0013-7227/90/1276-2744$02.00/0 Endocrinology Copyright © 1990 by The Endocrine Society

Vol. 127, No. 6 Printed in U.S.A.

Insulin-Like Growth Factor Binding Protein-3 Secretion from Cultured Rat Sertoli Cells: Dual Regulation by Follicle Stimulating Hormone and Insulin-Like Growth Factor-I ERIC P. SMITH*, BRYAN A. DICKSON, AND STEVEN D. CHERNAUSEKf Division of Endocrinology, Children's Hospital Medical Center, Cincinnati, Ohio 45229

ABSTRACT. The insulin-like growth factors (IGFs) are found in extracellular fluids bound to carrier proteins which influence the biological activity of the IGFs. Three structurally different binding proteins (BPs) have been isolated and cloned; each has distinct tissue specific expression and unique properties. We report here that testicular cells synthesize a specific subset of these binding proteins. Ligand blot analysis and RNA blot hybridization indicates that cultured peritubular cells synthesize primarily IGFBP-2. In contrast, as determined by ligand blot, RNA blot hybridization and N-linked deglycosylated studies, IGFBP-3 is predominantly synthesized by the Sertoli cell. In a dose dependent fashion, FSH markedly reduces the levels of

T

HE male reproductive system is thought to be modulated by locally produced growth factors that include transforming growth factor (TGF)/? (1), TGF« (2), the insulin-like growth factors (IGFs) (3-6), seminiferous growth factor (6), nerve growth factor (7), basic fibroblast growth factor (basic FGF) (8), and interleukin-1 (9). Though the function of these factors remains unclear, the notion of the IGFs, in particular, being autocrine/ paracrine amplifiers of gonadotropin action is increasingly accepted. Recent evidence suggests that the biological activity of the IGFs produced within the testis can be influenced by IGF binding proteins (10-12). There are three sequenced and cloned IGF binding proteins designated: IGFBP-1 (13), IGFBP-2 (14) and IGFBP-3 (15) and evidence for two other distinct species (16,17). IGFBP-2 was isolated from rat liver-derived BRL-3A cells (18), and has a higher affinity for IGF-II than IGF-I (12). IGFBP-1 is synthesized ubiquitously but in apparent greatest abundance by the liver (19) and decidua (20). It is found in

IGFBP-3 in Sertoli cell conditioned medium. Similarly, isoproterenol, (Bu)2cAMP and cholera toxin also markedly reduce the abundance of IGFBP-3 in conditioned media. In contrast, IGFI increases the concentrations of IGFBP-3 with the concentration required for half-maximal stimulation, approximately 20 ng/ml. Consistent with a peritubular cell origin, IGFBP-2 may be the predominant species found in interstitial fluid. In summary, our data reveal that the IGFBPs are expressed in a cell type specific manner in the testis. The opposing effects of FSH and IGF-I on Sertoli cell IGFBP-3 expression suggests a mechanism by which the IGF-I biological activity on Sertoli cell might be influenced. (Endocrinology 127: 2744-2751, 1990)

high concentrations in amniotic fluid (21) and its levels in plasma are regulated by nutritional status and insulin (12). IGFBP-3 is GH dependent in plasma and circulates in serum complexed to IGF and an acid labile subunit (22). Evidence already exists that IGF binding proteins are present in the testis (23), in seminal fluid (24, 25), and are released into the conditioned medium of both rat Sertoli cell and peritubular cells (3, 26). Cailleau has shown that cultured Sertoli cell IGFBPs as measured by charcoal assay are reduced by FSH and other cAMPstimulating ligands (26). However, the specific binding proteins produced by the testis have not been characterized or their individual regulation investigated. Recent studies in the female reproductive system suggest an important role for IGFBP-3 as a local inhibitors of IGF actions on the granulosa cell (10, 27-29). We hypothesized that a similar BP species is present in the testis and that distinct factors may regulate its expression.

Materials and Methods Received July 17, 1990. Address requests for reprints to: Dr. Eric P. Smith, Division of Endocrinology, Children's Hospital Medical Center, Elland and Bethesda Avenues, Cincinnati, Ohio 45229. * Supported by Children's Hospital Trustee Grant 31-367-630. f Supported by NIH Grant 5-R29-NS25354.

Reagents Penicillin and streptomycin were obtained from Whittaker M.A. Bioproducts (Walkersville, MD) and gentamicin from GIBCO Laboratories (Grand Island, NY.) (Bu)2cAMP, insulin,

2744

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2017. at 15:01 For personal use only. No other uses without permission. . All rights reserved.

SERTOLI CELL IGFBP-3 and cholera toxin were obtained from Sigma (St. Louis, MO). Ovine FSH (LER-1575) preparations were provided by the Pituitary Hormone Distribution Program, NIDDK. Basic FGF and IGF-I were obtained from Amgen Biologicals (Thousand Oaks, CA), TGFjS from R & D Systems (Minneapolis, MN) and epidermal growth factor from Collaborative Research (Bedford, MA).

2745

for 5 min, the samples were centrifuged at 12,000 X g for 15 min, the aqueous phase was transferred to a clean microfuge tube, and an equal volume of isopropanol added. The samples were then allowed to stand at 4 C overnight, centrifuged at 12,000 X g for 15 min, washed in 75% ethanol, and dissolved in diethyl pyrocarbonate treated water. RNA analysis

Primary cell culture Sertoli cell-enriched cultures were prepared from testes of 15 day-old Sprague Dawley rats (Harlan, Indianapolis, IN) by a modification (30) of the method of Dorrington (31). Cells were plated into 60-mm dishes in Eagle's minimal essential medium supplemented with 1 mM pyruvate, 4 mM glutamine, nonessential amino acids, 250 U/ml penicillin, 250 U/ml streptomycin, 12.0 Mg/ml gentamicin, and 5% fetal bovine serum (GIBCO). After 24 h to allow for attachment, now confluent cell monolayers were washed once, and fresh serum-free medium was added. Media were changed on day 3 and on day 4 before initiating experiments to ensure that fetal calf serum contaminants would not influence the results of the binding protein experiments. Peritubular cell cultures were performed according to the methods of Hutson and Stocco (32) and Tung et al. (33). Supernatant from the collagenase digestion was centrifuged at 20 X g at room temperature for 10 min in 50 ml conical tubes to sediment Sertoli cell clusters. The pellet was discarded and the supernatant was centrifuged at 200 X g for 10 min. The pellet, containing single cells, was resuspended in the same serum-containing medium used for Sertoli cell cultures plated in 10-cm dishes at a concentration of approximately 5 million cells per dish. Medium was changed at 24 h and on day 3. Before experimentation, the cells were washed two times in 10 ml serum free Eagle's minimal essential medium before adding fresh defined medium without serum. For both Sertoli cell and peritubular cell culture experiments, growth factors were added as 100X aliquots in Hanks' balanced salt solution containing 1 mg/ml BSA (98-99% pure, fatty acid free, Sigma-A-7030). In the experiments where the effects of hormones were assessed, each time point represented a pooled media sample from duplicate 60-mm dishes. RNA isolation Poly(A)+ mRNA was isolated from cultured Sertoli cell and peritubular cells by the method of Chirgwin et al. (34) and as modified by Lund et al. (35). Other total RNA samples were isolated by the method of Chomczynsky and Sacchi (36). Briefly, Sertoli cells in 60-mm dishes or peritubular cells in 10cm dishes or 100-mg aliquots of —70 frozen tissue were lysed in 1 ml, 3 ml, and 2 ml RNAzol (Cinna/Biotecx, Friendswood, TX), respectively. For tissue, the samples were first pulverized in liquid nitrogen, transferred to a 12 X 75-mm polystyrene snap cap tubes (Becton Dickinson Labware, Lincoln Park, NJ), homogenized with an Ultra-Turrax tissumizer (Tekmar Company, Cincinnati, OH) and transferred to 1.5 microfuge tubes. One hundred microliters of chloroform were added per ml RNAzol followed by vortexing for 15 sec. After standing at 4 C

RNA was glyoxylated following the procedure of Pederson and Davis (37) and fractionated by electrophoresis through 1% agarose gels. Quantification of the total RNA was confirmed by electrophoretic separation and staining of the gel with ethidium bromide. RNA was transferred to Biotrans nylon membranes (ICN, Irvine, CA) and analyzed by hybridization with IGFBP cDNA probes labeled with [32P] by random priming. Specific activities of 5 x 108 cpm/Vg DNA were typically achieved. Rat IGFBP-1(19) cDNA was provided by Guck T. Ooi, NIH. The rat IGFBP-2 cDNA (14), provided by Matthew M. Rechler and Alexandra L. Brown, NIH, and corresponded to nucleotides 502-1087 from the coding region. IGFBP-3 (27, 38) was synthesized from rat Sertoli cell mRNA as described below. SKHEP poly (A)+ RNA was a generous gift from Gary D. Shipley, Oregon Health Sciences University (Portland, OR) and was derived from a human adenocarcinoma cell line. Polymerase chain reaction amplification of Sertoli cell IGFBP-3 mDNA IGFBP-3 cDNA was synthesized by the polymerase chain reaction using a cDNA template derived from rat Sertoli cell poly (A)+ RNA and the following primers: 5': 5'-TTGAATTCAACACCACTGAGTCTGAGGA-3' (containing ECORI restriction site) and 3': 5' CCTAGGGTTGGTGTCATAGCCTGGCAAT-3' (containing BAMHl restriction site) (27, 38) synthesized in the Molecular Biology DNA Core facility, Department of Molecular Genetics, Biochemistry and Microbiology, University of Cincinnati Medical School. Specifically, 1 jug rat Sertoli cell poly (A)+ RNA was added to a reaction mixture containing the following: 1 mM dCTP, dATP, dGTP, and dTTP, 200 U reverse transcriptase (Bethesda Research Laboratories, Gaithersburg, MD), 40 U RNasin (Promega Corporation, Madison, WI), IX PCR buffer (Gene Amp Kit, Perkin-Elmer-Cetus, Norwalk, CT), and 20 pmol 3'-primer in a total volume of 20 fd. The reaction mixture was heated for 10 min at 23 C, 60 min at 42 C, and 10 min at 95 C, and placed on ice. For amplification step, 80 n\ lOx PCR buffer, 25 pmol 5'-primer, and 2 U Taq DNA polymerase (Perkin-Elmer-Cetus) were added to the reaction mixture. Thirty cycles (one cycle: 1 min at 92 C, 2 min at 55 C, 2 min at 72 C) were carried out using a thermal cycler (Perkin-Elmer-Cetus). An approximately 440 base pair fragment was isolated that generated the expected size fragments of approximately 140 and 280 when restricted with HAEll. Collection of interstitial fluid The method of Turner was used for collection of interstitial fluid (39). Briefly, testes were removed from the rat and small x incision was made at the caudal pole. The gonad was placed

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2017. at 15:01 For personal use only. No other uses without permission. . All rights reserved.

SERTOLI CELL IGFBP-3

2746

Endo • 1990 Vol 127 • No 6

on a 200 fi\ pipette tip support in a 5 ml capped tube. The testis was allowed to stand at 4 C for 6 h, and then centrifuged for 15 min at approximately 100 X g. Approximately 3-4 /xl fluid were collected per testis, diluted in Hanks' balanced salt solution at a 1:3 dilution, and frozen at —20 C. ^

Ligand blotting Samples of conditioned media (2 ml) were concentrated in 30,000 mol wt cutoff Centricons (Amicon, Danvers, MA) and were analyzed by sodium dodecyl sulfate (SDS)-poly aery lamide gel electrophoresis under nonreducing conditions as described by Hossenlopp et al. (40) and modified by Yang et al. (41). The proteins were transferred to 0.2 /tm pore size nitrocellulose (Scheicher & Schuell, Keene, NH) by electroblotting using a model TE50 Transphor Power-lid apparatus (Hoefer Scientific Instruments, San Francisco, CA). Binding proteins were identified by incubation with [125I]IGF-I (500,000 cpm/blot) and subjected to autoradiography as previously described. Deglycosylation using n-glycanase Samples of conditioned medium were concentrated using Centricon 30 Microconcentrators. The concentrated samples (25 n\) were incubated for 24 h at 37 C in 0.1% SDS, 0.1 M Na2HPO4 pH 8.6, 5 mM phenanthroline, 0.8% NP-40 with 0.3 U iV-glycanase (Genzyme Corporation, Boston, MA) in a final volume of 55 /*1. At the end of the reaction, an equal volume of sample buffer was added and the samples were boiled and assessed by ligand blot analysis.

Results Identity and origin of gonadal cell IGFBPs To characterize the IGFBP produced by Sertoli and peritubular cells, conditioned media were subjected to ligand blot analysis as described in the methods. Sertoli cell medium contains a predominant doublet of 38-58 kilodaltons (kDa) and less intense bands of approximately 30 and 21 kDa (Fig. 1). In contrast, the major BP elaborated by peritubular cells is the 30 kDa species, though smaller amounts of the 38-58 and 21 kDa BPs are present (Fig. 1). For comparison (Fig. 1), the profile IGF binding proteins from different fluid compartments within the rat are shown. Adult rat serum contains predominantly the 38-58 kDa species (42-44), day 22 rat serum predominantly a 26-28 kDa species and interstitial fluid contains a 30 K moiety similar in size to the peritubular cell conditioned medium sample. Based on this size comparison and with other reports in the literature, the larger molecular weight doublet in Sertoli cell conditioned medium is characteristic of IGFBP-3 (19, 42), the 30 kDa band, IGFBP-2 (19, 42) and the 21 kDa band, a recently described carrier protein isolated by Mohan et al. (16). To further characterize the identity of the binding proteins synthesized by peritubular and Sertoli cells, the