Sami Ventelä, 1,2 Hiroshi Ohta, 3 Martti Parvinen, 2 and Yoshitake Nishimune 3

Transcription

1 BIOLOGY OF REPRODUCTION 66, (2002) Development of the Stages of the Cycle in Mouse Seminiferous Epithelium after Transplantation of Green Fluorescent Protein-Labeled Spermatogonial Stem Cells Sami Ventelä, 1,2 Hiroshi Ohta, 3 Martti Parvinen, 2 and Yoshitake Nishimune 3 Institute of Biomedicine, 2 Department of Anatomy, University of Turku, FIN Turku, Finland Research Institute for Microbial Diseases, 3 Osaka University, Osaka , Japan ABSTRACT To study the mechanism of male germ cell differentiation, testicular germ cells carrying green fluorescent protein (GFP) as a transgene marker were transplanted into infertile mouse testis. Fluorescence-positive seminiferous tubule segments colonized with GFP-labeled donor germ cells were isolated and measured, and differentiated germ cells were analyzed in living squashed preparations. Cell associations in normal stages of the seminiferous epithelial cycle were also studied and used as a reference. Two months after transplantation, the average length of the colonies was 1.3 mm. The cell associations of transplanted colonies were consistent with those of normal stages of the cycle. However, stages of the cycle were not necessarily identical in different colonies. Three months after transplantation, the average length of transplanted colonies was 3.4 mm, and the cell association in every portion of a colony was similar to that of the corresponding stage of the cycle. Even in long fused colonies made by transplantation of a higher concentration of male germ cells, the cell association patterns in various regions of a single colony were similar and consistent with those of some of the normal stages of the cycle. Development of different stages inside the colony was observed by 6 mo after transplantation. These results indicate that the commencement of spermatogonial stem cell differentiation occurs randomly to develop different stages of the cycle in different colonies. Then, each colony shows one single stage of the cycle for a long time, even if it becomes a very large colony or fuses with other colonies. These observations indicate the existence of some kind of synchronization mechanism. By 6 mo, however, normal development of the stages of the cycle appeared in seminiferous tubules. gene regulation, Sertoli cells, spermatid, spermatogenesis, testis INTRODUCTION Spermatogenesis is a highly regulated process that takes place in the seminiferous tubules, where complex morphologic alterations lead to the formation of differentiated sperm. Spermatogenesis can be subdivided into 3 main phases: spermatogonial proliferation, meiosis of spermatocytes, and spermiogenesis of haploid spermatids. Inside the seminiferous tubules, different spermatogenic cells form defined associations also called stages [1] that can be identified by morphologic criteria of spermatids classified into steps 1 through 16 in the mouse [2]. During spermatogonial proliferation, undifferentiated type A spermatogonia divide mitotically and form intermediate and type B spermatogonia. After the last mitosis of type B spermatogonia, prelep- 1 Correspondence: Sami Ventelä, Institute of Biomedicine, Department of Anatomy, University of Turku, Kiinamyllynkatu 10, FIN Turku, Finland. FAX: ; satuve@utu.fi Received: 2 July First decision: 16 August Accepted: 10 December by the Society for the Study of Reproduction, Inc. ISSN: totene spermatocytes at stages IV VIII are formed in the seminiferous tubules, where they initiate meiosis and give rise to leptotene and zygotene spermatocytes at stages IX XII. These cells differentiate into pachytene and diplotene spermatocytes at stages I X and XI, respectively, followed by two meiotic divisions at stage XII and formation of haploid step 1 spermatids at stage I. Haploid spermatids are morphologically classified into 16 steps. A transillumination combined with phase contrast microscopy of living cells from squashes of freshly isolated seminiferous tubules allows rapid recognition of the stages of the seminiferous epithelial cycle [3]. Many different stage-specific factors have been found that are very likely needed in the local regulation of spermatogenesis [4]. However, it has not been possible to accurately determine the role of these factors because of the lack of an effective in vitro or in vivo system in which the normal development of male germ cells could be studied. An effective in vivo assay system has recently been developed in which mouse spermatogonial stem cells are transplanted into infertile mouse testis to achieve normal spermatogenesis in the recipient testis [5]. In kinetic analyses of transplanted spermatogonial stem cells, it has been demonstrated that qualitatively complete spermatogenesis is observed after 2 3 mo [6]. Recently, fluorescent (enhanced green fluorescent protein [egfp]) mouse spermatogonia transfected with chicken -actin-egfp (pcxn-egfp [7]) and acrosin-egfp (Acr3-EGFP [8]) transgenes were transplanted into infertile recipient mouse testis [9]. The specific expression of these markers was then used to identify differentiated steps of male germ cells; pcxn-egfp fluorescence varies in the developmentally different male germ cells and is very active in spermatogonia and spermatocytes [10], and acrosin-egfp labels the developing acrosome and helps to identify the steps of spermiogenesis [11]. In the present study, we used living cell techniques and the expression of both egfp-labeled transgenes to study the germ cell composition of transplanted colonies and to analyze the kinetics of stage formation in the seminiferous tubules. MATERIALS AND METHODS Animals Male C57BL/6 mice 2 mo of age were purchased from Shizuoka Laboratory Animal Center (Hamamatsu, Japan). To destroy the spermatogenic cells, busulfan was injected i.p. at a dose of 40 mg/kg. The busulfantreated mice were used as recipients 4 wk after injection. Double transgenic mice (C57BL/6TgN[acro/act-EGFP]OsbN01) carrying both the Acr3-EGFP [8] and pcxn-egfp [7] transgenes were used as donors of testicular germ cells at 7 days of age, at which time progression of germ cell differentiation is easily identified by the morphologic characterization of egfp [9]. Transplantation of Green Germ Cells Donor testicular cell suspension for transplantation was prepared by using a 2-step digestion procedure [12]. Briefly, the testes of 7-day-old 1422

2 MALE GERM CELL DEVELOPMENT AFTER TRANSPLANTATION 1423 FIG. 1. A living cell map of the stages (I XII) of the seminiferous epithelial cycle of an adult (6-mo-old) transgenic mouse expressing egfp under chicken -actin and acrosin promoters. A) Phase-contrast and fluorescence photomicrographs of live individual cells that constitute the cell association of each stage. In, Intermediate spermatogonia (s); B, type B spermatogonia (s); Pl, preleptotene spermatocytes (L); L, leptotene spermatocytes (L); Z, zygotene spermatocytes (e); P, pachytene spermatocytes and early pachytene spermatocytes at stages I III (e); D, diplotene spermatocytes (l); and m2, second meiotic division (l). Also shown are midpachytene spermatocytes at stages IV VII (m) and late pachytene spermatocytes at stages VIII XII (l). Numbers label spermatids at each step and elongated spermatids (steps 12 16) (E). Symbols in parentheses were used in B and are expressed as the percentage of germ cells at each differentiation stage per total number of germ cells counted. mice were placed in Dulbecco modified Eagle medium buffered with 20 mm Hepes at ph 7.3 containing collagenase type I (1 mg/ml) and hyaluronidase (1 mg/ml). The testes were incubated for 15 min at 37 C, with manual agitation at 5-min intervals. The seminiferous tubules were washed twice in calcium-free PBS and then incubated in PBS containing 0.25% trypsin for 15 min at 37 C, with manual agitation at 5-min intervals. Fetal bovine serum (one half of the volume of the cell suspension) was added, and the resulting cell suspension was pipetted several times. The suspension was then filtered through a nylon mesh with 30- m pores to remove large clumps of cells. The filtrate was centrifuged at 600 g at 16 C for 5 min, and the cell pellet was resuspended in injection medium (138 mm NaCl, 8.1 mm Na 2 HPO 4, 2.7 mm KCl, 1.1 mm KH 2 PO 4, 0.1 mm EDTA, 5.5 mm glucose, 5 mg/ml BSA, 100 g/ml DNase I, and 0.4 mg/ml trypan blue) [5] at a concentration of 10 6 or 10 8 cells per milliliter. The transplantation was performed via the efferent ductules as previously described [13]. Approximately 10 l of the donor cell suspension was transplanted into the seminiferous tubules of the busulfan-treated mice. Altogether, 65 different resulting colonies were studied from nine mice at 2, 3, and 6 mo after transplantation. Microscopic Evaluation Characterization of fluorescent germ cells in different stages in adult transgenic mice. Seminiferous tubules of the adult double transgenic mice were examined by a transillumination technique [3] with a fluorescence stereomicroscope (Leica DM RXE; Leica Microskopie & System GmbH, Wetzlar, Germany). From all stages of the cycle of the seminiferous epithelium, 0.5- to 1.0-mm-long representative segments were selected and carefully squashed between a microscope slide and coverslip as described previously [10]. From each squashed preparation, 8 15 images were processed (Leica DC 200) for further analyses, and differentiation of germ cells was identified according to morphologic criteria [1, 3] with oil immersion phase-contrast optics at 1000 magnification (Fig. 1A). Specific patterns of egfp fluorescence were used as additional aids for classifying the differentiated germ cells. The cells were divided into 12 groups: 1) type A, intermediate, and type B spermatogonia, 2) preleptotene and leptotene spermatocytes, 3) zygotene and early pachytene spermatocytes, 4) midpachytene spermatocytes, 5) late pachytene and diplotene spermatocytes, secondary spermatocytes, and meiotic divisions, 6) step 1 sperma- FIG. 2. Fluorescence microscopic pictures of transplanted colonies. Positive seminiferous tubular segments were observed in whole testis 2 mo after germ cell transplantation (A) and were dissected for further study. Arrowhead indicates the fluorescence-negative intermediate area between two colonies, and arrow indicates the center of the colony. B) Selected areas of the fluorescence-positive colonies were cut into 0.5- to 1.0-mm segments and squashed. The isolated cells were identified by morphologic criteria observed by phase-contrast (c1, d1) and fluorescence (c2, d2) microscopy. When the faint egfp-positive area (i.e., the end of the colony) was squashed, the aligned and connected spermatogonia were isolated from the edge of the seminiferous tubule (C, c1, c2). When a squashed preparation was made from the center of the colony, more developed male germ cells were observed (D, d1, and d2). A, Type A spermatogonia; s8, step 8 spermatid. Bar 1mm(A, B) and 10 m (c2, d2).

3 1424 VENTELÄ ET AL. TABLE 1. Quantitative analyses of cell composition of the colonies 2 mo after transplantation of 10 6 cells per testis. a Colony b 1 (A) 2 (A) 3 (A) 4 (B) 5 (C) 6 (D) 7 (D) 8 (E) 9 (E) 10 (E) Length (mm) c s L e m l E d 32.6 d 36.0 d 3.5 d 22.4 d 14.6 d 20.4 d 19.2 d d 54.0 d 18.0 d d 24.6 d d 34.7 d 32.4 d 14.2 d 1.3 d a The lengths of 10 fluorescence-positive colonies were measured. The numbers of individual cell types were determined and are expressed as percentages of the total number of germ cells ( 100 germ cells in each colony). s, Type A, intermediate, and type B spermatogonia; L, preleptotene and leptotene spermatocytes; e, zygotene and early pachytene spermatocytes (stages XIIII); m, midpachytene spermatocytes (stages IV VII); l, late pachytene spermatocytes (pachytene spermatocytes at stages VIII-X, diplotene spermatocytes at stage XI, secondary spermatocytes, and meiotic cells at stage XII); 1, step 1 spermatids; 3, step 2 3 spermatids; 5, step 4 5 spermatids; 7, step 6 7 spermatids; 9, step 8 9 spermatids; 11, step spermatids; E, elongated spermatids. b Letters in parentheses (A E), correspond to the normal stages shown in Figure 1B according to the similarity of cell associations in each colony. c Mean colony length was 1.3 mm. d Corresponds to normal stages in control tubules. Progression of spermatid maturation seemed to be delayed compared with that in normal tubules, especially in colonies 1, 2, 3, and d 32.2 d 15.6 d d 76.0 d 30.6 d d 8.3 d 41.0 d d 31.1 d 13.8 d 6.0 d 6.5 d 11.0 d 21.0 d 5.3 d tids, 7) step 2 3 spermatids, 8) step 4 5 spermatids, 9) step 6 7 spermatids, 10) step 8 9 spermatids, 11) step spermatids, and 12) elongated spermatids (steps 12 16). Characterization of fluorescent germ cells in transplanted colonies in recipient mice. At 2, 3, and 6 mo after transplantation, the recipient mice were anesthetized, and the testes were removed and examined by fluorescence stereomicroscopy. After the seminiferous tubules were decapsulated, they were gently separated with forceps; fluorescent segments were identified, and their lengths was measured. To study the developmental stages of spermatogenesis, colonies were cut into 0.2- to 1.0-mm segments. Squashed preparations (1 15 slides) were made from each colony, and 8 15 representative areas were selected for imaging and quantitative analyses. Statistical Methods An independent-samples t-test was used for statistical analyses. A P value 0.05 was considered significant. RESULTS Analyses of the Seminiferous Epithelial Cycle in Double egfp Transgenic Mouse Testis Cells from intermediate-type spermatogonia to preleptotene spermatocytes had only a faint fluorescent intensity (Fig. 1A). From leptotene spermatocytes to late pachytene spermatocytes, the intensity of egfp fluorescence gradually increased. Accumulation of acrosin-egfp in the Golgi complex began in diplotene spermatocytes. During spermiogenesis, the stage-specific development of the acrosome system (Golgi, cap, and acrosome phases) was clearly seen under fluorescence microscopy. After we identified the stages of the cycle in the seminiferous epithelium, we scored the relative numbers of 12 classes of male germ cells and expressed the frequencies of individual cell types as percentages of the total number of cells at each differentiation step. The quantitative analyses were made in five groups of cell associations in squashed seminiferous tubular fragments (Fig. 1B). Synchronization of Germ Cell Differentiation in Transplanted Colonies Two months after transplantation, colonized cells in the seminiferous tubules were identified by their fluorescence (Fig. 2, A and B). The average length of the colony was 1.3 mm (Table 1). The most advanced germ cells were haploid spermatids that were typically found in the middle of egfp-positive colonies (Fig. 2D), whereas only spermatogonia sometimes together with spermatocytes were found at the ends of the colonies (Fig. 2C). When more than two different parts of a colony were studied, the cell association was usually the same throughout the colony and was similar to that of a corresponding normal stage in control tubules. However, the stage of the cycle was not the same in different colonies (Fig. 3). In some colonies, the majority of the most advanced type of spermatids were at steps 6 7 (e.g., colonies 5 and 7 in Table 1), but no elongated spermatids were present in the colonies, indicating that germ cell differentiation had still not occurred 2 mo after transplantation. Furthermore, a slight deviation from the cell association of normal seminiferous epithelial stages was observed in some colonies; that is, some spermatids in more advanced steps of differentiation seemed to appear later than normal (e.g., colonies 1, 2, 3, and 7 in Table 1). These results indicate that the cell association in one colony is similar but independent from that in other colonies, even in the same seminiferous tubules. Moreover, completion of the stage of the cycle requires more than 2 mo after transplantation. Synchronization of Germ Cell Differentiation in Elongated Colonies To further study the developmental process of transplanted germ cells in the colonies, recipient seminiferous tubules were examined 3 mo after transplantation (Table 2). The average length of the colonies had increased to 3.4 mm. Even in very long colonies, the cell association in every part of a colony was qualitatively similar to that of the normal stage (e.g., colony 10 in Table 2; Fig. 4 is comparable to A[I III] in Fig. 1B). However, there were significant quantitative differences between cell numbers in transplanted colonies and normal seminiferous epithelium. The average rate of increase in the length of transplanted colonies was estimated to be approximately 1.1 mm/mo, resulting in a mean length of 3.4 mm 3 mo after transplantation. Although this estimation is compatible with previous results [12], the rate of increase in colony length in the first

4 MALE GERM CELL DEVELOPMENT AFTER TRANSPLANTATION 1425 FIG. 3. A various number of colonies settled in one seminiferous tubule. Example of three separate colonies observed 2 mo after transplantation. A) Independent colonies in the same seminiferous tubule showed different stages of the cycle. Three independent colonies (a, b, c) were isolated and dissected. The cell association in a squashed preparation of each colony (inset: numbers indicate the differentiation steps of spermatids) was analyzed by phase-contrast and fluorescence microscopy as described in Figure 1A. Bar 1 mm. The relative numbers of differentiated cells in each colony is indicated by the shaded bars in B, C, and D. *,P In B, cell numbers in normal stages are indicated by open bars. TABLE 2. Quantitative analyses of cell composition of the colonies 3 mo after transplantation. a Colony b 1 (A) 2 (A) 3 (C) 4 (C) 5 (C) 6 (C) 7 (D) 8 (D) 9 (E) 10 (E) Length (mm) c s L e m l E d 19.4 d 33.5 d 29.9 d 33.0 d 29.1 d 25.9 d 18.5 d 42.9 d 29.0 d d 30.6 d d 27.3 d 30.4 d 22.5 d d 22.5 d 2.7 d 9.8 d a,b Footnotes are the same as those given in Table 1. c Mean colony length was 3.4 mm. d Corresponds to normal stages in control tubules. The differentiation of spermatids seemed to be delayed in colonies 3, 5, and 7, whereas more spermatids had progressed in colonies 4 and d 36.5 d 25.2 d 17.9 d d 3.9 d 5.6 d d 24.9 d 19.4 d d 0.6 d 18.4 d 1 d 17.1 d 12.4 d 4.2 d 13.2 d 3.2 d 18.5 d 5.8 d 19.8 d

5 1426 VENTELÄ ET AL. FIG. 4. Fluorescence microscopic picture of an isolated long colony in a seminiferous tubule and its cell association. Three months after transplantation, a long colony was identified and isolated (A). A 0.5- mm length of four dissected fragments at position (a) of the colony was squashed, and relative numbers of cells was scored in each fragment. B) All of the fragments showed a similar cell association (shaded bars), which corresponded to that of stage I III in normal adult seminiferous tubules (open bars). *, P White arrows, The ends of the colony; a, the analyzed positions inside the colony. Bar 1.0 mm. 2 mo was approximately half of this value (0.6 mm/mo). Furthermore, at 3 mo, many colonies were more than two to three times longer than the shorter colonies (Table 2). The average colony length should be approximately 3 mm or a little shorter at 3 mo after transplantation if the colonies remained separated. Actually, the length of some of the observed colonies was more than 3.4 mm, suggesting that some colonies were derived from the fusion of neighboring colonies, which then developed synchronously. Also, local differences in spermatogonial proliferation may explain the existence of abnormally long colonies. Furthermore, spermatogenesis seemed to be qualitatively completed at 3 mo after transplantation, since we observed substantial numbers of elongated spermatids in all colonies (Table 2). Synchronization of Germ Cell Differentiation in Fused Colonies Three months after injection of 10 l of a cell suspension containing 10 8 cells per milliliter, we observed many long colonies presumed to be fused colonies. Ten selected, very large colonies were examined. Their average length was 12.4 mm (Table 3). After dissecting the long fluorescent tubules into many short segments (mean length, mm), we analyzed the type of germ cells in each squashed segment. Almost all segments in one colony showed very similar associations of differentiated germ cells, although some colonies deviated a little from the normal pattern (e.g., colonies 7 10 in Table 3). These results indicate that the stage of the seminiferous tubules in the whole length of a large colony is synchronized, although some colonies begin to show desynchronization before physiologic differentiation of the stages is evident. Six months after transplantation, however, segmental dissection of the colonies revealed more prominent development of different cell associations within colonies (Fig. 5). In long colonies, cell associations in one part of the colony were different from those in other parts of the colony. When these tubular segments were quantitatively and qualitatively compared with normal seminiferous tubule segments, only minor differences were found. These results suggest that the appearance of the normal stages of the seminiferous epithelial cycle start at the time of colony fusion. DISCUSSION A technique in which spermatogonial stem cells are successfully transplanted into infertile mouse testis has previously been described [5]. Recently, the transplantation technique has improved by using labeled donor spermatogonia either from LacZ [6] or from double-labeled egfp [9] TABLE 3. Quantitative analyses of cell composition of the fused colonies 3 mo after transplantation of 10 8 cells per testis. a Colony b 1 (A) 2 (A) 3 (A) 4 (C) 5 (C) 6 (C) 7 (CD) 8 (DE) 9 (DE) 10 (ABE) Length (mm) c s L e m l E d 6.8 d 5.3 d 0.5 d 1.2 d d 10.9 d 14.0 d 12.0 d 15.0 d 10.6 d 13.4 d 15.4 d 21.6 d 12.1 d d 10.6 d 9.6 d d 15.3 d 11.4 d 3.6 d d d 10.5 d 16.0 d d 11.2 d d 0.9 d 17.0 d 25.9 d 13.2 d 23.0 d a,b Footnotes are the same as those given in Table 1. Data are from 10 arbitrarily selected colonies. c Mean colony length was 12.4 mm. d Corresponds to normal stages in control tubules d d 24.3 d 18.4 d 12.1 d d 17.9 d 18.1 d 6.7 d d 13.3 d 25.3 d 22.2 d 26.3 d 25.0 d 19.8 d 21.0 d 27.6 d 20.8 d 10.5 d 16.6 d

6 MALE GERM CELL DEVELOPMENT AFTER TRANSPLANTATION 1427 FIG. 5. Representative figures of a colony 6 mo after transplantation. A) An isolated colony in the testis 6 mo after transplantation. White arrows indicate the ends of the colony. B, C) Relative numbers of the differentiated cells at regions (a) and (b), respectively. Open bars in B and C indicate the cell associations of stage VIII IX and IV V, respectively, of normal control seminiferous tubules; *, P Phase-contrast micrographs (D and F) and corresponding fluorescence micrographs (E and G) of squashed preparations from regions (a) and (b), respectively. H8, Step 8 spermatids; 9, step 9 spermatids. Bar 1.0 mm (A) and 10 m (D, F). transgenic mice. By these techniques, the stem cell settlement and differentiation and also the kinetics of colonization can be studied either after fixation and 5-bromo-4-chloro-3-indolyl- -D-galactopyranoside (X-gal) incubation (LacZ) or with real-time observation under excitation light (egfp) in recipient seminiferous tubules. However, no earlier quantitative or qualitative observations of spermatogenesis in transplanted colonies are available. In the double transgenic egfp donor mouse, all germ cells were labeled with egfp; that is, spermatogonia or spermatocytes were labeled with chicken -actin egfp, whereas spermatids were detected with acrosin-egfp transgene marker, which labels the acrosome system during spermiogenesis. Application of these useful markers facilitated the detailed analysis of the stage of the seminiferous epithelial cycle together with the identification of the differentiated germ cells. Thus, the cell association in normal seminiferous tubules was quantitatively defined in every stage of the cycle (even in living tissue) under fluorescence microscopy more accurately than has been possible with a transillumination microdissection technique [3]. By transillumination of the seminiferous tubules [3], no difference in transillumination could be observed between colonized and noncolonized areas for a long time after transplantation, although qualitatively complete spermatogenesis was observed in most of the colonies with fluorescence micros-

7 1428 VENTELÄ ET AL. copy. In contrast, the transillumination pattern starts to develop at the age of days, and different stages can be identified at least 2 mo after birth in normal germ cell differentiation in male mice (unpublished data). The homogeneously pale transillumination pattern in transplanted seminiferous tubules is actually due to a low number of maturation phase spermatids. Alternatively, some barrier may exist to cross meiotic division and to cause some retardation of spermiogenesis. Even after elongation of a colony to more than 4 mm in 3 mo, only one type of cell association was found within the colony (Table 2 and Fig. 4), although the length or size of one stage segment of normal seminiferous tubules has been estimated to be mm in rats [14]. These findings indicate that development within the colony is synchronized. Even in long fused colonies (mean length, 12.4 mm; Table 3) achieved by the injection of a high concentration of germ cells, we observed a single cell association, indicating synchronization of fused colonies. These observations indicate that some regulatory mechanism(s) should exist to maintain the synchronized development of male germ cells in the transplanted colonies. In our study, more than half of the colonies showed the cell association of stage VI IX (C and D in Table 2). As this phase is the point of nuclear elongation of round spermatid, it might be sensitive to the local factors secreted by Sertoli cells, which might cause some retardation at this stage and may work as a synchronizer in transplanted colonies. This notion is supported by an in vitro observation that elongation of step 8 spermatids is impossible in culture conditions in the rat [15]. This might also explain the low numbers of elongated spermatids in transplanted mouse testis even 1 yr after transplantation [16]. Previously, it has been demonstrated that vitamin A deficiency leads to an arrest of spermatogenesis and a loss of advanced male germ cells [17]. After spermatogenesis was restored by retinol administration, synchronized development of male germ cells occurred in the whole testis [18, 19]. In our study, male germ cell development was synchronized in each colony, but cell associations between colonies varied, in contrast to the synchronization noted in the whole testis in vitamin A-deficient mice given retinol. As Sertoli cells are known to regulate the proliferation and differentiation of germ cells [20], synchronized development of transplanted colonies may also be controlled by Sertoli cells. The interaction between newly settled spermatogonia and the recipient Sertoli cells may first cause the proliferation of spermatogonia and then promote their differentiation [9]. This is supported by the observation that differentiation mainly occurs in the middle portion of the colony, while spermatogonia proliferate in the periphery. It is obvious that a certain number of spermatogonia are needed to form colonies long enough to allow differentiation. It has previously been demonstrated that Sertoli cell function is cyclically controlled by maturing male germ cells [4] and also by differentiated spermatogonia [21]. Glial cell-line derived neurotrophic factor [22] and some other factors secreted by Sertoli cells may regulate the proliferation and differentiation of spermatogonia and may also be responsible for the synchronization of germ cell differentiation in the fused colonies. It takes at least 6 mo to develop the normal cyclic function of Sertoli cells, and the fusion of adjacent colonies may be an important prerequisite for development of the normal cycle of the seminiferous epithelium. In the present study, we have developed a new method for studying the differentiation of male germ cells in normal and transplanted testes. This method permits more accurate identification of male germ cells by morphologic criteria with phase-contrast microscopy and in combination with egfp fluorescence in living tissue. We have also demonstrated that each colony transplanted into seminiferous tubules develops independently and maintains typical stagedependent cell associations for a very long time after transplantation. Furthermore, if more than two colonies fuse, each stage of the cycle is adjusted to one stage and synchronized. These phenomena may be controlled by some local regulators among the many currently identified testisspecific genes or by some stage-specific factors. The combination of double-labeled egfp transgenic spermatogonia and accurate identification of living male germ cells is more sensitive than the usual simple transillumination technique for identifying differentiation of spermatogonial stem cells. Furthermore, this combination is a good tool for studying the mechanism of testicular germ cell differentiation and the development of the stages of the cycle in the seminiferous epithelium. REFERENCES 1. Oakberg EF. Duration of spermatogenesis in the mouse and timing of stages of the cycle of the seminiferous epithelium. Am J Anat 1956; 99: Russell LD, Ettlin RA, Sinha Hikim AP, Clegg ED. Histological and Histopathological Evaluation of Testis. Clearwater, FL: Cache River Press; 1990: Parvinen M, Hecht NB. Identification of living spermatogenic cells of the mouse by transillumination-phase contrast microscopic technique for in situ analyses of DNA polymerase activities. Histochemistry 1981; 71: Parvinen M. Cyclic function of Sertoli cells. In: Russell LD, Griswold MD (eds.), The Sertoli Cell. Clearwater, FL: Cache River Press; 1993: Brinster RL, Zimmermann JW. Spermatogenesis following male germ-cell transplantation. Proc Natl Acad Sci U S A 1994; 91: Nagano M, Avarbock MR, Brinster RL. Pattern and kinetics of mouse donor spermatogonial stem cell colonization in recipient testes. Biol Reprod 1999; 60: Okabe M, Ikawa M, Kominami K, Nakanishi T, Nishimune Y. Green mice as a source of ubiquitous green cells. FEBS Lett 1997; 407: Nakanishi T, Ikawa M, Yamada S, Parvinen M, Baba T, Nishimune Y, Okabe M. Real-time observation of acrosomal dispersal from mouse sperm using GFP as a marker protein. FEBS Lett 1999; 449: Ohta H, Yomogida K, Yamada S, Okabe M, Nishimune Y. Real-time observation of transplanted green germ cells : proliferation and differentiation of stem cells. Dev Growth Differ 2000; 42: Ventelä S, Okabe M, Tanaka H, Nishimune Y, Toppari J, Parvinen M. Expression of green fluorescent protein under beta-actin promoter in living spermatogenic cells of the mouse: stage-specific regulation by FSH. Int J Androl 2000; 23: Ventelä S, Mulari M, Okabe M, Tanaka H, Nishimune Y, Toppari J, Parvinen M. Regulation of acrosome formation in mice expressing GFP under acrosin promoter. Tissue Cell 2001; 32: Ohta H, Yomogida K, Dohmae K, Nishimune Y. Regulation of proliferation and differentiation in spermatogonial stem cells: the role of c-kit and its ligand SCF. Development 2000; 127: Ogawa T, Arechaga JM, Avarbock MR, Brinster RL. Transplantation of testis germinal cells into mouse seminiferous tubules. Int J Dev Biol 1997; 41: Perey B, Clermont Y, Leblond P. The wave of the seminiferous epithelium in the rat. Am J Anat 1961; 108: Toppari J, Parvinen M. In vitro differentiation of rat seminiferous tubular segments from defined stages of the epithelial cycle: morphologic and immunolocalization analysis. J Androl 1985; 6: Russell LD, Franca LR, Brinster RL. Ultrastructural observations of spermatogenesis in mice resulting from transplantation of mouse spermatogonia. J Androl 1996; 17: Van Thiel DH, Gavaler J, Lester R. Ethanol inhibition of vitamin A

8 MALE GERM CELL DEVELOPMENT AFTER TRANSPLANTATION 1429 metabolism in the testes: possible mechanism for sterility in alcoholics. Science 1974; 186: Ismail N, Morales C, Clermont Y. Role of spermatogonia in the stagesynchronization of the seminiferous epithelium in vitamin-a-deficient rats. Am J Anat 1990; 188: Gaemers IC, van Pelt AM, van der Saag PT, de Rooij DG. All-trans- 4-oxo-retinoic acid: a potent inducer of in vivo proliferation of growth-arrested A spermatogonia in the vitamin A-deficient mouse testis. Endocrinology 1996; 137: Russell LD. Morphological and functional evidence for Sertoli-germ cell relationship. In: Russell LD, Griswold MD (eds.), The Sertoli Cell. Clearwater, FL: Cache River Press; 1993: de Rooij DG, Lok D, Weenk D. Feedback regulation of the proliferation of the undifferentiated spermatogonia in the Chinese hamster by the differentiating spermatogonia. Cell Tissue Kinet 1985; 18: Meng X, Lindahl M, Hyvonen ME, Parvinen M, de Rooij DG, Hess MW, Raatikainen-Ahokas A, Sainio K, Rauvala H, Lakso M, Pichel JG, Westphal H, Saarma M, Sariola H. Regulation of cell fate decision of undifferentiated spermatogonia by GDNF. Science 2000; 287: