WB-F344 cells were plated in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% (v/v) fetal bovine serum (FBS) (both from Gibco, Grand Island, NY, USA) (complete medium). When the cells were 80–90% confluent, they were detached using trypsin and reseeded at a density of 1×10⁵ cells/ml in 100-mm dishes. WB-F344 cells were cultured overnight in serum-free medium prior to the experiments. Cells were either stimulated, or not stimulated, with different concentrations of CTGF (1 or 5 ng/ml; PeproTech, Rocky Hill, NJ, USA), a CTGF inhibitor (WP631, 0.01 nM; Alexis Biochemical, San Diego, CA, USA) and cultured for 24 h. Total mRNA and proteins were then extracted.

WB-F344 cells (6,000 cells/well) were seeded into 96-well plates and treated with CTGF or WP631 as described above. A cell viability assay was performed using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT). Three duplicate wells were used for each sample. Twenty microliters of MTT (5 mg/ml) was added to each well and incubated for 4 h. The supernatant was then removed and 150 μl DMSO was added to dissolve the formazan crystals. Plates were then transferred to a plate reader and the absorbance was measured at 450 nm.

Quantitative real-time PCR was performed using an ABI PRISM 7300 Sequence Detection (Applied Biosystems, Foster City, CA, USA) and the ABI Power SYBR-Green PCR Master Mix kit (Applied Biosystems, Warrington, UK). The primers used for amplification (Table I) were designed to be between 60 and 150 bp in length according to the PE Applied Biosystems guidelines for the comparative C_T method. A final primer concentration of 300 nM was assessed as the optimum for use under all conditions to ensure that no nonspecific amplification occurred in the sample wells. RT was carried out for 2 min at 5°C before incubation for 10 min at 95°C to inactivate the reverse transcriptase, which otherwise interferes with the DNA polymerase. Forty cycles at 95°C for 15 sec followed by 60°C for 60 sec were performed, and the fluorescence was measured at the end of each extension cycle. A SYBR-Green dissociation curve was constructed at the end of the reaction to ensure specificity; the temperature was increased from 65 to 95°C in 0.1°C/sec increments, and the fluorescence signals were measured and plotted against the temperature. The amount of target, normalized to an endogenous reference [glyceraldehyde 3-phosphate dehydrogenase (GAPDH)] and relative to a calibrator (WB-F344 cells cultured untreated), was given by the formula 2^−ΔΔCt, as determined by the ABI PRISM 7300 System Software’s (Applied Biosystems) built-in algorithm using an adaptive baseline to determine the Ct. The relative amounts were expressed as the means ± SD from three independent experiments.

Total proteins were extracted using a cellular protein extraction solution, and protein concentrations were measured using the BCA protein assay (Pierce, Rockford, IL, USA). After boiling for 10 min, the lysates were separated on 12% SDS-PAGE gels. Proteins were transferred to nitrocellulose membranes (Amersham Biosciences, Piscataway, NJ, USA), which were blocked in TTBS containing 5% non-fat dried milk for 2 h at room temperature. The membranes were then incubated with anti-ALB, anti-AFP, anti-CK-19 (all from R&D System, USA) and anti-CTGF (Abcam, Hong Kong, China) primary antibodies (Table II) at 4°C overnight. After three washes, the membranes were incubated with the corresponding horseradish peroxidase-conjugated rabbit anti-mouse or goat anti-rabbit secondary antibodies (diluted 1:8,000 and 1:9,000, respectively) for 1 h. The membranes were then subjected to chemiluminescence (Pierce Chemical Co.) detection and fluorography using X-ray film. After signal detection, the membranes were re-incubated with an anti-β-actin antibody (Sigma-Aldrich, St. Louis, MO, USA) as a loading control. The experiment was performed three times.

Cells were fixed by 4% paraformaldehyde for 30 min and then washed with dH₂O. They were then oxidized in 0.5% periodic acid for 15 min at room temperature, rinsed in dH₂O for 1 min and treated with Schiff’s reagent (Sigma-Aldrich) for 30 min at room temperature. After rinsing in dH₂O for 2 min, counterstaining was performed with haematoxylin for 2 min, rinsed in running tap water and observed under an inverted microscope.

Medium levels of albumin and cytochrome P450 were determined using an albumin enzyme-linked immunosorbent assay system and cytochrome P450 enzyme-linked immunosorbent assay system (both from Blue Gene, China) according to the instructions supplied by the manufacturer.

Full-length CTGF was amplified from the cDNA of WB-F344 cells using the forward, 5′-atagatctgaccatgctcgcctccgtcg-3′ (BglII site italicized) and the reverse primer, 5′-atctcgagctttacgccatgtctcc ata-3′ (XhoI site italicized). This fragment was inserted into the BglII and XhoI sites of the vector, dl6-95/neo to yield the recombinant plasmid, dl6-95/CTGF/neo. The recombinant plasmids were amplified in competitive cells, purified using a Wizard^® Plus SV Miniprep DNA Purification System (Promega) and verified by DNA sequencing.

Prior to the transfection experiments, monolayers were trypsinized and the cells were plated 1×10⁵ cells/well in a 6-well format. After overnight culture, cells were 60–70% confluent. The cells were divided into three groups: a WB-F344 cell control group; a dl6-95/neo group and a dl6-95/CTGF/neo group, and transfected with plasmid DNA using FuGene^® HD transfection reagent (9 μl transfection reagent to 3 g DNA; Roche, Mannheim, Germany) according to the manufacturer’s protocol. The medium was replaced with medium containing 10% FBS after 8 h of transfection. Forty-eight hours later, cells were trypsinized, plated in 2 x 100-mm dishes, and cultured overnight. The medium was then replaced with fresh complete medium containing G418 (100 μg/ml) for the stable transfection experiments. The medium was replaced every 3–5 days. Cells were incubated for ~20 days, during which the G418 killed all the control cells, and the dl6-95/neo and dl6-95/CTGF/neo groups developed stable clones. The medium was then replaced with medium containing 50 μg/ml G418 for continued culture until total mRNA or proteins were extracted. The expression of CTGF mRNA was detected by reverse transcription-polymerase chain reaction. The sequences of the primers used for PCR are shown in Table III. The PCR conditions were: 30 cycles at 94°C for 40 sec, annealing for 40 sec (temperature shown in Table I) and 72°C for 60 sec. Hepatic stellate cells were used as a control. PCR products were analyzed by electrophoresis on 1% agarose gels.

Data are expressed as means ± SD from three independent experiments. Differences between mean values of multiple groups were analyzed using the nonparametric ANOVA test (SPSS, Inc., Chicago, IL, USA). Comparison between 2 groups was made using the Student’s t-test. P<0.05 was indicative of a statistically significant result.

WB-F344 cells were treated with different doses of rCTGF (0.5, 1, 2, 4, 8 or 16 ng/ml), resulting in cell viabilities of 99, 99, 95, 97, 95 and 91%, respectively, after 24 h of incubation. The percentage of viable cells gradually decreased to 91, 91, 88, 83, 73 and 56%, respectively, after 48 h of incubation (Fig. 1A). To determine the dose which had the lowest effect on progenitor proliferation, we selected doses of 1 and 5 ng/ml CTGF and a treatment time of 24 h.

Cultured WB-F344 cells were small and polygonal with a high nuclear/cytoplasmic ratio. After treatment with CTGF (5 ng/ml) for 24 h, the cells were enlarged and the nuclear/cytoplasmic ratio was decreased (Fig. 1B).

The WB-F344 cell line is an in vitro model of hepatic progenitor cells. These progenitors express both AFP and ALB, which are hepatocyte markers, highly expressed either in the liver carcinoma cell line HepG2 or in primary isolated rat hepatocytes. WB-F344 cells also express CK-19, which is a cholangiocyte marker, also highly expressed by HepG2 cells (Fig. 2A).

To study whether CTGF alters the expression of the hepatic progenitor cell markers to induce differentiation, we observed changes in their expression levels after CTGF treatment. Treatment with CTGF at 5 ng/ml in serum-free medium for 24 h resulted in a 0.2-fold decrease in the AFP mRNA level, a 1.6-fold increase in the ALB mRNA level and a 0.6-fold decrease in the CK-19 mRNA level (in all P<0.05) (Fig. 2B). This was accompanied by a 0.6-fold decrease in AFP protein levels, a 3.5-fold increase in the ALB protein level, and a 1.6-fold increase in the CK-19 protein level, respectively (in all P<0.05) (Fig. 2C). CTGF at 1 ng/ml caused only minor changes in the expression levels of the hepatic progenitor markers at the protein level. These results indicate that recombinant CTGF induces WB-F344 cell differentiation into hepatocytes rather than cholangiocytes.

Finally, to confirm whether the differentiated cells exhibit the functional properties of hepatocytes, we examined the specific function of hepatocytes, including glycogen storage, albumin production and cytochrome P450 production. WB-F344 cells stimulated with CTGF gained the capability of glycogen storage (Fig. 3A), which was absent in cells without CTGF. These results were in correlation with the albumin (Fig. 3B) and cytochrome P450 (Fig. 3C) production.

To further investigate the role of CTGF in the differentiation of hepatic progenitor cells, we focused on whether a chemical inhibitor that blocks CTGF expression suppresses the differentiation of WB-F344 cells. A chemical inhibitor of CTGF (WP631) was used.

We first determined the optimal concentration of WP631 by an MTT assay. Treatment with WP631 at ≥0.5 nM for 24 or 48 h reduced cell viability. Thus, we selected 0.1 nM WP631 for the subsequent experiments, which resulted in cell viability of 90% after 24 h (Fig. 4A), without significantly affecting WB-F344 cell morphology (Fig. 4B).

WB-F344 cells were incubated with the inhibitor for 24 h (preliminary time-response experiments indicated that this afforded maximal inhibition of CTGF expression; data not shown). Incubation with 0.1 nM WP631 in serum-free medium decreased CTGF expression 0.4-fold (P<0.05) (Fig. 4C).

Finally, we examined whether inhibition of CTGF affects WB-F344 cell differentiation towards hepatocytes. Treatment with 0.1 nM WP631 decreased ABL expression 0.6-fold (P<0.05), but increased AFP expression 2.1-fold (P<0.05) (Fig. 4D).

We next examined whether endogenous CTGF also induces the differentiation of WB-F344 cells. We isolated CTGF cDNA from WB-F344 cells and constructed a plasmid carrying the full-length CTGF gene (details in ‘Materials and methods’) (Fig. 5A). WB-F344 cells were transfected with the CTGF plasmid (dl6-95/CTGF/neo) or a blank plasmid (dl6-95/neo) using the FuGene HD reagent (Fig. 5D). Following the selection of clones in medium containing G418, the expression of CTGF by stably transfected WB-F344 cells was analyzed by PCR and western blotting. The expression of CTGF mRNA and protein was increased in WB-F344 cells transfected with the CTGF plasmid when compared with that in the control (Fig. 5B and C).

After stable transfection, endogenous CTGF also altered the phenotype of hepatic progenitor cells. The expression of AFP decreased in the dl6-95/CTGF/neo group (0.7-fold; P<0.05) when compared with the dl6-95/neo group. In contrast, expression of the mature hepatocyte maker, ALB, increased (1.4-fold compared with the blank plasmid; P<0.05). CK-19 expression did not change significantly (P>0.05) (Fig. 5E). Endogenous CTGF stimulated WB-F344 cell differentiation to a lesser degree than recombinant CTGF.