egf responsiveness of hepatocytes after partial hepatectomy

Cell. Signal. Vol. 8, No. 8, pp. 555-559, 1996 ISSN 0898-6568/96 $15.00 Copyright © 1996 Elsevier Science Inc. Pll S0898-6568(96)001114

ELSEVIER

EGF Responsiveness of Hepatocytes After Partial Hepatectomy

Maria Marino,* Silvana Spagnuolo, Matteo Cavallini, Fulvia Terenzi, Maria Teresa Mangiantini and Silvia Leoni

DIPARTIMENTO DI BIOLOGIA CELLULARE E DELLO SVILUPPO, UNIVERSITA "LA SAP1ENZA," P.LE A. MORO, 5--00185--ROMA, ITALY.

ABSTRACT. We investigate the effect of EGF on IP3 production, PLCy phosphorylation, calcium transients in rat hepatocytes isolated in quiescent liver (Go phase of cell cycle) and at 4 h (Gl phase of cell cycle) and 24 h (M phase of cell cycle) after partial hepatectomy. Our results show that EGF does not utilize IP3 and calcium as its signal transduction molecules when the hepatocytes are in vivo stimulated to entry in the cell cycle. In particular the growth factor does not phosphorylate PLC',/and induces a decrease in IP3 content. These data sug- gest that EGF utilizes different signal transduction to send information from receptor to nucleus during PH with respect to the quiescent liver. Copyright © 1996 Elsevier Science Inc. CELt SmNAL 8;8:555--559, 1996.

KEY WORDS. Signal transduction, Growth factor, PLCy, Inositol phosphates metabolism, Calcium transients, Partial hepatectomy (rat hepatocytes)

I N T R O D U C T I O N

DNA synthesis of quiescent hepatocytes can be induced in

vivo by various conditions, including hepatic injury, expo- sure to chemical agents, and partial hepatectomy (PH) [1]. After PH in rat there is an extremely rapid and synchro- nized cell proliferation that culminates 16 h after surgery in DNA synthesis and 24 h after surgery in mitosis. The entry of the hepatocytes in cell cycle is followed by increase in the expression of a number of genes including c-fos, c-jun, c-myc and c-H-ras genes [2]. Furthermore, each phase of cell cycle is characterized by a different production of inositol phosphates [3, 4]. The nature and origin of the factors that induce these early changes in protooncogene expression are unknown. It has been postulated that EGF plays a role in promoting the earlier stages of liver regeneration, and a recent work demonstrated that the removal of salivary glands, the main EGF source, before PH prevents rat liver regeneration with a complete inhibition of DNA synthesis and an alteration in c-myc expression. The addition of EGF restored both DNA synthesis and c-myc expression [5].

In addition to the effect of EGF on DNA synthesis [6], several studies demonstrate that it regulates various metabolic events as gluconeogenesis [7]; furthermore, a role in regulating amino acid transport has been demonstrated in hepatocytes [8].

In cultured hepatocytes Yang et al. [9] showed that the occupancy of receptors by EGF triggers a rapid PLCy-medi-

* Author to whom correspondence should be addressed. Abbreviations: PLCy-phospholipase C; lP~-inositol trisphosphate; DAG-

diacyl glicerol; PH-partial hepatectomy. Received 20 December 1995; and accepted 17 June 1996.

ated hydrolysis of PIP2 interacting witlh a Gi~ protein. Hy- drolysis of PiP2 produces IP3 and DAG, both of which have second messenger roles in cell function. However, at the present, in cells induced to in vivo proliferation, our knowl- edge of the regulation by growth factors of IP3 production is limited. In fact, the studies are carried out in cultured cells stimulated by growth factors and these may be misleading in some aspects.

The aim of this work was to study the EGF effect on second messenger production in some steps of first cell cycle in rat hepatocytes; in particular we studied the production and levels of IP3, Ca + + transients and the activation of PLCy at 4 and 24 h after PH which represents respectively G1 and M phases of first cell cycle.

MATERIALS A N D METHODS Animals and Hepatocyte Isolation

Adult (150-200g) Wistar rats were used. PH was performed on male adult rats under light ether anesthesia as described by Higgins and Anderson [10]. The animals were sacrificed 4 and 24 h later. Livers were rapidly removed and cooled on ice. Hepatocytes were isolated by a perfusive method with collagenase [4]. The cell yield was 50450 × 106 hepatocytes/ g tissue and the viability index was 80%. Hepatocyte con- tamination by non-parenchimal cells was minimal.

Inositol Derivative Separation

10 X 106cells were suspended in Krebs-Bicarbonate medium and continuously oxygenated with CO2/O2 5/95% and were

556 M. Marino et al.

incubated with 1 IxCi of myo-[2-3H]inositol ( 18.7 Ci/mmol; Amersham, UK) [4]. A time course showed that the maxi- mum of myo-inositol incorporation was reached after 1 h of incubation and declined after 4 h (data not shown). After 1 h at 37°C, cells were washed twice with 0.9% NaC1, incubated for 15 min with 10 -8 M EGF; to some sample 100mM LiC1 was added 5 min before the addition of EGF. The growth factor incubation was stopped on ice, and cells were treated with 1 ml 10% TCA containing 0.1 mM EDTA. This fraction washed free of the acid with diethyl ether was analysed by anionic exchange chromatography on Dowex X1-8 resin formiate form; free inositol, IP, IP2, IP3 and IP4 were eluted from small glass column with, respectively, water, 0.2 M, 0.4 M, 0.8 M and 1.0 M ammonium formiate in 0.l M formic acid [4]. Radioactivity was counted on aliquots.

[3H]-labelled inositol lipids were extracted from TCA insoluble material with chloroform/methanol/0.1 M HC1 (3/ 1/1). The resulting water soluble glycerophospholipids, obtained after deacylation by mild alkaline hydrolysis of phosphoinositides, were separated as previously described on a small glass column on Dowex anion exchange resin with, respectively, water, 0.3 M nd 0.75 M ammonium formiate in 0.1 M formic acid [4]. The content of IP3 in TCA soluble fraction was determined by radioreceptor assay kit (AMER- SHAM, UK) on 10 × 106 cells/ml incubated with 10 s M EGF for 15 min.

Calcium Transients

The cytosolic free calcium concentration was assayed resus- pending 106 cells/ml with 5 IxM Fura 2/AM. After loading for 15 min at 37°C the cells were washed three times with Krebs-Henseleit solution supplemented with 0.2% bovine serum albumin. The fluorescence was monitored using a Perkin Elmer spectrofluorimeter (360 nm excitation wavelength and 510 nm emission wavelength) before and after the addition of 10 -8 M EGF [8]. The calibration values for Fmax and Fmin were obtained by permeabilization of the hepatocytes with digitonin at the final concentration of 5 I~M followed by the addition of EGTA at the final concentration of 10 mM in 20 mM Tris-HC1 (pH 8-8.5). The con-

centration of free intracellular calcium was calculated using the equation [8]

[Ca2+]i = Kd(F-Fmin)/(Fmax-F)

Phospholipase C ? Immunoblotting

Minced wet tissue (100 mg) was pretreated for 15 min with 10 -~ M EGF as described by Donaldson and Cohen [11]. After growth factor stimulation the tissue was sonicated in 600 ill of 0.125 M Tris-HC1 (pH 6.8) containing 10% SDS, 1 mM phenyl-methanesulphonyl fluoride and 5 I~l/ml leupeptin and boiled 2 rain in presence of 80 t*1 bromophenol blue/f3-mercaptoethanol/glicerol (2/2/5). 200 txg of solubilized proteins were subjected to SDS-7.5% PAGE at 200 V for 4 h. The proteins were then transferred to nitrocellulose for 2 h at 1 A. The nitrocellulose was treated with 1% bovine serum albumin (in 138 mM NaC1, 25 mM Tris/HC1, pH 8,0) and then probed at room temperature for 2 h with specific monoclonal antibody-antiphospholipase C (1 I~g/ ml, Affiniti, UK). Positive antibody reaction was visualized with alkaline phosphatase reaction.

Minced wet tissue (100 mg) pretreated with 10 -s M EGF as described before was sonicated in RIPA [50 mM NaC1, 50 mM Tris (pH 8.0), 1 mM EDTA, 1 mM EGTA, 3 mM sodium ortovanadate, lmM ammonium molibdate, 1% Non- idet NP-40, 0.5% deoxycholate, 0.1% SDS, 1 mM phenyl- methanesulphonyl fluoride and 5 bd/ml leupeptin]. 600 txg of solubilized proteins were treated with 10 ~1 of specific monoclonal antibody-antiphospholipase C and incubated on ice for 1 h. 100 I~1 Pansorbin were added and the samples rocked overnight at 4°C. After washing, proteins were subjected to SDS-7.5% PAGE as previously described, transferred to nitrocellulose and probed with specific monoclonal anti phosphotyrosine, PY20 (11xg/ml, Affiniti, UK). Positive antibody reaction was visualized with alkaline phosphatase reaction.

The amount of PLC and its phosphorylation were quanti- tated by scanning densitometry of the dried nitrocellulose.

R E S U L T S

The addition of EGF produced strong alteration in inositol phosphate production (Table 1). In particular, an increased

TABLE 1. EGF effect on myo-[2-3H]inositol incorporation in inositol phosphates

O h 4 h 24h

none + E G F none + E G F none +EGF

Ins 20637 ± 1030 21634 ± 739 18030 ± 1100 17000 ± 6119 21297 ± 993 20678 ± 3930 InsP 339 ± 70 231 ± 31 480 ± 97 260 ± 123 665 ± 155 670 ± 22 InsP2 237 ± 52 144 ± 18 771 ± 46A 550 ± 300 788 ± 126A 888 ± 212 InsP3 135 ± 19 332 ± 40* 2047 ± 266A 607 ± 302* 2012 ± 258A 803 ± 30* InsP4 161 ± 33 187 ± 22 1796 ÷ 509A 436 ± 21" 935 ± 148A 893 ± 68

10 X 106 rat hepatocytes isolated from non-operated animals (0 h), and at 4 and 24 h after partial hepatectomy (respectively 4 h, and 24 h) were incubated with 1 IxCi of myo-[2-~H]inositol in absence (none) or in presence of 10 -s M EGF (+ EGF), The reaction was stopped with 1 ml 10% TCA, inositol phosphates were analysed by anionic exchange chromatography on Dowex Xl-8 resin formiate form as described in Material and Methods section. Data expressed as dpm/106 cells/h are averages of 5 different experiments -+ S.D.

* P < 0.01 was calculated by Student's t-test with respect to non-operated (A) and unstimulated (*) samples.

EGF Signal Transduction During Proliferation

TABLE 2. EGF effect on myo-[2-3H]inositol incorporation in phosphoinositides

557

Oh 4 h 2 4 h

none +EGF none +EGF none +EGF

PIP 1869 ± 754 1221 ± 152 1331 ± 310 1506 ± 270 1425 ± 756 1269 ± 989 PIP2 2446 ± 778 1251 ± 15" 1446 ± 411 1465 ± 135 1031 ± 756 772 ± 497

l0 X 106 rat hepatocytes isolated from non operated animals (0 h), and at 4 and 24 h after partial hepatectomy were incubated with 1 btCi of myo-[2-~H]inositol in absence (none) or in presence of 10 s M EGF (+EGF). pH]-labelled inositol lipids were extracted from TCA insoluble material with chloroform/methanol/0.1 M HCI (3/1/1). The water soluble glycerophospholipids, obtained after deacylation by mild alkaline hydrolysis of phosphoinositides, were separated on a small glass column on Dowex anion exchange resin as described in Material and Methods section. Data are the mean of at least 5 different experiments.

* P < 0.01 with respect to the unstimulated samples, was calculated with the Student's t-test.

product ion of IP3 was present in non-opera ted animals (0 h) 15 min after addi t ion of 10 -~ M EGF. This t ime and this EGF concentra t ion gave the maximal response in prelimi- nary experiments.

Surprisingly, the addi t ion of growth factor inhibi ted the IP~ product ion in both phases of cell cycle considered, and IP4 produced at 4 h after surgery. O n the contrary, EGF addi t ion was unable to change the precursor incorporat ion in phosphoinosi t ides whereas a net decrease of PIP and PIP2 was present in non-opera ted animals after growth factor addi t ion (Table 2).

The Dowex chromatography is a useful method to separate inositol phosphates of several samples, but it does not allow us to discriminate between several IP3 isomers present in the cells. To obta in more information on second messenger production, we measured EGF effect on Ins(1,4,5)- P3 conten t in T C A soluble extracted from rat hepatocytes isolated in non operated animals (0 h) and at 4 and 24 h after PH.

Table 3 shows that PH is followed by an accumulat ion of this second messenger. The EGF addi t ion st imulated the Ins (1,4,5)P3 accumulat ion only in non-opera ted animals, whereas the growth factor, according to the data on its production, caused a net decrease of IP3 content in hepatocytes isolated in G, and M phases.

To evaluate if the effect of EGF was elicited with the ac- t ivat ion of specific phosphatases, we followed the incorporat ion of myoinositol in IP3 in presence of 100mM LiCI, a well known inhibi tor of inositol phosphate phosphatases.

TABLE 3. Content of inositol (1,4,5)P3 in rat hepatocytes isolated at 0, 4 and 24 h after P H

None +EGF pmol/106 cells

0 h 0.39 + 0.04 1.25 + 0.03* 4 h 1.45 _+ 0.01 0.57 -+ 0.04* 24 h 1.45 + 0.05 0.23 + 0.03*

10 x 10 ° rat hepatocytes isolated from non- operated animals (0 h), and at 4 and 24 h after partial hepatectomy were incubated at 37°C in absence (none) or in presence of 10 -8 M EGF (+EGF). The content of inositol (1,4,5)P~ in TCA soluble fraction was determined by radioreceptor assay kit. Data are averages of 5 different experiments -+ S.D. * P < 0.01 was calculated by Student's t-test.

Fig. 1 shows that the litium addit ion to hepatocytes isolated at 4 and 24 h after partial hepatectomy before exposing hepatocytes to EGF, blocked completely the growth factor effect on this second messenger production.

In spite of the EGF inability to produce IP3 during hepatic regeneration, we evaluated the growth factor effect on calcium cytosolic levels, in that Ca + + may be also released from IP~ insensitive stores, or may enter into the cells via direct act ivat ion of Ca+ +-specific channels. Fig. 2 shows that EGF addi t ion increases the Ca-- + cytosolic levels only in quiescent hepatocytes and, transiently, at 4 h after PH, whereas it was without effect 24 h after PH.

To estimate directly the role of PLC~/in EGF action during liver regeneration, the presence and the phosphoryla- t ion of PLCy was studied using monoclonal antibody.

r~ 0

ml >

o~

40

20

0

-20

-40

-60

-80

-I O0

1 O0 {

8o I

60 * . . . . * I {

I -!iiiii!ii'-~ ! i i::i ii ii J7

i : : : : : i ~ [ i :: :: i :. :::: iilliiiiiiiiiii i] i i i i i i i i i [i 1 4 H

J : i : i : i : i : i i i { ! ! i i i

- 12ili . . . .

+ , I

i * t '

i

+LiCI +EGF LiCI+EGF

FIGURE 1. Effect of LiCI on IP3 production. 10 x 106 rat hepatocytes isolated at 4 h and 24 h after partial hepatectomy (respectively 4 h and 24 h) were incubated with 1 p.Ci of myo- [2-3H]inositol. 100mM LiCI was added for 5 rain before the addition of 10 -8 M EGF (+EGF) for 10 min. The stimulation was stopped with 1 ml 10% TCA and inositol phosphates were analysed by anionic exchange chromatography on Dowex X1-8 resin formiate forms as described in Material and Methods section. Data, expressed as % variation with respect to the control _+ S.D., are the mean of at least 5 different experiments. *P < 0.01 was calculated by Student's t-test.

558

7O0

600 E

SO0

A

+EGF

400 i /

3 0 0 '

• Oh} C

seconds

700

6O0

"O s0o

4O0

3 0 0

+ E G F ~ B i

i i

seconds

# 4 h i i

700

60O

-~ so0

40O

3017

[ C

l +EGF

_, _. a.a,~._. _~ ._ t~rka .~a~ . -~ -" a.

L

x 24 h

FIGURE 2. EGF effect on calcium release. The cytosolic free calcium concentration was assayed in I06 cells/ml of Fura 2/AM. After loading for 15 min at 37"C the cells were washed and the fluorescence was monitored before and after the addition of 10 -s M EGF. Non-operated animals (O, panel A) and at 4 (panel B) and 24 h (panel C) after PH. Data are the mean of at least 4 different experiments. Standard Deviation did not exceed 12%.

2 4 h 4 h 0

1 4 8 k D a

M. Marino et al.

TABLE 4. Densitometric analysis of PLC~/ distribution in rat liver at O, 4 and 24 h after PH

Arbitrary units

0 h 0.60 _+ 0.01 4 h 0.79 -+ 0.01" 24 h 0.57 -- 0.02

After growth factor stimulation the tissue was sonicated and 200 Ixg of solubilized proteins were subjected to SDS-7.5% PAGE. The proteins were then transferred to nitrocellulose and then probed at room temperature for 2 h with specific monoclonal antibody-antiphospholipase C. Positive antibody reaction was visualized with alkaline phosphatase reaction. Data are averages of 4 separate immuno blots + S.D. * P < 0.01 was calculated by Student's t- test.

Fig. 3 shows the Western blot of PLCy in rat liver. A net increase of enzyme was present 4 h after PH, the enzyme level returned near the non-operated values 24 h after surgery (Table 4). EGF did not change the enzyme levels (data not shown) and was unable to stimulate the enzyme phosphorylat ion (Fig. 4) ei ther at 4 h or at 24 h after surgery, whereas enzyme phosphorylat ion was present in non operated animals after EGF addi t ion (Table 5).

D I S C U S S I O N

Our data show that in quiescent freshly isolated hepatocytes the EGF addi t ion causes PLC~/ phosphorylat ion and increases the IP3 levels. Furthermore, the growth factor induces a rapid calcium release. These data are in good agreement with the data obtained in cultured hepatocytes by Yang et al. [9].

As the hepatocytes are in vivo stimulated to enter into cell cycle, strong variations are present in the inositol phosphate metabolism. In fact, the results show an accumulat ion of second messenger in both steps of cell cycle considered, probably related to the effect of various mitogenic stimuli present in the serum of partially hepatectomized rats.

2 4 h + 24h 4 h + 4 h O+ O

q 148 kDa

FIGURE 3. PLC~/distribution in rat liver in non operated animals (0), 4 (4 h) and 24 h (24 h) after PH. Minced wet tissue of I00 nag was pretreated for 15 min with 10 -s M EGF. After growth factor stimulation the tissue was sonicated and 200 rag of sotubilized proteins were subjected to SDS-7.5% PAGE. The proteins were then transferred to nitrocellulose and then probed at room temperature for 2 h with specific monoclonal antibody- antiphospholipase C. Positive antibody reaction was visualized with alkaline phosphatase reaction.

FIGURE 4. EGF effect (+) on PLC~/ phosphoryhtion in rat liver in non-operated animals (0), 4 (4 h) and 24 h (24 h) after PH. Minced wet tissue of 100 mg was pretreated with 10 -s M EGF and sonicated in RIPA. Solubilized proteins of 600 mg were treated with specific monoclonal antibody-antiphospholi- vase C and incubated on ice for 1 h. Pansorbin were added and the samples rocked overnight at 4°C. After washing, proteins were subjected to SDS-7.5% PAGE as previously described, transferred to nitrocellulose and probed with specific monoclonal antiphosphotyrosine, PY20. Positive antibody reaction was visualized with alkaline phosphatase reaction.

EGF Signal Transduction During Proliferation 559

TABLE 5. Densitometric analysis of EGF effect on PLC3' phosphorylation in rat liver at O, 4 and 24 h after PHH

None +EGF lO-SM arbitrary units

0 h 0.190 _+ 0.04 0.510 + 0.03* 4 h 0.030 _+ 0.06 0.020 + 0.04 24 h 0.015 -+ 0.05 0.020 - 0.06*

After growth factor stimulation (+EGF) the tissue was sonicated in RIPA. 600 v-g of solubilized proteins were treated with specific monoclonal antibody-antiphospholipase C and incubated on ice for 1 h. Pansor- bin was added and the samples rocked overnight at 4°C. After washing proteins were subjected to SDS- 7.5% PAGE, transferred to nitrocellulose and probed with specific monoclonal anti phosphotyrosine, PY20. Positive antibody rreaction was visualized with alkaline phosphatase reaction. Data are averages of 4 separate immuno blots -+ S.D. * P < 0.01 was calculated by Student's t-test.

Furthermore, the behaviour of EGF in producing second messengers is quite different in hepatocytes isolated in G1 and M phases with respect to the control. In fact, during both G I and M phases of cell cycle EGF did not phosphorylate PLCy. This uncoupling between EGF and signal trans- ducer enzyme does not seem related to its receptor. In fact, it has been reported that the receptor binding affinity for EGF does not change during liver regeneration, and the number of receptors present in hepatocyte membranes var- ies in a biphasic pattern after PH. In the first 3 h after surgery in rats, the number of receptors doubles and remains constant until 6-8 h after operation; after this stage there is a large decrease in receptor number that persists until at least 72 h after the operation. The initial increase in receptor number occurs in parallel with the increase of EGF receptor-dependent tyrosine kinase activity [12]. Even if EGF receptor and PLC~/activity were uncoupled, our data show that the hepatocytes isolated at 4 h and 24 h after PH are not insensitive to EGF; in fact, the growth factor addition decreased IP3 levels. The presence of 100 mM of LiC1 completely abolished the EGF decrease of IP3, suggesting the involvement of inositol phosphate phosphatases. Even if the effect of EGF in decreasing 1P3 levels, the addition of growth factor stimulated at these steps of cell cycle the calcium release, probably operating on IP3 insensitive calcium stores or opening opportune calcium gates.

A different hormone responsiveness during hepatic regeneration has also been reported for other substances such as vasopressin and norepinephrine. A decreased sensitivity of PLC activity to vasopressin was reported [4] in hepato-

cytes only at 4 h after surgery; the enzyme sensitivity to hormone was restored 24 h after PH. Moreover, the norepinephrine blood concentration and the receptor number increase in the first hours after PH, but the hormone was unable to stimulate inositol phosphate production via PLC[3 [11.

The effect of EGF in decreasing inositol phosphates production has never been reported in proliferating hepatocytes. The physiological role played by EGF in decreasing IP3 levels in in vivo proliferating hepatocytes remains to be defined. Probably this growth factor could act as negative modulator on signal transduction stimulated by other hor- mones. This role of EGF has been reported in other models. In fact, it has been reported that EGF can modulate the IPs and cAMP accumulation in LH-stimulated ovarian cells [13]; other reports indicate that EGF is able to counteract the effect of insulin in hepatocytes [14].

In conclusion, the hepatocyte entry in cell cycle is followed by uncoupling between EGF receptor and PLCy activity. Further studies are in progress to establish if the reported uncoupling between EGF and its signal transduction molecules modifies other EGF-stimulated metabolic path- ways in hepatocytes, and how the growth factor regulates second messenger decrease.

This work was supported by grant 60% Ateneo. We would thank Dr. Fabio Pulcinelli for his helpful suggestions and assistance on calcium in- traceUular level assay.

References 1. Michalopoulos G. K. (t990) FASEB J. 4, 176-187. 2. Thompson N. L., Mead J. E., Braun L., Goyette M., Shank

P. R. and Fausto N. (1986) Cancer Res. 46, 3111-3117. 3. Marino M. (1990) Med. Sc/. Res. 18, 691-692. 4. Marino M., Mangiantini M. T. Sapgnuolo S., Luly P. and

Leoni S. (1992)J. CeU. Physiol. 152, 403-409. 5. Jones D. E., Tran-Patterson R., Cui D., Davin D., Estell K. P.

and Miller D. M. (1995) Am. J. Phisiol. 268, G872-G878. 6. McGowan J. A., Strain A. J. and Bucher N. L. R. (1981) J.

Cell. Physiol. 180, 353-363. 7. Quintana I., Grau M., Moreno F., Soler C., Ramirez I. and So-

ley M. (1995) Biochem. J. 308, 889-894. 8. Leoni S., Spagnuolo S., Marino M., Terenzi F., Massimi M.

and Conti Devirgiliis L. (1993) J. Cell. Physiol. 155,549-555. 9. Yang L. J., Rhee S. G. and Williamson J. R. (1994) J. Biol.

Chem. 269, 7156-7162. 10. Higgins G. and Anderson R. M. (1931 ) Arch. Pathol. 12, 186-

193. 11. Donaldson R. W. and Cohen S. (t992) Proc. Natl. Acad. Sci.

USA 89, 8477-8481. 12. Rubin R. A., O'Keefe E. J. and Earp H. S. (1982) Proc. Natl.

Acad. Sci. USA 79, 776-780. 13. Hubbard C. J. (1994) J. Cell. Physiol. 160, 227-232. 14. Peak M. and Angius L. (1994) Eur. J. Biochem. 221,529-536.

egf responsiveness of hepatocytes after partial hepatectomy

Documents