Vol. 131, No. 6 Printed in U.S.A.

Characterization Mouse Testis*

of Epidermal

Growth

B. RADHAKRISHNAN, BANKOLE RICHARD P. DIAUGUSTINE, AND

0. OKEt, VASSILIOS PAPADOPOULOS, CARLOS A. SUAREZ-QUIAN

Factor

in

Department of Anatomy and Cell Biology, Georgetown University School of Medicine (B.R., B.O.O., V.P., C.A.S.-Q.), Washington, DC 20007; Hormone and Cancer Workgroup, National Institutes of Health, National Institute of Environmental Health Sciences (R.P.D.), Research Triangle Park, North Carolina 27709 ABSTRACT Considerable evidence exists to suggest that epidermal growth factor (EGF) influences spermatogenesis directly. The tissue source of this EGF, however, is not yet clear. In this study we examine whether the testis itself can serve as a source of EGF. Gel filtration fractions of acid extracted testes exhibited the ability to displace “‘I-EGF from testis membranes. The testicular fractions containing the ‘Y-EGF displacement activity coeluted within the same range as those of submandibular gland (SG) fractions containing mature EGF and nrepared in an identical fashion. Next, we employed specific ant&era probes to investigate first, whether the testis synthesizes this EGF displacement activity and second, to determine the cell distribution of the testicular EGF. Two types of antisera probes were employed: 1) commercially available antisera to mature EGF (EGF,), i.e. the 6,000 M, peptide, and 2) polypeptide specific antisera to the C-terminus of the EGF precursor (EGF,), i.e. the 140,000 M, integral membrane molecule which exhibits seven EGF-like repeats in addition to the EGF,. Metabolic labeling of testis with ““S-methionine was performed, followed bv immunonrecinitation with the anti-EGF, antisera. Parallel . _ studies using kidney and SG were used as positive controls. Fluoro-

grams exhibited a prominent band at M, 140,000 for testis and kidney, corresponding to the EGF,. There was, in addition, a M, 50,000 band present for the testis. In SG, a band at M, 6,000, corresponding to EGF,, in addition to bands at M, 21,000 and 46,000 were observed also. Immunoblotting of testis, kidney, and SG membrane preparations with the specific antisera to either the EGF, or EGF, also resulted in identifying the EGF, at M, 140,000, as well as other lower mol wt bands. Preadsorption of anti-EGF, antisera with excess EGF, eliminated all of the specific bands that were immunoblotted. Peroxidase immunocytochemistry of testis, kidney, and SG was also performed using the specific antisera to either EGF, or EGF,. EGF, and EGF, staining in SG and kidney was identical to previously published results in which the distribution of EGF, in these tissues was established. In testis, EGF, immunostaining showed positive results in Sertoli cells, pachytene spermatocytes and round spermatids. In contrast, EGF, immunostaining was limited to pachytene spermatocytes and round spermatids. These results suggest that the testis must now be included in the list of tissues capable of synthesizing EGF,. Specifically, EGF, synthesis appears limited to the post meiotic germ cells. (Endocrinology 131: 3091-3099,1992)

E

EGF, without diminishing testosterone and FSH, and correlated with over a 50% decreaseof epididymal spermatozoa; in turn, the sialoadenectomy effects were specifically reversed by daily administration of EGF, (5). More recently, these studies have been corroborated in part (6, 7), although sialoadenectomy did not appear to lead to as great an effect on epididymal spermatozoa as previously observed (6). Further, in experimentally induced diabetic male mice, in which plasma insulin and EGF levels were abolished and infertility ensued, EGF administration was required to reverse the infertility, whereasinsulin alone was capable only of reversing the diabetes (8). Moreover, although the mechanismby which EGF, exerts an effect on spermatogenesisis not yet clear, several studies suggeststrongly that the effects are mediated via the testicular somatic cells. In Sertoli cells in vitro, for example, EGF, has been shown to affect androgen binding protein (9), lactate (lo), and inhibin (11, 12) production, as well as ornithine decarboxylase activity (13), and to both increase and alter polarized transferrin secretion (14). In Leydig cells, EGF, also affects androgen production (15-19). Consistent with these functional studies was the demonstration by immunocytochemistry that both Leydig cells and Sertoli cells in rodent, nonhuman primates, and human testesexhibit the

PIDERMAL growth factor (EGF) is a polypeptide of 53 amino acids that was first isolated and purified from the submandibular glands of male mice (1). The mature EGF (EGF,) is a proteolytic processed polypeptide of the EGF precursor (EGF,), an integral membrane protein of 140,000 M,

that

exhibits

eight

EGF-like

repeats

in

addition

to

the

EGF, (2, 3). Besidesthe submandibular gland, other tissues, including the kidney and the prostate, also produce EGF,, albeit whether processing of the EGF, to EGF, occurs in these tissuesis still not clear (3, 4). Further, although the normal physiological action of EGF, remainsobscure,several biological effects have been attributed to it, including growth and differentiation effects on cells in vitro and inhibition of gastric secretion (2, 3). EGF has been implicated also in the regulation of spermatogenesis. For example, ablation of the submandibular gland (SG) (sialoadenectomy) in adult male mice led to a marked decrease in circulating levels of immunoreactive Received April 9, 1992. Address all correspondence Suarez-Quian, Department of University School of Medicine, DC 20007. * Supported by NIH Grant t Recipient of a fellowship

and reprint requests to: Dr. Carlos A. Anatomy and Cell Biology, Georgetown 3900 Reservoir Road, N.W., Washington, HD-23484 (to C.A.S.-Q.). from the Rockefeller Foundation. 3091

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 06:59 For personal use only. No other uses without permission. . All rights reserved.

EGF

3092

IN TESTIS

EGF receptor (20-23). Taken together, the above studies suggestthat EGF, does indeed play a role in regulating spermatogenesisand that a possible tissue source of this EGF,, is likely to be the SG. However, not all investigators agree that sialoadenectomy leadsto a diminution in plasma EGF, levels (24); the majority, if not all, of mouse submandibular gland EGF, was reported to be released via an exocrine mechanism into the saliva (25). Additional evidence in support of a non endocrine role for SG EGF,, are several studies which describe other tissuesourcesof EGF,, although in these tissuesEGF,, levels are much reduced in comparison to those of the SG (reviewed in 2, 3). Nevertheless, the low concentration in these tissues may be overcome by EGF, acting in an autocrine/paracrine fashion and not undergoing a marked systemic dilution. The mouse and human testis are tissuesin which immunoreactive EGF was observed (26, 27). More recently, the presence of EGF, measured by a specific RIA for rat EGF, was demonstrated in the vitamin A synchronized rat testis (28). In the present study, we investigate further the presence of testicular EGF using different methodologies. We employ specific antisera probes raised against either the EGF, or the EGF,, from mouse to determine whether the testis produces EGF, aswell asto investigate the cellular distribution of EGF, and EGF, in the testis. Our results suggestthat the EGF, is produced in the testis, specifically in the more mature germ cells residing in the adluminal compartment of the bloodtestis barrier. Materials

and Methods

Animals Adult male BALB/c mice were viz testis, kidney, and SG, used in in the Research Resources Facility chow adlibitunl,and maintained

used for collection of tissue specimens, this study. These animals were housed at Georgetown University, fed rodent in a 12-h light, 12-h dark cycle.

Antibodies Two rabbit polyclonal antisera (lot no. 91-0193, Collaborative Research, Bedford, MA; lot no. 69F-4802, Sigma Chemical Co., St. Louis, MO) raised against EGF, from mouse were used in immunoprecipitation, immunoblotting, and immunocytochemical studies. The Collaborative Research antisera, diluted 1:8, is capable of neutralizing 2.5 Kg receptor grade EGF as measured by Ouchterlony. Five tenths milliliters of a 1:lOOO dilution of the Sigma antisera was able to bind at least 40% of 225 picograms of iodinated EGF with a specific activity of 100 wCi/ mg. In addition, a polyclonal rabbit antisera (25-75-4A), raised against a synthetic peptide (MilliGen/BioSearch, San Rafael, CA) corresponding to residues 1183-1201 (C-terminal sequence) of the intracellular domain of the mouse EGF,, (29), was also employed in similar methodology.

Membrane

preparations

Membranes from testis, liver, and SG were prepared as described previously (21) with modifications. The tissues were homogenized in buffer A (10 rnM Tris-HCI, 2 mM EGTA, 0.25 M Sucrose, pH 7.4) using a Polytron homogenizer (Brinkmann Industries, Westbury, NY) with three, IO-set pulses. The final pellet, after a three-step centrifugation process (800 X g for 10 min; 10,000 X g for 15 min; 100,000 X g for 45 min), was suspended either in buffer B [50 rnM Tris-HCl, 5 mM EGTA, 1 mM phenylmethylsulfonyl fluoride (I’MSF), pH 7.41 for immunoblotting studies, or in buffer C (50 mM Tris-HCI, 1 rnM PMSF, 1 mM N-

Endo. 1992 Vol 131. No 6

ethylmaleimide, pH 7.4) for receptor binding assays. Protein concentrations of the membrane preparations were by the method of Bradford (30) using BSA as a standard. Membranes from the kidney were prepared as described previously (31).

Displacementof EGF with testicular extract Testes were collected from ten adult mice and used for EGF extraction as described for SG (32). The extract was chromatographed at 4 C using a Bio-Gel P-10 column (1.7 cm x 42 cm) in HCI-NaCI buffer (0.15 M NaCl in 0.05 N HCl). One-milliliter fractions were collected and their absorbance at 280 nm determined using an LKB. Ultrospec II spectrometer (LKB Biochrom Ltd., Cambridge, England). Twelve one-ml fractions eluting after the void volume were pooled, lyophilized, and the resulting residues reconstituted with 500 ~1 buffer D (100 mM HEI’ES, 118 mM NaCl, 1.2 mM MgS04, 5 mM KCl, 8 mM dextrose, 0.1% BSA, pH 7.4). A competitive binding assay was performed by modification of assays described for liver membranes (33) to determine if the testicular extract contained any EGF displacement activity. Freshly prepared microsomal membranes of the testis were diluted in buffer D to a final concentration of 4 mg/ml. The assay was performed by incubating 2 nM ‘?EGF (Collaborative Research, Bedford, MA; SA, 93-124 bCi/Fg) with or without 100 ~1 resuspended eluates and 400 pg testis membrane proteins in a total volume of 200 ~1. A control assay was performed by incubating 2 nM “‘1-EGF with or without increasing concentrations (0.01 nM-600 nM) of unlabeled EGF with 400 rg testis membranes in a total incubation volume of 200 ~1. The incubation for both these assays were carried out at 10 C for 2 h. The reaction was stopped by centrifugation in a microfuge, the pellet was washed with buffer D at 4 C and the radioactivity counted in a y-counter (LKB Wallac, Turku, Finland).

Metabolic labelingand immunoprecipitation Testes, kidneys, and SG were first incubated in 2 ml Dulbecco’s modified eagle medium deficient in cysteine and methionine (Sigma Chemical Co.) either at 32 C or at 37 C, respectively, for 2 h. The medium was replaced with fresh Dulbecco’s modified eagle medium deficient in cysteine and methionine containing ‘?-methionine and ‘?Zcysteine (300 pCi/ml) (Tran?-label, ICN Biomedicals; SA, 1115 Ci/ mmol) and the incubation was carried out for an additional 4 h. Tissues were collected by gentle centrifugation and then homogenized in buffer E (100 mM NaCl, 20 rnM Tris-HCI, 1 mM EDTA, 0.5% NP-40, 0.5% Na deoxycholate, pH 7.4). The tissue homogenates were ultracentrifuged at 100,000 x g for 45 min at 4 C. To 700 ~1 supernatant, containing solubilized membrane associated proteins as well as crude cytosol, 20 ~1 1% rabbit immunoglobulin G (IgG) and 200 ~130% Protein A-sepharose (Pierce, Rockford, IL) were added and incubated for 15 min on ice followed by centrifugation at 12,500 X g for 5 min at 4 C. This “preclearing” procedure was repeated twice. To 125 ~1 supernatant obtained after the final centrifugation, 6 ~1 anti-EGF, antisera were added and incubated for 1 h on ice. At the end of the incubation, 50 ~1 30% Protein A were added and incubated for an additional 30 min. The immunoprecipitate was recovered after centrifugation at 12,500 X g for 3 min at 4 C and washed three times with buffer E and once with buffer F (50 mM Tris-HCI, 150 mM NaCl, 1 mM EDTA, 1 mM I’MSF, pH 7.4). The pellet was suspended in 50 ~1 sodium dodecyl sulfate (SDS)-sample buffer and incubated for 3-5 min at 100 C. Protein A was pelleted and the supernatant was subjected to either 7.5% or 15% SDS-polyacrylamide gel electrophoresis (PAGE) (34) and the immunoprecipitated proteins detected by fluorography.

Immunoblotting To corroborate the presence of EGF,, solubilized membrane proteins (20-30 pg) of testis, kidney, and SG were separated in 7.5% gel by SDSPAGE (34) and proteins Western blotted to nitrocellulose membrane using a Bio-Rad Mini Trans Blot electrophoretic transfer cell (Bio-Rad Laboratories, Richmond, CA). Immunoblotting was performed using either rabbit antisera raised against EGF, (Sigma or Collaborative Research) or a rabbit antisera to the EGF, (both antibodies were used at 1:300 dilution) (35). Detection of specific bands was performed by

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 06:59 For personal use only. No other uses without permission. . All rights reserved.

EGF incubating with peroxidase-conjugated goat lowed by substrate-chromogen mixture. Control for the anti-EGF,, antisera were performed by blotting with antisera preadsorbed with mature

IN TESTIS

antirabbit antibody folimmunoblotting studies repeating the immunoEGF,.

Immunocytochemistry

rated by gel filtration chromatography is shown in Fig. 1. A displacement curve obtained with increasing concentrations of unlabeled EGF also showed maximum displacement of about 40% (data not shown). Metabolic

Immunocytochemistry was performed as described recently (Oke, B. O., and C. A. Suarez-Quian, manuscript submitted). Adult male mice were anesthetized with pentobarbital (20 mg/kg), SC and fixed by perfusion with Bouin’s fluid through the ascending aorta for 15 min. Tissues were removed and further fixed at 4 C for an additional 24 h. The tissues were dehydrated in ascending grades of ice-cold ethanol (70%, 90%, and 99%) and at room temperature in absolute ethanol for 1 h each. The tissues were infiltrated in 50% polyester wax (BDH Ltd Poole, England) (36) (vol/vol in ethanol) for 1 h and 90% polyester wax for 30 min at 38 C. The infiltrated tissues were transferred to plastic embedding dishes containing 90% polyester wax and rapidly chilled on ice. The solidified tissue blocks were stored at 4 C until sectioned. Sections of 5 pm thickness were cut with a microtome at 4 C and allowed to dry at the same temperature. The sections were dewaxed and hydrated in descending grades of ethanol (lOO%, 90%, and 70%) for 3-5 min each. Before performing the immunocytochemistry, sections were treated first with 70% ethanol (vol/vol) containing 1% lithium carbonate (vol/vol) for 5 min, then with 300 rnM glycine solution for 5 min, and finally washed with 20 rnM Tris-HCl containing 0.1% BSA, pH 7.4. Streptavidin-biotin-peroxidase immunostaining was carried out using Histostain-SP kits (Zymed Laboratories, South San Francisco, CA) as described previously (21). Either the antisera to EGF,, (Sigma or Collaborative Research; diluted 1:200, 1:500, 1:1600 for testis, kidney, and SG, respectively) or the antisera to the EGF, (diluted 1:1600, 1:200, 1:500 for testis, kidney, and SG, respectively) were used as primary antibodies. Briefly, endogenous peroxidase activity was blocked by a 45-set treatment with periodic acid solution and endogenous avidin and biotin were blocked using the commercially available avidin and biotin blocking solutions (Zymed) per manufacturer’s instructions, followed by blocking nonspecific antibody binding with 10% nonimmune goat serum. Tissues were incubated at 37 C in a moist chamber with either normal rabbit serum, anti-EGF, antisera, or anti-EGF,, antisera, and immunostained using biotinylated goat antirabbit antibody, streptavidin-peroxidase, and substrate-chromogen mixture. The sections were counterstained with hematoxylin. The sections were examined with a Zeiss Axiophot microscope, fitted with a Planapo 63x, 1.4 N.A. objective using an 80A blue filter and images recorded on Kodacolor 100 ASA film. Photographs were processed by a professional photographic laboratory (CPI, Bethesda, MD).

labeling and immunoprecipitation

The presence of EGF displacement activity in testicular extracts prompted us to determine if EGF, is synthesized in testis. To this end, metabolic labeling and immunoprecipitation were carried out with antisera to EGF,. Fluorograms of immunoprecipitates of “S-labeled testis tissue exhibited a prominent band at M, 140,000 comparable to that of kidney, the tissueused as a positive control (Fig. 2). There were also lower molecular weight bands for testis and kidney, a M, 50,000 band and a M, 97,000 band, respectively. SG, the other tissue used as a positive control, exhibited the EGF, band at M, 6,000 in addition to bands at M, 46,000 and 21,000. The different bands reported here in the different tissueswere observed consistently in the four experiments performed. Immunoblotting

Evidence that EGF,, is synthesized in testis in the form of a high molecular weight precursor was examined further by immunoblotting. Testicular membrane preparations immunoblotted with the antisera to EGF, gave rise to a high M, band of about 140,000, corresponding exactly to that of a kidney membrane preparation (Fig. 3). In membrane preparations of SG, a very faint band at M, 140,000 could also be discerned, but the most intense band was in the region of M, 120,000. Identical results were obtained in four different experiments using different membranepreparations from the

Results

EGF activity in testicular

extract

The presence of EGF displacement activity in testis was examined by subjecting testicular acid extracts to gel filtration chromatography, pooling the fractions, and using these in a binding assay using testis membranes as an EGF receptor source. The procedure employed to prepare and partially purify the testicular EGF displacement activity was identical to the original procedure of Savage and Cohen (32) used to partially purify EGF,,, from submandibular glands. In the present investigation, testesand submandibular glands were processedin parallel and the acid extracts were subjected to identical chromatographic separation. The testicular EGF displacement activity, capable of displacing “‘I-EGF from testis membranesin the binding assay, coeluted within the samerange asthe SG EGF,, that was similarly prepared. The elution profile and the EGF displacement activity (approximately 40%) from one of three testicular acid extracts sepa-

a0

40 Volume

120

(ml)

FIG. 1. Absorbance profile of testis extract and displacement of ““IEGF from testis membrane EGF receptors by testis extract. Acid extracted testis was chromatographed on a Bio-Gel P-10 column (32). One milliliter fractions were collected and absorbance at 280 nm was determined. Only every fourth data point has been shown on the profile. Pooled fractions were lyophilized and used in the radioligand assay as described in Materials and Methods. The displacement activity is shown in the stippled area corresponding to the pooled fractions used.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 06:59 For personal use only. No other uses without permission. . All rights reserved.

EGF IN TESTIS

3094

Mrx10-3

K

T

r x10-~)

S

I

97-

Endo. 1992 Vol131 *No 6

T

K

s

-46

46-

-21 -14 -6

30’-

-3.4 21-

-2.3

144. Immunoblotting with anti-EGF, antisera. Results of membrane preparations of testis (T), kidney (K), and SC: (S) subjected to 7.5% SDS-PAGE and immunoblotted with antisera against EGF, are shown in panel A. Specific bands were detected at M, 140,000 for T and K, and at 95,000 and in the 95,000-140,000 range for S. Results of immunoblots after neutralization of the anti-EGF, antisera with excess EGF, are shown in panel B. Only the M, 70,000 band for testis remains, although the intensity of the band is much reduced.

FIG.

2. ““S-Methionine labeling and immunoprecipitation with antiEGF,,, antisera. Testis (T), kidney (K), and SG (S) were labeled with S-methionine in vitro and the processed tissue homogenates were immunoprecipitated with anti-EGF, antisera. The immunoprecipitates were subjected to either 7.5% (T and K) or 15% (S) SDS-PAGE. The gels were treated with 2,5-diphenyloxazole/dimethyl sulfoxide, dried, and fluorographed at -70 C for 8 days (T), 13 days (K), or 28 days (S). Specific bands were detected at M, 140,000 and 50,060 for T, at 140,000 and 97,000 for K, and at 46,000, 21,000, and 6,000 for S.

FIG.

intensity of this band was much reduced. Identical results were obtained in four separate experiments. Immunostaining

B,$x~O-~I

T

K

3. Immunoblotting with anti-EGF, antisera. Membrane preparations of testis (T), kidney (K), and SG (S) were subjected to 7.5% SDS-PAGE and immunoblotted using an antisera against a C-terminal sequence of EGF,,. Specific bands were detected at Mr 140,000 for T, K, and S, as well as at lower Mr for S.

FIG.

three tissues. When the anti-EGF,, antisera was used for immunoblotting, membrane preparations of testis and kidney exhibited high molecular weight bands in the same region as with antiEGF, antisera (Fig. 4). There was, in addition, a prominent band of M, 70,000 for testis. SG membrane preparations exhibited a prominent band at M, 95,000 and several bands in the region of M, 95,000-120,000. When immunoblotting was repeated with the anti-EGF, antisera which had been incubated with excess EGF,, all the bands described above were absent except for the M, 70,000 band, although the

The same anti-EGF, and anti-EGF, antisera used for immunoblotting were also used in immunocytochemistry studies. First, the immunostaining characteristics of the different antisera were examined in SG and kidney, two tissues in which EGF immunolocalization has been extensively characterized, to validate the usefulness of these antisera probes as specific markers of testicular EGF, and EGF,. In the SG, EGF, immunostaining was prominent in the granular convoluted tubule (GCT) cells (Fig. 5A). Immunostaining with the anti-EGF, antisera also gave rise to positive staining in the same cells, but the intensity of staining was much less compared to the results obtained with the anti-EGF, antisera (data not shown). In the kidney, positive staining with the anti-EGF,,, antisera was characteristically prominent and specific in the distal convoluted tubules and the straight segments of the distal tubules; further, the immunostaining appeared diffused throughout the cytoplasm (data not shown). Immunostaining with the anti-EGF, antisera resulted in positive staining of the same cells, but in this case the staining appeared specifically localized to the plasma membrane and was confined to the apical border of the cells (Fig. 5, B, C). Immunostaining of testis with either the anti-EGF, or antiEGF, antisera was examined as a function of the cycle of the seminiferous epithelium. Results revealed that the two antisera gave rise to two distinct immunostaining patterns, respectively (Fig. 6). Whereas positive results with the antiEGF, antisera were only obtained in germ cells, immunostaining with the antisera raised against the EGF, molecule gave rise to immunostaining in both the Sertoli cells and germ cells. These latter results were obtained with the two commercially available antisera to EGF,. Although staining of the interstitium was observed in some sections, this staining was interpreted as background noise. This interpretation

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 06:59 For personal use only. No other uses without permission. . All rights reserved.

-

*..

SOurn,

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 06:59 For personal use only. No other uses without permission. . All rights reserved.

FIG:. 6. EGF,, and EGF, immunostaining in testis. EGF, (panels A-C’) and EGF,,, (panels D-F) distribution as a function of different stages of the cycle of the seminiferous epithelium (reman numerals) are illustrated. Examples of diplotene (D) and pachytene (P) spermatocytes, round spermatids (HI and Sertoli cells (S) within the seminiferous epithelium are labeled. The thick arrows indicate the positive immunostaining in the (;olgi complex area of the germ cells. In C. orru&wods indicate the positive reaction product in the characteristic, spherical Golgi apparatus of diplotene spermatocytes. representing the culmination of meiutic prophase. Note that cells at metaphase (m) lack pusitive staining. In E and F, Sertoli cells 1s) are clearly positive for EGF,,,. Magnification: A--F, 7OUx.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 06:59 For personal use only. No other uses without permission. . All rights reserved.

EGF

IN TESTIS

is based on the results that interstitial staining was present in unperfused testis in which normal rabbit serum was used ascontrol. One likely possibility of this enhanced background is that red blood cells may lyse in the unperfused testis, releasing their abundant peroxidase activity, which in turn may overwhelm, the peroxidase blocking step in our protocol. In addition, prominent immunostaining of EGF, in germ cells was observed in the Golgi apparatus of pachytene spermatocytesand round spermatidsthroughout the cycle of the seminiferous epithelium, but no staining was detected in cells undergoing meiotic division or in elongated spermatids. Using the anti-EGF,, antisera, diffused EGF, immunostaining was readily observed in Sertoli cells, as well as in the Golgi apparatus of pachytene spermatocytesand round spermatids. Discussion

In the present investigation we provide evidence that the testis, like the SG and kidney, is likely to synthesize EGF,, an integral membrane protein of M, 140,000. The molecular massof the testicular EGF, is an agreement with the molecular massof the EGF,, reported for kidney in several studies (4, 37, 38). Further, additional specific protein bands of M, 50,000, 97,000, and 46,000 were observed in testis, kidney, and SG, respectively. The lower molecular weight bands in the kidney and SG have been suggestedto represent partially processedintermediates derived from the EGF, (3, 37, 38). Similarly, the M, 50,000 band in the testis may indicate an intermediate polypeptide of the EGF. Failure to detect the EGF, in the metabolic labeling experiment would appear to be at odds with the chromatography results presented in Fig. 1, in which testicular EGF, displacement activity coeluted with submandibular gland acid extracted EGF,,,. One possible explanation of these results is that processingof testicular EGF, to EGF,, may be impaired in culture. Alternatively, it is possible that EGF, processing in the testis is minimal and not detectable with immunoprecipitation/fluorography. For example, as discussed below, the processedEGF, may be rapidly internalized by Sertoli cells and not be available for immunoprecipitation. Becauseboth the anti-EGF, and anti-EGF, antisera have proven effective in identifying these molecules in female reproductive tissues (29, 39), the antisera were employed also to determine the specific cellular distribution of EGF, and EGF, in the testis. First, however, we performed extensive immunolocalization studies with the antisera probes to ascertain their usefulnessfor immunocytochemistry. In SG, as observed in previous studies (40), EGF, immunostaining was prominent in GCT cells, the cell site known to produce and store EGF,, (40, 41). In contrast, the immunostaining intensity of EGF, was negligible. This observation is consistent with the rapid processingof the EGF, to the EGF, peptide (2,3). In kidney, both anti-EGF, and anti-EGF, antiseragave rise to a strong positive reaction in the distal straight and convoluted tubule cells. These results are identical to results of others which have demonstrated the distribution of EGF positive cells in the kidney (40, 43). In addition, the immunostaining pattern detected with the anti-EGF, antisera

3097

was consistent with patterns observed for membrane-associated proteins. Our principal finding is that germ cells immunostain only for EGF, using the antisera probe specific for EGF,, whereas EGF, itself was localized in both germ cells and Sertoli cells using the antisera probe raised against EGF,. These results suggestthat the germ cellscontained within the seminiferous epithelium are the site of de novo synthesisof EGF,. Although in situ hybridization methodology with a complementary DNA (cDNA) probe to mouseEGF needsto be performed to corroborate our findings, we wish to emphasize that our conclusion is basedon a specific antisera probe which identifies the EGF, molecule, an integral membraneprotein. That the EGF, molecule was specifically immunolocalized within the Golgi complex compartment of pachytene spermatocytes and round spermatidslendsfurther support to our suggestion that the germ cells are the site of synthesis of EGF, in testis. In the kidney, for example, ultrastructural immunolocalization studies demonstrated positive staining in vesiclesof the trans Golgi complex in cells of the thick ascending limb of Henle and distal convoluted tubule, i.e. in the cells where it is produced (43). Interestingly, nerve growth factor-like proteins have also been detected in the samepopulation of germ cells that exhibit EGF, immunoreactivity (44). At the present time, however, we do not have an explanation for the intense EGF, immunostaining in Sertoli cells rendered by two different antisera. Since the EGF-EGF receptor system is the prototypical example of both ligand and receptor being internalized and degraded in lysosomes,we did not expect that, at steady state, internalized ligand would be immunostained to the extent that we observed in our study. Future studies will examine whether Sertoli cells in vivo processEGF differently from other cells. One possible mechanism of EGF action within the seminiferous epithelium is the recently describedjuxtacrine form of cell-cell interaction. In this pathway a membraneanchored growth factor precursor binds to its receptor on adjacent cells leading to transduction of a signal (45-47). That partially processedEGF, is biologically active could support this model (48). Interestingly, we have demonstrated previously that the EGF receptor is present exclusively in Sertoli cells within the seminiferous epithelium and that receptor distribution extends from the basal to the adluminal compartments of the blood-testis barrier (20, 21, 23). This localization was puzzling, since the EGF receptor is known to be distributed in the basolateraldomain of other polarized epithelial cells (49, 50). The present findings that EGF, is present in germ cells residing on the adluminal border of the blood-testis barrier may help explain our previous EGF receptor immunolocalization. That is, germ cell EGF, may serve as the ligand source for the apically localized EGF receptor in Sertoli cells. The results presentedherein that the testis is a site of EGF, production are consistent with previous studies demonstrating immunoreactive EGF in the testis. Using RIA, Byyny et al. (26) first reported the presenceof immunoreactive EGF in the mouse testis. Similar results were observed in human seminalplasma,as well asin testicular extracts (27). Whereas this immunoreactive EGF was observed to be immunologi-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 06:59 For personal use only. No other uses without permission. . All rights reserved.

EGF IN TESTIS

3098

tally and biochemically indistinguishable from a standard human EGF preparation, the EGF levels in testis were reported to be too low to account for this tissue origin (51). In this latter study, however, the method employed to prepare the tissue to measure immunoreactive EGF excluded all of the membrane bound forms of EGF,. More recently, using the vitamin A synchronized rat testis model and a specific RIA for rat EGF,, Bartlett et al. (28) provided compelling evidence that immunoreactive EGF increases in the testis after vitamin A replenishment. Significantly, higher levels of immunoreactive EGF were measured at stages IX-XIV of the cycle of the seminiferous epithelium than at other stages (28). In contrast, other studies have detected transforming growth factor-a (TGF-(u) in the testis by immunocytochemistry (52) and in cultured Sertoli cells (53), suggesting that this growth factor may be present there as well. We wish to emphasize, however, that the antisera used in the present investigation, as well as in the studies of Bartlett et al. (28), do not cross react with TGF-c~ thus, it is highly unlikely that this growth factor has been measured in the present investigation. Moreover, that EGF, has been detected in the testis does not preclude SG EGF,, from exerting an effect on Sertoli cells. Indeed, it is blood-borne EGF,, which was first demonstrated to modulate spermatogenesis directly (5-8). A means for both plasma EGF,, and germ cell EGF, to exert an effect on Sertoli cells would be if plasma EGF,,, acts directly on basolateral EGF receptors, whereas germ cell EGF, interacts with apical EGF receptors. Bartlett et al. (28) made the astute correlation that, since spermatogonial divisions occur at stages IX, XII, and XIV, their results indicating an increase in EGF,, levels during these stages after vitamin A replenishment are consistent with the possibility of EGF,, regulating spermatogonial cell division in the rat. However, other functions have been attributed to EGF,, that do not involve its well known mitogenie effect. One of these functions is immunosuppression (54,55). Interestingly, the cells in the seminiferous epithelium observed to immunostain for EGF,, pachytene spermatocytes, and round spermatids, are known to be antigenic (56). Thus, one possible role of germ cell EGF, we would like to suggest is that it may serve an immunosuppressive function in the event that the blood-testis barrier is breached. Future studies will address this role of germ cell EGF, in the control of spermatogenesis. References 1. Carpenter G, Cohen S 1990 Epidermal growth factor. J Biol Chem 265:7709-7712 2. Carpenter G, Wahl MI 1990 The epidermal growth factor family. In: Sporn MB, Roberts A (eds) Peptide Growth Factors and Their Receptors. Springer-Verlag, New York, vol 1:69-171 3. Fisher DA, Lakshmanan J 1990 Metabolism and effects of epiderma1 growth factor and related growth factors in mammals. Endocr Rev 11:418-442 4. Rail LB, Scott J, Bell GI, Crawford RJ, Penschow JD, Niall HD, Coghlan JP 1985 Mouse prepro-epidermal growth factor synthesis by the kidney and other tissues. Nature 313:228-231 5. Tsutsumi 0, Kurachi H, Oka T 1986 A physiological role of epidermal growth factor in male reproductive function. Science 233:975-977

Endo. 1992 Vol 131. No 6

6. Russell LD, Weiss T, Goh JC, Curl JL 1990 The effects of sialoadenectomy (submandibular gland removal) on testicular and epididymal parameters. Tissue Cell 22:263-268 7. Seethalakshmi L, Liu L, Kinkead T, Flares C, Carboni AA, Menon M, Davis R 1991 Epidermal growth factor (EGF): an important role in testicular function. J Androl [Suppl] 60 (Abstract) 8. Noguchi S, Ohba Y, Oka T 1990 Involvement of epidermal growth factor deficiency in pathogenesis of oligozoospermia in streptozotocin-induced diabetic mice. Endocrinology 127:2136-2140 9. Rich KA, Bardin CW, Gunsalus GL, Mather JP 1983 Age-dependent pattern of androgen-binding protein secretion from rat Sertoli cells in primary culture. Endocrinology 113:2284-2293 10. Mallea LE, Machado AJ, Navaroli F, Rommerts FFG 1986 Epiderma1 growth factor stimulates lactate production and inhibits aromatization in cultured Sertoli cells from immature rats. Int J Androl 9:201-208 11. Morris PL, Vale WW, Cappel S, Bardin CW 1988 Inhibin production by primary Sertoli cell-enriched cultures: regulation by folliclestimulating hormone, androgens, and epidermal growth factor. Endocrinology 122:717-725 12. Gonzales GF, Risbridger GP, deKretser DM 1989 Epidermal growth factor increases inhibin synthesis by isolated segments of iat seminiferous tubules. J Endocrinol 123:213-219 13. Smith FF. Tres LL. Kierszenbaum AL 1987 Ornithine decarboxylase act&y during rat spermatogenesis ir7 viuo and ill vitro: selective effect of hormones and growth factors. J Cell Physiol 133:305312 14. Spaliviero JA, Handelsman DJ 1991 Effect of epidermal and insulin-like growth factors on vectorial secretion of transferrin by rat Sertoli cells in vitro. Mol Cell Endocrinol 81:95-104 15. Hsueh AJW, Welsh TH, Jones PBC 1981 Inhibition of ovarian and testicular steroidogenesis by epidermal growth factor. Endocrinology 108:2002-2004 16. Welsh TH, Hsueh AJW 1982 Mechanism of the inhibitory action of epidermal growth factor on testicular androgen biosynthesis in vitro. Endocrinology 110:1498-1506 17. Lloyd CE, Ascoli M 1983 On the mechanisms involved in the regulation of the cell-surface receptors for human choriogonadotropin and mouse epidermal growth factor in cultured Leydig tumor cells. J Cell Biol 96:521-526 18. Verhoeven G, Cailleau J 1986 Stimulatory effects of epidermal growth factor on steroidogenesis in Leydig cells. Mol Cell Endocrinol 47:99-106 19. Sordoillet C, Chauvin MA, de Peretti E, Morera AM, Benahmed M 1991 Epidermal growth factor directly stimulates steroidogenesis in primary cultures of porcine Leydig cells: actions and sites of action. Endocrinology 128:2160-2168 20. Suarez-Quian CA, Dai M, Onoda M, Kris RM, Dym M 1989 Epidermal growth factor receptor localization in the rat and monkey testes. Biol Reprod 41:921-932 21. Suarez-Quian CA, Niklinski W 1990 Immunocytochemical localization of the epidermal growth factor receptor in mouse testis. Biol Reprod 43:1087-1097 22. Foresta C, Caretto A, Varotto A, Rossato M, Scandellari C 1991 Epidermal growth factor receptors (EGFR) localization in human testis. Arch Androl 27:17-24 23. Radhakrishnan B, Suarez-Quian CA 1992 Characterization of epidermal growth factor receptor (EGFR) in testis, epididymis and vas deferens of non-human primates. J Reprod Fertil in press 24. Tokida N, Shinoda I, Kurobe M, Tatemoto Y, Mori M, Hayashi K 1988 Effect of sialoadenectomy on the level of circulating mouse epidermal growth factor (mEGF) and on the reproductive function in male mice. J Clin Biochem Nutr 5:221-229 25. Murphy RA, Watson AY, Metz J, Forssmann WG 1980 The mouse submandibular gland: an exocrine organ for growth factors. J Histochem Cytochem 28:890-902 26. Byyny RL, Orth DN, Cohen S 1972 Radioimmunoassay of epiderma1 growth factor. Endocrinology 90:1261-1266 27. Elson SD, Browne CA, Thorburn GD 1984 Identification of epidermal growth factor-like activity in human male reproductive tissues and fluids. I Clin Endocrinol Metab 58:589-594 28. Bartlett JM, Spit&i-Grech J, Nieschlag E 1990 Regulation of

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 06:59 For personal use only. No other uses without permission. . All rights reserved.

EGF IN TESTIS insulin-like growth growth factor in 127:747-758 29

30

31.

32.

33.

34. 35.

36. 37.

38.

factor I and stage-specific levels of epidermal stage synchronized rat testes. Endocrinology

Brown CF, Teng CT, Pentecost BT, DiAugustine

RP 1989 Epider-

mal growth factor precursor in mouse lactating mammary gland alveolar cells. Mol Endocrinol 3:1077-1083 Bradford MM 1976 A rapid and sensitive method for the quantitation of microgram quantities of protein using the principle of protein dye binding. Anal Biochem 72:248-254 Inui K-I, Okano T, Takano M, Kitazawa S, Hori R 1981 A simple method for the isolation of basolateral plasma membrane vesicles from rat kidney cortex. Biochim Biophys Acta 647:150-154 Savage CR, Cohen S 1972 Epidermal growth factor and a new derivative: rapid isolation procedures and biological and chemical characterization. J Biol Chem 247:7609-7611 Kashimata M, Hiramatsu M, Minami N 1988 Sex difference in epidermal growth factor receptor levels in rat liver plasma membrane. Endocrinology 122:1707-1714 Laemmli UK 1970 Cleavage of structural proteins during assembly of the head of bacteriophage T4. Nature 227:680-685 Johnson DA, Gautsch JW, Sportsman JR, Edler JH 1984 Improved technique utilizing nonfat dry milk for analysis of proteins and nucleic acid transferred to nitrocellulose. Gene Anal Tech 1:13-18 Steedman HF 1957 Polyester wax: a new riboning embedding medium for histology. Nature 4574:1345 Breyer JA, Cohen S 1990 The epidermal growth factor precursor isolated from murine kidney membranes. J Biol Chem 267:1656416570

Lakshmanan

J, Salido EC, Lam R, Barajas L, Fisher DA 1990

Identification of pro-epidermal growth factor and high molecular weight epidermal growth factors in adult mouse urine. Biochem Biophys Res Commun 173:902-911 39.

DiAugustine RP, Petrusz P, Bell GI, Brown CF, Korach KS, Mclachlan JA, Teng CT 1988 Influence of estrogens on mouse uterine epidermal growth factor precursor protein ribonucleic acid. Endocrinology 122:2355-2363

40.

and messenger

Salido EC, Lakshmanan J, Shapiro LJ, Fisher DA, Barajas L 1990

Expression of epidermal growth factor in the kidney and submandibular gland during mouse postnatal development: an immunocytochemical and in situ hybridization study. Differentiation 45:38-43 41. Gresik EW, Gubits RM, Barka T 1985 In situ localization of mRNA for epidermal growth factor in the submandibular gland of the mouse. I Histochem Cvtochem 33:1235-1240 42. Fisher bA, Salido Ee, Barajas L 1989 Epidermal growth factor

and kidney. 43.

Annu

Rev Physiol

Salido EC, Lakshmanan

51:67-80

J, Fisher DA, Barajas L 1991 Immuno-

cytochemical localization in adult human kidney.

of epidermal growth factor prohormone Clin Res 39:llZA Abstract 44. Onoda M, Pflug B, Djakiew D 1991 Germ cell mitogenic activity is associated with nerve growth factor-like protein(s). J Cell Physiol 149:536-543 45. Wong ST, Winchell LF, McCune BK, Earp HS, Teixido J, Massague J, Herman B, Lee DC 1989 The TGF-a precursor expressed on the cell surface binds to the EGF receptor on adjacent cells, leading to signal transduction. Cell 56:495-506 46. Anklesaria P, Teixido J, Laiho M, Pierce JH, Greenberger JS, Massague J 1990 Cell-cell adhesion mediated by binding of membrane-anchored transforming growth factor alpha to epidermal growth factor receptors promotes cell proliferation. Proc Nat1 Acad Sci USA 87:3289-3293 47. Massague J 1990 Transforming growth factor-a: a model for membrane-anchored growth factors. J Biol Chem 265:21393-21396 48. Mroczkowski B, Reich M, Chen K, Bell GI, Cohen S 1989 Recombinant human EGF precursor is a glycosylated membrane protein with biological activity. Mol Cell Biol 9:2771-2778 49. Dunn WA, Connolly TP, Hubbard AL 1986 Receptor-mediated endocytosis of epidermal growth factor by rat hepatocytes: receptor pathway. J Cell Biol 102:24-36 50. Brandlit AW, Adamson ED, Simons K 1991 Transcytosis of epidermal growth factor. J Biol Chem 266:8560-8566 51. Hirata Y, Uchihashi M, Hazama M, Fujita T 1987 Epidermal growth factor in human seminal plasma. Horm Metab Res 19:3537 52. Teerds KJ, Rommerts FFG, Dorrington JH 1990 Immunohistochemical detection of transforming growth- factor-a in Leydig cells during the development of the rat testis. Mol Cell Endocrinol69:RlR6 53. Skinner MK, Takacs K, Coffey RJ 1989 Transforming growth factor-a gene expression and action in the seminiferous tubule: peritubular cell-Sertoli cell interactions. Endocrinology 124:845-854 54. Acres RB, Lamb JR, Feldman M 1985 Effects of platelet-derived growth factor and epidermal growth factor on antigen-induced proliferation of human T-cell lines. Immunology 54:9-16 55. Koch JH, Fifis T, Bender VJ, Moss BA 1984 Molecular species of epidermal growth factor carrying immunosuppressive activity. J Cell Biochem 25:45-59 56. O’Rand MG, Romrell LJ 1977 Appearance of cell surface auto- and isoantigens during spermatogenesis in the rabbit. Dev Biol 55:347358

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 06:59 For personal use only. No other uses without permission. . All rights reserved.