population in both pup and neonate testis cells (Figs. 3 and 4, respectively). In pups, the staining pattern of v-integrin in the Thy-1+ cells (Fig. 3B, G1) shifted slightly compared with a control sample, which was stained with Thy-1 antibody alone (data not shown), indicating that Thy-1+ pup testis cells could be described as v-integrindim. Three fractions, Thy-1+ v-integrin–/dim (G1), Thy-1– v-integrin–/dim (G2), and v-integrin+ (G3), were isolated by FACS from pup or neonate testis cells and were transplanted into recipient testes. Donor-derived spermatogenic colonies were generated almost exclusively from Thy-1+ v-integrin–/dim cells in both pup and neonate testes (Figs. 3C and 4C). Pup and neonate Thy-1+ v-integrin–/dim cells produced 124 and 17 colonies per 105 cells transplanted, respectively. Thy-1+ v-integrin–/dim cell fractions did not contain ß2M+ cells (data not shown); therefore, the SSC is ß2M– in pup and neonate. In addition, all ß2M– Thy-1+ v-integrin–/dim cells were 6-integrin+ (data not shown). These results indicate clearly that ß2M– Thy-1+ 6-integrin+ subpopulations contain the majority of SSCs in wild-type adult, pup, and neonate testes as well as cryptorchid adult testis.
|
|
Stem Cell Enrichment by Thy-1 Antibody-Conjugated Magnetic Microbeads
FACS analysis of cryptorchid adult and wild-type adult, pup, and neonate testes showed that Thy-1+ cells other than the SSC-enriched subpopulations were very few in the testes (Figs. 1–4). Therefore, we anticipated that SSCs could be enriched by MACS with Thy-1 antibody-conjugated magnetic microbeads, which would greatly simplify stem cell enrichment at all ages. Unfractionated testis cells and Thy-1+ cells that were isolated by Thy-1 microbeads (MACS Thy-1+) from cryptorchid adult and wild-type adult, pup, and neonate testis cells were transplanted into recipient testes to determine the stem cell activity of MACS Thy-1+ cells. The number of donor-derived spermatogenic colonies in the recipient testes is shown in Figure 5. MACS Thy-1+ cells of cryptorchid adult and wild-type adult, pup, and neonate produced 192, 48, 70, and 22 colonies per 105 cells transplanted, respectively. Compared with unfractionated cell populations before MACS separation, the stem cells in the MACS Thy-1+ fraction were enriched 6-, 30-, 4-, and 5-fold for cryptorchid adult and wild-type adult, pup, and neonate, respectively. These results indicate that Thy-1 microbeads enriched SSCs efficiently and that the highest concentration of stem cells was achieved when cryptorchid adult testes were used.
|
Establishment of a Culture System to Support SSC Expansion
After methods were developed to obtain an enriched population of stem cells, experiments were initiated to evaluate in vitro conditions to maintain and/or expand stem cell number ex vivo. A major objective was to develop a culture system that consisted of defined conditions to facilitate studying the biology of the stem cell and identifying external factors for in vitro maintenance. In a previous study with primitive hepatic progenitor cells, a combination of serum-free hormonally defined medium and STO feeders supported clonal growth of purified progenitors, and it was possible to study fate determination of the progenitors in vitro by modifying the culture condition [19]. Therefore, the basic culture system that consisted of enriched stem cells, serum-free defined culture medium, and mitotically inactivated STO feeders was employed in this study. Because large numbers of contaminating nonstem cells are likely to influence SSC behavior in culture, an enriched stem cell population was believed to be critical to evaluate different culture conditions. Therefore, ß2M– Thy-1+ cells isolated by FACS from cryptorchid testes were chosen as a starting SSC population because the stem cell activity was highest among the Thy-1+ populations examined.
Initially, the effects of FBS, basal medium type, and feeder cells were evaluated on stem cell activity. Four different concentrations of FBS, 0%, 0.1%, 1%, and 10%, were compared. Previous studies had shown that MEM was better than DMEM for maintenance of SSCs [10]; therefore, two basal media, MEM and F10, were examined. Originally, F10 was developed for clonal culture under serum-free condition [31]. In addition, it has been reported that only F10, but not other basal media tested, was able to maintain immature stages of mesenchymal cells [32]. STO cells, which are able to maintain various types of stem and progenitor cells [9, 14, 19], as well as NB testis cells were used as feeder cells. NB testis feeder cells may produce critical factors for expansion of SSCs because the number of SSCs increases dramatically after birth [24].
In total, 16 conditions (four serum concentrations x two basal media x two types of feeders) were compared. In each condition, 2 x 104 ß2M– Thy-1+ cells, which were expected to generate about 280 spermatogenic colonies per 105 cells transplanted (Fig. 1C), were cultured in two wells of a six-well plate (1000 cells/cm2), in which feeder cells were seeded. After 8–10 days of culture, all cells were harvested from the two wells and transplanted into recipient testes to determine whether SSCs increased or decreased during the culture periods by comparing the number of spermatogenic colonies generated in recipient testes to the 280 value in Figure 1C. In the absence of FBS, between 42% (MEM/NB testis) and 85% (F10/STO) of SSCs were maintained during the 8–10-day culture period (Fig. 6). However, in the presence of serum, the SSC activity markedly decreased in combination with NB testis feeders. In the presence of 10% serum, there were no colonies generated in either the MEM/NB testis or F10/NB testis combination. Interestingly, SSCs expanded about 1.8-fold (500/ 283) in the presence of 10% FBS when MEM and STO feeders were used, while the number of SSCs in F10/STO/ 10% FBS decreased about 0.4-fold (111/283). These results indicated that combinations of different basal medium, FBS concentration, and types of feeder cells had a dramatic effect on self-renewal of SSCs in vitro.
|
Identification of Growth Factor Effect on SSCs In Vitro
Although F10 maintained SSCs in serum-free medium, SSC expansion was observed only in the MEM/STO/10% FBS condition. This result suggested that additional soluble factors (i.e., from FBS) could support self-renewal of SSCs in the MEM/STO serum-free condition. Therefore, we used the MEM/STO combination for further experiments to develop a serum-free defined culture condition to support expansion of SSCs. Because the protocol for obtaining subpopulations of SSCs is more rapid and flexible using MACS than FACS, we confirmed the results of the culture experiments with FACS-enriched SSCs (Fig. 6) by using MACS Thy-1+ cells in MEM/STO/serum-free and 10% FBS culture medium (Fig. 7). Freshly isolated MACS Thy-1+ cells, 1-wk cultured cells, and 2-wk cultured cells were transplanted into recipient testes for subsequent analysis of colony formation. In 10% FBS, a 2.5-fold (478/191) expansion of SSCs was observed following 1-wk culture, but no change occurred from the first to the second week. In the serum-free condition, SSCs expand slightly (1.3-fold; 261/191) in 1 wk, although the number of SSCs decreased to 0.7-fold (169/191) in the second week of the culture. These results indicate that maintenance and proliferation of SSCs in the MACS Thy-1+ cells was similar to that in FACS-enriched ß2M– Thy-1+ cells; therefore, MACS was used in subsequent experiments to provide SSC populations.
|
On the basis of the above experiments, in which SSCs were maintained without loss of activity for the first week in the serum-free culture condition, a system was established to study SSC maintenance and proliferation in vitro. This system consisted of STO cell feeders and MEM, and the SSC-enriched population was Thy-1+ cells from MACS. Using this system, we examined the effect of six growth factors at 2–3 concentrations on SSC proliferation during a 7-day period in vitro (Fig. 8). Two factors, LIF and bFGF, did not have a significant effect on SSC activity during the 7-day period at the lowest concentrations. However, stem cell activity was significantly decreased in 10 000 U/ml (P < 0.009) LIF and in 10 ng/ml (P < 0.01) or 100 ng/ml (P < 0.0003) bFGF. The other four factors (EGF, IGF-I, SCF, and GDNF) did not demonstrate a significant effect on SSC activity during the culture period at the concentrations examined. These results clearly suggest that this assay system, using defined serum-free culture conditions, provides a precise assessment of specific biological actions of individual growth factors on SSCs, and the approach can be readily extrapolated to multifactor analyses.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Because previous studies suggested that testicular somatic cells and differentiating germ cells were detrimental to maintenance of SSCs in vitro [10], an essential initial objective was to define characteristics of the stem cell that would facilitate its identification and purification. We reasoned that the characteristics would comprise one or more surface antigens and that the antigens might be conserved throughout life, as has been found for hematopoietic stem cells [33]. In earlier studies, it was demonstrated that SSCs in cryptorchid testes were MHC-I– Thy-1+, almost all the stem cells were contained in this surface antigenic profile, and about 1 in 15 cells of this population behaved as a fully functional stem cell when transplanted to recipient testes [17]. However, cryptorchid testes are an enriched source of SSCs, and it could be argued that these testes represent a special or unusual physiological environment. Therefore, in initial experiments, we used FACS analysis of testis cells and a transplantation assay to demonstrate that stem cell activity is found almost exclusively in ß2M– Thy-1+ cells of neonatal, pup, and adult testes. Because ß2M is a light chain of MHC-I, the surface phenotype of SSCs is MHC-I– Thy-1+ throughout postnatal life of the mouse. Moreover, MHC-I– Thy-1+ cells of neonate, pup, and adult testis share the surface phenotype of 6-integrin+ and v-integrin–/dim [17]. Although the surface phenotype of other testis cells varies during development, it was possible to use specific purification techniques to prove that the SSCs have a distinctive surface phenotype that allows enrichment for culture and other studies. Furthermore, the continuity of surface phenotype suggests that other unique biochemical molecules of SSCs can be identified beginning with these MHC-I– Thy-1+ 6-integrin+ v-integrin–/dim cells.
Stem cell activity in the MHC-I– Thy-1+ cell population changed during development; it was low (17 colonies/105 cells transplanted) in the neonate but increased dramatically (124 colonies/105 cells transplanted) in the pup. Others have found a similar increase in stem cell activity during this developmental period [3, 24, 34, 35]. Perhaps some of the differences found result from a gradual differentiation of gonocytes into true stem cells that can self-renew and produce spermatogenesis. The only germ cells in the neonate testis are gonocytes, located in the center of the seminiferous tubule [25, 36]. By 6 dpp, most gonocytes have migrated to the basement membrane and become stem cells or type A spermatogonia [37, 38]. In this study, 0.5–1.5-dpp neonates and 4.5–5.5-dpp pups were used. Cell number collected from the testes increases about 2-fold during this period (5.1 x 105 ± 0.5 cells/testis at 0.5–1.5 dpp and 9.6 x 105 ± 1.5 cells/testis at 4.5–5.5 dpp; mean ± SEM, n = 4 or 5; FACS and MACS data not shown). Therefore, the dramatic increase of stem cell activity in the MHC-I– Thy-1+ v-integrin–/dim cells represents about four doublings in 4–5 days. A similar doubling rate can be obtained from values for unfractionated cells in Figure 5 (pup = 14.4 vs. neonate = 1.7). Therefore, the stem cells are dividing approximately every day, which is about four times more rapid than the 4–5-day doubling time estimated in adult testis [37, 39]. Subsequently, little change in stem cell concentration occurs in the cell population from pup to wild-type adult. Such a dynamic alteration of stem cell activity in this phenotypically identical cell population provides a unique opportunity to investigate SSC development.
The stem cell activity of MHC-I– Thy-1+ 6-integrin+ cells of wild-type adult testis (162 colonies/105 cells transplanted) was only slightly higher than in the 5-day pup, which could indicate that the relationship between stem cell and number of associated primitive spermatogonia with similar surface phenotype is established soon after birth. Interestingly, MHC-I– Thy-1+ cells of the cryptorchid testis contained a much higher stem cell activity (283 colonies/ 105 cells transplanted). The difference in the microenvironment of the cryptorchid and wild-type testis must result in a different ratio of stem cell to primitive spermatogonia with MHC-I– Thy-1+ surface phenotype. A negative feedback system to inhibit the proliferation of undifferentiated spermatogonia by differentiating spermatogonia has been reported in wild-type testis [40]. However, experimental cryptorchid testes are devoid of all differentiated spermatogonia; only undifferentiated spermatogonia remain in the testis [41]. Thus, there is no such feedback system present in the cryptorchid testis, and under these conditions, turnover of undifferentiated spermatogonia may be more active [42]. Moreover, the loss of differentiating germ cells profoundly influences the balance between the germ cell and endocrine compartments of the testis, and absence of differentiating germ cells interferes with the reestablishment of spermatogenesis in irradiated rat testes even when stem cells remain [43]. These physiological differences between cryptorchid and wild-type adult testes must influence the ratio of stem cells to nonstem MHC-I– Thy-1+ cells and result in the difference we observed in stem cell activity. Although stem cell activity and the microenvironment surrounding SSCs change dramatically after birth and are quite different in adult wild-type and cryptorchid testis, Thy-1 is expressed on SSCs constitutively in each of these conditions. Moreover, Thy-1 has recently been identified as a marker on rat SSCs (unpublished data) and may represent a characteristic surface phenotype of SSCs for all species. The biological function of Thy-1 on SSCs and primitive germ cells is unknown, and studies on hematopoietic and neuronal systems have also failed to elucidate the exact role of this surface antigen.
Following identification of the surface antigens for SSCs by FACS, the phenotype was exploited to obtain enriched populations of SSCs using microbeads. This later technique is simple, less expensive, and can produce more cells in less time, which facilitates assessment of various culture conditions. Although there are several known antigenic markers of SSCs, including 6-integrin, CD24, or ß1-integrin [17, 18, 30], Thy-1 expression is relatively unique for the SSC population. For example, in the pup testis, a number of 6-integrin+ v-integrin+ or CD24+ v-integrin+ cells, which are not SSCs, were identified (data not shown). However, selective expression of Thy-1 on SSCs in the testis allowed us to use Thy-1 microbeads for enrichment of SSCs by MACS. In fact, significant enrichment of SSCs from cryptorchid adult and wild-type adult, pup, and neonate testis was achieved (Fig. 5). In both FACS and MACS systems, the Thy-1+ fraction of cryptorchid adult testes contained the highest stem cell activity. Therefore, SSC populations of cryptorchid testes enriched by FACS and MACS were used for all culture experiments, which minimized the effect of nonstem cells on the culture environment.
While it is desirable to minimize the effect of nonstem cells of the testis on culture conditions, neonatal testis cells are different from those from adult wild-type or cryptorchid males. The rapid increase in stem cell concentration in the MHC-I– Thy-1+ v-integrin–/dim cell population during the first 5 days following birth suggests that testis somatic cells are producing growth factors to stimulate the stem cell population. Therefore, a comparison of NB testis to STO feeders on SSC maintenance in vitro was undertaken. When SSCs enriched by FACS were cultured with NB testis feeders in the presence of serum at a concentration of 1% or 10%, the number of stem cells decreased dramatically during 8–10 days of culture, and no stem cell activity was detected in 10% FBS using NB testis feeders (Fig. 6). A possible explanation is that stimulation of NB testis feeders by FBS resulted in production of factors that induce differentiation or cell death in SSCs, an effect not anticipated from the increase in stem cell activity found in pups. These results suggest that reduction of the testicular somatic cell population improves culture conditions for ex vivo expansion of SSCs in the presence of serum. However, a combination of serum-free medium and NB testis feeder cells did not show a detrimental effect on SSC maintenance, suggesting that serum-free conditions dramatically alter the NB testis cell effect on SSCs.
The combination of serum with NB testis feeder was detrimental to SSC maintenance in vitro, and serum in combination with F10 and STO feeders also appeared to have a negative effect. However, serum added to MEM with STO feeders enhanced the maintenance and proliferation of SSCs in vitro. These results clearly indicate that serum is a complicating factor in developing culture conditions for SSCs and in understanding the environmental and growth factor requirement for SSC proliferation. Nonetheless, the very positive effect of 10% FBS in MEM using STO feeders suggested that FBS contained factors beneficial for SSC proliferation and that MEM and STO feeders could serve as a culture condition in which to test the effect of individual growth factors when added to supplemented serum-free MEM.
When we used the MEM serum-free medium and STO feeders, the stem cell number did not decrease appreciably during the 7-day culture period (Fig. 7). Therefore, the culture system allowed us to examine the effect of individual growth factors on replication of SSCs in vitro. Six growth factors (IGF-I, LIF, bFGF, EGF, SCF, and GDNF) were chosen to evaluate the system for studying SSC proliferation and biology in vitro. Some of the growth factors were used individually for in vitro maintenance of SSC in serum-supplemented conditions [10, 44]. However, there is no study in which the effects of single growth factors on SSCs were investigated in a serum-free condition. Three growth factors, LIF, bFGF, and IGF-I, appeared to have a more negative than positive effect on SSC maintenance in vitro at concentrations often employed for other cells in culture. Particularly surprising was the absence of a supporting role for LIF, which can replace the STO feeder effect when maintaining ES cells in vitro [45, 46]. The bFGF also showed an inhibitory effect at 10 ng/ml that is used to support primordial germ cells in vitro [47]. Another study, however, suggested that a peak response of bFGF for primordial germ cell proliferation is around 1 ng/ml [48], which showed no negative effect on SSCs. IGF-I did not improve SSC maintenance at any concentration examined. The influence of EGF is equivocal. EGF appeared neutral up to a level of 1–10 ng/ml but may be inhibitory at 100 ng/ml. The two growth factors that appeared to have a beneficial effect on SSC maintenance were SCF and GDNF. Addition of SCF seemed to increase slightly the number of SSCs; however, there is no detectable expression of the receptor, c-kit, on the cell surface of SSCs by FACS analysis [17], and another study has not shown an effect of SCF on long-term proliferation of stem cells [11]. In the presence of all concentrations of GDNF, the number of stem cells was increased approximately 1.5-fold, indicating that proliferation of SSCs may have occurred during the 1-wk culture period. In vivo studies with transgenic animals [49] or transfection by electroporation [50] have strongly suggested that GDNF is involved in SSC proliferation. Furthermore, previous studies using GDNF at 100 ng/ml in serum-containing medium also showed a significant increase in SSC activity after 1 wk [10]. Likewise, GDNF forms part of a growth factor cocktail added to the serum-supplemented condition that supports SSC proliferation from neonatal ICR or C57/BL/6 x DBA/2 mouse testes [11], although LIF, EGF, and bFGF, which are the remaining growth factors in the cocktail, had no positive effect in the culture studies described here. In another study, a cocktail of seven growth factors, not including GDNF, in serum-supplemented DMEM supported SSC survival during culture periods up to 3 mo [44]. Thus, growth factor supplementation is likely beneficial for SSC maintenance and proliferation, and the effect of GDNF on SSCs in our previous study, in which serum was used, and in this study in a well-defined serum-free medium strongly suggests that GDNF should be a component of the basal medium for SSC. The next step will be to use media supplemented with GDNF and examine the effect of growth factors, individually or in combinations, as well as other environmental modifications to determine the requirements for long-term SSC in vitro proliferation. Identification of crucial factors for replication of SSCs in a defined condition will establish the foundation for understanding the regulatory mechanisms that determine the critical cell fate determination of self-renewal or differentiation.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: R.L. Brinster, School of Veterinary Medicine, University of Pennsylvania, 3850 Baltimore Ave., Philadelphia, PA 19104. FAX: 215 898 0667; cpope{at}vet.upenn.edu
Received: 3 March 2004.
First decision: 28 March 2004.
Accepted: 19 April 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  Y.-H. Huang, C.-C. Chin, H.-N. Ho, C.-K. Chou, C.-N. Shen, H.-C. Kuo, T.-J. Wu, Y.-C. Wu, Y.-C. Hung, C.-C. Chang, et al. Pluripotency of mouse spermatogonial stem cells maintained by IGF-1- dependent pathway FASEB J, July 1, 2009; 23(7): 2076 - 2087. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Golestaneh, E. Beauchamp, S. Fallen, M. Kokkinaki, A. Uren, and M. Dym Wnt signaling promotes proliferation and stemness regulation of spermatogonial stem/progenitor cells Reproduction, July 1, 2009; 138(1): 151 - 162. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Schmidt, M. R. Avarbock, J. W. Tobias, and R. L. Brinster Identification of Glial Cell Line-Derived Neurotrophic Factor-Regulated Genes Important for Spermatogonial Stem Cell Self-Renewal in the Rat Biol Reprod, July 1, 2009; 81(1): 56 - 66. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Oatley, M. J. Oatley, M. R. Avarbock, J. W. Tobias, and R. L. Brinster Colony stimulating factor 1 is an extrinsic stimulator of mouse spermatogonial stem cell self-renewal Development, April 1, 2009; 136(7): 1191 - 1199. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. M Aponte, T. Soda, K. J Teerds, S C. Mizrak, H. J G van de Kant, and D. G de Rooij Propagation of bovine spermatogonial stem cells in vitro Reproduction, November 1, 2008; 136(5): 543 - 557. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Kim, M. Seandel, I. Falciatori, D. Wen, and S. Rafii CD34+ Testicular Stromal Cells Support Long-Term Expansion of Embryonic and Adult Stem and Progenitor Cells Stem Cells, October 1, 2008; 26(10): 2516 - 2522. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Goel, M. Fujihara, N. Minami, M. Yamada, and H. Imai Expression of NANOG, but not POU5F1, points to the stem cell potential of primitive germ cells in neonatal pig testis Reproduction, June 1, 2008; 135(6): 785 - 795. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-B. Stukenborg, J. Wistuba, C. M. Luetjens, M. A. Elhija, M. Huleihel, E. Lunenfeld, J. Gromoll, E. Nieschlag, and S. Schlatt Coculture of Spermatogonia With Somatic Cells in a Novel Three-Dimensional Soft-Agar-Culture-System J Androl, May 1, 2008; 29(3): 312 - 329. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. R Yeh, X. Zhang, and M. C Nagano Establishment of a Short-Term In Vitro Assay for Mouse Spermatogonial Stem Cells Biol Reprod, November 1, 2007; 77(5): 897 - 904. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Goel, M. Sugimoto, N. Minami, M. Yamada, S. Kume, and H. Imai Identification, Isolation, and In Vitro Culture of Porcine Gonocytes Biol Reprod, July 1, 2007; 77(1): 127 - 137. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. L. Brinster Male Germline Stem Cells: From Mice to Men Science, April 20, 2007; 316(5823): 404 - 405. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. K. Hamra, K. M. Chapman, D. Nguyen, and D. L. Garbers Identification of Neuregulin as a Factor Required for Formation of Aligned Spermatogonia J. Biol. Chem., January 5, 2007; 282(1): 721 - 730. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Fouchecourt, M. Godet, O. Sabido, and P. Durand Glial cell-line-derived neurotropic factor and its receptors are expressed by germinal and somatic cells of the rat testis. J. Endocrinol., July 1, 2006; 190(1): 59 - 71. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Oatley, M. R. Avarbock, A. I. Telaranta, D. T. Fearon, and R. L. Brinster From the Cover: Identifying genes important for spermatogonial stem cell self-renewal and survival PNAS, June 20, 2006; 103(25): 9524 - 9529. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. R. Lee, K.-S. Kim, Y. H. Yang, H. S. Oh, S. H. Lee, T. G. Chung, J. H. Cho, H. J. Kim, T. K. Yoon, and K. Y. Cha Isolation of male germ stem cell-like cells from testicular tissue of non-obstructive azoospermic patients and differentiation into haploid male germ cells in vitro Hum. Reprod., February 1, 2006; 21(2): 471 - 476. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. K. Hamra, K. M. Chapman, D. M. Nguyen, A. A. Williams-Stephens, R. E. Hammer, and D. L. Garbers Self renewal, expansion, and transfection of rat spermatogonial stem cells in culture PNAS, November 29, 2005; 102(48): 17430 - 17435. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Buageaw, M. Sukhwani, A. Ben-Yehudah, J. Ehmcke, V. Y. Rawe, C. Pholpramool, K. E. Orwig, and S. Schlatt GDNF Family Receptor alpha1 Phenotype of Spermatogonial Stem Cells in Immature Mouse Testes Biol Reprod, November 1, 2005; 73(5): 1011 - 1016. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B.-Y. Ryu, H. Kubota, M. R. Avarbock, and R. L. Brinster Conservation of spermatogonial stem cell self-renewal signaling between mouse and rat PNAS, October 4, 2005; 102(40): 14302 - 14307. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Riou, H. Bastos, B. Lassalle, M. Coureuil, J. Testart, F. D. Boussin, I. Allemand, and P. Fouchet The Telomerase Activity of Adult Mouse Testis Resides in the Spermatogonial {alpha}6-Integrin-Positive Side Population Enriched in Germinal Stem Cells Endocrinology, September 1, 2005; 146(9): 3926 - 3932. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Kanatsu-Shinohara, H. Miki, K. Inoue, N. Ogonuki, S. Toyokuni, A. Ogura, and T. Shinohara Long-Term Culture of Mouse Male Germline Stem Cells Under Serum-or Feeder-Free Conditions Biol Reprod, April 1, 2005; 72(4): 985 - 991. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Kubota, M. R. Avarbock, and R. L. Brinster Growth factors essential for self-renewal and expansion of mouse spermatogonial stem cells PNAS, November 23, 2004; 101(47): 16489 - 16494. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
