JOURNAL OF CELLULAR PHYSIOLOGY 152:403-409 (1992)

Signal Transduction During liver Regeneration: Role of Insulin and Vasopressin M. MARINO,* M.T. MANGIANTINI, S. SPAGNUOLO, P. LULY, AND S. LEON1 Department of Cellular and Developmental !?io/ogy, University of Rome ”La Sapienza,” 00785 Rome, (M.M., M.T.M., S.S., S.L.), arid Department of Biology, University of Rome “Tor Vergata,” 00773 Rome (P.L.), ltaly The relationship between cell proliferation and inositol lipid turnover has been studied by comparing the steady state of inositol derivative metabolism in quiescent and regenerating rat hepatocytes isolated at 4 h (GI phase of first cell cycle) and 24 h (onset of M phase) after partial hepatectomy. The effect of two hormones able to regulate hepatic regeneration, insulin and vasopressin, has been considered, and the results can be summarized as follows: (i)at 4 h after partial hepatectomy, the precursor incorporation into inositol polyphosphates and the particulate phospholipase C activity increase with respect to quiescent hepatocytes, whereas the content of II,4,5P3 does not change, suggesting an increased turnover of this molecule in this step of cell cycle priming; (ii)24 h after partial hepatectomy, the radioactivity linked to IP, and IP, as well as soluble and particulate phospholipase C activity, and IP, content increase, suggesting the presence, at the onset of M phase, of second messenger accumulation; (iii)only 24 h after partial hepatectomy, the inositol derivative metabolism is affectcd by vasopressin; and (iv) insulin exerts a modulatory role on inositol polyphosphate production without involving membrane-bound PLC activity or phosphoino5itide hydrolysis. These data suggest that inositol-derived signal molecules are associated with hepatic regeneration; moreover, the metabolic pathway of such compounds seems to be regulated so that only specific inositol phosphates are present in each step of the cell cycle. 0 1992 Wilev-Liss, Inc

Control of cell proliferation by growth factors depends upon a n orderly transfer of information from the surface into the nucleus. A vast body of evidence has emerged in the last few years, indicating a relationship between metabolism of inositol derivatives and DNA synthesis (Carney et al., 1985; Matuoka et al., 1988; Michell, 1989). Several mitogens, such as platelet-derived growth factor (PDGF) (Hepler et al., 1990), bombesin (Rozengurt, 1989), epidermal growth factor (EGF) (Moolenaar et al., 1987), and vasopressin (Williamson and Hansen, 1987) stimulate phosphatidylinositol (PI) turnover through phosphatidylinositol 4,5-bisphosphate (PIP,) breakdown, via phospholipase C (PLC) when added to cultured cells, yielding the putative second messengers, inositol 1,4,5-trisphosphate (IP,) and diacylglicerol (DAG). IP, elevates cytosolic calcium by stimulating its release from endoplasmic reticulum stores, and DAG directly activates a calcium and phospholipid-dependent protein kinase C (Rana and Hokin, 1990). Even if these data support that the phosphoinositide turnover constitutes a crucial step in cell growth, it is not well understood whether this biosynthetic pathway is a n obligatory or regulatory event in stimulating normal or oncogenic growth. But the difficulty in clarifying this issue depends on the need to obtain cellular population, which differs only for mitotic activity. 0 1992 WILEY-LISS. INC.

A useful proliferative in vivo model which allows the study of metabolic activity of dividing cells with respect to quiescent ones is represented by regenerating rat liver. Hepatocytes of adult animals can be easily induced to enter the replicative cycle after removal of two-thirds of the liver. The high degree of synchronization of the first cell cycle, in this model, allows a good correlation between metabolic variations and cell cycle phases (Bucher and Malt, 1971). Moreover, liver regeneration is characterized by the appearance of factors in the peripheral blood which stimulate DNA synthesis, utilizing various second messengers (cyclic AMP, Calcium ions, IP3). During liver regeneration, few peptide growth factors (epidermal growth factor, EGF; transforming growth factor, TGFru; hepatic growth factor, HGF) induce DNA synthesis (Michaelopoulos, 1990) in the presence of other substances which can modulate their effect on DNA synthesis.

Received October 1,1991; accepted February 25, 1992.

*To whom reprint requestsicorrespondence should be addressed at Dipartimento di Biologia Cellulare e dello Sviluppo, Universita’ di Roma “La Sapienza”, P.le A. Moro, 5-00185-Rome,Italy.

404

MARINO ET AL

Particularly intriguing is the role of insulin and vasopressin in liver proliferation. In fact, liver regeneration is impaired in r a t strains congenitally deficient in the production of vasopressin (Russel and Bucher, 19831, and pancreatectomy decreases DNA synthesis during liver regeneration (Leffert et al., 1979). These two hormones seem to work at different steps of the cell cycle: vasopressin probably cooperates with EGF and HGF in the transition between GO and G1 phase of first cell cycle, while insulin works in the presence of TGFa for hepatocyte transition into the S phase (DNA synthesis) (Leffert et al., 1979). The effects of both hormones in transition between G2 and M phase (mitosis) is unknown. In experiments reported herein, we observed the steady state metabolism of inositol derivatives in quiescent and proliferating hepatocytes 4 and 24 h after partial hepatectomy (G1 and onset of M phase of first cell cycle) (Trentalance e t al., 1984). Furthermore, we analyzed the effect of two potent comitogens for hepatocytes, insulin and vasopressin. For these hormones, in quiescent hepatocytes, differing behaviours have been reported in regulating second messenger production: while vasopressin utilizes IP, for transducing its signal (Williamson and Hansen, 19871, insulin does not (Marino et al., 1991).

MATERIALS AND METHODS Animals and hepatocyte isolation Male Wistar rats (150-200 g) were housed under a standard lighting regimen and allowed free access to food and water ad libitum. The animals were allowed 7-10 days acclimatization before surgery. Sham operations and partial hepatectomy (removal of approximately two-thirds of liver mass) were performed under light ether anaesthesia (Higgins and Anderson, 1931). Four and 24 h after surgery, the rats were anaesthetized with intraperitoneal injection of Farmotal(10 mg/ 100 g body wt. Farmitalia, Italy). Hepatocytes were obtained from livers of control and partially hepatectomized rats according to the method of Moldeus et al. (1978) using collagenase perfusion (Boehringer, Germany). The cell yield was 40-50 x lo6 hepatocytesig of liver and the viability index was 90%. Contamination of hepatocytes by nonparenchimal cells was minimal. Myo-inositol incorporation Isolated hepatocytes were suspended a t a concentration of lo7 cellsiml in Krebs-bicarbonate medium containing 2% bovine serum albumin, and continuously oxygenated in a rotatory bath at 37°C. Cells were incu-

Abbrevaations

I IP, IP.3

FAG PIP, PIP PI PLC

inositol inositol 1,4-bisphosphate and its isomers inositol 1,4,5-trisphosphate and its isomers inositol 1,3,4,5-tetrakisphosphateand its isomers sn-1,2-diacylglycero1 phosphatidylinositol 4,5-bisphosphate phosphatidylinositol 4-phosphate phosphatidylinositol phospholipase C-phosphatidylinositol 4,5-bisphosphate specific

bated with myo12-3HHlinositol (1 pCiil0 x lo6 cells, 18.7 Ciimmol; Amersham, UK).After 1h of incubation, the reaction was stopped, the supernatant was removed, cells were washed with NaCl 0.9% and equilibrated for 15 min before hormone addition. Cells were exposed to various concentrations of [Arg8]-vasopressin (Sigma Chemical Co., St. Louis, MO) and insulin (Sigma Chemical Co., St. Louis, MO) for up to 1 and 15 min, respectively. Cells were rinsed twice very rapidly with NaCl 0.9% and treated with 1ml of 10% trichloroacetic acid (TCA) containing 0.1 mM EDTA. After 15 min in ice, the cells were centrifuged and the extraction with TCA repeated.

Separation of inositol derivatives The TCA soluble fraction washed free of the acid with diethyl ether and neutralized with NaOH (1M), was analyzed by anionic exchange chromatography on Dowex X1-8 resin (Sigma Chemical Co., MO), formiate form; free inositol, IP, IP,, IP,, and IP, were eluted from small glass columns with water, 0.2 M, 0.4 M, 0.8 M, and 1 M of ammonium formiate in formic acid (0.1 M), respectively (Downes and Michell, 1981). Radioactivity was counted on aliquots. The content of IP, in TCA soluble fractions of isolated hepatocytes was determined by radioreceptor assay kit (NEN, UK). ['HI-labelled inositol lipids were extracted from TCA insoluble material with chloroformimethanol/O.Ol M HCl (31111, vh). The resulting water soluble glycerophospholipids, obtained after deacylation by mild alkaline hydrolysis of phosphoinositides, were separated on a small glass column on Dowex anion exchange resin (Creba et al., 1983). Phospholipase C activity Two x lo6 hepatocytes were homogenized in 2 ml of ice-cold buffer containing 0.25 M sucrose, 0.5 mM DTT, and 10 mM Hepes (pH 7.4). The homogenate was centrifuged at 105,000 x g for 30 min. The membrane pellet was resuspended in sucrose buffer. Phospholiphase C activity was determined in the soluble and membrane fractions according to the procedure of Pittner and Fain (1990). Each assay contained, in a final volume of 100 pl, 100 mM Tris-HC1 (pH 7.0), 100 mM NaC1, 2 mM sodium cholate, sodium deoxycholate (1mM for soluble o r 2 mM for membrane enzyme), and 20,000 dpm (300 nM) of phosphatidylinositol[2-'H]4,5-bisphosphate (3 Ciimmol; NEN, UK) plus 20 pl of proteins (approx. 20 pg). Incubations a t 37°C were terminated after 10 min by the addition of 1.25 ml chloroformimethanol (112, v1v). The phases were separated after the addition of 0.5ml of cloroform and 0.5 ml of 0.25 M HC1. The aqueous phase, analyzed by anionic chromatography, showed that the main product was IP,. RESULTS Effect of partial hepatectomy on inositol derivative production No significant difference was detectable between sham operated and nonoperated animals. Therefore, the nonoperated animals were considered as controls.

LlVER REGENERATION AND INOSITOL DERIVATIVE METABOLISM TABLE 1. Effect of partial hepatectomy on phosphoinositide metabolism','

C 4 PH 24 PH

PI

PI!?

PIPg

21,563+ 115 39,761"? 375 41,274"2 2,156

1,048 ? 622 1,7122 543 1,232? 1,056

354 ? 207 208 2 125 144 -t 132

'Hepatocytes isolated 4 h (4 PHI and 24 h 124 PHI after partial hepatedomy, and from the control ( C )were incubated with [3H]myo-inosltol; phosphoinositides were separated after dcncylation of lipid extract as described in the text. 'Data. expressed as dpm/107 cells, are averages of at least 10 independent experiments ? SD. *P .C 0.01 was calculated with Student's t-test with respect ta the control.

TABLE 3. Effect of partial hepatectomy on inos1tol-1,4,5trlsphosphate level in control (C) rat liver and 4 h (4PHI and 24 h (24PH) after partial hepatectomy' IP3 prnolk wet liver

C

275 75 160 t 40 YO0 -t 194% +

4 PH 24 PH

*

'The data are averages of at least 5 independent experimcnts SD. *P < 0.01 was calculated with Student's t-test with respect to the control

Partial hepatectomy causes increased precursor incorporation into phosphatidylinisitol without significant effect on other phosphoinositide synthesis in the first cell cycle of isolated hepatocytes (Table 1). Significant changes were present in the inositol phosphate profiles during early steps of liver regeneration (Table 2). In particular, a t 4 h after partial hepatectomy (G1 phase), the precursor incorporation into IP,, IP,, and IP, increased with respect to the control, and the radioactivity linked to I P decreased significantly. The profile distribution of radioactivity present 24 h after surgery (M phase) was: IPS and IP, increased, IP decreased with respect to the control. The IP, levels in regenerating rat liver were assayed with a rapid radioreceptor assay (Table 3). Increased level of this second messenger is present only 24 h after surgery, whereas no significant changes are present after 4 h (G1 phase) from partial hepatectomy.

Effect of insulin and vasopressin on myo-inositol incorporation Hepatocytes were exposed to hormones for various times: the optimal response of precursor incorporation in quiescent liver was obtained after hepatocyte exposure to vasopressin for 1min (according to Pittner and Fain, 1990), and to insulin for 15 min (data not shown). The effect of several concentrations of insulin on [3Hlmyo-inositol incorporation into TCA soluble and lipid extract of rat hepatocytes isolated a t 4 and 24 h

405

after partial hepatectomy is shown in Table 4.Incubation with M insulin for 15 min increases the precursor incorporation into TCA soluble fractions only 4 h after surgery (Table 4). Vasopressin did not change the inositol incorporation on either TCA soluble or lipid extracts (Table 5).

Effect of insulin and vasopressin on inositol derivative production During hepatic regeneration, insulin and vasopressin did not change the precursor incorporation into phosphoinositides at any of the hormone concentrations considered (data not shown). The analysis of TCA soluble fractions on Dowex 1-X8 resin showed that insulin changes the radioactivity distribution profile of inositol phosphates (Fig. 1). In particular, hormone concentrations from 10 -9 to lop7M decrease the radioactivity linked to IP, and IP, in regenerating liver, whereas increased radioactivity is linked to free inositol. At 4 h after partial hepatectomy, vasopressin increases myo-inositol incorporation only into IP, while a t 24 h after surgery, the radioactivity linked to inosito1 polyphosphates increases (Fig. 2). Phospholipase C activity The effect of hepatic proliferative processes and hormones on inositol derivatives was directly studied with the assay of phospholipase C-PIP, specific activity (PLC, Table 6). Membrane fractions obtained from regenerating hepatocytes show an increased PIC-PIP, specific activity, whereas cytosolic PLC activity increases only 24 h after surgery. Insulin (10-'M) has no effect on membrane-bound enzyme activity either in quiescent or in regenerating hepatocytes; on the contrary, the hormone deeply decreases the soluble enzyme activity in normal and regenerating liver 24 h after partial hepatectomy. According to Pittner and Fain (1990) arginine-vasopressin (lopsM) activates both soluble and membrane-bound PLC activity in quiescent hepatocytes. An increased enzyme activity is present in membrane fractions isolated at 24 h after partial hepatectomy. None of the hormones taken into consideration were able to change PLC activity of hepatocytes isolated a t 4 h after surgery.

DISCUSSION The results reported in this study indicate that the incorporation of myo-inositol into second messengers IP, and IP, is higher at the early stages of liver regeneration; furthermore, the reported variations of inositol polyphosphate distribution suggest that hepatocytes

TABLE 2. Effect of partial heoatectomv on inositol phosphates metabolism'.'

C 4 PH 24 PH

I

IP

IP,

IP3

IP4

26,026t 1,765 23,6472 555 27,241* 497

3,135f 943 625" t 19 828" 2 132

517 t- 240 1,000"? 56 927 + 198

426 ? 40 2,333"t 130 1,735"-t 210

456 % 29 1,249"2 39 925% i 54

'Hepatocytes isolated 4 h (4 I'H) and 24 h (24 PHI aRer partial hepatectomy. and from the control (C) were incubated with !BHlmyo-inositol;inositol phosphates were separated from TCA soluble fraction as described in the text. 'Data, expressed as dpm/107cells, are averages of at least 10 independent experiments ? SD '*P< 0.01 was calculated with Student's t-test with respect to the control.

MARINO ET AL.

406

TABLE 4. Effect of insulin treatment (15 min) on myo-inositol incorporation into TCA soluble fraction (TCA) and lipid extract (Lipids)',' 4 PH

Insulin IM) None 10-9 10-8 10

24 PH

TCA

Liuids

TCA

20,811 4 10,144 30,083 ? 1,163 29,470 t 10,925 40,860" 2 1,981

41,766 + 3,076 47,672 ? 15,107 41,628 ? 9,626 40,940 i 8.760

33,140 23,540 29,830 25.340

-t +

? +

Linids

8,430 11,570 15,000 9,879

44,003 799 40,019 ? 7,132 37,015 t 3,415 58.175 k 18.401 +

'Hepatwytes. isolated 4 h (4 PHI and 24 h (24 PJl) after partial hepatectomy. were incubated with [3Hlmyo-inositol as described in the text. 'Data, expressed as dpm/lO' cells, are averages or at leash 5 independent experiments f SD. *P :< 0.01 was calculated with Student's t-test w i t h respect lo corresponding controls in the absence of insulin.

TABLE 5. Effect of vasopressin (AVP) treatment i 1 min) on myo-inositol incorporation into TCA soluble fraction iTCA) and lipid extract (Lipids)',2 24 PH

4 PH

AVP (M) None 10-9 10-8

10-7

TCA

Lipids

TCA

Linids

20,811 ? 10,144 19,600 t 963 24,240 ? 9,925 28,480 f 1,981

41,766 t 3,076 51,824 2 3,267 49,865 f 13,065 46,492 ? 5,431

33,140 t 8,430 39,540 i- 9.975 41,620 t 11,180 40,324 .t 1,981

44,003 799 45,762 i_ 1,342 48,660 t 2,114 48,560 t 16,514

*

'Heyatocytes isolated 4 h (4 PH) and 24 h (24 PHI after partial hepatectomy. were incubated with 13Hlmyo-inositol as described in the text. 'Data, expressed a s dpmIlO' cells, are averages of a t least 5 independent experiments + SD.

a 90

I

*

4 Pt

I

k?

loo

*

I

90

TI

10

TI

IP

I

IP3

IP2

IPA

7 1

0

s 10

0

0

0

lo-' M

M

lo-'

M

Insulin

0

lo-'

M

M

lo-' M

Insulin

Fig. 1. Effect of various concentrations (M) of insulin on percent distribution radioactivity into inositol phosphates a t (left)4 (4 PH) a n d (right) 2 4 h (24 PHI after partial hepatectomy. Data a r e averages of a t least 10 independent experiments. Standard deviation never exceeded 10%. *P

Signal transduction during liver regeneration: role of insulin and vasopressin.

The relationship between cell proliferation and inositol lipid turnover has been studied by comparing the steady state of inositol derivative metaboli...
687KB Sizes 0 Downloads 0 Views