Immediate cell signal induced by laminin in rat Sertoli cells

Matrix Biology 19 Ž2000. 11]18

Immediate cell signal induced by laminin in rat Sertoli cells Anna Tarantab, Anna Teti c , Mario Stefanini a , Angela D’Agostinoa,U a

Department of Histology and Medical Embryology, Uni¨ ersity ‘La Sapienza’ of Rome, ¨ ia A. Scarpa 14, 00161 Rome, Italy b Istituto Dermopatico dell’Immacolata, Rome, Italy c Department of Experimental Medicine, Uni¨ ersity of L’Aquila, L’Aquila, Italy Received 5 May 1999; received in revised form 27 July 1999; accepted 27 July 1999

Abstract Rat Sertoli cells in primary culture have been studied for their ability to respond to extracellular matrix macromolecules by increases of wCa2q x i . We observed that cells seeded on glass coverslips, loaded with the intracellular Ca2q indicator fura-2, responded to laminin, but not to fibronectin, with an immediate wCa2q x i raise, with a peak followed by a prolonged plateau. wCa2q x i increases were dependent upon Ca2q influx across the plasma membrane and Ca2q release from intracellular Ca2q pools. Ca2q influx was inhibited by extracellular Ca2q removal by EGTA, and by treatment with La3q, or with the L-type voltage operated Ca2q channel blocker, nifedipine. Ca2q release from intracellular Ca2q storing organelles, was inhibited by the microsomal Ca2q-ATPase blocker thapsigargin. Responses were mimicked by synthetic peptides carrying the Arg]Gly]Asp adhesion sequence, but not by the control Arg]Gly]Glu-containing peptide, in which aspartic acid was replaced by glutamic acid. Laminin-dependent wCa2q x i increases were down-regulated by the follicle-stimulating hormone. However, this occurred only when cells were not subjected to homotypic cell]cell contact, and responded to the hormone with a significant wCa2q x i elevation. These results indicate that laminin may regulate Sertoli cells by intracellular signals that perturb Ca2q homeostasis. This role may be related to an effect exerted by the seminiferous epithelium basement membrane on the regulation of spermatogenesis. Q 2000 Elsevier Science B.V.rInternational Society of Matrix Biology. All rights reserved. Keywords: Sertoli cells; Laminin; Ca2q; Cell signal; FSH

1. Introduction Sertoli cells are the somatic elements of the seminiferous tubules forming a columnar monolayer spanning the seminiferous epithelium from the basis to the lumen. Sertoli cells are in contact with germ cells, which in turn are organized in several layers corresponding to sequential differentiative stages

U

Corresponding author. Tel.: q39-06-4976-6803; fax: q39-06446-2854.

ŽRussell, 1977a,b; Bardin et al., 1988.. In the mammalian testis, Sertoli cells provide nutritional and mechanical supports to germ cells, and regulate spermatogenesis by extensive surface relationships with developing germ cells, and by secreting a Kq-enriched fluid in the tubular microenviroment ŽMeans et al., 1980; Bardin et al., 1988; Griswold, 1998.. Sertoli cells and spermatogonia, the most undifferentiated germ cell stage, are attached to a basement membrane, which, in turn, is surrounded by myoid cells ŽPalombi et al., 1988.. Sertoli cells contribute to the production of the basement membrane, both in

0945-053Xr00r$ - see front matter Q 2000 Elsevier Science B.V.rInternational Society of Matrix Biology. All rights reserved. PII: S 0 9 4 5 - 0 5 3 X Ž 9 9 . 0 0 0 5 4 - 2

12

A. Taranta et al. r Matrix Biology 19 (2000) 11]18

vivo and in vitro, secreting abundant extracellular matrix ŽECM. components ŽDym, 1994.. The basement membrane, in turn, plays a primary role in the control of the seminiferous epithelium function ŽRavindranath et al., 1996., since it: Ži. provides a mechanical support for the somatic and the undifferentiated germ cells; Žii. influences Sertoli cell morphology; Žiii. contributes to the blood]testis barrier; and Živ. represents the vehicle for factors regulating the spermatogenesis. Immunocytochemical studies on tissue sections have demonstrated that the basement membrane immediately adjacent to the seminiferous epithelium is composed of laminin, type IV collagen, heparan sulfate proteoglycan, and nidogenrentactin ŽTryggvason, 1993.. Fibronectin has not been detected in the basement membrane, but is localized in the connective tissue external to the myoid cells, implying that the seminiferous epithelium does not directly interact with this extracellular matrix component. Increasing evidence indicates that ECM is not an inert structure that surrounds cells, but it accomplishes a number of specific functions that play crucial roles in the regulation of cell activities. The interaction of cells with the ECM is mediated by cell surface receptors belonging to the integrin superfamily. It is well known from the literature ŽHynes, 1992. that integrin receptors not only serve as structural receptors connecting the extracellular matrix to the cytoskeleton in order to attach the cell to the substrate, but also function as signaling receptors that lead to changes in second messenger molecule levels, determine modification of cell pH and cytosolic free calcium concentration wCa2q x i 4 , and induce activation of tyrosine kinases andror gene expression. Thus, the ability of ECM to influence growth, differentiation and other cell functions is likely to be related to their effects on signaling pathways inside the cells. With the aim of understanding the mechanism, whereby the basement membrane constituents influence Sertoli cell function, we examined whether laminin, a major basement membrane macromolecule, caused prompt cell signal transfer into rat Sertoli cells in culture. To this end we performed wCa2q x i studies in Sertoli cells obtained from 18 to 21-day-old rat testis, using the fura-2 microfluorometry method. We observed that laminin, but not fibronectin, elicited a rapid wCa2q x i raise, dependent upon both Ca2q release from intracellular Ca2q storing organelles, and Ca2q influx across the plasma membrane. Such intracellular signaling pathway was regulated by factors that support spermatogenesis via indirect action on Sertoli cells, i.e. the follicle-stimulating hormone ŽFSH. produced by the pituitary gland.

2. Experimental procedures 2.1. Materials Minimum Essential Medium ŽMEM., reagents and sterile plastic ware for cell culture were from Flow Laboratory ŽIrvine, UK.. Fura-2 acetoxymethyl ester ŽFura-2rAM. was from Molecular Probes ŽEugene, OR, USA.. Ionomycin and fatty acid-free Bovine Serum Albumin ŽBSA. were from Calbiochem ŽLa Jolla, CA, USA.. Laminin Žfrom basement membrane of Engelbreth]Holm]Swarm mouse sarcoma. and fibronectin Žfrom rat plasma. were from Sigma Chemicals Co. ŽSt. Louis, MO, USA.. All other reagents were from Sigma Chemical Co. ŽSt. Louis, MO, USA.. 2.2. Cell cultures Sertoli cells were isolated from 18 to 21-day-old Wistar rats. Briefly, testes were removed, cleaned from the tunica albuginea and reduced in small fragments. Fragments were digested with 99% trypsin 1:250, 1% DNAase, for 30 min at 328C, then washed Ž2 = . with Hank’s solution. Enzymatic digestion was repeated twice, then the cell suspension was centrifuged at 15 =g for 3 min. Pellets were resuspended 1:10 in MEM, then cells were seeded in 3.5-cm Petri dishes containing 24-mm diameter round glass cover slips, and incubated at 328C in a water-saturated atmosphere containing 95% air and 5% CO 2 . After 4 days, germ cells were eliminated by gentle treatment with hypotonic solution Ž20 mM Tris]HCl, pH 7.2. for 1 min. Sertoli cells were then incubated for a further 24 h. All the experiments were performed on day 5 of culture. 2.3. Buffer Bathing medium through all experiments was a modified Krebs]Henseleit solution ŽKHH., buffered with 20 mM N-2hydroxyethylpiperazine-N92-ethanesulfonic acid ŽHEPES. and containing 1.25 mM Ca2q and 0.2% BSA. Ca2q-free solutions were obtained on addition of ethyleneglycol-bis- Žb-aminoethyl-ether . N, N9-tetraacetic acid ŽEGTA.. 2.4. Measurement of [Ca 2 q ] wCa2q x i of Sertoli cells was measured by dual wavelength fluorescence of cells loaded with the Ca2q-sensitive intracellular probe, fura-2 ŽGrynkiewicz et al., 1985.. Cells were loaded with 3 mM fura-2rAM in serum-free medium, at 378C for 60 min. Measurements were performed at 340 and 380 nm excitation

A. Taranta et al. r Matrix Biology 19 (2000) 11]18

13

wavelengths, with an AR-CM microfluorometer ŽSpex Industries, Inc. Edison, NJ, USA. connected with a Diaphot TMD inverted microscope ŽNikon, Corporation, Tokyo, Japan. equipped with a Nikon CF= 40 objective. Emissions were collected by a photomultiplier carrying a 505 nm cut-off filter and recorded by an ASEM Desk 2010 computer, which automatically calculated real-time 340r380 rations. Calibration of the signal was obtained at the end of each observation by adding 15 mM ionomycin to saturate the dye to maximal fluorescence, followed by 7.5 mM EGTA plus 60 mM Tris]HCl, pH 10.5, to release Ca2q from fura-2, and obtain minimal fluorescence. wCa2q x i was calculated according to previously described formulas ŽGrynkiewicz et al., 1985.. 2.5. Statistics Data are presented as average " S.E. Statistical analysis was performed by analysis of variance ŽANOVA.. P- 0.05 was conventionally considered to indicate statistical significance.

3. Results In order to study a potential modulation of wCa i by extracellular matrix proteins, cultured rat Sertoli cells have been exposed to various concentrations of fibronectin and mouse-laminin-1 at various intervals of time. As shown in Fig. 1 and Table 1, although fibronectin had no effect on wCa2q x i , laminin rapidly increased intracellular ion concentration. Laminin produced a prompt increase in cytosolic Ca2q followed by a plateau in 90% of cells tested, whereas in 10% of cells we observed an initial rapid decrease in wCa2q x i , followed by an increase and a plateau. Responses were concentration-dependent. The dose]response curve constructed for laminin was biphasic. Maximal stimulation was observed at a concentration of laminin of 12 mgrml, which increased wCa2q x i up to three times over the basal level ŽFig. 2., whereas at higher doses wCa2q x i responses significantly declined. To examine the mechanism inducing laminin-de2q x

Table 1 Effect of inhibitors of Ca2q influx on wCa2q x i of Sertoli cells treated with laminin a Conditions

Lanthanum 10y3 M Nifedipine 10y5 M EGTA 3 mM

n

6 5 4

wCa2q xi nM Basal

Laminin Ž12 mgrml.

90 " 17 175 " 16 53 " 17

96 " 17 182 " 18 41 " 15

Values are means " S.E. wCa2q x i , intracellular free Ca2q concentration. a

Fig. 1. Effect of fibronectin and laminin on cytosolic free Ca2q concentration wCa2q x i 4 of single Sertoli cells. The output of a microspectrofluorometer has been calibrated as wCa2q xi as described in Section 2. The traces represent the effect of 100 mgrml fibronectin Ža. and 12 mgrml laminin Žb.. Arrows indicate the addition of agonist by direct dilution into the slide chamber. Fibronectin and laminin were diluted into the cell buffer from a stock solution Ž1 mgrml. immediately before use. Calibration was obtained at the end of each experiment on addition of 5 mM of the Ca2q ionophore ionomycin, followed by 7.5 mM EGTA. wCa2q x i was calculated according to Grynkiewicz et al. Ž1985.. Medium wCa2q x s 1.25 mM. All experiments were repeated 5]20 times.

pendent wCa2q x i increase we used several experimental protocols. To clarify the relevance of Ca2q entry from the extracellular environment into the cell, we removed extracellular Ca2q by the Ca2q-chelating agent EGTA. Fig. 3a and Table 1 show that the wCa2q x i raise in response to laminin was significantly inhibited in Sertoli cells incubated in Ca2q-free KHH. This indicates that Ca2q influx across the plasma membrane contributes to the cellular response to laminin. This issue is further confirmed by experiments performed in Sertoli cells treated with lanthanum, a competitive blocker of both receptor and voltage-gated Ca2q channels. Similar to EGTA, lanthanum ŽFig. 3b, Table 1. was active in inhibiting wCa2q x i increase induced by laminin. Sertoli cells possess voltage-gated Ca2q channels ŽVOCCs. sensitive to the dihydropyridine derivatives nifedipine and nicardipine, and to the v-conotoxin ŽD’Agostino et al., 1992.. To investigate whether

14

A. Taranta et al. r Matrix Biology 19 (2000) 11]18

Fig. 2. Concentration-dependent laminin-induced wCa2q x i elevations in Sertoli cells. Illustrated is the size of maximal wCa2q x i elevations triggered by laminin at the doses indicated on abscissa. Results are mean " S.E. from at least 15 independent determinations for each point. Note that the effect of laminin was biphasic, with maximal activation at 12 mgrml, followed by a decline at higher doses.

VOCCs played a role in the control of Ca2q homeostasis by laminin, we treated Sertoli cells with nifedipine. Fig. 4 and Table 1 show that similar to EGTA and lanthanum, nifedipine reduced the wCa2q x i response, thus indicating a contribution of VOCCs activity to the laminin-dependent wCa2q x i raise. In order to verify whether cytosolic Ca2q changes were due not only to the opening of membrane divalent ion channels but also to the discharge of intracellular Ca2q stores, Sertoli cells were treated with thapsigargin, an inhibitor of microsomal Ca2q-ATPase. Fig. 5 and Table 2 show that thapsigargin significantly inhibited the response of Sertoli cells to laminin, thus indicating that wCa2q x i increases were due to both, Ca2q influx across the plasma membrane, and Ca2q release from intracellular Ca2q pools. Laminin contains several adhesion sequences that may mediate interaction with the surface receptors. Among these, the best known adhesion motif, Arg]Gly]Asp ŽRGD., contained in the isoform used in this study, although criptic, may play a physiological role upon exposure under certain conditions Ži.e. proteolytic digestion.. To investigate the role of this motif in the regulation of Sertoli cell Ca2q homeostasis, we used two synthetic peptides, the active Gly]Arg]Gly]Glu]Ser]Pro ŽGRGDSP. peptide, and the control Gly]Arg]Gly]Asp]Ser]Pro ŽGRGESP. peptide. Exposure to the GRGDSP sequence induced a significant wCa2q x i increase, shown in Fig. 6. Similar to integral laminin, the GRGDSP peptide produced a rapid wCa2q x i peak followed by a plateau in 90% of cells tested, whereas the peak was preceded by an initial wCa2q x i decrease in 10% of cells. The responses

Fig. 3. Mechanism inducing laminin-dependent wCa2q x i elevations. Ža. A single cell was treated with 3 mM EGTA to chelate extracellular Ca2q, then 12 mgrml laminin was added. Žb. A single cell was treated with 10y3 M La3q, then with 12 mgrml laminin. Note that both traces reveal inhibition of laminin-induced wCa2q x i increases.

were dose-dependent ŽFig. 7a., with maximal stimulation observed at 24 mgrml. In contrast, treatment of the cells with the control GRGESP peptide, in which, aspartic acid was replaced with glutamic acid, did not result in significant wCa2q x i elevations vs. basal levels ŽFig. 7b.. We next investigated whether cell signal induced in Sertoli cells by laminin was subjected to regulation by

Fig. 4. Role of voltage-dependent Ca2q channels ŽVOCCs. in laminin-induced wCa2q x i raise. The role of VOCCs has been examined by treatment with the dihydropyridine derivative nifedipine, known as L-type VOCC blocker. Nifedipine was added at concentration of 10y5 M, then was followed by 12 mgrml laminin.

A. Taranta et al. r Matrix Biology 19 (2000) 11]18

15

Table 2 Effect of thapsigargin on wCa2q x i of Sertoli cells treated with laminin a n

Conditions

y6

Thapsigargin 2x10 a

M

Values are means " S.E. wCa

10 2q x

i,

wCa2q xi ŽnM. Basal

Peak

Plateau

Laminin

108 " 19

389 " 68

124 " 17

183 " 44

intracellular free Ca

2q

concentration.

FSH. Sertoli cells are known to express functional FSH receptors coupled to the adenylate cyclaser cAMP pathway. In Fig. 8a Sertoli cells were treated with 200 nM ovine-FSH immediately before addition of laminin. We observed that FSH significantly increased wCa2q x i in isolated Sertoli cells and down-regulated the response of the cell to laminin. By contrast, when FSH was given to Sertoli cells clustered in large cell aggregates, no wCa2q x i increases, nor regulation of the response to laminin was observed Žnot shown.. Finally, when FSH was added to Sertoli cells after exposure to laminin, it failed to further modify wCa2q x i . 4. Discussion Sertoli cells are known to produce ECM macromolecules. In vitro studies have demonstrated that Sertoli cells synthesize laminin and type IV collagen, but not fibronectin ŽDym, 1994; Richardson et al., 1995.. In vivo, while 1- and 3-laminin-isoforms ŽVirtanen et al., 1997. and type IV collagen are immunolocalized in the basement membrane underlying the seminiferous epithelium, and are in contact with both Sertoli cells and spermatogonia, fibronectin is localized in the connective tissue external to the myoid cells, and appears to be secreted by this cell type in a polarized manner ŽThompson et al., 1995.. In this study we present evidence indicating that, according to this tissue organization of the two macromolecules,

Fig. 5. Effect of intracellular Ca2q store depletion on laminin-induced wCa2q x i elevations. A single Sertoli cell was treated with endoplasmic reticulum Ca2q-ATPase inhibitor thapsigargin. This produced a transient wCa2q xi elevation. When wCa2q x i stabilized, 12 mgrml laminin were added.

Fig. 6. Effect of GRGDSP and GRGESP synthetic peptides; on Sertoli cell wCa2q x i . Single cells were challenged with 12 mgrml GRGDSP Ža. and GRGESP Žb.. Note that while GRGDSP produced a wCa2q x i elevation similar to that of laminin, the effect of GRGESP was almost undetectable.

laminin, but not fibronectin, shows the ability of eliciting signal transduction into Sertoli cells. Laminin, in fact, prompts a wCa2q x i increase in Sertoli cells cultured on glass, characterized by a biphasic Žmaximal activation at 12 mgrml followed by a decline. dosedependent pattern. This effect is similar to that observed in other cell types treated with other components of ECM ŽPaniccia et al., 1993., indicating that the concentration of the macromolecule in the matrix is crucial to the cellular activity to be modulated. Although the addition of this protein in soluble form cannot be constructed as equal to recognition of matrix protein in the basement membrane, the use of this experimental approach does not subtract from the results, since receptors for the protein may be expressed even in the dorsal cell membrane and guarantee interaction with the soluble laminin. In addi-

16

A. Taranta et al. r Matrix Biology 19 (2000) 11]18

Fig. 7. Concentration-dependent wCa2q x i increases elicited by GRGDSP peptide. Ža. maximal size of wCa2q x i elevations triggered in Sertoli cells by addition of GRGDSP at doses indicated in abscissa. Žb. Effect of GRGESP at concentration of 12 mgrml. Results are mean " S.E. of at least 15 independent experiments per each treatment.

tion, it has been formerly reported that laminin has a EGF-like activity, including the capability of trigging intracellular calcium mobilization ŽPanayoutou et al., 1989., which fits with our findings in Sertoli cells. The effect elicited by laminin on Sertoli cells wCa2q x i appears to be due to two canonic Ca2q transport pathways, Ca2q influx across the plasma membrane, and Ca2q release from intracellular Ca2q storing organelles. These two pathways are generally activated by phospholipase C-associated receptors, which metabolize membrane phosphoinositol lipids to yield the Ca2q-release signal, inositol, 1,4,5 trisphosphate ŽIP3 . and the protein kinase C-activating element, diacylglycerol. A substantial body of evidence indicates that IP3 not only directly activates endoplasmic reticulum IP3-receptor Ca2q channels, but also indirectly stimulates Ca2q influx across the plasma membrane to provide Ca2q for refilling the depleted Ca2q pool. The shape of the wCa2q x i traces recorded in this study upon stimulation with laminin Ža peak followed by a plateau., as well as the inhibitory roles of extracellular Ca2q depletion by EGTA, and competitive inhibition of receptor operated Žby lanthanum.. and voltage operated Ca2q channels Žby lanthanum and nifedipine. of laminin-induced wCa2q x i raise, indicate a relevant role of this influx pathway in the cell function. Furthermore, great inhibition of laminin-induced wCa2q x i raise following thapsigargin treatment, indicates that intracellular Ca2q store depletion may be secondarily responsible for gating of plasma membrane Ca2q channels as well. Laminin is a large glycoprotein Ž Mr 800 000. which contains multiple active sites for receptor recognition. The isoform. used in this study Žlaminin-1., which, together with the isoform-3, faces the Sertoli cell

surface ŽVirtanen et al., 1997. includes the RGD motif as a criptic site which, however, can be exposed under certain conditions Ži.e. proteolitic digestion. ŽRuoslahti et al., 1987; Aumailley et al., 1990.. The RGD sequence is recognized by most integrin receptors. Synthetic peptides, based on this sequence, mimic some activities of the intact RGD-containing ECM macromolecules, producing competitive binding to the receptor. Therefore, in order to investigate whether the laminin RGD motif was relevant for laminin function in Sertoli cells, we challenged the cells with the synthetic GRGDSP peptide, and the control GRGESP peptide. The RGD-containing peptide mimicked the action of integral laminin, whereas the RGE-containing peptide showed a significantly lower activity, indicating: Ži. that the action of laminin on wCa2q x i signaling in Sertoli cells may be RGD-dependent; and Žii. that replacement of a single amino acid interferes with such response. However, a number of issues remains to be clarified, i.e. whether or not the RGD sequence supports other cell functions in cultured Sertoli cells and whether or not laminins lacking the RGD sequence or containing this sequence as

Fig. 8. Effect of follicle-stimulating hormone ŽFSH. on response elicited by laminin. Ža. An isolated Sertoli cell was treated with 200 nM FSH. In this circumstance, FSH triggered rapid wCa2q x i transient elevation. The cell was then treated with 12 mgrml laminin. Note that laminin did not further increase wCa2q x i . Žb. An isolated Sertoli cell was treated with 12 mgrml laminin followed by 200 nM FSH.

A. Taranta et al. r Matrix Biology 19 (2000) 11]18

a criptic motif affect Sertoli cell calcium homeostasis. Based on our results, however, we cannot exclude that the RGD sequence affects Sertoli cell independently of its insertion into the laminin protein. Like many other ECM adhesion molecules, laminin binds cell surface receptors of the integrin superfamily. There is an increasing number of laminin receptors, including the b 1 subunit Žthat may be associated to a 1 , a 2 , a 3 , a 6 or a 7 ., and the b 4 subunit Žeither associated to a v or a 6 .. Furthermore, Sertoli cells are known to express laminin-binding proteins ŽLustig et al., 1998.. By affinity chromatography and w 125 I x-laminin binding to Sertoli cell plasma membrane, Davis et al. Ž1991. identified and partially characterized six binding proteins in the 25 000] 110 000 Mr range. Laminin binding proteins with Mr 67 000 and 110 000 are phosphorylated. The 67 000, 45 000 and 36 000 binding proteins react with antibodies, which stain the basolateral plasma membrane of Sertoli cells in vivo. The 67 000 peptide has the property of a peripheral membrane protein, which functions as a galactoside-binding protein and recognizes not only laminin but also elastin. Our study does not provide insights into the receptorrbinding protein involved in laminin-dependent signaling in Sertoli cells. Preliminary results obtained in our laboratory indicate that the a 2 as well as the a 3 integrin subunits are not responsible for laminin signaling, due to the fact that specific anti-a 2 and -a 3 subunit antibodies fail to inhibit the wCa2q x i increase produced by laminin Ždata not shown.. Further studies are in progress in our laboratory to address this important issue. Sertoli cells are regulated in vivo by two hormones, the testosterone released by the Leydig cells of the testis, and the FSH produced by the pituitary gland. FSH is recognized by a cell surface receptor associated to the adenylate cyclasercAMP pathway. Although several studies have been performed in order to establish whether FSH receptor is also coupled to the phospholipase Crintracellular Ca2q pathway, this issue remains, to date, unsolved. In our study we observed that FSH was never active in inducing immediate modifications of wCa2q x i in Sertoli cells organized in monolayers, in which not only cell-to-substrate, but also homotypic cell]cell interactions were established. On the contrary, when Sertoli cells were isolated, with plasma membrane borders not perfectly sticking to the culture plate, and in the absence of cell]cell contacts, FSH stimulated an increase of wCa2q x i similar to that induced by other Ca2q mobilizing hormones. The interpretation of these results is at present only speculative. It is well established that both cell-to-ECM and cell]cell interactions may modulate specific cell functions ŽJaconi et al., 1991., therefore, the functional state of Sertoli cells when they

17

interact with other Sertoli cells may be different from their state when isolated. This is supported by recent results indicating that FSH fails to stimulate immediate wCa2q x i increases in monolayers of Sertoli cells attached to coverslips ŽD’Agostino et al., 1992., whereas it induces wCa2q x i transients in freshly prepared Sertoli cells maintained in cell suspension ŽGorczynska et al., 1991.. Most interestingly, however, pretreatment of isolated Sertoli cells with FSH, not only produced immediate wCa2q x i raise, but also down-regulated the response of the cells to laminin. In clustered cells, in which FSH failed to induce modifications of wCa2q x i , it also failed to modulate laminin-dependent signaling. Laminin, in turn, caused down-regulation of FSH-induced wCa2q x i raise in isolated cells. These results indicate that: Ži. the effect of laminin may be subjected to regulation by a hormone crucial for supporting spermatogenesis, thus, suggesting a relevant role for Sertoli cell]basement membrane interaction in reproduction; and Žii. the effect of FSH on Sertoli cells may greatly be dependent upon the specific interactions of these cells with both the substrate and the other cells of the seminiferous epithelium. In conclusion, our study provides an insight to the mechanism underlying Sertoli cell]basement membrane interaction and signaling. Thus, evidence that Ca2q mobilization accompanies recognition of laminin is provided. Such a response depends upon the interaction of the cells with the RGD adhesion motif and is subjected to hormonal regulation only when homotypic cell]cell interactions are not operating. Acknowledgements This work was supported by grants from the Ministero del Lavoro e della Previdenza Sociale to M.S. and from the National Council of Research Žcontract no. 97.04285.04 to M.S... We are thankful to the National Hormone and Pituitary Program, which kindly donated to us ovine FSH, and to Mr Roberto Pezzotti for the excellent technical assistance. References Aumailley, M., Gerl, M., Sonnenberg, A., Deutzmann, R., Timpl, R., 1990. Identification of the Arg]Gly]Asp sequence in laminin A chain as a latent cell-binding site being exposed in fragment P1. FEBS Lett. 262, 82]86. Bardin, C.W., Yan Cheng, C., Musto, N.A., Gunsalus, G.L., 1988. The Sertoli cell. In: Knobil, E., Neill, J. ŽEds.., The Physiology of Reproduction. Raven Press Ltd, New York, pp. 933]973. D’Agostino, A., Mene, ` P., Stefanini, M., 1992. Voltage-gated calcium channels in rat Sertoli cells. Biol. Reprod. 46, 414]418. Davis, C.M., Papadopulos, U., Jlia, M.C., Yamada, Y., Klinman, H.K., Dym, M., 1991. Identification and partial characterization of laminin binding proteins in immature rat Sertoli cells. Exp. Cell Res. 193, 262]273.

18

A. Taranta et al. r Matrix Biology 19 (2000) 11]18

Dym, M., 1994. Basement membrane regulation of Sertoli cells. Endocr. Rev. 15, 102]115. Gorczynska, E., Handelsman, D.J., 1991. The role of calcium in follicle-stimulating hormone signal transduction in Sertoli cells. J. Biol. Chem. 266, 23739]23744. Griswold, M.D., 1998. The central role of Sertoli cells in spermatogenesis. Semin. Dev. Biol. 9, 411]416. Grynkiewicz, G., Poenie, M., Tsien, R.Y., 1985. A new generation of Ca2q indicators with greatly improved fluorescence properties. J. Biol. Chem. 260, 3440]3450. Hynes, R.O., 1992. Integrins: versatility, modulation and signaling in cell adhesion. Cell 69, 11]25. Jaconi, M.E., Theler, J.M., Schlegel, W., Appel, D., Wright, S.D., Lew, P.D., 1991. Multiple elevations of cytosolic-free Ca2q in human neutrophils: initiation by adherence receptors of the integrin family. J. Cell Biol. 112, 1249]1257. Lustig, L., Meroni, S., Cigorraga, S., Casanova, M.B., Vianello, S.E., Denduchis, B., 1998. Immunodetection of cell adhesion molecules in rat Sertoli cell cultures. Am. J. Immunol. 39, 399]405. Means, A.R., Dedman, J.R., Tindall, J.S., van Sickle, M., Welsh, M.J., 1980. Regulation of the testis Sertoli cell by follicle-stimulating hormone. Annu. Rev. Physiol. 42, 59]70. Palombi, F., Di Carlo, C., 1988. Alkaline phosphatase is a marker for myoid cells in cultures of rat peritubular and tubular tissue. Biol. Reprod. 39, 1101]1109. Panayoutou, G., End, P., Aumailley, M., Timpl, R., Engel, J., 1989. Domains of laminin with growth-factor activity. Cell 56, 93]101.

Paniccia, R., Colucci, S., Grano, M., Serra, M., Zambonin Zallone, A., Teti, A., 1993. Immediate cell signal by bone-related peptides in human osteaclast-like cells. Am. J. Physiol. 265, C1289]C1297. Ravindranath, N., Papadopoulos, V., Brooker, G., Dym, M., 1996. Rat Sertoli cell calcium response to basement membrane and follicle-stimulating hormone. Biol. Reprod. 54, 130]137. Richardson, L.L., Kleinman, H.K., Dym, M., 1995. Basement membrane gene expression by Sertoli and peritubular myoid cells in vitro in the rat. Biol. Reprod. 52, 320]330. Ruoslahti, E., Pierschbachen, M.D., 1987. New perspectives in cell adhesion: RGD and integrins. Science 238, 491]497. Russell, L., 1977a. Desmosome-like junctions between Sertoli and germ cells in the rat testis. Am. J. Anat. 148, 301]312. Russell, L., 1977b. Observations on rat Sertoli cell ectoplasmic specialisation’s in their association with germ cells of the rat testis. Tissue Cell 9, 475]498. Thompson, E.W., Blackshaw, A.W., Raychoudhury, S.S., 1995. Secreted products and extracellular matrix from testicular peritubular myoid cells influence androgen-binding protein secretion by Sertoli cells in culture. J. Androl. 16, 28]35. Tryggvason, K., 1993. The laminin family. Curr. Opin. Cell Biol. 5, 877]882. Virtanen, I., Lohi, J., Tani, T., Korhonen, M., Burgeson, R.E., Lehto, V.P., Leivo, I., 1997. Distinct changes in the laminin composition of basement membranes in human seminiferous tubules during development ad degeneration. Am. J. Pathol. 150, 1421]1431.