Human Placenta-Derived Mesenchymal Stem Cells Promote Hepatic Regeneration in CCl4-Injured Rat Liver Model via Increased Autophagic Mechanism

Authors


  • Author contributions: J.J.: collection and analysis of data, data interpretation, and manuscript drafting; J.H.C.: data interpretation and analysis of revision data; Y.L.: data interpretation and analysis of data; J.W.P. and I.H.O.: conception and critical discussion; S.G.H.: financial support and critical discussion; K.S.K.: data interpretation, critical discussion, and manuscript drafting; G.J.K.: conception and design, manuscript drafting, financial support, and final approval of manuscript.

Abstract

Mesenchymal stem cells (MSCs) have great potential for cell therapy in regenerative medicine, including liver disease. Even though ongoing research is dedicated to the goal of bringing MSCs to clinical applications, further understanding of the complex underlying mechanisms is required. Autophagy, a type II programmed cell death, controls cellular recycling through the lysosomal system in damaged cells or tissues. However, it is still unknown whether MSCs can trigger autophagy to enhance regeneration and/or to provide a therapeutic effect as cellular survival promoters. We therefore investigated autophagy's activation in carbon tetrachloride (CCl4)-injured rat liver following transplantation with chorionic plate-derived MSCs (CP-MSCs) isolated from placenta. The expression markers for apoptosis, autophagy, cell survival, and liver regeneration were analyzed. Whereas caspase 3/7 activities were reduced (p < .05), the expression levels of hypoxia-inducible factor-1α (HIF-1α) and factors for autophagy, survival, and regeneration were significantly increased by CP-MSCs transplantation. Decreased necrotic cells (p < .05) and increased autophagic signals (p < .005) were observed in CCl4-treated primary rat hepatocytes during in vitro coculture with CP-MSCs. Furthermore, the upregulation of HIF-1α promotes the regeneration of damaged hepatic cells through an autophagic mechanism marked by increased levels of light chain 3 II (LC 3II). These results suggest that the administration of CP-MSCs promotes repair by systemically concomitant mechanisms involving HIF-1α and autophagy. These findings provide further understanding of the mechanisms involved in these processes and will help develop new cell-based therapeutic strategies for regenerative medicine in liver disease. STEM Cells 2013;31:1584–1596

Introduction

Regenerative medicine using human stem cells is a new and promising field for treating various intractable diseases and damaged organs, including difficult-to-treat liver diseases. Many scientists have demonstrated the abilities of various stem cells to ameliorate liver damage [1–3] in animal models of hepatic failure.

Human placenta-derived stem cells (PDSCs) have recently been classified and have become the focus of attention in stem cell research [4] because they are the first of the adult stem cells to appear, they have great potential for proliferation, differentiation, and self-renewal, they are readily available and they are easily procured without invasive procedures [5]. In contrast to other (MSCs), including bone marrow-derived MSCs (BM-MSCs) and adipose-derived MSCs (AD-MSCs), PDSCs can be secured in massive numbers and have strong immunologic privileges [1, 6]. Previously, we reported that in a full-term placenta there are several types of PDSCs, such as amniotic MSCs (AMSCs), chorionic plate-derived MSCs (CP-MSCs), chorionic villi-derived MSCs (CV-MSCs), and Wharton's jelly-derived MSCs (WJ-MSCs) [7]. We also showed that CP-MSCs have the ability to differentiate into a variety of cell types, including hepatocytes, and also demonstrate therapeutic effects in carbon tetrachloride (CCl4)-injured rat liver through an antifibrotic mechanism [8].

Generally, the balance between cell death and survival is critical for several cellular processes, including embryogenesis, organogenesis, repair systems, and carcinogenesis [9–11]. Tissue and organ repair consists of complex, multicellular processes involving the coordinated efforts of inflammation, survival, and/or cell death regulation [12, 13]. In particular, a microenvironmental stress that frequently occurs at sites of tissue damage is hypoxia. The hypoxic lesions have been well described in sites of damaged tissues associated with microvascular damage, leading to induction of the hypoxia-inducible factor (HIF). Interestingly, it was reported that hypoxia induces autophagy in an HIF-dependent manner in normal and cancer cell lines, resulting in promotion of cell survival [14]. Also, Fujiyoshi and Ozaki reported that, in liver regeneration, cell proliferation and cell growth are achieved through interleukin (IL)-6/STAT3 and phosphatidylinositol 3-kinase (PI3K)/PDK1/Akt pathways via downstream molecules, and protect against cell death by upregulating antiapoptotic factors [15]. Recently, it was reported that proliferation or apoptosis mechanisms are regulated by autophagy [16, 17], which is considered a key regulating factor of disease-repair processes [18, 19].

Autophagy is a type of programmed cell death and is a degradation process involving the digestion of a cell's own organelles through fusion with the lysosomal machinery. It is referred to as “type II programmed cell death,” as it is a cell death process distinct from apoptosis, which is referred to as “type I programmed cell death.” Because it is difficult to separate the independent roles of autophagy and apoptosis, the physiological role of autophagy in developmental cell death has been difficult to define. The complex crosstalk between autophagy (self-digestion) and apoptosis (self-killing) might be a key in the diverse aspects of development and disease pathogenesis [20–22]. Furthermore, although autophagy is sometimes associated with cell death, it is generally considered to be a survival mechanism [23] because autophagy, unlike apoptosis and necrosis, is activated in response to such stresses as starvation, hypoxia, mitochondrial dysfunction, and infection [21, 24]. Therefore, it has been argued that autophagy is a crucial cellular pathway that regulates development, differentiation, survival, and homeostasis [25].

The autophagic signal is rapidly regulated through the inhibition of mTOR phosphorylation [22, 26]. The process of autophagosome formation involves three major steps: (a) initiation by the uncoordinated 51-like kinase one complex via mTOR inhibition, (b) nucleation of autophagic vesicles by the Beclin1 PI3K class III complex, which includes Atg6, and (c) elongation of the isolation membrane upon the lipidation of microtubule-associated protein light chain 3 (LC3), leading to the formation of a double membrane-limited autophagic vacuole or autophagosome [26, 27]. Recently, the implications of autophagy in hepatology have been demonstrated [28, 29]. It displays Janus-faced properties; although it is primarily a survival mechanism, under certain conditions, it can also lead to autophagic cell death. Clearly, however, one of the major functions of autophagy is to keep cells alive under the stressful, cell-damaging conditions that exist in most liver diseases [30]. Although the usage of autophagy has been studied in cancer therapies [31, 32], to our knowledge, the autophagic mechanism has not yet been evaluated in stem cell-related therapy.

Therefore, the aims of this study were to investigate the balance between the survival and death of damaged hepatic cells in a CCl4-injured rat liver model through the transplantation of CP-MSCs and to determine whether CP-MSCs transplantation influences the autophagic mechanism to improve the regeneration of the injured liver.

Materials and Methods

Cell Culture

The collection of samples and their use for research purposes were approved by the Institutional Review Board of CHA General Hospital, Seoul, Korea. All participants provided written, informed consent prior to sample collection. Placentas were collected from women who were free of medical, obstetrical, and surgical complications and who delivered at term (≥37 gestational weeks). CP-MSCs were harvested as described previously [33]. Briefly, CP-MSCs were collected from the inner side of the chorioamniotic membrane of the placenta. The cells scraped from the membrane were treated with 0.5% collagenase IV (Sigma-Aldrich, St. Louis, http://www.sigmaaldrich.com) and cultured in Ham's F-12/ Dulbecco's modified Eagle's medium (F12/D-MEM) (Invitrogen, Camarillo, CA, http://www.invitrogen.com) supplemented with 10% fetal bovine serum (FBS; Invitrogen), 1% penicillin/streptomycin (Invitrogen), 25 ng/ml human fibroblast growth factor-4 (hFGF-4) (Peprotech, Inc., NJ, www.peprotech.com), and 1 μg/ml heparin (Sigma-Aldrich). WI-38 cells, normal fibroblast, were purchased from the American Type Culture Collection (CCL-75), and the WB-F344 cells, donated by Dr. J. Jang (Seoul National University, Seoul, Korea), were cultured at 37°C in α-MEM (Invitrogen) supplemented with 10% FBS and 1% P/S. Rat primary hepatocytes were isolated from Sprague-Dawley rats (Orient Bio Inc., Korea) by enzymatic isolation protocol. Briefly, rats were perfused with Hanks' balanced salts solution (HBSS) (Invitrogen) containing 150 μg/ml collagenase (Sigma-Aldrich), 2 mg/ml DNase (Promega, WI, http://www.promega.com), 1.3 mM CaCl2, and 2% penicillin/streptomycin for 10–15 minutes. After perfusion and enzyme digestion, the cells filtered through a 100 μm mesh (BD Falcon, CA, http://www.bdbiosciences.com) and cultured according to the experimental schemes with William's E medium (Lonza, Switzerland, http://www.lonza.com) containing 10% FBS and 2% penicillin/streptomycin at 37°C.

CCl4-Injured Liver Rat Model and Transplantation of CP-MSCs

Six-week-old male Sprague-Dawley rats (Orient Bio Inc.) were maintained in an air-conditioned animal house under specific pathogen-free conditions. Liver failure was induced by intraperitoneal (i.p.) injection of CCl4 (1.6 g/kg; DAEJUNG chemicals & metals Co., Ltd., Republic of Korea, www.daejungchem.co.kr) dissolved in corn oil twice a week for 9 weeks. Control rats (n = 5) were injected with an equal volume of corn oil alone. CP-MSCs (2 × 106 cells, 8–10 passages) were transplanted into the spleen (TP; n = 19). Nontransplanted (NTP; n = 19) rats were maintained as sham controls. To prevent immune rejection, rats were intraperitoneally injected daily with an immunosuppressant, FK506 (0.2 mg/kg per day; Prograf; Astellas Pharma, Inc., Republic of Korea, www.astellas.com), beginning 1 day after the CP-MSCs transplantation. Liver tissues were collected 1, 2, and 3 weeks post-transplantation from animals in the TP groups and at the same time points from NTP rats. We conducted all animal experimental procedures using protocols consistent with the National Institutes of Health Guidelines.

Enzyme-Linked Immunosorbant Assay

IL-10 concentration and HIF-1α activities were determined by enzyme-linked immunosorbant assay (ELISA). Equal amounts of protein from individual animals were pooled from control rats (n = 5), TP 1 week (n = 6), TP 2 weeks (n = 6), TP 3 weeks (n = 7), NTP 1 week (n = 6), NTP 2 weeks (n = 6), and NTP 3 weeks (n = 7). All reactions were performed in triplicate. Data are expressed as the mean ± SD of triplicate experiments. The concentration of IL-10 was measured using rat IL-10 Quantikine ELISA kit (R&D systems, MN, www.rndsystems.com) according to the manufacturer's instructions. The activation of HIF-1α transcription factor was measured using a TransAM HIF-1α kit (Active Motif, CA, www.activemotif.com). The tissue lysates (50 μg of each sample) were added to 96-well plates coated with specific double-stranded oligonucleotide and following the manufacturer's protocols.

In Vitro Coculture Experiments

For the detection of autophagy, CP-MSCs or WI-38 cells (3 × 104) were seeded onto Transwell inserts (BD Falcon). WB-F344 cells (1.5 × 104) were exposed to 6 mM CCl4 and rat primary hepatocytes were exposed to 3 mM CCl4 for 1 hour and cocultured with Transwell inserts containing CP-MSCs or WI-38 cells in α-MEM supplemented with 10% FBS and 1% penicillin/streptomycin under normoxia or hypoxia (1% oxygen). For HIF-1α inhibition, siHIF-1α transfection (40 nM; forward, 5′-AGUUAGUUCAAACUGAGUUAAUCCCUU-3′, reverse, 3′-GGGAUUAACUCAGUUUGAACUAACUUU-5′) was performed using Lipofectamine 2000 (Invitrogen) or YC-1 (20 μM; A.G. Scientific Inc., www.agscientific.com) and treated at for 24 hours under hypoxic culture.

Immunostaining

Liver samples were embedded in Tissue-Tek OCT compound (Sakura Finetechnical Co. Ltd. Japan, www.sakura-finetek.com). Cryostat sections (10-μm thick) were fixed in 100% methanol and then incubated in protein block solution (Dako, CA, www.dako.com) for 30 minutes at room temperature (RT). The primary antibodies used in this study are summarized in supporting information Table S1. To observe tissue leukocytes infiltration following transplantation with CP-MSCs or in the NTP group, mouse anti-CD45 antibody followed by incubation with a biotinylated secondary anti-mouse antibody, and incubation with horseradish peroxidase-conjugated streptavidin–biotin complex (Dako), and 3,3-diaminobenzidine (DAB) (Vector Laboratories, http://www.vectorlabs.com) was used to generate a chromatic signal. To investigate the activation of HIF-1α in tissues via translocalization into the nuclei, anti-HIF-1α was used and a rat primary hepatocytes was used as anti-albumin at 4°C overnight followed by a 30 minutes incubation with an Alexa 568 (Invitrogen)-conjugated secondary antibody at RT. To investigate autophagic vacuoles in tissues or cells, mouse anti-LC3 antibody was used at 4°C overnight, followed by a 30 minutes incubation with an Alexa 568 (Invitrogen)-conjugated secondary antibody at RT. 4′,6-diamidino-2-phenylindole (DAPI) (Invitrogen) or propidium iodide (Sigma-Aldrich) staining was used as a counterstain. To analyze the proliferation activities, liver tissues were fixed and embedded in paraffin as mentioned above. The sections were treated with 3% H2O2 in methanol to block endogenous peroxidase activity. After antigen retrieval, the slides were next incubated with an anti-Ki67 monoclonal antibody at 4°C overnight, followed by 30 minutes incubation with the secondary antibody at RT. The sections were counterstained with Mayer's Hematoxylin. The Ki67 labeling index (Ki-LI) represented the percentage of hepatocytes with Ki67-positive nuclei relative to the total number of hepatocytes in randomly selected sections (six fields per rat at ×100 magnification). Images were detected using a Zeiss Axioskop2 MAT microscope (Carl Zeiss MicroImaging).

Quantitative Real-Time Polymerase Chain Reaction Analysis

The mRNA level of vascular endothelial growth factor (VEGF) was determined by quantitative real-time PCR amplification using the SYBR ExScript RT-PCR Kit (TaKaRa) followed by detection in StepOne equipment (Applied Biosystems, www.appliedbiosystems.com). VEGF gene expression was normalized to that of an internal reference (GAPDH). The sequences of the primers: rat VEGF forward 5′-ACTGGACCCTGGCTTTACTG-3′, rat VEGF reverse 5′-ACGCACTCCAGGGCTTCATC-3′, rat GAPDH forward 5′-GGAAAGCTGTGGCGTGAT-3′, and rat GAPDH reverse 5′-AAGGTGGAAGAATGGGAGTT-3′. Target sequences were amplified using the following thermal conditions: 2 minutes at 95°C, and 40 cycles of 5 seconds at 95°C and 30 seconds at 59°C. All reactions were performed in triplicate.

Western Blot Analysis

Liver tissues and cocultured cells were lysed in protein lysis buffer (Sigma-Aldrich). The protein lysates were loaded onto 8%–15% sodium dodecyl sulfate polyacrylamide gels, and the separated proteins were transferred to polyvinylidene difluoride (PVDF) membranes, blocked, and then incubated overnight at 4°C with primary antibodies. supporting information Table S1 lists the primary antibodies used. Membranes were washed and then incubated with a secondary antibody (horseradish peroxidase-conjugated anti-mouse IgG [1:10,000, Bio-Rad Laboratories, CA, www.bio-rad.com]) or anti-rabbit IgG (1:5,000, Bio-Rad Laboratories) or anti-goat IgG (1:5,000, Santa Cruz Biotechnology) for 1 hour at RT using an orbital shaker. After washing, the bands were detected using an enhanced-chemiluminescence reagent (Pierce, IL, http://www.piercenet.com).

Cell Apoptosis and Necrosis Assay

WB-F344 rat liver epithelial cells and rat primary hepatocytes were seeded in 24-well culture plates at 1.5 × 104 cells per well. CP-MS or WI-38 (3 × 104) cells were seeded onto Transwell inserts. The WB-F344 cells and rat primary hepatocytes were exposed to 6 mM CCl4 for 1 hour and cocultured with CP-MSCs or WI-38 cells in α-MEM supplemented with 10% FBS and 1% P/S under normoxia or hypoxia (1% oxygen), respectively. To assess the level of apoptosis, rat hepatic cells and liver tissues were lysed with protease inhibitor-free lysis buffer (culture lysis reagent; Promega), and the caspase 3/7 activities were measured using the Caspase-Glo 3/7 assay kit (Promega). An equal volume of reagents and protein (100 μg) were added into each well of a white-walled 96-well plate and incubated at RT for 2 hours. The luminescence of each sample was measured using a luminometer plate reader (TECAN). To estimate necrosis, coculture supernatants were harvested, and the release of lactate dehydrogenase (LDH) into the supernatants was estimated using a CytoTox 96 assay system (Promega) following the manufacturer's protocol. All of the samples were assayed in duplicate.

Statistical Analysis

Student's t test was used, and p values less than .05 were considered statistically significant.

Results

CP-MSCs Transplantation Has an Anti- inflammatory Effect in CCl4-Injured Rat Liver

Generally, the inflammatory response coincides with damage and repair processes in tissues, causing a hypoxic condition that acts as a triggering factor for cell damage and also as a necessary factor during liver regeneration [34, 35]. To investigate potential anti-inflammatory effects of CP-MSCs transplantation, we analyzed leukocyte infiltration in liver by immunohistochemical staining for the leukocyte common antigen CD45. As shown in Figure 1A, in CCl4-injured NTP group, the leukocytes (CD45 positive) massively infiltrate the liver up to 3 weeks, whereas the transplantation of CP-MSCs into rat liver tissues resulted in a dramatic reduction of leukocytic infiltration (Fig. 1A). Additionally, we investigated whether CP-MSCs induce expression of the anti-inflammatory cytokine IL-10. The expression of IL-10 in CP-MSCs-transplanted liver tissue was significantly increased compared to the NTP group at 1 week and 3 weeks post-transplantation (p < .05, Fig. 1B). These findings suggest that transplantation of CP-MSCs improves injured liver tissues conditions, at least in part through anti-inflammation processes.

Figure 1.

Anti-inflammatory effect on rat injured liver according to CP-MSCs transplantation. (A): Expression of CD45 in rat injured liver tissues depends on CP-MSCs transplantation. Massive leukocytes infiltration was observed in the NTP group, whereas reduced CD45-positive cells were displayed in the TP group. Scale bar = 100 μm (×100 original magnification). (B): Quantitative enzyme-linked immunosorbant assay (ELISA) analysis of the anti-inflammatory cytokine IL-10. Equal amounts of protein from individual animals were pooled from control rats (n = 5), TP 1 week (n = 6), TP 2 weeks (n = 6), TP 3 weeks (n = 7), NTP 1 week (n = 6), NTP 2 weeks (n = 6), and NTP 3 weeks (n = 7) and analyzed by ELISA. All reactions were performed in triplicate. Data are expressed as the mean ± SD of triplicate experiments. Significant differences were observed at 1 week and 3 weeks post-transplantation (*, p < .05). Abbreviations: CTL, control rat; NTP, nontransplanted group; TP, CP-MSCs-transplanted group; wk, week(s); IL-10, Interleukin 10.

CP-MSCs Transplantation Induces Nuclear Translocation and Upregulates Expression of HIF-1α

Since HIF-1α is known as a trigger factor for adaptation and survival-related signal transduction [14], we investigated whether CP-MSCs induce its expression and activation. We found that, in CP-MSCs-transplanted liver tissue, HIF-1α was translocalized to the nuclei compared to NTP or control group (Fig. 2A). Furthermore, compared to the NTP group, CP-MSCs transplantation robustly enhanced expression of HIF-1α (Fig. 2B). Moreover, HIF-1α activity was significantly increased in the TP group compared to NTP group (p < .05, Fig. 2C). These findings suggest that transplantation of CP-MSCs elevate HIF-1α activity via its nuclear translocation as well as upregulated expression.

Figure 2.

Expression and activity of HIF-1α in rat injured liver after CP-MSCs transplantation. (A): Expression and localization of HIF-1α in liver tissues. In the TP group active HIF-1α translocated into the nuclei (red = HIF-1α, blue = nuclei, DAPI). Scale bar = 50 μm (×400 original magnification). (B): HIF-1α expression in liver tissues. Western blot analysis of HIF-1α expression in rat livers at 1, 2, and 3 weeks from TP, NTP, and control rats. Equal amounts of protein from individual animals were pooled at 1, 2, and 3 weeks and loaded onto the gel. Control rats (CTL; n = 5), TP 1 week (n = 6), TP 2 weeks (n = 6), TP 3 weeks (n = 7), NTP 1 week (n = 6), NTP 2 weeks (n = 6), and NTP 3 weeks (n = 7) were used. Actin was used as loading control. (C): HIF-1α activity from liver tissues. Determination of HIF-1α concentration by enzyme-linked immunosorbant assay (ELISA). Equal amounts of protein from individual animals were pooled from the following groups: control rats (CTL; n = 5), TP 1 week (n = 6), TP 2 weeks (n = 6), TP 3 weeks (n = 7), NTP 1 week (n = 6), NTP 2 weeks (n = 6), and NTP 3 weeks (n = 7) and analyzed by ELISA. All reactions were performed in triplicate. Data are expressed as the mean ± SD of triplicate experiments. Significant differences were observed between the TP and NTP groups (*, p < .05). Abbreviations: CTL, control rat; NTP, nontransplanted group; TP, CP-MSCs-transplanted group; wk, week(s); HIF-1α, hypoxia inducible factor-1α.

CP-MSCs Transplantation Dynamically Regulates Apoptotic Signaling in CCl4-Injured Rat Liver

To elucidate the effect of CP-MSCs on hepatic cell survival/death in CCl4-injured rat liver, expression of apoptosis-related factors and poly(ADP-ribose) polymerase (PARP) and the activity of caspase 3/7 were analyzed by Western blot or ELISA analysis. As shown in Figure 3A, the expression level of Bcl-2, a well-known antiapoptotic factor, was dramatically increased in the TP group in comparison to the control and NTP groups. Interestingly, the expression level of Bax, a well-known proapoptotic factor, was greatly increased at 1 week in the NTP group but was subsequently decreased to the control level at 2 and 3 weeks while it was upregulated and sustained in the TP group until 3 weeks compared to the control group. The expression of the cleaved form of PARP, an indicator of apoptosis, was downregulated in the TP group than the NTP group (Fig. 3A). Furthermore, the activity of caspase 3/7 was significantly lower in the TP group compared with the NTP group at all time points (p < .05, Fig. 3B). Our results suggest that transplantation of CP-MSCs controls hepatic cell apoptosis/survival through dynamically regulating the balance between proapoptotic and antiapoptotic factors in CCl4-injured rat liver.

Figure 3.

Apoptosis-related factors, caspase 3/7 activities, autophagy-related factors and LC3 expression from liver tissues of the TP and NTP groups. (A): Bcl2, Bax, and PARP cleavage levels were detected by Western blot analysis using pooled protein samples from rat livers (TP, NTP, and control rats) at 1, 2, and 3 weeks. Actin was used as loading control. (B): The activities of caspase 3/7 by enzyme-linked immunosorbant assay (ELISA) in the liver tissues of the TP and NTP groups. Equal amounts of protein from individual animals were pooled from control rats control rats (CTL; n = 5), TP 1 week (n = 6), TP 2 weeks (n = 6), TP 3 weeks (n = 7), NTP 1 week (n = 6), NTP 2 weeks (n = 6), and NTP 3 weeks (n = 7) and loaded each well of ELISA kit. All reactions were performed in triplicate. Data are expressed as the mean ± SD of triplicate experiments. *Depicts significant differences between the TP and NTP groups (p < .05). (C): Western blot analysis was performed at 1, 2, and 3 weeks in the TP and NTP groups and control rats. Equal amounts of protein from individual animals were pooled at 1, 2, and 3 weeks and loaded onto the gel. Control rats (CTL; n = 5), TP 1 week (n = 6), TP 2 weeks (n = 6), TP 3 weeks (n = 7), NTP 1 week (n = 6), NTP 2 weeks (n = 6), and NTP 3 weeks (n = 7) were used. The expression of total- (t-), phosphorylated- (p-) mTOR, PI3K III, Beclin1, ATG7, ATG5-12, LC3 I, and LC3 II in the liver tissues of the TP and NTP groups. Actin was used as a loading control. (D): LC3, which is the final molecule induced during autophagy, was detected by immunofluorescence in the liver tissues of the TP and NTP groups (green = LC3; red = nuclei, PI). Enhanced LC3-positive autophagosomes were detected in the rat liver of the TP group, whereas low frequencies of LC3 were observed in the NTP group. Inset highlights the structure of LC3-puncta with higher magnifications. Scale bar = 50 μm (×400 original magnification). Abbreviations: CTL, control rat; LC3, light chain 3; NTP, nontransplanted group; TP, CP-MSCs-transplanted group; wk, week(s).

CP-MSCs Transplantation Activates the Autophagic Pathway in CCl4-Injured Rat Liver

Given that there are different types of cell death such as necrosis, apoptosis, and autophagy, their dynamic regulation may critically influence cell repair/regeneration mechanisms. Our findings that CP-MSCs transplantation controls apoptotic factors prompted us to hypothesize that CP-MSCs transplantation could improve liver regeneration by regulating the balance between different forms of cell death, in particular by triggering the autophagy pathway in CCl4-injured rat liver. To address this hypothesis, we examined the expression of autophagic signaling markers by Western blot and immunofluorescence analyses of CCl4-injured rat livers in the TP and NTP groups. Remarkably, we found that the expression levels of autophagic signaling factors, such as PI3K class III, Beclin1, ATG7, ATG5-12, and LC3 II (active form of LC3), were all dramatically increased in the TP group compared to the NTP group (Fig. 3C). In contrast, phosphorylated mTOR (p-mTOR), a well-known negative regulator of autophagy, was lower in the TP group compared to the NTP group, whereas total-mTOR (t-mTOR) expression was similar in both groups. Furthermore, we observed that cells that are positive for the microtubule-associated protein 1A/1B-LC3, a critical marker of autophagy, markedly increased in the TP group compared to the NTP group in all weeks (Fig. 3D). Taken together, these data show that transplantation of CP-MSCs promotes the autophagic mechanism through upregulation of autophagy-inducing factors (e.g., PI3K class III, Beclin1, ATG7, ATG5-12, and LC3 II) as well as downregulation of negative regulator (e.g., p-mTOR) of autophagy in CCl4-injured rat liver.

CP-MSCs Transplantation Increases Liver Regeneration, Cell Survival, and Cell Proliferation in CCl4-Injured Rat Liver

Next, we investigated whether CP-MSCs transplantation affects liver regeneration, cell survival, and cell proliferation in CCl4-injured rat liver tissues. As shown in Figure 4A, the expression levels of proliferation factors (e.g., Jak1, PI3K p110α, phosphorylated ERK1/2, and Smad2/3) were robustly increased in the TP group compared to the NTP group. Furthermore, the expression of cyclin A, cyclin E, pituitary tumor transforming gene1 (PTTG1), and stem cell factor (SCF),which are related to the survival of stem cells and liver regeneration [36–38], were also increased in the TP group compared to the NTP group. Consistent with these results, cell proliferation rate, as measured by Ki-67 expression, was significantly increased in the TP group when compared with the NTP group (p < .001, Fig. 4B), and was comparable to the control group. Additionally, the expression levels of liver regeneration factors IL-6 and gp130 in the TP group were markedly higher than in the NTP group. The expression level of albumin, a liver synthetic function parameter, was higher in the TP group than in the NTP group (Fig. 4C). Interestingly, the expression levels of ATP-binding cassette, subfamily G (ABCG)1 and ABCG2, which are known as somatic stem cell markers and tissue regeneration-related factors [39], were also higher in the TP group compared to the NTP group (Fig. 4D). These findings support the conclusion that transplantation of CP-MSCs promotes liver regeneration by increasing several factors involved in survival, proliferation, and/or regeneration of hepatic cells in CCl4-injured rat liver.

Figure 4.

Expression levels of cell cycle-, liver regeneration-, and tissue regeneration-related factors in liver tissues from the TP and NTP groups. (A): The expression levels of Jak1, PI3K p110α, phosphorylated ERK1/2, Smad2/3, cyclin A, cyclin E, and PTTG1, which are known as factors related with cell proliferation activities, were detected by Western blot analysis. Equal amounts of protein from individual animals were pooled at 1, 2, and 3 weeks and loaded onto the gel. Control rats (CTL; n = 5), TP 1 week (n = 6), TP 2 weeks (n = 6), TP 3 weeks (n = 7), NTP 1 week (n = 6), NTP 2 weeks (n = 6), and NTP 3 weeks (n = 7) were used. GAPDH was used as loading control. (B): The proliferative activity of liver tissues from rats in the control, TP, and NTP groups assessed through immunohistochemical analysis of Ki67 and the Ki67 labeling index (Ki-LI) and presented as the percentage of Ki67-positive nuclei in the total number of hepatocytes. Data expressed as mean ± SD. *Depicts significant differences between the TP and NTP groups (p < .001). (C): The protein levels of liver IL-6 and gp130, which are known liver regeneration-related factors, and albumin, a marker for liver synthetic functions, were analyzed by Western blot analysis from control, TP, and NTP groups. GAPDH was used as loading control. (D): The expressions of ABCG1 and ABCG2, which are known somatic stem cell markers and tissue regeneration markers, were detected. GAPDH was used as loading control. Abbreviations: CTL, control rat; NTP, nontransplanted group; TP, CP-MSCs-transplanted group; wk, week(s), Ki-LI, Ki-67 labeling index.

Effects of In Vitro Coculture of CCl4-Treated Rat Primary Hepatocytes with CP-MSCs on Necrosis and Apoptosis

Coculture experiments of CP-MSCs with CCl4-treated rat primary hepatocytes were performed to further analyze their mode of action. Since in vivo hypoxic conditions critically affect liver damage/regeneration processes [33], we used normoxic and hypoxic in vitro culture conditions. Isolated rat primary hepatocytes were positive for albumin using immunofluorescence, indicating that their liver function is intact (Fig. 5C left). First, we sought to analyze necrotic cell death by measuring LDH secreted into the supernatants of rat primary hepatocytes. As shown in Figure 5A, we observed that CCl4-treatment prominently increased LDH activity in rat primary hepatocytes under both normoxic and hypoxic conditions. Supporting the significance of hypoxic conditions, LDH activity was significantly higher under hypoxic conditions compared to normoxic conditions, regardless of the CCl4 treatment (*, p < .05, Fig. 5A). The levels of LDH were markedly decreased when these CCl4-treated rat primary hepatocytes were cocultured with CP-MSCs and WI-38 cells (§, p < .05), suggesting that both coculture systems similarly affected necrosis. However, the levels of LDH in rat primary hepatocytes were significantly lower when cocultured with CP-MSCs, compared to coculturing with WI-38 cells, under both hypoxic and normoxic conditions (#, p < .05). Interestingly, unlike rat primary hepatocytes, coculture of rat liver epithelial WB-F344 cells with CP-MSCs or WI-38 cells did not affect the LDH levels (supporting information data 1).

Figure 5.

Alteration of necrosis and apoptosis in CCl4-treated rat primary hepatocytes cocultured with CP-MSCs. (A): To assess necrotic cells, LDH analysis was performed on rat primary hepatocytes with or without CCl4 treatment and cocultured with CP-MSCs or WI-38 cells; a coculture-free condition was used as a sham control under normoxia and hypoxia. All reactions were performed in triplicate. Data are expressed as the mean ± SD of triplicate experiments. Significant differences were observed between normoxia and hypoxia (1% oxygen) (*, p < .05) and between coculturing with CP-MSCs and WI-38 (#, p < .05), compared to the coculture-free system (§, p < .05). (B): To analyze apoptotic cells, caspase 3/7 enzyme-linked immunosorbant assay analysis was performed on rat primary hepatocytes using the same conditions as described (A). All reactions were performed in triplicate. Data are expressed as the mean ± SD of triplicate experiments. Significant differences were depicted between normoxia and hypoxia (1% oxygen) (*, p < .05) and between coculturing with CP-MSCs and WI-38 (#, p < .05), as compared to the coculture-free system (§, p < .05). (C): Rat primary hepatocytes were identified by albumin immunostaining (left). Rat primary hepatocytes showing apoptotic phenotypes were identified using DAPI staining. Arrows indicate apoptosomes. Scale bar = 20 μm (×400 original magnification). CP-MSCs, CP; WI-38, WI. Abbreviations: CCL4, carbon tetrachloride; DAPI, 4′,6-diamidino-2-phenylindole; LDH, lactate dehydrogenase.

We next examined whether the above in vitro parameters regulate caspase 3/7 activity. Interestingly, caspase 3/7 activity was significantly higher in rat primary hepatocytes under hypoxia than under normoxia, regardless of the CCl4 treatment (*, p < .05, Fig. 5B). Furthermore, the caspase 3/7 activity was significantly increased when CCl4-treated rat primary hepatocytes were cocultured with CP-MSCs, but not with WI-38 cells (#, p < .05). Thus, it appears that coculturing rat primary hepatocytes with CP-MSCs increase the apoptotic activity while in vivo CP-MSCs transplantation decreased it (Fig. 3B). The reasons underlying these apparently conflicting effects await further investigation. Notably, caspase 3/7 activity was similarly induced by hypoxic condition in rat liver epithelial WB-F344 cells (supporting information data 1).

In Vitro Coculture of CCl4-Treated Rat Primary Hepatocytes with CP-MSCs Prominently Induces Autophagy

Next, we investigated the effects of CP-MSCs on autophagy and its correlation of HIF-1α in rat primary hepatocytes as well as in hepatic epithelial cells. We found that coculture of hepatic epithelial WB-F344 cells with CP-MSCs, but not with control fibroblast WI-38 cells, prominently induced autophagy under hypoxic condition, as examined by LC3 immunocytochemistry (Fig. 6A). To investigate whether HIF-1α is involved in autophagic activation, we treated hepatic epithelial WB-F344 cells with small interference HIF-1α RNA (siHIF-1α) or YC-1 [lsqb]3-(5′-hydroxymethyl-2′-furyl)-1-benzylindazole[rsqb], an HIF-1α inhibitor [40], under the hypoxic coculture system. As shown in Figure 6B, both treatments prominently reduced HIF-1α mRNA expression levels. In addition, LC3 II and VEGF, well known as HIF-1α-downstream genes, were significantly reduced (Fig. 6B, upper). Induced expression of VEGF mRNA was also greatly diminished (Fig. 6B, lower). We also examined the effect of YC-1 on autophagy activation in WB-F344 cells cocultured with CP-MSCs or WI-38 cells. As shown in Figure 6C, 6D, the expression of LC3 II was activated by hypoxic condition (*, p < .05), while it was diminished by combined treatment with YC-1 and CCl4 (#, p < .05). Interestingly, under these same conditions LC3 II was significantly recovered when cells were cocultured with CP-MSCs, compared to conditions without coculture or coculture with WI-38 cells (, p < .05; §, p < .05). However, additional YC-1 treatment under hypoxic conditions decreased LC3 II activation but did not completely inhibited LC 3 II activation.

Figure 6.

Alteration of autophagy in CCl4-treated rat hepatic epithelial cells (WB-F344) cocultured with CP-MSCs under hypoxic condition. (A): LC3 immunofluorescence staining of WB-F344 cells under the indicated conditions (red = LC3; blue = nuclei, DAPI). CCl4 (6 mM) and hypoxia (1% oxygen) were used. Scale bar = 20 μm. (B): To inhibit HIF-1α, siHIF-1α (40 nM) or YC-1 (20 μM) were added to WB-F344 cells under hypoxia (1% oxygen) for 24 hours. HIF-1α, VEGF, and LC3 protein levels were estimated by Western blot analysis. GAPDH was used as loading control (upper). VEGF's mRNA expression level was determined by real-time PCR (lower). (C): Representative Western blot showing HIF-1α, LC3 I, and LC3 II expression levels in WB-F344 cells under the indicated conditions. CCl4 (6 mM), hypoxia (1% oxygen), and YC-1 (20 μM) were used. GAPDH was used as loading control. (D): Quantification of the LC3 II blots shown in (C). Results are the mean ± SD (n = 2), *, p < .05, normoxia versus hypoxia; #, p < .05, hypoxia versus hypoxia with YC1; , p < .05, between coculture-free and cocultured with CP-MSCs and WI-38 conditions; §, p < .005, between coculture with CP-MS and WI-38 cells. CP-MSCs, CP; WI-38, WI. Abbreviations: CCL4, carbon tetrachloride; LC3, light chain 3; VEGF, vascular endothelial growth factor; HIF-1α, hypoxia inducible factor-1α.

In rat primary hepatocytes, similar to our findings in hepatic epithelial WB-F344 cells, LC3 was prominently activated under hypoxic condition compared to normoxia condition (Fig. 7A). Furthermore, in CCl4-treated rat primary hepatocytes, coculture with CP-MSCs prominently enhanced LC3 activation under hypoxic condition (Fig. 7B, 7C). In contrast, coculture with WI-38 cells did not induce any LC3 activation, again demonstrating differential effects of CP-MSCs and WI-38 cells. When rat primary hepatocytes were treated with YC-1 in hypoxic condition, HIF-1α expression levels were reduced regardless of CCl4-treatment and coculture with CP-MSCs or WI-38 cells (Fig. 7B). However, LC3 II expression was significantly induced by coculture with CP-MSCs under hypoxia, compared normoxic conditions (*, p < .05), to those not cocultured with CP-MSCs (, p < .05), and to those cocultured with WI-38 cells (§, p < .05). In addition, LC3 II activation was significantly reduced by YC-1 treatment, in parallel with similar HIF-1α expression patterns (#, p < .05, Fig. 6C, 6D). We also found that LC3 II activation was significantly recovered when cocultured with CP-MSCs (§, p < .05). Taken together, our results strongly suggest that CP-MSCs transplantation/coculture induces autophagic mechanisms in CCl4-damaged hepatic cells in a hypoxia-dependent manner and that HIF-1α is critically involved in the downstream pathway.

Figure 7.

Alteration of autophagy in CCl4-treated rat primary hepatocytes cocultured with CP-MSCs under hypoxic condition. (A): LC3 immunofluorescence staining of rat primary hepatocytes under the conditions indicated (green = LC3; blue = nuclei, DAPI). CCl4 (3 mM) and hypoxia (1% oxygen) were used. Scale bar = 20 μm. (B): Representative Western blot showing expression levels of HIF-1α, LC3 I, and LC3 II in rat primary hepatocytes under the conditions indicated. CCl4 (3 mM), hypoxia (1% oxygen), and YC-1 (20 μM) were used. GAPDH was used as loading control. (C): Quantification of the LC3 II blots shown in (B). The results are the mean ± SD (n = 2), *, p < .05, normoxia versus hypoxia; #, p < .05, hypoxia versus hypoxia with YC1; , p < .05, between coculture-free and cocultured with CP-MSCs and WI-38 conditions; §, p < .05, between coculture with CP-MS and WI-38 cells. CP-MSCs, CP; WI-38, WI. Abbreviations: CCL4, carbon tetrachloride; LC3, light chain 3; HIF-1α, hypoxia inducible factor-1α.

Discussion

MSCs hold great promise as therapeutic agents for tissue regeneration as they offer several advantages, including the ability for self-renewal, the ability to differentiate into multiple cell types as well as their immunomodulatory properties and absence of ethical problems [41, 42]. Indeed, MSCs have shown promising therapeutic potential in several animal models (e.g., liver injury [8], lung injury [43], and myocardial infarction [44]), acting through multiple signaling pathways and several cellular events [1, 8, 43–46]. However, the molecular and/or cellular mechanisms underlying their therapeutic effects are poorly understood.

Hepatic inflammation and fibrosis are prominent features of chronic liver diseases [47]. During hepatic regeneration, inflammation activates Kupffer cells leading to activation of hepatic stellate cells, the major cell type causing fibrosis and collagen synthesis [47, 48]. Recently, we have reported that hypoxia promotes CP-MSCs' self-renewal activities, which was accompanied with induction of the SCF/c-kit pathway as well as the autophagic marker LC3 II, via HIF-1α expression [49]. Based on these findings, we hypothesized that CP-MSCs play a role in the regenerative potential of damaged liver cells through activation of autophagy via hypoxia-related pathways. In this study, we provide several lines of evidence that support our hypothesis. First, we confirmed that CP-MSCs transplantation to CCl4-injured rat liver exhibited antifibrotic activity [8] and anti-inflammatory effects, as examined by leukocyte infiltration and induction of anti-inflammatory cytokine expression (Fig. 1). Furthermore, we found that CP-MSCs transplantation promotes liver regeneration, cell survival, and cell proliferation, as evidenced by comprehensive gene expression analyses of involved factors (Fig. 4). Second, CP-MSCs transplantation to CCl4-injured rats triggers a robust induction of autophagy, as examined by immunostaining of the autophagy marker, LC3 II, and Western blotting analyses (Fig. 3). In addition, in vitro coculture of rat primary hepatocytes and rat hepatic epithelial WB-F344 with CP-MSCs, but not with control human fibroblast WI-38 cell, similarly triggered robust induction of LC3 II expression (Figs. 6 and 7), strongly suggesting that CP-MSCs exert a specific autophagy-inducing effect. Third, our results demonstrate that hypoxic condition plays a critical role in autophagy induction by CP-MSCs by comparing in vitro culture conditions under hypoxia and normoxia. Notably, CP-MSCs transplantation induced not only robust nuclear translocation of HIF-1α but also its expression levels. Furthermore, using both siRNA and a chemical inhibitor of HIF-1α, we also showed that HIF-1α participates in autophagy activation by CP-MSCs. However, inhibition of HIF-1α did not completely abolish LC3 II activation, indicating that additional factors may be involved in the process. In agreement with our results, previous studies showed that hypoxia and inflammation are important features of different progenitor/stem cells and their therapeutic application [50–52].

Although tissue hypoxia and HIF-α signaling pathways have been implicated in tissue injury and repair mechanisms [53, 54], functional roles and regulation of HIF-1α during liver regeneration are not completely known. In this regard, our results shed further insights into the roles of hypoxia and HIF-1α during liver regeneration induced by CP-MSC transplantation through antifibrotic, anti-inflammatory, proregenerative factors as well as autophagic mechanisms.

In regenerative medicine, appropriately controlled balance between damaged cell death and functional cell survival is of utmost importance [55, 56]. In tissue repair, activation of the cell survival pathway might promote cellular adaptation to stresses, whereas it can also convey signals favoring apoptosis in irreversibly injured cells [57]. To investigate the dynamic regulation of different types of cell death, we also investigated the effects of CP-MSCs on necrosis and apoptosis. We found that CCl4-treatment of rat primary hepatocytes induced necrotic cell death under both normoxic and hypoxic conditions, as assessed by LDH activity, although the effect is significantly higher under hypoxic conditions. These results strongly suggest that the regenerative mechanisms induced by CP-MSCs involve reduction of necrotic death. Additionally, in general agreement with previous studies reporting antiapoptotic effects by MSCs transplantation [58, 59], we observed reduced apoptotic signals in CP-MSCs-transplanted liver (Fig. 3). However, coculturing of CCl4-treated rat hepatic epithelial cells and rat primary hepatocytes with CP-MSCs under hypoxia enhanced caspase 3/7 activities and apoptosomes (Fig. 5B, 5C), inferring that CP-MSCs induce different apoptotic responses in vivo and in vitro.

Autophagy is a Janus-faced function as it acts as a cellular recycling function through cell death in the cellular lysosomal system and protects cells from apoptosis by eliminating damaged mitochondria and other organelles that have the potential to trigger apoptosis [60, 61]. However, sustained over-activity or dysfunction of the autophagic pathway in pathologic states can lead to a caspase-independent form of cell death, which shares certain features with apoptosis. It is suggested that well-managed autophagy is critical for protecting weakly damaged cells and eliminating extensively damaged cells for energy supply and/or to provide building blocks to newly born cells in injured tissues [62]. Recently, it was reported that tissue repair occurs after experimentally induced inflammation of the lacrimal gland, programmed cell death after the injury triggers cell proliferation, cell differentiation, and the recruitment of MSCs [63] and is associated with wound repair cytokines regulation [64]. Additionally, a study showed that suppression of autophagy in rat liver might contribute to functional failure of the liver during polymicrobial sepsis [65]. However, in spite of the importance of autophagy in organ homeostasis, there is no report on the regulation of autophagic mechanisms by stem cell-based therapeutic intervention.

To our knowledge, this study is the first showing that CP-MSCs transplantation induces hepatic cell regeneration in injured livers in an animal model, via upregulation of autophagy-related signaling molecules and in an hypoxia-dependent manner. In addition, our results support the notion that CP-MSCs stimulate autophagy in damaged hepatic cells and trigger liver regeneration, at least in part through nuclear localization and induction of HIF-1α. We propose that the repair mechanism triggered by CP-MSCs transplantation in injured liver acts through multiple events as follows: (a) environmental rehabilitation through antifibrosis [66] and anti-inflammation, (b) functional cells repopulation through activation of cell proliferation, (c) damaged cells clearance through apoptosis and autophagy, and (d) cellular protection and recycling of cellular products through autophagy in injured hepatic cells. Our findings that CP-MSC transplantation involves autophagy activation and participation in the repair mechanism offer new insights into further understanding of stem cell-based therapeutic mechanisms and should provide new avenues to develop more efficient therapies.

Conclusion

These results suggest that the administration of CP-MSCs promotes repair by systemically concomitant mechanisms involving HIF-1α and autophagy. These findings provide further understanding of the mechanisms involved in these processes and will help develop new cell-based therapeutic strategies for regenerative medicine in liver disease.

Acknowledgements

We thank Dr. J. Jang (Seoul National University, Seoul, Korea) for the gift of the WB-F344 cells. This work was supported by the Korea Research Foundation Grant funded by the Korean Government (MEST) (KRF-2011-0019610).

Disclosure of Potential Conflicts of Interest

The authors indicate no potential conflicts of interest.

Ancillary