Effect of bone marrow-derived mesenchymal stem cells on hepatic fibrosis in a thioacetamide-induced cirrhotic rat model

Human BM-MSCs were obtained from healthy persons who voluntarily donated their bone marrow stem cells. Approximately 10-20 mL of BM was aspirated from the posterior iliac crests of humans under local anesthesia. BM mononuclear cells were isolated through density-gradient centrifugation (Histopaque-1077, Sigma-Aldrich, St. Louis, MO, USA). Mononuclear cells (2-3 × 10⁵ cells/cm²) were plated in a 75-cm² flask (Falcon, Franklin Lakes, NJ, USA) with Dulbecco's modified Eagle's medium (DMEM; Gibco, Grand Island, NY, USA) containing 10% fetal bovine serum (FBS; Gibco) and 1% penicillin/streptomycin (Gibco) and then cultured at 37°C in a 5% CO₂ atmosphere. When the cultures approached 80% confluence, the cells were harvested by treatment with a trypsin/EDTA solution (Gibco) and replated at a density of 4-5 × 10³ cells/cm² in 175-cm² flasks. Cells for injection were serially subcultured to passage four or five. The human procedures and protocols were approved by the Institutional Review Board (IRB) at Yonsei University Wonju Severance Hospital (CR109021), and conducted according to the principles of the Declaration of Helsinki. All participants provided written informed consent prior to participation in the study.

The immunophenotypes of the BM-MSCs (CD14, CD34, CD45, CD73, and CD105) were analyzed on the day of injection, and their differentiation potentials were determined (osteogenesis and adipogenesis; Figure 1). For the immunophenotype analysis, BM-MSCs were stained with antibodies conjugated with fluorescein isothiocyanate (FITC) or phycoerythrin (PE): CD14-FITC, CD34-FITC, CD45-FITC, CD73-PE, and CD105-PE (BD Biosciences, San Jose, CA, USA). Briefly, 5 × 10⁵ cells were resuspended in 0.2 mL of phosphate-buffered saline (PBS) and incubated with FITC- or PE-conjugated antibodies for 20 min at room temperature. FITC- or PE-conjugated mouse IgGs were used as the control isotype at the same concentration as the specific primary antibodies. The fluorescence intensity of the cells was evaluated through flow cytometry (Epics XL; Beckman Coulter, Miami, FL, USA). Osteogenic differentiation was determined by first plating the cells at 2 ± 10⁴ cells/cm² in six-well plates and then incubating them in the following osteogenic medium for 2-3 weeks: DMEM medium supplemented with 10% FBS, 10 mM β-glycerophosphate, 10^-7 M dexamethasone, and 0.2 mM ascorbic acid (Sigma-Aldrich) [33]. The osteogenic differentiation was quantified from the release of p-nitrophenol from p-nitrophenyl phosphate by the enzyme alkaline phosphatase [34]. For adipogenic differentiation, BM-MSCs were plated at 2 × 10⁴ cells/cm² in six-well plates and cultured for 1 week; then, differentiation was induced with adipogenic medium (10% FBS, 1 μM dexamethasone, 0.5 mM 3-isobutyl-1-methylxanthine, 10 μg/mL insulin, and 100 μM indomethacin in high-glucose DMEM) for an additional 3 weeks. The differentiated cells were fixed in 4% paraformaldehyde for 10 min and stained with fresh Oil Red-O solution (Sigma-Aldrich) to display lipid droplets (Figure 1B). The criteria regarding the use of MSCs included viability greater than 80%; the absence of microbial contamination (bacteria, fungus, virus, or mycoplasma) when tested 3-4 days before administration; CD73 and CD105 expression in more than 90% of the cells; and the absence of CD14, CD34, and CD45 in less than 3% of the cells, as assessed by flow cytometry.

Seven-week-old male Sprague–Dawley rats (Orient Bio Inc.) were maintained at a room temperature of 25°C with a 12/12-h light/dark cycle and with free access to food and water throughout the 12-week experiment. Hepatic fibrosis was induced in Sprague–Dawley rats by intraperitoneal (i.p.) injections of TAA (Sigma, St. Louis, MO, USA; 300 mg/kg body weight) twice a week for 12 weeks. The animals were randomly allocated into three groups (each group, n =18) as follows: Group I (G1, sham group); Group II (G2, untreated cirrhotic group), which received the TAA injection; and Group III (G3, BM-MSC treated group), which received the TAA injection and the BM-MSC treatment. The rats were anesthetized by intramuscular administration with a mixture of Zoletil® (Virbac Laboratories, Carros, France) and Rompun® (Bayer Korea, Seoul, Korea). With aseptic techniques, a 1-cm incision was made caudal to the costal arch on the right flank to expose the right lobe of the liver. With a syringe, 2 ± 10⁶ human BM-MSCs were injected directly into the right lobe of the liver at 6 and 8 weeks during the 12-week TAA administration in the BM-MSC treated group (Figure 2). At 12 weeks, the animals were sacrificed after taking blood samples, and liver tissue specimens were collected, fixed, and immediately snap-frozen and stored at -80°C for analysis. The animal experimental procedures and protocols were approved by the Institutional Animal Care and Use Committee (IACUC) at Yonsei University Wonju College of Medicine (YWC-131008-1).

Five-micrometer-thick sections of paraffin-embedded liver tissues were prepared and stained with hematoxylin and eosin (H&E), Masson's trichrome (MTC), α-smooth muscle actin (α-SMA), and Picrosirius Red. Fibrosis was evaluated using the Laennec fibrosis scoring system (Table 1). In the Laennec system, the thickness of the predominant type of septae in each specimen is chosen, and the smallest nodule is selected for scoring. A liver pathologist who was blinded to the clinical data evaluated fibrosis simultaneously and prospectively. The Laennec fibrosis scoring system was used because this system divides cirrhosis into three subclasses, allowing for a more detailed estimation of fibrosis after intervention [35]. In addition, to estimate any treatment-induced changes in liver fibrosis, the fibrosis area was quantified as a percentage of the total area that was positive for MTC stain in the digital photomicrographs using a computerized image analysis system (Analysis 3.0, Soft Imaging System, Münster, Germany). To quantify the fibrosis area, microscopic areas were selected randomly at an original magnification of 100X. For immunohistochemical analysis, the tissue sections were incubated with primary antibody against α-SMA (diluted 1:800; Neomarkers, Fremont, CA, USA) for 90 min at room temperature after washing with buffer. The tissue sections were incubated with the chromogen 3-amino-9-ethylcarbazole (BioGenex, San Ramon, CA, USA) for 5-7 min. Prior to mounting, the sections were counterstained with hematoxylin and then dehydrated. An UltraVision LP Large Volume Detection System (Lab Vision, Runcorn, UK) was used as the detection system. Morphological analysis of immunopositive cells was also performed using a computerized image analysis system.

Picrosirius Red staining was performed to quantify the total amount of collagen. Five-micrometer-thick sections of paraffin-embedded liver tissues were deparaffinized and rehydrated with distilled water and stained with a Picrosirius Red staining kit (Polysciences, Warrington, PA, USA) according to the manufacturer's instructions. In addition, the amount of collagen (the main component of fibrous tissue) was estimated from the collagen proportionate area and expressed as the percentage of the total area that was positive for Picrosirius Red stain on microscopy (Olympus BX51, Tokyo, Japan) using a computerized image analysis program (IMT i solution, Vancouver, Canada). While measuring the collagen proportionate area, we eliminated image artifacts and structural collagen in large portal tracts and blood vessel walls [36].

The expression of hepatocyte specific marker (Albumin) was analyzed at 7 and 14 days after 2 × 10⁶ human BM-MSCs were injected directly into the right lobe of the liver in TAA-induced hepatic fibrosis rats. Five-micrometer-thick sections of liver tissues were incubated with human anti-albumin antibody (A6684; Sigma) at 4°C overnight. The slides were washed and incubated with Alexa Fluor® 488 (Invitrogen, Carlsbad, CA) secondary antibody for 1 hr in dark. The nuclei were stained with Hoechst. Fluorescence images were showed under a laser scanning confocal microscope (TCS SPE, Leica Microsystems GmbH, Wetzlar, Germany). To identify injected BM-MSCs in the hepatic fibrosis rat liver, BM-MSCs were labeled with fluorescent spherical silica nanoparticles with CELL-STALKER™-CSR (Biterials) that contained Rhodamine B isothiocyanate (RITC) according to manufacturer's protocol. CELL-STALKER was centrifuged at 12,000 rpm for 10 min then supernatant was discarded except pellet and sonicated for 5 min at 40 khz-300 w to evenly distribute particles, placed into BM-MSCs culturing 75 T flask and incubated at 37°C for 24 hrs. After incubation, the cells were harvested by treatment with a trypsin/EDTA solution. The 2 ± 10⁶ BM-MSCs labeled with CELL-STALKER were injected directly into the right lobe of the liver in TAA-induced hepatic fibrosis rats and fluorescence images for BM-MSCs were showed under a laser scanning confocal microscope at 0, 3, 7, 14 days.

Measurement of alanine transaminase (ALT), aspartate transaminase (AST), total bilirubin, and albumin levels were carried out using commercially available kits (Asan Pharmaceutical, Republic of Korea) according to manufacturer's instructions.

Total RNA was isolated from liver tissues using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's protocol. RNA purity and concentration were determined using a spectrophotometer (Ultrospec 2100 pro UV/Visible, Amersham Bioscience, Freiburg, Germany). cDNA was synthesized from total RNA (1 μg) using the GeneAmp RNA PCR Kit (Applied Biosystems, Foster City, CA, USA) with oligo-dT (Applied Biosystems). For the real-time polymerase chain reaction (PCR), amplification was performed to measure the mRNA levels of TGF-β1, type 1 collagen (collagen-1), and α-SMA using sequence-specific primers (Table 2). Quantitative real-time PCR using SYBR Green PCR Master Mix (Applied Biosystems) was performed in an ABI PRISM 7900HT Sequence Detection System (Applied Biosystems) according to the manufacturer's instructions. The data were analyzed using SDS 2.2.2 software (Applied Biosystems). The cycle threshold (Ct) values of the target genes were normalized to the Ct values of the endogenous control (glyceraldehyde-3-phosphate dehydrogenase). Relative changes were calculated using the equation 2^–ΔΔCt.

Liver tissues were hydrolyzed with 6 N HCl at 120°C for 16 hrs. The hydrolysate was then cooled, neutralized with 6 N NaOH, and centrifuged at 13,000 g for 10 min. The supernatants were supplemented with 7% chloramine T, acetate/citrate buffer (sodium acetate·3H₂O, trisodium citrate·2H₂O, citric acid, with isopropanol). Ehrlich's solution (dimethylaminobenzaldehyde with perchloric acid and isopropanol) was then added and incubated at 60°C for 35min. After cooling, absorbance was measured at 560 nm with an Emax Precision Microplate Reader (Molecular Devices, Sunnyvale, CA, USA). Hydroxyproline concentration was calculated from a standard curve prepared with hydroxyproline (Sigma, H5534). The results were expressed as micrograms of hydroxyproline per gram of liver tissue.

For total protein extracted from liver tissues, tissues were homogenized using a TissueLyser II (QIAGEN GmbH, Haan, Germany) with a tissue protein extraction reagent (T-PER; Pierce, Rockford, IL, USA). The lysates were centrifuged at 13,000 rpm for 15 min at 4°C, and the protein concentrations of the supernatants were determined using a protein assay kit (Bio-Rad Laboratories Inc., Hercules, CA). Thirty micrograms of each liver protein was electrophoresed on a 10% sodium dodecyl sulfate-polyacrylamide (SDS-PAGE) gel and then transferred to polyvinylidene difluoride (PVDF) membranes (Millipore, Bedford, MA, USA). The membranes were blocked with 5% skim milk in Tris-Buffered Saline (TBS) containing 0.1% Tween-20 for 1 hr at room temperature, and the membranes were then incubated with primary antibodies at 4°C overnight. The primary antibodies were as follows: TGF-ß1 (Abcam, Cambridge, MA, USA), α-SMA (Abcam), Smad3 (Cell Signaling Technology, Danvers, MA, USA), and phospho Smad3 (Cell Signaling Technology). Horseradish peroxidase (HRP)-conjugated secondary antibodies against either mouse IgG (Abcam) or rabbit IgG (Cell Signaling Technology) were incubated for 1 hr at room temperature. Specific protein bands on the membranes were developed using an enhanced chemiluminescence (ECL) detection kit (Amersham, Piscataway, NJ, USA). To normalize between experiments, the membranes were probed with a β-actin antibody (Abcam), and the intensity of each protein band was normalized to the intensity of β-actin.

The values are expressed as means ± standard deviations. Nonparametric analysis was performed with the Kruskal-Wallis H test. Statistical analysis was performed using SPSS software version 20.0 (SPSS, Chicago, IL, USA). For all tests, P values <0.01 were considered significant.

The immunophenotypes for CD14, CD34, CD45, CD73, and CD105 cells were determined, and osteogenic or adipogenic differentiation was induced on the day of BM-MSC injection (Figure 1). In both the first and second injected cell populations, CD73 or CD105 (which are positive markers of BM-MSCs) were expressed in more than 98% of the cells. However, CD14, CD34, and CD45 (which are known to be negative markers of BM-MSCs) were expressed in less than 1% of the cells (Figure 1). Therefore, the BM-MSCs were differentiated into osteocytes and adipocytes (Figure 1B).

The Laennec fibrosis scoring system revealed detailed individual changes within the cirrhotic tissue (F4A–F4C; Table 1). Histological analysis was evaluated by H&E and MTC staining (Figure 3). In the untreated cirrhotic group, the liver sections were strongly stained by H&E and MTC, showing definite cirrhosis (stage 4 fibrosis) with regenerating nodules and fibrous septae (Figure 3A), whereas the BM-MSC treated group showed mild fibrosis (Figure 3B). To compare these findings, the degree of fibrosis was scored according to the Laennec fibrosis scoring system. The BM-MSC treated group had a significantly lower mean score than that of the untreated cirrhotic group. According to the Laennec fibrosis system, histologically, the BM-MSC treated group showed significant improvement in hepatic fibrosis compared to the untreated cirrhotic group (P <0.01). These results were further confirmed by immunohistochemical staining revealing α-SMA expression and Picrosirius Red stain (Figure 3). The relative expression of the collagen proportionate area stained by Picrosirius Red was analyzed using an image analysis program. The percentage of the collagen proportionate area significantly decreased from 16.72 ± 5.51 to 5.06 ± 1.27 after BM-MSC treatment (P <0.01; Figure 3G). In addition, the analysis of histopathological fibrosis scoring confirmed that cirrhosis was significantly reduced by BM-MSCs treatment, compared to the untreated cirrhotic group (Table 3).

To investigated whether BM-MSCs are capable of undergoing hepatic differentiation, the expression of albumin was detected by immunofluorescence staining. At 7 days after BM-MSCs injection, expression of albumin was weakly detected. However, intensity of albumin staining was stronger at 14 days (Figure 4).

To identify injected BM-MSCs to the hepatic fibrosis rat liver, fluorescence red labeled BM-MSCs were showed at 0, 3, 7, 14 days after direct injection in hepatic fibrosis rat liver. Fluorescence intensity from labeled BM-MSCs was maintained for 14 days in liver tissue. However fluorescence intensity from labeled BM-MSCs was gradually reduced and disappeared at 14 days (Additional file 1: Figure S1).

Differentiation of BM-MSCs into hepatocytes was determined by immunofluorescence staining using albumin antibody and identification of BM-MSCs in hepatic fibrosis was determined by fluorescent microscopy using CELL STALKER-CSR dye staining. Consequently, we showed that fluorescence intensity from labeled BM-MSCs with CELL-STALKER was gradually disappeared and albumin-stained areas had gradually expanded at 14 days after injection, considering that BM-MSCs had transdifferentiated into hepatocytes after injection. Based on these results, it may suggested that BM-MSCs had transdifferentiated into hepatocytes after injecting hepatic fibrosis rats.

The serum levels of ALT, AST, and total bilirubin were significantly decreased in BM-MSCs treated group. In addition, serum level of albumin in the BM-MSC treated group were increased compared to the untreated cirrhotic group (P <0.01; Figure 5).

After BM-MSC administration, the mRNA relative expression levels of TGF-β1, collagen-1, and α-SMA were determined through quantitative real-time reverse transcription-PCR (RT-PCR) analysis. The collagen-1 mRNA relative expression levels were 1 ± 0.39, 14.04 ± 4.54, and 4.74 ± 2.46 (P <0.01) in G1, G2, and G3, respectively (Figure 6B). In addition, the TGF-β1 and α-SMA mRNA relative expression levels were 1 ± 0.27, 4.61 ± 0.61, and 2.14 ± 0.35 (P <0.01) and 1 ± 0.32, 8.83 ± 3.12, and 2.62 ± 0.41 (P <0.01) in G1, G2, and G3, respectively (Figure 6A, 6C). The TGF-β1, collagen-1, and α-SMA mRNA relative expression levels in the BM-MSC treated group were significantly decreased compared to the untreated cirrhotic group (P <0.01).

The content of hepatic hydroxyproline was quantified colorimetrically from liver tissues. Quantitative analysis showed that the contents of hepatic hydroxyproline of the liver tissue were 16.44 ± 4.07, 85.81 ± 17.62, and 46.25 ± 13.19 (P <0.01) in G1, G2, and G3, respectively (Figure 6D). The content of hepatic hydroxyproline was significantly lower in the BM-MSC treated group than in the untreated cirrhotic group.

The TGF-β1 and α-SMA protein expression levels were measured through western blot assays (Figure 7). The TGF-β1 and α-SMA protein expression levels were 1 ± 0.15, 2.5 ± 0.39, and 1.41 ± 0.41 (P <0.01) and 1 ± 0.14, 1.8 ± 0.49, and 1.16 ± 0.25 (P <0.01) in G1, G2, and G3, respectively (Figure 7). The TGF-β1 and α-SMA protein expression levels in the BM-MSC treated group were significantly decreased compared to the untreated cirrhotic group (P <0.01).

We examined the effect of BM-MSCs on the status of Smad3 and found that Smad3 phosphorylation was markedly increased in the untreated cirrhotic group but was significantly decreased after BM-MSC treatment. These results demonstrate that BM-MSCs modulated the TGF-β1/Smad signaling pathway by attenuating TGF-β1 and Smad3 expression and inhibiting Smad3 phosphorylation.