Liver fibrosis results from an imbalance in the normal wound-healing response resulting in an abnormal continuation of fibrogenesis (connective tissue production and deposition).1 The most common causes of liver fibrosis are infection with the hepatitis B or C virus, autoimmune diseases, and metabolic disorders, but it can also be caused by alcohol abuse or drug toxicity.2 Cirrhosis is an advanced stage of liver fibrosis that is accompanied by distortion of the hepatic vasculature, and it often requires liver transplantation. However, the limited supply of donors necessitates the development of alternative therapies.3, 4
We and others5–10 have demonstrated that bone marrow (BM)–derived stem cells can engraft in injured livers and restore function, raising the possibility that stem cell–based therapy has therapeutic potential for liver disease. BM is a recognized source of progenitor cells of several cell types,11 including endothelial cells,12 epithelial cells,13 mesenchymal stem cells,14 hepatocytes,15 and cardiocytes.16 However, engrafting is a rare event and may not be sufficient to induce therapeutic effects alone.
Endothelial progenitor cells (EPCs) have high proliferative potential17 and can migrate to regions of the circulatory system with injured endothelia, including traumatic, degenerative, or ischemic injury sites, thus promoting the repair or formation of new vessels.18–20 BM-derived cells assumed to be EPCs and monocyte-lineage cells contributed to hepatic sinusoid reconstruction during liver regeneration after partial hepatectomy in mice.21 Circulating endothelial cells play a role in regulating liver regeneration, and treatment with vascular endothelial growth factor (VEGF) can mobilize endothelial cells to accelerate this process. The progression of liver fibrosis has been linked to injuries associated with hypoxia and neovascularization; thus, transplantation of these cells may promote fibrosis resolution and liver regeneration. These cells can repair tissue damage and are more homogeneous than BM.22 Therefore, in this study, we evaluated whether BM-derived EPC transplantation could reduce liver fibrosis after carbon tetrachloride (CCl4)–induced liver cirrhosis.
α-SMA, α-smooth muscle actin; Alb, albumin; ALT, alanine aminotransferase; AST, aspartate aminotransferase; BM, bone marrow; CCl4, carbon tetrachloride; COL III, collagen III; DAPI, 4′,6-diamidino-2-phenylindole; Dil-acLDL, acetylated low-density lipoprotein with fluorescent Dil dye; EPC, endothelial progenitor cell; FCS, fetal calf serum; FITC, fluorescein isothiocyanate; FL1-H, fluorescent intensity; Flk-1, fetal liver kinase 1; H&E, hematoxylin and eosin; HGF, hepatocyte growth factor; HSC, hepatic stellate cell; INR, international normalized ratio; M, molecular size marker; M199, Medium 199; PBS, phosphate-buffered saline; PCR, polymerase chain reaction; RT-PCR, reverse-transcription polymerase chain reaction; Sry, sex-determining region of the Y chromosome; TBIL, total bilirubin; TGF, transforming growth factor; UEA, Ulex europaeus agglutinin; VEGF, vascular endothelial growth factor.
MATERIALS AND METHODS
Inbred Sprague-Dawley rats (Academy of Military Medical Science, China), with an initial body weight of 150 ± 10 g, were housed in a standard animal laboratory. They were kept at 25°C with a 12-hour light/dark cycle and allowed standard chow and water ad libitum until the time of the study. The experiments were performed in accordance with the National Research Council's Guide for the Care and Use of Laboratory Animals.
EPC Isolation and Culture from Rats
Male Sprague-Dawley rats were anesthetized with ether. Their BM was harvested from the femurs and tibias with a syringe, collected into Medium 199 (M199; Clonetics, San Diego, CA), filtered (40 μm), and centrifuged at 1200 rpm for 10 minutes to pellet the cells. After suspension in phosphate-buffered saline (PBS), the BM cells were loaded onto 5 mL of Histopaque-1083 (Sigma Chemical Co., St. Louis, MO) and centrifuged at 1600 rpm for 30 minutes. Mononuclear cells were collected at the Histopaque-1083 interface. After washing three times, the cells were suspended in Medium 199 supplemented with 20% fetal calf serum (FCS; PAA Laboratories, Austria), penicillin (100 U/ml), streptomycin (100 μg/ml), and VEGF (10 ng/mL; R&D Systems, Minneapolis, MI), and then seeded onto 6-well tissue culture plates (2 × 107 cells per well) precoated with rat-derived fibronectin (10 μg/mL; Sigma Chemical Co., St. Louis, MO). After 48 hours, the nonadherent cells were removed via washing with PBS, and new medium was added. On the ninth day, the adherent cells were harvested by trypsinization, washed, and resuspended at 5 × 105 cells/mL in saline for transplantation.
Characterization and Counting of Rat EPCs
Cells were first characterized by the presence of the cobblestone morphology that is typical of confluent endothelial cells under phase-contrast microscopy. Direct fluorescent staining was then used to detect the binding of fluorescein isothiocyanate (FITC)–Ulex europaeus agglutinin I (UEA-I; Vector Laboratories, Burlingame, CA) and the incorporation of acetylated low-density lipoprotein tagged with the fluorescent dye Dil (Molecular Probes, Eugene, OR). Cells showing double positive fluorescence were identified as differentiating EPCs. To further characterize the EPCs, the following antibodies were used for immunofluorescence and flow cytometry analysis: mouse monoclonal anti-CD34, goat polyclonal anti-CD133 (Santa Cruz Biotechnology, Santa Cruz, CA), rabbit polyclonal anti–fetal liver kinase 1 (anti–Flk-1), and mouse monoclonal anti-CD31 (Chemicon, Temecula, CA). The samples were evaluated with a Zeiss confocal fluorescence microscope with epifluorescence and a BD FACSCalibur flow cytometer. Two or three independent investigators evaluated the EPC densities by counting 15 randomly selected high-power fields (×200).
For the analysis of capillary tube formation, 200 μL of Matrigel (Becton Dickinson, Heidelberg, Germany) was added to the wells of a 48-well plate and allowed to solidify at 37°C for 30 minutes, after which 3 × 104 EPCs were added in 200 μL of M199 supplemented with 10% FCS or 50 ng/mL VEGF. Capillary tube formation on Matrigel was observed under an Olympus microscope after 5 or 18 hours of incubation.
Cell Labeling with PKH26
The cells were harvested and labeled with PKH26 (Sigma Chemical) according to the manufacturer's instructions. In brief, after the removal of M199, 1 mL of diluent C (PKH26 fluorescent cell linker kit, Sigma Chemical) was added to the cells. The cells were then suspended and immediately mixed with an equal volume of a 2× PKH26 stock solution in diluent C and further incubated at room temperature. After 5 minutes, an equal volume of FCS was added to stop the staining reaction. The cells were again pelleted, transferred to a fresh centrifuge tube, and washed 3 times with PBS. Cell viability was over 90%, as determined by Trypan blue exclusion.
CCl4-Induced Liver Injury and Administration of EPCs
Induction of Liver Cirrhosis in Rats
Female Sprague-Dawley rats were divided into 2 groups: cirrhotic rats and controls. Cirrhosis was induced with CCl4 (Merck, Darmstadt, Germany). CCl4 was given by gavage twice a week on a full stomach. The initial dose of CCl4 (diluted 1:1 in olive oil) was 5 mL/kg, but each subsequent dose was 3 mL/kg (diluted 1:1 in olive oil) for 12 weeks. The control group received olive oil for the same period.
The experimental design of the transplantation experiments is shown in Fig. 1. Twelve weeks after the administration of CCl4 and water, 2 rats from each group (cirrhotic and control) were killed to confirm the presence of cirrhosis. The rats surviving after 12 weeks of CCl4 administration were divided into 3 experimental groups of similar body weights (10% variation). First, the survival rate in the rats receiving EPC transplantation (n = 10) or saline (n = 10) for 12 weeks was calculated 16 weeks after CCl4administration. Second, to track the EPCs in the cirrhotic liver, the liver tissues of the rats that had been transplanted with PKH26-labeled EPCs at 12 weeks were collected to track the PKH26-labeled EPCs. Third, to determine the role of the EPCs in the cirrhotic liver, the rats were divided into 3 groups after CCl4 administration for 12 weeks. The CCl412w+EPC group (n = 10) received CCl4 plus intraportal transplantation of EPCs (5 × 105 cells). The CCl412w+saline group (n = 10) received CCl4 plus an intraportal injection of the same volume of saline without cells. CCl4 administration was not discontinued in the cirrhotic group after transplantation to avoid spontaneous cirrhosis recovery.23 The CCl412w group (n = 10) was administered only CCl4 for 12 weeks. Four weeks after transplantation (at week 16), rats in the CCl412w+EPC group and CCl412w+saline group were killed under anesthesia, and rats in the CCl412w group were killed 12 weeks after the initial CCl4 administration. The left lobe of the liver was fixed in 4% paraformaldehyde and paraffin-embedded for sectioning. Other portions of the liver were frozen in liquid nitrogen and kept at −80°C for the preparation of genomic DNA and total RNA. Blood was obtained from the inferior vena cava to evaluate the parameters of liver function and hepatic fibrosis.
Cell transplantation and surgical procedures were performed under ether inhalation anesthesia. An upper midline incision was made inferiorly from the xiphoid, and the abdominal cavity was exposed with a retractor. The portal vein was gently exposed with a moistened cotton swab. Freshly harvested EPCs (5 × 105), suspended in 1 mL of saline, were injected into the portal vein with a 25-gauge needle connected to a 2-mL syringe. After surgery, the rats were kept in an incubator that provided an environmental temperature above 26°C and were allowed free access to food and tap water.
In Situ Hybridization of the Sex-Determining Region of the Y Chromosome (Sry) and Costaining with Anti-CD31 and Anti–α-Smooth Muscle Actin (Anti–α-SMA) Antibodies
In situ hybridization was performed as described previously.5 A DNA probe complementary to the rat Sry gene was labeled with digoxigenin by polymerase chain reaction (PCR). Sections of the livers from rats transplanted with EPCs were selected for the in situ hybridization assay. After deparaffinization and rehydration, the sections were fixed again in 4% paraformaldehyde. The hybridization (in 10% dextran sulfate, 5× standard saline sodium citrate, 5× Denhardt's solution, 100 μg/mL salmon sperm DNA, 50% deionized formamide, and the appropriate digoxigenin-labeled probe) was performed overnight at 37°C. The hybrids were then visualized with an alkaline phosphatase–conjugated anti-digoxigenin antibody and detected with a detection system from Boehringer Mannheim. The sections were rinsed again, counterstained with 0.1% nuclear fast red, and analyzed under a light microscope. Some tissue sections were processed without a probe as negative controls.
The hybridized slides were then washed and incubated with 10% normal serum. Mouse antibodies directed against rat α-SMA (Sigma Chemical) and CD31 were used as the primary antibodies. The sections were then washed in PBS at 25°C. The secondary antibody reaction and peroxidase labeling were performed with Envision Plus (Dako, Denmark) with diaminobenzidine as the chromogen and analyzed under a light microscope.
Immunofluorescent Detection of Sry+ and PKH26+/CD31+ Cells
The liver tissues were sectioned vertically to a 6-μm thickness. Fluorescent immunolabeling was performed with the indirect immunofluorescence method. The primary antibodies were goat anti-Sry polyclonal antibody (1:40 dilution; Santa Cruz Biotechnology) and mouse monoclonal anti-CD31 (1:50 dilution; Chemicon). The sections were incubated with primary antibody overnight at 4°C. After the sections were washed with PBS, they were incubated with secondary antibody, FITC-labeled donkey anti-goat immunoglobulin G (1:100 dilution; Rockland Immunochemicals, Gilbertsville, PA), and FITC-labeled rabbit anti-mouse immunoglobulin G (1:200 dilution; Dako). After they had been washed in PBS, the sections were mounted with a medium containing 4′,6-diamidino-2-phenylindole as a counter stain (Vectashield, Vector Laboratories, Inc., Burlingame, CA), coverslipped, and examined under a Zeiss confocal fluorescence microscope with epifluorescence. The controls were obtained by the omission of the primary or secondary antibody during incubation.
Evaluation of Fibrosis, Histology, and Regeneration
In all experimental groups, 6-μm-thick sections were routinely processed with hematoxylin and eosin staining or Masson trichrome staining to determine the extent of liver necrosis/degeneration or liver fibrosis, respectively.24, 25 For immunohistochemistry, the liver sections were mounted on silane-covered slides and deparaffinized, and the endogenous peroxidase activity was quenched with 0.03% H2O2 in absolute methanol. The liver sections were incubated overnight at 4°C with a mouse monoclonal antibody against rat α-SMA (Sigma Chemical), collagen III (COL III; Sigma Chemical), and transforming growth factor β1 (TGF-β1; Santa Cruz Biotechnology). The secondary antibody reaction and peroxidase labeling were performed with Envision Plus (Dako) with diaminobenzidine as the chromogen. All sections were counterstained with hematoxylin. The histopathology was interpreted by 2 independent board-certified pathologists who were blinded to the study. There was a 5% margin of difference in their analysis.
The quantitative analysis was performed with image analysis software (Leica QWIN, Germany). Video images were captured with a 10× objective lens under constant exposure control. Threshold tools were used to precisely define and measure the total positively stained area and average gray density. The fields were edited manually to eliminate nonspecific artifacts. The average intensity score of the 4 fields analyzed in each case represented the intensity expression. Collagen was measured with Masson staining (green). After 4 fields of the sample had been measured, the final percentage of the fibrotic area was generated automatically with the following algorithm: total positive area/total field area. For α-SMA, COL III, and TGF-β1 staining, the average gray density of the positive area was measured. The final immunoreaction indices were generated automatically from 4 fields with the following algorithm: total positive area × average density. The hepatic labeling index was calculated as the average number of Ki67-labeled nuclei per 1000 hepatocyte nuclei.
Detection of Hepatocyte Growth Factor (HGF) and TGF-α1 by PCR and Western Blotting
The steady-state level of HGF and TGF-α1 messenger RNA was assessed by semiquantitative PCR using β-actin as the housekeeping gene. Total RNA was prepared with the Trizol reagent (Gibco Life Technologies, Gaithersburg, MD) according to the manufacturer's protocol and quantified by ultraviolet spectrophotometry. First-strand complementary DNA was synthesized from 4 μg of total RNA primed with oligo-dT18 primer with the First-Strand complementary DNA synthesis kit (Gibco Life Technologies). One-fiftieth of the reverse-transcription product was amplified in a GeneAmp PCR System 2400 thermal cycler (PerkinElmer Corp., Boston, MA) in a total reaction volume of 20 μL containing 10 pmol of each primer, 200 μM deoxyribonucleotide triphosphate, and 2 units of Taq DNA polymerase. The oligonucleotide primers were as follows: TGF-α (186-bp product) forward, 5′-GCAGTGGTGTCTCACTTCAA-3′, and reverse, 5′-CACTGCCAGGAGATCTGCATGCTC-3′; HGF (101-bp product) forward, 5′-CCAGCTAGAAACAAAGACTTGAAAGA-3′, and reverse, 5′-GAAATGTTTAAGATCTGTTTGCGTT-3′; and β-actin (265-bp product) forward, 5′-CCTACAACTCCATCATGAAGTGTG-3′, and reverse, 5′-ACTGACTTCATCGTACTCCTGCTT-3′. The PCR procedure consisted of an initial denaturation of the complementary DNA for 5 minutes at 95°C and then amplified for 35 cycles of denaturation for 30 s at 95°C, annealing for 30 s at 52°C, and extension for 90 s at 72°C followed by additional extension for 7 minutes at 72°C after the completion of cycling. The PCR products were separated on 2% agarose gels containing ethidium bromide (0.5 μg/mL) and quantified by scanning densitometry with bio-image analysis software (Quantity-One 4.1.0, Bio-Rad, Hercules, CA).
The liver tissues were homogenized in a lysis buffer [50 mM Tris-HCl (pH 8.0) 1% Nonidet P40, 150 mM NaCl, 0.1% sodium dodecyl sulfate, 0.02% sodium azide, and 100 μg/mL phenylmethylsulfonyl fluoride]. The tissue proteins (20 mg) were separated electrophoretically by 12% sodium dodecyl sulfate–polyacrylamide gel electrophoresis and electrotransferred to a nitrocellulose membrane (40 V overnight). Then, the membranes were blocked with nonfat dried milk in Tris-buffered saline containing 0.2% Tween 20 for 1 hour at room temperature. The membranes were then incubated overnight at 4°C with rabbit polyclonal anti-HGF antibody (1:500 dilution; Santa Cruz Biotechnology), rabbit polyclonal anti–TGF-α1 (1:500 dilution; Abcam, Cambridge, United Kingdom), or anti–β-actin (1:1000 dilution; Santa Cruz Biotechnology). After the membranes had been washed 3 times in Tris-buffered saline containing 0.2% Tween 20, they were reacted with horseradish peroxidase–labeled goat anti-rabbit immunoglobulin G (1:2000 dilution; Santa Cruz Biotechnology) for 1 hour at room temperature. Immunodetection was performed with the ECL Plus kit (Pierce Biotechnology) according to the manufacturer's instructions. The signals were visualized with the enhanced chemiluminescence system (Amersham, United States) and exposed to film. The results were quantified by the measurement of the intensity of the hybridization signals.
Evaluation of the Serum Parameters and International Normalized Ratio (INR) for Liver Function
Alanine aminotransferase, aspartate aminotransferase, albumin, total bilirubin, and INR were assessed in our hospital with routine laboratory methods. INR is a measure of the extrinsic pathway of coagulation. It is used to determine the clotting tendency of blood and to monitor liver damage.
The results are expressed as means ± standard deviation. Differences in continuous parameters among groups were analyzed with a 1-way analysis of variance followed by Scheffe's post hoc test. P < 0.05 was considered significant. Animal survival was evaluated with a Kaplan-Meier plot. Significance was defined as P < 0.05. The data were analyzed with SPSS software, version 11.0 (SPSS, Inc., Chicago, IL).
Characteristics of BM-Derived Rat EPCs
After the in vitro induction and differentiation of mononuclear cells from rat BM, the adhesive cells grew out of an angioblast on the fifth day (Fig. 2A) and reached confluence on the ninth day (Fig. 2B).The cells were positive for FITC labeling, UEA adsorption, and DiI-acLDL internalization after 9 days of culture (Fig. 2D), and the EPCs showed positive staining for CD34, CD133, Flk-1, and CD31 (Fig. 2E). After 9 days of culture, 40.12% ± 6.28% of the cells were positive for CD34, 15.62% ± 4.31% were positive for CD133, 44.05% ± 8.20% were positive for Flk-1, and 76.1% ± 7.20% were positive for CD31 (Fig. 2F). The EPCs formed capillary-like structures within 18 hours when plated onto Matrigel (Fig. 2C). The spontaneous formation of capillary-like structures was even observed when the EPCs were seeded at low cell numbers on fibronectin-coated (10 μg/mL) cell culture plates. These observations suggest the high capacity of EPCs to form new vessels.
EPCs Transplanted into CCl4-Treated Rats Engrafted and Differentiated
PKH26-labeled EPCs were transplanted via the portal vein into the cirrhotic livers and were observed around the portal tracts (Fig. 3A), in the fibrotic areas (Fig. 3B), and around inflammatory sites after transplantation. Moreover, PKH26-labeled EPCs were also detected within CD31+ blood vessel walls and formed vessels (Fig. 3C,D).
In the livers of the female recipients killed 4 weeks after transplantation, the engraftment of male EPCs was confirmed along the vasolateral surfaces of blood vessels, with immunofluorescence and in situ hybridization used to detect the Y chromosome (Fig. 3E,F). These EPCs were positive for CD31 and negative for α-SMA 4 weeks after EPC transplantation (Fig. 3G,H). These data suggest that EPCs may contribute to vascular repair and are capable of accelerating the recovery of liver injury. PCR analysis of the male rat–specific Sry gene confirmed the presence of male-derived cells in the liver (Fig. 3I).
EPCs Alleviate CCl4-Induced Liver Cirrhosis
After 12 weeks of CCl4 treatment, liver cirrhosis was observed in Masson-stained sections. EPC transplantation reduced this fibrotic area (P < 0.05) and made the thickened septal fibrosis thinning or disappear. In comparison with the CCl412w group, the expression of COL III increased in the CCl412w+saline group (P < 0.05) but did not differ in the CCl412w+EPC group. α-SMA, a marker of activated hepatic stellate cells (HSCs), and TGF-β1 showed similar changes (Table 1 and Fig. 4). Thus, EPC transplantation prevented and ameliorated the formation of liver fibrosis.
Table 1. Effect of EPC Transplantation on Liver Fibrosis
Fibrosis Area (%)
NOTE: Data are means ± standard deviation (n = 10).
EPC Transplantation Suppresses Liver Inflammation and Promotes Liver Regeneration
After CCl4 administration for 16 weeks, 5 rats remained alive in the saline group (50% survival), but only 1 rat died in the EPC group (90% survival; Fig. 5). At 12 weeks after CCl4 treatment, no ascites or retroperitoneal varices were present in any animal. At 16 weeks after CCl4 treatment, ascites and retroperitoneal varices were present in 4 of the 5 animals in the CCl412w+saline group and in 1 of the 9 animals in the EPC group. The CCl4 treatment also increased the levels of alanine aminotransferase, aspartate aminotransferase, and total bilirubin (Table 2) and reduced the albumin levels below normal (27.7 ± 2.83 versus 32.9 ± 2.8 g/L, respectively, P < 0.05). In contrast, EPC transplantation restored serum albumin levels to normal (32 ± 5.1 g/L for transplanted animals versus 32.9 ± 2.8 g/L for control animals, P > 0.05) and restored INR levels; this suggests that EPCs improve liver function and restore albumin production. EPC transplantation also restored the liver architecture (as seen with hematoxylin and eosin staining), reduced inflammation, and partially resolved the pseudolobules. EPCs significantly increased Ki67 expression (Fig. 6). HGF and TGF-α can stimulate hepatocytes to replicate, and EPC transplantation increased HGF and TGF-α messenger RNA and protein levels (P < 0.05; Fig. 7).
Table 2. Effects of EPC Transplantation on Liver Function
NOTE: Data are means ± standard deviation (n = 10).
Liver cirrhosis, which is characterized by hepatic dysfunction with extensive accumulation of fibrous tissue, is the end stage of chronic liver diseases and a leading cause of death throughout the world. To date, there is no proven therapy for hepatic fibrosis, and in most instances, the only treatment that patients receive is for the complications of the disease. Ideal strategies for the treatment of liver cirrhosis would involve the prevention of fibrogenesis, the stimulation of hepatocyte regeneration, and the reorganization of the liver architecture.26
EPCs originating from BM play a significant role in the neovascularization of ischemic tissues and in the re-endothelialization of injured blood vessels.18–20 During the inflammatory and fibrotic processes of liver fibrosis, intrahepatic hypoxia and ischemia may affect hepatocyte regeneration. Therefore, EPC transplantation may be antifibrogenic and stimulate hepatocyte regeneration.
Here we evaluated the engraftment of EPCs into liver tissue by detecting PKH26-positive and Sry-positive cells in liver tissues after sex-mismatched transplantation. PKH26-labeled EPCs and Sry-positive EPCs were confirmed in the injured livers of female recipient rats. The PKH26-labeled EPCs were detected within CD31+ blood vessel walls and along fibrotic areas, and this suggests that EPC transplantation might promote neovascularization and the reconstruction of the microvasculature.
The EPCs also reduced liver fibrosis. Hepatic myofibroblasts play a central role in the development of liver cirrhosis and are the main extracellular matrix–producing cells in the injured liver. EPCs might inhibit liver fibrosis by affecting activated myofibroblasts. α-SMA is a reliable marker of activated HSCs, which precede the deposition of fibrous tissue.27 Reductions in α-SMA reflect a reduction in the number of activated HSCs.28 EPC transplantation reduced the expression of α-SMA, perhaps by removing activated HSCs, thus inhibiting fibrosis. EPC transplantation also prevented activated HSCs from secreting TGF-β1, a soluble mediator that promotes the fibrogenic response. EPCs may also alter the microenvironment of the liver at the engraftment sites by producing cytokines, such as HGF, VEGF, or TGF-α.29 EPC transplantation increased HGF and TGF-α levels in liver tissue, which modulated HSC proliferation, collagen formation, and TGF-β1 expression.30–33 EPCs could therefore transform into resident mature endothelial cells and diminish liver fibrosis by affecting HSCs via a paracrine mechanism.
EPCs have high proliferative potential,17 and EPC transplantation may be a feasible treatment for patients with cirrhosis or to promote the regeneration of other organs and tissues. Our data indicate that EPC transplantation promoted liver regeneration and improved liver dysfunction, as reported by other researchers.21 The increased Ki67 expression may indicate that rapid regeneration resulting from proliferating hepatic cells is involved in this process, providing a direct mechanism for the stimulation of hepatic cells by EPCs. Although EPC transplantation facilitates ischemia-induced neovascularization and microvasculature reconstruction and might promote hepatocyte regeneration, EPCs can also express several growth factors, such as HGF, that elicit mitogenic, antiapoptotic, and anti-inflammatory effects in hepatocytes.
In summary, our data suggest that EPCs, given their potential to engraft into liver tissue and their ability to ameliorate CCl4-induced liver injury, might provide a new approach to the treatment of liver diseases. However, further work is required to determine whether EPCs directly or indirectly reduce fibrosis.