Biologically active kit ligand growth factor is produced by mouse Sertoli cells and is defective in Sl^d mutant mice

A double gene dose of mutant alleles at either the SI or W locus of the mouse produces the pleiotropic effects of macrocytic anemia, a deficiency of mast cells and a defect in gametogenesis and pigmentation (Russell, 1979; Kitamura et al. 1989). However, the mechanisms responsible for these abnormalities in SI or W mutant mice are different. Many lines of evidence indicate that the abnormality in W mutant animals, such as the W/W^v mouse, reflects a defect in the stem cells that generate the affected lineages, whereas that of SI mutants, such as Sl/Sl^d mouse, lies in a defect of microenvironmental factors promoting the proliferation and differentiation of the developing cells (Russell et al. 1959; McCulloch et al. 1964; Kitamura et al. 1978; Mayer and Green, 1968).

The W locus encodes the proto-oncogene c-kit (Chabot et al. 1988; Geissler et al. 1988; Nocka et al. 1989), normal cellular homolog of v-kit, the oncogene of the HZ4 feline sarcoma virus. Its sequence predicts a transmembrane tyrosine kinase receptor showing homology with CSF-1 and PDGF receptors (Besmer et al. 1986; Yarden et al. 1987; Qiu et al. 1988). Characterization of the molecular defects of several mutant alleles at the W locus has shown that they affect kinase activity and other aspects of c-kit receptor function (Tan et al. 1990; Reith et al. 1990; Nocka et al. 1990). By contrast, the SI locus could encode a putative factor that interacts with the c-kit receptor or a product affecting the activity of such a factor (Russell, 1979; Chabot et al. 1988). In fact, wild-type fibroblasts can produce a factor that interacts with the c-kit receptor, but Sl/Sl^d fibroblasts do not (Jarboe and Huff, 1989; Fujita et al. 1989). Recently, a novel growth factor, which can exist in both soluble and membrane-bound forms and is a ligand for c-kit (KL), has been-purified and the corresponding cDNA has been cloned and shown to map to the SI locus (Huang et al. 1990; Copeland et al. 1990; Zsebo et al. 1990; Anderson et al. 1990).

SI and W gene functions are necessary for migration and/or proliferation of primordial germ cells (Bennett, 1956; Mintz and Russell, 1957) and also for testicular germ cell differentiation (Nishimune et al. 1980; Tajima et al. 1991). Biologically active KL growth factor has not been identified in the testis. Sl/Sl^d:+/+ mouse aggregation chimera experiments have shown that some S1/S1° germ cells differentiated into functional spermatozoa (Nakayama et al. 1988) suggesting Sertoli cells of +/+ genotype could help mutant germ cells differentiate and that no intrinsic defects exist in mutant germ cells. Therefore, Sertoli cells would play an important role in regulating germ cell development by determining the microenvironment of seminiferous tubules.

To study the functional role of the c-kit receptor and its ligand on testicular germ cell development, we applied the mast cell proliferation assay (Nocka et al. 1990) by co-culture with Sertoli cells. In this paper, we report that biologically active KL growth factor is present in primary Sertoli cell cultures derived from +/+ and W^v/W^v mice but not from Sl^d/Sl^d mice and that it can exist in membrane-bound forms.

C57BL/6-Sl^d/+ and —W^v/+ mice as well as their normal littermates (+/+) were raised in our animal facilities using parental stocks originally obtained from The Jackson Laboratory (Bar Harbor, Maine). Mast-cell-deficient adult (WB-W/+ –C57BL/6-W^v/+) Fj -W/W’’ mice (W/W*) were purchased from the Shizuoka Laboratory Animal Center (Hamamatsu, Japan).

Homogeneous populations of cultured mast cells were obtained by culturing bone marrow cells at 10⁶ cells ml^-1 in α- MEM (Flow Laboratories, Irvine, UK) supplemented with 10^-4M 2-mercaptoethanol (Sigma), 10% fetal calf serum (FCS) (Irvine Scientific, Santa Ana, CA), l00IUml^-1 penicillin, 100 μg ml^-1 streptomycin, and 10% pokeweed mitogen-stimulated spleen cell conditioned medium (PWM-SCM) (Nakahata et al. 1982). Half the medium was replaced every 7 days, and those cultured for more than four weeks were used in the experiments as described previously (Nakano el al. 1985).

Primary cultures of Sertoli cells were prepared from 16- to 20-day-old mice by sequential enzymatic digestion (Dorrington et al. 1975) with a modified procedure as previously described (Tung et al. 1984). Briefly, the decapsulated testes were minced into small fragments and placed in Eagle’s medium (EMEM) containing Imgml^-1 collagenase (Sigma) and incubated in a shaking water bath (80cycle min”) at 37°C for 1 h to remove Leydig cells and other interstitial tissues. Subsequently, the seminiferous tubules were digested with 1mg ml”¹ hyaluronidase (Type 1-S; Sigma) at 37°C for 1 h in a shaking water bath to remove the peritubular cells. The Sertoli aggregates and single cells were then passed through a nylon mesh (100,ton) and washed three times by centrifugation (500 g). Phase-contrast microscopy was used to monitor the extent of the digestion. Sertoli cells were plated in 24-well culture plate (Corning Glass, Coming, NY) at 5 ×10^s cells ml^-1 per well (2 cm²). The cells were cultured in serum-free defined medium (SFDM) of α-MEM containing 1mg ml^-1 bovine serum albumin (Sigma), 5 μg ml^-1 Insulin (Sigma) and 5 μgml^-1 transferrin (Sigma) at 37°C in a humidified atmosphere containing 5 % CO2 and 95 % air. The next day after plating, the medium was changed. Sertoli cells thus prepared readily attached, whereas most of the germ cells remained free floating in the medium. Some germ cells were trapped in the aggregates But they tended to degenerate during culture. Sertoli cells were cultured for an additional 3-4 days in fresh medium.

For co-culture with Sertoli cells, 5×10⁴ mast cells suspended in 1 ml of SFDM were added to a culture of Sertoli cells in 24-well plates and incubated at 37°C in a humidified atmosphere containing 5 % CO₂ and 95 % air. The culture medium was aspirated and replaced with fresh SFDM every 2 days. In some experiments, direct cell contact between mast cells and Sertoli cells was inhibited using a bicamera culture chamber (Millicell, Millipore Corp) as previously described (Fujita et al. 1988). Before co-culture, +/+ Sertoli cells were seeded in the outer dish, then an equal number of +/+ mast cells were placed in both the outer and inner dish.

The proportion of mast cells in S-phase that incorporated 5-bromodeoxyuridine (BrdU; Sigma Chemical Co., St Louis, MO) was measured as described (Fujita et al. 1988). Briefly, BrdU (3 μg ml^-1) was added to the culture dishes 48 h after initiation of co-culture and cells were incubated at 37°C for 30min. Cytocentrifuge preparations of cells were fixed in Camoy’s fluid and first stained with alcian blue. They were then treated with 0.07 N NaOH for 2 min to denature DNA, neutralized with 0.1M borate buffer (pH8.5) and incubated with mouse anti-BrdU monoclonal antibody (Becton-Dickinson, Mountain View, CA) for 30 min at room temperature (Grantzner, 1982). Bound monoclonal antibody was visualized using the Vectastain elite ABC™, anti-mouse IgG, biotin-avidin-peroxidase kit (Vector Laboratories, Burlingame, CA). Mast cells were identified by the presence of alcian blue-positive granules in the cytoplasm, and the incorporation of BrdU was recognized by the presence of dark brown granules of diaminobenzidine on the nucleus.

The cultured Sertoli cells were examined both by phase-contrast microscopy and by light microscopy after fixation and staining with hematoxylin.

Primary Sertoli cell cultures were set up following the method described by Dorrington et al. and modified by Tung et al. as described in Materials and methods. This procedure resulted in a Sertoli cell culture with greater than 95 % purity (Tung et al. 1984). Fig. 1 illustrates the morphological appearance of primary mouse Sertoli cell culture on the second day after plating.

Wild-type (+/+) mast cells, but not W/W^v mutant mast cells with a defective kit receptor, can proliferate in the presence of KL growth factor (Nocka et al. 1990). Using the mast cell culture, biologically active KL growth factor can be assessed. To study the spermato-genic microenvironment, production of KL growth factor by testicular supporting cells was examined. We prepared Sertoli cells from normal (+/+) and mutant (SL/Sl¹¹) immature mice. Indicator mast cells derived from +/+ or W/W^v mice were seeded onto a Sertoli cell culture and cultured in SFDM. As shown in Table 1, when Sertoli cells derived from +/+ mice were used, 4.6 % of +/+ mast cells incorporated BrdU at 48 h after the onset of co-culture, while W/VU mutant mast cells did not incorporate BrdU under the same conditions. In contrast, Sertoli cell cultures derived from Sl^d/Sl^d mice could not support the transition of either +/+ or W/W^v mast cells into S-phase (Table 1). Furthermore, Sertoli cells derived from W^v/W^v mice could support the entry of + /+ mast cells into S-phase (Table 1). This result supports the notion that +/+ and W^v/W^v Sertoli cells expressed a biologically active KL growth factor but not Sl^d/Sl^d Sertoli cells.

The number of +/+ mast cells was maintained in the co-culture with Sertoli cells derived from + /+ but not Sl^d/Si^d mice, whereas the number of W/W’’ mast cells decreased to less than 4 % that of +/+ mast cells two weeks after seeding (Fig. 2).

To assess whether the stimulative effect of Sertoli cells was mediated by KL growth factor of membrane-bound or soluble forms, we used a bicamera culture chamber. Stimulation of DNA synthesis was observed only in mast cells in the outer dish 48 h after the onset of coculture (Table 2). Furthermore, medium conditioned with Sertoli cells did not show any stimulative effects on the proliferation of +/+ mast cells (data not shown). These results indicate that cell-cell contact is indispensable for the activity of KL growth factor produced by factor appears to be associated with the cell surface of Sertoli cells.

Biologically active KL growth factor encoded at the SI locus was expressed by Sertoli cells in the mouse testis. Sertoli cells derived from + /+ but not Sl^d/Sl^d mice supported the proliferation and maintenance of +/+ mast cells in vitro (Table 1; Fig. 2). This effect was not observed on mast cells having a kinase-deficient form of the receptor derived from W/W^v mice (Table 1; Fig. 2). Thus, sterility in Sl^d/Sl^d mice may depend on the impairment of Sertoli cell production of biologically active KL growth factor.

In agreement with the notion that a cell membranebound form of KL binds to the cell surface of BALB/c 3T3 cells but not to fibroblasts derived from Sl/Sl mice (Flanagan and Leder, 1990), the stimulant activity of the Sertoli cells required close-range interaction with mast cells (Table 2).

Seminiferous tubule transplantation has shown that the environmental effect of Sl/Sl^d mice is endogenous to the tubules (Kuroda et al. 1989). Furthermore, in seminiferous tubules of Sl/Sl^d:+/+ chimeras, normal regions alternate with undifferentiated regions, presumably corresponding to the clones of Sertoli cells of wild-type and mutant origin (Nakayama et al. 1988). These findings suggest that the KL growth factor expressed by Sertoli cells is directly affected by the SI mutation and exists as a surface protein, which fits well with the biology of developmental processes affected by W and SI mutations. All these processes can be interpreted as involving direct cell-cell contact between the affected cells and a separate type of support cell during embryogenesis and in the adult. In gametogenesis, the proliferation defect of the primordial germ cells in W/W^v or Sl/Sl^d mutant mice is expressed during migration from the yolk sac splanchnopleure to the genital ridges during early development (Mintz and Russell, 1957; Bennett, 1956). Around midgestation, transcripts of the c-kit gene have been detected in cells of the developing germ line of days old genital ridge (Orr-Urtreger et al. 1990). In contrast, KL transcripts are present in stroma cells and are not localized in the primordial germ cells (Matsui et al. 1990). In postnatal gametogenesis, c-kit transcripts has been detected in proliferating type A and type B spermatogonia (Manova et al. 1990), in immature and mature oocytes and in interstitial tissue (Manova et al. 1990; Orr-Urtreger et al. 1990). Although the mechanism of KL growth factor and its receptor interaction on spermatogenesis is not well known at present, the transition from type A to type B spermatogonia is defective in the cryptorchid conditions in Si/+ mice (Nishimune et al. 1980), and maintenance of type A spermatogonia in Sl^d/+ testis is difficult under the same conditions (Tajima et al. 1991). These findings together with our results indicate that Sertoli cells are the source of the biologically active KL growth factor (SI gene product), and that KL growth factor and its receptor could play an important role in the testicular germ cell proliferation and differentiation by signal transduction.