| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Obstetrics and Gynecology and Division of Experimental Medicine, McGill University, Montreal, Québec, Canada H3A 1A1
ABSTRACT
Spermatogonial stem cells (SSCs) are responsible for life-long, daily production of male gametes and for the transmission of genetic information to the next generation. Unequivocal detection of SSCs has relied on spermatogonial transplantation, in which functional SSCs are analyzed qualitatively and quantitatively based on their regenerative capacity. However, this technique has some significant limitations. For example, it is a time-consuming procedure, as data acquisition requires at least 8 weeks after transplantation. It is also laborious, requiring microinjection of target cells into the seminiferous tubules of individual testes. Donor-recipient immunocompatibility for successful transplantation and large variations in data obtained represent further limitations of this technique. In the present study, we provide evidence that a recently developed SSC culture system can be employed as a reliable, short-term in vitro assay for SSCs. In this system, donor cells generate three-dimensional structures of aggregated germ cells (clusters) in vitro within 6 days. We show that each cluster originates from a single cell. Thus, by counting the clusters, cluster-forming cells can be quantified. We observed a strong linear correlation between the numbers of clusters and SSCs over extended culture periods. Therefore, cluster numbers faithfully reflect SSC numbers. These results indicate that by simply counting the number of clusters, functional SSCs can be readily detected within 1 week in a semi-quantitative manner. The faithfulness of this in vitro assay to the transplantation assay was further confirmed under two experimental situations. This in vitro cluster formation assay provides a reliable short-term technique to detect SSCs.
growth factors, spermatogenesis, stem cells, testis
Spermatogonial stem cells (SSCs) are the foundation of spermatogenesis, a process that produces numerous spermatozoa throughout life after puberty [1]. The presence of SSCs allows regeneration of spermatogenesis and restoration of male fertility following various testicular insults, such as cancer chemotherapy. Compared to other stem cell types, SSCs are unique, since they are dispensable for the life of an individual but critical for the continuation and evolution of the species.
Since stem cells are defined by their functions of self-renewal and differentiation, thereby regenerating and maintaining a complete adult tissue [2], the detection of SSCs has relied on an in vivo functional assay, spermatogonial transplantation [3]. In this technique, a single cell suspension prepared from donor testes is injected into the seminiferous tubules of infertile recipient testes, leading to donor-derived spermatogenesis reconstitution. Since injected nonstem germ cells lack the potential for continuous self-renewal, these cells are eliminated through differentiation or death over time. Thus, spermatogonial transplantation selectively allows SSCs to re-establish and maintain long-term, donor-derived spermatogenesis in recipient testes [3, 4], and represents an unequivocal assay of functional SSCs.
Spermatogonial transplantation is also an effective method to quantify SSCs. Following transplantation, donor-derived spermatogenesis is regenerated in the form of distinctive colonies along the recipient seminiferous tubules [4]. Recent studies have convincingly shown that each colony arises from a single stem cell [5, 6]. Therefore, by simply counting the number of colonies, the number of functional SSCs can be determined. As such, spermatogonial transplantation is a powerful technique to analyze qualitatively and quantitatively functional SSCs that successfully regenerate spermatogenesis after transplantation.
Although the in vivo transplantation technique has proven crucial for the study of SSCs, it is not without limitations. The greatest problem associated with this technique is its time-consuming nature; the preparation of recipients requires 4 wk or more, and the transplantation results are obtained at least 8 wk after transplantation, at which time spermatogenic regeneration is complete [4]. In addition, the microinjection of donor cells into recipient seminiferous tubules is a laborious procedure, and transplantation experiments require careful planning to ensure immunocompatibility between donors and recipients. It is also of note that the data obtained using this in vivo assay show large variability, requiring a large number of recipient testes to obtain consistent results.
These problems could be circumvented by the establishment of a reliable, short-term in vitro assay. For example, stem cells of the central nervous system (CNS) can be quantified in vitro using the neurosphere assay [7, 8]. In this assay, neural stem cells (NSCs) actively divide in a chemically defined culture condition, which results in the formation of three-dimensional, floating, spheroid cell clusters, called neurospheres. Thus, NSCs are quantified by counting the number of neurospheres [9]. However, this type of in vitro assay system has not been developed for SSCs.
Recently, it has become possible to culture SSCs in the long-term, during which time these stem cells can be expanded extensively [10, 11]. In this system, donor testis cells are cultured on a feeder layer with glial cell line-derived neurotrophic factor (GDNF) and fibroblast growth factor 2 (FGF2) in a serum-free medium [10, 11]. Under these conditions, the donor cells form groups of aggregated germ cells on the feeder layer within 1 wk, which are termed clusters to distinguish them from colonies established in recipient testes after transplantation. The transplantation of clusters results in the regeneration of spermatogenesis, which confirms that the clusters contain SSCs [10, 11]. Furthermore, SSCs can be maintained virtually indefinitely in vitro by serially passaging the clusters [12].
Since germ cell clusters have a distinct three-dimensional structure, it may be possible to analyze quantitatively SSCs in vitro by counting clusters, a procedure analogous to the neurosphere assay for NSCs. To this end, two important criteria need to be addressed. Since quantification based on cluster numbers can only be valid if the clusters are clonal, the first criterion is whether each cluster is generated by a single "cluster-forming cell." The second criterion is whether, over an extended period of time, the number of clusters directly correlates with the number of spermatogenic colonies established by SSCs after transplantation. This issue is known to be one of the caveats of the neurosphere assay. Since a transplantation assay to detect the regenerative activity of NSCs is not possible in the CNS, the fidelity of the neurosphere assay relies on the long-term self-renewal activity of sphere cells, which is evaluated by the proliferation kinetics of the spheres over prolonged passaging periods [13, 14]. However, since committed progenitors can also form spheres, the proliferation kinetics of neurospheres can differ from those of NSCs [13, 15]. In contrast, the long-term proliferation kinetics of germ cell clusters can be directly correlated with that of SSCs by comparing cluster numbers with the SSC numbers determined by transplantation. Nonetheless, a correlation between these two parameters has not been demonstrated clearly.
In the present study, we provide evidence that the SSC culture system satisfies both of these criteria, and that functional SSCs can therefore be determined semiquantitatively by simply counting the number of clusters formed in vitro within 1 wk. Furthermore, we demonstrate that the results of SSC quantification are nearly identical between the in vitro assay and the transplantation assay under experimental conditions. Importantly, this in vitro assay overcomes most of the problems encountered with the transplantation assay. Thus, the development of this reliable, short-term, in vitro assay provides a powerful technique for SSC studies and should facilitate the progress of this research field.
Donor Mice and Cell Preparation
Donor mice were the F1 progeny of C57BL/6 (B6) females and B6.129S7-Gtrosa26Sor (designated ROSA26; The Jackson Laboratory) males. These mice were designated B6/ROSA. ROSA26 mice express the Escherichia coli lacZ transgene in virtually all cell types, including all types of postnatal male germ cells, allowing the discrimination of donor cells from recipient cells after transplantation in vivo and from feeder cells in culture [4, 16]. In some experiments, B6 mice were also used as donors. Donor cells were obtained from 6- to 8-day-old pup testes, and a single-cell suspension was prepared using a two-step enzymatic digestion protocol described previously [17, 18]. All animal procedures were approved by the Institutional Animal Care and Use Committee of McGill University.
A single suspension of testis cells was passed through a 40-µm mesh to remove undigested testis fragments and subsequently enriched for SSCs through immunomagnetic cell sorting, using a double selection protocol, as described previously [19] with modifications. Briefly, 6 x 106 testis cells were resuspended in 1 ml Dulbecco modified Eagle medium (DMEM) that was supplemented with 1% fetal bovine serum (FBS). These cells were first incubated with primary antibodies against two cell surface molecules that are expressed on non-SSCs in the testis (negative markers) [19, 20]: biotin-conjugated anti-
V-integrin antibody (clone RMV-7BD; Biosciences) and rat anti-ß2-microglobulin antibody (clone Ly-m11; BD Biosciences). The antibodies were incubated at a concentration of 5 µg/ml each at 4°C for 30 min with gentle agitation. Unbound primary antibodies were removed by washing twice with PBS that was supplemented with 1% FBS, and the cells were subsequently resuspended in 1 ml DMEM plus 1% FBS. Secondary antibodies conjugated to magnetic beads were then added as follows: M450 anti-rat antibody and M280 streptavidin (Dynal), at 20 µl each per 1 ml of cell suspension. The secondary antibodies were allowed to react for 30 min at 4°C with gentle agitation. The antibody-bound fraction was sorted using a Magnetic Particle Concentrator (MPC; Dynal) for 5 min. The antibody nonbound, V integrin-negative, ß2-microglobulin-negative fraction was subsequently resuspended in 1 ml DMEM plus 1% FBS and incubated for 30 min at 4°C with gentle agitation with biotinylated anti-Thy1.2 antibody (clone 52–2.1, 5 µg/ml; BD Biosciences), which targets a cell-surface molecule that has been demonstrated to be a positive marker for SSCs [21]. Unbound antibody was removed by two washes in PBS plus 1% FBS. Then, M280 streptavidin was added at 5 µl per 1 ml of cell suspension. Sorting was accomplished using the MPC for 1 min. The antibody-bound, Thy1.2-positive, V-integrin-negative, ß2-microglobulin-negative fraction constituted 
3% of the initial pup testis cell populations of both the B6 and B6/ROSA mice, and it was enriched 6-fold for pup SSCs versus unselected control testis cells in B6/ROSA mice. These SSC-enriched cells were used for all the experiments, unless otherwise indicated.
Cell Culture and Cluster Analysis
Testis cells sorted using immunomagnetic cell sorting were cultured on a feeder layer of STO (SIM mouse embryo-derived thioguanine and ouabain resistant) embryonic fibroblasts [22]. STO feeder cells were mitotically inactivated after mitomycin C treatment and seeded into a 24-well tissue culture plate at 5 x 104 cells/cm2 in 1 ml of medium. Sorted testis cells were placed on the STO feeder layer at 1.25 x 104 cells/cm2, unless otherwise indicated. The SSC culture medium was composed of Minimum Essential Medium (Invitrogen) with 0.2% BSA (Sigma), and supplements, as described previously [10]. The growth factors used were recombinant human GDNF (R&D Systems), recombinant rat GFRA1 (R&D Systems), and FGF2 (Invitrogen) at 40 ng/ml, 300 ng/ml, and 1 ng/ml, respectively [10]. The medium was changed every 3–4 days, and all cultures were digested with 0.25% trypsin-EDTA and subcultured at a 1:2 to 1:4 dilution every 6–7 days. All cultures were maintained at 37°C in a humidified atmosphere of 5% CO2.
For cluster quantification, cultures were fixed with 0.5% glutaraldehyde and reacted with 5-bromo-4-chloro-3-indolyl-ß-D-galactopyranoside (X-gal), to distinguish B6/ROSA-derived cells from the STO feeder layer and B6-derived cells. All clusters in a well were visually counted. Data were obtained from two to four experiments, and two wells of a 24-well culture plate were counted in each experiment. The data are expressed as mean ± SEM. Significance was determined using ANOVA followed by the Fisher test for Least Significant Difference or Student t-test.
To generate single-cell cultures, individual clusters from established clusters (more than five passages) were picked using a flame-polished Pasteur pipette connected via tubing to a mouth piece. Isolated clusters were gently digested to single cells with 0.05% trypsin-EDTA in PBS and subsequently placed at a clonal concentration in three 96-well culture plates for visual inspection. The contents of each well that contained only one cell were transferred onto STO feeder cells with growth factors in a single 96-well plate.
Recipient Mice and Transplantation Analysis
The recipients were 129/SvEv x B6 F1 hybrid mice. To deplete endogenous spermatogenesis, recipient mice were injected i.p. with busulfan (50 mg/kg body weight) at 4 wk of age [17, 18]. Approximately 4–6 wk after busulfan treatment, these mice were used as recipients for transplantation. The transplantation assay was used to quantify SSCs in donor testis cell preparations and in a population of cultured cells. For SSC quantification in donor testis cell preparations, SSC-enriched cells were resuspended at 1.0–1.2 x 106 cells/ml, and injected into recipient seminiferous tubules through the efferent duct [17, 18]. Cultured cells were collected using 0.25% trypsin-EDTA, resuspended at a concentration of 1.0–1.2 x 106 cells/ml, and transplanted as described above. For SSC quantification, recipient testes were harvested 2 mo after transplantation, and stained with X-gal to count visually the colonies of donor-derived spermatogenesis; the number of colonies indicates that of SSCs [5]. Data were obtained from two to four experiments, involving at least 10 recipient testes, and expressed as the mean ± SEM. Significance was determined using ANOVA followed by the Fisher test for Least Significant Difference or Student t-test.
Approximately 105 sorted pup testis cells were incubated in 1 ml of 20 mM Tris-HCl buffer (hypotonic solution, pH 7.0) for 5 min or 20 min at room temperature [23, 24]. The hypotonic treatment was terminated by adding 9 ml of Hanks Balanced Salt Solution (HBSS). The cells in the control group were incubated in HBSS, without hypotonic treatment. Each of these groups was then divided into two groups; one group was cultured with growth factors to determine cluster numbers, and the other group was transplanted into recipient mice to quantify SSCs. Data were collected from four experiments, with two to seven recipient testes and two wells of a 24-well culture plate per experiment. To quantify the survival of all the cells in the seminiferous tubules following hypotonic treatment, a single-cell suspension of unselected B6/ROSA pup tubules were treated with hypotonic solution as described above. Trypan blue exclusion and total cell recovery were used to determine the numbers of viable cells. Experiments were conducted four times and the data are expressed as the mean percentage ± SEM versus an untreated control. Significance was determined through ANOVA, followed by the Fisher test for Least Significant Difference.
Each Cluster Is Derived from a Single Cluster-Forming Cell
Testis cells were obtained from B6/ROSA transgenic mice (6–8 days old) and enriched for SSCs in the Thy1.2+ V-integrin– ß2-microglobulin– fraction using two-step immunomagnetic cell sorting (referred to hereinafter as ‘sorted cells') [19, 20]. Sorted cells were cultured on a STO feeder layer in the presence of GDNF, soluble GFRA1, and FGF2, which are required for the long-term maintenance and expansion of SSCs in vitro [10]; this growth factor cocktail is designated hereinafter as ‘growth factors.’ As reported previously [10], clusters of donor cells developed by the sixth day of culture (Fig. 1A); no clusters appeared in the absence of growth factors. These clusters were typically nonspherical, but rather appeared as globular three-dimensional structures attached to the feeder layer. We define a cluster as a group of at least six cells. Occasionally, we observed chains of cells, which are indicative of differentiating germ cells [25, 26], and they were not counted as clusters (Fig. 1B).
|
We first investigated whether or not a cluster arises from a single cell. To this end, sorted testis cells from B6/ROSA mice were first seeded in a limiting dilution, which ranged from 5 to 25 000 cells/cm2 (Fig. 1C). After 6 days of culture, the numbers of clusters were visually determined and correlated with the numbers of cells seeded. The results clearly show a positive linear relationship between these two parameters, which suggests that each cluster is derived from one ‘cluster-forming cell.’
To confirm that individual cells can form germ cell clusters, single-cell cultures were prepared from established cluster cells (more than five passages). In these experiments, individual clusters were isolated from feeder cells using gentle aspiration and digested to single cells. These cluster cells were first placed at a clonal concentration into 96-well culture plates, and wells that contained one cell were visually identified. The contents of single cell-containing wells were transferred onto STO feeders, in individual wells, with growth factors. After 1 wk, we observed a germ cell cluster (Fig. 1D), at a frequency of 1 per 40 cells seeded. These results demonstrate that a single cell can form a cluster.
To substantiate these results and detect the occurrence of polyclonal clusters, a cell chimerism experiment was performed. In this experiment, sorted cells derived from B6/ROSA mice were mixed at a 1:1 ratio with those from wild-type C57BL/6 (B6) mice at increasing total cell densities and cultured for 6 days under cluster-forming conditions (Fig. 2). Subsequently, the clusters that formed were incubated with X-gal, to distinguish B6/ROSA cells from B6 cells. Three cluster staining patterns emerged: clusters comprised of ß-galactosidase-positive cells alone (i.e., B6/ROSA origin; Fig. 2A, bottom cluster), ß-galactosidase-negative cells alone (B6 origin; Fig. 2A, top cluster indicated by arrow), and a mixture of the two cell types (Fig. 2B). The data show that most of the clusters produced were either of B6/ROSA origin or B6 origin, rather than a mixture of both at all the cell densities examined, further suggesting that the majority of clusters are exclusively of one genotype (Fig. 2C).
|
These results collectively indicate that each germ cell cluster generated in vitro arises from a single cluster-forming cell, and therefore, the number of clusters corresponds to the number of functional cluster-forming cells in vitro.
Cluster Numbers Correlate Linearly with Colony Numbers
We next compared the proliferative properties of cluster-forming cells to those of SSCs. After one week of culture, 1.63 ± 0.16 x 104 (n = 8) clusters were generated per 106 sorted cells initially placed in culture. When germ cell clusters were enzymatically dispersed to single cells on Day 6 or Day 7 in vitro and subcultured on a fresh STO feeder layer, secondary clusters formed. Following serial passaging, the number of clusters increased exponentially over a long period of time. As shown in Figure 3A, the increase in cluster number was 40 000-fold in 12 wk (11 passaged generations). Since one cluster arises from one cluster-forming cell, this result indicates that cluster-forming cells are long-term self-renewing cells and proliferate with a population doubling time of 5.5 days (84 days/log2 40 000) under the culture conditions used.
|
To determine the in vitro SSC activity, we transplanted cultured cells into recipient testes at various time-points during the 12-wk culture, which resulted in the production of colonies of regenerated spermatogenesis (Fig. 3); colony numbers were determined 2 mo after transplantation. We observed that the number of colonies increased exponentially from 2.91 ± 0.38 x 103 (n = 24) per 106 sorted cells initially placed in culture to 1.09 ± 0.32 x 108 (n = 10) after 12 wk of culture. Thus, the SSCs proliferated approximately 38 000-fold in vitro over the 12-wk time span, which is almost identical to the expansion kinetics of cluster-forming cells. Importantly, the proliferation kinetics of cluster-forming cells paralleled those of SSCs, as no difference was detected between the two regression coefficients (Fig. 3A). These results demonstrate that cluster numbers correlate linearly with colony numbers throughout the long-term multiple passaging period, and that the number of clusters detected within 1 wk of culture faithfully reflects the number of SSCs determined 2 mo after transplantation. These results suggest that SSCs can be measured using this cluster-forming assay by counting the number of clusters established in vitro. Although it is not clear at present whether every cluster implies a stem cell (see Discussion), this in vitro assay at least allows the semiquantitative detection of SSCs.
The Cluster-Forming Assay Results Faithfully Reflect the Transplantation Assay Results under Experimental Conditions
To verify further the fidelity of the cluster-forming assay, we applied it to two experimental conditions, and compared the results with those obtained using the transplantation assay. We cultured sorted cells without growth factors, and measured the stem cell activity remaining in the culture using both the transplantation assay and the cluster-forming assay (Fig. 4A); sorted cells were obtained from B6/ROSA pup testes. The transplantation assay was performed at Day 0 (no culture), Day 2, and Day 5 in vitro, and colonies were counted after 2 mo (Fig. 4A). At each time-point, these cells were also passaged onto fresh STO feeder cells and cultured with growth factors to generate clusters; subsequently, the number of clusters was determined 6 days after passaging. We expected that the SSC number would decline with time due to the lack of growth factors.
|
The two assays gave similar results (Fig. 4B). Compared to Day 0, the numbers of colonies and clusters were 1.2-fold and 1.1-fold higher on Day 2 and Day 5, respectively. However, no significant difference was detected with either assay. Therefore, these findings demonstrate that the results of the cluster-forming assay faithfully recapitulate those of the transplantation assay in a semiquantitative manner, and that despite the absence of growth factors, SSC activity or cluster-forming activity does not decline during the initial 5 days of culture of sorted cells.
In a second experiment, we used spermatogonial transplantation and the cluster-forming assay to measure the numbers of SSCs and cluster-forming cells following hypotonic treatment of sorted cells. Hypotonic lysis has been widely used to eliminate selectively germ cells from heterogeneous mixtures of the seminiferous epithelium. Accordingly, hypotonic treatment has been used to prepare Sertoli cell cultures [23, 24]. However, it has not been clarified whether SSCs are also eliminated by this treatment. Therefore, we used hypotonic treatment as a means of potentially reducing SSC and cluster-forming cell numbers. Following immunomagnetic cell sorting, B6/ROSA pup testis cells were treated with a hypotonic solution for 5 min or 20 min, or left untreated. Then, the cells were transplanted into recipient testes or cultured to induce cluster formation. Colony numbers were determined 2 mo after transplantation, while cluster numbers were assessed after 6 days of culture.
None of the assays detected any significant decrease in colony or cluster numbers after a 5-min exposure (Fig. 5, A and B). In contrast, a 50% decrease in colony and cluster numbers was detected after treatment for 20 min. These results further demonstrate that cluster numbers are representative of colony numbers. The 50% survival rate of SSCs after a 20-min exposure to hypotonic solution was somewhat surprising, since the solution represents such a harsh environment for these cells. Thus, we compared the survival kinetics of whole testes cells and SSCs in a hypotonic solution. As shown in Figure 5C, a significant proportion of the total testis cells died by 5 min, and only 33.4% were viable after 20 min. Since the SSCs did not decrease in number at 5 min and 50% of these cells were viable after 20 min, these results indicate that SSCs are more resistant to a hypotonic environment and survive these conditions better than non-SSC testis cells.
|
The results obtained under these two experimental conditions collectively demonstrate that the number of clusters faithfully reflects the number of spermatogenic colonies established after transplantation.
In the present study, we show that a recently established SSC culture system [10] allows reliable detection in a semiquantitative manner of functional SSCs within 1 wk. We established this in vitro SSC assay, the cluster-forming assay, based on two important stem cell characteristics: clonogenicity and long-term self-renewal ability, and confirmed the faithfulness of this assay to the definitive SSC assay, spermatogonial transplantation, using two experimental conditions. The data presented in Figures 1–3 demonstrate that most clusters originate from a single cluster-forming cell (clonogenicity), and that the cluster-forming activity is sustained for at least 12 wk over 11 passaging generations (long-term self-renewal). Significantly, this robust self-renewal ability of clusters clearly paralleled that of definitive SSCs, which was measured using the in vivo transplantation assay (Fig. 3). Therefore, these results indicate that we can now faithfully detect and quantify functional SSCs by simply counting the number of clusters formed in vitro within 1 wk. A similar methodology has been used in the neurosphere assay for NSCs [8, 9], except that this method is not directly associated with the regenerative ability of stem cells, since a transplantation assay is not feasible for the CNS. In contrast, we have demonstrated that the results of the short-term in vitro assay directly correlate with the results of spermatogonial transplantation, which is the unequivocal functional assay for SSCs.
The in vitro cluster-forming assay offers five clear advantages. First, as described above, the assay can be completed in a short period of time (6 days). Second, the procedure is simpler to perform than transplantation, allowing greater ease of use. Since spermatogonial transplantation is a laborious procedure, it has been technically demanding to compare five or more experimental conditions at once; the in vitro assay overcomes this limitation. Third, while immunocompatibility between donors and recipients is required for successful transplantation, it is not an issue in the in vitro cluster-forming assay. Fourth, in the latter context, the in vitro assay can be used for SSCs derived from wild-type mice. As shown in Figure 2, we found that when donor B6 and B6/ROSA cells were cocultured at a 1:1 ratio, the number of clusters with a single genotype was similar for either cell population, which indicates that the frequency of SSCs is equivalent in B6 and B6/ROSA cells. In the transplantation assay, quantification of SSCs derived from wild-type mice has been technically demanding, since the lack of genetic labels makes it difficult to distinguish donor-derived spermatogenesis from the endogenous spermatogenesis of a recipient. As such, the cluster-forming assay allows us to circumvent this problem. However, the application of the cluster-forming assay to various strains of mice may be necessary to evaluate the versatility of this in vitro assay. Finally, this in vitro assay appears to involve less variability of the data, as compared to the in vivo transplantation assay. Based on the data presented in Figures 4 and 5, we calculate that the coefficient of variation is 2-fold higher with the transplantation assay than with the cluster-forming assay; the former gives 50.1% of the coefficient, while the latter gives 25.2%. This lower level of variance should ensure greater consistency in data acquisition and justify the reduction of sample numbers.
Three disadvantages are associated with the cluster-forming assay. First, to induce cluster formation, donor testis cells need to be sorted for SSC enrichment. This is probably because without cell sorting, contaminating somatic cells interfere with the action of SSCs in vitro in the presence of growth factors. However, this disadvantage is restricted to primary cells that are prepared freshly from testes, and it is not a significant problem when established clusters are used to study biological properties of SSCs in vitro. Second, SSC detection using this in vitro technique is still retrospective, as it is with spermatogonial transplantation. Third, the cluster-forming assay does not entirely replace the transplantation assay for SSC detection. Since spermatogenesis cannot currently be reconstituted in vitro, the cluster-forming assay is not based directly on the regenerative activity of stem cells, and thus, is not an unequivocal stem cell assay on its own.
In this context, it should also be noted that counting cluster numbers does not present SSC numbers in an absolute manner. For example, it cannot be convincingly determined that there are 100 SSCs when one observes 100 clusters. This is because the quantification of SSCs in absolute numbers is currently not possible using a biological assay. In the cluster-forming assay, cluster chimerism at higher cell seeding densities can lead to underestimation of cluster numbers. In addition, cluster-forming cells may die or be physically lost when cells are replated to induce secondary clusters. Similarly, in the transplantation assay, the proportion of SSCs that survives transplantation and colonizes recipient testes (seeding efficiency) has not been determined unequivocally [27–29]. At present, therefore, caution is necessary when directly correlating cluster numbers and SSCs in absolute values. In this regard, our calculation based on the results shown in Figure 3 gives an estimate that the number of clusters is 70% of SSC number, assuming a 12% seeding efficiency after transplantation [27, 28]. Nevertheless, the cluster-forming assay clearly has significant advantages as described above, and this in vitro assay should provide a powerful assay tool for SSC research, particularly when used in combination with spermatogonial transplantation to substantiate the results of the cluster-forming assay.
We applied both in vitro and in vivo SSC assays to two experimental conditions, and in both cases, the assays gave almost identical results, further attesting to the fidelity of the cluster-forming assay. In one experiment, we quantified SSCs following culturing without growth factors, which are required for long-term SSC maintenance and expansion in vitro [10]. The numbers of SSCs remaining after 5 days in culture were similar, irrespective of the presence of growth factors. This was contrary to our expectation that the lack of growth factors would reduce the number of surviving SSCs. However, this experiment was conducted in the short-term and used cells freshly obtained from donor testes. Since in vitro SSC expansion is exponential and SSCs eventually disappear over time without growth factors in vitro [10] (Fig. 3), the effect of growth factors may be difficult to observe at an early stage of culture, and the results may simply imply that the culture condition is sufficient for SSC survival over the short-term, despite the lack of growth factors.
In the second experiment, we counted the number of SSCs that survived a cytotoxic hypotonic environment. The results of both the transplantation and cluster-forming assays showed that SSCs survived well after a 5-min hypotonic treatment, whereas 50% of them died after 20 min of treatment. Interestingly, although the SSC numbers did not change after a 5-min hypotonic treatment, the number of viable total testis cells declined significantly (Fig. 5). After a 20-min hypotonic treatment, the survival of total testis cells, including somatic cells, was significantly lower than the survival of SSCs measured with the cluster-forming assay (P 
0.03; Fig. 5, B and C). This difference was not detected with the transplantation assay (P 0.20; Fig. 5, A and C), probably due to the large variability of the data. Therefore, SSCs appear to be more robust and less susceptible to osmotic changes than non-SSC testis cells, and hypotonic treatment may be effective in partially enriching testis cells for SSCs.
In conclusion, this study provides evidence that the cluster-forming assay reliably and reproducibly detects functional SSCs within 1 wk in a semiquantitative manner. This efficient and versatile assay should facilitate progress in studies of SSCs in the future.
ACKNOWLEDGMENTS
The authors thank Kevin Ebata for his careful reading of the manuscript and suggestions.
FOOTNOTES
1Supported by CIHR (MOP-49444) and FRSQ (Bourse de Career). 
Correspondence: 2Makoto Nagano, Royal Victoria Hospital, F3.07, 687 Pine Avenue West Montreal, Québec, Canada H3A 1A1. FAX: 514 843 1662; e-mail: makoto.nagano{at}muhc.mcgill.ca
Received: 23 May 2007.
First decision: 11 June 2007.
Accepted: 8 August 2007.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  D. G. de Rooij and S. C. Mizrak Deriving multipotent stem cells from mouse spermatogonial stem cells: a new tool for developmental and clinical research Development, July 1, 2008; 135(13): 2207 - 2213. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
