EXPERIMENTAL
CELL
RESEARCH
200,
295-300
(19%)
Phosphoenolpyruvate Carboxykinase and Glucose-6-phosphate Dehydrogenase Expression in Fetal Hepatocyte Primary Cultures under Proliferative Conditions CRISTINAMOLERO,ANGELAM.VALVERDE,MANUELBENITO,ANDMARGARITALORENZO' Departamento
de Bioqu;mmica
y Biologh Molecular, Ciudad Uniuersitaria,
Centro Mixto C.S.I.C.IU.C.M., 28040 Madrid, Spain
Press,
de Farmacia,
zyme involved in providing ribose-&phosphate for DNA synthesis, has been related to the proliferative state of hepatocytes. In fact, EGF caused an induction of GGPD in cultured hepatocytes from adult rats [2]. However, the role of this enzyme in fetal hepatocyte proliferation has not yet been studied. On the other hand, fetal hepatocytes in culture may express specific genes characteristic of the adult state. Thus, maturation of fetal hepatocytes can be monitored by the induction of enzymes of postnatal expression as phosphoenolpyruvate carboxykinase (PEPCK). PEPCK expression is virtually absent in the fetus: enzyme activity [3] and mRNA [4] accumulate shortly after birth, when transcription of the gene is activated by CAMP [5]. Studies on fetal rat hepatocytes have shown a PEPCK-specific activity induction by epinephrine and glucagon [6]. Also, an accumulation of PEPCK mRNA in the presence of dexamethasone, CAMP, and triiodothyronine has been described on embrionic rat hepatocytes [7]. However, the effect of factors that induce hepatocyte proliferation on PEPCK expression has not yet been explored. Accordingly, the aim of this work was to establish the possible relationship between GGPD expression (as a marker of hepatocyte proliferation) and PEPCK expression (as a marker of hepatocyte maturation) in 20-day fetal hepatocyte primary cultures.
Fetal hepatocytes cultured for 64 h in the presence of glucagon and dexamethasone maintain a quiescent state, showing a low expression of glucose-6-phosphate dehydrogenase (GGPD) and a high induction of phosphoenolpyruvate carboxykinase (PEPCK). Under these culture conditions, the presence of EGF produced hepatocyte proliferation, with a concomitant increase of DNA synthesis, DNA content, and GGPD expression, meanwhile the expression of PEPCK was drastically reduced. The presence of forskolin plus IBMX nearly suppressed the increase in DNA synthesis and GBPD expression induced by EGF, showing a very high expression of PEPCK. Accordingly, it is possible to establish an inverse relation between GBPD, highly expressed in proliferating fetal hepatocytes, and PEPCK expression, highly expressed in quiescent fetal hepatocytes under specific hormonal stimulation. Q 1992 Academic
Facultad
Inc.
INTRODUCTION Liver has been subjected to many studies of proliferation, and it has served as a model for differentiation, mostly in the adult. However, during fetal development, growth factors and hormones must play a significant role in regulating the orderly progression of liver growth and differentiation. Hepatocytes of fetal rats have been shown to undergo division in primary culture. Under defined conditions, EGF plus insulin initiate DNA synthesis, liver-specific functions as albumin or a-fetoprotein expression remaining unaffected [l]. Moreover, the expression of glucose-6-phosphate dehydrogenase (G6PD),’ an en-
MATERIALS
AND METHODS
Chemicals. [3H]Thymidine, [a-32P]dCTP (3006 Ci/mmol) and the multiprimer DNA labeling system kit were purchased from Amersham. Cycle test reagent kit was from Becton-Dickinson (San Jose, CA). Rat GGPD cDNA (pKs-GGPD), PEPCK cDNA (pcKlO), and rat &-microglobulin cDNA (pGEM-&-MG) were kindly provided by Dr. Ho (Durham, NC), Dr. Hanson (Cleveland, OH), and Dr. Soprano (Philadelphia, PA), respectively. Insulin, adrenaline, glucagon, dexamethasone, forskolin, IBMX, EGF, and BSA were from Sigma (St. Louis, MO). IGF-I was from Amgen Biologic& (Thousand Oaks, CA). FCS, trypsin-EDTA, and culture media were from Flow Laboratories. Methods. Hepatocytes were isolated from 20-day fetuses of Wistar rats as described [8] and plated (1 X 10’ ceils/60-mm diam plastic dishes) in 2.5 ml of medium 199 with Earle’s salts in the presence of
’ To whom correspondence and reprints should be addressed. Fax: 34-1-5494616. ’ Abbreviations used: PEPCK, phosphoenolpyruvate carboxykinase; GGPD, glucose-6-phosphate dehydrogenase; FCS, fetal calf serum; PBS, phosphate-buffered saline; BSA, bovine serum albumin; IBMX, 3-isobutyl-1-methylxantine. 295
All
Copyright 0 1992 rights of reproduction
0014-4827/92 $5.00 by Academic Press, Inc. in any form reserved.
296
MOLERO
10% FCS and dexamethasone (100 nM). After 4 h of culture, cells were washed twice with PBS, and 70% of the initial cells were attached forming a monolayer that under light microscopy represents around 25% of confluency. So, cell density was estimated as 2.5 X 10’ cells/cm*. At this time, cells were cultured for 64 h in a serum free medium supplemented with 0.2% BSA and dexamethasone in the presence of various hormones and growth factors. The culture medium and additives were changed every 48 h. At the end of the culture, cells were washed with PBS and collected for further determinations of RNA, DNA, and enzymatic activities. Routine control experiments to check the purity of our cultures revealed the presence of albumin by immunocytochemistry, as described in a previous work [Q]. Albumin-expressing fetal hepatocytes were distinguished from fibroblastlike cells, which comprised less than 5% of the total cells. DNA synthesis was determined in duplicated dishes by [3H]thymidine incorporation (0.2 &i/ml) during the last 4 h of culture. At the end of the incubation period, radioactivity in TCA-precipitable material was determined as described [lo]. Results were expressed as a percentage of the radioactivity incorporated by the untreated cells. DNA content was determined in duplicated dishes by a fluorimetric method [ll] using calf thymus DNA as standard, and results were expressed as pgldish. Cell cycle analysis was performed in cell suspensions after trypsinEDTA detachment of hepatocytes from the culture dishes and staining of cell nuclei with ice-cold propidium iodide (125 @g/ml) in 0.1% Na, citrate for 15 min. Cells were examined on a FACScan flow cytometer (Becton-Dickinson). For determination of enzyme activities, two dishes were pooled for each individual determination. Cells were sonicated for 45 s at 1.5 mA in 1 mM EDTA, 1 mM dithiothreitol, 0.25 M sucrose, 25 mM TrisHCl, pH 7.5, and centrifuged at 12,000g for 4 min. Cytosolic supernatants were used for the radiometrical determination of PEPCK [3] and the spectrophotometrical determination of GGPD 1121, respectively. Proteins were determined by [13] with BSA as standard. PEPCK was expressed as mU/mg of protein. A milliunit was a nmole of CO, incorporated to malate/min. GGPD was expressed as mU/mg of protein. A milliunit was a nmole of NADPH formedimin. For Northern-blot analysis of RNA, four dishes were pooled for each individual determination at the end of the time of culture, after washing cells with PBS and direct lysis of the monolayer with guanidinium thiocyanate-phenol reagent [14]. Total cellular RNA was isolated as 1141, denatured in 50% formamide, 2.2 M formaldehyde, 20 n-&f Mops, pH 7.0,5 mM sodium acetate, 1 mM EDTA at 65°C for 15 min. RNAs (20 fig) were electrophoresed on 1% agarose gels containing 2.2 M formaldehyde. Resolved RNAs were blotted onto nitrocellulose filter [15], baked at 80°C for 2 h, prehybridized for 4 h, and hybridized 40 h at 42‘C in a buffer containing 50% formamide, 2X Denhardt’s solution, 5~ SSC (0.75 M NaClI0.075 M Na, citrate), 0.1% SDS, denatured salmon-testis DNA (100 pglml) with denatured labeled cDNAs. cDNA labeling was carried out with [o-32P]dCTP to a specific activity of 0.5-l X lo9 cpm/pmol by using a multiprimer DNA labeling system kit. Then, filters were washed three times with 1X SSC containing 0.1% SDS at room temperature and twice at 42°C with 0.1X SSC containing 0.1% SDS. Filters were exposed to Kodak X-Omat AR films. Relative densities of the hybridization signals were determined by densitometric scanning of the autoradiograms, being linear at all the densities studied. At the end of the procedure, filters were stained with 0.05% methylene blue to verify the positions of 28 S and 18 S rRNAs. Northern-blot analysis was performed in triplicate from three different cell preparations. The variability in the measurement of the mRNA levels after the densitometryc analysis of the filters was not greater than 15%. RESULTS
AND
DISCUSSION
Effects of Hormones and Growth Factors on Proliferation of Fetal Hepatocyte Primary Cultures
Fetal hepatocytes were cultured for 64 h in a serumfree medium supplemented with dexamethasone to re-
ET AL.
strain growth of unwanted cells [16] and prolong the survival of the hepatocytes in the absence or presence of growth factors and hormones [l]. At the end of this time, the analysis of the cell cycle in cells cultured in the only presence of dexamethasone showed the quiescence of our cells, since >94% of the cells were arrested in the GO/G1 phase of the cell cycle, as assessed by flow cytometry. DNA synthesis can be induced in these quiescent cells after 64 h of treatment with different mitogens as determined by [3H]thymidine incorporation during the last 4 h of culture (Table 1). We tested the effect of EGF (the main known growth factor for the liver) on [3H]thymidine incorporation at different cell densities. At low cell density (2.5 X lo4 cells/cm2), EGF (3.3 nM) produced a 4-fold increase in DNA synthesis as compared with untreated cells (Table 1). However, at 5 X lo4 cells/ cm2 EGF increased DNA synthesis 2-fold and at 10 x lo4 cells/cm2 no significant effect of EGF on [3H]thymidine incorporation was observed. So, as growthrelated functions are stimulated at low cell density, we chose a cell density of 2.5 X lo4 for further experiments described in this paper. We cultured fetal hepatocytes in the presence of 10% FCS as a positive control for cell proliferation. Under FCS stimulation a 3-fold increase on [3H]thymidine incorporation was observed as compared with that of untreated cells (Table 1). Both EGF and FCS produced an increase of 2.5fold and 2-fold in the percentage of cells in S phase of cell cycle, respectively, as compared with that of untreated cells. Under EGF or FCS stimulation, 8% of fetal hepatocytes were in mitosis, as determined by DNA staining with propidium iodide and flow cytometry analysis. We investigated the effect of insulin (40 nM), IGF-I (1.4 nM), vasopressin (18 r&f), or glucagon (1 j&f) on the [3H]thymidine incorporation in fetal hepatocyte primary cultures. These factors, individually, did not modify the [3H]thymidine incorporation observed in untreated cells (Table 1). The combination of these factors with EGF did not further increase [3H]thymidine incorporation as compared with the effect induced by EGF alone (Table 1). The presence of forskolin (10 PM), a direct activator of adenylate cyclase, together with IBMX (50 pLM), a CAMP-phosphodiesterase inhibitor, did not modify the [3H]thymidine incorporation in fetal hepatocyte primary cultures as compared with untreated cells. However, forskolin plus IBMX decreased [3H]thymidine incorporation in EGF-stimulated hepatocytes almost to the levels observed in untreated cells (Table 1). Moreover, DNA content at the end of the time of culture under different experimental conditions has also been included in Table 1. In the untreated cells, there was a good correlation between the DNA content and the cell number that gave a mean +_SEM of 27.6 + 3.0 pg DNA/lo’ cells and indicated that cell density in the absence of serum or signals and in the only presence of dexamethasone was constant throughout the time of
PEPCK
AND
GGPD
TABLE
297
EXPRESSION
1
Effects of Hormones and Growth Factors on [3H]Thymidine Incorporation in Fetal Rat Hepatocyte Primary Cultures
and DNA Content
[sH]Thymidine incorporation (percentage) Addition None Glucagon (1 p&f) Insulin (40 nM) IGF-I (1.4 n&f) Vasopressin (18 nM) Forskolin (10 ).rAf) + IBMX 10% FCS
None
(50 &f)
100 112 t 2 119 + 6 118? 6
CELL
RESEARCH
200,
295-300
(19%)
Phosphoenolpyruvate Carboxykinase and Glucose-6-phosphate Dehydrogenase Expression in Fetal Hepatocyte Primary Cultures under Proliferative Conditions CRISTINAMOLERO,ANGELAM.VALVERDE,MANUELBENITO,ANDMARGARITALORENZO' Departamento
de Bioqu;mmica
y Biologh Molecular, Ciudad Uniuersitaria,
Centro Mixto C.S.I.C.IU.C.M., 28040 Madrid, Spain
Press,
de Farmacia,
zyme involved in providing ribose-&phosphate for DNA synthesis, has been related to the proliferative state of hepatocytes. In fact, EGF caused an induction of GGPD in cultured hepatocytes from adult rats [2]. However, the role of this enzyme in fetal hepatocyte proliferation has not yet been studied. On the other hand, fetal hepatocytes in culture may express specific genes characteristic of the adult state. Thus, maturation of fetal hepatocytes can be monitored by the induction of enzymes of postnatal expression as phosphoenolpyruvate carboxykinase (PEPCK). PEPCK expression is virtually absent in the fetus: enzyme activity [3] and mRNA [4] accumulate shortly after birth, when transcription of the gene is activated by CAMP [5]. Studies on fetal rat hepatocytes have shown a PEPCK-specific activity induction by epinephrine and glucagon [6]. Also, an accumulation of PEPCK mRNA in the presence of dexamethasone, CAMP, and triiodothyronine has been described on embrionic rat hepatocytes [7]. However, the effect of factors that induce hepatocyte proliferation on PEPCK expression has not yet been explored. Accordingly, the aim of this work was to establish the possible relationship between GGPD expression (as a marker of hepatocyte proliferation) and PEPCK expression (as a marker of hepatocyte maturation) in 20-day fetal hepatocyte primary cultures.
Fetal hepatocytes cultured for 64 h in the presence of glucagon and dexamethasone maintain a quiescent state, showing a low expression of glucose-6-phosphate dehydrogenase (GGPD) and a high induction of phosphoenolpyruvate carboxykinase (PEPCK). Under these culture conditions, the presence of EGF produced hepatocyte proliferation, with a concomitant increase of DNA synthesis, DNA content, and GGPD expression, meanwhile the expression of PEPCK was drastically reduced. The presence of forskolin plus IBMX nearly suppressed the increase in DNA synthesis and GBPD expression induced by EGF, showing a very high expression of PEPCK. Accordingly, it is possible to establish an inverse relation between GBPD, highly expressed in proliferating fetal hepatocytes, and PEPCK expression, highly expressed in quiescent fetal hepatocytes under specific hormonal stimulation. Q 1992 Academic
Facultad
Inc.
INTRODUCTION Liver has been subjected to many studies of proliferation, and it has served as a model for differentiation, mostly in the adult. However, during fetal development, growth factors and hormones must play a significant role in regulating the orderly progression of liver growth and differentiation. Hepatocytes of fetal rats have been shown to undergo division in primary culture. Under defined conditions, EGF plus insulin initiate DNA synthesis, liver-specific functions as albumin or a-fetoprotein expression remaining unaffected [l]. Moreover, the expression of glucose-6-phosphate dehydrogenase (G6PD),’ an en-
MATERIALS
AND METHODS
Chemicals. [3H]Thymidine, [a-32P]dCTP (3006 Ci/mmol) and the multiprimer DNA labeling system kit were purchased from Amersham. Cycle test reagent kit was from Becton-Dickinson (San Jose, CA). Rat GGPD cDNA (pKs-GGPD), PEPCK cDNA (pcKlO), and rat &-microglobulin cDNA (pGEM-&-MG) were kindly provided by Dr. Ho (Durham, NC), Dr. Hanson (Cleveland, OH), and Dr. Soprano (Philadelphia, PA), respectively. Insulin, adrenaline, glucagon, dexamethasone, forskolin, IBMX, EGF, and BSA were from Sigma (St. Louis, MO). IGF-I was from Amgen Biologic& (Thousand Oaks, CA). FCS, trypsin-EDTA, and culture media were from Flow Laboratories. Methods. Hepatocytes were isolated from 20-day fetuses of Wistar rats as described [8] and plated (1 X 10’ ceils/60-mm diam plastic dishes) in 2.5 ml of medium 199 with Earle’s salts in the presence of
’ To whom correspondence and reprints should be addressed. Fax: 34-1-5494616. ’ Abbreviations used: PEPCK, phosphoenolpyruvate carboxykinase; GGPD, glucose-6-phosphate dehydrogenase; FCS, fetal calf serum; PBS, phosphate-buffered saline; BSA, bovine serum albumin; IBMX, 3-isobutyl-1-methylxantine. 295
All
Copyright 0 1992 rights of reproduction
0014-4827/92 $5.00 by Academic Press, Inc. in any form reserved.
296
MOLERO
10% FCS and dexamethasone (100 nM). After 4 h of culture, cells were washed twice with PBS, and 70% of the initial cells were attached forming a monolayer that under light microscopy represents around 25% of confluency. So, cell density was estimated as 2.5 X 10’ cells/cm*. At this time, cells were cultured for 64 h in a serum free medium supplemented with 0.2% BSA and dexamethasone in the presence of various hormones and growth factors. The culture medium and additives were changed every 48 h. At the end of the culture, cells were washed with PBS and collected for further determinations of RNA, DNA, and enzymatic activities. Routine control experiments to check the purity of our cultures revealed the presence of albumin by immunocytochemistry, as described in a previous work [Q]. Albumin-expressing fetal hepatocytes were distinguished from fibroblastlike cells, which comprised less than 5% of the total cells. DNA synthesis was determined in duplicated dishes by [3H]thymidine incorporation (0.2 &i/ml) during the last 4 h of culture. At the end of the incubation period, radioactivity in TCA-precipitable material was determined as described [lo]. Results were expressed as a percentage of the radioactivity incorporated by the untreated cells. DNA content was determined in duplicated dishes by a fluorimetric method [ll] using calf thymus DNA as standard, and results were expressed as pgldish. Cell cycle analysis was performed in cell suspensions after trypsinEDTA detachment of hepatocytes from the culture dishes and staining of cell nuclei with ice-cold propidium iodide (125 @g/ml) in 0.1% Na, citrate for 15 min. Cells were examined on a FACScan flow cytometer (Becton-Dickinson). For determination of enzyme activities, two dishes were pooled for each individual determination. Cells were sonicated for 45 s at 1.5 mA in 1 mM EDTA, 1 mM dithiothreitol, 0.25 M sucrose, 25 mM TrisHCl, pH 7.5, and centrifuged at 12,000g for 4 min. Cytosolic supernatants were used for the radiometrical determination of PEPCK [3] and the spectrophotometrical determination of GGPD 1121, respectively. Proteins were determined by [13] with BSA as standard. PEPCK was expressed as mU/mg of protein. A milliunit was a nmole of CO, incorporated to malate/min. GGPD was expressed as mU/mg of protein. A milliunit was a nmole of NADPH formedimin. For Northern-blot analysis of RNA, four dishes were pooled for each individual determination at the end of the time of culture, after washing cells with PBS and direct lysis of the monolayer with guanidinium thiocyanate-phenol reagent [14]. Total cellular RNA was isolated as 1141, denatured in 50% formamide, 2.2 M formaldehyde, 20 n-&f Mops, pH 7.0,5 mM sodium acetate, 1 mM EDTA at 65°C for 15 min. RNAs (20 fig) were electrophoresed on 1% agarose gels containing 2.2 M formaldehyde. Resolved RNAs were blotted onto nitrocellulose filter [15], baked at 80°C for 2 h, prehybridized for 4 h, and hybridized 40 h at 42‘C in a buffer containing 50% formamide, 2X Denhardt’s solution, 5~ SSC (0.75 M NaClI0.075 M Na, citrate), 0.1% SDS, denatured salmon-testis DNA (100 pglml) with denatured labeled cDNAs. cDNA labeling was carried out with [o-32P]dCTP to a specific activity of 0.5-l X lo9 cpm/pmol by using a multiprimer DNA labeling system kit. Then, filters were washed three times with 1X SSC containing 0.1% SDS at room temperature and twice at 42°C with 0.1X SSC containing 0.1% SDS. Filters were exposed to Kodak X-Omat AR films. Relative densities of the hybridization signals were determined by densitometric scanning of the autoradiograms, being linear at all the densities studied. At the end of the procedure, filters were stained with 0.05% methylene blue to verify the positions of 28 S and 18 S rRNAs. Northern-blot analysis was performed in triplicate from three different cell preparations. The variability in the measurement of the mRNA levels after the densitometryc analysis of the filters was not greater than 15%. RESULTS
AND
DISCUSSION
Effects of Hormones and Growth Factors on Proliferation of Fetal Hepatocyte Primary Cultures
Fetal hepatocytes were cultured for 64 h in a serumfree medium supplemented with dexamethasone to re-
ET AL.
strain growth of unwanted cells [16] and prolong the survival of the hepatocytes in the absence or presence of growth factors and hormones [l]. At the end of this time, the analysis of the cell cycle in cells cultured in the only presence of dexamethasone showed the quiescence of our cells, since >94% of the cells were arrested in the GO/G1 phase of the cell cycle, as assessed by flow cytometry. DNA synthesis can be induced in these quiescent cells after 64 h of treatment with different mitogens as determined by [3H]thymidine incorporation during the last 4 h of culture (Table 1). We tested the effect of EGF (the main known growth factor for the liver) on [3H]thymidine incorporation at different cell densities. At low cell density (2.5 X lo4 cells/cm2), EGF (3.3 nM) produced a 4-fold increase in DNA synthesis as compared with untreated cells (Table 1). However, at 5 X lo4 cells/ cm2 EGF increased DNA synthesis 2-fold and at 10 x lo4 cells/cm2 no significant effect of EGF on [3H]thymidine incorporation was observed. So, as growthrelated functions are stimulated at low cell density, we chose a cell density of 2.5 X lo4 for further experiments described in this paper. We cultured fetal hepatocytes in the presence of 10% FCS as a positive control for cell proliferation. Under FCS stimulation a 3-fold increase on [3H]thymidine incorporation was observed as compared with that of untreated cells (Table 1). Both EGF and FCS produced an increase of 2.5fold and 2-fold in the percentage of cells in S phase of cell cycle, respectively, as compared with that of untreated cells. Under EGF or FCS stimulation, 8% of fetal hepatocytes were in mitosis, as determined by DNA staining with propidium iodide and flow cytometry analysis. We investigated the effect of insulin (40 nM), IGF-I (1.4 nM), vasopressin (18 r&f), or glucagon (1 j&f) on the [3H]thymidine incorporation in fetal hepatocyte primary cultures. These factors, individually, did not modify the [3H]thymidine incorporation observed in untreated cells (Table 1). The combination of these factors with EGF did not further increase [3H]thymidine incorporation as compared with the effect induced by EGF alone (Table 1). The presence of forskolin (10 PM), a direct activator of adenylate cyclase, together with IBMX (50 pLM), a CAMP-phosphodiesterase inhibitor, did not modify the [3H]thymidine incorporation in fetal hepatocyte primary cultures as compared with untreated cells. However, forskolin plus IBMX decreased [3H]thymidine incorporation in EGF-stimulated hepatocytes almost to the levels observed in untreated cells (Table 1). Moreover, DNA content at the end of the time of culture under different experimental conditions has also been included in Table 1. In the untreated cells, there was a good correlation between the DNA content and the cell number that gave a mean +_SEM of 27.6 + 3.0 pg DNA/lo’ cells and indicated that cell density in the absence of serum or signals and in the only presence of dexamethasone was constant throughout the time of
PEPCK
AND
GGPD
TABLE
297
EXPRESSION
1
Effects of Hormones and Growth Factors on [3H]Thymidine Incorporation in Fetal Rat Hepatocyte Primary Cultures
and DNA Content
[sH]Thymidine incorporation (percentage) Addition None Glucagon (1 p&f) Insulin (40 nM) IGF-I (1.4 n&f) Vasopressin (18 nM) Forskolin (10 ).rAf) + IBMX 10% FCS
None
(50 &f)
100 112 t 2 119 + 6 118? 6
