| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
RESEARCH |
1 Department of Woman and Child Health, Astrid Lindgren Children’s Hospital, Pediatric Endocrinology Unit, Q2:08 and 2 Children’s Cancer Research Unit, Karolinska Insitutet and University Hospital, SE-171 76 Stockholm, Sweden, 3 Department of Medicine, Division of Hematology, Karolinska Institutet and University Hospital, Stockholm, Sweden, 4 Second Hospital of ShanDong University, Jinan, China and 5 Departments of Pediatrics and Physiology, University of Turku, Turku, Finland
Correspondence should be addressed to M Hou; Email: mi.hou{at}ki.se
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and MethodsAcknowledgementsReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and MethodsAcknowledgementsReferences |
|---|
Hematological spread of pediatric tumors results in a significant risk for intravascular and interstitial infiltration of testicular tissue by cancer cells. This is substantial in the case of acute lymphoblastic leukemia (ALL), where 20% of newly diagnosed patients exhibit microscopic infiltration of leukemic cells into their testes (Kim et al. 1981). Similar phenomena have been observed in association with experimental leukemia and lymphoma in rodents (Shaw et al. 1996, Jahnukainen et al. 2001).
An additional clinical concern is the relatively low number of stem cells present in the immature testis (Jahnukainen et al. 2006). Thus, in connection with the future development of clinical transplantation of germ cells, procedures designed to obtain testicular stem cell spermatogonia free from contaminating cancer cells and in good yield are urgently needed.
Flow cytometry (FACS) in association with selection on the basis of multiple parameters results in successful enrichment of murine SSC (Shinohara et al. 2000, Kubota et al. 2003, Ryu et al. 2004). The stem cells that are isolated in this fashion displayed low side-scatter (SSClow) and high forward-scatter (FSChigh) and expressed epithelial cell adhesion molecule (Ep-CAM), 
6-integrin/ ß1-integrin (CD49f), CD24, and thy-1 (CD90), but not v-integrin, c-kit, and major histocompatibility complex class I (MHC Cl I) on their surfaces (Shinohara et al. 1999, 2000, Giuili et al. 2002, Kubota et al. 2003, Ryu et al. 2004).
Promisingly, FACS has also recently been utilized to isolate germ cells from leukemic mice free from contamination by malignant cells (Fujita et al. 2005). In this case, cells of the myeloblastic leukemic (C1498) line, which cause leukemia in C57BL/6 mice, were eliminated from testicular cells removed from the leukemic donor on the basis of the expression of CD45 and MHC Cl I (Fujita et al. 2005). When testicular cells isolated in this manner were injected into the seminiferous tubules of healthy C57BL/6 mice, no transmission of leukemia was observed and the offspring originating from these cells were born healthy. More recently, the same surface markers were used successfully in seven of eight cases to remove cells of human leukemia and lymphoma lines from single-cell suspensions of human testicular cells with which the tumor cells had been mixed (Fujita et al. 2006).
In contrast, following positive selection for germ cells in human testicular samples employing the spermatogonial marker CD49f in combination with the removal of human B-cell acute lymphoblastic leukemic cells on the basis of their surface expression of HLA by FACS, the contamination by malignant cells was 0.58% in 10 of 11 samples sorted in this manner (Geens et al. 2007). This observation indicates that sorting by FACS is unlikely to be capable of completely depleting malignant cells from testicular samples.
To date, no systematic evaluation of the limitations associated with the removal of cancer cells from testicular samples on the basis of surface markers has been reported. In this context, it would be especially important to examine purification of testicular cell samples that have been infiltrated with leukemia cells by the natural spreading of this disease. Therefore, in the present investigation, Roser’s T-cell leukemia was employed to obtain infiltration of the testis by lymphoblasts and leukemic cells.
Roser’s leukemia is an acute T-lymphoblastic leukemia induced in piebald variegated (PVG) rats by irradiation and has been maintained by serial transmissions ever since its initial development (Dibley et al. 1975). This leukemia infiltrates immature testicular tissue (Jahnukainen et al. 1993) and the leukemic cells express the surface markers characteristic of immature T-lymphoblasts (Nestvold et al. 2004), properties that closely resemble those of human ALL. Following injection of more than 20 leukemic T-cells into a syngeneic rat, uninterrupted progression of leukemia occurs during a period of 14–19 days. The time of death is dependent on the number of cells injected, which allows this model to be used as a sensitive indicator of contamination of testicular samples by leukemic cells (Jahnukainen et al. 2001).
The goals of the present study were to identify surface markers expressed specifically by leukemic or germ cells in testicular samples from rats with Roser’s leukemia and, thereafter, to employ these markers to delete leukemic cells from and/or select for germ cells in these samples by FACS. Special emphasis was placed on the identification of cellular factors that may limit the efficacy of such surface marker-based purification. Moreover, the yield of testicular cells and the time required for sorting were examined in order to evaluate the clinical feasibility of this approach.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and MethodsAcknowledgementsReferences |
|---|
and Fig. 1. Among the surface markers examined, Ep-CAM was the only one present on spermatogonia, but not on the leukemic cells. Thus, germ cells located at the basal membrane of seminiferous tubules exhibited strong positive staining for Ep-CAM (Fig. 1A), whereas antibodies toward this protein did not bind to leukemic cells in cytospin slides (Fig. 1B) or to such cells that had infiltrated into the interstitial compartment (Fig. 1A). MAB EE2 recognizing a glycoprotein expressed on the surface of mouse spermatogonia (Falciatori et al. 2004) was not found to react with the rat epitopes in the present study.
|
|
). The corresponding values were 15.8 and 39% for normal testicular cells (Fig. 2G and I) and 15.1 and 34.1% for leukemic testicular cells (Fig. 2D and F) respectively. An increase of 6.2 and 3.7% in the numbers of cells staining strongly for CD4 and MHC Cl I was exhibited by leukemic testicular cells in comparison with control testicular cells. With respect to CD45, 93% of the lymphoblasts from the peripheral blood of leukemic donors expressed this marker (Fig. 2B, M2, green curve), although at a lower level than CD4 and MHC Cl I (Fig. 2A and C), whereas leukemic and normal testicular cells exhibited no staining whatever for this surface marker (Fig. 2E and H; Table 1
). Moreover, no expression of CD5 and CD43 by testicular cells could be detected, but leukemic lymphoblasts from the peripheral blood stained significantly for these markers (Table 1). In addition, leukemic cells in the interstitial tissue immuno-stained positively for CD4 and MHC Cl I, whereas there was no staining for these surface markers in the intratubular compartment (Fig. 1C and D).
|
Cell sorted by FACS: transmission of leukemia, phenotypic analysis, and yield
Protocol 1 (Ep-CAM–Alexa)
One of the rats that received testicular cells from a leukemic donor that were selected for on the basis of their expression of Ep-CAM–Alexa did not develop leukemia, whereas the other two died 24 and 33 days after transplantation of these cells (Fig. 3). As summarized in Table 2, no cells with the leukemic CD4+/MHC Cl I+phenotype were detected in these samples. Altogether, 5.1% of the cells expressed both leukemic CD4 and MHC Cl I and non-leukemic Ep-CAM cell surface markers, while 1.5% of the testicular cells from healthy control animals demonstrated this same mixed phenotype. Earlier observations concerning co-selection of both cells expressing Ep-CAM and lymphoid markers (e.g., T-lymphocytes, antigen-presenting cells, epithelial cells; Armstrong & Eck 2003) indicate that our present findings in this context reflect aggregates of normal testicular and leukemic cells. Indeed, this conclusion is supported by our observation in immunostained cytospin slides of a CD4-positive leukemic cell that had aggregated with an Ep-CAM-positive testicular cell (Fig. 1E).
|
|
).
|
). Moreover, no difference in the survival of recipients injected with two distinct subpopulations of germ cells as described earlier by Ryu et al. (2004), expressing low and high levels of Ep-CAM was observed (Table 3, experiment 1; Fig. 4, D1). Repetition of this same experiment employing a higher concentration of germ cells and more rapid sorting resulted in a prolongation of the sorting time and a reduction in the viability of the testicular cells obtained (Table 3, experiment 2). Moreover, no improvement in purification was achieved when Ep-CAMhigh-positive cells were sorted a second time using the more narrow gating area R2 (Fig. 4, D2; Table 3, experiment 3). After FACS sorting, the proportion of the cells that expressed both of the spermatogonial surface markers CD90 and Ep-CAM was 0.8% (Table 2).
|
). Neither the length of survival nor the course of leukemia development was altered when the gate for deletion by FACS was enlarged by reducing the threshold for identification of CD4- and MHC CI I-positive cells (Fig. 4B, gate R3, and Fig. 3). Cell preparation obtained in this way revealed 0.2% of the cells expressing the leukemic markers CD4 and MHC Cl I, and the same proportion expressed both of the spermatogonial markers CD90 and Ep-CAM (Table 2). A relatively high number of cells (2.1 and 1.3x106 respectively; Table 3, experiments 1 and 2) were retrieved after this 2-h sorting procedure.
Protocol 4 (Ep-CAM–Alexa+CD4–PE/MHC-1–PE)
When testicular cells prepared both by positive selection and by negative deletion were injected into PVG rats, both recipients survived for at least 4 months without developing leukemia (Fig. 3). As documented in Table 2, no cells expressing the leukemic markers CD4, MHC Cl I, or CD90 could be detected after sorting in this manner. In contrast, when the cells were used after only one sorting cycle, all five recipients died within 20.6±0.40 days after injection (data not shown). In total, 0.6% of the cells recovered with this procedure expressed both of the spermatogonial markers CD90 and Ep-CAM. Unfortunately, only 0.7x106 of the original 300x106 testicular cells subjected to this 4-h procedure were recovered (Table 3, experiment 2).
Immunocytochemical identification of testicular cells
Table 4 summarizes the results of the immunocytochemical analyses performed on the cell preparations sorted by FACS according to the various protocols. All the preparations selected on the basis of Ep-CAM expression included a high percentage of Oct-4-positive cells, with direct labeling with antibodies toward this surface marker yielding the purest fraction of such cells. Deletion of cells expressing CD4 and MHC Cl I produced a heterogeneous population of testicular cells, 5% of which expressed GATA-4. Examination under the light microscope revealed that many of these GATA-4-positive cells were elongating spermatids (data not shown).
|
). Upon incubation of 100 testicular cells per leukemic cell for 1, 2, and 3 h, the mean numbers of aggregates obtained per 100 cells examined were 1.1 (±0.31), 1.8 (±0.33), and 1.3 (±0.28) respectively. With the lower ratio of 10:1, the corresponding values were 2.7 (±0.83), 2.5 (±0.55), and 2.7 (±0.19). Thus, neither the length of the period of incubation nor the relative number of leukemic cells appeared to exert any pronounced influence on the formation of aggregates. When leukemic cells were labeled with antibodies toward Ep-CAM, CD4, and MHC Cl I, and the gate for sorting of cells exhibiting the SSClow and FSChigh characteristic for spermatogonia by FACS (Fig. 1G, R1), 
2.4% of the total cells recovered in this area were found to be positive for all three of these markers (Fig. 1H, upper right quadrant). We assume that the majority of these cells were aggregates of germ and leukemic cells.
Effects of enzymatic digestion on cell surface markers
With and without pre-treatment by enzymatic digestion, the fluorescence intensities of leukemic lymphoblasts incubated with antibodies directed against CD4, CD90, MHC Cl I, CD45, and CD5 were 198 vs 596, 2317 vs 2298, 174 vs 143, 12 vs 16, and 121 vs 133 respectively. Only the decrease in the fluorescence intensity observed with the CD4 antibody was statistically significant (P = 0.018).
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and MethodsAcknowledgementsReferences |
|---|
On the basis of promising findings with the mouse and cell lines, sorting by FACS has been proposed as a potential removal of tumor cells from testicular samples (Fujita et al. 2005, 2006). However, the efficacy of such selection is entirely dependent on the availability of specific surface markers for cancer and/or normal testicular cells. In the present investigation, poor specificity of the markers for leukemic and germ cells evaluated, aggregation of germ and leukemic cells, and heterogeneity of the leukemic cell population were found to seriously impair the efficiency of purification of testicular samples by FACS. Positive selection of germ cells or deletion of leukemic cells alone was insufficient to decontaminate testicular cell preparations, in agreement with previous observations (Geens et al. 2007). Only the combination of positive selection and deletion prevented transmission of leukemia in association with testicular cell transplantation in rats.
Here, detailed analysis of four known surface markers for spermatogonia revealed that poor specificity seriously limits their efficacy in separating germ cells from leukemic lymphoblasts. To date, only a few surface markers that can be used to identify the spermatogonial subpopulation with stem cell potency have been identified (Shinohara et al. 2000, Viglietto et al. 2000, Falciatori et al. 2004, Kanatsu-Shinohara et al. 2004). Of these markers, RET, 6/ß1 integrin, and CD90 were detected here on the surface of rat leukemic T-lymphoblasts as well.
These observations strongly suggest that leukemic lymphoblasts and spermatogonia share stem cell-like characteristics, making their separation highly difficult. Indeed, Ep-CAM was the only spermatogonial marker not detected on the surface of Roser’s rat leukemic lymphoblasts. Ep-CAM, a calcium-independent homophilic adhesion molecule expressed by most epithelia and carcinoma (Litvinov et al. 1997), has recently been utilized to isolate murine gonocytes and type A spermatogonia for culture (van der Wee et al. 2001, Moore et al. 2002).
Although considered to be a marker for SSC, our present immunohistochemical examination revealed staining for Ep-CAM in a continuous ring of cells on the basement membrane, without changes in intensity or differences between seminiferous tubules (Fig. 1A). This observation, which is consistent with an earlier report (Anderson et al. 1999; Fig. 2C), suggests that Ep-CAM is not a unique marker for SSC in the mature rat testis, but is probably expressed by germ cells in general. The patterns of expression of Ep-CAM in the immature and adult rat testis may differ significantly, since 90% of the germ cells from rat pup testis exhibiting SSClow and FSChigh and expressing Ep-CAM demonstrated stem cell potential in a functional assay involving germ cell transplantation. This latter fraction was also found to express Thy-1 (CD90), a marker for a variety of stem cells, including hematopoietic and SSC (Baum et al. 1992, Ryu et al. 2004).
In the present study, sorting based on a combination of FSChigh and SSClow together with expression of Ep-CAM (protocols 1, 2, and 4, Fig. 4) yielded cells which expressed Ep-CAM (data not shown) and also Oct-4 (Table 4; Hofmann et al. 2005). The cell population obtained with protocol 1 contained no cells that expressed the leukemic makers CD4 and MHC Cl I. However, a distinctive cell population expressing markers for both leukemic and germ cells was detected, and two of the three recipients injected with the cells from this protocol died from relapsing leukemia (Table 2).
This finding indicated the presence of leukemic cells that had aggregated with Ep-CAM-positive cells, an explanation that was confirmed by our observation of CD4-positive cell aggregates in immunostained cytospin slides. Rapid aggregation that was independent of the ratio of normal testicular to leukemic cells also occurred when these two types of cells (labeled with different fluorescence dyes) were mixed together (Fig. 1F). In this case, the cell aggregates formed displayed SSClow and FSChigh and were located in the gated FACS containing germ cells. It appears highly likely that adhesion of this sort represents a serious risk for contamination of all germ cells separated on the basis of surface markers by malignant cells. Deletion of leukemic cells on the basis of their surface markers is mandatory to eliminate such contamination.
On the other hand, removal of leukemic cells on the basis of their specific surface markers did reduce the contamination of testicular cell preparations, but to a lesser extent than that achieved by positive selection on the basis of Ep-CAM. None of the recipients of the former cell preparations survived. This result is very different from that described by Fujita et al. (2005) who used negative selection by FACS based on one leukemia specific together with one somatic cell marker (MHC Cl I) to obtain total decontamination of leukemic testicular samples.
One obvious reason for these different outcomes is the difference in the models employed for leukemia. The myeloblastic leukemic (C1498) cells studied by Fujita et al. (2005) uniformly express CD45 and MHC Cl I on their surface, whereas variations in leukemic cell immunophenotype drastically limited the efficacy of negative selection in our study. Roser’s rat leukemic T-cells were found to express a number of lymphoid markers, including CD3, CD4, CD8, CD45, and CD43, but there was a significant immunophenotypic variation, especially with respect to the markers expressed at low levels, CD8 and CD45 (Table 1). In the clinical situation as well, immunophenotypic variation and formation of subsets of leukemic cells occur regularly in connection with the development of human acute leukemias (Lay et al. 1971, Catovsky et al. 1974).
Moreover, we also demonstrate unequivocally here that the testicular micro-environment may contribute to immunophenotypic variation among testicular leukemic cells in this organ. The leukemic surface markers CD45, CD43 and CD5 could not be detected in testicular samples in which 4% of the cells were leukemic, despite the fact that these markers were expressed on lymphoblasts in the peripheral blood of the same animals. In addition, heterogeneous expression of surface markers on testicular leukemic cells was also detected in our previous study on magnetic cell sorting involving the same model system (Hou et al. 2007b). In that case, weakly CD4-positive leukemic lymphoblasts were recovered in the flow-through fraction, whereas strongly CD4-positive leukemic cells were deleted by the magnetic cell sorting. Either the leukemic lymphoblasts alter their expression of surface markers after infiltrating testicular tissue or only a specific leukemic subclone infiltrates the testis.
Our observation that enzymatic dispersion decreases the fluorescence intensity of labeled leukemic cells suggests that use of such a procedure to prepare testicular cells mayalso result in digestion of surface marker proteins. Here, the fluorescence intensity of leukemic cells labeled with antibodies toward CD4, but not with antibodies toward the weak markers CD5, CD45 or CD43, was reduced by enzymatic dispersion, indicating that our inability to detect these latter markers in our testicular cell samples was not due to sample preparation.
The tissue- and preparation-dependent expression of leukemic cell surface markers observed here raises serious questions concerning the relevance of certain previous purification studies (Fujita et al. 2006, Geens et al. 2007). Thus, FACS sorting of artificial mixtures of cancer cell lines and normal testicular cells may not reflect the clinical situation, since the malignant cells employed do not originate from testicular tissue and have not been exposed to enzymatic digestion. In addition, the present findings suggest that surface markers expressed on leukemic cells in peripheral blood will probably not be suitable for clinical FACS sorting of leukemic testicular samples.
Only by employing positive selection of germ cells in combination with deletion of leukemic cells did we succeed in isolating testicular cells free from leukemic cells. No testicular somatic cells were detected in such preparations and all rats injected with them survived. The results obtained by Geens in a previous comparable investigation were less promising. In that case, after one cycle of combined positive selection and deletion, 0.4 and 0.6% contamination by murine lymphoma and human acute B-lymphoblastic leukemia cells respectively was detected (Geens et al. 2007). However, these findings by Geens are strikingly similar to ours after only one cycle of purification by FACS, where all recipients of such cells developed leukemia. A second sorting cycle was required to abolish the contamination by leukemic cells, suggesting that repeated FACS cycles may enhance the efficacy of testicular cell sorting.
One clear disadvantage associated with the use of a second sorting cycle was the serious loss of cells. Thus, only 0.7 x 106 of the initial 300 x 106 cells sorted were recovered. This low yield of 0.23%, together with the lack of a specific marker for monitoring donor-derived spermatogenesis, prevented us from confirming the stem cell capacity of the isolated cells by germ cell transplantation. Nor could we evaluate the risk of leukemia relapse associated with injection of larger numbers of testicular cells. When 0.1 x 106 testicular cells purified in this manner were injected here, no leukemia was detected, but injection of a 10- to 100-fold greater number of cells would have resulted in the transmission of leukemia. One major limiting factor in connection with assessment of cell purification by FACS may be the relatively low ability of this approach to detect contaminating cancer cells, which is at best one cell among 104–105 cells sorted (Hu et al. 2005).
The cells expressing both CD90 and Ep-CAM, but no leukemic surface markers, detected here may be SSC (Ryu et al. 2004). The level of these cells in our testicular samples (0.5–1.1%) was well within the low level of SSC described previously in the rodent testis (Tegelenbosch & de Rooij 1993). The proportion of CD90- and Ep-CAM-positive cells was the highest in the unsorted testicular cell preparations, indicating that none of the sorting protocols applied here was capable of selection for these putative SSC. Only 3000–9000 CD90- and Ep-CAM-positive cells could be collected from 6 to 10 adult rat testes by the procedures employed here.
Morphological studies indicate that one testis of a 10-year-old prepubertal boy contains ~83 x 106 germ cells, with the corresponding value for a boy < 1 year of age being 13 x 106 germ cells (Muller & Skakkebaek 1983). It therefore seems unlikely that small testicular biopsies from young boys will provide a sufficient number of cells for combined positive selection and deletion by FACS.
Our present use of experimental lymphoblastic leukemia as a model provides the first insights concerning the limitations associated with surface marker-based selection of testicular cells for clinical transplantation in case of ALL. Patients with this disease present the largest pediatric group that could benefit from novel strategies designed to assure future fertility by germ cell transplantation. Here, the poor specificity of spermatogonial surface markers, aggregation of germ and leukemic cells, and significant variations in the expression of specific leukemic surface markers were found to seriously limit the efficacy of both positive selection of normal testicular cells and deletion of leukemic testicular cells by FACS. Future development of functional deletion of malignancy, i.e., by culturing in vitro or xenotransplantation into an intermediate host (Hou et al. 2007a), may provide new tools to guarantee the absence of tumor cells from clinical samples of testicular cells. The present investigation helps provide a platform for planning such future studies.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and MethodsAcknowledgementsReferences |
|---|
The lymphoblasts utilized for examination of surface markers, cell aggregation, and immunophenotypic variation were isolated from the enlarged cervical lymph nodes (Jahnukainen et al. 2001) and on Ficoll from peripheral blood of terminally ill leukemic rats. Cells obtained from the peripheral blood or testis were labeled shortly after isolation either with antibodies for FACS analysis (see below) or with fluorescence dyes for evaluation of cell adhesion or else used to make cytospin slides for immunocytochemical analysis (see below). For the preparation of frozen sections, the testes removed from three healthy and three leukemic rats at 40 days of age were immersed in a solution of OCT (HistoLab, Västra Frölunda, Sweden), frozen in dry ice, and stored at – 70 ° C. The frozen testes were cut into 12 µ m thick sections with a microtome, mounted on Superfrost slides, and maintained thereafter at – 70 ° C until use.
Screening for surface markers that could potentially allow efficient sorting by FACS
Lymphoblasts obtained from leukemic donors or testicular cells from control animals were first incubated with mouse anti-rat CD5, CD43 and CD45 antibodies (kindly supplied by Dr Olle Lidman at Karolinska Institute, Stockholm, Sweden) at a dilution of 1:100 (using 10 µ l/106 cells) for 30 min; then washed twice with PBS containing 0.5% fetal calf serum (FCS; designated hereafter as PBS/FCS); and thereafter incubated with PE-goat anti-mouse F(ab')2 (diluted 1:20, 10 µ l/106 cells; DakoCytomation, Glostrup, Denmark) for 30 min. Following two additional washes with PBS/FCS, these cells were subjected to FACS analysis. For direct labeling, the cell preparations were incubated instead with mouse anti-rat CD3– PE (5 µ l/106 cells), CD4– PE (5 µ l/106 cells), CD90– Percp (10 µ l/106 cells), MHC Cl I– PE (5 µ l/106 cells; all purchased from BD Biosciences, San Jose, CA, USA), CD8– FITC (10 µ l/106 cells; Serotec, Oxford, UK), or Ep-CAMAlexa (0.4 µ g/106 cells; BioVendor Gmbh, Heidelberg, Germany) antibodies for 30 min; washed twice with PBS/FCS; and subsequently subjected to FACS analysis. As negative controls, identical cell preparations were incubated with irrelevant mouse IgG1-PE (Chemicon, Boronia, VIC, Australia), IgG1-FITC (Chemicon), IgG1-Alexa 488 (Serotec), and IgG1– PerCP (BD Biosciences) antibodies respectively at the same concentrations. Gating for FACS analysis was designed to exclude dead cells, cell debris, and granulocytes and a FACS Calibur flow cytometer (BD Bioscience) employing Cell quest Pro acquisition software (BD Bioscience) was used.
Cell sorting
For purposes of cell sorting, four FACS sorting protocols were designed (Fig. 4). Sorting was carried out following labeling of testicular cells with fluorescence antibodies (Fig. 4). For protocols 1, 3, and 4, testicular samples were incubated directly with Ep-CAM– Alexa (0.4 µ g/107 cells; BioVendor) or CD4– PE and MHC-Cl I– PE (5 and 10 µ l/107 cells respectively; BD Biosciences) or with all three antibodies for 30 min; thereafter washed twice with cold PBS/FCS containing 2 mM EDTA; and then diluted to a concentration of 15 x 106 to 20 x 106 cells/ml for subsequent passage through a filter with pores 40 µ m in diameter. In order to amplify the labeling signal, an indirect labeling procedure was performed, which was designated as protocol 2 (Fig. 4). In this case, testicular cells were first incubated with primary mouse anti-rat Ep-CAM antibodies (0.1 µ g/12 x 106 cells; BioVendor) for 30 min and, following two washes with PBS, further incubated with PE-conjugated goat anti-mouse secondary antibody (10 µ l/107 cells; DakoCytomation) for 30 min, followed by two more washes with PBS.
Following the cell labeling, the testicular cells displaying spermatogonial characteristics (Ryu et al. 2004) with SSClow and FSChigh were subjected to FACS sorting according to four different protocols as described in Fig. 4. In the case of protocol 1, cells labeling positively for Ep-CAM– Alexa in gate R1 (Fig. 4A) were collected. In protocol 2, testicular cells labeled with both antibodies toward Ep-CAM and secondary PE-antibodies (Ep-CAM + 2nd Ab– PE) in gates Ep-CAM– high and Ep-CAM– low (Experiments 1 and 2, Table 3) and R2 (Experiment 3, Table 3) were collected and analyzed separately (Fig. 4, D1 and D2). In protocol 3, cells expressing CD4– PE and MHC Cl I– PE in gate R3 (Fig. 4B) were deleted, while the unlabeled testicular cells in area R4 were collected. Finally, cells treated according to protocol 4 were sorted by a combination of the approaches used in connection with protocols 1 and 3. Accordingly, cells expressing CD4 and MHC Cl I in gate R6 (Fig. 4C) were deleted, and cells expressing Ep-CAM in gate R5 (Fig. 4C) were simultaneously selected. All sorting was performed on a Moflo high-speed cell sorter (DakoCytomation) using two cycles, except that only one sorting cycle was carried out for experiments 1 and 2 in association with protocol 2 (Table 3).
Evaluation of the sorting procedures
The testicular cell subpopulations obtained with protocols 1–4 were either used for transplantation studies (see below) or further labeled for additional FACS analysis. In the case of FACS analysis, 0.3 x 106 to 0.5 x 106 Ep-CAM + cells obtained by protocol 1 were labeled with CD4– PE, MHC Cl I– PE, and CD90– PerCP antibodies, while the cells from protocols 2 and 4 were labeled with CD90– PerCP antibodies. The CD4/MHC Cl I-negative cells collected following protocol 3 were labeled with CD90– PerCP and Ep-CAM– Alexa antibodies as described above. As a presorting control, testicular cells from leukemic rats and equal number of cells from control animals were placed immediately after isolation into three separate tubes (106 cells per tube) and subsequently labeled as follows: tube 1: Ep-CAM– Alexa, CD4– PE, and MHC Cl I– PE antibodies; tube 2: Ep-CAM– Alexa and CD90– PerCP antibodies; and tube 3: CD4– PE, MHC Cl I– PE, and CD90– PerCP antibodies (at concentrations 0.4 µ g, 5 µ l, 5 µ l, and 10 µ l per 106 cells respectively). The phenotype of cell fractions that are labeled with these antibodies was analyzed by FACS (Table 2). In all cases, at least 20 000 cells from each sample were analyzed by FACS. In addition, cytospin slides prepared from these samples were analyzed immunocytochemically (see below).
Testicular cell injection
For the transplantation studies, 25 syngeneic PVG rats at 40 days of age were injected intratesticularly (into the left testis) with 0.1 x 106 testicular cells sorted according to one of the four protocols (Table 3). As a positive control for transmission of leukemia, six animals received the same number of FACS-sorted CD4- and MHC Cl I-positive leukemic cells intratesticularly. Furthermore, three rats received the same number of unsorted cells as a standard control. Animals that survived for more than 120 days after transplantation were considered as being non-leukemic.
Immunocytochemical staining
The cryopreserved cytospin slides and testicular sections were air-dried at room temperature (RT) for 10 min; fixed in methanol/acetone (1:1) at – 20 ° C for 10 min; washed twice with PBS; and then incubated overnight at 4 ° C with antibodies against the lymphoid markers CD3-
(Santa Cruz Bio-technology, Santa Cruz, CA, USA), CD4 and /ß TCR (Biosource, Camarillo, CA, USA), CD90 (Abcam, Cambridge, UK), and MHC Cl I (supplied by Dr Olle Lidman), all at dilutions of 1:100; or antibodies against the spermatogonial surface markers Ep-CAM (Ryu et al. 2004; BioVendor), 6/ß 1 integrin (Shinohara et al. 1999; Serotec, Kidlington, Oxford, UK), Ret (Viglietto et al. 2000; R & D Systems, Abingdon, UK), EE2 (Falciatori et al. 2004; a kind gift from Prof. Yoshitake Nishimune, Japan), or a nuclear protein marker of spermatogonia Oct-4 (Pesce et al. 1998, Ohbo et al. 2003; Santa Cruz Biotechnology) using dilutions of 1:100, 1:200, 1:20 1:104, and 1:100 respectively. In addition, testicular cytospin samples were immunolabeled with antibodies directed against the macrophage marker ED1 (Dijkstra et al. 1985; dilution 1:100; Acris, Hiddenhausen, Germany) and the Sertoli cell marker GATA-4 (Ketola et al. 2002; dilution 1:200; Santa Cruz Biotechnology, kindly supplied by Prof. Markku Heikinheimo, University of Helsinki, Finland). The cytospin samples were then washed twice with PBS and incubated with biotinylated horse anti-mouse, horse anti-goat, or goat anti-rat IgG antibodies (all purchased from Vector Laboratories, Burlingame, CA, USA) at dilutions of 1:250, 1:300, and 1:500 respectively for 30 min at RT. After two more washes with PBS, the slides were incubated with the ABC reagent (ABC kit, Vector Laboratories) for 30 min and, finally, with a solution of 3,3'-diaminobenzidine (Vector Laboratories) for 0.5–1 min until color developed, followed by counterstaining with hematoxylin (Zymed Laboratories, San Francisco, CA, USA) and mounting with an appropriate medium (Vector Laboratories).
Staining of testicular cytospin samples for the myoid cell marker, alkaline phosphatase, was carried out as described by Palombi & Di Carlo (1988). 3ß-Hydroxysteroid-dehydrogenase, which is expressed solely by Leydig cells, was detected by incubating the cytospin slides with a solution containing 2.5 mg ß-NAD, 0.6 mg NBT, and 1.5 mg etiocholan-3ß-ol-17-one per milliliter of methanol at 37 ° C for 1 h, followed by two washes with PBS and fixation in 10% formalin (Payne et al. 1980). Under a light microscope, 500 cells on each slide were examined for positive immunostaining. Negative controls were prepared either by omitting the primary antibody or by incubating the slides with irrelevant primary antibodies.
Labeling of cells with PKH dyes
In accordance with the manufacture’s instructions, 40 x 106 testicular cells isolated from healthy 40-day-old PVG rats or the same number of cells from the lymph nodes of leukemic rats were labeled with PKH 26 or PKH 67 (Sigma) respectively. Following dilution to obtain 20 x 106 labeled cells/ml medium, the testicular and leukemic cells were mixed at ratios of 100:1 or 10:1 and these mixtures subsequently incubated for 1, 2, or 3 h. Thereafter, four smears prepared from each such mixture and at each time point were analyzed under a fluorescence microscope (Nikon Eclipse 800; Tokyo, Japan), with leukemic cells emitting green fluorescence (analyzed using a FITC filter) and germ cells red fluorescence (a CY-3 filter). All image processing was performed utilizing the Image-J software (NIH, Bethesda, MD, USA, http://rsb.info.nih.gov/ij/).
In addition, testicular cells from leukemic rats were also labeled with mouse anti-rat Ep-CAM– Alexa, CD4– PE, and MHC Cl I– PE antibodies at the same concentrations and as described above. Subsequently, aggregation of leukemic and germ cells was analyzed by FACS by gating for FSC– high and SSC– low (Fig. 1G gate R1) and on the basis of the ratio of expression of CD4 and MHC Cl I versus that of Ep-CAM (as the setting for FACS sorting; Fig. 1H).
Enzymatic digestion of leukemic lymphoblasts
After separation on Ficoll, peripheral blood lymphoblasts from leukemic rats were either maintained in a medium at 34 ° C; subjected to enzymatic digestion employing a procedure that has been used to obtain single-cell suspension from testicular tissue (Jahnukainen et al. 2001); or subjected to this same procedure with the omission of the enzyme. Thereafter, the cells from all three groups were labeled separately with mouse anti-rat CD4, CD8, CD45, CD90 and MHC Cl I antibodies as described above, and FACS analysis performed.
Statistical analyses
The quantitative data in the figures and tables are presented as mean values ± S.E.M. The one-way ANOVA and t-tests were employed for statistical comparison of independent groups of samples, with a P value of <0.05 being considered to indicate a significant difference.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and MethodsAcknowledgementsReferences |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and MethodsAcknowledgementsReferences |
|---|
Anderson R, Schaible K, Heasman J & Wylie C 1999 Expression of the homophilic adhesion molecule, Ep-CAM, in the mammalian germ line. Journal of Reproduction and Fertility 116 379–384.
Armstrong A & Eck SL 2003 EpCAM: a new therapeutic target for an old cancer antigen. Cancer Biology & Therapy 2 320–326.[Web of Science][Medline]
Baum CM, Weissman IL, Tsukamoto AS, Buckle AM & Peault B 1992 Isolation of a candidate human hematopoietic stem-cell population. PNAS 89 2804–2808.
Catovsky D, Goldman JM, Okos A, Frisch B & Galton DA 1974 T-lymphoblastic leukaemia: a distinct variant of acute leukaemia. BMJ 2 643–646.
Dibley M, Dorsch S & Roser B 1975 T cell leukaemia in the rat: the pathophysiology. Pathology 7 219–235.[Web of Science][Medline]
Dijkstra CD, Dopp EA, Joling P & Kraal G 1985 The heterogeneity of mononuclear phagocytes in lymphoid organs: distinct macrophage subpopulations in the rat recognized by monoclonal antibodies ED1, ED2 and ED3. Immunology 54 589–599.[Web of Science][Medline]
Falciatori I, Borsellino G, Haliassos N, Boitani C, Corallini S, Battistini L, Bernardi G, Stefanini M & Vicini E 2004 Identification and enrichment of spermatogonial stem cells displaying side-population phenotype in immature mouse testis. FASEB Journal 18 376–378.
Fujita K, Ohta H, Tsujimura A, Takao T, Miyagawa Y, Takada S, Matsumiya K, Wakayama T & Okuyama A 2005 Transplantation of spermatogonial stem cells isolated from leukemic mice restores fertility without inducing leukemia. Journal of Clinical Investigation 115 1855–1861.[CrossRef][Web of Science][Medline]
Fujita K, Tsujimura A, Miyagawa Y, Kiuchi H, Matsuoka Y, Takao T, Takada S, Nonomura N & Okuyama A 2006 Isolation of germ cells from leukemia and lymphoma cells in a human in vitro model: potential clinical application for restoring human fertility after anticancer therapy. Cancer Research 66 11166–11171.
Geens M, Van de Velde H, De Block G, Goossens E, Van Steirteghem A & Tournaye H 2007 The efficiency of magnetic-activated cell sorting and fluorescence-activated cell sorting in the decontamination of testicular cell suspensions in cancer patients. Human Reproduction 22 733–742.
Giuili G, Tomljenovic A, Labrecque N, Oulad-Abdelghani M, Rassoulzadegan M & Cuzin F 2002 Murine spermatogonial stem cells: targeted transgene expression and purification in an active state. EMBO Reports 3 753–759.[CrossRef][Web of Science][Medline]
Hofmann MC, Braydich-Stolle L, Dettin L, Johnson E & Dym M 2005 Immortalization of mouse germ line stem cells. Stem Cells 23 200–210.
Hou M, Andersson M, Eksborg S, Soder O & Jahnukainen K 2007a Xenotransplantation of testicular tissue into nude mice can be used for detecting leukemic cell contamination. Human Reproduction 22 1899–1906.
Hou M, Andersson M, Zheng CY, Sundblad A, Söider O & Jahnukainen K 2007b Immunomagnetic separation of normal rat testicular cells from Roser’s T-cell leukemia cells is ineffective. International Journal of Andrology [in press]. DOI:10.1111/j.1365-2605.2007.00819.x.
Hu X, Bessette PH, Qian J, Meinhart CD, Daugherty PS & Soh HT 2005 Marker-specific sorting of rare cells using dielectrophoresis. PNAS 102 15757–15761.
Jahnukainen K, Morris I, Roe S, Salmi TT, Makipernaa A & Pollanen P 1993 A rodent model for testicular involvement in acute lymphoblastic leukaemia. British Journal of Cancer 67 885–892.[Web of Science][Medline]
Jahnukainen K, Hou M, Petersen C, Setchell B & Soder O 2001 Intratesticular transplantation of testicular cells from leukemic rats causes transmission of leukemia. Cancer Research 61 706–710.
Jahnukainen K, Ehmcke J, Soder O & Schlatt S 2006 Clinical potential and putative risks of fertility preservation in children utilizing gonadal tissue or germline stem cells. Pediatric Research 59 40R–47R.[CrossRef][Web of Science][Medline]
Kanatsu-Shinohara M, Toyokuni S & Shinohara T 2004 CD9 is a surface marker on mouse and rat male germline stem cells. Biology of Reproduction 70 70–75.
Ketola I, Anttonen M, Vaskivuo T, Tapanainen JS, Toppari J & Heikinheimo M 2002 Developmental expression and spermatogenic stage specificity of transcription factors GATA-1 and GATA-4 and their cofactors FOG-1 and FOG-2 in the mouse testis. European Journal of Endocrinology 147 397–406.[Abstract]
Kim TH, Hargreaves HK, Brynes RK, Hawkins HK, Lui VK, Woodard J & Ragab AH 1981 Pretreatment testicular biopsy in childhood acute lymphocytic leukaemia. Lancet 2 657–658.[Web of Science][Medline]
Kubota H, Avarbock MR & Brinster RL 2003 Spermatogonial stem cells share some, but not all, phenotypic and functional characteristics with other stem cells. PNAS 100 6487–6492.
Lay WH, Mendes NF, Bianco C & Nussenzweig V 1971 Binding of sheep red blood cells to a large population of human lymphocytes. Nature 230 531–532.[CrossRef][Medline]
Litvinov SV, Balzar M, Winter MJ, Bakker HA, Briaire-de Bruijn IH, Prins F, Fleuren GJ & Warnaar SO 1997 Epithelial cell adhesion molecule (Ep-CAM) modulates cell–cell interactions mediated by classic cadherins. Journal of Cell Biology 139 1337–1348.
Moore TJ, de Boer-Brouwer M & van Dissel-Emiliani FM 2002 Purified gonocytes from the neonatal rat form foci of proliferating germ cells in vitro. Endocrinology 143 3171–3174.[Abstract]
Muller J & Skakkebaek NE 1983 Quantification of germ cells and seminiferous tubules by stereological examination of testicles from 50 boys who suffered from sudden death. International Journal of Andrology 6 143–156.[Web of Science][Medline]
Nestvold J, Stokland A, Naper C & Rolstad B 2004 Phenotype and natural killer cell sensitivity of a radiation-induced acute T-cell leukaemia (Roser leukaemia) in PVG rats. Scandinavian Journal of Immunology 60 153–158.[CrossRef][Web of Science][Medline]
Ohbo K, Yoshida S, Ohmura M, Ohneda O, Ogawa T, Tsuchiya H, Kuwana T, Kehler J, Abe K, Scholer HR et al. 2003 Identification and characterization of stem cells in prepubertal spermatogenesis in mice small star, filled. Developmental Biology 258 209–225.[CrossRef][Web of Science][Medline]
Palombi F & Di Carlo C 1988 Alkaline phosphatase is a marker for myoid cells in cultures of rat peritubular and tubular tissue. Biology of Reproduction 39 1101–1109.[Abstract]
Payne AH, Downing JR & Wong KL 1980 Luteinizing hormone receptors and testosterone synthesis in two distinct populations of Leydig cells. Endocrinology 106 1424–1429.
Pesce M, Wang X, Wolgemuth DJ & Scholer H 1998 Differential expression of the Oct-4 transcription factor during mouse germ cell differentiation. Mechanisms of Development 71 89–98.[CrossRef][Web of Science][Medline]
Ryu BY, Orwig KE, Kubota H, Avarbock MR & Brinster RL 2004 Phenotypic and functional characteristics of spermatogonial stem cells in rats. Developmental Biology 274 158–170.[CrossRef][Web of Science][Medline]
Shaw JM, Bowles J, Koopman P, Wood EC & Trounson AO 1996 Fresh and cryopreserved ovarian tissue samples from donors with lymphoma transmit the cancer to graft recipients. Human Reproduction 11 1668–1673.
Shinohara T, Avarbock MR & Brinster RL 1999 Beta1- and alpha6-integrin are surface markers on mouse spermatogonial stem cells. PNAS 96 5504–5509.
Shinohara T, Orwig KE, Avarbock MR & Brinster RL 2000 Spermatogonial stem cell enrichment by multiparameter selection of mouse testis cells. PNAS 97 8346–8351.
Tegelenbosch RA & de Rooij DG 1993 A quantitative study of spermatogonial multiplication and stem cell renewal in the C3H/101 F1 hybrid mouse. Mutation Research 290 193–200.[CrossRef][Web of Science][Medline]
Viglietto G, Dolci S, Bruni P, Baldassarre G, Chiariotti L, Melillo RM, Salvatore G, Chiappetta G, Sferratore F, Fusco A et al. 2000 Glial cell line-derived neutrotrophic factor and neurturin can act as paracrine growth factors stimulating DNA synthesis of Ret-expressing spermatogonia. International Journal of Oncology 16 689–694.[Web of Science][Medline]
van der Wee KS, Johnson EW, Dirami G, Dym TM & Hofmann MC 2001 Immunomagnetic isolation and long-term culture of mouse type A spermatogonia. Journal of Andrology 22 696–704.[Abstract]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
