Adipose-tissue derived porcine mesenchymal stem cells efficiently ameliorate CCl₄-induced acute liver failure in mice

Abstract

Adipose tissue derived mesenchymal stem cells (ADMSCs) may be an attractive therapeutic source for acute liver injury because of their high accessibility and non-invasiveness. Here, we investigated the therapeutic potentials of porcine ADMSCs for acute liver failure (ALF). The morphology, differentiation potential, expression patterns of cell surface markers and liver-specific genes were compared between the ADMSCs derived from the pigs with or without ALF. For therapeutic studies, the expanded porcine ADMSCs from either ALF pig (ALF-ADMSCs) or healthy control pig (Nor-ADMSCs) of passage 3 were transplanted into CCl₄-induced ALF mice, and the liver histology and functional tests were performed at days 1, 7, 14, and 21 after cell transplantation. ALF-ADMSCs expressed higher mRNA level of hepatic growth factor (HGF) than the Nor-ADMSCs. Both ALF-ADMSCs and Nor-ADMSCs improved liver histology, functions, and mouse survival rate. Higher level of porcine hepatocyte-specific genes was seen in the livers of ALF-ADMSCs transplanted mice as compared to the Nor-ADMSCs transplanted mice. In particular, ALF-ADMSCs transplanted mice expressed significantly higher level of albumin and cytokeratin 18 in the liver tissues as compared to the Nor-ADMSCs transplanted mice. ALF-ADMSCs might be superior to Nor-ADMSCs in the treatment of ALF as the former possesses stronger hepatic differentiation potential.

Introduction

Acute liver failure (ALF) refers to a group of severe clinical syndromes in which rapid deterioration of liver functions leads to jaundice, coagulopathy, hepatic encephalopathy, and a high mortality rate. Current treatments include pharmacotherapy and artificial liver support, but their therapeutic efficacy is limited. The only valid treatment is liver transplantation, however it is associated with multiple limitations such as donor shortage, high surgical risk, high cost, immune rejection after transplantation, and the necessity of life-long immune-suppressant together with the associated poor quality of life. Clearly, other alternative therapies are needed for ALF.

Mesenchymal stem cell (MSC) transplantation has shown great potential in the treatment of many fatal diseases (Duran et al. 2013; Laffey and Matthay 2017; Ashour et al. 2016; Gutierrez-Fernandez et al. 2011), including acute myocardial infraction, acute respiratory distress syndrome, acute kidney injury, and acute ischemic stroke. MSCs can be derived from multiple sources such as bone marrow, umbilical cord blood, and adipose tissues. However, acquisition of bone marrow-derived mesenchymal stem cells (BMSCs) is an invasive process, and yield is usually very low. Umbilical cord blood mesenchymal stem cells (UCMSCs) have low immunogenicity and low rejection, and isolation of UCMSCs poses no harm to the mother and fetus. However, isolation of UCMSCs is technically challenging, and the allogeneic UCMSCs possess tumorigenic risk to the recipients due to the fact that these transplanted cells have evaded the immune surveillance from the donors. In contrast, adipose tissue derived mesenchymal stem cells (ADMSCs) have several advantages over the MSCs of other sources such as they can be produced with high yield, they have greater proliferative capacity and low immunogenicity. Hence, ADMSCs constitute a promising source of MSCs for therapeutic purposes.

ADMSCs was first isolated from liposuction surgery (Zuk et al. 2001), and was subsequently reported to be of therapeutic potential for liver injuries (Zhu et al. 2013; Saidi et al. 2015; Deng et al. 2016). In this aspect, serums from rats with liver injury were shown to improve hepatic differentiation of BMSCs more effectively than hepatocyte growth factor (HGF) (Yang et al. 2009). Pretreatment of MSCs with injured liver tissue led to a significant increase in the expression of hepatocyte specific genes in these cells (Mohsin et al. 2011). ADMSCs derived from pigs with ALF (ALF-ADMSCs) maintained the characteristics and biological activity of stem cells, and expressed higher level of several liver specific genes than the ADMSCs from normal pigs (Nor-ADMSCs) (Hu et al. 2016). Studies have confirmed that BMSCs obtained from stroke rats are superior to those derived from the normal rats in the neurorestorative treatment of stroke, possibly due to the ability of stroke rats-derived BMSCs in promoting angiogenesis and release of hepatic trophic factors (Zacharek et al. 2010). In translational medicine, an important question to be answered is whether the ADMSCs from patients with severe liver injury are functionally viable for therapeutic purposes.

In the present study, ADMSCs from a pig with acute liver failure (ALF-ADMSCs) were compared with the ADMSCs from a normal healthy pig (Nor-ADMSCs) and their therapeutic effects were investigated in acute liver failure in mice.

Materials and methods

Animals

A total of 80 male BALB/c mice (6–8 weeks, 18–22 g) and two male Bama miniature pigs were used in the study. Mice were housed under standard conditions with a 12-h circadian rhythm at an ambient temperature with free access to food and water. Pigs and mice were allowed to adapt to the laboratory environment for 7 days before the initiation of the experiments. All procedures were approved by Zhejiang University of Traditional Chinese Medicine Laboratory Animal Management and Ethics Committee [Approval No: SYXK (Zhe) 2013-0184].

Generation of acute liver failure in pig and isolation of ADMSCs

Acute liver failure (ALF) was induced in a Bama miniature pig by injecting d-galactosamine (d-GalN) (Sigma, St. Louis, MO, USA) at a dose of 1.5 g/kg via jugular vein catheterization under general anesthesia (Shi et al. 2017). ALF was confirmed by histology and blood biochemistry 24 h (h) after d-GalN injection. The pig injected with phosphate-buffered saline (PBS) was used as a control.

Subcutaneous adipose tissue was aseptically collected from the neck and abdomen of the experimental pigs. In brief, pigs were sacrificed with exsanguination. Adipose tissues were surgically collected, washed with equal volumes of PBS (pH 7.2 ± 0.1) (Genom Sciences, Hangzhou, China) containing 3% penicillin and streptomycin, and cut into small pieces. The minced adipose tissues were digested with collagenase (0.6 mg/mL) (Gibco, Carlsbad, CA, USA) in low glucose-Dulbecco’s Modified Eagle’s Medium (L-DMEM) (Gibco, Carlsbad, CA, USA) for 90 min (min) at 37 °C, with periodic mixing to avoid over-digestion. The digested suspension was centrifuged at 700g for 10 min at room temperature and the supernatant was discarded. The cell pellet was resuspended in L-DMEM medium, filtered through a 100 μM nylon mesh, centrifuged again at 700 g for 10 min. The resultant cells were cultured at 37 °C with 5% CO₂ in L-DMEM containing 10% FBS and 1% penicillin and streptomycin. Thereafter, the medium was replaced every 3 days. When the adipose cells reached 80–90% confluency, they were detached with trypsin–EDTA solution (Gibco, Carlsbad, CA, USA), and passaged to expand. The cells of 3rd to 5th passages were used in this study.

Transplantation of porcine ADMSCs into mice with acute liver failure

ALF in mice were induced by intraperitoneal injection of 4.0 mL/kg CCl₄ mixed with olive oil (both from Sinopharm Chemical Reagent Co., Shanghai, China), as previously described (Liang et al. 2018). ALF was confirmed by liver histology and blood biochemistry 24 h after CCl₄ injections.

Mice (n = 80) were divided into the following four groups: Group 1 Normal control group (n = 5); Group 2 ALF mice (n = 25) were injected with 200 μL of PBS via tail vein 24 h after CCl₄ injection; Group 3 ALF mice (n = 25) were injected with 1 × 10⁶ porcine ALF-ADMSCs (in 200 μL of PBS) via tail vein 24 h after CCl₄ injection; Group 4 ALF mice (n = 25) were injected with 1 × 10⁶ porcine Nor-ADMSCs (in 200 μL of PBS) via tail vein 24 h after CCl₄ injection.

Flow cytometry

A total of 1 × 10⁶ porcine ALF-ADMSCs and Nor-ADMSCs of 3rd passage were resuspended in 100 µL of PBS containing 0.3% (w/v) FBS, respectively incubated for 30 min with a mouse monoclonal antibody for CD90-FITC, a mouse monoclonal antibody for CD105-FITC, and a mouse monoclonal antibody for CD14-phyoerythrin. Cells incubated with mouse IgG2a-PE and mouse IgG1-FITC were used as isotype controls. Incubated cells were washed with PBS, and assessed by flow cytometry. Data were analyzed with a BD Accuri™ C6 software (Becton, Dickinson and Company, Piscataway, NJ, USA). All antibodies were purchased from Abcam (Cambridge, UK).

Confirmation of the differentiating potential of the porcine ALF-ADMSCs and Nor-ADMSCs

In order to confirm the differentiating potential of the cultured porcine ADMSCs (both ALF-ADMSCs and Nor-ADMSCs), we evaluated the ability of these cells to differentiate into osteoblast, adipocyte and chondroblast linages. In all experiments, ALF-ADMSCs and Nor-ADMSCs of passage 3 were used for differentiation studies, and the same cells cultured in normal undifferentiation growth medium were used as controls.

For osteogenic differentiation, ALF-ADMSCs and Nor-ADMSCs were plated at a density of 1 × 10⁴/well in six-well plates, and incubated in L-DMEM supplemented with 10% FBS and 1% penicillin and streptomycin solution for 24 h. Culture medium was then changed to osteogenic medium (1:1 preparation of StemPro^® Osteogenesis Supplement and StemPro^® Osteocyte/Chondrocyte Differentiation Basal Medium) (Gibco Life Technologies, New York, USA). The medium was changed every 3 days. On the 21st day of culture, mineral deposits were quantitatively analyzed by Alizarin Red Solution (Sigma-Aldrich, St. Louis, MO, USA) staining. Total cellular RNA was extracted and the expression level of osteogenic genes including osteocalcin, alkaline phosphatase, osteopontin and Runx2 was determined by quantitative real-time PCR (qPCR).

To test the ability of ALF-ADMSCs and Nor-ADMSCs to differentiate into adipose cells, ALF-ADMSCs and Nor-ADMSCs were grown in 6-well plates. The cells with 100% confluency (Day 0) were incubated in the adipogenic differentiation medium (L-DMEM supplemented with 10% FBS, 1% penicillin and streptomycin, 5 μg/mL of insulin, 1 nM of 3,3′,5-Triiodo-l-thyronine, 125 μM of indomethacin, 2 μg/mL of dexamethasone, 0.5 mM of 3-Isobutyl-1-methylxanthine, and 0.5 μM of rosiglitazone) for 2 days. The medium was then changed to induction medium (L-DMEM supplemented with 5 μg/mL of insulin, 1 nM of 3,3′,5-Triiodo-l-thyronine, and 1 μM of rosiglitazone) (Day 2). After additional 48 h of culture, cells were incubated with fresh induction medium (L-DMEM supplemented with 5 µg/mL of insulin, 1 nM of 3,3′,5-Triiodo-l-thyronine, and 1 nM of rosiglitazone) for additional 2 days (Day 4). On Day 6, the same induction medium as the Day 4 was replaced. On Day 8 of induction, adipogenesis was confirmed by Oil Red O staining (Sigma-Aldrich, St. Louis, MO, USA) and detection of adipogenesis markers (aP2, peroxisome proliferation-activated receptor y2(PPARy2), and lipoprotein A) by qPCR.

For chondrogenic differentiation, ALF-ADMSCs and Nor-ADMSCs were plated at a density of 1 × 10⁴/well in six-well plates, incubated in L-DMEM supplemented with 10% FBS and 1% penicillin and streptomycin solution for 24 h. The medium was then changed to chondrogenic medium (1:1 preparation of StemPro^® Osteogenesis Supplement and StemPro^® Osteocyte/Chondrocyte Differentiation Basal Medium) (Gibco Life Technologies, New York, USA). The medium was changed every 3 days. Chondrogenesis was assessed by Alcian blue (Shyuanye, Shanghai, China) staining. At Day 21 of differentiation, the expression of chondrogenic markers (Aggrecan and Col II) was detected by qPCR.

Analysis of cell proliferation

Cell proliferation was measured by CCK8 assay. Briefly, the 5th passaged ALF-ADMSCs and Nor-ADMSCs were plated into 96-well plates at a density of 2000 cells per well, cultured in an incubator at 37 °C with 5% CO₂ for 24 h, and then incubated with 10 µL of CCK8 for 2 h before the absorbance was measured at 450 nm. Cell proliferation was measured daily for 7 days and a cell growth curve was constructed by plotting the culture time on the X-axis and OD values on the Y-axis. All experiments were repeated for three times.

Hepatic differentiation of porcine ALF-ADMSCs and Nor-ADMSCs

Hepatic differentiation was induced as previously reported (Zhang et al. 2018). In brief, ALF-ADMSCs and Nor-ADMSCs of passage 3 were cultured in 6-well plates at a density of 1 × 10⁴/well in Step 1 differentiation medium (L-DMEM supplemented with 20 ng/mL of FGF-4, 20 ng/mL of HGF, 40 nmol/mL of dexamethasone and 1% of ITS containing Insulin, Transferrin, and Selenium). Medium was changed every 3 days. Successful differentiation was verified by periodic acid-Schiff (PAS) (Sigma-Aldrich, Runnymede Malthouse, UK) staining after 21 days, and the hepatic genes ALB, G6PD, HGF, HNF1A, and HNF1B were detected by qPCR.

Western blot analysis

The 3rd passaged ALF-ADMSCs and Nor-ADMSCs were lysed with protein lysis buffer containing a protease inhibitor. Total lysates (50 µg) were loaded onto SDS-PAGE and transferred onto a PVDF membrane (Millipore, Billerica, USA) by electro-blotting. The membranes were blocked with 5% nonfat milk in TBS for 1 h and incubated with the following primary antibodies overnight at 4 °C: anti-HGF (1:5000) (OriGene Technologies, Rockville, UAS) or anti-GAPDH (1:1000) (Beyotime, Shanghai, China). The immunoblotted membranes were incubated with secondary antibodies goat anti-rabbit IgG (1:2000) (Abcam, Cambridge, UK) or goat anti-mouse IgG1 (1:2000) (Abcam, Cambridge, UK) for 2 h at room temperature, respectively. The protein bands were visualized using enhanced chemiluminescence. Quantitative analysis was performed using AlphaView Software (ProteinSmple).

Animal survival and biochemical analysis

Ten mice in each group were used for survival analysis. Survival of the CCl₄-induced ALF mice with or without treatment with porcine ADMSCs (both ALF-ADMSCs and Nor-ADMSCs) was assessed after 21 days of cell transplantation. Serum samples were obtained by centrifugation of the blood samples within 1 h of collection, and were used to detect the specific markers of liver injury including albumin (ALB), alanine transaminase (ALT), aspartate transaminase (AST), total bilirubin (TBiL) and direct bilirubin (DBiL) on the days of 1, 7, and 14 of transplantation.

Histological analysis of liver tissues

On days 1, 7, 14, and 21, mice were euthanized by cervical dislocation, liver tissues were obtained, fixed in 4% paraformaldehyde, embedded in paraffin blocks, and cut into sections of 5-μm thicknesses. Routine hematoxylin and eosin (H&E) staining was performed for histological evaluation of the liver tissues. Immunohistochemical analysis was used to examine the expression level of albumin (ALB) and cytokeratin 18 (CK18) in liver tissues. Briefly, sections were treated with 0.3% hydrogen peroxide in methanol at room temperature for 10 min to block endogenous peroxidase, and then heated in citrate buffer (pH 6.0) for antigen retrieval. Non-specific binding was blocked by 10% bovine serum albumin in PBS. Sections were incubated with a mouse anti-CK18 (1:200) (Abcam, Cambridge, UK) or a goat anti-pig ALB (1:200) (Bethyl, Montgomery, USA) at 4 °C overnight. The sections were washed, and incubated with goat anti-mouse IgG1 (1:500) (Abcam, Cambridge, UK) or rabbit anti-goat IgG (H + L) (1:500) (Bethyl, Montgomery, USA), respectively. Peroxidase activity was visualized by DAB Peroxidase Substrate Kit (Vector Labs, Burlingame, CA, USA). Quantification of CK18 and ALB positive areas was based on five randomly selected fields in each section. The results were visualized and quantified using a microscope (Olympus, Tokyo, Japan). The expression level of ALB and CK18 was semi-quantitatively assessed using Image J (Image J, Sun Microsystems, and Inc. USA) software on microscopy-harvested images.

Tracking the transplanted cells in the recipient mouse livers and qPCR

Evaluation of the engrafted ALF-ADMSCs and Nor-ADMSCs in the recipient liver tissues was performed by examining the expression of porcine hepatic-specific genes in mouse liver tissues. In brief, total RNA was extracted from approximately 30 mg of frozen liver tissue using the RNAiso Plus (TaKaRa, Tokyo, Japan) according to the manufacturer’s protocol. The quantity and purity of RNA were evaluated by the ratio of the absorbance at 260 and 280 nm, and cDNA was synthesized using TaKaRa-Prime Script RT reagent kit (TaKaRa, Tokyo, China). qPCR reactions were run in duplicate with the UltraSYBR Mixture (Low ROX) and the results analyzed using the 7500 Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). All experiments were performed in replicate using three different samples in each group. PCR amplification conditions were as follows: 95 °C for 10 min, followed by 40 cycles at 95 °C for 15 s and 60 °C for 60 s. After 40 cycles, a dissociation curve was generated to verify the specificity of each primer. The relative level of gene expression was normalized to that of GAPDH and calculated using the 2^−ΔΔCT method. The sequences of all primers (Sangon, Shanghai, China) used in the study are listed in Table 1.

Table 1.

Primers for RT-qPCR

Gene name	Forward	Reverse
Osteocalcin	5′-CAGGAGGGAGGTGTGTGAG-3′	5′-TGCGAGGTCTAGGCTATGC-3′
Osteopontin	5′-AAGGACAGTCAGGAGACGAG-3′	5′-TCAATCACATTGGAATGCTC-3′
Alkaline phosphatase	5′-CCAAAGGCTTCTTCTTGCTG-3′	5′-TGTACCCGCCAAAGGTAAAG-3′
Ap2	5′-AACCCAACCTGATCATCACTG-3′	5′-TCTTTCCATCCCACTTCTGC-3′
LPA	5′-GCAGGAAGTCTGACCAATAAG-3′	5′-GGTTTCTGGATGCCAATAC-3′
PPARy2	5′-AGGAGCAGAGCAAAGAGG-3′	5′-AGAGTTACTTGGTCATTCAGG-3′
Aggrecan	5′-TTCCCTGAGGCCGAGAAC-3′	5′-GGGCGGTAATGGAACACAAC-3′
Col II	5′-CTGGAGCTCCTGGCCTCGTG-3′	5′-CAGATGCGCCTTTGGGACCAT-3′
ALB	5′-GCCTCTTGTGGATGAGCCTA-3′	5′-GTTCAGGACCAGGGACAGAT-3′
G6PD	5′-GATCTACCGCATCGACCACT-3′	5′-TGTTGTCTCGGTTCCAGATG-3′
HNF1A	5′-AGGCTCGTGATTCTGCACTT-3′	5′-TTTGGCCTTACTGCCTTCTG-3′
HNF1B	5′-TGTCTACCTTGTGCTCCTTCG-3′	5′-CAGTGTGTTTGGCTCAGTTCA-3′
HGF	5′-ATGCGAGGGAGATTATGGTG-3′	5′-GCAGATTTGGAATGGCACA-3′
GAPDH	5′-GGGCATGAACCATGAGAAGT-3′	5′-AAGCAGGGATGATGTTCTGG-3′

Statistical analysis

GraphPad Prism version 6.02 (GraphPad Software, San Diego, CA, USA) was used for data processing and analysis. All measurements were repeated at least three times, and results are expressed as mean ± SD. Group comparisons were performed using two-way ANOVA or Student’s t-test. Survival analysis was compared between groups with log-rank tests. A p < 0.05 was considered statistically significant.

Results

Serum biochemistry and histopathology in normal and ALF pigs

Serum biochemistry of normal and ALF pigs are shown in Table 2. Compared with normal pig, the serum levels of ALT, AST, TBiL, DBiL in ALF pig were significantly increased, whereas the levels of ALB and TP were significantly decreased. H&E staining of normal pig liver sections showed intact hepatic lobules without hepatocyte enlargement and necrosis. In contrast, liver histology of ALF pig showed unclear hepatocyte boundaries, large-scale necrosis, inflammatory cell infiltration in the necrotic area, and hepatic sinus congestion (Fig. 1).

Table 2.

Partial biochemical indicators of serum in normal pig and ALF pig

Serum biochemistry	Normal pig	ALF pig
ALT (IU/L)	52.1	437.3
AST (IU/L)	53.8	13,805.0
TBiL (μmol/L)	1.74	32.57
DBiL (μmol/L)	1.03	24.65
ALB (g/L)	37.98	37.84
TP (g/L)	70.80	61.50

Fig. 1.

H&E staining of liver tissues of normal and ALF pigs. a Histology for normal pig liver. b Histology for ALF pig liver. Original magnification × 200

Characterization of the stem cell features of ADMSCs

Morphology

Porcine ALF-ADMSCs were morphologically similar to porcine Nor-ADMSCs. The first generation of ADMSCs were small in size with a spindle-shaped morphology. When cultured to the fifth generation, these cells exhibited fibroblast-like morphology with typical fingerprint-like patterns (Fig. 2a).

Fig. 2.

Characterization of porcine ALF-ADMSCs and Nor-ADMSCs. a Morphology of ALF-ADMSCs and Nor-ADMSCs. Scale bars 20 μM. b Analysis for cell surface markers CD90, CD105, and CD14 by flow cytometry. c Multilineage differentiation potential of ALF-ADMSCs and Nor-ADMSCs. These cells can be differentiated into osteogenic, adipogenic and chondrogenic linages, as demonstrated by accumulation of extracellular calcium deposits (alizarin-red staining), intracellular lipid droplets (Oil-red staining), and oval chondrocytes (Alcain blue staining), respectively. Scale bars 20 μM. d Increased mRNA expression levels of osteogenic, adipogenic and chondrogenic markers in differentiated ALF-ADMSCs and Nor-ADMSCs

Expression of stem cell markers

Using flow cytometry, the isolated porcine ALF-ADMSCs and Nor-ADMSCs of passage 3 were confirmed to contain high level of mesenchymal stem cell markers (CD90 and CD105). However, very little expression of the hematopoietic markers CD14 was observed in these cells (Fig. 2b). These expression patterns are consistent with previous reports (Hu et al. 2016).

Differentiation potential

The differentiation potential of the isolated porcine ALF-ADMSCs and Nor-ADMSCs was evaluated by examining the ability of these cells to differentiate into osteoblast, adipocytes and chondroblast. As shown in Fig. 1d, e, ADMSCs cultured in osteogenic inductive medium for 21 days showed calcium deposits (Fig. 2c) and expressed rich level of osteogenetic markers including osteocalcin, alkaline phosphatase and osteopontin (Fig. 2d). The adipogenic differentiation of the porcine ALF-ADMSCs or Nor-ADMSCs was verified by the presence of lipid droplets in cells (Fig. 2c) and adipocyte markers including aP2, PPARy2 and LPA at the Day 8 of differentiation (Fig. 2d). ALF-ADMSCs and Nor-ADMSCs cells cultured in chondrogenic inductive medium for 21 days exhibited typical morphology of oval chondrocytes (Fig. 2c) and expressed high level of chondrocyte markers including Aggrecan and Col II (Fig. 2d).

Importantly, we tested the differentiation potential of ALF-ADMSCs and Nor-ADMSCs into mature hepatocytes. After 21 days of differentiation towards hepatic lineage, ALF-ADMSCs and Nor-ADMSCs showed typical feature of mature hepatocytes such as polygonal or oval cells with double nuclei (Fig. 3a). Some of the differentiated cells of both group were positive for PAS (Fig. 3a, b). These differentiated cells expressed high mRNA level of albumin (ALB), glucose-6-phosphate dehydrogenase (G6PD), hepatocyte growth factor (HGF), hepatocyte nuclear factor-1α (HNF1A), and hepatocyte nuclear factor-1β (HNF1B). However, there was no difference in the expression level of the hepatocyte specific genes between the ALF-ADMSCs and Nor-ADMSCs (Fig. 3c).

Fig. 3.

Hepatic differentiation and proliferation ability of porcine ALF-ADMSCs and Nor-ADMSCs. Positive staining for PAS in ALF-ADMSCs (a) and Nor-ADMSCs (b) cultured in hepatic differentiation medium for 21 days. Scale bars 50 μM. c Expression of porcine hepatocyte specific genes in hepatic differentiated ADMSCs at day 21. d qPCR analysis of liver-specific genes in 3rd passage undifferentiated ALF-ADMSCs and Nor-ADMSCs. Data were expressed as mean ± SD. ****p < 0.0001. e Western blot analysis of HGF protein in 3rd passage and 5th passage undifferentiated ALF-ADMSCs and Nor-ADMSCs. f Proliferative capacity of 5th passage ALF-ADMSCs and Nor-ADMSCs

We also compared the expression level of liver-specific genes between the naïve 3rd passage ALF-ADMSCs and Nor-ADMSCs (i.e., undifferentiated). As shown in (Fig. 3d), the HGF gene expression level was significantly higher in undifferentiated ALF-ADMSCs than in the undifferentiated Nor-ADMSCs. However, no difference was observed in ALB, HNF1A, HNF1B and G6PD between the undifferentiated ALF-ADMSCs and undifferentiated Nor-ADMSCs. At the protein level, there was no difference between 3rd and 5th passage undifferentiated ALF-ADMSCs and undifferentiated Nor-ADMSCs (Fig. 3e).

Collectively, these data showed that both ALF-ADMSCs and Nor-ADMSCs are of multipotency, and the hepatocytes derived from both ALF-ADMSCs and Nor-ADMSCs were functionally viable.

Proliferative capacity

To test the proliferative potential of the ALF-ADMSCs and Nor-ADMSCs, CCK8 assay was performed in the 5th generation of these cells (Fig. 3f). Both ALF-ADMSCs and Nor-ADMSCs showed a very similar growth pattern with a very slow growth rate at the first 1–3 days followed by a logarithmic growth phase after the 4th day of culture and then slowed down after Day 7 (Fig. 3f). No significant difference was observed in the growth rate between the ALF-ADMSCs and Nor-ADMSCs.

Therapeutic effects of porcine ALF-ADMSCs and Nor-ADMSCs on ALF mice

Blood biochemistry

Marked increase in serum ALT, AST, DBiL and TBiL, but a significant reduction in ALB levels were observed in ALF mice. After 7 and 14 days of treatment with either porcine ALF-ADMSCs or porcine Nor-ADMSCs, a significant reduction in ALT, AST, DBiL and TBiL but a marked increase in ALB were seen, whereas mice receiving PBS did not show any improvement of these biochemical parameters (Fig. 4).

Fig. 4.

Serum levels of liver injury markers in mice after cell transplantation. Serum level of ALB a, ALT b, AST c, DBiL d, and TBiL e at 1, 7, and 14 days after cell transplantation. Data represent the mean ± SD of five separate animals. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001

Liver histopathology

By H&E staining, CCl₄ induced ALF mice showed a marked loss of normal hepatic lobule structure, extensive hepatocyte necrosis, large areas of sinusoid congestion, inflammatory cell infiltration, and periportal hemorrhage (Fig. 5a, e and i). Treatment with porcine ALF-ADMSCs or Nor + ADMSCs led to a time-dependent improvement in liver histology. Thus, at the 7, 14, and 21 days of cell transplantation, a significant amelioration in inflammatory cell infiltration and regeneration of new hepatic lobules were observed (Fig. 5b–d, f–h). In contrast, in mice treated with PBS, CCl₄-induced hepatic damage and inflammation did not shown any improvement after 7, 14, and 21 days (Fig. 5j, k and l).

Fig. 5.

Histology of the ALF mice. Representative photographs of mouse liver tissues treated with porcine ALF-ADMSCs a–d or Nor + ADMSCs e–h for 1 day (a, e), 7 days (b, f), 14 days (c, g), and 21 days (d, h) are shown. i–l Mouse livers from Group 2 for 1,7,14 and 21 days, respectively. Data represent the mean ± SD of five separate animals. Original magnification × 100

Tracing of transplanted cells

To investigate whether porcine ADMSCs are capable of engrafting injured mouse liver, ALF mice were sacrificed at days 1, 7, 14, and 21 after treatment, and the expression of porcine hepatic-specific genes was examined in liver tissues. A significant increase in the hepatic expression of HGF was observed in ALF mice receiving porcine ALF-ADMSCs as compared to mice receiving porcine Nor-ADMSCs at day 1 but not days 7, 14, and 21 (p < 0.05). ALB mRNA expression was significantly increased in the ALF-ADMSCs group compared to the Nor-ADMSCs group on days 7 (p < 0.0001) and 14 (p < 0.0001) of treatment. Similarly, a more significant increase in the hepatic expression of HNF1A mRNA was seen in mice receiving the porcine ALF-ADMSCs than in mice receiving the porcine Nor-ADMSCs on days 7 (p < 0.01), 14 (p < 0.05) and 21 (p < 0.01) (Fig. 6). No significant difference in the expression of G6PD was observed between the mice receiving porcine ADMSCs and those receiving PBS treatments at days 1, 7, 14, and 21.

Fig. 6.

Expression of hepatocyte-specific genes in mouse livers following treatment with porcine ADMSCs. Liver tissues of the ALF mice treated with ALF-ADMSCs or Nor-ADMSCs for 1, 7, 14 and 21 days were used to examine the expression of hepatocyte specific genes. Data are expressed as mean ± SD. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001

Hepatic differentiation of porcine ADMSCs

Apart from improved liver histology, there was a significant increase in the hepatic expression of ALB and CK18 in mice receiving porcine cell transplantation (Fig. 7a, b; Fig. 8a, b). Noticeably, the improvement of hepatic ALB expression was more significant in mice receiving 7-day treatment with porcine ALF-ADMSCs than the mice receiving 7-day treatment with porcine Nor-ADMSCs (p < 0.0001) (Fig. 7). After 14 days of cell therapy, a significant increase in the number of CK18 positive cells was observed in mice receiving porcine ALF-ADMSCs (p < 0.05) (Fig. 8a, b).

Fig. 7.

Immunohistochemical staining for ALB in ALF mouse liver tissues treated with porcine ADMSCs. a Expression for porcine ALB in mice treated with porcine ALF-ADMSCs and Nor-ADMSCs for 7, 14 and 21 days. No positive staining for porcine ALB was observed in the livers of the control mice. Scale bars 50 μM. b Quantitative analysis of the ALB positive areas in liver sections. Five randomly selected areas positive for ALB were analyzed. Data are expressed as mean ± SD. ****p < 0.0001

Fig. 8.

Immunohistochemical staining for CK18 in ALF mouse liver tissues treated with porcine ADMSCs, and survival analysis. a Expression for porcine hepatocyte-specific markers CK18 in mice treated with porcine ALF-ADMSCs and Nor-ADMSCs for 7, 14 and 21 days. No positive staining for porcine CK18 was observed in the livers of the control mice. Scale bars 50 μM. b Quantitative analysis of the CK18 positive areas in liver sections. Five randomly selected areas positive for CK18 were analyzed. Data are expressed as mean ± SD. *p < 0.05. c Survival rate of animal in various groups

Survival rate

To test the therapeutic efficiency of ALF-ADMSCs and Nor-ADMSCs on ALF mice, CCl₄-induced ALF mice were infused with either ALF-ADMSCs or Nor-ADMSCs or PBS. After 21 days of treatment, nine out of ten animals died after PBS treatment compared to mice receiving ALF + ADMSCs or Nor + ADMSCs treatment groups (p < 0.05). Six of ten animals treated with ALF + ADMSCs and five of ten animals treated with Nor + ADMSCs (p > 0.05) recovered from ALF (Fig. 8c).

Discussion

Mesenchymal stem cells are involved in a variety of tissue injury repairs due to their self-renewal and multiple differentiation potential. Of the multiplex mesenchymal stem cell lines, ADMSCs have been considered the best seed cells for cell transplantation due to their abundance and availability. In ALF, cell therapy may serve as a bridging treatment for liver transplantation. In experimental cell therapy, the cellular source and the route of cell transplantation may directly affect the therapeutic outcomes. Allogenic cell transplantation may incur immune attack by the recipients. However, ADMSCs are an exception because they are immunoregulatory cells and are capable of suppressing immune responses (Wheat et al. 2017). In a previously published study, ADMSCs delivered through tail vein were shown to exert a significant therapeutic effect for acute liver failure (Kim et al. 2011).

Here, we utilized a mouse model of CCl₄-induced ALF to test the therapeutic potential of porcine ADMSCs. We demonstrated that the allogenic ADMSCs from the pig exerted a significant therapeutic effect for mice with acute hepatic injury, as evidenced by the significant improvement in liver histology, reduction of serum concentration of ALT, AST, DBiL and TBiL, as well as an increased expression of ALB and mouse survival. Notably, allogenic ADMSCs derived from the pig with acute liver failure (ALF-ADMSCs) might exerted a better therapeutic effect than the allogenic ADMSCs derived from the normal pig (Nor-ADMSCs) in ALF mice.

Transplanted MSCs play a therapeutic role through several mechanisms. MSCs can colonize in liver, proliferate and differentiate into hepatocyte-like cells in the microenvironment of liver injury, thereby exerting its therapeutic effects (Li et al. 2013). In addition, MSCs can stimulate the proliferation of hepatic stem cells or hepatocytes by secreting multiple pro-proliferative cytokines (Zhang and Wang 2013) or directly fusing with hepatocytes (Hao et al. 2015). It was reported that MSCs can secrete high level of matrix metalloproteinase and directly degrade the excessive extracellular matrix deposited in the diseased liver, thus reducing the liver fibrosis (Higashiyama et al. 2007). Furthermore, MSCs have immunoregulatory properties, which can alleviate liver inflammation and damage (Singh et al. 2011). In our study, we found that porcine ALF-ADMSCs significantly enhanced the expression of porcine hepatic-specific genes ALB, HGF and HNF1A in ALF mice, confirming the differentiating potential of porcine ADMSCs into hepatocytes, although the expression level of the porcine hepatocyte-specific genes declined with the prolongation of transplantation time.

It has been reported that hepatogenic differentiation after transplantation of BMSCs occurred at a low frequency in the acute liver failure model, and the transplanted BMSCs could promote fibrosis (di Bonzo et al. 2008). For liver damage, pro-fibrotic factors such as TGF-β, platelet-derived growth factor (PDGF), IL-13 and IL-4 play important roles in the activation and proliferation of hepatic stellate cells (HSCs) (Kim et al. 2017). MSCs can achieve anti-fibrotic ability by acting directly or indirectly on HSC. On the one hand, MSCs can directly inhibit the activity of HSC by inducing apoptosis of HSCs; on the other hand, MSCs can indirectly inhibit the activity of HSCs by inhibiting the activity of immune cells. The in vitro hepatic differentiation of ADMSCs usually takes 2–3 weeks. In our study, we observed that 14 to 21 days after the treatment with porcine ADMSCs, mouse liver expressed high level of porcine hepatocyte-specific markers such as ALB and CK18, suggesting that transplanted ADMSCs had migrated to the injured mouse livers and had undergone hepatic differentiation.

Currently, it is believed that MSCs secret HGF, an essential factor for hepatic differentiation. It also has anti-apoptotic, anti-fibrotic and proliferation-promoting functions. Thus, HGF plays an irreplaceable role in the development and regeneration of hepatocytes, and provides a microenvironment for the differentiation of stem cells into hepatocytes. There have been numerous reports on the synergistic use of HGF in the treatment of experimental liver failure. For example, ectopic expression of HGF in UCMSCs was shown to mitigate liver damage and improve the survival rate of acetaminophen-induced acute liver failure in mice through anti-apoptosis and anti-oxidation properties (Tang et al. 2016). Furthermore, HGF-overexpressing ADMSCs showed a greater ability of stimulating hepatocyte regeneration and could significantly ameliorate radiation-induced liver damage in rats (Zhang et al. 2014). In our study, it found that the HGF gene expression level was significantly higher in undifferentiated ALF-ADMSCs than in the undifferentiated Nor-ADMSCs, however there was no difference in the expression level of HGF between the ALF-ADMSCs and Nor-ADMSCs differentiated into mature hepatocytes. Probably because that mature hepatocytes might produce a minimal amount of HGF, however, adipose stem cells can synthesize and secrete large amounts of HGF. We demonstrated that the allogenic ADMSCs from the pig might exerted a therapeutic potential for mice with acute hepatic injury, as evidenced by the improvement in liver histology, reduction of serum concentration of ALT, AST, DBiL and TBiL, as well as an increased expression of ALB and mouse survival. Compared with Nor-ADMSCs transplanted ALF mice, ALF-ADMSCs might be superior in the treatment of CCl₄ induced liver failure in mice owing to ALF-ADMSCs transplanted mice expressed significantly higher level of albumin and cytokeratin 18 in the liver tissues as compared to the Nor-ADMSCs transplanted mice.

The variability of gene profiling in mini pigs is significantly higher than in human, hence the significant high expression of HGF genes in undifferentiated ALF-ADMSCs is most likely due to individual differences in mini pigs. Moreover, it’s necessary to analysis gene expression profiling on adipose derived mesenchymal stem cells from mini pigs. The deficiency of this study is the limited number of mini pigs. In future, we will enlarge the number of mini pigs to study the effect of ADMSCs on ALF, and acquire human ALF-ADMSCs from liver transplantation patients. In summary, porcine ALF-ADMSCs may constitute an advantageous treatment potential for acute liver failure. Further studies are necessary to ascertain the effectiveness and therapeutic mechanisms of ADMSCs.