Role of Class B Scavenger Receptor Type I in Phagocytosis of Apoptotic Rat Spermatogenic Cells by Sertoli Cells*

Rat Sertoli cells phagocytose apoptotic spermatogenic cells, which consist mostly of spermatocytes, in primary culture by recognizing phosphatidylserine (PS) exposed on the surface of degenerating spermatogenic cells. We compared the mode of phagocytosis using spermatogenic cells at different stages of spermatogenesis. Spermatogenic cells were separated into several groups based on their ploidy, with purities of 60–90%. When the fractionated spermatogenic cell populations were subjected to a phagocytosis assay, cells with ploidies of 1n, 2n, and 4n were almost equally phagocytosed by Sertoli cells. All the cell populations exposed PS on the cell surface, and phagocytosis of all cell populations was similarly inhibited by the addition of PS-containing liposomes. Class B scavenger receptor type I (SR-BI), a candidate for the PS receptor, was detected in Sertoli cells. Overexpression of the rat SR-BI cDNA increased the PS-mediated phagocytic activity of Sertoli cell-derived cell lines. Moreover, phagocytosis of spermatogenic cells by Sertoli cells was inhibited in the presence of an anti-SR-BI antibody. Finally, the addition of high density lipoprotein, a ligand specific for SR-BI, decreased both phagocytosis of spermatogenic cells and incorporation of PS-containing liposomes by Sertoli cells. In conclusion, SR-BI functions at least partly as a PS receptor, enabling Sertoli cells to recognize and phagocytose apoptotic spermatogenic cells at all stages of differentiation.

Most physiological cell death is caused by apoptosis, and apoptotic cells are believed to immediately undergo heterophagic elimination by surrounding phagocytic cells, such as macrophages (reviewed in Refs. [1][2][3]. However, the molecular basis underlying the phagocytosis of apoptotic cells remains to be clarified. Apoptosis and subsequent phagocytosis also occur in areas where macrophages do not infiltrate, such as the brain and the testis. In the testis, more than half of the differentiating sper-matogenic cells die, probably by apoptosis, before they mature into spermatozoa (reviewed in Refs. 4 -8), although the mechanism and meaning of this phenomenon are unknown. The occurrence of spermatogenic cell apoptosis at various stages of differentiation has been reported (Refs. 9 -13 and reviewed in Refs. 7 and 8). Only a limited number of apoptotic spermatogenic cells, however, are detectable when testis sections are histochemically examined. This may be due to the elimination of degenerating spermatogenic cells by testicular phagocytes at the early stage of apoptosis. Electron microscopic studies with rodent testis sections have shown that degenerating spermatogenic cells are engulfed by Sertoli cells, a type of testicular somatic cell (14 -17). Some murine Sertoli cell lines show phagocytic activity against latex beads (18 -20). Sertoli cells are thus likely to be responsible for eliminating apoptotic spermatogenic cells in the testis (21). However, little is known about the regulation of this Sertoli cell function.
We previously reported that primary cultured rat spermatogenic cells of 20-day-old rats, which mostly consisted of spermatocytes, underwent apoptosis and were phagocytosed by Sertoli cells (22). In the apoptotic spermatogenic cells, phosphatidylserine (PS), 1 which is otherwise confined to the inner leaflet of the membrane bilayer, was translocated to the outer leaflet, and phagocytosis was inhibited in the presence of liposomes containing PS (23). In the present study, we examined whether spermatogenic cells at various stages of differentiation are phagocytosed by Sertoli cells in a manner similar to spermatocyte phagocytosis and tried to identify the PS receptor presumed to be present on the surface of Sertoli cells and to be responsible for recognition and subsequent phagocytosis of apoptotic spermatogenic cells.
Phagocytosis Assay-Phagocytosis of spermatogenic cells by Sertoli cells was performed as described previously (23) with a few modifications. In brief, spermatogenic cells that had been cultured without Sertoli cells for 14 -15 h were labeled with biotin (NHS-LS-Biotin, * This work was supported by a grant-in-aid for scientific research from the Ministry of Education, Science, Sports, and Culture of Japan; a grant from the Organized Research Combination System of the Science and Technology Agency of Japan; a grant from the Japan Society for the Promotion of Science; a Sasakawa scientific research grant from the Japan Science Society; and a grant from Ofukai (the Alumnae Association of Japan Women's University). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
ʈ To whom correspondence and reprint requests should be addressed: Faculty of Pharmaceutical Sciences, Kanazawa University, Takaramachi, Kanazawa, Ishikawa 920-0934, Japan. Tel.: 81-76-234-4481; Fax: 81-76-234-4480; E-mail: nakanaka@kenroku.kanazawa-u.ac.jp. Pierce) and added to Sertoli cell cultures maintained either in 96-well plates (Corning, Cambridge, MA) for phagocytosis of fractionated spermatogenic cells or in Lab-Tek Chamber Slides (Nalge Nunc, Naperville, IL) for phagocytosis of unfractionated spermatogenic cells of 20-day-old rats at a ratio of 10:1 spermatogenic cells/Sertoli cells. The culture was kept at 32.5°C for 2 h, and unreacted spermatogenic cells were removed first by pipetting with phosphate-buffered saline and then by trypsin (0.5 mg/ml) treatment for 3 min at room temperature. The remaining cells were fixed, supplemented with fluorescein isothiocyanate (FITC)labeled avidin (fluorescein-avidin D, Vector Labs, Inc., Burlingame, CA), and examined under a fluorescence/phase-contrast microscope (IX70, Olympus, Tokyo, Japan). The ratio of the number of Sertoli cells having fluorescent signals to total Sertoli cells was determined in each microscopic field. Six to eight fields from different culture wells were examined in each experiment, and the results were analyzed statistically. The means Ϯ S.D. of a typical example from at least two independent experiments were taken as the phagocytic index.
Since Sertoli cell-derived cell lines only weakly attached to the culture container, the phagocytosis assay with these cells was slightly modified. The phagocytic cells and cultured spermatogenic cells were mixed and maintained at 32.5°C for 2-3 h, and the cells were all detached from the culture container by treatment with 0.1% (w/v) trypsin and 0.02% (w/v) EDTA, placed on poly-D-lysine-coated glass slides, and further treated as described above.
Phospholipid Externalization Assay-Cell-surface PS was detected using FITC-conjugated annexin V (Bender MedSystems, Vienna, Austria) as described previously (27). Spermatogenic cells were simultaneously treated with propidium iodide and FITC-conjugated annexin V and analyzed with a flow cytometer (EPICS XL, Coulter Corp., Hialeah, FL). The cells negative for propidium iodide staining, which were considered to retain integrity of the plasma membrane, were gated and analyzed for binding of the FITC-labeled probe.
Liposome and High Density Lipoprotein (HDL) Preparation-Liposomes were prepared as described previously (28). PS-containing liposomes were composed of phosphatidylcholine (PC) and PS at a molar ratio of 7:3. Fluorescence-labeled liposomes were prepared with N-(lissamine rhodamine B sulfonyl)-L-␣-phosphatidylethanolamine (Avanti Polar Lipids) at 1% of total phospholipids. Human HDL was prepared in the density range 1.063-1.21 g/ml from plasma by ultracentrifugation according to standard procedures (29).
Northern Blots-Total RNA was prepared using acid guanidinium thiocyanate (30), and poly(A)-containing RNA was enriched by affinity chromatography (oligo(dT)-cellulose type 2, Collaborative Biomedical Products, Bedford, MA). The RNA was separated on a formaldehydecontaining 1.2% (w/v) agarose gel and blotted onto a nitrocellulose membrane (BA85, Schleicher & Schuell, Dassel, Germany). The membrane was probed with a DNA fragment derived from the human ␤-actin pseudogene (31) or a cDNA of the hamster class B scavenger receptor type I (SR-BI) (32). The hybridization signals were visualized by autoradiography using x-ray film (X-AR, Eastman Kodak Co.). Electrophoresis, transfer, blotting, and hybridization were conducted according to standard procedures (33). The probes were labeled with 32 P by random priming (33) using a commercial kit (Takara Shuzo, Otsu, Shiga, Japan).
Cloning of Rat Sertoli SR-BI cDNA-Poly(A)-containing RNA (ϳ6 g) was prepared from Sertoli cells (1.8 ϫ 10 8 ) of 20-day-old rats and used to make a cDNA library using a commercial kit (Great Lengths cDNA synthesis kit, CLONTECH). The cDNA was ligated with the ZAPII vector (Stratagene, La Jolla, CA) and packaged using Giga-packIII Gold (Stratagene). The library, containing 1 ϫ 10 5 independent clones, was screened by hybridization with a probe of the hamster SR-BI cDNA. Final positive clones were sequenced using an automated DNA sequencer (ABI Prism 377, Perkin-Elmer).
Transfection of Sertoli Cell-derived Cell Lines with Rat Sertoli SR-BI cDNA-Rat SR-BI cDNA was recloned into the pRc/CMV vector (Invitrogen, NV Leek, The Netherlands) and introduced into Sertoli cellderived cultured cell lines 15P-1 (a gift from F. Cuzin) (19) and TM4 (obtained from the American Type Culture Collection, Rockville, MA) (34) by the calcium phosphate method using a commercial kit (Invitrogen), and the cells were selected by maintenance in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum and G418 (0.5 mg/ml) (Geneticin, Life Technologies, Inc.) at 32.5°C for 7-14 days. The selected cells were cloned and used for further analyses.
Liposome Incorporation Assay-Fluorescence-labeled liposomes were added to cultured cells at 0.2 mM, and the mixture was incubated for 1 h at 32.5°C. The cells were washed with phosphate-buffered saline, and the extent of liposome incorporation was determined using either a flow cytometer or a fluorescence/phase-contrast microscope.
Latex Bead Incorporation Assay-Sertoli cell-derived cell lines were mixed with fluorescence-labeled latex beads (Polybead Microparticles (ø ϭ 0.75 m), Polysciences, Warrington, PA) for 1 h at 37°C, and incorporation of the beads was examined using either flow cytometry or microscopy.
Anti-SR-BI Antibody Preparation-Peptides corresponding to amino acid residues 76 -95 and 110 -132 of hamster SR-BI (32) with an extra Cys residue at the carboxyl terminus were synthesized. The peptides were coupled to keyhole limpet hemocyanin, emulsified with Freund's adjuvant, and injected into the backs of New Zealand White rabbits. Anti-SR-BI antibodies were affinity-purified from the rabbit sera as described previously (35). Anti-SR-BI-76 and anti-SR-BI-110 stand for antibodies raised against peptides corresponding to amino acid residues 76 -95 and 110 -132, respectively. Anti-SR-BI-76 was used throughout the study, except that immunohistochemical analysis was done with anti-SR-BI-110.
Western Blotting-Membrane fractions were prepared from liver and primary cultured Sertoli cells and spermatogenic cells of 20-day-old rats as described previously (36). The fractions were denatured under reducing conditions and separated on an 8% SDS-polyacrylamide gel. The proteins were electrophoretically transferred onto a polyvinylidene difluoride membrane (Millipore Corp., Bedford, MA), and the membrane was blocked with 5% dry skim milk. The membrane was incubated with an anti-SR-BI antibody in a buffer consisting of 10 mM Tris-HCl (pH 8.0), 0.15 M NaCl, and 0.5% Tween 20; washed; reacted with an alkaline phosphatase-conjugated anti-rabbit IgG antibody (Bio-Rad); and subjected to a chemiluminescence reaction using the Immun-Star system (Bio-Rad).
Immunohistochemistry-Cultured cell lines or Sertoli cells of 20-dayold rats were fixed with 4% paraformaldehyde/phosphate-buffered saline for 20 min at room temperature and blocked with 3% bovine serum albumin for 1 h at room temperature. The fixed cells were then mixed with an anti-SR-BI antibody and left at room temperature for 1 h. To detect the transcription factor Ad4-binding protein (Ad4BP) (37), the fixed Sertoli cells were further treated with 0.1% Triton X-100/phosphate-buffered saline, then with methanol for permeabilization of the plasma membrane, and finally with an anti-Ad4BP antibody (37) for 1 h at room temperature. The cells were supplemented with an FITCconjugated anti-rabbit IgG antibody (Immunotech, Marseilles, France) for 30 min at room temperature and examined under a fluorescence/ phase-contrast microscope.

Preparation of Spermatogenic Cell Populations with Distinct
Ploidy-The cells present in seminiferous tubules of either 20or 45-day-old rats were dispersed by successive treatment with trypsin and collagenase and subjected to a Percoll density gradient. The morphology of the separated cells was examined by microscopy, and cells with similar morphology were combined. When the combined cells were subjected to DNA flow cytometry, cell populations with distinct ploidy were found at 60 -90% purity (Fig. 1A). The cells were eluted from the gradient in the order of 4n-, 1n-, 2n-, and 1n-rich populations; 1n-rich cells were recovered in two different populations. 2nand 4n-rich cell populations were obtained from 20-day-old rats, and 1n-and 2n-rich cell populations from 45-day-old rats. The separated cell populations were distinctive in appearance, and morphological examination allowed us to identify particular spermatogenic cell types, i.e. 1n-rich cells (first elution) consisted of round spermatids, 1n-rich cells (second elution) contained many elongated spermatids, most 2n-rich cells were spermatogonia, and 4n-rich cells were spermatocytes (Fig. 1B). Testicular somatic cells, including Sertoli cells, were not recovered as a major population in any fraction, and they probably remained attached to the chamber wall of the separation apparatus.
Phagocytosis of Fractionated Spermatogenic Cells by Sertoli Cells-We first determined how efficiently the fractionated spermatogenic cells were phagocytosed by Sertoli cells. The cells were cultured without Sertoli cells for 14 -15 h to remove residual contamination of somatic cells that adhered to the culture container. The cultured spermatogenic cells, the viability of which was 80 -90% as determined by trypan blue exclu-sion, were subjected to a phagocytosis assay with isolated rat Sertoli cells. Sertoli cells prepared from 20-day-old rats phagocytosed 2n-and 4n-rich cells of 20-day-old rats as efficiently as they did unfractionated spermatogenic cells ( Fig. 2A). Similarly, 1n-and 2n-rich cells prepared from 45-day-old rats were almost equally phagocytosed by Sertoli cells from 20-day-old rats (Fig. 2B). These results showed that spermatogenic cells at various stages of differentiation were phagocytosed by Sertoli cells with similar efficiencies.
Involvement of PS in Phagocytosis of Fractionated Spermatogenic Cells-We previously showed that PS was translocated from the inner to the outer leaflet of the membrane bilayer of unfractionated spermatogenic cells of 20-day-old rats during culture without Sertoli cells and that liposomes containing PS inhibited phagocytosis of the spermatogenic cells by Sertoli cells (23). We thus examined whether PS externalization occurs in fractionated spermatogenic cells. Translocation of PS to the surface of the spermatogenic cells was determined using flow cytometry with FITC-labeled annexin V, which specifically binds to PS (38). All spermatogenic cell populations examined showed the presence of annexin V-bound cells, although the content differed among the populations (Fig. 3A), indicating that PS externalization occurs in apoptotic spermatogenic cells at all stages of differentiation.
We then examined the effect of PS-containing liposomes on phagocytosis of fractionated spermatogenic cells by Sertoli cells. As shown in Fig. 3B, the addition of PS-containing liposomes significantly reduced phagocytosis of 1n-and 2n-rich cells of 45-day-old rats as well as of 4n-rich cells of 20-day-old rats, whereas liposomes composed of PC alone had little effect. The inhibition by PS-containing liposomes of phagocytosis of 2n-rich cells from 20-day-old rats was not significant, but a decrease in the phagocytic index in the presence of the lipo-  (40,41) and is present in the testis (32,(42)(43)(44). We first examined whether Sertoli cells contain SR-BI mRNA. Oligo(dT)-selected RNA from Sertoli cells and spermatogenic cells of 20-day-old rats was blot-hybridized with a hamster SR-BI cDNA probe. The RNA from both cell types showed a discrete signal whose size roughly corresponded to that of the rat ovary SR-BI mRNA (45) (Fig. 4). This indicated that rat Sertoli cells express a gene coding for SR-BI. We then isolated an SR-BI cDNA by screening a library prepared from Sertoli cell mRNA using the hamster SR-BI cDNA as a hybridization probe. Three positive clones were obtained, and sequence analyses of two clones revealed one to be a part of the other. The longer clone included the entire coding region, and the primary sequence of rat Sertoli SR-BI 2 was found to be identical to that of rat ovary SR-BI (45).
To determine the function of Sertoli cell SR-BI, the cDNA was introduced into Sertoli cell-derived cultured cell lines 15P-1 and TM4. From the cells that remained alive in G418containing medium, 14 clones of 15P-1 and 30 clones of TM4 cells were isolated. They were first examined for the ability to incorporate fluorescence-labeled PS-containing liposomes by flow cytometry. The five 15P-1 clones and the 10 TM4 clones tested incorporated PS-containing liposomes more efficiently than the corresponding parental cells (Fig. 5A). The SR-BI mRNA was detectable in both parental cell lines (Fig. 5B). When parental cells and transfectants were immunohisto-chemically examined for the presence of SR-BI with anti-SR-BI-110, extranuclear localization of the protein was observed in all of them, and all the transfectants examined were found to contain more SR-BI than the corresponding parental cells (Fig.  5C). These cell clones were then subjected to a phagocytosis assay with apoptotic spermatogenic cells. Both parental cell lines possessed activity for phagocytosing spermatogenic cells in a PS-dependent manner (Fig. 5D). When the selected transfectants were tested, all showed higher activity levels of PSmediated phagocytosis than the parental cells; the extent of the increase in the activity caused by SR-BI expression was not  1 and 4) and poly(A)-containing RNA (lanes 2, 3, 5, and 6) (2.5 g each) from Sertoli (lanes 1, 2, 4, and 5) and spermatogenic (lanes 3 and  6) cells of 20-day-old rats were analyzed on Northern blots with probes of the hamster SR-BI cDNA (lanes 1-3) and human ␤-actin pseudogene DNA (lanes 4 -6). The positions of 28 S and 18 S rRNAs are shown with arrowheads.
large, but was significant (Fig. 5D). The levels of activity of engulfing latex beads were comparable among parental cells and transfectants (data not shown). The above results indicated that overexpression of SR-BI confers the PS-mediated phagocytic activity of apoptotic spermatogenic cells on Sertoli cell-derived cell lines.
We next studied whether SR-BI is actually involved in spermatogenic cell phagocytosis by Sertoli cells. Sertoli cells were first examined for the presence of SR-BI protein. Membrane fractions prepared from primary cultured Sertoli cells and spermatogenic cells of 20-day-old rats were analyzed by Western blotting with anti-SR-BI-76 (Fig. 6A). A discrete signal with a molecular mass of ϳ70 kDa was detected in Sertoli cell proteins, and it disappeared in the presence of the antigen peptide, but not a peptide corresponding to another region of SR-BI. Moreover, the migration of the signal was almost the same as that of a signal detectable with rat liver proteins. SR-BI protein seemed much less abundant in spermatogenic cells than in Sertoli cells. When Sertoli cells were immunohistochemically examined with anti-SR-BI-110, many of the cells showed extranuclear signals, but at differing intensities (Fig.  6B). On the other hand, the nuclei of Sertoli cells appeared to be uniformly stained with an antibody specific to the transcription factor Ad4BP, which exists in Sertoli and Leydig cells of the testis (37). Treatment with control normal rabbit IgG did not produce signals (data not shown). These results indicated that SR-BI exists in rat Sertoli cells. The antibodies used here do not distinguish SR-BII (46) from SR-BI. Most of the signal IgG. The extent of phagocytosis was measured and is shown relative to that in a reaction with no added antibodies, which was taken as 100. Right panel, anti-SR-BI-76 (3 g) and PS-containing liposomes (0.5 mM) were added to the phagocytosis reaction either by themselves or simultaneously. Significance was calculated using Student's t test. *, p Ͻ 0.001. was, however, likely to be derived from SR-BI since SR-BII is much less abundant than SR-BI in the testis (46). We then determined the effect of anti-SR-BI antibodies on phagocytosis of spermatogenic cells by Sertoli cells prepared from 20-day-old rats. The addition of anti-SR-BI-76 inhibited the phagocytosis reaction, whereas control normal rabbit IgG had a minimal effect (Fig. 6C). Anti-SR-BI-110 showed a similar inhibitory effect (data not shown). Significant levels of the phagocytic activity always remained in the presence of maximal amounts of the antibody, and the residual activity was reduced by the addition of PS-containing liposomes (Fig. 6D), indicating that some part of PS-mediated phagocytosis by Sertoli cells is resistant to anti-SR-BI antibodies. We next examined the effect of HDL, a specific ligand for SR-BI (39). HDL inhibited both phagocytosis of spermatogenic cells (Fig. 7A) and incorporation of PS-containing liposomes (Fig. 7B) by Sertoli cells of 20-dayold rats in a dose-dependent manner. All of the above results indicated that SR-BI acts as a phagocytosis-inducing PS receptor of Sertoli cells, but that a small portion of PS-mediated phagocytosis seems to be executed by a molecule(s) other than SR-BI. DISCUSSION As has been suggested from histochemical examination, apoptotic spermatogenic cells at all stages of differentiation were phagocytosed by Sertoli cells in vitro. This indicated that Sertoli cells are capable of phagocytosing all types of spermatogenic cells undergoing apoptosis during spermatogenesis. Moreover, the present results showed that all the apoptotic spermatogenic cells examined were recognized by Sertoli cells via PS exposed on the surface of the dying cells. This suggests that the apoptotic pathway and mode of subsequent recognition by Sertoli cells are common to all degenerating spermatogenic cells regardless of their state of differentiation.
Membrane phospholipids of normal cells are localized asymmetrically on the membrane bilayer, i.e. PC and sphingomyelin are mostly present in the outer leaflet, whereas other phospholipids, including PS and phosphatidylethanolamine, are restricted to the inner leaflet (reviewed in Refs. 47 and 48). If such asymmetry is lost upon induction of apoptosis, phospholipids that normally exist on the cytoplasmic side of the plasma membrane would appear on the cell surface, serving as a marker of apoptotic cells (49,50). Translocation of PS from the inner to the outer membrane leaflet has been reported in a variety of apoptotic cells such as thymocytes (51), vascular smooth muscle cells (52), neutrophils (53), spermatocytes (23), and some cultured cell lines (38, 50, 54 -56); and in cytoplasts treated with an apoptosis-inducing reagent (57). We showed here that PS externalization occurs in apoptotic spermatogenic cells at all stages of differentiation. It is thus possible that a loss of membrane phospholipid asymmetry, at least with regard to PS, is a general feature of apoptotic cells.
It is still unclear how phagocytes discriminate between apoptotic and normal cells. Several approaches have been taken to resolve this issue, and it is presumed that certain molecules present on the surface of phagocytic cells are responsible for the recognition of apoptotic cells (reviewed in Refs. 3, 58, and 59). The externalization of PS is one of the earliest changes in apoptosing cells (reviewed in Refs. 48 and 60), and a role for externalized PS in the recognition of apoptotic cells by phagocytes has been demonstrated (reviewed in Refs. 3, 48, 58, and 59). The presence of a phagocyte receptor(s) that recognizes PS exposed on the surface of apoptotic cells has thus been postulated, and several molecules have been identified as candidates (Ref. 61 and reviewed in Ref. 60); among them are class B scavenger receptors. We have provided evidence here that SR-BI, a member of the class B scavenger receptor family, is a Sertoli cell PS receptor responsible for phagocytosis of apoptotic spermatogenic cells. SR-BI has been shown to function as the PS-recognizing phagocytosis receptor in some cultured cell lines such as Chinese hamster ovary cells (41) and a human monocyte-derived cell line, THP-1 (62). It is thus likely that this particular member of the scavenger receptor family is the phagocytosis-inducing PS receptor common to the non-macrophage-type phagocytic cells. Since antibody inhibition of phagocytic activity of Sertoli cells was only partial and the residual activity was inhibitable by PS-containing liposomes, the presence of another phagocytosis-inducing PS receptor is presumed. We previously showed that integrin is not involved in the phagocytosis reaction by Sertoli cells (23). Lectins are not likely to participate in the recognition between Sertoli cells and spermatogenic cells either since the addition of GlcNAc, GlcN, GalN, or Glc at 20 mM did not affect the phagocytosis reaction (data not shown). Other approaches should be taken to identify this additional phagocytosis receptor of Sertoli cells. The level of SR-BI expression did not appear to be uniform among Sertoli cells of 20-day-old rats. This suggests that their phagocytic activity varies at different spermatogenic stages.