The role of multipotent cancer associated fibroblasts in hepatocarcinogenesis

For human samples, written informed consent was obtained from patient or by a legal representative and patient anonymity has been preserved. Investigation was conducted according to the principles expressed in the Declaration of Helsinki. For animal study, the experimental procedure study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals and all efforts were made to minimize suffering. The Ethical Committee Ateneo of the University of Trieste and responsible administration of the Ministry of Health of the Republic of Italy approved the protocol (Permit number: 107/2010).

Paired fresh HCC and distant cirrhotic liver tissues were obtained from nine HCC patients undergoing partial hepatic resection. The ratio of female:male was 5:4, mean age 73 ± 7 years; three were hepatitis C virus (HCV) positive and six were metabolic-related HCC. None of the patients received previous liver surgery, radiofrequency, and trans-arterial chemoembolization.

Tissues were finely minced with scalpel in a tissue culture dish and enzymatically dissociated in 1 mg/mL collagenase type IV (Sigma-Aldrich, St Louis, MO, USA) at 37°C for 1 hour with frequent shaking. The activity of collagenase was blocked using phosphate saline buffer (PBS) supplemented with 10% fetal bovine serum (FBS). Single cells suspension was washed and filtered through a 40 μm cell strainer (BD Biosciences, Milan, Italy). Cells were plated on a 100 mm dish in MyeloCult® medium (StemCell Technologies, Vancouver, BC, Canada) in the presence of 1 μM hydrocortisone sodium succinate and 1% antibiotics. They were maintained in a controlled CO₂ incubator with 37°C, 95% humidity, 5% CO₂ with medium change every three days and sub-cultured with 0.05% trypsin in PBS when they reached 80% - 90% of confluence. Morphological homogeneity of the cells were noticed along subcultures. Cells from passages 2–6 were used for all experiments. The primary cells from HCC nodules were identified as CAF while cells from distant non-tumoral tissues as NTF (non-tumoral fibroblast).

Human HCC cell lines HuH-7 (JCRB0403) and JHH-6 (JCRB1030) were obtained from the Japan Health Science Research Resources Bank (HSRRB, Tokyo, Japan). HuH-7 cells were grown in DMEM medium (high glucose) and JHH-6 in Williams’ E medium. Both media were supplemented with 10% FBS, 1% L-glutamine, and 1% antibiotics. All cells were maintained at 37°C in a humidified 5% CO₂ incubator and were routinely sub-cultured with 0.05% trypsin in PBS when they reached 85% - 95% confluence.

The presence of surface marker antigens was detected using antibodies CD90/THY1 (Stem Cell Technologies, Vancouver, BC, Canada), CD133/PROM1, CD45 (Miltenyi Biotec GmbH, Bergisch Gladbach, Germany), CD44 (Abcam, Cambridge, UK), and STRO-1 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA). After detachment, at least two million cells per mL were incubated with specific first antibodies for 60 minutes on ice in the dark. After two washings with PBS containing 0.5% bovine serum albumin (BSA), when necessary, the cells were incubated with fluorescence-conjugated secondary antibody for 60 minutes on ice in the dark. Flow cytometric analysis was performed immediately in a FACSCalibur flow cytometer (Becton Dickinson, NJ, USA). Ten thousands events were analysed per sample.

The presence and localization of the proteins CD90, CD44, Vimentin/Vim (Abcam), and alpha smooth muscle actin/αSMA (Dako, Glostrup, Denmark) were visualized with immunofluorescence using fluorescence microscope Leica DM2000 (Leica Camera AG, Solms, Germany). The nucleus of the cells was stained with Hoechst 33342 dye (Sigma-Aldrich).

Total RNA was extracted using TriReagent (Sigma–Aldrich) according to the manufacture’s protocol. RNA was quantified at wavelength 260 nm in a DU®730 spectrophotometer (Beckman Coulter, Fullertone, CA, USA) and RNA purity was evaluated according the MIQE guidelines [7] by measuring the ratio A260/A280 with an appropriate purity value between 1.8 and 2.0. The integrity of RNA was assessed on standard 1% agarose/formaldehyde gel. Reverse transcription of 1 μg of total RNA was done using iScript cDNA synthesis Kit (Bio-Rad Laboratories, Hercules, CA, USA).

RTqPCR was performed according to the iQ SYBR Green Supermix protocol. PCR amplification was carried out in a 15 μL reaction volume containing 25 ng of cDNA, 1x iQ SYBR Green Supermix, and 250 nM gene specific sense and anti-sense primers and reaction was run on a Bio-Rad iQ5 real-time PCR detection system (IQ5 software version 3.1; Bio-Rad Laboratories), together with reference genes RNA18S and β-actin (ACTB). Cycling parameters were determined and analyzed using the Pfaffl modification of ΔΔCt equation with taking account to the efficiency of the reaction [8,9]. The primers for PCR were designed using software Beacon Designer Version 7.9 (Premier Biosoft International, Palo Alto, CA, USA). Primer sets were built across two exons to avoid the contamination of genomic DNA. Nucleotide BLAST was performed to check the specificity of the sequences. Melting curve analysis and agarose gel electrophoresis were carried out to assess the size of the template products. The list of the primers sequences was listed in Supplemental File (Additional file 1: Table S1). Control for of primers: total RNA extract from IHH cells for CD90; HuH-7 cells for CD133 [10,11]; Jurkat cells for CD34; and HCC tissues and blood samples for CD44, CD45, CD29, CD31, CD105, CD166, CD11B, CD13, CD19, CD29, and CD79.

The clonogenic capacity of the CAF in a 3-dimensional anchorage-independent matrix was performed by growing the cells at low density of 5000 cells/mL in 67% Matrigel (BD Biosciences) in growth medium without serum on a 12-well cell culture plate. The clones were observed under light microscope every three days (Nikon Eclipse TS100, Nikon Instruments, Campi Bisenzio, Italy).

For adipocyte differentiation, cells were plated in AdipoDiff medium (Miltenyi Biotec GmbH, Bergisch Gladbach, Germany) for 3 weeks with medium change every 4 days. The accumulation of fat deposit was stained using Nile Red, a specific intercellular lipid staining, and the up-regulation of gene PPARG (peroxisome proliferator-activated receptor gamma) was quantified using RTqPCR. For osteoblast differentiation, cells were plated in OsteoDiff medium (Miltenyi Biotec) for 2 weeks. Alkaline phosphatase activity was detected using Sigma Fast BCIP/NBT substrate (Sigma-Aldrich) according to the manufacture’s protocol. The up-regulations of bone-specific genes were quantified using RTqPCR. The protocol for pancreatic cells differentiation was performed according previous reports in the presence of 10 mM nicotinamide [12,13]. The identification of cells was performed using RTqPCR on gene somatostatin (SST) and gastric inhibitory polypeptide (GIP). Direct DNA sequencing was carried out to confirm the result (automated sequencer ABI Prism 3500xl genetic analyser, Applied Biosystems, Monza, Italy).

A total 50,000 cells/mL of CAF were co-cultured together with 100,000 cells/ml of HuH-7 and JHH-6 cells for 7 days in a 6-well plate of Transwell system (Corning Costar, Milan, Italy). The alteration of tumor promoting factors was evaluated by way of RTqPCR. For comparison, co-culture between NTF and both HCC cell lines was also performed in the similar condition. The experiment was conducted in duplicates in two independent sets.

Male 7 weeks athymic nude (Foxn1(nu/nu)) homozygotes and NOD/SCID (NOD.CB17-Prkdcscid/NCrHsd) mice for in vivo xenotransplantation studies were obtained from Harlan Laboratories, Srl (Udine, Italy). All animals were maintained in the animal facility of the University of Trieste. After detachment, CAF was suspended in 100 μL cold PBS and placed in ice. A total of 50,000 to 1 million cells were injected subcutaneously into abdomen of the nude mouse in duplicates or orthotopic in the liver of NOD/SCID mice. Viability of the cells was checked by tryphan blue staining dye after injection. The xenotransplantated mice together with control were observed for four months after injection. Mouse body weight was measured every week. Serum alanin transferase (ALT) and aspartate transferase (AST) levels on sacrifice day were measured by photometric enzymatic test (Cobas, Roche, Mannheim, Germany).