Biochemical characterization of integral membrane heparan sulfate proteoglycans in Sertoli cells from immature rat testis.

(35)S-Radiolabeled cultured Sertoli cells from immature rat testis were extracted with detergent and the different proteoheparan sulfate (HSPG) forms of the extract were discriminated and quantified on the basis of their high anionic charge, hydrodynamic size, lipophilic properties, susceptibility to trypsin and phosphatidylinositol phospholipase C (PI-PLC). Trypsin released 50% of total cellular HSPG corresponding to 80% of total hydrophobic HSPG. Trypsin-accessible HSPG were presumed to be integral membrane species. Trypsin-resistant HSPG, probably intracellular, distributed into non-lipophilic (37.5%) and lipophilic (12.5%) populations. Biochemical analysis of PG copurified with plasma membrane confirmed the existence of hydrophobic HSPG integrated into this structure. Among hydrophobic HSPG accessible to trypsin, 35% were PI-PLC released and radiolabeled by [(3)H]inositol indicating that about one third of integral membrane HSPG were intercalated into the plasma membrane through a phosphatidylinositol anchor (glypican type). PI-PLC-resistant forms represented HSPG inserted into the membrane through a hydrophobic segment of the core protein (syndecan type). No lipophilic PG was present in other cell compartments (culture medium, cell periphery, extracellular matrix). (125)I-Iodinated hydrophobic HSPG were deglycanated and submitted to SDS-polyacrylamide gel electrophoresis. In the glypican family, a core protein (64--65 kDa) was detected, whereas in the syndecan family, bands of 60 and 68 kDa were observed which may correspond to self-association of different core proteins. In Sertoli cell, specific functional attributes of different integral membrane HSPG forms remain to be investigated.

Activation of protein kinase C increases proteoglycan synthesis in immature rat Sertoli cells.

In order to determine the signal transduction pathways involved in the regulation of proteoglycan (PG) synthesis in immature rat Sertoli cells (SC), we have examined the effect of the tumor promoter phorbol ester PMA (phorbol myristate acetate) on [35S]sulfate and [3H]glucosamine incorporation into PG molecules neosynthesized by cultured rat SC. PMA induced a dose- and time-dependent stimulation of labeled cell-associated PG as determined by quantitative solid phase assay. The overall effect of PMA resulted from enhancement of both glycosylation and catabolism of cell PG, this latter effect leading to a drastic decrease of their residence time in the membrane. Besides these quantitative effects, activation of protein kinase C by PMA induced qualitative changes as reflected by increase in relative proportion of heparan sulfate PG (HSPG) in cell membrane PG. In light of our previous results suggesting an inverse relationship between PG synthesis and FSH responsiveness in immature rat Sertoli cells, the PMA-induced upregulation of cell membrane PG, and particularly HSPG, could constitute one mechanism involved in the repression of FSH-stimulated steroidogenesis induced by PKC activation.

Role of proteoglycans on testosterone synthesis by purified Leydig cells from immature and mature rats.

In order to characterize an involvement of proteoglycans (PG) in the regulation of Leydig cell function, we have examined the effects of para-nitrophenyl-beta-D-xyloside (PNPX), a specific inhibitor of PG synthesis and para-nitrophenyl-beta-D-galactoside (PNPG), an inefficient structural analogue, on testosterone production by purified Leydig cells from immature and mature rats, in the presence or not of various concentrations of hCG during 24 h. Whatever the age, the addition of PNPX induces a decrease of [35S] and [3H] incorporations into cell layer associated-PG; these latter being less numerous (-50 and -25%, respectively in immature and mature rat), and less sulfated (-40%) when compared to control Leydig cells. In immature Leydig cells, the inhibition of PG synthesis decreases both the basal and weakly stimulable-hCG or -(Bu)2cAMP or -LH testosterone synthesis. In mature Leydig cells, the PG inhibition has no effect on testosterone production both in the absence of hCG and in the presence of weak amounts of hCG but increases it in the presence of subsaturating hCG concentrations. Whatever the age, the inhibition of PG synthesis is ineffective in the presence of saturating amounts of either hCG or (Bu)2cAMP. These effects are maintained in the presence of MIX, PMA, but are not observed in the presence of 22R-hydroxycholesterol. Therefore, our results suggest that in rat Leydig cells, the inhibition of PG synthesis affects the signal transduction at a step distal to cyclic AMP and more precisely, the cholesterol supply to the mitochondria by acting on its cellular distribution (free and esterified cholesterol).

Sodium chlorate induces undersulfation of cellular proteoglycans and increases in FSH-stimulated estradiol production in immature rat Sertoli cells.

The functional influence of cell proteoglycan (PG) undersulfation on estradiol synthesis by immature rat Sertoli cell cultures was investigated by using sodium chlorate, an inhibitor of the active sulfate donor for sulfotransferases. The addition of sodium chlorate to 20-day-old rat Sertoli cell cultures abolished [35S]-sulfate incorporation into neosynthesized PG and consequently reduced the residence time of undersulfated PG in cell membrane. Simultaneously, follicle-stimulating hormone (FSH)-stimulated estradiol synthesis was increased by 45%. The effects of sodium chlorate upon Sertoli cell PG synthesis and steroidogenesis were not reproduced with the addition of sodium chloride. Addition of phosphodiesterase inhibitors (MIX or Ro20-1724) decreased the magnitude of the chlorate effect on FSH-stimulated steroidogenesis, suggesting that part of chlorate's effect on steroidogenesis resulted from a decrease in adenosine cyclic 3',5'-phosphate (cAMP)-specific phosphodiesterase activity. Additionally, chlorate 1) increased Sertoli cell steroidogenesis at a step located beyond cAMP (restricted to Sertoli cell cultures exhibiting moderate steroidogenic response to (Bu)2cAMP) and 2) abolished the inhibition of steroidogenesis induced by transforming growth factor-beta. These results support our previous data, which showed that alteration in PG synthesis and the consequent decrease in cell membrane PG content induce an increase in FSH-stimulated estradiol synthesis in Sertoli cell cultures. The identification of cAMP-specific phosphodiesterase activity as a signal transduction step modified by PG undersulfation suggests the possible involvement of cell PG in the regulation of phosphodiesterase activity and, therefore, of FSH responsiveness during testicular development.

Activation of protein kinase C pathway by phorbol ester results in a proteoglycan synthesis increase in peritubular cells from immature rat testis.

In cultured peritubular cells (PT) from rat testis, protein kinase C (PKC) was activated by phorbol 12-myristate 13-acetate (PMA). PMA enhanced the synthesis of proteoglycans (PG) and to a lesser extent their catabolism; the stimulation of the synthesis appeared to be due to an increase in PG protein moiety production and, at the same time, to an increase in the glycanation process as revealed by the use of an exogenous acceptor, p-nitrophenyl-beta-d-xyloside. In the presence of PMA, the molecular weight of neosynthesized PG and the length of their constitutive glycosaminoglycan chains were not modified. Moreover, the distribution of proteochondroitin sulfate and proteoheparan sulfate in medium and in cell layer remained unchanged. However, PMA reduced the sulfation level of chondroitin sulfate and heparan sulfate chains, suggesting that PKC activation resulted in an independent modulation of the sugar chain formation and of the sulfate residue transfer. PMA effect on the synthesis of hyaluronan was also determined: PMA dramatically enhanced its production by PT cells.

Inhibition of transmembrane calcium influx induces decrease in proteoglycan synthesis in immature rat Sertoli cells.

Beyond increased cAMP synthesis, calcium influx has been involved in signal transduction triggered by the gonadotropin follicle-stimulating hormone (FSH), the main regulator of Sertoli cells functions. In order to delineate a possible involvement of calcium in the regulation of proteoglycan synthesis, we have examined the effect of low-voltage-activated calcium channel blocker verapamil on both [(35)S]-sulfate and [(3)H]-glucosamine incorporation into proteoglycan molecules neosynthesized by cultured Sertoli cells from 20-day-old rats. Verapamil induced a dose- and time-dependent decrease in labeling of both secreted and cell-associated proteoglycans, as determined by quantitative solid-phase assay. This effect was mimicked by the addition of the calcium chelator EGTA, suggesting that verapamil effect resulted from the inhibition of transmembrane calcium influx. The decrease in apparent proteoglycan synthesis appeared to be attributable primarily to a lowering of the glycanation process, as shown by experiments using an exogenous acceptor for glycosaminoglycan synthesis. Moreover, verapamil induced a decrease in relative proportion of heparan sulfate proteoglycans in the cell layer. Pulse-chase kinetics demonstrated that verapamil also altered proteoglycan catabolism, leading to glycosaminoglycan retention in the cell layer and inhibiting the proteoglycan desulfation step. We conclude that intracellular calcium is essential to maintain Sertoli cell proteoglycan expression and could thus be involved in the repression of Sertoli cell cAMP-dependent syntheses such as estradiol production.

Regulation of proteoglycan and hyaluronan synthesis by elevated level of intracellular cyclic adenosine monophosphate in peritubular cells from immature rat testis.

The effects of an increase in intracellular cAMP concentration on proteoglycan (PG) synthesis by peritubular (PT) cells from immature rat testis were investigated. In the presence of dBcAMP for 72 h, the [3H]-hexosamine incorporation in secreted PG and in cell-associated PG was reduced, whereas [35S]-sulfate radioactivity was enhanced in secreted PG and not affected in cell-associated PG. Cholera toxin and IBMX, known to generate high intracellular cAMP levels, induced similar changes. Cyclic AMP did not alter PG protein moiety synthesis but enhanced PG turnover. Cholera toxin and dBcAMP profoundly modified PG characteristics: (1) Apparent molecular weight of PG was increased. (2) This was due to an increase in glycosaminoglycans (heparan sulfate (HS) and chondroitin sulfate (CS)) length. (3) The number of glycosaminoglycan chains was presumably reduced. (4) Heparan sulfate and chondroitin sulfate chains of medium and cell layer-associated PG appeared oversulfated. (5) The pattern of cell layer associated PG was modified with a decrease in HSPG and a correlative increase in CSPG. Cholera toxin and dBcAMP also dramatically stimulated hyaluronan synthesis by possible phosphorylation induced activation of hyaluronan synthase(s).

IGF-1 stimulates synthesis of undersulfated proteoglycans and of hyaluronic acid by peritubular cells from immature rat testis.

The exposure of confluent peritubular (PT) cells from immature rat testis to insulin-like growth factor-1 (IGF-1) induced a time and dose-dependent increase of [35S]-sulfate and [3H]-D-glucosamine incorporations in newly synthesized proteoglycans (PG). This increased content of PG was the result of an enhancement of PG synthesis rather than a decreased rate of degradation. IGF-1 had no effect on the molecular weight of synthesized PG nor on the nature and distribution of the constitutive glycosaminoglycan chains, both in medium and in cell layer. The stimulation of PG synthesis by IGF-1 appeared to be due, at least partially, to an increase of glycosylation processes. IGF-1 effect was mediated by the classical tyrosine kinase signalling process, since IGF-1 action on PG synthesis was abolished by genistein and tyrphostin A9, two well known tyrosine kinase inhibitors. The increase of PG synthesis was accompanied with an undersulfation of constitutive glycosaminoglycan (GAG) chains (chondroitin sulfate and heparan sulfate chains) since the [35S]/[3H] ratio was reduced by about 20-25% in presence of IGF-1. Although the mechanism of hyaluronic acid synthesis was completely different from those of other GAG, IGF-1 also dramatically enhanced its production by PT cells.

[Synthesis of proteoglycans by purified Leydig cells in immature rats and mature rats]. / Synthèse de protéoglycanes par les cellules de Leydig purifiées du rat immature et du rat mature.

In Percoll purified Leydig cells from mature rat incubated for 24 h in Ham F12/DME medium, we demonstrated that the testosterone production (6 ng/10(6) Leydig cells/24 h) is increased 11 fold by a saturating amount of hCG whereas rat Leydig cells the basal output of testosterone is 4.5 ng and increased 2.5 fold by hCG. In the sulfate-free Ham F12/DME medium used for the proteoglycan (PG) studies, the productions of testosterone are lower but the cells remain sensitive to hCG. Whatever the age, the Leydig cells synthesize labeled PG after 24 h of incubation in the presence of [3H] glucosamine and [35S] sulfate. In immature rat, the quantity of synthesized PG is 20% higher than in mature animal; the secreted PG represented respectively 70 and 85% of the immature and mature Leydig cells PG outputs. In addition, the immature Leydig cell HSPG represents 15% of the secreted PG whereas it is 5% in mature rat; irrespective of the age the associated-HSPG account for 10% of the PG. The addition of hCG improves the production of HSPG especially in the immature Leydig cells. Therefore we evidenced a rat Leydig cell synthesis of PG, especially HSPG which are likely involved in the regulation of the Leydig cell function.

Inhibition of proteoglycan synthesis induces an increase in follicle stimulating hormone (FSH)-stimulated estradiol production by immature rat Sertoli cells.

In order to define the possible involvement of proteoglycans (PG) in the regulation of Sertoli cell functions, we have examined the effect of para-nitrophenyl-beta-D-xyloside (PNPX), a specific inhibitor of PG synthesis, on follicle stimulating hormone (FSH)-dependent estradiol production by immature rat Sertoli cells. Addition of PNPX to the culture medium induced a dose-dependent inhibition of 35S-labeled PG synthesis in Sertoli cells both in the medium and the cell layer. Simultaneously there was a drastic increase in 35S-labeled secreted glycosaminoglycans. By 1 mM PNPX, syntheses of chondroitin sulfate proteoglycans released into culture medium and of heparan sulfate proteoglycans associated with the cell layer were 35% of values from untreated cells. Simultaneously, PNPX induced a twofold (mean of seven experiments, range 17-250%) enhancement of FSH (100 ng/ml)-stimulated estradiol production. In each individual experiment, there was an inverse relationship between the amplitude of PNPX-induced increase in FSH responsiveness and the FSH capability to stimulate basal estradiol production in cultured rat Sertoli cells. The effect of PNPX on FSH-stimulated aromatase activity was not mimicked by para-nitrophenyl-beta-D-galactoside, a structural analog of PNPX that has no effect on PG synthesis. The (Bu)2cAMP-stimulated estradiol synthesis was not modified in the presence of PNPX. Moreover, PNPX enhancement of FSH-stimulated estradiol synthesis disappeared when Sertoli cells were cultured in the presence of 1-methyl-3-isobutylxanthine, an inhibitor of phosphodiesterase activity. These findings suggest that inhibition of PG synthesis under PNPX conditions did not affect signal transduction steps distal to cAMP but rather decreased the phosphodiesterase activity in Sertoli cells.(ABSTRACT TRUNCATED AT 250 WORDS)

Uterine luteinizing hormone/human chorionic gonadotropin-binding sites in the early pregnant rat uterus: evidence for total occupancy in the periimplantation period.

We have investigated the presence of high affinity LH/hCG-binding sites (RLH) in crude membranes from early pregnant rats uteri. The uterine concentration of available RLH increased from day 1 to day 3 (1.3 +/- 0.2 vs. 2.8 +/- 0.4 fmol/mg protein) without a change in the affinity constant (approximately 5 x 10(10) M-1). However, unoccupied uterine RLH disappeared in the periimplantation period (days 4-6). To determine if the drop in available RLH was consecutive to their occupancy, uterine membranes were treated with acidified medium (25 mM Tris-HCl, and 5 mM MgCl2, pH 2.5) to remove endogenous ligand. The number and affinity of total (occupied plus available) RLH in acid-eluted membranes were estimated by Scatchard analysis of [125I]hCG binding and compared with those of available RLH in untreated membranes from the same uterine preparation. The uterine concentration of total RLH increased first between days 1 and 2 (2.2 +/- 0.5 vs. 4.2 +/- 0.8 fmol/mg protein), then between days 3 and 4 (4.2 +/- 0.6 vs. 6.5 +/- 0.8 fmol/mg protein), before plateauing until day 6. Thus, the reduction in the available uterine RLH in the periimplantation period is largely due to occupancy, rather than down-regulation, of RLH. The occupancy of uterine RLH 1) increased during early pregnancy (day 1, approximately 20%; days 2-3, approximately 40%; days 4-6, approximately 100%), 2) paralleled the increase in total RLH number, and 3) was probably due to pituitary LH only. However, the blastocyst itself seemed to influence uterine RLH occupancy, since available uterine RLH were detected on day 5 of pseudopregnancy. The increase in total uterine RLH as well as the perfect synchrony between their occupancy and the previously described pattern of uterine cAMP concentration during rat early pregnancy suggest that the response of uterine (and more precisely luminal epithelium) adenylate cyclase to LH (and/or related substance originating from embryo) may determine uterine receptivity for ovoimplantation and subsequent decidualization.

Changes of progesterone content of rat uterine flushings in relation to serum concentrations of progesterone during the oestrous cycle.

Uterine fluid was collected from four-day cyclic rats at each stage of the oestrous cycle and assayed for progesterone and protein content. Progesterone was determined by radioimmunoassay either after ethanol (or 2.5% NaOH) denaturation of proteins from uterine flushings ('total' progesterone) or without protein denaturation ('ether-extractable' progesterone). The amount of 'ether-extractable' progesterone in the lumen was constant from metoestrus to pro-oestrus (340 pg per uterus) but lower in oestrus (200 pg per uterus). However, 'total' progesterone content of uterine fluid was subject to cyclic variations and was highest in dioestrus (890 pg per uterus) and lowest in oestrus (350 pg per uterus), in contrast to serum progesterone which is lowest in dioestrus and highest in oestrus. Protein content of uterine flushings peaked to 780 micrograms per uterus in pro-oestrus then fell to about 140 micrograms per uterus until the end of the oestrous cycle. Changes in protein content of the lumen were followed by qualitative variations since the mean amount of 'bound' progesterone ('total' progesterone minus 'ether-extractable' progesterone) released per milligram of denatured lumen protein rose from 1.8 pmol in pro-oestrus to 18.2 pmol in dioestrus. The changes of luminal 'bound' progesterone during the oestrous cycle suggest that progesterone binding to luminal proteins could be an important modulator of progesterone action in rat uterus. Moreover, the variations in progesterone content of the lumen, irrespective of serum progesterone concentrations, are consistent with the hypothesis that progesterone synthesis occurs in the uterus.

Estrous cycle-related changes of high affinity luteinizing hormone/human chorionic gonadotropin binding sites in the rat uterus.

LH/human CG (LH/hCG) high affinity binding sites were detected in crude membrane preparations of rat uteri. There was little competition for receptor occupancy between hCG and ovine FSH (oFSH) (0.05%) and no competition between hCG and ovine PRL (less than 0.01%). No similar binding sites were detected in crude membrane preparation of heart, kidney, skeletal muscle, liver, and lung tissues. Concentrations of uterine unoccupied binding sites (RLH) were determined for each stage of the 4-day estrous cycle. The RLH were found in all preparations of metestrus uteri (n = 10) but only in some preparations from the other stages of the estrous cycle (1 of 7 on proestrus, 3 of 4 on estrus, 5 of 7 on diestrus). The concentration of uterine RLH varied throughout the estrous cycle with highest values during the metestrus (1.50 +/- 0.15 fmol/mg protein) and lowest values during the proestrus (less than 0.2 fmol/mg protein). The affinity constant for hCG of uterine RLH remained constant during the estrous cycle (about 0.8 x 10(11) M-1) and was nearly identical to that of rat ovarian receptors. On metestrus, RLH concentration appeared to be approximately 35-fold lower in the uterus than in the ovaries when expressed per mg protein (1.50 +/- 0.15 vs. 52.83 +/- 3.61 fmol/mg protein) but only 20 times lower when expressed per organ (2.2 vs. 48.3 fmol/organ). The estrous cycle-related changes of uterine RLH concentration, together with our data establishing an in vitro influence of hCG on progesterone metabolism in rat uterus, suggest that some uterine functions could be directly regulated either by LH from the pituitary or during early pregnancy by an LH-like substance originating from the embryo.

Involvement of the phospholipase C second messenger system in the regulation of steroidogenesis in small bovine luteal cells.

We have previously suggested that the interaction between luteinizing hormone (LH) and its receptor, in addition to stimulating adenylate cyclase, is able to trigger a negative regulatory signal at a step beyond cAMP synthesis (Benhaim et al. (1987) FEBS Lett. 223, 321-326). The present study was conducted to determine whether the phospholipase C system is involved in this phenomenon. Small bovine luteal cells from pregnant cows were incubated with phospholipase C, A23187, an ionophore of calcium and/or phorbol 12-myristate 13-acetate (PMA), an activator of protein kinase C (PKC), in the presence or absence of bovine luteinizing hormone or dibutyryl cyclic AMP (dbcAMP). A23187 associated with PMA was able to mimic the stimulatory effect of phospholipase C on basal progesterone production, whereas neither A23187 nor PMA alone had any effect. In the presence of high doses of LH, phospholipase C inhibited progesterone and cAMP production in a dose-dependent manner. A23187 and PMA were able to mimic the inhibition of progesterone synthesis but stimulated LH-induced cAMP accumulation. When cells were stimulated by high doses of dbcAMP, phospholipase C and A23187 but not PMA inhibited progesterone synthesis. These observations suggest that (1) phospholipase C can mimic the post-cAMP negative regulatory signal induced in vitro by high doses of LH, in the presence of an activation of PKC; (2) phospholipase C is also able to mimic in vitro the luteolytic properties of prostaglandin F2 alpha that we previously described (Benhaim et al. (1987) Prostaglandins 33, 227-239); and (3) under basal conditions or in the presence of low doses of LH, the phospholipase C system slightly stimulates steroidogenesis.

Human chorionic gonadotropin affects tissue levels of progesterone and cyclic adenosine 3',5'-monophosphate in the metestrus rat uterus in vitro.

The effect of human chorionic gonadotropin (hCG) on in vitro progesterone (P) level in the uterus was investigated by using short-term incubations of uterine tissue taken from 4-day cyclic rats at different stages of the estrous cycle. Control incubations resulted in a decrease of endogenous P content in uteri removed from rats in proestrus (PRO), estrus (EST), and metestrus (MET): -25% (n = 6), -60% (n = 6), and -45% (n = 8), respectively. The amount of P found after incubation of MET tissue in the presence of hCG was significantly higher (p less than 0.001) than that found after control incubations. The hCG effect was dose-dependent and was not observed with PRO or EST tissue. Although the mean P level found in MET tissue after incubation with 10 IU/ml hCG was not significantly different from the mean level found in unincubated tissue (1562 +/- 341 vs. 1470 +/- 174 pg/mg protein, n = 8), an obvious synthesis was observed in two experiments. It thus seems likely that observed hCG effect would involve a de novo P synthesis rather than a decrease of P catabolism. Furthermore, hCG induced a dose-dependent increase of cyclic adenosine 3', 5'-monophosphate (cAMP) uterine levels in MET tissue. N,O'-dibutyryl cyclic AMP [Bu)2 cAMP) at 5 mM induced a significative increase (p less than 0.01) of P uterine level in EST and MET tissue compared to control incubations, but had no effect on PRO tissue. Our results suggest the progressive maturation throughout the rat estrous cycle of a luteinizing hormone/hCG- and cAMP-dependent process able to regulate uterine P content.(ABSTRACT TRUNCATED AT 250 WORDS)

In vitro action of PG F2 alpha on progesterone and cAMP synthesis in small bovine luteal cells.

The action of prostaglandin F2 alpha (PG F2 alpha) on incubated small bovine luteal cells in the presence or in the absence of bovine luteinizing hormone (LH) or dibutyryl cyclic adenosine monophosphate (db cAMP) was investigated. In the absence of LH and db cAMP, PG F2 alpha stimulated progesterone synthesis at concentrations of 10 ng/ml and 100 ng/ml but had no effects at concentrations below 1 ng/ml. PG F2 alpha partially inhibited the LH or db cAMP stimulated progesterone synthesis. This inhibition was maximal for PG F2 alpha concentrations around 100 pg/ml whereas distinctly higher or lower concentrations were without effect. At the concentration of 100 pg/ml, PG F2 alpha partially inhibited the LH induced cAMP accumulation. These results demonstrate an "in vitro" action of PG F2 alpha on bovine luteal cells. They indicate that the luteolytic action of PG F2 alpha in the bovine species could involve, as already suggested for the rat, both an inhibition of the LH induced synthesis of cAMP and an inhibition of the action of cAMP.