Vol. 181, No. 3, 1991 December 31, 1991
BIOCHEMICAL
Comparison
of Effects
AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages 1173-1180
of HGF and EGF on Cellular
Calcium
in Rat Hepatocytes Tetsuya Mine,’
* Etsuro Ogata, and Toshikazu
Itaru Kojima,
Fourth
Department
of Internal
Tokyo
School of Medicine,
Medicine,
3-28-6 Mejirodai,
Nakamura**
University
of
Bunkyo-ku
Tokyo 112, Japan *
Institute
of Endocrinology,
Gunma
University,
Maebashi
371, Japan **
Department
of Biology, University,
Received
November
Faculty
Fukuoka
of Science,
Kyushu
812, Japan
1, 1991
We compared the effects of HGF and EGF on cytoplasmic free calcium concentration, [Ca2+l,, and inositol trisphosphate production in rat hepatocytes. HGF induced a prompt and transient elevation of [Ca2+lc. EGF also induced an immediate increase in [Ca2+lc, the magnitude of which was greater than that by HGF. In contrast, in the presence of 1 pM extracellular calcium EGF increased [Ca2+l, to a lesser extent than HGF. When cells were pretreated with EGF, the effect of HGF on [Ca2+l, was greatly enhanced. However, such enhancement was not observed in medium containing 1 pM extracellular calcium. In hepatocytes prelabeled with PHIinositol, both HGF and EGF increased [3Hlinositol trisphosphate. HGF and EGF acted synergistically to stimulate production of inositol trisphosphate. These results indicate that both HGF and EGF increase [Ca2+lc by a mechanism involving phosphoinositide turnover and that the actions of HGF 0 1991 and EGF on hepatocyte calcium metabolism are not totally identical. Academic
Press,
Inc.
The activity of hepatocyte growth factor (HGF) was first convincingly demonstrated in serum and platelets (l-3). Subsequently, the factor was purified from several sources including rat platelets (4), human plasma (5,6) and rabbit serum (6). The structure of HGF revealed by molecular cloning has significant homology with plasminogen and related serine proteases (7). Despite that HGF is a potent stimulator of cell growth in hepatocytes, expression of HGF mBNA is observed in several organs (8-11). This raises a 1 To whom correspondence
should
be addressed. ooo6-291x/91 1173
$1.50
Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
Vol.
181, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
possibility that HGF may also have an action as a paracrine growth factor in various cells. In target cells, HGF induces tyrosine phosphorylation of cellular proteins (8) and it is shown recently that HGF receptor possesses an activity of tyrosine kinase (12,13). Epidermal growth factor (EGF) has been known to stimulate proliferation of hepatocytes. When added concomitnatly, the actions of HGF and EGF are additive (3,14). Since EGF receptor also possesses tyrosine kinase activity, it is of interest whether or not the signal transduction mechanisms for HGF and EGF are identical. In this regard, EGF causes hydrolysis of phosphoinositides and increases cytoplasmic free calcium concentration, [Ca2+lc (15). In the present study, we compared the actions of EGF and HGF in isolated rat hepatocytes. Our results demonstrate that HGF, as in the case for EGF, increases [Ca2+le by causing hydrolysis of phosphoinositides but that the action of HGF on cellular calcium metabolism differs in some respects.
Methods
and Materials
Preparation
of Parenchvmal
Cells
Male Wistar rats, wei hin about 200g were used. The parenchymal cells were prepared using t%e coEilagenase perfusion technique of Berry and Friend (16) as previously described (17,18). Then, parenchymal cells were suspended in modified Hanks’ solution (in mM): Na, 137; K, 3.5; KH2pO4, 0.44; NaHC03,4.2; Na$IP04, 0.33; CaCl2 1.0, 1.0; and HEPES (pH 7.4), 20 (equilibrated with 02 gas). Measurement
of Cstoplamic
Free Ca2+ Concentration
in Heuatocvtes
Aequorin was loaded into hepatocytes by making plasma membrane reversibly permeable by the method of Borle et al (19), as described previously (20,21). More than 95% aequorin-loaded cells extruded trypan blue. Aequorin bioluminescence was calibrated in terms of free calcium concentration, assuming a magnesium concentration of 1 mM and an even distribution of calcium in cytosol(22). Measurement
of Production
of Inositol
Trisphosuhate
Inositol phosphates were measured as previously described (15,16). Hepatocytes (lo7 cellslml) were labeled with [ Hlinositol by incubating cells with 10 pCi/ml [3H]inositol for 120 min. After the labeling period, cells were washed and incubated in modified Hanks’ solution containmg 10 mM LiCl for 10 min. Cells were stimulated for 20 set with HGF or EGF. The reaction was stopped by adding perchloric acid (final concentration: 10%). Cells were homo enized by repetitive aspirations through a 26-gauge needle and centri f uged at 800Xg for 5 min. The supernatant was neutralized with 5M KOH and applied to an anion exchange column. Inositol phosphates were separated as described by Berridge and Irvine (23). Isomers of inositol trlsphosphate were not measured. 1174
Vol.
181, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
Materials EGF was purchased from Sigma (St. Louis , MI). HGF (Human recombinant HGF: h-r HGF) was purified from culture fluid from transformed CHO cells (mouse fibroblast) as described previously (7). A.&q? was purchased from Dr. J.R. Bhnks, Mayo Foundation (Rochester, . Results Effects of HGF and EGF on [Ca2+l, in Hepatocvtes
First, the effect of HGF on [Ca2+lc was investigated. Immediately after the addition of lo-lo M HGF, a prompt elevation of [Ca2+], was observed (Fig. 1A). A rise of [Ca2+lc was composed of an immediate sharp single peak followed by a gradual decline. Approximately 1 min after the addition of HGF, [Ca2+], returned to the resting levels. The effect of HGF on ICa2+lc was detected at lo-” M. The magnitude of the response was increased in a dosedependent manner and the maximum response of [Ca2+], was obtained at lo-lo M. Second, the effect of EGF on [Ca2+l, was also investigated. [Ca2+], was elevated after the addition of 10s8 M EGF. The initial peak was higher than that of HGF (Fig. 2A). About 1 min after the addition of EGF, [Ca2+lc
m.
Comparisonof Effect of HGF on [Ca2+], in the Presenceof 1 mM and 1 uM Extracellular Calcmm Aequorin-loaded hepatocytes were stimulated by 10.lo M HGF in the presenceof 1 mM (Al or 1 pM (B) extracellular calcium. When extracellular calcium was reduced to 1 pM, the concentration of calcium was fixed by using Ca2+-EGTA buffer. Results are the representative of four experiments with similar results. Fip. 2. Effect of EGF on [Ca2+l, in the Presence of 1 mM and 1 uM Extracellular Calcium Aequorin-loaded hepatocytes were stimulated by 10-s M EGF in the presenceof 1 mM (A) or 1 pM (B) extracellular calcium. Results are the representative of four experiments with similar results. 1175
Vol.
181, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
A
B EGF
EGF HGF
I
HGF
Fip. 3. Effect of Pretreatment of EGF on HGF-induced [Ca2+]c in the Presence of 1 mM and 1 uM Extracellular Calcium Aequorin-loaded cells were stimulated b lo-lo HGF after the pretreatment with lo-@ M EGF in the presence oP 1 mM (A) or 1 pM (B) extracellular calcium. Results are the representative of three experiments.
returned to the resting level. The effect of EGF on [Ca2+], was detected at 10-r’ M and the magnitude of the response was increased dose-dcpcndcntly. Maximum response of [Ca2+], was obtained at 10-e M EGF. Then, a role of release from internal Ca2+ pool(s) in the action of IIGF and EGF was compared. In the presence of 1 pM extracellular calcium concentration, the rises of [Ca2+], induced by HGF and EGF were determined. In medium containing this concentration of calcium, basal as well as stimulated calcium influx is negligible (24). Compared to the action observed in 1 mM extracellular calcium, the peak of [Ca2+lc induced by lo-r0 M HGF was not altered in medium containing of 1 pM extracelluIar calcium (Fig. 1B). In contrast to the action of HGF, the peak of [Ca2+lc induced by 10S8M EGF was markedly decreased in medium containing 1 pM extracellular calcium (Fig. 2B). Similar results were obtained at concentrations of UPS M and lO-‘O M (data not shown). However, the rise of [Ca2+], was abolished in the presence of 1 pM extracellular calcium when the concentration of EGF was reduced to lo-” M (data not shown) . Interaction
of [Ca2+lr induced by HGF and EGF
EGF pretreatment on HGF-induced increase in [Ca2+lc was examined. When aequorin-loaded cells were first stimulated by loss M EGF and then lo-lo M HGF was added, the HGF-induced increase in aequorin bioluminescence was greater than that observed in the absence of EGF pretreatment (Fig. 3A). The same experiment was done in the presence of 1 pM extracellular calcium. In this condition, however, the HGF-induced 1176
Vol.
181, No. 3, 1991
BIOCHEMICAL
I 04
’ 10-I'
AND BIOPHYSICAL
I lo-'0 hwtiil
I to-9 CM)
104
RESEARCH COMMUNICATIONS
I 10-7
Dose-Response Relationship for EGF-and HGF.-induced Inositol Firr. 4. Trisphosphate Production Hepatoc:ytes (lo7 cells/ml) were labeled with [3Hlinositol by incubating cells with 10 pCi/ml [3H]inositcl for 120 min. After washing, cells were stimulated for 20 seewith various concentrations of HGF (0) or EGF (0) in the presenceof 10 mM LiCI. Inositol trisphosphates were determined as described in Materials and Methods. Values are the mean f SE for four determinations.
increase in aequorin bioluminescence was similar absence of EGF pretreatment (Fig. 3B). Effects of HGF and EGF on Inositol
Trisphosphate
Effects of HGF and EGF on inositol
to that observed in the
Production
trisphosphate production were cells. Changes in C3H]inositol trisphosphate were detected at 10s and was maximal at 20s (data not shown). As shown in Fig. 4, when changes in [3H]inositol trisphosphates were measured at 2Os, lo-lo M HGF induced a maximal 30% increased in [3HIinositol trisphosphate. Effect of HGF on inositol t&phosphate production was detected at 10-” M. In the case of EGF, time course of changes in C”H]inositol trisphosphate was also examined and production of inositol trisphosphate was determined at 20s. EGF, at doses higher than lo-lo M, increased inositol trisphosphate in a dose-dependent manner (Fig. 4). At lOma M, EGF-induced production of inositol trisphosphate was much greater than that induced by lo-lo M HGF. measured
using [3H]inositol-labelled
1177
Vol.
181, No. 3, 1991
BIOCHEMICAL
AND BtOPHYSlCAL
RESEARCH COMMUNICATIONS
Effect of EGF on HGF induced Inositol Trisphosphate W&ion Hepatocytes (lo7 cells/ml) were labeled with [YH]inositol as described in Materials and Methods. Inositol trisphosphates were determined at 20 set after the addition of various peftides. ml@10 M HGF; 10-g EGF; m 10-r” M HGF and lo- M EGF. Results are expresse!F= aspercent of basal response.Values are the mean f SE for four determinations.
Interaction
of Inositol
T&phosphate
Production
induced by EGF and HGF
The effect of simultaneous addition of EGF and HGF on inositol trisphosphate was examined. Compared with single addition of lo-lo M EGF or lo-lo M HGF, the actions of EGF and HGF were additive (Fig. 5). Even in the presence of 1 pM extracellular concentration, similar additivity was seen (data not shown).
Discussion The present results clearly demonstrate that HGF increases [Ca2+lc in rat hepatocytes. The elevation of [Ca2+], is due largely to mobilization of calcium from an intracellular pool and this reaction is probably brought about by HGF-induced production of inositol 1,4,5-trisphosphate. Hence, HGF activates the calcium messenger system in hepatocytes. The present results together with a recent report that HGF receptor contains tyrosine kinase activity (12,131 indicate that EGF and HGF shares similar transduction system in hepatocyes. With regard to EGF action on inositol phospholipid metabolism in hepatocytes, there still remains some controversy: Bosch et al. (25) reported that EGF did not increase production of inositol trisphosphate; whereas 1178
Vol.
181, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
Johnson et al. (26) reported that EGF elicited increases in both inositol1,4,5trisphosphate and inositol 1,3,4-trisphosphate. Our present results confirm the previous report by Johnson et al. (26). In other cell systems, EGF induces phosphoinositide turnover by activating a specific type of phospholipase C (PLC), PLC-7. It is shown that binding of EGF to its receptor leads to an association of PLC-7 to the activated receptor. Then, PLC-7 is tyrosylphosphorylated and the activity of PLC-7 is augmented by phosphorylation (27,28). The action of EGF in hepatocytes may also involve the same mechanism, However, Liang et al. (29) have shown that pertussis toxin attenuates EGF-induced production of inositol 1,4,5-trisphosphate. This raises a possibility that a pertussis toxin-sensitive G protein may be involved in a certain step of signal transduction mechanism of EGF in hepatocytes. Whatever the mechanism of EGF action is, HGF action on hepatocyte calcium metabolism resembles that of EGF. The present results also point out that actions of EGF and HGF are different in some respects. First, despite that two factors increase inositol trisphosphate and [Ca2+lc, EGF induced two responses to greater extent. Second, reduction of extracellular calcium reduced [Ca2+l, response to EGF but not to HGF. We interpret these data that sources of calcium mobilized by EGF are both intracellular pool and extracellular origin while the action of HGF on [Ca2+l, depends largely on intracellular pool. Our results do not exclude a possibility that HGF stimulates calcium entry, but suggest that HGF has only a small effect on hepatocyte calcium gating apparatus at least in an early time point. These results also raise a possibility that EGF increases calcium entry by a mechanism independent of inositol trisphosphate since these two growth factors increases inositol trisphosphate. It is noteworthy that HGF-induced elevation of [Ca2+], was enhnaced in the presence of EGF. This is presumably due to augmentation of calcium entry since such enhancement is not observed in low calcium-containing medium. Therefore, EGF and HGF act synergistically to stimulate calcium entry. This is of interest since calcium entry is critical in the action of EGF on cell-cycle progression in hepatocytes (Kojima, I. and Mine, T., manuscript in preparation).
References 1. 2. 3.
Nakamura, T., Nawa, K., Ichihara, A. (1984) Biochem. Biophys. Res. Commun. 122,145O Russel, W.E., McGowan, J.A., Bucher, N.L.R. (1984) J. Cell Physiol. 119, 183 Nakamura, T., Teramoto, H., Ichihara, A. (1986) Proc. Natl. Acad. Sci. USA 83,6489 1179
Vol.
181, No. 3, 1991
4. 5.
t: 8.
9 10. 11. 12. 13.
19.
;;*
28: 29.
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
Nakamura, T., Nawa, K., Ichihara, N., Kaise, N., Nishino, T. (1987) FEBS Lett. 224,311 Gohda, E., Tsubouchi, H., Nakayama, H., Hirono, S., Sakiyama, O., Takahashi, K., Miyazakl, H., Hashimoto, S., Daikuhara, Y. (1988) J. Clin. Invest. 81! 414 Zarnegar, R., Mchalopoulos, G. (1989) Cancer Research 49,3314 Nakamura, T., Nishizawa, T.., Hagiya, M., Seki, T., Shlmonishi, M., Sugimura, A., Tashiro, K., Shnnizu, S. (1989) Nature 342,440 Rubin, J.S., Chan, A.M.L., Bottaro, D.P., Burgess, W.H., Taylor, W.G., Cech, A.C., Hirschfield, D.W., Won , J., Miki, T., Finch, P.W., Aaronson, S.A. (1991) Proc. Natl. Acad. Sci. U 8 A 88,415 Tashiro, K., Hagiya, M., Nishizawa, T., Seki, T., Shimonishi, M., Shimizu, S., Nakamura, T. (1990) Proc. Natl. Acad. Sci. USA. 87,320O Zarne ar, R., Muga, S., Rahija, R., Michalopoulos, G. (1990) Proc. Natl. Acad. 5 ci. USA 87,1252 Kinoshita, T., Tashiro, K., Nakamura, T. (1989) Biochem. Biophys. Res. Commun. 1651229 Bottaro, D.P., Rubin, J.S., Faletto, D.L., Chan, A.M.L., Kmiecik, T.E., Vande Woude, G.F., Aaronson, S.A. (1991) Science 251,802 Naldini, L., Vi na, E., Narsimhan, R., Gaudino, G., Zarnegar, R., Michalopoulos, 8 .K., Comoglio, P.M. (1991) Oncogene 6,501 Carpenter, G., Cohen, S. (1979) Annu. Rev. Biochem. 48,193 Johnson, RX, Garrison, J.C. (1987) J. Biol. Chem. 262,17285 Berry, M.N., Friend, D.S. (1969) J. Cell Biol. 43,506 Mine, T., Kojima, I., Kimura, S., Ogata, E. (1986) Biochem. Biophys. Res. Commun. 140,170 Mine, T., Kojima, I., Kimura, S., Ogata, E. (1987) Biochim. Biophys. Acta 927,792 f;;l;tf., Freudenlich, C.C., Snowdowne, K.W. (1986) Am. J. Physiol. Mine, T., Kojima, I., Ogata, E. (1989) Endocrinology 125,586 Mine, T., Kojima, I., Ogata, E. (1990) Endocrinology 126,283l Snowdowne, K.W., Borle, A.B. (1984) Am. J. Physiol. 246, El98 Berridge, M.J., Irvine, R.F. (1984) Nature 312,315 Mauger, J.P., Poggioli, J., Guesdon, F.j Claret, M. (1984) Biochem. J. 221: 121 Bosch, F., Bouscarel, B., Slaton, J., Blackmore, P.F., Exton, J.M. (1986) Biochem. J. 239,523 Johnson, R.M., Garrison, J.C. (1987) J. Biol. Chem. 262,17285 Czech! M.P. (1985) Annu. Rev. Physiol. 47,357 Magm, M., Pandiella, A., Helin, K., Meldolesi, J., Bequinot, L. (1991) Biochem. J. 277: 305 Liang, M., Garrison, J.C. (1991) 266,13342
1180
BIOCHEMICAL
Comparison
of Effects
AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages 1173-1180
of HGF and EGF on Cellular
Calcium
in Rat Hepatocytes Tetsuya Mine,’
* Etsuro Ogata, and Toshikazu
Itaru Kojima,
Fourth
Department
of Internal
Tokyo
School of Medicine,
Medicine,
3-28-6 Mejirodai,
Nakamura**
University
of
Bunkyo-ku
Tokyo 112, Japan *
Institute
of Endocrinology,
Gunma
University,
Maebashi
371, Japan **
Department
of Biology, University,
Received
November
Faculty
Fukuoka
of Science,
Kyushu
812, Japan
1, 1991
We compared the effects of HGF and EGF on cytoplasmic free calcium concentration, [Ca2+l,, and inositol trisphosphate production in rat hepatocytes. HGF induced a prompt and transient elevation of [Ca2+lc. EGF also induced an immediate increase in [Ca2+lc, the magnitude of which was greater than that by HGF. In contrast, in the presence of 1 pM extracellular calcium EGF increased [Ca2+l, to a lesser extent than HGF. When cells were pretreated with EGF, the effect of HGF on [Ca2+l, was greatly enhanced. However, such enhancement was not observed in medium containing 1 pM extracellular calcium. In hepatocytes prelabeled with PHIinositol, both HGF and EGF increased [3Hlinositol trisphosphate. HGF and EGF acted synergistically to stimulate production of inositol trisphosphate. These results indicate that both HGF and EGF increase [Ca2+lc by a mechanism involving phosphoinositide turnover and that the actions of HGF 0 1991 and EGF on hepatocyte calcium metabolism are not totally identical. Academic
Press,
Inc.
The activity of hepatocyte growth factor (HGF) was first convincingly demonstrated in serum and platelets (l-3). Subsequently, the factor was purified from several sources including rat platelets (4), human plasma (5,6) and rabbit serum (6). The structure of HGF revealed by molecular cloning has significant homology with plasminogen and related serine proteases (7). Despite that HGF is a potent stimulator of cell growth in hepatocytes, expression of HGF mBNA is observed in several organs (8-11). This raises a 1 To whom correspondence
should
be addressed. ooo6-291x/91 1173
$1.50
Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
Vol.
181, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
possibility that HGF may also have an action as a paracrine growth factor in various cells. In target cells, HGF induces tyrosine phosphorylation of cellular proteins (8) and it is shown recently that HGF receptor possesses an activity of tyrosine kinase (12,13). Epidermal growth factor (EGF) has been known to stimulate proliferation of hepatocytes. When added concomitnatly, the actions of HGF and EGF are additive (3,14). Since EGF receptor also possesses tyrosine kinase activity, it is of interest whether or not the signal transduction mechanisms for HGF and EGF are identical. In this regard, EGF causes hydrolysis of phosphoinositides and increases cytoplasmic free calcium concentration, [Ca2+lc (15). In the present study, we compared the actions of EGF and HGF in isolated rat hepatocytes. Our results demonstrate that HGF, as in the case for EGF, increases [Ca2+le by causing hydrolysis of phosphoinositides but that the action of HGF on cellular calcium metabolism differs in some respects.
Methods
and Materials
Preparation
of Parenchvmal
Cells
Male Wistar rats, wei hin about 200g were used. The parenchymal cells were prepared using t%e coEilagenase perfusion technique of Berry and Friend (16) as previously described (17,18). Then, parenchymal cells were suspended in modified Hanks’ solution (in mM): Na, 137; K, 3.5; KH2pO4, 0.44; NaHC03,4.2; Na$IP04, 0.33; CaCl2 1.0, 1.0; and HEPES (pH 7.4), 20 (equilibrated with 02 gas). Measurement
of Cstoplamic
Free Ca2+ Concentration
in Heuatocvtes
Aequorin was loaded into hepatocytes by making plasma membrane reversibly permeable by the method of Borle et al (19), as described previously (20,21). More than 95% aequorin-loaded cells extruded trypan blue. Aequorin bioluminescence was calibrated in terms of free calcium concentration, assuming a magnesium concentration of 1 mM and an even distribution of calcium in cytosol(22). Measurement
of Production
of Inositol
Trisphosuhate
Inositol phosphates were measured as previously described (15,16). Hepatocytes (lo7 cellslml) were labeled with [ Hlinositol by incubating cells with 10 pCi/ml [3H]inositol for 120 min. After the labeling period, cells were washed and incubated in modified Hanks’ solution containmg 10 mM LiCl for 10 min. Cells were stimulated for 20 set with HGF or EGF. The reaction was stopped by adding perchloric acid (final concentration: 10%). Cells were homo enized by repetitive aspirations through a 26-gauge needle and centri f uged at 800Xg for 5 min. The supernatant was neutralized with 5M KOH and applied to an anion exchange column. Inositol phosphates were separated as described by Berridge and Irvine (23). Isomers of inositol trlsphosphate were not measured. 1174
Vol.
181, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
Materials EGF was purchased from Sigma (St. Louis , MI). HGF (Human recombinant HGF: h-r HGF) was purified from culture fluid from transformed CHO cells (mouse fibroblast) as described previously (7). A.&q? was purchased from Dr. J.R. Bhnks, Mayo Foundation (Rochester, . Results Effects of HGF and EGF on [Ca2+l, in Hepatocvtes
First, the effect of HGF on [Ca2+lc was investigated. Immediately after the addition of lo-lo M HGF, a prompt elevation of [Ca2+], was observed (Fig. 1A). A rise of [Ca2+lc was composed of an immediate sharp single peak followed by a gradual decline. Approximately 1 min after the addition of HGF, [Ca2+], returned to the resting levels. The effect of HGF on ICa2+lc was detected at lo-” M. The magnitude of the response was increased in a dosedependent manner and the maximum response of [Ca2+], was obtained at lo-lo M. Second, the effect of EGF on [Ca2+l, was also investigated. [Ca2+], was elevated after the addition of 10s8 M EGF. The initial peak was higher than that of HGF (Fig. 2A). About 1 min after the addition of EGF, [Ca2+lc
m.
Comparisonof Effect of HGF on [Ca2+], in the Presenceof 1 mM and 1 uM Extracellular Calcmm Aequorin-loaded hepatocytes were stimulated by 10.lo M HGF in the presenceof 1 mM (Al or 1 pM (B) extracellular calcium. When extracellular calcium was reduced to 1 pM, the concentration of calcium was fixed by using Ca2+-EGTA buffer. Results are the representative of four experiments with similar results. Fip. 2. Effect of EGF on [Ca2+l, in the Presence of 1 mM and 1 uM Extracellular Calcium Aequorin-loaded hepatocytes were stimulated by 10-s M EGF in the presenceof 1 mM (A) or 1 pM (B) extracellular calcium. Results are the representative of four experiments with similar results. 1175
Vol.
181, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
A
B EGF
EGF HGF
I
HGF
Fip. 3. Effect of Pretreatment of EGF on HGF-induced [Ca2+]c in the Presence of 1 mM and 1 uM Extracellular Calcium Aequorin-loaded cells were stimulated b lo-lo HGF after the pretreatment with lo-@ M EGF in the presence oP 1 mM (A) or 1 pM (B) extracellular calcium. Results are the representative of three experiments.
returned to the resting level. The effect of EGF on [Ca2+], was detected at 10-r’ M and the magnitude of the response was increased dose-dcpcndcntly. Maximum response of [Ca2+], was obtained at 10-e M EGF. Then, a role of release from internal Ca2+ pool(s) in the action of IIGF and EGF was compared. In the presence of 1 pM extracellular calcium concentration, the rises of [Ca2+], induced by HGF and EGF were determined. In medium containing this concentration of calcium, basal as well as stimulated calcium influx is negligible (24). Compared to the action observed in 1 mM extracellular calcium, the peak of [Ca2+lc induced by lo-r0 M HGF was not altered in medium containing of 1 pM extracelluIar calcium (Fig. 1B). In contrast to the action of HGF, the peak of [Ca2+lc induced by 10S8M EGF was markedly decreased in medium containing 1 pM extracellular calcium (Fig. 2B). Similar results were obtained at concentrations of UPS M and lO-‘O M (data not shown). However, the rise of [Ca2+], was abolished in the presence of 1 pM extracellular calcium when the concentration of EGF was reduced to lo-” M (data not shown) . Interaction
of [Ca2+lr induced by HGF and EGF
EGF pretreatment on HGF-induced increase in [Ca2+lc was examined. When aequorin-loaded cells were first stimulated by loss M EGF and then lo-lo M HGF was added, the HGF-induced increase in aequorin bioluminescence was greater than that observed in the absence of EGF pretreatment (Fig. 3A). The same experiment was done in the presence of 1 pM extracellular calcium. In this condition, however, the HGF-induced 1176
Vol.
181, No. 3, 1991
BIOCHEMICAL
I 04
’ 10-I'
AND BIOPHYSICAL
I lo-'0 hwtiil
I to-9 CM)
104
RESEARCH COMMUNICATIONS
I 10-7
Dose-Response Relationship for EGF-and HGF.-induced Inositol Firr. 4. Trisphosphate Production Hepatoc:ytes (lo7 cells/ml) were labeled with [3Hlinositol by incubating cells with 10 pCi/ml [3H]inositcl for 120 min. After washing, cells were stimulated for 20 seewith various concentrations of HGF (0) or EGF (0) in the presenceof 10 mM LiCI. Inositol trisphosphates were determined as described in Materials and Methods. Values are the mean f SE for four determinations.
increase in aequorin bioluminescence was similar absence of EGF pretreatment (Fig. 3B). Effects of HGF and EGF on Inositol
Trisphosphate
Effects of HGF and EGF on inositol
to that observed in the
Production
trisphosphate production were cells. Changes in C3H]inositol trisphosphate were detected at 10s and was maximal at 20s (data not shown). As shown in Fig. 4, when changes in [3H]inositol trisphosphates were measured at 2Os, lo-lo M HGF induced a maximal 30% increased in [3HIinositol trisphosphate. Effect of HGF on inositol t&phosphate production was detected at 10-” M. In the case of EGF, time course of changes in C”H]inositol trisphosphate was also examined and production of inositol trisphosphate was determined at 20s. EGF, at doses higher than lo-lo M, increased inositol trisphosphate in a dose-dependent manner (Fig. 4). At lOma M, EGF-induced production of inositol trisphosphate was much greater than that induced by lo-lo M HGF. measured
using [3H]inositol-labelled
1177
Vol.
181, No. 3, 1991
BIOCHEMICAL
AND BtOPHYSlCAL
RESEARCH COMMUNICATIONS
Effect of EGF on HGF induced Inositol Trisphosphate W&ion Hepatocytes (lo7 cells/ml) were labeled with [YH]inositol as described in Materials and Methods. Inositol trisphosphates were determined at 20 set after the addition of various peftides. ml@10 M HGF; 10-g EGF; m 10-r” M HGF and lo- M EGF. Results are expresse!F= aspercent of basal response.Values are the mean f SE for four determinations.
Interaction
of Inositol
T&phosphate
Production
induced by EGF and HGF
The effect of simultaneous addition of EGF and HGF on inositol trisphosphate was examined. Compared with single addition of lo-lo M EGF or lo-lo M HGF, the actions of EGF and HGF were additive (Fig. 5). Even in the presence of 1 pM extracellular concentration, similar additivity was seen (data not shown).
Discussion The present results clearly demonstrate that HGF increases [Ca2+lc in rat hepatocytes. The elevation of [Ca2+], is due largely to mobilization of calcium from an intracellular pool and this reaction is probably brought about by HGF-induced production of inositol 1,4,5-trisphosphate. Hence, HGF activates the calcium messenger system in hepatocytes. The present results together with a recent report that HGF receptor contains tyrosine kinase activity (12,131 indicate that EGF and HGF shares similar transduction system in hepatocyes. With regard to EGF action on inositol phospholipid metabolism in hepatocytes, there still remains some controversy: Bosch et al. (25) reported that EGF did not increase production of inositol trisphosphate; whereas 1178
Vol.
181, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
Johnson et al. (26) reported that EGF elicited increases in both inositol1,4,5trisphosphate and inositol 1,3,4-trisphosphate. Our present results confirm the previous report by Johnson et al. (26). In other cell systems, EGF induces phosphoinositide turnover by activating a specific type of phospholipase C (PLC), PLC-7. It is shown that binding of EGF to its receptor leads to an association of PLC-7 to the activated receptor. Then, PLC-7 is tyrosylphosphorylated and the activity of PLC-7 is augmented by phosphorylation (27,28). The action of EGF in hepatocytes may also involve the same mechanism, However, Liang et al. (29) have shown that pertussis toxin attenuates EGF-induced production of inositol 1,4,5-trisphosphate. This raises a possibility that a pertussis toxin-sensitive G protein may be involved in a certain step of signal transduction mechanism of EGF in hepatocytes. Whatever the mechanism of EGF action is, HGF action on hepatocyte calcium metabolism resembles that of EGF. The present results also point out that actions of EGF and HGF are different in some respects. First, despite that two factors increase inositol trisphosphate and [Ca2+lc, EGF induced two responses to greater extent. Second, reduction of extracellular calcium reduced [Ca2+l, response to EGF but not to HGF. We interpret these data that sources of calcium mobilized by EGF are both intracellular pool and extracellular origin while the action of HGF on [Ca2+l, depends largely on intracellular pool. Our results do not exclude a possibility that HGF stimulates calcium entry, but suggest that HGF has only a small effect on hepatocyte calcium gating apparatus at least in an early time point. These results also raise a possibility that EGF increases calcium entry by a mechanism independent of inositol trisphosphate since these two growth factors increases inositol trisphosphate. It is noteworthy that HGF-induced elevation of [Ca2+], was enhnaced in the presence of EGF. This is presumably due to augmentation of calcium entry since such enhancement is not observed in low calcium-containing medium. Therefore, EGF and HGF act synergistically to stimulate calcium entry. This is of interest since calcium entry is critical in the action of EGF on cell-cycle progression in hepatocytes (Kojima, I. and Mine, T., manuscript in preparation).
References 1. 2. 3.
Nakamura, T., Nawa, K., Ichihara, A. (1984) Biochem. Biophys. Res. Commun. 122,145O Russel, W.E., McGowan, J.A., Bucher, N.L.R. (1984) J. Cell Physiol. 119, 183 Nakamura, T., Teramoto, H., Ichihara, A. (1986) Proc. Natl. Acad. Sci. USA 83,6489 1179
Vol.
181, No. 3, 1991
4. 5.
t: 8.
9 10. 11. 12. 13.
19.
;;*
28: 29.
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
Nakamura, T., Nawa, K., Ichihara, N., Kaise, N., Nishino, T. (1987) FEBS Lett. 224,311 Gohda, E., Tsubouchi, H., Nakayama, H., Hirono, S., Sakiyama, O., Takahashi, K., Miyazakl, H., Hashimoto, S., Daikuhara, Y. (1988) J. Clin. Invest. 81! 414 Zarnegar, R., Mchalopoulos, G. (1989) Cancer Research 49,3314 Nakamura, T., Nishizawa, T.., Hagiya, M., Seki, T., Shlmonishi, M., Sugimura, A., Tashiro, K., Shnnizu, S. (1989) Nature 342,440 Rubin, J.S., Chan, A.M.L., Bottaro, D.P., Burgess, W.H., Taylor, W.G., Cech, A.C., Hirschfield, D.W., Won , J., Miki, T., Finch, P.W., Aaronson, S.A. (1991) Proc. Natl. Acad. Sci. U 8 A 88,415 Tashiro, K., Hagiya, M., Nishizawa, T., Seki, T., Shimonishi, M., Shimizu, S., Nakamura, T. (1990) Proc. Natl. Acad. Sci. USA. 87,320O Zarne ar, R., Muga, S., Rahija, R., Michalopoulos, G. (1990) Proc. Natl. Acad. 5 ci. USA 87,1252 Kinoshita, T., Tashiro, K., Nakamura, T. (1989) Biochem. Biophys. Res. Commun. 1651229 Bottaro, D.P., Rubin, J.S., Faletto, D.L., Chan, A.M.L., Kmiecik, T.E., Vande Woude, G.F., Aaronson, S.A. (1991) Science 251,802 Naldini, L., Vi na, E., Narsimhan, R., Gaudino, G., Zarnegar, R., Michalopoulos, 8 .K., Comoglio, P.M. (1991) Oncogene 6,501 Carpenter, G., Cohen, S. (1979) Annu. Rev. Biochem. 48,193 Johnson, RX, Garrison, J.C. (1987) J. Biol. Chem. 262,17285 Berry, M.N., Friend, D.S. (1969) J. Cell Biol. 43,506 Mine, T., Kojima, I., Kimura, S., Ogata, E. (1986) Biochem. Biophys. Res. Commun. 140,170 Mine, T., Kojima, I., Kimura, S., Ogata, E. (1987) Biochim. Biophys. Acta 927,792 f;;l;tf., Freudenlich, C.C., Snowdowne, K.W. (1986) Am. J. Physiol. Mine, T., Kojima, I., Ogata, E. (1989) Endocrinology 125,586 Mine, T., Kojima, I., Ogata, E. (1990) Endocrinology 126,283l Snowdowne, K.W., Borle, A.B. (1984) Am. J. Physiol. 246, El98 Berridge, M.J., Irvine, R.F. (1984) Nature 312,315 Mauger, J.P., Poggioli, J., Guesdon, F.j Claret, M. (1984) Biochem. J. 221: 121 Bosch, F., Bouscarel, B., Slaton, J., Blackmore, P.F., Exton, J.M. (1986) Biochem. J. 239,523 Johnson, R.M., Garrison, J.C. (1987) J. Biol. Chem. 262,17285 Czech! M.P. (1985) Annu. Rev. Physiol. 47,357 Magm, M., Pandiella, A., Helin, K., Meldolesi, J., Bequinot, L. (1991) Biochem. J. 277: 305 Liang, M., Garrison, J.C. (1991) 266,13342
1180
