EXPERIMENTAL
CELL RESEARCH
198,130-134
(1992)
Cross-Talk Regulation between Cyclic AMP Production and Phosphoinositide Hydrolysis Induced by Prostaglandin E2 in Osteoblast-like Cells OSAMU KOZAWA,*~’ HAFWHIKO TOKUDA,? MASAICHI MIWA,? JUN KOTOYORI,P AND YUTAKA OISO~ *Department of Biochemistry, Znstitute for Developmental Research, Aichi Prefectural Colony, Kasugai, Aichi 480-03, Japan; and fFirst Department of Internal Medicine, Nagoya University School of Medicine, Nagoya 466, Japan
In cloned osteoblast-like MC3T3-El cells, PGE, stimulated both CAMP accumulation and the formation of inositol trisphosphate (IPJ dose dependently. The CAMP accumulation showed the peak value at 5 min and decreased thereafter, whereas the IP, formation reached a plateau almost within 10 min and sustained it up to 30 min. The effect of PGEp on CAMP accumulation (E&, was 80 n&f) was more potent than that on IP, 12-G-Tetradecanoylformation (EC& was 0.8 a). phorbol-13-acetate (TPA), a protein kinase C (PKC)activating phorbol ester, reduced the PGE,-induced CAMP accumulation, whereas la-phorbol 12,13didecanoate, a PKC-nonactivating phorbol ester, had little effect on the CAMP accumulation. 1-Oleoyld-acetylglycerol, a specific activator for PKC, inhibited PGEainduced CAMP accumulation. TPA had little effect on CAMP accumulation induced by forskolin or NaF, a GTP-binding protein activator. So, the effect of TPA is presumed to be exerted at the point between the PGEl receptor and G,. On the other hand, forskolin and dibutyryl CAMP had little effect on the IP, formation stimulated by PGE,. H-7, a PKC inhibitor, enhanced the PGE,-induced CAMP accumulation in comparison with HA1004, a control for H-7. Our data suggest that PGEa regulates CAMP production through self-induced activation of PKC. These results strongly suggest that there is an autoregulatory mechanism in PGEs signaling, and PGE, modulates osteoblast functions through a crosstalk interaction between CAMP production and phosphoinositide hydrolysis in osteoblast-like cells. Q 1992 Academic Prese, Inc.
INTRODUCTION
Prostaglandins are considered to be important regulators of osteoblasts as autacoids [l, 21. Among them, PGE, is known as a potent bone resorbing agent [3]. In osteoblasts, the actions of PGE, have been thought to be 1 To whom reprint
requests should be addressed.
0014~4%?7/92 $3.00 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
mediated through the activation of adenylate cyclase [4-61. It is well known that GTP-binding proteins (G, and Gi) function as intermediaries in transmembrane signaling from the receptor to adenylate cyclase [7]. In addition, it has recently been reported that phosphoinositide (PI) hydrolysis is also stimulated by PGE, in rat osteosarcoma UMR-106 cells [S]. It is generally accepted that in response to a variety of agonists, phosphoinositides are hydrolyzed by phospholipase C, resulting in the formation of diacylglycerol and inositol phosphates. Among these products, diacylglycerol and inositol trisphosphate (IP,) serve as messengers for the activation of protein kinase C (PKC) and the mobilization of intracellular Ca2+, respectively [9, lo]. In this signaling system, as well as the adenylate cyclase-CAMP system, evidence is accumulating that GTP-binding protein(s) is involved in the coupling of the receptor to phospholipase C [ 71. In a previous report [ 111, we showed that the PGE,-induced PI hydrolysis is mediated by a pertussis toxin-sensitive GTP-binding protein in cloned osteoblast-like MC3T3-El cells which have been derived from newborn mouse calvaria [12, 131. So, it is now accepted that the effects of PGE, are mediated through both CAMP production and PI hydrolysis in osteoblasts. Recently, it has been reported that a tumor-promoting phorbol ester known as a potent stimulator of PKC inhibits the PGE,-inducedcAMP production in UMR-106 cells and mouse osteoblast-like MOB 3-4 cells [14,15]. This evidence suggests that so-called “cross-talk” between these signal transduction pathways does indeed exist in osteoblasts. However, the precise mechanism of cross-talk between these two signaling pathways of PGE2 has not yet been elucidated. In the present study, therefore, we investigated the interaction between two PGE,-induced signaling systems, CAMP production and PI hydrolysis in osteoblast-like MC3T3-El cells. Herein, we demonstrate that PGE, regulates CAMP production through self-induced activation of PKC in MC3T3-El cells. In contrast, the adenylate cyclase-CAMP signaling pathway had little effect on the PGE,-induced PI hydrolysis in these cells. These
130
PGE,-INDUCED
SIGNAL
TRANSDUCTION
results strongly suggest that PGE, modulates osteoblast functions through a cross-talk interaction between CAMP production and PI hydrolysis in osteoblast-like cells. MATERIALS
AND METHODS
Materials. myo-[2-3H]Inositol (81.5 Ci/mmol) was purchased from Amersham International. PGE,, 12-O-tetradecanoylphorbol13-acetate (TPA), la-phorbol 12,13didecanoate (la-PDD), forskolin, and NaF were purchased from Sigma. 1-Oleoyl-2-acetylglycerol (OAG) was purchased from Nacalai Tesque, Inc. (Kyoto, Japan). l-(5-Isoquinolinesulfonyl)-2-methylpiperazine dihydrochloride (H7) and N-(2-guanidinoethyl)-5-isoquinolinesulfonamide hydrochloride (HA1004) were purchased from Seikagaku Kogyo Co. (Tokyo, Japan). The CAMP radioimmunoassay kit and dibutyryl CAMP (Bt,cAMP) were obtained from Yamasa Shoyu Co. (Chiba, Japan). Other materials and chemicals were obtained from commercial sources. TPA and 4a-PDD were dissolved in dimethyl sulfoxide. PGE, and forskolin were dissolved in ethanol. The maximum concentration of dimethyl sulfoxide or ethanol in the culture medium was O.l%, and this affected neither the assay for CAMP nor the measurement of IP, formation. Cell culture. Cloned osteoblast-like cells, MC3T3-El, were generously provided by Dr. M. Kumegawa (Meikai University, Sakado, Japan) and maintained in a-minimum essential medium (o-MEM) containing 10% fetal calf serum (FCS) at 37’C in a humidified atmosphere of 5% CO,/95% air. The cells (5 X 10’) were seeded into 35-mm diameter dishes in 2 ml of ol-MEM containing 10% FCS. After 5 days, the medium was exchanged for 2 ml of LU-MEM containing 0.3% FCS. The cells were used for experiments after 48 h. In experiments for the formation of IP,, the medium was exchanged for inositol-free (YMEM containing 0.3% FCS. Assay for CAMP. The cultured cells were pretreated with 0.5 mM 3-isobutyl-1-methylxanthine for 10 min in an assay buffer (5 mM 4-(2-hydroxyethyl)l-piperazineethanesulfonic acid, pH 7.4,150 mM NaCl, 5 mM KCl, 0.8 mM MgSO,, 1 mM CaCl,, and 5.5 mM glucose) containing 0.01% bovine serum albumin (BSA). When indicated, TPA, 4o-PDD, OAG, H-7, or HA1004 was added 10 min prior to 3isobutyl-l-methylxanthine treatment. The cells were then stimulated by PGE,, forskolin, or NaF at 37°C. The reaction was terminated by aspiration of the medium, then the intracellular CAMP was extracted with 1 ml of 90% n-propanol [16]. CAMP in the medium and extracts was measured by a radioimmunoassay kit. Measurement of the formation of 1P3. The cultured cells were labeled with myo-[2-3H]inositol (3 &i/dish) for 48 h. The labeled cells were pretreated with 10 mM LiCl at 37°C for 10 min in 1 ml of the assay buffer containing 0.01% BSA. When indicated, B&CAMP or forskolin was added 10 min prior to LiCl treatment. The cells were then stimulated by PGE,. The reaction was terminated by 15% trichloroacetic acid. The acid supernatant was treated with diethyl ether to remove the acid and neutralized with NaOH. The supernatant was applied to a column of Dowex AGl-X8 formate form. To elute inositol monophosphate and inositol bisphosphate, 8 ml of 0.1 M formic acid containing 0.4 M ammonium formate was applied to the column. The radioactive IP, was then eluted from the column with 8 ml of 0.1 M formic acid containing 1 M ammonium formate [17, 181. The data were analyzed by Student’s t test. Statistical analysis.
RESULTS
PGE, stimulated both CAMP accumulation and IP, formation time dependently in osteoblast-like MC3T3El cells (Fig. 1). The CAMP accumulation showed the peak at 5 min and decreased thereafter, whereas the
IN MOUSE
OSTEOBLASTS
131
formation of IP, reached a plateau almost within 10 min and sustained it up to 30 min. In the same cell cultures examined for intracellular CAMP accumulation, the CAMP in the medium was undetectable and not influenced by PGE,. Both the CAMP accumulation and the IP, formation were stimulated by PGE, in a dosedependent manner in the range between 1 niV and 10 pA4 (Fig. 2). The effect of PGE, on CAMP accumulation was more potent (EC, was 80 nM) than that on IP, formation (EC, was 0.8 PM). Pretreatment with TPA, a PKC-activating phorbol ester, which by itself had no effect on CAMP accumulation in these cells (data not shown), suppressed the 10 p&f PGE,-induced CAMP accumulation in a dose-dependent manner in the range between 0.1 ti and 1 pM (Fig. 3). TPA at 1 pM resulted in about 50% suppression of 10 pM PGE,-induced CAMP accumulation. 4a-PDD, which is known to be inactive for PKC [9], was ineffective in this capacity (Fig. 3). OAG, a synthetic diacylglycerol known to be a specific activator of PKC [9], which alone had no effect on CAMP accumulation, like TPA (data not shown), also suppressed the CAMP accumulation induced by 10 PM PGE, in a dose-dependent manner in the range between 1 pM and 0.1 mM (Fig. 4). To investigate the point exerted by TPA on the inhibitory effect, we next examined the effect of TPA on forskolin- or NaF-induced CAMP accumulation. The CAMP accumulation induced by 50 pM forskolin, which activates adenylate cyclase directly [19], was not affected by TPA (Table 1). NaF, a GTP-binding protein activator [20,21], induced CAMP accumulation dose dependently between 5 and 40 mM in these cells (data not shown). NaF (40 mA4) induced CAMP accumulation was not affected by TPA (Table 1). In addition, to test the effect of the adenylate cyclaseCAMP system on PGE,-induced PI hydrolysis, we examined the effect of forskolin or B&CAMP on PGE,-induced formation of IP,. B&CAMP showed little effect on the formation of IP, induced by 10 pM PGE, (Fig. 5). Forskolin (50 PM) had little effect on the IP, formation induced by 10 PM PGE, (data not shown). Moreover, to elucidate the role of PKC in the PGE,induced CAMP accumulation, we examined the effect of H-7, a PKC inhibitor [22], on CAMP accumulation. H-7 or HA1004 by itself had little effect on CAMP accumulation (data not shown). When the cells were incubated for 10 min with PGE,, the pretreatment of H-7 (20 pM) had little effect on CAMP accumulation compared with those pretreated with HA1004, a control for H-7. (PGE,-induced CAMP accumulations after the pretreatment with H-7 and HA1004 were 518 + 20 pmol/ dish and 532 + 20 pmol/dish, respectively. Each value represents the mean + SD of triplicate determinations.) However, when incubated for 20 or 30 min with PGE,, H-7 significantly enhanced the CAMP accumulation compared with HA1004 (Fig. 6).
132
KOZAWA ET AL.
09 0
’
’
0
10
20
Time
30
(min)
0
20
10
Time
30
(mln)
FIG. 1. Time-dependent effects of PGE, on the CAMP accumulation (A) and the formation of IP, (B) in MC3T3-El cells. (A) The cultured cells were pretreated with 0.5 n&f 3-isobutyl-1-methylxanthine for 10 min and stimulated by 10 pM PGE, for the indicated periods, and then the CAMP accumulation was determined. (B) The cultured cells were pretreated with 10 m&f LiCl for 10 min and stimulated by 10 pM PGE, for the indicated periods, and then the formation of IP, was determined. Each value represents the mean + SD of triplicate determinations. Similar results were obtained with two additional and different cell preparations.
DISCUSSION
In the present study, we showed that TPA inhibited the PGE,-induced CAMP accumulation, but 4a-PDD was inactive in this capacity in osteoblast-like MC3T3El cells. TPA is known to activate PKC, whereas 4aPDD is known to be inactive for PKC [9]. Therefore, it is most likely that the action of TPA in the inhibition of the PGE,-induced CAMP production is mediated through the activation of PKC. In support of this notion, OAG, a synthetic diacylglycerol which is known as a specific activator for PKC [9], mimicked TPA by in-
0
9
6
7
6
5
-Log (PGEP), M FIG. 2. Dose-dependent effects of PGE, on the CAMP accumulation and the formation of IPs in MC3T3-El cells. The cultured cells were stimulated by various doses of PGE, for 5 min, and then the CAMP accumulation (0) and the formation of IPI (0) were determined. Values are expressed as a net increase compared with control. Each value represents the mean f SD of triplicate determinations. Similar results were obtained with two additional and different cell preparations.
hibiting the PGE,-induced CAMP accumulation. Our results are consistent with previous reports that PGE,induced CAMP production is attenuated by TPA or OAG observed in osteoblast-like UMR-106 cells and MOB 3-4 cells [14, 151. It is well established that G, is involved in the signaling between the receptor and adenylate cyclase [7]. To clarify the focus of PKC, we examined the effect of TPA on NaF- or forskolin-induced CAMP accumulation. We demonstrated that TPA showed little effect on the CAMP accumulation induced by these agents. Since forskolin is known to activate adenylate cyclase directly [19] and NaF is an activator of GTP-binding protein [20, 211, these results indicate
o?~
J
-Log (Phorbol Ester), M
FIG. 3. Effects of phorbol esters on PGE,-induced CAMP accumulation in MC3T3-El cells. The cultured cells were pretreated with various doses of TPA (0) or 4a-PDD (0) for 20 min, then stimulated by 10 PM PGE, for 5 min. Each value represents the mean f SD of triplicate determinations. Similar results were obtained with two additional and different cell preparations.
PGEJNDUCED
SIGNAL TRANSDUCTION
133
IN MOUSE OSTEOBLASTS
1 041 0
-Log (OAG), Y
03.
1
3
BE& (mu)
FIG. 4. Effect of OAG on PGE,-induced CAMP accumulation in MC3T3-El cells. The cultured cells were pretreated with various doses of OAG for 20 min, then stimulated by 10 &f PGE, for 5 min. Each value represents the mean f SD of triplicate determinations. Similar results were obtained with two additional and different cell preparations.
that the activation of PKC does not affect post-G, signaling in the PGE,-induced CAMP production. So, it seems certain that the effect of PKC exerts at the point between PGE, receptor and G, in MC3T3-El cells. In addition, we showed that forskolin and B&CAMP had little effect on the PGE,-induced IP, formation. Namely, in MC3T3-El cells, it seems likely that there is a cross-talk between CAMP and PKC pathways, and the CAMP signaling pathway is modulated by PKC action but PI hydrolysis is not modulated by the adenylate cyclase-CAMP system. In addition, it has recently been demonstrated that PGE, stimulates not only CAMP production but also PI hydrolysis in UMR-106 cells [8]. In MC3T3-El cells, we
TABLE
01.
FIG. 6. Effect of B+AMP on PGE,-induced IP, formation in MC3T3-El cells. The cultured cells were pretreated with various doses of Bt+AMP for 20 min, then stimulated by 10 @ PGE? for 10 min. Each value represents the mean + SD of triplicate determinations. Similar results were obtained with two additional and different cell preparations.
have also shown that PI hydrolysis is stimulated by PGE, in a pertussis toxin-sensitive manner [ll]. It is well established that PI hydrolysis by phospholipase C results in the formation of two second messengers, diacylglycerol and IP3, and among these products, diacylglycerol is responsible for the activation of PKC [9, lo]. Therefore, it is considered that PGE, by itself activates PKC through PI hydrolysis in osteoblasts. In turn, this evidence that TPA inhibited the PGE,-induced CAMP accumulation might allow the speculation that PGE, regulates CAMP production through self-induced activation of PKC.
1
Effects of TPA on the CAMP Accumulation Induced by PGE,, Forskolin, or NaF in MC3T3-El Cells CAMP accumulation Effector
Control
TPA-treated pmol/dish
None PGEz (10 df)
Forskolin (50 PM) NaF (40 mi%f)
35+ 3 575 k 25
29Lk 5 280 + 25*
1263 3f:90 512 + 30
1320 + 95 547 f 35
Note. The cultured cells were pretreated with 1 @TPA for 20 min, then stimulated by 10 fl PGEa, 50 M forskolin, or 40 mM NaF for 5 min. Each value represents the mean rt SD of triplicate determinations. Similar results were obtained with two additional and different cell preparations. * Significantly different from control (P < 0.01).
Time
(mln)
FIG. 6. Effect of H-7 or HA1004 on PGE,-induced CAMP accumulation in MC3T3-El cells. The cultured cells were pretreated with 20 fl H-7 or 20 piU HA1904 for 20 min, then stimulated by 10 & PGE, for the indicated periods. Each value represents the mean f SD of triplicate determinations. Similar results were obtained with two additional and different cell preparations.
134
KOZAWA ET AL.
We first compared the time-course of PGE,-induced CAMP accumulation with that of PGE,-induced IP, formation in MC3T3-El cells (Fig. 1). We demonstrated that the CAMP accumulation showed the peak value at 5 min and decreased thereafter, whereas the IP, formation reached a plateau almost within 10 min and sustained it up to 30 min. As for CAMP accumulation, PGE,-induced intracellular CAMP levels fell after 5 min despite our use of 3-isobutyl-1-methylxanthine, an inhibitor of phosphodiesterase, and the CAMP of the cultured medium was undetectable. These data suggest that CAMP excretion is unlikely and some degradation of CAMP occurs. Nevertheless, it seems that both of the time-dependent curves are consistent with our notion that CAMP production is inhibited by the activation of PKC through PI hydrolysis but that IP, formation is not inhibited by the adenylate cyclase-CAMP system. So, we next compared the PGE,-induced CAMP accumulation with the PGE,-induced IP, formation in dose dependency (Fig. 2) and showed that EC,, levels of PGE, affecting IP, formation were lo-fold higher than those affecting CAMP accumulation. This evidence suggests that PGE, promotes CAMP production at a lower dose and modulates the production by self-induced activation of PKC at a higher dose. Therefore, both the time and dose dependency of PGE,-induced CAMP accumulation and IP, formation seem to be reasonable for the existence of interaction between two signaling pathways which we have shown as above. PGE, is considered to act as an autocrine or paracrine factor; thus, the local concentration of PGE, close to the cells is important. So, it seems likely that the local concentration reaches a sufficient level affecting IP, formation under a physiological state. In addition, we demonstrated that H-7, a PKC inhibitor [22], significantly enhanced the PGE,-induced CAMP accumulation compared with HA1004 in MC3T3-El cells. Since HA1004 is known to be useful as a control for H-7 [22], this evidence strongly suggests that the activation of PKC through PI hydrolysis by PGE, truly causes the decrease of CAMP production observed in the experiment of time-course. So, it seems certain that PGE, regulates CAMP production through self-induced activation of PKC in these cells. Received June 27,199l Revised version received September 13, 1991
In conclusion, our results strongly suggest that PGE, modulates osteoblast functions through a cross-talk interaction between CAMP production and PI hydrolysis in osteoblast-like cells. REFERENCES Nijweide, P. J., Burger, E. H., and Feyen, J. H. M. (1986) Physiol. Rev. 66,855-8&L 2. Smith, W. L. (1989) Biochem. J. 259,315-324. 3. Martin, T. J., and Partridge, N. C. (1981) Excerpta Med. 1471.
156. 4. 5.
Atkins, D., and Martin, T. J. (1977) Prostagkmdins 13,861-871. Partridge, N. C., Alcom, D., Michelangeli, V. P., Kemp, B. E., Ryan, G. B., and Martin, T. J. (1981) Endocrinology lOf3,213219.
15.
Hakeda, Y., Nakatani, Y., Hiramatau, M., Kurihara, N., Tsunoi, M., Ikeda, E., and Kumegawa, M. (1985) J. Biochem. 97,97-104. Gilman, A. G. (1987) Annu. Rev. Biochem. S&615-649. Yamaguchi, D. T., Hahn, T. J., Beeker, T. G., Kleeman, C. R., and MuaIlem, S. (1988) J. Biol. Chem. 263,10,745-10,753. Nishizuka, Y. (1986) Science 233,305-312. Berridge, M. J., and Irvine, R. F. (1984) Nature 312, 315-321. Tokuda, H., Kozawa, O., Yoneda, M., Oiso, Y., Takatsuki, K., Asano, T., and Kato, K. (1991) J. Biochem. 109,229-233. Kodama, H., Amagai, Y., Sudo, H., Kasai, S., and Yamamoto, S. (1981) Jpn. J. Oral Biol. 93,899-901. Sudo, H., Kodama, H., Amagai, Y., Yamamoto, S., and Kasai, S. (1983) J. Cell Biol. 96, 191-198. Dobson, P. R. M., Brown, B. L., Michelangeli, V. P., Short, A. D., Moseley, J. M., Russel, R. G. G., and Martin, T. J. (1999) Btichim. Biophys. Acta 1052,323-326. Kawase, T., Orihara, M., and Suzuki, A. (1991) Calcif. Tissue Znt.
16.
Robey, P. G., and Termine, J. D. (1985) Calcif. Z’i.ssueZnt. 37,
6. 7. 8. 9. 10. 11. 12. 13. 14.
48,167-175. 453-458. 17. 18. 19.
Berridge, M. J., Dawson, R. M. C., Downes, C. P., Healop, J. P., and Irvine, R. F. (1983) Biochem. J. 212,473-482. Berridge, M. J., Heslop, J. P., Irvine, R. F., and Brown, K. D. (1984) Biochem. J. 222,195-201. Seamon, K., and Daly, J. W. (1981) J. Biol. Chem. 256,97999801.
20.
Ross, E. M., and Gilman, A. G. (1986) Annu. Reu. Biochem. 49, 533-564.
Sternweis, P. C., and Gilman, A. G. (1982) Proc. Natl. Acad. Sci. USA 79,4888-4891. 22. Hidaka, H., Inagaki, M., Kawamoto, S., and Sasaki, Y. (1984) Biochemistry 23,5036-5041.
21.
CELL RESEARCH
198,130-134
(1992)
Cross-Talk Regulation between Cyclic AMP Production and Phosphoinositide Hydrolysis Induced by Prostaglandin E2 in Osteoblast-like Cells OSAMU KOZAWA,*~’ HAFWHIKO TOKUDA,? MASAICHI MIWA,? JUN KOTOYORI,P AND YUTAKA OISO~ *Department of Biochemistry, Znstitute for Developmental Research, Aichi Prefectural Colony, Kasugai, Aichi 480-03, Japan; and fFirst Department of Internal Medicine, Nagoya University School of Medicine, Nagoya 466, Japan
In cloned osteoblast-like MC3T3-El cells, PGE, stimulated both CAMP accumulation and the formation of inositol trisphosphate (IPJ dose dependently. The CAMP accumulation showed the peak value at 5 min and decreased thereafter, whereas the IP, formation reached a plateau almost within 10 min and sustained it up to 30 min. The effect of PGEp on CAMP accumulation (E&, was 80 n&f) was more potent than that on IP, 12-G-Tetradecanoylformation (EC& was 0.8 a). phorbol-13-acetate (TPA), a protein kinase C (PKC)activating phorbol ester, reduced the PGE,-induced CAMP accumulation, whereas la-phorbol 12,13didecanoate, a PKC-nonactivating phorbol ester, had little effect on the CAMP accumulation. 1-Oleoyld-acetylglycerol, a specific activator for PKC, inhibited PGEainduced CAMP accumulation. TPA had little effect on CAMP accumulation induced by forskolin or NaF, a GTP-binding protein activator. So, the effect of TPA is presumed to be exerted at the point between the PGEl receptor and G,. On the other hand, forskolin and dibutyryl CAMP had little effect on the IP, formation stimulated by PGE,. H-7, a PKC inhibitor, enhanced the PGE,-induced CAMP accumulation in comparison with HA1004, a control for H-7. Our data suggest that PGEa regulates CAMP production through self-induced activation of PKC. These results strongly suggest that there is an autoregulatory mechanism in PGEs signaling, and PGE, modulates osteoblast functions through a crosstalk interaction between CAMP production and phosphoinositide hydrolysis in osteoblast-like cells. Q 1992 Academic Prese, Inc.
INTRODUCTION
Prostaglandins are considered to be important regulators of osteoblasts as autacoids [l, 21. Among them, PGE, is known as a potent bone resorbing agent [3]. In osteoblasts, the actions of PGE, have been thought to be 1 To whom reprint
requests should be addressed.
0014~4%?7/92 $3.00 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
mediated through the activation of adenylate cyclase [4-61. It is well known that GTP-binding proteins (G, and Gi) function as intermediaries in transmembrane signaling from the receptor to adenylate cyclase [7]. In addition, it has recently been reported that phosphoinositide (PI) hydrolysis is also stimulated by PGE, in rat osteosarcoma UMR-106 cells [S]. It is generally accepted that in response to a variety of agonists, phosphoinositides are hydrolyzed by phospholipase C, resulting in the formation of diacylglycerol and inositol phosphates. Among these products, diacylglycerol and inositol trisphosphate (IP,) serve as messengers for the activation of protein kinase C (PKC) and the mobilization of intracellular Ca2+, respectively [9, lo]. In this signaling system, as well as the adenylate cyclase-CAMP system, evidence is accumulating that GTP-binding protein(s) is involved in the coupling of the receptor to phospholipase C [ 71. In a previous report [ 111, we showed that the PGE,-induced PI hydrolysis is mediated by a pertussis toxin-sensitive GTP-binding protein in cloned osteoblast-like MC3T3-El cells which have been derived from newborn mouse calvaria [12, 131. So, it is now accepted that the effects of PGE, are mediated through both CAMP production and PI hydrolysis in osteoblasts. Recently, it has been reported that a tumor-promoting phorbol ester known as a potent stimulator of PKC inhibits the PGE,-inducedcAMP production in UMR-106 cells and mouse osteoblast-like MOB 3-4 cells [14,15]. This evidence suggests that so-called “cross-talk” between these signal transduction pathways does indeed exist in osteoblasts. However, the precise mechanism of cross-talk between these two signaling pathways of PGE2 has not yet been elucidated. In the present study, therefore, we investigated the interaction between two PGE,-induced signaling systems, CAMP production and PI hydrolysis in osteoblast-like MC3T3-El cells. Herein, we demonstrate that PGE, regulates CAMP production through self-induced activation of PKC in MC3T3-El cells. In contrast, the adenylate cyclase-CAMP signaling pathway had little effect on the PGE,-induced PI hydrolysis in these cells. These
130
PGE,-INDUCED
SIGNAL
TRANSDUCTION
results strongly suggest that PGE, modulates osteoblast functions through a cross-talk interaction between CAMP production and PI hydrolysis in osteoblast-like cells. MATERIALS
AND METHODS
Materials. myo-[2-3H]Inositol (81.5 Ci/mmol) was purchased from Amersham International. PGE,, 12-O-tetradecanoylphorbol13-acetate (TPA), la-phorbol 12,13didecanoate (la-PDD), forskolin, and NaF were purchased from Sigma. 1-Oleoyl-2-acetylglycerol (OAG) was purchased from Nacalai Tesque, Inc. (Kyoto, Japan). l-(5-Isoquinolinesulfonyl)-2-methylpiperazine dihydrochloride (H7) and N-(2-guanidinoethyl)-5-isoquinolinesulfonamide hydrochloride (HA1004) were purchased from Seikagaku Kogyo Co. (Tokyo, Japan). The CAMP radioimmunoassay kit and dibutyryl CAMP (Bt,cAMP) were obtained from Yamasa Shoyu Co. (Chiba, Japan). Other materials and chemicals were obtained from commercial sources. TPA and 4a-PDD were dissolved in dimethyl sulfoxide. PGE, and forskolin were dissolved in ethanol. The maximum concentration of dimethyl sulfoxide or ethanol in the culture medium was O.l%, and this affected neither the assay for CAMP nor the measurement of IP, formation. Cell culture. Cloned osteoblast-like cells, MC3T3-El, were generously provided by Dr. M. Kumegawa (Meikai University, Sakado, Japan) and maintained in a-minimum essential medium (o-MEM) containing 10% fetal calf serum (FCS) at 37’C in a humidified atmosphere of 5% CO,/95% air. The cells (5 X 10’) were seeded into 35-mm diameter dishes in 2 ml of ol-MEM containing 10% FCS. After 5 days, the medium was exchanged for 2 ml of LU-MEM containing 0.3% FCS. The cells were used for experiments after 48 h. In experiments for the formation of IP,, the medium was exchanged for inositol-free (YMEM containing 0.3% FCS. Assay for CAMP. The cultured cells were pretreated with 0.5 mM 3-isobutyl-1-methylxanthine for 10 min in an assay buffer (5 mM 4-(2-hydroxyethyl)l-piperazineethanesulfonic acid, pH 7.4,150 mM NaCl, 5 mM KCl, 0.8 mM MgSO,, 1 mM CaCl,, and 5.5 mM glucose) containing 0.01% bovine serum albumin (BSA). When indicated, TPA, 4o-PDD, OAG, H-7, or HA1004 was added 10 min prior to 3isobutyl-l-methylxanthine treatment. The cells were then stimulated by PGE,, forskolin, or NaF at 37°C. The reaction was terminated by aspiration of the medium, then the intracellular CAMP was extracted with 1 ml of 90% n-propanol [16]. CAMP in the medium and extracts was measured by a radioimmunoassay kit. Measurement of the formation of 1P3. The cultured cells were labeled with myo-[2-3H]inositol (3 &i/dish) for 48 h. The labeled cells were pretreated with 10 mM LiCl at 37°C for 10 min in 1 ml of the assay buffer containing 0.01% BSA. When indicated, B&CAMP or forskolin was added 10 min prior to LiCl treatment. The cells were then stimulated by PGE,. The reaction was terminated by 15% trichloroacetic acid. The acid supernatant was treated with diethyl ether to remove the acid and neutralized with NaOH. The supernatant was applied to a column of Dowex AGl-X8 formate form. To elute inositol monophosphate and inositol bisphosphate, 8 ml of 0.1 M formic acid containing 0.4 M ammonium formate was applied to the column. The radioactive IP, was then eluted from the column with 8 ml of 0.1 M formic acid containing 1 M ammonium formate [17, 181. The data were analyzed by Student’s t test. Statistical analysis.
RESULTS
PGE, stimulated both CAMP accumulation and IP, formation time dependently in osteoblast-like MC3T3El cells (Fig. 1). The CAMP accumulation showed the peak at 5 min and decreased thereafter, whereas the
IN MOUSE
OSTEOBLASTS
131
formation of IP, reached a plateau almost within 10 min and sustained it up to 30 min. In the same cell cultures examined for intracellular CAMP accumulation, the CAMP in the medium was undetectable and not influenced by PGE,. Both the CAMP accumulation and the IP, formation were stimulated by PGE, in a dosedependent manner in the range between 1 niV and 10 pA4 (Fig. 2). The effect of PGE, on CAMP accumulation was more potent (EC, was 80 nM) than that on IP, formation (EC, was 0.8 PM). Pretreatment with TPA, a PKC-activating phorbol ester, which by itself had no effect on CAMP accumulation in these cells (data not shown), suppressed the 10 p&f PGE,-induced CAMP accumulation in a dose-dependent manner in the range between 0.1 ti and 1 pM (Fig. 3). TPA at 1 pM resulted in about 50% suppression of 10 pM PGE,-induced CAMP accumulation. 4a-PDD, which is known to be inactive for PKC [9], was ineffective in this capacity (Fig. 3). OAG, a synthetic diacylglycerol known to be a specific activator of PKC [9], which alone had no effect on CAMP accumulation, like TPA (data not shown), also suppressed the CAMP accumulation induced by 10 PM PGE, in a dose-dependent manner in the range between 1 pM and 0.1 mM (Fig. 4). To investigate the point exerted by TPA on the inhibitory effect, we next examined the effect of TPA on forskolin- or NaF-induced CAMP accumulation. The CAMP accumulation induced by 50 pM forskolin, which activates adenylate cyclase directly [19], was not affected by TPA (Table 1). NaF, a GTP-binding protein activator [20,21], induced CAMP accumulation dose dependently between 5 and 40 mM in these cells (data not shown). NaF (40 mA4) induced CAMP accumulation was not affected by TPA (Table 1). In addition, to test the effect of the adenylate cyclaseCAMP system on PGE,-induced PI hydrolysis, we examined the effect of forskolin or B&CAMP on PGE,-induced formation of IP,. B&CAMP showed little effect on the formation of IP, induced by 10 pM PGE, (Fig. 5). Forskolin (50 PM) had little effect on the IP, formation induced by 10 PM PGE, (data not shown). Moreover, to elucidate the role of PKC in the PGE,induced CAMP accumulation, we examined the effect of H-7, a PKC inhibitor [22], on CAMP accumulation. H-7 or HA1004 by itself had little effect on CAMP accumulation (data not shown). When the cells were incubated for 10 min with PGE,, the pretreatment of H-7 (20 pM) had little effect on CAMP accumulation compared with those pretreated with HA1004, a control for H-7. (PGE,-induced CAMP accumulations after the pretreatment with H-7 and HA1004 were 518 + 20 pmol/ dish and 532 + 20 pmol/dish, respectively. Each value represents the mean + SD of triplicate determinations.) However, when incubated for 20 or 30 min with PGE,, H-7 significantly enhanced the CAMP accumulation compared with HA1004 (Fig. 6).
132
KOZAWA ET AL.
09 0
’
’
0
10
20
Time
30
(min)
0
20
10
Time
30
(mln)
FIG. 1. Time-dependent effects of PGE, on the CAMP accumulation (A) and the formation of IP, (B) in MC3T3-El cells. (A) The cultured cells were pretreated with 0.5 n&f 3-isobutyl-1-methylxanthine for 10 min and stimulated by 10 pM PGE, for the indicated periods, and then the CAMP accumulation was determined. (B) The cultured cells were pretreated with 10 m&f LiCl for 10 min and stimulated by 10 pM PGE, for the indicated periods, and then the formation of IP, was determined. Each value represents the mean + SD of triplicate determinations. Similar results were obtained with two additional and different cell preparations.
DISCUSSION
In the present study, we showed that TPA inhibited the PGE,-induced CAMP accumulation, but 4a-PDD was inactive in this capacity in osteoblast-like MC3T3El cells. TPA is known to activate PKC, whereas 4aPDD is known to be inactive for PKC [9]. Therefore, it is most likely that the action of TPA in the inhibition of the PGE,-induced CAMP production is mediated through the activation of PKC. In support of this notion, OAG, a synthetic diacylglycerol which is known as a specific activator for PKC [9], mimicked TPA by in-
0
9
6
7
6
5
-Log (PGEP), M FIG. 2. Dose-dependent effects of PGE, on the CAMP accumulation and the formation of IPs in MC3T3-El cells. The cultured cells were stimulated by various doses of PGE, for 5 min, and then the CAMP accumulation (0) and the formation of IPI (0) were determined. Values are expressed as a net increase compared with control. Each value represents the mean f SD of triplicate determinations. Similar results were obtained with two additional and different cell preparations.
hibiting the PGE,-induced CAMP accumulation. Our results are consistent with previous reports that PGE,induced CAMP production is attenuated by TPA or OAG observed in osteoblast-like UMR-106 cells and MOB 3-4 cells [14, 151. It is well established that G, is involved in the signaling between the receptor and adenylate cyclase [7]. To clarify the focus of PKC, we examined the effect of TPA on NaF- or forskolin-induced CAMP accumulation. We demonstrated that TPA showed little effect on the CAMP accumulation induced by these agents. Since forskolin is known to activate adenylate cyclase directly [19] and NaF is an activator of GTP-binding protein [20, 211, these results indicate
o?~
J
-Log (Phorbol Ester), M
FIG. 3. Effects of phorbol esters on PGE,-induced CAMP accumulation in MC3T3-El cells. The cultured cells were pretreated with various doses of TPA (0) or 4a-PDD (0) for 20 min, then stimulated by 10 PM PGE, for 5 min. Each value represents the mean f SD of triplicate determinations. Similar results were obtained with two additional and different cell preparations.
PGEJNDUCED
SIGNAL TRANSDUCTION
133
IN MOUSE OSTEOBLASTS
1 041 0
-Log (OAG), Y
03.
1
3
BE& (mu)
FIG. 4. Effect of OAG on PGE,-induced CAMP accumulation in MC3T3-El cells. The cultured cells were pretreated with various doses of OAG for 20 min, then stimulated by 10 &f PGE, for 5 min. Each value represents the mean f SD of triplicate determinations. Similar results were obtained with two additional and different cell preparations.
that the activation of PKC does not affect post-G, signaling in the PGE,-induced CAMP production. So, it seems certain that the effect of PKC exerts at the point between PGE, receptor and G, in MC3T3-El cells. In addition, we showed that forskolin and B&CAMP had little effect on the PGE,-induced IP, formation. Namely, in MC3T3-El cells, it seems likely that there is a cross-talk between CAMP and PKC pathways, and the CAMP signaling pathway is modulated by PKC action but PI hydrolysis is not modulated by the adenylate cyclase-CAMP system. In addition, it has recently been demonstrated that PGE, stimulates not only CAMP production but also PI hydrolysis in UMR-106 cells [8]. In MC3T3-El cells, we
TABLE
01.
FIG. 6. Effect of B+AMP on PGE,-induced IP, formation in MC3T3-El cells. The cultured cells were pretreated with various doses of Bt+AMP for 20 min, then stimulated by 10 @ PGE? for 10 min. Each value represents the mean + SD of triplicate determinations. Similar results were obtained with two additional and different cell preparations.
have also shown that PI hydrolysis is stimulated by PGE, in a pertussis toxin-sensitive manner [ll]. It is well established that PI hydrolysis by phospholipase C results in the formation of two second messengers, diacylglycerol and IP3, and among these products, diacylglycerol is responsible for the activation of PKC [9, lo]. Therefore, it is considered that PGE, by itself activates PKC through PI hydrolysis in osteoblasts. In turn, this evidence that TPA inhibited the PGE,-induced CAMP accumulation might allow the speculation that PGE, regulates CAMP production through self-induced activation of PKC.
1
Effects of TPA on the CAMP Accumulation Induced by PGE,, Forskolin, or NaF in MC3T3-El Cells CAMP accumulation Effector
Control
TPA-treated pmol/dish
None PGEz (10 df)
Forskolin (50 PM) NaF (40 mi%f)
35+ 3 575 k 25
29Lk 5 280 + 25*
1263 3f:90 512 + 30
1320 + 95 547 f 35
Note. The cultured cells were pretreated with 1 @TPA for 20 min, then stimulated by 10 fl PGEa, 50 M forskolin, or 40 mM NaF for 5 min. Each value represents the mean rt SD of triplicate determinations. Similar results were obtained with two additional and different cell preparations. * Significantly different from control (P < 0.01).
Time
(mln)
FIG. 6. Effect of H-7 or HA1004 on PGE,-induced CAMP accumulation in MC3T3-El cells. The cultured cells were pretreated with 20 fl H-7 or 20 piU HA1904 for 20 min, then stimulated by 10 & PGE, for the indicated periods. Each value represents the mean f SD of triplicate determinations. Similar results were obtained with two additional and different cell preparations.
134
KOZAWA ET AL.
We first compared the time-course of PGE,-induced CAMP accumulation with that of PGE,-induced IP, formation in MC3T3-El cells (Fig. 1). We demonstrated that the CAMP accumulation showed the peak value at 5 min and decreased thereafter, whereas the IP, formation reached a plateau almost within 10 min and sustained it up to 30 min. As for CAMP accumulation, PGE,-induced intracellular CAMP levels fell after 5 min despite our use of 3-isobutyl-1-methylxanthine, an inhibitor of phosphodiesterase, and the CAMP of the cultured medium was undetectable. These data suggest that CAMP excretion is unlikely and some degradation of CAMP occurs. Nevertheless, it seems that both of the time-dependent curves are consistent with our notion that CAMP production is inhibited by the activation of PKC through PI hydrolysis but that IP, formation is not inhibited by the adenylate cyclase-CAMP system. So, we next compared the PGE,-induced CAMP accumulation with the PGE,-induced IP, formation in dose dependency (Fig. 2) and showed that EC,, levels of PGE, affecting IP, formation were lo-fold higher than those affecting CAMP accumulation. This evidence suggests that PGE, promotes CAMP production at a lower dose and modulates the production by self-induced activation of PKC at a higher dose. Therefore, both the time and dose dependency of PGE,-induced CAMP accumulation and IP, formation seem to be reasonable for the existence of interaction between two signaling pathways which we have shown as above. PGE, is considered to act as an autocrine or paracrine factor; thus, the local concentration of PGE, close to the cells is important. So, it seems likely that the local concentration reaches a sufficient level affecting IP, formation under a physiological state. In addition, we demonstrated that H-7, a PKC inhibitor [22], significantly enhanced the PGE,-induced CAMP accumulation compared with HA1004 in MC3T3-El cells. Since HA1004 is known to be useful as a control for H-7 [22], this evidence strongly suggests that the activation of PKC through PI hydrolysis by PGE, truly causes the decrease of CAMP production observed in the experiment of time-course. So, it seems certain that PGE, regulates CAMP production through self-induced activation of PKC in these cells. Received June 27,199l Revised version received September 13, 1991
In conclusion, our results strongly suggest that PGE, modulates osteoblast functions through a cross-talk interaction between CAMP production and PI hydrolysis in osteoblast-like cells. REFERENCES Nijweide, P. J., Burger, E. H., and Feyen, J. H. M. (1986) Physiol. Rev. 66,855-8&L 2. Smith, W. L. (1989) Biochem. J. 259,315-324. 3. Martin, T. J., and Partridge, N. C. (1981) Excerpta Med. 1471.
156. 4. 5.
Atkins, D., and Martin, T. J. (1977) Prostagkmdins 13,861-871. Partridge, N. C., Alcom, D., Michelangeli, V. P., Kemp, B. E., Ryan, G. B., and Martin, T. J. (1981) Endocrinology lOf3,213219.
15.
Hakeda, Y., Nakatani, Y., Hiramatau, M., Kurihara, N., Tsunoi, M., Ikeda, E., and Kumegawa, M. (1985) J. Biochem. 97,97-104. Gilman, A. G. (1987) Annu. Rev. Biochem. S&615-649. Yamaguchi, D. T., Hahn, T. J., Beeker, T. G., Kleeman, C. R., and MuaIlem, S. (1988) J. Biol. Chem. 263,10,745-10,753. Nishizuka, Y. (1986) Science 233,305-312. Berridge, M. J., and Irvine, R. F. (1984) Nature 312, 315-321. Tokuda, H., Kozawa, O., Yoneda, M., Oiso, Y., Takatsuki, K., Asano, T., and Kato, K. (1991) J. Biochem. 109,229-233. Kodama, H., Amagai, Y., Sudo, H., Kasai, S., and Yamamoto, S. (1981) Jpn. J. Oral Biol. 93,899-901. Sudo, H., Kodama, H., Amagai, Y., Yamamoto, S., and Kasai, S. (1983) J. Cell Biol. 96, 191-198. Dobson, P. R. M., Brown, B. L., Michelangeli, V. P., Short, A. D., Moseley, J. M., Russel, R. G. G., and Martin, T. J. (1999) Btichim. Biophys. Acta 1052,323-326. Kawase, T., Orihara, M., and Suzuki, A. (1991) Calcif. Tissue Znt.
16.
Robey, P. G., and Termine, J. D. (1985) Calcif. Z’i.ssueZnt. 37,
6. 7. 8. 9. 10. 11. 12. 13. 14.
48,167-175. 453-458. 17. 18. 19.
Berridge, M. J., Dawson, R. M. C., Downes, C. P., Healop, J. P., and Irvine, R. F. (1983) Biochem. J. 212,473-482. Berridge, M. J., Heslop, J. P., Irvine, R. F., and Brown, K. D. (1984) Biochem. J. 222,195-201. Seamon, K., and Daly, J. W. (1981) J. Biol. Chem. 256,97999801.
20.
Ross, E. M., and Gilman, A. G. (1986) Annu. Reu. Biochem. 49, 533-564.
Sternweis, P. C., and Gilman, A. G. (1982) Proc. Natl. Acad. Sci. USA 79,4888-4891. 22. Hidaka, H., Inagaki, M., Kawamoto, S., and Sasaki, Y. (1984) Biochemistry 23,5036-5041.
21.
