A neurotransmitter, norepinephrine (NE), amplifies the mitogenic effect of epidermal growth factor (EGF) in the liver by acting on the α1-adrenergic receptor coupled with G protein, Gαh. However, the molecular mechanism is not well understood. Gαh is known as a transglutaminase 2 (TG2), a cross-linking enzyme implicated in hepatocyte proliferation. We investigated the effect of NE on EGF-induced cell proliferation and TG2 activity using hepatocytes isolated in periportal and perivenous regions of the liver, which differ in proliferative capacity. Periportal hepatocytes (PPH) and perivenous hepatocytes (PVH) were isolated by the digitonin-collagenase perfusion technique. EGF or NE receptor binding was analyzed by Scatchard analysis. Changes in NE-induced DNA synthesis, EGF receptor (EGFR) dimerization and phosphorylation, and TG2 activity were measured. NE enhanced EGF-induced DNA synthesis, EGF-induced EGFR dimerization, and its phosphorylation in PVH but not in PPH. [3H]NE binding studies indicated that PVH was found to have a greater affinity and number of receptors than PPH. Furthermore, NE treatment decreased TG2 activity and increased phospholipase C activity in PVH although TG2 level showed no change. These results suggest that NE-induced amplification of EGF-induced DNA synthesis especially in PVH is caused by upregulation of EGFR activation through the switching of function from TG2 to Gαh.
- epidermal growth factor
- DNA synthesis
- α1 adrenergic receptor
hepatocytes are classified into periportal hepatocytes (PPH) and perivenous hepatocytes (PVH) on the basis of previous studies that demonstrated zonal differences in metabolism (13, 22) and cellular proliferation (12, 29, 36). In an experimental model induced by 70% partial hepatectomy, PPH and PVH show different growth capacities, and this regenerative growth proceeds in a characteristic pattern and results in several sequential waves of DNA synthesis and mitosis starting in the periportal area and ending in the perivenous area (36). This process is regulated by endogenous growth factors and cytokines (11, 33). We and others have reported that cultured PPH and PVH show different responses to various mitogens such as epidermal growth factor (EGF) and hepatocyte growth factor (HGF) (14, 21, 29). However, the molecular mechanisms underlying the zonal differences in proliferative response are not well understood.
In addition, the liver regeneration is thought to be innervated by both the sympathetic and parasympathetic nerves. A sympathetic and/or parasympathetic denervation has previously been reported to cause a delay in the DNA synthesis and cellular proliferation after partial hepatectomy (15, 35, 42). A neurotransmitter, norepinephrine (NE), rises rapidly in the plasma within 1 h after hepatectomy (5) and induces secretion of EGF from Brunner's glands of the duodenum (38). In hepatocyte culture, NE amplifies the mitogenic effects of EGF by acting on the α1-adrenergic receptor (4). The specificity to this receptor type was determined by the use of a selective antagonist. However, the molecular mechanism by which NE has this effect remains unclear.
Adrenergic agonists such as epinephrine and phenylephrine have been shown to enhance hepatocyte proliferation through binding to the α1B-adrenergic receptor that is coupled with Gαh (32, 50). Gαh has the role of guanosine triphosphatase (GTPase) like other G proteins, which bind and hydrolyze GTP (20, 27). Gαh is also known as a transglutaminase 2 (TG2), which catalyzes a calcium-dependent transamidation reaction, resulting in protein-protein cross-linking through the formation of ε-(γ-glutamyl)lysine isopeptide bonds (34). Thus Gαh (TG2) is a bifunctional enzyme with GTPase and cross-linking activities. The GTPase function of Gαh has been shown to be involved in regulation of cell cycle progression and receptor-mediated signaling (31). On the other hand, the cross-linking function of TG2 has been shown to be involved in cross-linking of the extracellular matrix (48), apoptosis (30), and downregulation of cellular proliferation (50). We reported that inhibition of de novo synthesis of TG2 resulted in promoting the growth of cultured rat hepatocytes in the presence of EGF (23) or HGF (24). Moreover, we have recently reported that TG2 is involved in the difference in growth capacities between PPH and PVH through downregulation of EGF receptor (EGFR) activation (29).
These findings raise the possibility that NE may modulate the zonal difference in proliferative capacity between PPH and PVH through regulation of switching between GTPase function of Gαh and cross-linking function of TG2.
In the present study, to test this possibility, we investigated the influence of NE on EGF-induced EGFR activation and DNA synthesis and TG2 activity in cultured PPH and PVH.
MATERIALS AND METHODS
Animals and materials.
Male Wistar rats (SLC, Hamamatsu, Japan) were kept at a controlled temperature (23 ± 1°C) under a 12-h light-dark cycle and were maintained on a standard diet and water. All animal experiments were approved by the Animal Care and Use Commitee of the Tohoku Pharmaceutical University. [Methyl-3H]thymidine, [3H]NE, [125I]EGF, [1,4-14C]putrescine, and inositol 1,4,5-triphosphate [3H]radioreceptor assay kit were obtained from Perkin-Elmer Life Sciences (Boston, MA). Collagenase was obtained from Nitta Gelatin (Osaka, Japan). Digitonin, NE, prazosin, yohimbine, metoprolol, and butoxamine were obtained from Sigma-Aldrich (St. Louis, MO). Mouse EGF was obtained from Biomedical Technologies (Stoughton, MA). Anti-ErbB-1 (EGFR) polyclonal antibody, anti-phospho-EGFR at Y1173 antibody and goat anti-rabbit IgG antibody conjugated with peroxidase were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-TG 2 antibody was obtained from Neomarker (Fremont, CA). NE (g-l-glutamyl)-l-lysine (Gln-Lys) isopeptide mouse monoclonal antibody (81D4) was obtained from Covalab (Villeurbanne, France).
PPH and PVH isolation and culture.
PPH and PVH were isolated from separate animals by the digitonin-collagenase perfusion technique; a detailed procedure has been described previously (36). Separation of PPH and PVH was confirmed by measuring enrichment in two specific marker enzymes, alanine aminotransferase for PPH (41) and glutamine synthetase for PVH (49). Viability of hepatocytes was determined by Trypan blue staining and was at a level of more than 90%. PPH and PVH were placed in 12-well collagen-coated plates (Iwaki, Tokyo, Japan) at a density of 0.8 × 105 cells/cm2 in Williams' E medium containing 10% fetal bovine serum, 10−9 M insulin, 10−9 M dexamethasone, 1% (vol/vol) antibiotics (GIBCO, Grand Island, NY). After 3-h incubation, the medium was replaced with a serum-free medium. The hepatocytes were cultured with a serum-free medium for 24 h and were then treated with or without 10−8 M EGF.
Measurement of DNA synthesis.
DNA synthesis was assessed by [methyl-3H]thymidine incorporation into hepatocytes. [Methyl-3H]thymidine was included from 24 to 48 h after EGF treatment. In the case of treatment with NE and/or adrenoceptor antagonists (NE, 10−4 M; prazosin, 10−5 M; yohimbine, 10−5 M; metoprolol, 10−5 M; butoxamine, 10−5 M), the hepatocytes were treated 1 h before EGF treatment. The radioactivity of [methyl-3H]thymidine incorporated into cells was measured with a Beckman LS 6500 liquid scintillation counter (Beckman Coulter, Fullerton, CA).
Assay for binding of radiolabeled ligand to cultured hepatocytes.
Hepatocytes were plated in 12-well collagen-coated plates at a density of 0.8 × 105 cells/cm2. After attachment, hepatocytes were incubated for 24 h at 37°C with serum-free medium. Culture medium was then replaced with binding buffer containing 50 mM HEPES (pH 7.4), 128 mM NaCl, 5 mM KCl, 1.2 mM CaCl2, 5 mM MgSO4, and 0.5% BSA. Hepatocytes were then incubated at 4°C with binding buffer containing various concentrations of [125I]EGF (specific activity; 81.4 TBq/mmol) or [3H]NE (specific activity; 1.8315 TBq/mmol). Radioactivity of radiolabeled ligand bound to hepatocytes was measured with a gamma counter (Aloca ARC-370M, Tokyo, Japan). Nonspecific binding was determined in the presence of a 100-fold concentration of unlabeled ligand and amounted to less than 10% of the total binding.
Cross-linking of EGFRs.
Cross-linking of EGFRs on hepatic membrane was performed by using bis-(sulfosuccinimidyl) suberate (BS3) (Pierce), a cross-linking reagent; the detailed procedure has been described previously (37). Briefly, after treatment with EGF for indicated times, the hepatocytes were incubated for 1 h at 4°C with 2 mM BS3. The quench solution (0.5 M Tris·HCl buffer, pH 7.4) was then added to a final concentration of 0.25 M and incubated for 10 min at 4°C. The hepatocytes were subjected to hepatic membrane fractionation.
Plasma membrane isolation.
Plasma membranes from cultured hepatocytes were isolated by using the Percoll density gradient method described by Cefaratti et al. (1) with minor modification. Cultured hepatocytes were homogenized in an ice-cold isolation medium containing 5 mM HEPES, 250 mM sucrose, and 1 mM EGTA supplemented with 1% protease inhibitor cocktail (Sigma-Aldrich) with 30 strokes on a loose-fitting Dounce homogenizer. The homogenates were then centrifuged at 600 g for 10 min, and the resulting pellets were resuspended in the isolation medium. A volume (10.4 ml) of this was mixed with 1.4 ml of Percoll (GE Healthcare) in a 15-ml Cortex tube and centrifuged at 34,540 g for 30 min. A distinct layer close to the top of the tube containing plasma membranes was collected and washed in 50 mM Tris·HCl, pH 7.4. The isolated plasma membranes were subjected to Western blot analysis.
Western blot analysis.
The hepatic membranes were mixed with sample buffer for SDS-PAGE containing 62.5 mM Tris·HCl buffer (pH 6.8), 10% glycerol, 2% SDS, 5% β-mercaptoethanol, and 0.025% bromophenol blue. The mixture was boiled at 100°C for 3 min and was size separated by SDS-PAGE on 6% polyacrylamide gradient gel. The separated membrane proteins were transferred onto polyvinylidene difluoride (PVDF) membranes (GE Healthcare) and immunoblot analysis was carried out with use of the appropriate antibodies. Immunoreactive bands were detected with ECL Western blotting detection reagents (GE Healthcare) and exposed with X-ray film (Fujifilm).
Cultured hepatocytes were lysed in a buffer composed of 50 mM Tris·HCl (pH 8.0), 1% Nonidet P-40, 10% glycerol, and 1% protease inhibitor cocktail, and the cell debris was removed by centrifugation at 15,000 g for 5 min. The cell lysates were incubated overnight with anti-EGFR antibody at 4°C. The immunocomplexes were precipitated with protein G-Sepharose (GE Healthcare). The immunoprecipitates were subjected to SDS-PAGE and subsequent Western blotting with anti-isopeptide antibody.
Measurement of transglutaminase activity.
Transglutaminase activity was assessed by [1,4-14C]putrescine incorporation into proteins of hepatocytes. [1,4-14C]putrescine was added at the time of EGF treatment in cultured hepatocytes. After 1-h incubation, hepatocytes were harvested with distilled water. The cell suspension was homogenized by a loose-fitting Dounce homogenizer and was precipitated by 10% trichloroacetic acid. The precipitates were spotted onto Whatman 3MM filter paper, which was then immersed in 10% trichloroacetic acid solution. The radioactivity of each filter paper was measured with a Beckman LS6500 liquid scintillation counter.
Assay for PLC activity.
Phospholipase C (PLC) activity was assessed by production of inositol 1,4,5-triphosphate (IP3) in hepatocytes using IP3 [3H]radioreceptor assay kit. The production of IP3 was determined in the presence and absence of 10−4 M NE. IP3 production in hepatocytes was stopped at 1 h by adding ice-cold 20% trichloroacetic acid after removing the media as recommended by the manufacturer. The radioactivity was measured with a Beckman LS6500 liquid scintillation counter.
For the binding assay of [125I]EGF or [3H]NE, Kd values of PPH and PVH were compared as correlation coefficients. Student's t-test was used for the statistical analysis of [1,4-14C]putrescine incorporation into PPH and PVH and IP3 production as PLC activity.
Effect of NE and adrenoceptor antagonists on EGF-induced DNA synthesis in cultured PPH and PVH.
To examine the effect of NE on zonally different hepatocyte proliferation, we investigated DNA synthesis induced by EGF treatment with or without NE in cultured PPH and PVH (Fig. 1). NE markedly potentiated the EGF-induced DNA synthesis in PVH but not in PPH. Next, we investigated the effect of selective adrenoceptor antagonists on the ability of NE to stimulate DNA synthesis in hepatocytes in the presence of EGF (Fig. 1). Only the specific α1-adrenoceptor antagonist, prazosin, prevented the effect of NE on DNA synthesis in PVH. Yohimbine, an α2-adrenoceptor antagonist; metoprolol, a β1-adrenoceptor antagonist; and butoxamine, a β2-adrenoceptor antagonist, were almost completely ineffective. The addition of 10 ng/ml of aphidicolin (a specific inhibitor of the replicative enzyme, DNA polymerase α) to cultured hepatocytes completely abolished EGF stimulated [methyl-3H]thymidine incorporation without any effect on cell viability (data not shown).
[3H]NE binding to cultured PPH and PVH.
To characterize the zonal differences of NE-induced response between PPH and PVH in EGF-induced DNA synthesis, characteristics of [3H]NE binding to PPH and PVH were assessed. Scatchard plots comparing [3H]NE binding in both subpopulations are shown in Fig. 2. In both subpopulations, the Scatchard plots were curvilinear, indicating the presence of two classes of NE binding sites; one of high affinity and one of low affinity. Figure 2 (inset) shows saturation curves of [3H] NE-specific binding to its receptor. In PPH, dissociation constants (Kd) were 147 (high-affinity) and 1,500 (low-affinity) pM, and the numbers of binding sites (Bmax) were 2.52 (high-affinity) and 8.91 (low-affinity) pM, respectively. On the other hand, in PVH, Kd values were 112 (high-affinity) and 938 (low-affinity) pM, and Bmax values were 8.74 (high-affinity) and 31.9 (low-affinity) pM. PVH was found to have a greater affinity and number of receptors compared with PPH for NE.
Effect of NE on [125I]EGF binding to cultured PPH and PVH.
To identify the locus of the NE-dependent alterations in EGF-induced DNA synthesis, the characteristics of [125I]EGF binding to PPH and PVH were assessed. As shown in Fig. 3A, a Scatchard plot of the binding data for control PPH was curvilinear and yielded two apparent Kd of 9.12 pM (high-affinity) and 107 pM (low-affinity). Bmax was 0.95 (high-affinity) and 6.15 (low-affinity) pM, respectively. In NE-treated PPH, Kd values were 8.06 (high-affinity) and 101 (low-affinity) pM, and Bmax values were 0.82 (high-affinity) and 6.71 (low-affinity) pM, respectively. In PPH, NE treatment showed no significant change in affinity of EGF for either of these sites or in receptor number. In control PVH, Kd values were 30.1 (high-affinity) and 212 (low-affinity) pM, and Bmax values were 1.36 (high-affinity) and 7.02 (low-affinity) pM, respectively (Fig. 3B). On the other hand, Kd values in NE-treated PVH were 9.04 (high-affinity) and 195 (low-affinity) pM, and Bmax values were 0.87 (high-affinity) and 6.27 (low-affinity) pM, respectively (Fig. 3B). In PVH, NE treatment led to a significant increase in the affinity of EGF to its receptor, whereas there was no significant difference in Bmax values.
Effect of NE on EGFR activation in cultured PPH and PVH.
Next, we investigated the effect of NE on EGFR dimerization. Addition of EGF induced redistribution of the receptors from the 170- to 175-kDa band to the 340- to 350-kDa band, corresponding to the monomeric and dimeric forms of EGFR, respectively. As the control for EGFR activation, hepatic membranes from cultured hepatocytes without EGF treatment were used (Fig. 4A, lane 1). As shown in Fig. 4, the dimerization of EGFRs of both subpopulations greatly increased in response to EGF stimulation. In the control group, the dimerization of EGFRs following EGF treatment was greater in PPH than in PVH. In comparison, in the NE-treated group, the dimerization of EGFRs in PVH significantly increased in response to EGF stimulation, but not in PPH.
EGF-induced EGFR dimerization promotes autophosphorylation at the intracellular tyrosine residue, thereby initiating an intracellular signaling pathway. The effect of NE on EGFR phosphorylation at Y1173 on cell membrane obtained from hepatocytes is shown in Fig. 5. The phosphorylation of EGFR was examined in cell membranes of cultured hepatocytes at various time points following EGF treatment. As the control for EGFR phosphorylation, cell membranes from cultured hepatocytes without EGF treatment were used (Fig. 5A, lane 1). As shown in Fig. 5, the phosphorylation of EGFR on hepatic membranes from control and NE-treated groups showed good correlations with its dimerization (Fig. 4).
Effect of NE on TG2 expression and its cross-linking activity.
TG2 is a bifunctional enzyme possessing transglutaminase cross-linking and GTPase activities. The cross-linking activity has been implicated to be involved in extracellular matrix organization and in inhibition of cell growth and proliferation. On the other hand, the GTPase function as G protein (Gαh) has been shown to be involved in regulation of cell cycle progression and receptor-mediated signaling. The Gαh-coupled receptors include the α1-adrenoceptor. Therefore, we speculated that NE stimulation of EGF-induced hepatocyte proliferation in the perivenous region might be occurring through control of TG2's bifunctional activities in cross-linking and of being a GTPase. To investigate the effect of NE on TG2 expression and its cross-linking activity, we performed Western blot analysis using a TG2 antibody to show its protein expression and [1,4-14C]putrescine incorporation assay to show TG2 activity. As shown in Fig. 6, in control cells, TG2 expression and its activity in PVH were slightly higher than those in PPH. In NE-treated cells of both subpopulations, TG2 expression was the same as control level (Fig. 6A), whereas TG2-catalyzed cross-linking activity was significantly decreased in PVH but not in PPH (Fig. 6B). In all cases, EGF treatment showed no changes in TG2 expression and activity.
Effect of NE on TG2-catalyzed isopeptide cross-linking formation.
Next, to investigate the effect of NE on TG2-catalyzed cross-linking formation of ε-(γ-glutamyl)lysine isopeptide bonds, we performed immunoprecipitation with anti-EGFR antibody and subsequent Western blotting with anti-isopeptide antibody. As shown in Fig. 7, the distribution of ∼170 kDa cross-linked isopeptide bonds in EGFR was observed. In the NE-untreated group, the formation of isopeptide bonds in EGFR in PVH was higher than that in PPH. By comparison, in the NE-treated group, the level of isopeptide bond in EGFR in PVH significantly decreased, but not in PPH. In both cases, EGF treatment showed no changes in formation of isopeptide bonds in EGFR (data not shown). Data showed a good correlation between TG2 activity and TG2-catalyzed cross-linked products, isopeptide bonds.
Effect of NE on PLC activity.
PLC-δ1 is identified as an effector molecule in Gαh-mediated signaling. Activated PLC catalyzes the hydrolysis of phosphatidylinositol 4,5-bisphosphate to produce two second messengers: IP3 and diacylglycerol. Thus, to investigate whether Gαh-mediated PLC-δ1 is activated by NE treatment, we determined the production of IP3 by PLC. As shown in Fig. 8, in PPH, NE treatment showed almost no changes in PLC activity. By comparison, in PVH, its activity was significantly increased to seven times the level of NE-untreated cells. In both subpopulations, EGF treatment showed a slight increase in its activity (data not shown).
Numerous studies using in vivo and in vitro models have provided evidence that a neurotransmitter, NE, amplifies the mitogenic effect of growth factor, including EGF in the liver. However, the molecular mechanism of NE-induced comitogenic effect at the proliferative state of the liver is not well understood. To understand this phenomenon in detail, we examined hepatocytes isolated in periportal and perivenous regions of the liver, which differ in proliferative capacity.
Several previous reports have indicated that NE has been shown to act via the α1-adrenoceptor on parenchymal hepatocytes to enhance their responsiveness to EGF in a serum-free primary cultured system (3, 19). These findings are consistent with our present data on DNA synthesis using selective adrenoceptor antagonists (Fig. 1). In the present results, we found here that NE significantly increased EGF-induced DNA synthesis only in PVH. To confirm this zonal specificity, we provide data on the binding of [3H]NE to both subpopulations. Our data indicate that PVH has a greater affinity and number of receptors than PPH for NE (Fig. 2). This higher affinity for NE in PVH may be involved in enhancement of EGF-induced DNA synthesis. We and others reported that PPH responded to EGF with higher sensitivity than PVH (10, 12, 29). However, the present data indicated that the zonal differences in EGF-induced DNA synthesis in both subpopulations disappeared with NE treatment. A possible reason for a drastic change in EGF-induced DNA synthesis in PVH is that the characteristics of EGFR may be changed by NE treatment. Thus, to investigate this, we studied the characteristics of [125I]EGF binding to PPH and PVH. Our results indicated that only in PVH did NE significantly increase in the affinity of EGF to its receptor, without a significant change in the number of receptors (Fig. 3). These results suggest that a subclass of high- but not low-affinity EGFR is upregulated by NE treatment. However, Cruise et al. (3) reported that NE decreased the number of EGFR, without a change in receptor affinity to its ligand, regardless of the enhancement of EGF-induced DNA synthesis. A possible reason for the discrepancy is the different conditions used for culture: Cruise et al. cultured hepatocytes under low-density conditions, whereas we cultured them under high-density but not confluent conditions. It has been reported that culturing hepatocytes at low density gives rise to a rapid loss of liver-specific functions (9). In fact, the previous study indicated that specific markers in PPH and PVH were abolished in cells cultured under low-density conditions (29). Therefore, we think that hepatocytes cultured under high-density conditions reflect the in vivo conditions to some extent.
In general, it is assumed that EGF preferentially binds to high-affinity EGFR, which is in equilibrium with low-affinity EGFR, and thereafter the EGF-EGFR complexes undergo dimerization, resulting in activation of the downstream signaling pathway (7, 17). Indeed, the present results show that the expression of EGF-induced dimerization was highly correlated with the appearance of high-affinity EGFR in both cases: the control group and the NE-treated group (Fig. 4). Moreover, the pattern of phosphorylation of EGFR at tyrosine 1173, which is essential for EGF-induced hepatocyte growth, was nearly parallel to that of dimerization of its receptor (Fig. 5). Therefore, we think that NE-induced EGFR upregulation as well as affinity for its ligand, dimerization, and phosphorylation must be related to hepatocyte proliferation. In the present study, NE stimulates EGFR dimerization and tyrosine phosphorylation in the absence of EGF stimulation and that this effect is greater in PVH than in PPH. Previous reports have suggested that the EGFR may be utilized by G protein-coupled receptor (GPCR) as intermediate signaling protein (8). GPCR-induced activation of EGFR signaling is the so-called EGFR transactivation that has been originally designated as ligand-independent tyrosine phosphorylation of EGFR (6). Moreover, GPCR has been demonstrated to stimulate the activity of a membrane-spanning matrix metalloprotease, resulting in the release of heparin-binding EGF and transforming growth factor-α and, subsequently, in a ligand-mediated autocrine and/or paracrine activation of EGFR (40, 44). Therefore, the present results suggest that NE treatment alone may stimulate EGFR dimerization and phosphorylation by GPCR/EGFR cross talk as a mechanism for NE potentiation of EGFR signaling, but this effect is only modest in the present primary cultured system.
In the previous study, we have found that TG2, which catalyzes the cross-linking reaction between proteins by the formation of isopeptide bonds, regulates the binding affinity of high-affinity EGFR but not that of low-affinity receptor (29). Moreover, we have shown that TG2 downregulates EGFR activation through intramolecular cross-linking of its receptor, resulting in inhibition of hepatocyte proliferation (28). Thus, to investigate the effect of NE on TG2 activity, we measured its protein expression and cross-linking activity. As a result, TG2-catalyzed cross-linking activity represented by [1,4-14C]putrescine incorporation and by detection of isopeptide bonds was decreased in PVH but not in PPH, whereas TG2 protein expression showed no change in both subpopulations (Figs. 6 and 7). The present data suggest that this decrease in TG2 cross-linking activity in PVH may lead to amplification of the DNA synthesis. In addition to its cross-linking activity, TG2 has a role performing the GTPase function of a G protein (Gαh). It has been reported that α-adrenergic stimulation of hepatocyte proliferation appears to act, at least in part, through activated GTPase performing as Gαh (50). Adrenergic signaling occurs through activation of PLC, which in turn produces the two intracellular messengers, diacylglycerol and IP3. These intermediate messengers mediate the activation of protein kinase C and intracellular calcium elevation. The downstream events include activation of cell cycling genes and cell proliferation in malignant hamster fibrosarcoma (31). Wu et al. reported that Gαh-dependent activation of PLC-δ1 by phenylephrine was preferentially involved in the enhancement of rat hepatocytes proliferation (50). In the present study, NE treatment alone significantly increased the PLC activity in PVH but not in PPH (Fig. 8) and did not influence the DNA synthesis in either subpopulation (Fig. 1). Kimura and Ogihara (25, 26) indicated that α- and β-adrenoceptor agonists alone did not significantly influence hepatocyte DNA synthesis and proliferation. Thus, considering the results of previous reports in addition to our present findings, it is possible that NE-induced switching from TG2 to Gαh, i.e., decrease of TG2 cross-linking activity, is far more important for the early stage of cell proliferation than Gαh-dependent downstream signaling events.
We have demonstrated that the level of TG2 in PVH is higher than that in PPH, and this higher level of TG2 in PVH leads to lower EGF-induced DNA synthesis than PPH through downregulation of EGFR by TG2 (28, 29). The present results suggest that NE-induced amplification of EGF-induced DNA synthesis in PVH is caused by upregulation of EGFR activation through decrease of TG2 cross-linking activity, i.e., NE-induced switching from TG2 to Gαh.
It has been reported that EGF induces the PLC-γ activation by the autophosphorylation of EGFR at Y1173 (2). Indeed, EGF slightly increased IP3 production in both subpopulations (data not shown). IP3 mediates calcium release from intracellular stores, affecting a host of calcium-dependent enzymes such as Ral (18) and nuclear factor-κB (45). However, in the present study, the activity of TG2 that catalyzes a calcium-dependent cross-linking reaction was decreased by NE treatment (Fig. 6B). Therefore, IP3-mediated calcium oscillation may not be related to the activation of TG2.
Recent studies have shown that the alcohol-induced switching of function from Gαh to TG2 may play an important role in ethanol-induced liver disease (50, 51). Tatsukawa et al. (46) reported that ethanol-induced enzymatic TG2 induced hepatocyte apoptosis via transcription factor Sp1 cross-linking and inactivation, with resultant downregulation of c-Met expression required for hepatocyte viability. Moreover, TG2, acting as a G protein, has recently been reported to have a protective role against Fas-mediated death pathway in hepatocytes (43). These reports together with ours suggest switching from enzymatic to nonenzymatic TG2 function may be involved in the recovery program of the liver after various kinds of hepatic injury. In the present study, the switching was mediated by NE, a neurotransmitter. Thus the nervous system may play an important role in homeostasis or disease onset of the liver.
The manipulation of TG2/Gαh activity will modulate the NE-EGFR pathway and may thus be exploited therapeutically. A variety of compounds such as cysteamine, 5-(biotinamido)pentylamine, and gluten peptides have been used to suppress TG2-catalyzed protein cross-linking. However, since several of these compounds contain primary amines or potential inhibitory motifs, it remains unclear whether the observed effects are due to an excess of competing amine donor or by blocking TG2 turnover (16). On the other hand, TG2 inducers such as ethanol and retinoic acid have not only inducible effect of TG2, but also multiple possible effects (39, 47). Thus resolution of this issue will require further detailed studies using the highly selective reagents to manipulate the TG2 activity or a more direct method.
Adrenergic regulation is now thought to be involved in the hepatic regenerative process in vivo. Until now, effect of NE on liver regeneration was not well understood. In the present study, we demonstrate the possibility that NE affects hepatocyte proliferation by the modification of EGF receptor activation through the switching of function between TG2 and Gαh. Moreover, these results raise the possibility that NE may regulate the characteristic pattern of cell growth for PPH and PVH. However, the increase in GTPase activity as Gαh may be associated with a concomitant decrease in the TG2 cross-linking activity. What is more clear is that NE-mediated change in relative amount of TG2 vs. GTPase acting as Gαh may well play a significant role in the regulation of zonally different hepatocyte proliferation.
No conflicts of interest, financial or otherwise, are declared by the author(s).
- Copyright © 2010 the American Physiological Society