Growth-related expression of the ectonucleotide pyrophosphatase PC ...

3 downloads 3953 Views 600KB Size Report
expression of both PC-1 mRNA and protein in rat liver and in hepatoma cells is ..... Collectively, the above data suggest that the loss of PC-1. mRNA that is ...
Growth-Related Expression of the Ectonucleotide Pyrophosphatase PC-1 in Rat Liver CRISTIANA STEFAN, WILLY STALMANS,

Plasma cell differentiation antigen-1 (PC-1) is a 58ectonucleotide pyrophosphatase that has been implicated in various processes including insulin- and nucleotidemediated signaling and cell growth. We show here that the expression of both PC-1 mRNA and protein in rat liver and in hepatoma cells is strictly growth-related. Thus, the level of PC-1 in FAO hepatoma cells increased with the cell density. PC-1 was not expressed in the neonatal rat liver, but gradually appeared in the first weeks of age, to reach adult levels around the weaning period. Furthermore, PC-1 protein and mRNA largely disappeared from the liver within 24 hours following a hepatectomy of 70%, but re-appeared in the later phases (3-15 days) of the ensuing regeneration period. An equally rapid loss of PC-1 protein and mRNA could also be provoked in normal livers by the administration of the translational inhibitor, cycloheximide, but the transcriptional inhibitors, actinomycin D and a-amanitin, did not show these effects. Nuclear run-on assays revealed that the loss of PC-1 mRNA after hepatectomy or after the administration of cycloheximide was not caused by a decreased transcription of the PC-1 gene, suggesting that the level of PC-1 is controlled by an mRNA-stabilizing protein that is lost after hepatectomy and has a high turnover. (HEPATOLOGY 1998;28:1497-1503.) Plasma cell differentiation antigen-1 (PC-1) is an integral protein of the plasma membrane that is expressed in various differentiated cell types including B lymphocytes, hepatocytes, and chondrocytes.1-3 A fraction of PC-1 may also be associated with the endoplasmic reticulum.4 In addition, there is preliminary evidence for the existence of a soluble and secreted form of PC-1 that is generated by intracellular proteolysis.5 The membrane form of PC-1 is a disulfidebonded dimer of polypeptide(s) of about 100 kd.3,5 In the plasma membrane, each monomer consists of a short aminoterminal intracellular domain, a single transmembrane domain, and a large extracellular carboxy-terminal domain. The extracellular part of PC-1 displays a nucleotide pyrophosphatase/phosphodiesterase-I activity that releases nucleosides-58-

Abbreviations: PC-1, plasma cell differentiation antigen-1; ATP, adenosine triphosphate; SDS, sodium dodecyl sulfate. From the Afdeling Biochemie, Faculteit Geneeskunde, Katholieke Universiteit Leuven, Leuven, Belgium. Received January 21, 1998; accepted July 8, 1998. Supported by the Fund for Medical Scientific Research (grant G.0237.98). Address reprint requests to: Mathieu Bollen, Afdeling Biochemie, Campus Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium. Fax: 32-16-34-59-95. Copyright r 1998 by the American Association for the Study of Liver Diseases. 0270-9139/98/2806-0008$3.00/0

AND

MATHIEU BOLLEN

monophosphate from a variety of substrates, including adenosine triphosphate (ATP).6,7 The catalysis occurs via a nucleotidylated enzyme, involving a threonine in the catalytic site. The same threonine is also autocatalytically phosphorylated at submicromolar concentrations of ATP, and this phosphorylation is associated with an inactivation of PC-1 as a nucleotide pyrophosphatase.3,6,8 An increase in the concentration of nucleotides results in an autocatalytic dephosphorylation and re-activation of PC-1. The autophosphorylation of PC-1 has been interpreted as an autoregulatory mechanism that prevents the complete hydrolysis of nucleotides.3,6,8 PC-1 has been implicated in the regulation of various cellular processes. The tight control of PC-1 by nucleotides (see above) has been taken as evidence for a role of PC-1 in the regulation of the extracellular concentration of nucleotides such as ATP, adenosine diphosphate, and di-adenosine polyphosphates.6,8 Uridine diphosphate–sugars and 38phosphoadenosine 58-phosphosulfate, which act as sugar and sulfate donors in protein glycosylation and glycosaminoglycan sulfation, respectively, have been proposed as substrates for intracellular PC-1.9,10 In bone and cartilage, PC-1 appears to be required for the generation of pyrophosphate from nucleoside triphosphates.2 PC-1 has also been described as an inhibitor of insulin signaling.11 However, PC-1 is not a direct inhibitor of the insulin receptor,12 and its effect on insulin signaling does not require nucleotide pyrophosphatase activity.13 A structurally related nucleotide pyrophosphatase, known as gp130RB13-6 or B10, has an RGD-tripeptide sequence for attachment to integrins, suggesting a role in cell adhesion.14,15 Autotaxin, yet another structurally related nucleotide pyrophosphatase, has been isolated as a secreted cytokine that stimulates both chemotactic and chemokinetic responses in tumor cells.16 The effect of autotaxin is blocked by pertussis toxin, providing a link to G-protein–mediated signaling. We have started to analyze the role of PC-1 in the liver, where it is mainly localized at the basolateral surface of the hepatocytes, facing the blood and neighboring cells.15 Several lines of evidence suggest that PC-1 could have a role in liver growth. Thus, the combined action of PC-1 and 58nucleotidase results in the production of growth-inhibiting adenosine16 from mitogenic ATP,17 which is produced extracellularly as a paracrine-secreted signaling molecule.18 PC-1 also inhibits signaling via insulin,11 a hepatic co-mitogen.19 Furthermore, the expression of PC-1 has been shown to be controlled by transforming growth factor b1,2,20 a known hepatic growth suppressor. PC-1 may also inhibit liver growth more indirectly, e.g., through its extracellular somatomedin B domain, by the binding and stabilization of the plasminogen activator inhibitor-1,21,22 thus preventing plas-

1497

1498

STEFAN, STALMANS, AND BOLLEN

minogen-mediated matrix degradation.23 To further evaluate the role of PC-1, we have analyzed the expression of PC-1 in various models of liver growth. These include the postnatal switch in the hepatic protein expression pattern,24 liver regeneration after hepatectomy,19,25 and the proliferation of Reuber H35 hepatoma cells.26 We show here that the expression of PC-1 is tightly correlated with the growth and differentiation phase of hepatocytes. MATERIALS AND METHODS Materials. Cycloheximide was purchased from Janssen Chimica (Beerse, Belgium), actinomycin D from Merck, Sharp, and Dohme (Brussels, Belgium), and a-amanitin from Fluka (Buchs, Switzerland). The full-length murine p53 cDNA in the pSP65 vector was a gift from Dr. C. Desaintes (Paris, France). The complete coding region of murine PC-1 was obtained from the pSVL/PC-1 plasmid4 by digestion with XhoI/BamH1. Polyclonal antibodies against a recombinant fragment of PC-1, referred to as clone 5 by Rebbe et al.,7 were made in rabbits. A synthetic peptide encompassing the 14 C-terminal residues of murine PC-1 (LRLKTHLPIFSQED), plus an additional N-terminal cysteine, was coupled to keyhole limpet hemocyanin using the Pierce immunogen conjugation kit. Rabbit polyclonal antibodies against the hemocyanin-coupled peptide were purified by chromatography on Affi-T Sepharose. PC-1 was purified from rat liver as described by Uriarte et al.8 Treatment of Animals and Liver Fractionation. A hepatectomy of 70% was performed under ether anesthesia on Wistar rats weighing approximately 200 g.25 Cycloheximide (50 mg/kg body weight) and actinomycin D (2 mg/kg body weight) were administered by an intraperitoneal injection. a-Amanitin (0.5 mg/kg body weight) was injected intravenously in the penile vein under ether anesthesia. All animals were decapitated. Subsequently, the liver was rinsed in saline and either freeze-clamped in liquid nitrogen or homogenized immediately. The livers were homogenized in 4 volumes of a buffer containing 5 mmol/L Tris/HCl at pH 7.6, 0.5 mmol/L MgCl2, and 0.25 mol/L sucrose (buffer A), further supplemented with 0.5 mmol/L phenylmethanesulfonyl fluoride, 0.5 mmol/L benzamidine, 0.1 mmol/L L-1-tosylamido-2-phenylethyl chloromethyl ketone, 0.1 mmol/L 1-chloro-3-tosylamido-7-amino-2-heptanone, aprotinin (1 g/mL), pepstatin (0.7 µg/mL), and leupeptin (1 µg/mL). Following a low-speed centrifugation (5 minutes at 280g), the supernatant was centrifuged during 10 minutes at 1,500g.27 The resulting pellet was used for the preparation of plasma membranes by centrifugation in a discontinuous sucrose gradient, as described by Ozols.27 The 1,500g supernatant was further centrifuged, first for 15,000g at 30 minutes and subsequently for 60 minutes at 100,000g. The 100,000g supernatant was used as the ‘‘cytosolic’’ fraction. The 15,000g pellet, further referred to as the crude membrane fraction, was washed once in 10 volumes of a buffer containing 5 mmol/L Tris/HCl at pH 7.6 and 0.5 mmol/L MgCl2, and then resuspended in 20 mmol/L Tris/HCl at pH 8, 100 mmol/L KCl, and 2% (vol/vol) Triton X-100 at a final protein concentration of 10 to 20 mg/mL. Intact liver nuclei were prepared from whole homogenates as described by Jagiello et al.28 The outer nuclear membrane was then solubilized by resuspending the nuclei in buffer A plus 0.5% (vol/vol) Triton X-100 and separated from the insoluble nuclear fraction by centrifugation for 10 minutes at 1,500g. Cell Culture. FAO rat hepatoma cells were cultured in a mixture (1:1) of Ham’s F12 and NCTC-135 media, with 2 mmol/L glutamine, 10% (vol/vol) fetal calf serum, 100 U/mL penicillin, and 100 U/mL streptomycin. After washing with phosphate-buffered saline, the cells were harvested by scraping in either 20 mmol/L Tris/HCl at pH 7.5, 150 mmol/L NaCl, and 2% (vol/vol) Triton X-100 for Western analysis, or in 25 mmol/L sodium acetate at pH 5.5, 4 mol/L guanidinium isothiocyanate, 0.1 mol/L b-mercaptoethanol, and 0.5% (wt/vol) sodium N-lauroylsarcosine for Northern analysis.

HEPATOLOGY December 1998 Western and Northern Analysis. The PC-1 antisera were used for Western analysis at a final dilution of 1:5,000, and the peroxidaselabeled secondary antibodies were detected by enhanced chemiluminescence. Total RNA was isolated from 100 to 500 mg of frozen pulverized liver by CsCl gradient centrifugation.29 The RNA (30 µg/lane) was fractionated by electrophoresis in formaldehyde-agarose (1%) gels and transferred overnight by capillary diffusion to nylon membranes. The membranes were hybridized with cDNAs encoding PC-1 or 18-S ribosomal RNA, labeled to a specific radioactivity of 2 3 109 cpm/µg by random priming (Life Technologies, Inc.). Hybridizations were performed between 6 and 16 hours at 65°C in 53 SSC, 20 mmol/L phosphate (pH 6.5), 13 Denhardt’s solution, 8% dextran sulfate, and 100 µg/mL salmon sperm DNA. Subsequently, the membranes were washed twice for 15 minutes at room temperature in 23 SSC/0.1% sodium dodecyl sulfate (SDS) and twice for 25 minutes at 65°C in 0.23 SSC/0.1% SDS. The hybridized cDNA was localized and quantified by phosphor image analysis. Transcriptional Run-on Assays. Liver nuclei were obtained as described by Marzluff,30 with slight modifications. Briefly, 2 g of freeze-clamped and pulverized liver was homogenized in 6 mL of an ice-cold buffer containing 10 mmol/L HEPES at pH 7.6, 2 mol/L sucrose, 10% glycerol, 3 mmol/L MgCl2, 5 mmol/L dithiothreitol, 25 mmol/L KCl, 0.5 mmol/L phenylmethanesulfonyl fluoride, 0.5 mmol/L benzamidine, 50 µmol/L L-1-tosylamido-2-phenylethyl chloromethyl ketone, and 50 µmol/L 1-chloro-3-tosylamido-7-amino-2heptanone. The homogenate was then diluted to 22.5 mL with homogenization buffer and layered over a cushion of 7.5 mL of homogenization buffer. The nuclear pellet was obtained by centrifugation (80,000g for 1 hour) and resuspended in 200 µL Tris/HCl at pH 8.0, 40% (wt/vol) glycerol, 5 mmol/L MgCl2, 1 mmol/L dithiothreitol, 0.25 mmol/L phenylmethanesulfonyl fluoride, and 0.25 mmol/L benzamidine. The transcriptional elongation was performed according to Kren et al.31 The elongation was arrested by addition of 5 mL of a mixture containing 25 mmol/L sodium acetate at pH 5.5, 4 mol/L guanidinium isothiocyanate, 0.1 mol/L b-mercaptoethanol, and 40 µg/µL yeast tRNA. The 32P-labeled run-on transcripts were isolated as described by Chirgwin et al.29 and hybridized for 3 days with 24 g of the bamH1-linearized pSVL/PC1 plasmid or with 5 µg of p53 cDNA, both slot-blotted onto nylon membranes. The hybridization conditions were the same as those for Northern analysis. Within each experiment, every membrane was hybridized with the same amount of 32P-labeled RNA. After hybridization, the membranes were first washed for 15 minutes at room temperature in 23 SSC/0.1% SDS, and then for 15 minutes at 65°C in 0.23 SSC/0.1% SDS. Subsequently, the membranes were incubated for 30 minutes at 37°C with 10 µg/mL RNase A and then for 30 minutes at 37°C with proteinase K (100 µg/mL). The retained label was visualized by phosphor image analysis. Other Assays. PC-1 was assayed as phosphodiesterase-I at 30°C in the presence of 50 mmol/L Tris/HCl at pH 9.0, with 0.9 mmol/L p-nitrophenyl thymidine 58-monophosphate as substrate.6 The protein concentration was assayed colorimetrically,32 with bovine serum albumin as standard.

RESULTS The Subcellular Localization of PC-1. As shown in Fig. 1, purified rat liver PC-1 is visualized8 by Western analysis with polyclonal antibodies against recombinant PC-1 as a doublet of 118/128 kd, as well as a polypeptide of 197 kd, which represents nonreduced PC-1 dimer. The same polypeptides were also detected in purified liver plasma membranes and in the outer nuclear membrane fraction (Fig. 1), which is continuous with the endoplasmic reticulum. The ratio between the PC-1 polypeptides of 118 and 128 kd was variable, however, which may reflect different extents of proteolysis. Indeed, unlike the 128-kd polypeptide, the 118-kd polypep-

HEPATOLOGY Vol. 28, No. 6, 1998

STEFAN, STALMANS, AND BOLLEN

FIG. 1. Localization of PC-1 in rat liver. Polyclonal antibodies against recombinant PC-1 were used for Western analysis of purified PC-1 (a), the plasma membrane fraction (b), the outer nuclear membrane fraction (c), and the cytosolic fraction (d), all derived from rat liver.

tide was not recognized by antibodies directed against the C-terminal 14 residues of PC-1 (not shown), indicating that the latter represents a C-terminally nicked form of the 128-kd polypeptide. The antibodies against recombinant PC-1 also recognized polypeptides of 175 kd and 80 kd in the outer nuclearmembrane fraction and a major polypeptide of 72 kd in the cytosolic liver fraction. The cytosolic 72-kd polypeptide was not detected in lysates of isolated hepatocytes (not shown) or FAO hepatoma cells (see below), suggesting that it stemmed from nonparenchymal cells. None of the above-mentioned polypeptides were recognized by antibodies against the nucleotide pyrophosphatase, B10/gp130RB13-6, which did recognize, however, a polypeptide of 135 kd in rat liver

A FIG. 2. Age-related expression of PC-1 in the liver. PC-1 in liver homogenates of rats of the indicated age was visualized by Western analysis (A). An aliquot of the same livers was also used for the preparation of total RNA for Northern analysis, using cDNAs encoding PC-1 or 18-S rRNA as probes (B). The lanes contained 17 µg protein (A) or 30 µg total RNA (B).

1499

homogenates (not shown), in agreement with its recent localization at the canalicular surface of the hepatocytes.15 Age-Dependent Expression of PC-1. Western analysis showed that PC-1 was absent in neonatal rat liver homogenates, but gradually appeared in the postnatal period to reach adult levels at the age of 3 to 6 weeks (Fig. 2A). In contrast, the expression of the cross-reacting polypeptides of 175 and 72 kd was not clearly age-dependent. Unexpectedly, the PC-1 antibodies also detected polypeptides of 77 to 88 kd in the neonatal liver homogenates. Their gradual loss in the first weeks after birth could be related to the gradual appearance of PC-1. It cannot be excluded, however, that these polypeptides stem from hematopoietic cells, which gradually disappear from the liver in the postnatal period. The polypeptides of 77 to 88 kd were also present in freeze-clamped livers that were homogenized in Laemmli buffer (not shown), indicating that they were not generated by proteolysis of PC-1 during liver fractionation. Northern analysis revealed PC-1 transcripts of 3.2 and 4.1 kb (Fig. 2B). Both transcripts were virtually absent in the neonatal liver, but their level increased in parallel with the level of PC-1 mRNA protein. It should also be noted that the PC-1 transcripts did not hybridize with a B10/gp130Rb13-6specific probe (not shown), indicating that both transcripts encode PC-1. Fractionation of liver homogenates from rats of 8 weeks on DEAE-Sephadex revealed three overlapping peaks of phosphodiesterase-I activity, using p-nitrophenyl thymidine phosphate as substrate (Fig. 3A). The second and third activity peaks were largely absent in the corresponding fractions from rats of 1 week (Fig. 3B). Western analysis (Fig. 3) indicated that the latter peaks correspond to PC-1, which is virtually absent in rats of 1 week. These data suggest that PC-1 accounts for a significant fraction of the phosphodiesterase-I activity in rat liver. It is also shown in Fig. 3B that the distribution of the cytosolic 72-kd polypeptide that is recognized by the PC-1 antibodies differs markedly from that of the p-nitrophenyl thymidine phosphatase activity, suggesting that the 72-kd polypeptide is not a phosphodiesterase-I. Transient Decrease of PC-1 After Hepatectomy. As a second model of hepatic growth, we analyzed the fate of PC-1 during liver regeneration following a hepatectomy of 70%. Immunodetectable PC-1 in liver homogenates (not shown) or in the crude membrane fraction (Fig. 4A and 4C) started to decrease by the first hours after hepatectomy and reached a minimal level, i.e., 24% 6 4% of the control value (n 5 7), after 24 hours. The concentration of PC-1 started to increase again in the course of the first week and approached normal levels 15

B

1500

STEFAN, STALMANS, AND BOLLEN

HEPATOLOGY December 1998

FIG. 3. Fractionation of PC-1 on DEAE-Sephadex. Liver homogenates (6 mL) from rats of 8 weeks (A) or 1 week (B) were supplemented with 2% (vol/vol) Triton X-100 for 1 hour and subsequently centrifuged for 30 minutes at 10,000g. The supernatants were applied to a DEAE-Sephadex column (10 3 1 mL), equilibrated in 50 mmol/L Tris/HCl at pH 7.5 and 0.1% (vol/vol) Triton X-100. The retained proteins were eluted with a 100-mL gradient of NaCl (0-0.5 mol/L). The fractions were assayed for PC-1 by Western blotting (upper panels) and for phosphodiesterase-I activity, using p-nitrophenyl thymidine phosphate as substrate (lower panels).

days after hepatectomy, when the regeneration of the liver, as assessed by its weight (not shown), was virtually complete. The transient decrease in the level of PC-1 protein was correlated with changes in the level of PC-1 mRNA, as detected by Northern analysis (Fig. 4B and 4C). However, the 4.1-kb transcript decreased to lower levels (24% 6 4%; n 5 2) than the 3.2-kb transcript (58% 6 10%; n 5 2). The rates of disappearance of the PC-1 protein and mRNA were very similar, indicating that the protein has a short half-life. It is also shown in Fig. 4 that the levels of PC-1 mRNA and protein were not measurably affected by a sham operation. The Effect of Inhibitors of Protein Synthesis. To check more accurately the turnover of PC-1, its rate of disappearance was measured by Western analysis, following the injection of an amount of cycloheximide that blocked translation in rat liver by 91% to 99%.33 These experiments yielded a half-life for the PC-1 protein of about 3 hours (Fig. 5). Unexpectedly, the administration of cycloheximide also caused a loss of PC-1 mRNA, suggesting that the stability of the PC-1 transcripts is controlled by a protein with a high turnover. As was also seen after hepatectomy (Fig. 4C), the 4.1-kb transcript of PC-1 was lost more extensively than the 3.2-kb transcript after the administration of cycloheximide (Fig. 5B). Moreover, the rates at which the transcripts disappeared were similar for both treatments (Figs. 4C and 5B), and the combination of both treatments did not result in a more rapid loss of the transcripts (not shown). As a control for the treatment with cycloheximide, the Northern blots were also hybridized with p53 cDNA (Fig. 5A). In agreement with published data,34 cycloheximide was found to induce a manifold increase in the mRNA level of this tumor suppressor (Fig. 5A).

The in vivo administration of the transcriptional inhibitors, actinomycin D plus a-amanitin, did not measurably affect the steady-state level of PC-1 mRNA within 8 hours, but did cause a pronounced decrease in the level of the p53 transcript (Fig. 6), in accordance with findings reported by Kren et al.34 Collectively, the above data suggest that the loss of PC-1 mRNA that is caused by hepatectomy or by translational inhibition results from a common mechanism. Additional evidence for this view came from transcriptional run-on assays, showing that neither of these treatments had an effect on the rate of synthesis of PC-1 mRNA (Fig. 7). This implies that the loss of PC-1 mRNA is caused by an increased rate of degradation. Hepatectomy or a treatment with cycloheximide did not affect the transcription rate of p53 either (not shown), in agreement with previous investigations.34 The Expression of PC-1 in FAO Hepatoma Cells. We have also investigated the expression of PC-1 in FAO rat hepatoma cells. A striking difference with intact liver preparations (Fig. 1) or freshly isolated hepatocytes (not shown) was the absence of the 118-kd partner of the PC-1 doublet in hepatoma cells (Fig. 8A). Yet both the 3.2-kb and 4.1-kb transcripts were present in FAO cells (Fig. 8B). The level of PC-1 protein, but not the cross-reacting 175-kd polypeptide, was found to increase with the cell density (Fig. 8A). At 90% confluency, when the cells have stopped dividing, the concentration of PC-1 protein was fourfold higher than at 20% confluency. This increase was not associated, however, with similar changes in the level of PC-1 mRNA (Fig. 8B), implying that the different level of PC-1 protein resulted from a (post)translational control mechanism.

HEPATOLOGY Vol. 28, No. 6, 1998

STEFAN, STALMANS, AND BOLLEN

1501

the so-called ‘‘late-suckling’’ cluster of proteins that become only maximally expressed around the weaning period (Fig. 2). Liver regeneration represents another well-established physiological model of growth regulation.19,25 While virtually all hepatocytes in the adult liver are in interphase, they synchronously re-enter the cell cycle after a partial hepatectomy and undergo one to two divisions, resulting in a full restoration of the original liver mass in 2 to 3 weeks. Liver regeneration is associated with substantial changes in the expression of a large number of proteins, and the time scale of these changes is protein-specific. PC-1 belongs to a small group of proteins, also including the CCAAT/enhancer binding protein a,35 whose level initially decreases after hepatectomy and only shows a maximal expression again after the major growth phase (Fig. 3). It should therefore be investigated whether PC-1, like the CCAAT/enhancer binding protein a, has a role in terminating the regenerative response to hepatectomy, once the recovery of the liver mass has been accomplished. FAO cells are a subline of the Reuber H35-hepatoma cells that share many properties of the regenerating liver.26 Accordingly, we found that the level of PC-1 increases with the cell density (Fig. 8). These data are in agreement with other reports showing that the level of the plasma membrane phosphodiesterase-I activity is five- to sevenfold increased in contact-inhibited 3T3 cells.36 Similarly, Clark et al.37 reported that Walker 256 carcinoma cells have an increased phospho-

A

B

C

A

FIG. 4. The levels of PC-1 protein and of mRNA after a hepatectomy of 70%. The crude membrane fraction from the liver of rats at the indicated times following a hepatectomy of 70% or a sham operation was analyzed for the presence of PC-1 by Western blotting (A). Samples of the same livers were also used for the preparation of total RNA for Northern analysis, using the cDNAs for PC-1 and 18-S RNA as probes (B). (C) The means 6 SEM (n 5 2-7), except for the 7-day time point (n 5 1), of the levels of PC-1 protein (derived from both the 118-kd and 128-kd polypeptides) and mRNA, as calculated from scanning of the blots. The data were normalized for the same protein or RNA loading, as assessed by scanning of Coomassie-stained blots or 18-S rRNA–probed membranes.

B

DISCUSSION Hepatic PC-1 Is Maximally Expressed After the Major Growth Phase. We have used several well-established models to

investigate whether the expression pattern of PC-1 in liver is growth-related. One model involves the changing protein expression profile in the postnatal liver, reflecting new functional requirements.24 Three main clusters in protein expression have been discerned, according to whether the proteins are beginning to be expressed in the late gestational period, in the first neonatal day, or just before weaning. We found that PC-1 is not yet expressed at birth and belongs to

FIG. 5. Effect of cycloheximide on the steady-state levels of PC-1 and p53. At the indicated times after an intraperitoneal injection of cycloheximide, the livers were freeze-clamped. PC-1 protein levels were measured by Western analysis of liver homogenates. The mRNA levels of PC-1 and p53 were visualized by Northern blot analysis on total RNA. The graphs in (B) represent the means 6 SEM of three experiments.

1502

STEFAN, STALMANS, AND BOLLEN

HEPATOLOGY December 1998

A

B

FIG. 6. Effect of actinomycin D plus a-amanitin on the mRNA level of PC-1 and p53. At the indicated times after the administration of actinomycin D plus a-amanitin, total RNA was prepared from freeze-clamped livers. Northern blot analysis was performed with the indicated probes.

diesterase-I activity when induced to grow as a solid tumor, which involves a closer cellular contact and interaction with the environment. In conclusion, PC-1 is a protein that is maximally expressed after the major period of growth or regeneration. While our results demonstrate that the level of PC-1 can be used as a marker for the quiescent state of the cell cycle, the cause/effect relationship between the expression of PC-1 and cell growth remains to be explored. The Expression of PC-1 Is Controlled at Various Levels. The increase in the PC-1 protein level with the cell density was not associated with an increased steady-state level of PC-1 mRNA (Fig. 8), suggesting a (post)translational control mechanism. In addition, nuclear run-on assays (Fig. 7) showed that the loss of PC-1 mRNA after hepatectomy cannot be explained by a decreased transcription rate, implying a control on the transcript stability. Alteration in transcript stability is a key posttranscriptional regulatory mechanism used by the regenerating liver to modulate the steadystate mRNA levels.36 In fact, regulation of gene expression following the first 3 hours after a hepatectomy of 70% occurs predominantly at the posttranscriptional level. Preliminary data suggest that this posttranscriptional control of PC-1 involves a short-lived mRNA-stabilizing protein that is lost after hepatectomy. The main evidence is that a similar loss of PC-1 protein and mRNA could also be provoked with the translational inhibitor, cycloheximide (Fig. 5), and that the

FIG. 7. Transcriptional run-on assays after hepatectomy (A) or cycloheximide (B). At the indicated times after a hepatectomy of 70% or after an injection of cycloheximide, the PC-1 gene expression rate was measured by nuclear run-on assays, as detailed in Materials and Methods. The result of hybridization with the empty pSVL/PC-1 plasmid is also shown. A second hybridization with the 32P-labeled transcripts did not yield any signal (not shown).

A

B

FIG. 8. The translation of PC-1 in FAO cells depends on the cell density. FAO cells were grown until 20% or 90% confluency and then analyzed by Western analysis (A) or Northern analysis (B) for the level of PC-1 protein or mRNA, respectively, using total cell lysates (24 µg protein/lane) or total RNA (30 µg/lane).

HEPATOLOGY Vol. 28, No. 6, 1998

effects of cycloheximide and hepatectomy were not additive. A labile mRNA-stabilizing protein has also been proposed to control the level of the c-fms proto-oncogene product.39 Likewise, it has been shown that c-myc mRNA is stabilized by a protein that binds to the coding region and protects the mRNA from degradation by a ribosome-associated endoribonuclease.40 Acknowledgment: A. Hoogmartens and C. Verwichte provided expert technical assistance. The authors thank Dr. Veerle Vulsteke for advice on the culture of FAO cells, Dr. S. Hickman (St. Louis, MO) for the generous gift of the PC-1 expression vector and antibodies, and Dr. H. Deissler (Essen, Germany) for the generous gift of B10/gp130 cDNA and antibodies. REFERENCES 1. van Driel IR, Goding JW. Plasma cell membrane glycoprotein PC-1. Primary structure deduced from cDNA clones. J Biol Chem 1987;262: 4882-4887. 2. Huang R, Rosenbach M, Vaughn R, Provvedini D, Rebbe N, Hickman S, Goding J, et al. Expression of the murine plasma cell nucleotide pyrophosphohydrolase PC-1 is shared by human liver, bone, and cartilage cells. Regulation of PC-1 expression in osteosarcoma cells by transforming growth factor-b. J Clin Invest 1994;94:560-567. 3. Uriarte M, Stalmans W, Hickman S, Bollen M. Phosphorylation and nucleotide-dependent dephosphorylation of hepatic polypeptides related to the plasma cell differentiation antigen PC-1. Biochem J 1993;293:93-100. 4. Rebbe NF, Tong BD, Hickman S. Expression of nucleotide pyrophosphatase and alkaline phosphodiesterase I activities of PC-1, the murine plasma cell antigen. Mol Immunol 1993;30:87-93. 5. Belli SI, van Driel IR, Goding JW. Identification and characterization of a soluble form of the plasma cell membrance glycoprotein PC-1 (58nucleotide phosphodiesterase). Eur J Biochem 1993;217:421-428. 6. Stefan C, Stalmans W, Bollen M. Threonine autophosphorylation and nucleotidylation of the hepatic membrane protein PC-1. Eur J Biochem 1996;241:338-342. 7. Rebbe NF, Tong BD, Finley EM, Hickman S. Identification of nucleotide pyrophosphatase/alkaline phosphodiesterase I activity associated with the mouse plasma cell differentiation antigen PC-1. Proc Natl Acad Sci U S A 1991;88:5192-5196. 8. Uriarte M, Stalmans W, Hickman S, Bollen M. Regulation of purified hepatic PC-1 (phosphodiesterase-I/nucleotide pyrophosphatase) by threonine auto(de)phosphorylation and by binding of acidic fibroblast growth factor. Biochem J 1995;306:271-277. 9. Rebbe NF, Hickman S. Modulation of nucleotide pyrophosphatase in plasmacytoma cells. Biochem Biophys Res Commun 1991;175:637-644. 10. Yamashima I, Yoshida H, Fukui S, Funakoshi I. Biochemical studies on Lowe’s syndrome. Mol Cell Biochem 1983;52:107-124. 11. Maddux B, Sbraccia P, Kumakura S, Sasson S, Youngren J, Fisher A, Spencer S, et al. Membrane glycoprotein PC-1 and insulin resistance in non–insulin-dependent diabetes mellitus. Nature 1995;373:448-451. 12. Stefan C, Wera S, Stalmans W, Bollen M. The inhibition of the insulin receptor by the membrane protein PC-1 is not specific and results from the hydrolysis of ATP. Diabetes 1996;45:980-983. 13. Grupe A, Alleman J, Goldfine ID, Sadick M, Stewart TA. Inhibition of insulin receptor phosphorylation by PC-1 is not mediated by the hydrolysis of adenosine triphosphate or the generation of adenosine. J Biol Chem 1995;270:22085-22088. 14. Deissler H, Lottspeich F, Rajewsky MF. Affinity purification and cDNA cloning of rat neural differentiation and tumor cell surface antigen gp130RB13-6 reveals relationship to human and murine PC-1. J Biol Chem 1995;270:9849-9855. 15. Scott LJ, Delautier D, Rajho Meerson N, Trugnan G, Goding JW, Maurice M. Biochemical and molecular identification of distinct forms of alkaline phosphodiesterase I expressed on the apical and basolateral plasma membrane surfaces of rat hepatocytes. HEPATOLOGY 1997;25:995-1002.

STEFAN, STALMANS, AND BOLLEN

1503

16. Hargrove JL, Granner DK. Inhibition of hepatoma cell growth by analogs of adenosine and cyclic AMP and the influence of enzymes in mammalian sera. J Cell Physiol 1982;111:232-238. 17. Heppel LA, Wang D, Huang N, Gonzalez FA, Ahmed AH, Alfonzo RG, Desheesh M. Extracellular ATP is a mitogen for 3T3, 3T6, A431, DDT1-MF2, BALB/MK, NIE-115, and HFF cells. Ann N Y Acad Sci 1990;603:432-434. 18. Schlosser SF, Burgstahler AD, Nathanson MH. Isolated hepatocytes can signal to other hepatocytes and bile duct cells by release of nucleotides. Proc Natl Acad Sci U S A 1996;93:9948-9953. 19. Michalopoulos GK, DeFrances MC. Liver regeneration. Science 1997;276: 60-66. 20. Lotz M, Rosen F, McCabe G, Quach J, Blanco F, Dudler J, Solan J, et al. Interleukin 1 suppresses transforming growth factor–induced inorganic pyrophosphate (PPi) production and expression of the PPi-generating enzyme PC-1 in human chrondrocytes. Proc Natl Acad Sci U S A 1995;92:10364-10368. 21. Stracke ML, Clair T, Liotta LA. Autotaxin, tumor motility stimulating exophosphodiesterase. Adv Enzyme Regul 1997;37:135-144. 22. Seiffert D, Ciambrone G, Wagner NV, Binder BR, Loskutoff DJ. The somatomedin B domain of vitronectin. Structural requirements for the binding and stabilization of active type 1 plasminogen activator inhibitor. J Biol Chem 1994;269:2659-2666. 23. Kim T-H, Mars WM, Stolz DB, Petersen BE, Michalopoulos GK. Extracellular matrix remodelling at the early stages of liver regeneration in the rat. HEPATOLOGY 1997;26:896-904. 24. Greengard O. Enzymic differentiation of human liver: comparison with the rat model. Pediatr Res 1977;11:669-676. 25. Alison MR. Regulation of hepatic growth. Physiol Rev 1986;66:499-541. 26. Mohn KL, Laz TM, Hsu J-C, Melby AE, Bravo R, Taub R. The immediate-early growth response in regenerating liver and insulinstimulated H-35 cells: comparison with serum-stimulated 3T3 cells and identification of 41 novel immediate-early genes. Mol Cell Biol 1991;11: 381-390. 27. Ozols J. Preparation of membrane fractions. Methods Enzymol 1990;182: 225-235. 28. Jagiello I, Beullens M, Stalmans W, Bollen M. Subunit structure and regulation of protein phosphatase-1 in rat liver nuclei. J Biol Chem 1995;270:17257-17263. 29. Chirgwin JM, Przybyla AE, MacDonald RJ, Rutter WJ. Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 1979;18:5294-5299. 30. Marzluff WF. Preparation of active nuclei. Methods Enzymol 1990;181: 30-36. 31. Kren BT, Teel AL, Steer CJ. Transcriptional rate and steady-state changes of retinoblastoma mRNA in regenerating rat liver. HEPATOLOGY 1994;19: 1214-1222. 32. Bradford MM. A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976;72:248-254. 33. Bollen M, Dopere´ F, Goris J, Merlevede W, Stalmans W. The nature of the decreased activity of glycogen synthase phosphatase in the liver of the adrenalectomized starved rat. Eur J Biochem 1984;144:57-63. 34. Kren BT, Trembley JH, Steer CJ. Alterations in mRNA stability during rat liver regeneration. Am J Physiol 1996;270:G763-G777. 35. Cereghini S. Liver-enriched transcription factors and hepatocyte differentiation. FASEB J 1996;10:267-282. 36. Harshman S, Stoller D, Harshman DL. Increase in plasma membrane phosphodiesterase activity in contact-inhibited 3T3 cells and in phenotypically reverted SV-3T3 cells. J Membr Biol 1979;50:165-175. 37. Clark CM, Goodlad GAJ. Protein and enzyme content of plasma membranes derived from Walker 256 carcinoma cells grown as ascitic or solid tumours. Cancer Lett 1993;68:91-94. 38. Kren BT, Steer CJ. Posttranscriptional regulation of gene expression in liver regeneration: role of mRNA stability. FASEB J 1996;10:559-573. 39. Weber B, Horiguchi J, Luebbers R, Sherman M, Kufe D. Posttranscriptional stabilization of c-fms mRNA by a labile protein during human monocytic differentiation. Mol Cell Biol 1989;9:769-775. 40. Prokipcak RD, Herrick DJ, Ross J. Purification and properties of a protein that binds to the C-terminal coding region of human C-myc mRNA. J Biol Chem 1994;269:9261-9269.