A total of 20 female Wistar rats (weight, 200–220 g) were used in the present study. The rats were obtained from the Centre of Experimental Medicine, The Medical University of Bialystok (Bialystok, Poland). The rats were housed in pairs in standard polysulfone cages (59×38×20 cm) containing wood-chip bedding material in a temperature (22±1°C) and humidity-controlled (50±10%) room with a 12-h light/dark cycle. The animals had ad libitum access to standard rat pellet food and tap water. All procedures were approved by the Ethics Committee for Animal Experimentation at the University of Life Sciences (Lublin, Poland; approval number 81/2015).

Following a 1-week period of acclimatisation the 10-week-old rats were divided into two groups for OC induction: i) Non-neoplastic PH, PH was performed and PBS was administered in the drinking water (n=10); and ii) neoplastic PH/DEN, a two-step rat model of hepatocarcinogenesis was performed following the Solt-Farber protocol (n=10) (16). A PH (two-thirds) was performed to the rats in each group according to the Higgins and Anderson method via excision of left lateral (~38%) and right (~30% lobes as previously described (17). All surgical procedures were performed under ketamine and xylazine anaesthesia (90 mg/kg and 10 mg/kg, respectively) between 9.00 a.m. and 12.00 p.m. to minimise the diurnal effect of liver regeneration. At 7 days following PH, DEN (50 mg/l; Sigma Aldrich; Merck KGaA, Darmstadt, Germany) was added to the drinking water of the neoplastic group of rats and supplied ad libitum for a period of 10 weeks. The non-neoplastic group were administered PBS. Following this 10-week period, an exploratory laparotomy was performed in the PH/DEN rats and the development of tumours on the liver surface was confirmed.

After 10 weeks OCs were isolated from each group as described previously (4). OC cells were isolated during the exploratory laparotomy in the PH/DEN group and during a regular laparotomy in the PH group. Prior to the laparotomy, rats were anesthetised with intramuscular administration of ketamine (90 mg/kg) and xylazine (10 mg/kg). The liver was perfused in situ through the portal vein with a Krebs-Ringer buffer (Sigma Aldrich; Merck KGaA) made up of three parts as follows: i) Containing EGTA, ii) without Ca²⁺ and chelating agent and iii) with type IV collagenase. Following perfusion, the liver was transferred to Ham's F-12/Dulbecco's modified Eagle's medium (DMEM; 1:1 v/v; Sigma Aldrich; Merck KGaA), minced, filtered through a double gauze and digested for 1 h at 37°C in PBS containing 0.1% collagenase, 0.25% trypsin, 0.004% DNase I and 0.1% pronase E (all Sigma Aldrich; Merck KGaA), which selectively digested mature hepatocytes into undetectable debris. The mixture was subsequently decanted through a 70 µm-nylon mesh followed by a 40 µm-nylon mesh and then loaded on a discontinuous gradient of Percoll and centrifuged at 420 × g for 20 min at 4°C. Density gradient centrifugation is an effective method of separating OCs from the remaining liver cell fractions (12). In comparison with the other methods, including flow cytometry and immunomagnetic sorting, density gradient centrifugation in Percoll, followed by specific enzymatic digestion leads to the isolation of pure OCs with a high viability (18–19).

The collected cell suspension was washed with antibiotic supplemented (100 U/ml penicillin, 0.1 mg/ml streptomycin, 10 mg/ml gentamycin) Ham's F-12/DMEM medium containing 10% fetal calf serum (Sigma Aldrich; Merck KGaA) and plated at 250,000 cells/well in 1,000 µl Ham's F-12/DMEM (1:1 v/v) medium in a 24-well plate at 37°C with 5% CO₂. The viability of the cells was estimated using the Trypan-blue exclusion method and was between 75–85%. A total of 180 µl Trypan blue (Sigma Aldrich; Merck KGaA) was mixed with 20 µl cell suspension and 50 µl of the mixture was dropped on a Bürker counting chamber. The number of viable cells (with a clear cytoplasm) and dead cells (with a blue cytoplasm) was determined in 24 squares (0.25 mm²) of the Bürker chamber as previously described (20). Adhered OCs from the PH and PH/DEN groups were cultured in HAMS-12/DMEM (v/v) medium at 37°C with 5% CO₂. The medium and cells were analysed at 24, 48 and 72 h, and the first and second week of incubation.

Serum levels of alanine aminotransferase (ALT; cat. no. A11A01627), aspartate aminotransferase (AST; cat. no. A11A01629), γ-glutamyl transferase (GGT; cat. no. A11A01630) and alkaline phosphatase (AP; cat. no. A11A01626) were quantified using a kinetic methods with an ABX Pentra 400 (HORIBA Ltd., Kyoto, Japan) using commercially available tests from HORIBA Ltd. All procedures were performed according manufacture's protocol.

Liver tissues were fixed for 24 h at room temperature in 10% phosphate-buffered formalin, embedded in paraffin and sectioned at 3–5 µm using a rotary microtome (Leica SR-200; Leica Microsystems, Ltd., Milton Keynes, UK) prior to histological examination. The sections were stained at room temperature with haematoxylin (5 min) and eosin (5 min) for histological observation. The histological analyses were performed via light microscopy (Eclipse E-600; Nikon Corporation, Tokyo, Japan) at magnification, ×200. The hepatic injures were assessed according to the World Health Organisation Histological Classification of Tumours (21).

The OCs were cultured at 36°C for 2 weeks in Ham's F-12/DMEM mixture (1:1 v/v). On each day of culture the OCs were subjected to microscopic analysis of the morphology under a phase-contrast microscope (Olympus CK40; Olympus Corporation, Tokyo, Japan). The appearance of the cells, colony formation, spread of cell culture and the construction or lack of a monolayer was analysed.

At various time points (24, 48, 72 h and 1st and 2nd week) the OCs were harvested in a lysis buffer composed of 0.2% Nonidet-P40, 150 mmol/l NaCl, 20 mmol/l Tris, 2 mmol EDTA, 0.1% glycerol, 10 mmol/l dithiothreitol and a cOmplete^™ protease inhibitor cocktail (Roche; Sigma-Aldrich; Merck KGaA). Concentration of obtained protein was analysed by Lowry method (22). Equal amounts (30 µg/lane) of protein were separated by 12% SDS-PAGE, transferred to polyvinylidene difluoride membranes (Bio-Rad Laboratories, Inc., Hercules, CA, USA) and blocked for 1 h at 20°C with PBS containing 5% low-fat milk. Following washing with Tris-buffered saline and 0.1% Tween-20 (Sigma Aldrich; Merck KGaA), the membranes were incubated overnight at 4°C with primary antibodies directed against cytokeratine-19 (CK19; 1:200; cat. no. M0888), albumin (Alb; 1:1,000; cat. no. F0117; both Dako; Agilent Technologies, Inc., Santa Clara, CA, USA), α-fetoprotein (Afp; cat. no. ab46799), RIPK1 (cat. no. ab106393) and RIPK3 (cat. no. ab56164) (all 1:200; Abcam, Cambridge, UK). The washed membranes were subsequently incubated for 3 h at 4°C with alkaline phosphatase-labelled goat anti-mouse (cat. no. P0447) or anti-rabbit (cat. no. P0448) (both 1:15,000; Dako; Agilent Technologies, Inc.) secondary antibodies. The protein bands were visualised using colorimetric detection (GelDog XR system; Bio-Rad Laboratories, Inc.) and analysed using Quantity One 1-D analysis software version 4.6.3 (Bio-Rad Laboratories, Inc.) (4).

Apoptosis detection was performed using an Annexin V/FITC assay kit (Bio-Rad Laboratories, Inc.) according to the manufacture's protocol. The OCs were washed in PBS and centrifuged at 420 × g for 10 min at 4°C and the supernatant was discarded. The cell pellet was suspended in the binding buffer and adjusted to a concentration of 5×10⁵ cells/ml. A total of 5 µl Annexin V-fluorescein isothiocyanate (FITC) solution was added and the cells were incubated for 15 min at room temperature in the dark. A total of 10 µl propidium iodide (PI) was subsequently added and the probes were incubated for 15 min in the dark at room temperature. The suspension was subsequently centrifuged at 300 × g for 5 min at room temperature, re-suspended in 200 µl binding buffer and analysed using System XL-II software with a Coulter Epics XL flow cytometer (Beckman Coulter, Inc., Brea, CA, USA) (23).

To detect the proliferation rate of PH and PH/DEN OCs, an ELISA-Bromodeoxyuridine (BrdU) kit (Abcam; cat. no. ab126556) was used according to the manufacturer's protocol. This test measures the BrdU incorporated into newly synthesised cellular DNA. The cells were seeded in a 96-well plate at a density of 2×10⁵ cells/well in 100 µl Ham's F-12/DMEM (1:1 v/v) culture medium. A total of 20 µl diluted 1:500 BrdU was added to the OCs in the plates and incubated at 37°C for 4 h (BrdU incorporation). Following removal of the culture medium the cells were fixed for 30 min at room temperature with FixDenat (provided with the test kit) followed by incubation with peroxide-conjugated monoclonal antibodies to the thymidine-analogue 5-bromo-2deoxyuridine Fab fragments, which bind with the incorporated DNA. Following the addition of substrate solution, the immune complexes were detected at 450 nm using an ELx800^™ ELISA reader (BioTek Instruments, Inc., Winooski, VT, USA).

Superoxide anion production was measured as previously described (23). OCs were incubated with 0.1% nitroblue tetrazolium (Sigma-Aldrich; Merck KGaA) solution at room temperature for 10 min and the absorbance was read at 545 nm. Nanomoles of superoxide produced over the incubation period were calculated using the extinction coefficient of 21.1 nmol (24).

All experiments were repeated a minimum of three times with different batches of cell samples. Data are presented as the mean ± standard deviation. Data were analysed using one-way analysis of variance (ANOVA) or Student's t-test as appropriate. When ANOVA indicated significant differences within groups, comparisons were made using Tukey's post hoc test. P<0.05 was considered to indicate a statistically significant difference. Statistical analyses were performed using STATISTICA version 13.1 software (StatSoft Ltd., Bedford, UK).

Rats in the PH/DEN group had a slower growth rate compared with the PH group. At the time of sacrifice the average body weight was significantly (~20%) lower in the PH/DEN group compared with the PH group (Table I). Blood analyses indicated that AST, ALT, AP and GGT levels were significantly higher in the PH/DEN group compared with the PH group (Table I).

The livers of the PH (Fig. 1A) and PH/DEN (Fig. 1B) rats were examined. Normal liver histology was observed in the PH group (Fig. 1C). The liver histology of animals in the PH/DEN group was characterised by chronic damage and areas of cellular atypia (black arrow; Fig. 1D). The signs included lymphocytic infiltration, cells with enlarged nuclei and atypical hepatocytes.

Western blot analysis was performed to detect the stem cell markers Afp, CK19 and Alb to determine whether rat OCs were activated in each group (Fig. 2A). Each group was positive for Afp and CK19, the classical OC markers immediately following isolation. The isolated OCs also co-expressed Alb at a high level. One way of defining a cell population as a progenitor OC is to clarify its ability to multiply and differentiate into hepatocytes and cholangiocytes. The freshly isolated cells expressed hepatocytic (Alb) and cholangiocytic markers (CK19), indicating that they were bipotential hepatic stem cells. Afp and CK19 were not detected in PH OCs following 2 weeks of incubation (whereas Alb was detected at all time points of the experiment), which indicates that PH OCs successively gave rise to novel hepatocytes. By contrast, the expression of Alb by PH/DEN OCs gradually disappeared throughout incubation.

Freshly isolated OCs were round in shape and had a high nuclear to cytoplasmic ratio. PH OCs attached to the culture dish within 24 h and grew to confluence within 72 h as spindle-shaped cells (Fig. 2B). From the first week of incubation, the PH cells composed a flat monolayer and began to differentiate into cells of mature morphology, which was clearly visible in the second week of incubation (Fig. 2B). This maturation of cells was additionally confirmed by a strong Alb expression (Fig. 2A).

The PH/DEN group OCs expanded in the culture without creating a regular monolayer. It was observed that some of the cells were round mononuclear cells with an oval-like appearance, however some were longitudinal. Following 1 week of culture, the PH/DEN OCs aggregated and formed colonies. At 2 weeks incubation there was intensive multiplying of the OCs.

OCs isolated from PH and PH/DEN livers expressed RIP1 and RIP3 protein throughout culture (Fig. 3). The expression of RIP1 and RIP3 proteins by OCs remained constant in the PH group, however, a notable increase in RIP1 and RIP3 protein expression was observed in the PH/DEN group following 72 h incubation (Table II). In addition, the bands density of RIP1 and RIP3 obtained in the 1st and 2nd week of PH/DEN OCs incubation, were significantly higher (P<0.05) compared with the PH group at the same time points (Table II). At the 2nd week the PH/DEN OCs exhibited the strongest staining of RIP1 and RIP3 at 3.488,89 and 3.547,67 INT/mm², respectively (Table II). It appears that, conversely to non-neoplastic OCs, the time of incubation intensifies RIP1- and RIP3-dependent necroptotic death in neoplastic cells.

During the initial 48 h of OC incubation in the PH group, the mean number of apoptotic cells was 10.56±0.5% (Fig. 4). At 72 h the percentage of apoptotic cells significantly increased to 33.58±0.5%, which was markedly decreased at 2 weeks (30.08±0.83%). Compared with the PH OCs, the cells isolated from the PH/DEC group exhibited significantly less apoptosis between 72 h and 2 weeks. There were particularly few apoptotic cells (2.82±1.1%) in the PH/DEN group at 72 h. The level of apoptosis in the PH/DEN group gradually increased and was significantly increased at 2 weeks compared with 1 week, however at no time point did it exceed 15.94±0.9% of all cultured cells. Concurrent staining with Annexin V and PI revealed differences between the percentage of apoptotic PH and PH/DEN OCs. Whereas in PH-OCs apoptosis intensified concomitantly with the time of incubation, (bottom and upper right quadrants) live neoplastic OCs dominated independently of the time of cell culture (bottom left quadrant).

The proliferative activity of PH OCs gradually increased from 24 h (0.47±0.03) to the first week of incubation (1.1±0.3), however at the second week there was a slight reduction (0.98±0.1; Fig. 5). In contrast, PH/DEN OCs demonstrated increased proliferative activity at each time point, reaching the highest value at the second week (1.99±0.09). The proliferative activity of PH/DEN OCs was significantly increased compared with the PH OCs at the same time point between 72 h and the second week.

ROS production by PH OCs increased gradually from 0.70±0.03 to 2.50±0.37 nM/25×10⁴ cells from 24 h to the first week of incubation (Table III). PH/DEN OCs generated significantly increased amounts of ROS at each time point compared with the PH group, The highest ROS production occurred in the first week of cell incubation in each group and reached 2.50±0.37 nM/2.5×10⁴ cells in the PH group and 3.70±0.27 nM/2.5×10⁴ cells in the PH/DEN group. Despite the slight decrease in ROS production (3.2±0.37 nM/2.5×10⁴ cells) at the 2nd week in the PH/DEN group, this value was still significantly increased compared with the PH group.