|Home | About | Journals | Submit | Contact Us | Français|
Cholangiocytes are the target cells in cholestatic models of ductal hyperplasia including bile duct ligation (BDL). We have shown that: (i) cholangiocytes express VEGFR-2 and VEGFR-3; (ii) VEGF-A and VEGF-C stimulate cholangiocyte proliferation via an autocrine mechanism; and (iii) chronic administration of VEGF-A prevents cholangiocyte damage induced by hepatic artery ligation. Caffeic acid phenethyl ester (CAPE) induces growth inhibition in different cells. Taurocholic acid (TC) protects cholangiocytes against injury induced by parasympathetic or sympathetic denervation. The aims of this study were to determine if: (i) CAPE induces bile duct damage; and (ii) TC prevents CAPE-induced bile duct damage by increasing cholangiocyte VEGF expression. Methods: Normal and BDL rats (immediately after surgery) were fed 1% TC or control diet in the absence/presence of daily IP injections of CAPE (10 mg/Kg BW). One week later, we evaluated: (i) cholangiocyte apoptosis, proliferation and ductal mass in liver sections; (ii) functional activity by measuring secretin-stimulated bile and bicarbonate secretion; and (iii) VEGF-A/C and VEGFR-2/R-3 expression in liver sections. In vitro, BDL cholangiocytes were exposed to CAPE (40 μM) in the absence/presence of TC (40 μM) with and without pretreatment with VEGF receptor inhibitors before evaluating cholangiocyte apoptosis and proliferation. Results: Chronic CAPE administration to BDL rats increased cholangiocyte apoptosis and decreased ductal mass. This effect was associated with reduced expression of VEGF-A, VEGF-C, VEGFR-2 and VEGFR-3. In vivo, TC feeding partly prevented CAPE-induced changes in cholangiocyte apoptosis and growth and loss of ductal secretion. The protective effect of TC was associated with enhanced VEGF-A, VEGF-C, VEGFR-2 and VEGFR-3. In vitro, TC partially prevented CAPE-induced increases in apoptosis and decreases in cholangiocyte proliferation. These changes were reversed by pretreatment with VEGF-receptor inhibitors. Conclusion: Manipulation of cholangiocyte VEGF expression by bile acids may be important in preventing the impairment of cholangiocyte proliferation by exogenous agents.
The intrahepatic biliary tree consists of a complex, ramified network of interconnecting tubular conduits of different diameters and function, lined by cholangiocytes (1, 2). Constitutively, normal cholangiocytes have low replicative activity (3, 4), but they proliferate or undergo apoptosis in cholangiopathies that are characterized by an abnormal balance between cholangiocyte proliferation and death (5). Cholangiocytes are the target cells in a number of animal models of cholestasis including bile duct ligation (BDL), bile acid feeding or administration of carbon tetrachloride (CCl4), models that are characterized by changes in cholangiocyte growth and/or apoptosis (6–10). In rats with BDL, cholangiocyte hyperplasia is associated with enhancement of basal and secretin-stimulated ductal secretion (3, 10). On the other hand, reduced biliary growth and enhanced cholangiocyte apoptosis is coupled with decreased secretin-stimulated cholangiocyte secretory activity (8, 9). These features are typical of the terminal stages of different cholangiopathies including primary biliary cirrhosis (11).
Caffeic acid phenethyl ester (CAPE) is the active component of propolis, a honeybee hive product. CAPE exhibits a broad spectrum of biological properties including antiproliferative (12) and anti-metastatic activities (13). For example, CAPE is able to suppress the growth of HepG2 (a human hepatocellular carcinoma cell line) xenografts in nude mice (14). No information exists on the role of CAPE on the regulation of cholangiocyte proliferation in cholestasis.
Bile acids modulate cholangiocyte pathobiology by entering these cells via a specific transporter (Na+-dependent bile acid transporter, ASBT) localized at the apical pole of cholangiocytes (15, 16). Entry of bile acids into these cells by the ASBT induces changes in cholangiocyte secretion, proliferation and apoptosis (6, 17–19). Taurocholic acid (TC) plays an important role in the regulation of cholangiocyte function (19–21). For example, TC feeding to normal rats stimulates cholangiocyte proliferation and secretin-stimulated choleresis (22). Furthermore, depletion of endogenous bile acid pools in BDL rats reduces cholangiocyte proliferation and secretin-stimulated ductal secretion and ASBT expression and bile acid transport activity compared to controls (23). TC prevents the induction of cholangiocyte death by apoptosis, and the loss of proliferative and functional responses of cholangiocytes in response to toxins (e.g., CCl4) and cholinergic or adrenergic denervation (19–21).
Vascular endothelial growth factor (VEGF) is a mitogen for vascular endothelial cells and regulates vascular pathophysiology (24, 25). While VEGF-A binds to vascular endothelial growth factor receptor type 2 (VEGFR-2), VEGF-C exerts its function by either interaction with VEGFR-2 or VEGFR-3 (24, 26). VEGF is secreted by several epithelia, where it modulates cell growth by autocrine and paracrine mechanisms (27, 28). We have previously demonstrated that VEGF-A and VEGF-C are secreted by cholangiocytes and play an important role in modulating cholangiocyte proliferation and apoptosis in response to cholestasis (26) and hepatic artery ligation (29).
The aims of this study were to assess: (i) the chronic in vivo and acute in vitro effects of CAPE on cholangiocyte apoptosis, proliferation and ductal secretion in cholestatic BDL rats; (ii) if CAPE effects on cholangiocyte apoptosis and growth are associated with changes in VEGF-A/C and VEGF receptor expression; (iii) if TC partially prevents the effects of CAPE on cholangiocyte function by affecting cholangiocyte VEGF expression; and (iv) if in vitro, inhibition of VEGF receptors can block the protective effects of TC on CAPE-induced bile duct damage.
Reagents were purchased from Sigma Chemical (St. Louis, MO) unless otherwise indicated. Rat chow, containing 1% TC or AIN-76 (bile acid control diet), was prepared from Dyets Inc. (Bethlehem, PA). The substrate for γ-glutamyl transpeptidase (γ-GT), N-(-L-glutamyl)-4-methoxy-2-naphthylamide, was purchased from Polysciences (Warrington, PA).
Male Fischer 344 rats (150−175 gm) were purchased from Charles River (Wilmington, MA) and kept in a temperature-controlled environment (20−22°C) with a 12:12 hour light-dark cycle. Animals had free access to water and chow. The experimental groups used in the studies are shown in Table 1. Specifically, we used rats that, immediately after BDL (for preparation of liver blocks or isolation of cholangiocytes) (3, 30) or BDI (for bile collection during intravenous infusion of Krebs Ringer Henseleit, KRH or secretin), were fed AIN-76 (bile acid control (BAC) diet) or 1% TC (an approximate dose of 275 μmol/day) (6) in the absence or presence of daily IP injections of 0.1% DMSO or CAPE (10 mg/kg body weight) (31) in 1:1 DMSO:NaCl for 1 week. Since we have previously shown (6, 19) that daily IP injections of DMSO to BDL rats fed bile acid control diet or TC do not affect the functional activity of cholangiocytes, we did not include these groups of animals in the study. We also evaluated the effect of CAPE on apoptosis and proliferation of normal rats. Before each experimental procedure, the animals were anaesthetized with sodium pentobarbital (50 mg/kg body weight) according to the regulations of the panel on euthanasia of the American Veterinarian Medical Association. All experiments were conducted under the guidelines of the local IACUC.
Paraffin embedded liver sections (4 μm, 3 sections analyzed per group) from the selected groups of animals were stained with hematoxylin and eosin (H&E) before determining lobular damage, necrosis and the degree of portal inflammation (32). Ten randomly selected portal areas and surrounding lobular areas were evaluated using light microscopy of H&E-stained sections. Liver sections were examined in a coded fashion with the microscope ECLIPSE E600 (Nikon Eclipse, Tokyo, Japan) (using BX-51 light microscopy Olympus, Tokyo, Japan). Two hundred cells per slide were counted in a coded fashion in ten non-overlapping fields.
Cholangiocyte apoptosis was evaluated by the quantitative terminal deoxynucleotidyl transferase biotin-dUTP nick end labeling (TUNEL) kit (ApopTag; Chemicon International, Inc.) (9, 33) according to the vendor's instructions in paraffin-embedded liver sections (4 μm, 3 sections analyzed per group). The number of apoptotic cholangiocytes in liver sections was evaluated in a coded fashion using a BX-51 light microscopy (Olympus, Tokyo, Japan). Two hundred cells per slide were counted in ten non-overlapping fields. We also evaluated the expression of the apoptotic marker Bax (34) by immunohistochemistry in liver sections from all the studied groups (see below).
Liver pieces were fixed in 10% buffered formalin for 2 hours and paraffin embedded at 55°C. Liver sections (4 μm, 3 sections analyzed per group) were mounted on glass slides coated with acetone amino-propyltriehoxylan (2%) solution. After deparaffination, endogenous peroxidase activity was blocked by a 20-minute incubation in methanolic hydrogen peroxide (2.5%). Sections were hydrated in graded ethanol and rinsed in 1× phosphate-buffered saline (1× PBS, pH 7.4). Endogenous biotin activity was blocked by using the Biotin Blocking System (DakoCytomation code X0590, Glostrup, Denmark) and the samples were rinsed in 1× PBS. Sections were incubated overnight at 4°C with the following antibodies from Santa Cruz Biotechnology, Inc.: (i) CK-19 (G-14, code sc-33120, dilution 1:100); (ii) PCNA (PC10, code sc-56, dilution: 1:200); (iii) VEGF-A (code sc-152, dilution: 1:100); VEGF-C (code sc-9047, dilution: 1:100), VEGFR-2 (A-3, code sc-6251, dilution: 1:100) and VEGFR-3 (C-20, code sc-321, dilution: 1:100). Samples were rinsed with 1× PBS for 5 minutes, incubated for 20 minutes at room temperature with secondary biotinylated antibody (Dako-Cytomation LSAB Plus System-HRP, code K0690, Glostrup, Denmark), then with Dako ABC for 20 minutes (DakoCytomation LSAB Plus System-HRP, code K0690, Glostrup, Denmark) and developed with 3−3′-diaminobenzidine (DakoCytomation Liquid DAB Plus Substrate Chromogen System, code K3468, Glostrup, Denmark). For all immunoreactions, negative controls (non-immune serum with the omission of the primary antibody) were also included. Immunohistochemistry observations were taken by BX-51 light microscopy (Olympus, Tokyo, Japan) with a Videocam (Spot Insight; Diagnostic Instruments, Inc., Sterling Heights, MI) and processed with an Image Analysis System (IAS; Delta Sistemi, Rome, Italy).
The intrahepatic bile duct mass was measured as the area occupied by the number of CK-19-positive bile ducts/total area × 100 in liver sections (4 μm, 3 sections analyzed per group) from the selected group of animals. The numbers of PCNA, VEGF-A, VEGF-C, VEGFR-2 and VEGFR-3 positive cholangiocytes were assessed in sections (4 μm, 6 sections analyzed per group) for each group. Positive cells were counted in six non-overlapping fields (magnification ×20) for each slide and the data expressed as percentage of positive cells. Light microscopy and immunohistochemical observation was performed by three independent pathologists in a blinded manner.
To demonstrate the effects of TC feeding on CAPE-induced functional damage of bile ducts, we measured basal and secretin-stimulated choleresis (a functional marker of cholangiocyte proliferation) (9, 10, 35) in the selected groups of animals. Briefly, following anesthesia with sodium pentobarbital, the selected BDI rats were surgically prepared for bile collection as described (10). When steady-state bile flow was reached, the selected BDI animal was infused with secretin (100 nM) for 30 minutes and by a final infusion of KRH for 30 minutes. Bile was collected every 10 minutes in pre-weighed tubes and immediately stored at −70°C before determination of bicarbonate concentration. Bile bicarbonate concentration (measured as total CO2) was determined by an ABL™ 520 Blood Gas System (Radiometer Medical A/S, Copenhagen, Denmark).
In vitro experiments were performed in virtually pure (98% by γ-GT histochemistry) (36) BDL cholangiocytes isolated by immunomagnetic separation using a monoclonal antibody (IgM, kindly provided by Dr. R. Faris, Brown University, Providence, RI) against an unidentified antigen expressed by all intrahepatic cholangiocytes (37). Cell number and viability (greater than 97%) were assessed by standard Trypan blue exclusion.
BDL cholangiocytes were incubated at 37°C with 0.2% BSA (basal) or CAPE (40 μM) (38) for 5 hours in the absence or presence of TC (40 μM) (22), with or without pretreatment with a VEGFR-2 (SU5416: VEGF Receptor 2 Kinase Inhibitor III, Calbiochem, Gibbstown, NJ) (39) or VEGFR-3 (MAZ51: VEGF Receptor 3 Kinase Inhibitor, Calbiochem, Gibbstown, NJ) (40) inhibitor. Subsequently, we evaluated expression for PCNA and the pro-apoptotic protein Bax (34) by immunoblots (17) in protein (10 μg) from whole cell lysates. After stripping, the amount of protein loaded was normalized by immunoblots for β-actin (17). The intensity of the bands was determined by scanning video densitometry using the phospho-imager, Storm 860, (GE Healthcare, Piscataway, NJ) and the ImageQuant TL software version 2003.02 (GE Healthcare, Little Chalfont, Buckinghamshire, England).
All data are expressed as mean ± SEM. The differences between groups were analyzed by Student's t test when two groups were analyzed, or analysis of variance (ANOVA) if more than two groups were analyzed. A P value of less than 0.05 was used to indicate statistically significant differences.
There was no significant difference in lobular damage, necrosis and the degree of portal inflammation in the experimental groups found in Table 1. Figure 1 shows a representative section from each group of animals examined and no visible changes are noted. Administration of CAPE to BDL rats significantly increased the number of TUNEL-positive cholangiocytes compared to BDL controls and BDL rats fed TC (Fig. 2A and Table 2). TC feeding partly prevented CAPE-induced increases in cholangiocyte apoptosis (Fig. 2A and Table 2). The number of apoptotic cholangiocytes (Fig. 2A–B and Table 2) was significantly decreased in BDL rats fed TC and injected with CAPE compared to BDL rats treated with CAPE alone; however, this number was not as low as BDL rats treated with DMSO (controls) in the absence or presence of the TC rich diet. Consistent with previous studies (19), TC feeding did not alter cholangiocyte apoptosis compared to BDL controls (Fig. 2A and Table 2). We found similar changes in Bax expression (Fig. 2B). There was no positive staining for Bax in the BDL control or BDL + TC groups. However, after chronic administration with CAPE, Bax expression greatly increases and is consequently decreased in animals treated with CAPE and fed a TC enriched diet. These data suggest that TC partially prevents CAPE-induced cell death. CAPE did not induce damage of cholangiocytes in normal rats (not shown).
Coupled with increased apoptosis (Fig. 2A–B and Table 2), chronic administration of CAPE to BDL rats decreased the number of PCNA- and CK-19-positive cholangiocytes in liver sections compared with BDL rats fed a bile acid control diet in combination with DMSO injections (controls) (Figs. 3 and and44 and Table 2). BDL rats fed a TC rich diet and injected with DMSO exhibited similar levels of apoptotic-, PCNA-, and CK-19-positive cholangiocytes (Figs. 3 and and44 and Table 2) compared to control rats. Similarly, in rats fed a TC rich diet in combination with CAPE treatment, the levels of the number of positive PCNA- and positive CK-19-cholangiocytes were similar to controls (Figs. 3 and and44 and Table 2) demonstrating a partial protective effect of TC against CAPE treatment. CAPE did not change the number of PCNA-positive cholangiocytes and intrahepatic ductal mass of normal rats (not shown).
Basal and secretin-stimulated bile flow and secretion of control BDI rats were similar to that of previous studies (Fig. 5) (10). Consistent with the notion that CAPE induces functional damage of bile ducts, secretin failed to increase bile and bicarbonate secretion in BDI rats treated with CAPE (Fig. 5). Similar to other studies showing that TC increases biliary growth (6), we found that TC feeding to BDI rats significantly increased both basal and secretin-stimulated bile and bicarbonate secretion compared to BDI control rats (Fig. 5). Keeping with the concept that TC feeding ameliorates the functional damage of bile ducts by CAPE, in BDI rats treated with CAPE + TC there was enhanced basal and secretin-stimulated bile flow and bicarbonate secretion compared to control BDI rats (Fig. 5).
Immunohistochemistry in liver sections showed that the expression of VEGF-A and VEGF-C (Fig. 6A) and the receptors VEGFR-2 and VEGFR-3 (Fig. 6B) was lower in cholangiocytes from BDL rats treated with CAPE compared with BDL controls and rats fed a TC enriched diet (Fig. 6A–B). TC feeding prevented the CAPE-induced decrease in VEGF-A, VEGF-C, VEGFR-2 and VEGFR-3 expression that was similar to that of BDL control rats (Fig. 6A–B). Quantitative data for cholangiocyte VEGF protein and receptor expression is found in Figure 6C.
Similar to our in vivo studies, in BDL cholangiocytes treated with CAPE, we found increased Bax protein expression (Fig. 7A) and decreased PCNA protein expression (Fig. 7B) compared to cholangiocytes treated with BSA. In vitro stimulation with TC alone did not change the expression of Bax (Fig. 7A), but significantly increased PCNA protein expression (Fig. 7B). CAPE-induced changes in Bax and PCNA protein expression were partly prevented by pre-treatment with TC (Fig. 7A–B). The effects of TC on CAPE-induced changes in Bax and PCNA protein expression were partially blocked by pre-incubation with VEGFR-2 or VEGFR-3 inhibitors (Fig. 7A–B). VEGFR-2 or VEGFR-3 inhibitors alone did not change either Bax or PCNA protein expression (not shown).
Our study demonstrates that TC, both in vivo and in vitro, partially prevents CAPE-induced cholangiocyte functional damage by changes in cholangiocyte VEGF expression. Specifically, we have shown that: (i) chronic treatment of BDL rats with CAPE increases apoptosis, and reduces cholangiocyte proliferation and secretin-stimulated ductal secretion; (ii) CAPE-induced changes in cholangiocyte apoptosis and proliferation are associated with reduced expression of VEGF-A, VEGF-C, VEGFR-2, and VEGFR-3; (iii) TC feeding partly prevents CAPE-induced changes in cholangiocyte apoptosis and proliferation, and maintains VEGF-A, VEGF-C, VEGFR-2, and VEGFR-3 expression; (iv) TC feeding reverses the CAPE-induced functional damage of bile ducts; and (v) in purified BDL cholangiocytes, TC moderately prevents CAPE-induced effects on cholangiocyte apoptosis and proliferation by sustaining the VEGF system.
Cholangiocyte proliferation is a typical hallmark of cholangiopathies that induce chronic damage of the biliary tree with dysregulation of the balance between the survival and death of cholangiocytes (9, 33, 41). Cholangiocyte proliferation in the course of cholangiopathies compensates for the loss of injured ducts (9, 10, 33); in fact, proliferating cholangiocytes show higher secretory activity (4, 10, 42), which is fundamental for maintaining a normal ductal secretion despite the loss of damaged bile ducts (9, 33). The cholestatic BDL rat model, which mimics the pathological conditions observed in human cholangiopathies (3, 10), is characterized by enhanced cholangiocyte proliferation (3, 10) and is widely used for evaluating the intracellular mechanisms of cholangiocyte proliferation/loss.
In the rat liver, CAPE treatment prevents CCl4-induced damage (43), inhibits proliferation and collagen synthesis of stellate cells at lower concentrations, and induces stellate cell apoptosis at high concentrations (12). Similar to BDL vagotomized rats (33), CAPE induced an increase in cholangiocyte cell death that was coupled with decreased cholangiocyte proliferation after chronic and acute stimulation. CAPE-induced cholangiocyte damage was achieved without changes in portal inflammation, hepatic damage and necrosis. These data suggest that CAPE could be used therapeutically to treat pathological conditions of abnormal cholangiocyte proliferation such as those observed in autosomal dominant polycystic kidney disease (ADPKD) (44). The data also suggest that CAPE may be an important therapeutic approach for decreasing ductular reaction in chronic liver diseases as a strategy to delay biliary fibrosis given the link between ductular reaction and biliary fibrosis (45).
Many studies have shown that cholangiocyte biology is influenced by bile acids including TC, ursodeoxycholate (UDCA) and tauroursodeoxycholate (TUDCA) (6, 18, 22). TC is not only a potent mitogenic stimulus (6), but is required for the proliferative response by cholangiocytes to cholestasis, as demonstrated by the fact that depletion of endogenous pools of TC reduces cholangiocyte growth and secretin-stimulated ductal secretion (23). We have previously demonstrated that TC administration is protective against cholangiocyte apoptosis induced by acute administration of CCl4 (6), and sympathetic (20) or adrenergic denervation (19).
Cholangiocyte proliferation is also regulated by neuropeptides, hormones and growth factors, including VEGF (17, 29). Cholangiocytes secrete VEGF-A and VEGF-C and express VEGFR-2 (Flk-1) and VEGFR-3 (Flt-4), all of which are amplified in BDL cholangiocytes (26, 29). Therefore, VEGF participates in the adaptive proliferative and apoptotic response of cholangiocytes to cholestasis by autocrine mechanisms (26, 29). For example, Gaudio, et al. found that administration of anti-VEGF-A or anti-VEGF-C antibodies (which neutralizes VEGF and reduces circulating levels of VEGF) decreases BDL-induced hyperplasia, thus confirming the important role of VEGF in the balance between cholangiocyte proliferation and loss (26, 29). Important studies have revealed that VEGF regulates cholangiocyte growth, including a study that has shown that cholangiocytes are able to produce a VEGF gradient demonstrated by an increasing amount of VEGF expression as bile ducts develop (46).
The effects of bile acids on VEGF expression have been shown mainly in cancerous cell types where bile acids affect angiogenesis thus altering cell growth, such as in human colon cancer in which deoxycholic acid significantly enhances the angiogenic potential of the tumor (47) or the promoting role of chenodeoxycholic acid in the development of human esophageal cancer supported by a higher production of VEGF, via the COX-2 pathway (48). VEGF has been shown to play a role in the cholangiocyte proliferative response (26, 29); however, the effects of TC on VEGF expression have not been shown in the context of the regulation of cholangiocyte hyperplasia. Here, we propose that the in vivo and in vitro TC protective effects on CAPE-induced damage of the biliary epithelium are due to changes in cholangiocyte VEGF expression.
In our final set of experiments, we set out to determine if the observed in vivo effects of CAPE and TC + CAPE are specific to cholangiocytes and if acute treatment would elicit similar results. Similar (although not as dramatic) to in vivo results, we found that acute CAPE stimulation in vitro decreased proliferation coupled with increased apoptosis that is blocked by inhibition of VEGF receptor activation. Gaudio, et al. have previously shown that there is an autocrine regulation of cholangiocyte VEGF expression (26, 29). The findings in our study lend support to the concept that autocrine VEGF regulation may be modulated by bile acids in cholangiocytes and that our findings in vivo can be duplicated in vitro.
Other studies support the concept that the activation of the VEGF system represents a protective mechanism during experimental liver damage (29). Our data support these findings, in particular the concept that bile acids, like TC, and VEGF may play an important role to maintain the balance between loss and growth in cholangiopathies. In conclusion, our findings may provide novel perspectives in the regulation of cholangiocyte loss in chronic cholestatic liver diseases via manipulation of cholangiocyte VEGF expression.
We sincerely thank Dr. Gianfranco Alpini for his assistance and expert advice in the development of this manuscript. Also, we thank Ms. Julie Venter for her assistance with the animal studies.
The study was supported by a grant award from Scott & White to Heather Francis, the Dr. Nicholas C. Hightower Centennial Chair of Gastroenterology from Scott & White, a grant award from Health and Labour Sciences Research Grants for the Research on Measures for Intractable Diseases (from the Ministry of Health, Labour and Welfare of Japan) and from a Grant-in-Aid for Scientific Research C (19590744) from JSPS to Dr. Ueno, by University funds to Dr. Onori and PRIN 2007 and Federate Athenaeum funds from University of Rome “La Sapienza” to Prof. Gaudio and Prof. Alvaro, by MIUR (PRIN) grants: #2007, prot. HPT7BA_003 to Prof. Alvaro and an NIH K01 grant award (DK078532) to Dr. DeMorrow.