SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References
  8. Supporting Information

Although mesenchymal stem cells (MSCs) have been implicated in hepatic injury, the mechanism through which they contribute to diabetic liver disease has not been clarified. In this study, we investigated the effects of MSC therapy on diabetic liver damage with a focus on the role of bone-marrow–derived cells (BMDCs), which infiltrate the liver, and elucidated the mechanism mediating this process. Rat bone-marrow (BM)-derived MSCs were administered to high-fat diet (HFD)-induced type 2 diabetic mice and streptozotocin (STZ)-induced insulin-deficient diabetic mice. MSC-conditioned medium (MSC-CM) was also administered to examine the trophic effects of MSCs on liver damage. Therapeutic effects of MSCs were analyzed by assessing serum liver enzyme levels and histological findings. Kinetic and molecular profiles of BMDCs in the liver were evaluated using BM-chimeric mice. Curative effects of MSC and MSC-CM therapies were similar because both ameliorated the aggravation of aspartate aminotransferase and alanine aminotransferase at 8 weeks of treatment, despite persistent hyperlipidemia and hyperinsulinemia in HFD-diabetic mice and persistent hyperglycemia in STZ-diabetic mice. Furthermore, both therapies suppressed the abnormal infiltration of BMDCs into the liver, reversed excessive expression of proinflammatory cytokines in parenchymal cells, and regulated proliferation and survival signaling in the liver in both HFD- and STZ-diabetic mice. In addition to inducing hepatocyte regeneration in STZ-diabetic mice, both therapies also prevented excessive lipid accumulation and apoptosis of hepatocytes and reversed insulin resistance (IR) in HFD-diabetic mice. Conclusion: MSC therapy is a powerful tool for repairing diabetic hepatocyte damage by inhibiting inflammatory reactions induced by BMDCs and IR. These effects are likely the result of humoral factors derived from MSCs. (Hepatology 2014;59:1816–1829)

Abbreviations
Akt

protein kinase B

ALT

alanine aminotransferase

ANOVA

analysis of variance

AST

aspartate aminotransferase

Bcl2

B-cell lymphoma 2

BM

bone marrow

BMDC

BM-derived cell

BMT

BM transplantation

b.w.

body weight

Ccr2

C-C chemokine receptor 2

C/EBPα, CCAAT/enhancer-binding protein alpha; ERK

extracellular signal-regulated kinase

FABP4

fatty-acid–binding protein 4

FACS

fluorescence-activated cell sorting

Fizz1

found in inflammatory zone 1

FoxO1

forkhead box O1

GFP

green fluorescence protein

GLUT2

glucose transporter 2

HFD

high-fat diet

HNF-4

hepatocyte nuclear factor 4

HOMA-IR

homeostasis model assessment-estimated IR

HSECs

hepatic sinusoidal endothelial cells

ICAM-1

intracellular adhesion molecule 1

IL

interleukin

IP

intraperitoneal

IR

insulin resistance

Irs-2

insulin receptor substrate 2

JNK

c-Jun amino-terminal kinase

MAPK

mitogen-activated protein kinase

MCs

mononuclear cells

MCP

monocyte chemotactic protein

Mrc1

mannose receptor C type 1

MSC

mesenchymal stem cell

MSC-CM

MSC-conditioned medium

NAFLD

nonalcoholic fatty liver disease

NASH

nonalcoholic steatohepatitis

NF-κB

nuclear factor kappa light-chain enhancer of activated B cells

PBMCs

peripheral blood mononuclear cells

PBS

phosphate-buffered saline

RBCs

red blood cells

SE

standard error

SECs

sinusoidal endothelial cells

SREBP-1c

sterol response element-binding protein 1c

STZ

streptozotocin

T2D

type 2 diabetes

TGF-β1

transforming growth factor beta 1

Tg

transgenic

TLR4

Toll-like receptor 4

TNFα

tumor necrosis factor alpha

Trop2

trophoblast cell-surface antigen 2

TUNEL

terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling

The incidence of diabetes is second only to that of cancer. Type 2 diabetes (T2D) is associated with nonalcoholic fatty liver disease (NAFLD), which progresses to nonalcoholic steatohepatitis (NASH). Insulin resistance (IR), fatty acid accumulation in the liver, and proinflammatory cytokine expression are the main factors that increase susceptibility to hepatocyte damage in NASH.[1] In addition, the prevalence of elevated alanine aminotransferase (ALT) is also 3- to 4-fold higher in patients with type 1 diabetes, compared with the general population, with a marked accumulation of glycogen and steatohepatitis.[2] Therefore, the therapeutic approach for reversing hepatocyte damage as well as IR caused by diabetes is a matter of significance.

Previously, we detected proinsulin- and tumor necrosis factor alpha (TNFα)-producing abnormal cells in the bone marrow (BM) of high-fat diet (HFD)-induced and streptozotocin (STZ)-induced diabetic mice and showed that these cells migrated from the BM to infiltrate the liver.[3] In diabetes, BM-derived cells (BMDCs) infiltrate the liver excessively and subsequently produce cytotoxic chemokines or fuse with hepatocytes, causing parenchymal cells to produce proinsulin and cytotoxic TNFα, which leads to the degeneration or apoptosis of target cells.[4, 5] These findings, in the context of diabetes, suggest that hyperglycemia is the primary cause of abnormal cells in BM and that infiltration of abnormal BMDCs in the liver may be the secondary cause of hepatocyte degeneration.

Intravenous transplantation of BM-derived mesenchymal stem cells (MSCs) has been demonstrated to be effective for diabetes. However, most previous studies have investigated the potential role of BM-derived MSCs in regenerating β cells and for subsequently improving hyperglycemia in type 1 or insulin deficiency diabetes models.[6, 7] A limited number of reports have shown the effects of BM-derived MSCs on T2D with hepatocyte damage and IR.[8, 9] Previous studies have reported that MSC therapy reversed hepatic enzyme levels in HFD-induced diabetic mice and suppressed proinflammatory cytokine expression in the liver, but MSC therapy failed to reverse obesity, hypercholesterolemia, hyperglycemia, and IR.[8] Another study showed an improvement in IR in an HFD-induced diabetes model9; however, rats received an STZ injection in addition to a 2-week HFD and the model was also an insulin deficiency model, rather than a hyperinsulinemia T2D model, so the therapeutic effects of MSCs were derived from pancreatic β-cell regeneration. To date, no reports have shown the effects of MSCs on IR associated with NAFLD or NASH in T2D with persistent hyperglycemia and -insulinemia.

BM-derived MSCs are known to exert immunoregulatory and antiapoptotic effects.[10] The mechanism of action underlying MSC therapy is suggested to mediate the cell complement effect by cell differentiation or various paracrine effects through trophic factors secreted by MSCs.[11] In fact, because systemic infusion of MSC-conditioned medium (MSC-CM) has been shown to ameliorate hepatocellular death and stimulate hepatocyte regeneration in a rat model of fulminant hepatic failure,[12] the mechanism of action of systemically injected MSCs might be the result of their released humoral factors. This hypothesis is conceivable because the number of MSCs logged after systemic injection is remarkably small, considering their powerful therapeutic effects.

In this study, we aimed to investigate the following effects resulting from intravenous administration of MSCs, which contributed to a common, distinctive pathogenesis in liver of mice in an HFD-induced T2D model and an STZ-injected insulin deficiency model: (1) suppression of abnormal BMDC infiltration into the liver; (2) amelioration of hepatocyte degeneration induced by proinflammatory cytokine expression and subsequent apoptosis; (3) enhancement of hepatocyte regeneration; and (4) reduction of IR associated with lipid accumulation and proinflammatory cytokine expression in the liver. To clarify the therapeutic effects of MSC-derived trophic factors, effects of MSC infusion were compared in parallel with those of MSC-CM administration. The present findings may provide new perspectives on therapeutic approaches to diabetes-induced hepatic damage.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References
  8. Supporting Information
Animals and BM Transplantation

C57BL/6J mice and C57BL/6-Tg (CAG-EGFP; green fluorescence protein-transgenic [GFP-Tg]) mice were purchased from Japan SLC (Shizuoka, Japan). GFP-BM chimeric mice were produced by lethal irradiation (9 Gy) and systemic injection with 4-6 × 106 BM cells isolated from GFP-Tg mice. Four weeks after BM transplantation (BMT), hyperglycemia was induced by feeding a HFD containing 60% lard (High-Fat Diet 32; Clea Japan Inc., Tokyo, Japan) or by a single intraperitoneal (IP) injection of STZ (150 mg/kg; Wako, Osaka, Japan) dissolved in citrate buffer (pH 4.5). Controls were fed a normal diet or treated with an IP injection of buffer. After 28 weeks of HFD feeding, mice were administered 1 × 104 MSCs/g body weight (b.w.) four times every 2 weeks (HFD-MSC), whereas controls received vehicle (HFD-vehicle; Fig. 1A). At 4 weeks after STZ injection, mice were administered twice with 1 × 104 MSCs/g b.w. every 4 weeks (STZ-MSC), whereas controls received vehicle (STZ-vehicle; Fig. 2A). This study was performed with the approval of the animal experiment committee of Sapporo Medical University (Sapporo, Japan).

image

Figure 1. Rat MSC therapy for HFD-induced diabetic mice. Protocol for MSC (A) and MSC-CM (D) therapies in HFD-diabetic mice. (B and E) Changes in body-weight and serum blood-glucose levels, AST, and ALT after beginning rMSC therapy. Data are expressed as mean ± SE values of 5-10 animals. †P < 0.05, HFD-vehicle versus control; ‡P < 0.05, HFD-MSC or HFD-MSC-CM versus control; #P < 0.05, HFD-MSC or HFD-MSC-CM versus HFD-vehicle. (C) Distribution of rMSCs marked with PKH26 in liver and femur of HFD mice at 1 week after initial rMSC injection are shown in left panel. Changes in the ratio of PKH26-positive cells in liver and bone at 1, 2, and 4 weeks after initial rMSC injection are shown in right panel. Bar, 50 µm. (F) Histological findings of the periportal area in the hematoxylin and eosin–stained liver section 8 weeks after initial MSC therapy. Bar, 20 µm.

Download figure to PowerPoint

image

Figure 2. Rat MSC therapy for STZ-induced diabetic mice. Protocol for MSC (A) and MSC-CM (D) therapies in STZ-diabetic mice. (B and E) Changes in body-weight and serum blood-glucose levels, AST, and ALT after beginning rMSC therapy. Data are expressed as mean ± SE values of 5-10 animals. †P < 0.05, STZ-vehicle versus control; ‡P < 0.05, STZ-MSC or STZ-MSC-CM versus control; #P < 0.05, STZ-MSC or STZ-MSC-CM versus STZ-vehicle. (C) Distribution of rMSCs marked with PKH26 in liver and femur of STZ mice at 1 week after initial rMSC injection were shown in the left panel. Changes in the ratio of PKH26-positive cells in liver and bone at 1, 2, and 4 weeks after initial rMSC injection are shown in right panel. Bar, 50 µm. (F) Histological findings of the periportal area in the hematoxylin and eosin–stained liver section 8 weeks after the initial MSC therapy. Bar, 20 µm.

Download figure to PowerPoint

Isolation, Culture, and Characterization of Rat BM-Derived MSCs

Rat MSCs were harvested from BM of 8-week-old Lewis rats (Charles River Laboratories Japan Inc., Yokohama, Japan) and cultured as described previously.[13] Immunophenotype and differentiation potential of rat MSCs were then determined (Supporting Fig. 1A,B).

Preparation of MSC-CM From Rat MSCs

Rat MSCs (2 × 105 cells) were seeded in 150-cm2 culture dishes. When MSCs reached confluence, the medium was changed to serum-free medium and cultured for 24 hours. Then, the supernatant was collected and further concentrated (final concentration: 1 mg/mL) as MSC-CM by ultrafiltration using centrifugal filter units with a 10-kDa cutoff (Ultracel-10K; Millipore, Billerica, MA), following the manufacturer's instructions.

Intravenous Administration of MSCs and MSC-CM

MSCs (1 × 104 MSCs/g b.w. per animal suspended in 200 µL of phosphate-buffered saline [PBS]) or MSC-CM (2 mg/kg/day) were administered through the tail vein of mice after induction of diabetes (Fig. 1D). Vehicle administration was PBS for MSC therapy. We also administered MSC-CM to diabetic mice daily for 8 weeks (Fig. 2D). Vehicle administration was serum-free medium that was concentrated using the same procedure as that for the conditioned medium used for MSC-CM therapy.

Detection of Donor MSCs

HFD- and STZ-induced diabetic mice without BMT were administered MSCs labeled with PKH26 Red Fluorescent Cell Linker Kits (Sigma-Aldrich, Saint Louis, MO), sacrificed at 1, 2, or 4 weeks after MSC injection, and lung, liver, kidney, spleen, and bone were obtained. Organs were immersed in 4% paraformaldehyde, and bone was decalcified with 0.5 M of ethylenediaminetetraacetic acid (Wako, Osaka, Japan) for 2 days. Frozen sections of each organ were stained with 4',6-diamidino-2-phenylindole (Dojindo Laboratories, Kumamoto, Japan) at 0.1 mg/mL. Distribution of MSCs expressing red fluorescence in each organ was observed by confocal laser scanning microscopy (LSM 510; Carl Zeiss, Oberkochen, German). The ratio of MSCs distributed in each organ was determined by counting PKH26-positive cells in 20 randomly selected visual fields at 100× magnification per mouse (n = 3-5) and compensated by the number of MSCs given to each mouse.

Quantitative Analysis of the GFP-Positive Area

The mean percentage of the area occupied by GFP-positive cells in the liver of each animal was determined by sampling 20 randomly selected visual fields at 100× magnification using the NIS element BR 3.0 image analyzing system.

Isolation of GFP-Positive BMDCs From the Liver by Fluorescence-Activated Cell Sorting

GFP-positive BMDCs were isolated from the liver. Pieces of liver were removed, minced, and treated with 400 U/mL of collagenase (Wako) diluted in PBS for 30 minutes at 37°C. Dissolved tissue was further minced and filtered with a 100-µm pore cell strainer and centrifuged for 5 minutes at 300×g. The cell pellet was treated with red blood cell (RBC) lysis buffer (Qiagen, Venlo, the Netherlands) to remove remaining RBCs and washed with 2% fetal bovine serum in 0.1 M of PBS. GFP-positive BMDCs were isolated by fluorescence-activated cell sorting (FACS; Aria; BD Biosciences, Tokyo, Japan).

Isolation of Mononuclear Cells From Peripheral Blood

Peripheral blood mononuclear cells (PBMCs) were isolated by the Ficoll-Paque (GE Healthcare Japan, Tokyo, Japan) density-gradient separation method, following the manufacturer's instructions.

Statistical Analysis

Data are expressed as mean ± standard error (SE) values. Analysis of variance (ANOVA) was employed for multiple comparisons. Two-way repeated-measures (mixed between-within subjects) ANOVA, followed by Bonferroni's test, was used for serial assessment. Differences were considered significant at P < 0.05 in all two-tailed tests.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References
  8. Supporting Information
Differentiation of Rat BM-Derived MSCs Into Multiple Mesenchymal Lineages

Specific surface antigens for rat MSCs were detected by FACS. CD90 was positively detected, whereas CD11b, CD31, CD43, CD44, and CD45 were negative (Supporting Fig. 1A). MSCs exhibited osteogenic, adipose, and chondrogenic differentiation ability (Supporting Fig. 1B).

MSC Therapy Prevented Liver Damage in HFD- and STZ-Diabetic Mice

MSC treatment successfully reversed hepatocyte damage caused by diabetes, as indicated by recovery of aspartate aminotransferase (AST) and ALT levels after 8 weeks of treatment in both HFD-MSC (Fig. 1B) and STZ-MSC mice (Fig. 2B), compared to HFD- and STZ-vehicle mice, respectively. Body-weight and blood-glucose levels were mildly ameliorated in HFD-MSC mice, compared to HFD-vehicle mice (Fig. 1B); however, they were unchanged between STZ-vehicle and STZ-MSC (Fig. 2B).

Administered MSCs Detected in Liver and BM of HFD- and STZ-Diabetic Mice

PKH26-positive MSCs were detected in the periportal area and in the parenchyma of the liver, as well as around the marrow sinusoid and in the bone parenchyma in HFD-MSC and STZ-MSC mice (Figs. 1C and 2C). Detection frequency of administered MSCs was changed, depending on the time. The number of MSCs distributed in the liver decreased over time, whereas it was increased for 2 weeks after administration before decreasing thereafter in bone. However, the number of PKH26-positive MSCs detected was too small to account for the direct effect of MSCs, such as cell-cell contact or cell differentiation.

MSC-CM Therapy Prevented Liver Damage in HFD- and STZ-Diabetic Mice, Similar to MSCs

Similar to the effect of cell therapy, MSC-CM injections successfully reversed hepatocyte damage, as indicated by recovery of AST and ALT levels after 8 weeks of initial administration in HFD-diabetic mice and after 4 weeks of initial administration in STZ-diabetic mice (Figs. 1E and 2E). In contrast to cell therapy, there were no differences in body-weight or fasting blood-glucose levels between HFD-MSC-CM and HFD-vehicle mice (Fig. 1E). Body weight recovered in STZ-MSC-CM mice, whereas glucose level remained unchanged (Fig. 2E). The MSC-CM dose of 2 mg/kg/day was considered suitable because our preliminary study showed that 1 mg/kg/day was not effective, whereas 4 mg/kg/day proved toxic and caused loss of body weight (data not shown). Injection of MSC-CM caused no fever in mice.

MSC and MSC-CM Therapies Prevented Histopathological Damage in Liver of HFD- and STZ-Diabetic Mice

Significant infiltration of inflammatory cells into the periportal area and numerous deposition of lipid droplets, which had a round vacuolar appearance, in hepatocytes were detected in HFD-vehicle mice (Fig. 1F). These pathological findings were attenuated in HFD-MSC and HFD-MSC-CM mice. Whereas massive accumulation of inflammatory cells in the periportal area and destruction of the round ligament of the liver were observed in STZ-vehicle mice (Fig. 2F), MSC and MSC-CM therapy suppressed excessive infiltration of inflammatory cells and reversed the structure of hepatic tissue.

MSC and MSC-CM Therapies Suppressed Excessive Infiltration of BMDCs, Especially Macrophages, in Liver of HFD- and STZ-Diabetic Mice

Areas occupied by BMDCs (green) were significantly larger in HFD- and STZ-vehicle mice, compared to controls (Figs. 3A and 4A, left). BMDCs massively infiltrated not only the hepatic sinusoid, but also the liver parenchyma and sometimes encircled hepatocytes. MSC and MSC-CM therapies reversed the increase in the quantitative area occupied by infiltrated BMDCs in liver of HFD- and STZ-vehicle mice (Figs. 3A and 4A, right).

image

Figure 3. BMDCs infiltrating liver of GFP-chimeric HFD-diabetic mice. (A) GFP-expressing BMDCs in the liver. GFP-positive density in the liver is quantified in right panel (mean value of 20 panels per group). (B) Immunofluorescence staining of F4/80 (red) in the liver. (C) mRNA expression of Il-6, Cd11c, Fizz1, and Mrc1 on GFP-positive BMDCs in the liver. (D) Immunofluorescence staining of ICAM-1 (red) in the liver. mRNA expression of Ccr2 on GFP-positive BMDCs in the liver (E) and PBMCs (F). Relative amounts of mRNA are normalized to an internal control, β-actin. Bar, 50 µm. Data are expressed as mean ± SE values of 5-8 animals. *P < 0.05. mRNA, messenger RNA.

Download figure to PowerPoint

image

Figure 4. BMDCs infiltrating liver of GFP-chimeric STZ-diabetic mice. (A) GFP-expressing BMDCs in the liver. GFP-positive density in the liver is quantified in right panel (mean value of 20 panels per group). (B) Immunofluorescence staining of F4/80 (red) in the liver. (C) mRNA expression of Il-6, Cd11c, Fizz1, and Mrc1 on GFP-positive BMDCs in the liver. (D) Immunofluorescence staining of ICAM-1 (red) in the liver. mRNA expression of Ccr2 on GFP-positive BMDCs in the liver (E) and PBMCs (F). Relative amounts of mRNA are normalized to an internal control, β-actin. Bar, 50 µm. Data are expressed as mean ± SE values of 5-9 animals. *P < 0.05. mRNA, messenger RNA.

Download figure to PowerPoint

We investigated the population of macrophages, which expressed F4/80 in BMDCs of diabetic liver. Infiltration of BM-derived macrophages (Figs. 3B and 4B, yellow) was remarkably increased in HFD- and STZ-vehicle mice, but was reversed to normal levels by MSC and MSC-CM therapies.

MSC and MSC-CM Therapies Modified the Character of BMDCs Infiltrating Liver and Mononuclear Cells in Peripheral Blood of HFD- and STZ-Diabetic Mice

Expression of interleukin (Il)-6 and Cd11c, markers for classically activated macrophages, was unchanged in HFD-vehicle mice, compared to controls, although MSC and MSC-CM therapies down-regulated Cd11c expression (Fig. 3C). Il-6 and Cd11c expression was up-regulated in STZ-vehicle mice, but was suppressed by MSC and MSC-CM therapies (Fig. 4C). Expression of Fizz1 (found in inflammatory zone 1) and Mrc1 (mannose receptor C type 1), markers for alternatively activated macrophages, was significantly decreased in HFD- and STZ-vehicle mice, compared to controls, but was reversed to normal levels or up-regulated by MSC and MSC-CM therapies (Figs. 3C and 4C).

MSC and MSC-CM Therapies Regulated Interaction Between BMDCs and Endothelial Cells in the Liver

Intracellular adhesion molecule 1 (ICAM-1) was stained in hepatic sinusoidal endothelial cells (HSECs), particularly in the periportal area in controls. Staining intensity was notably increased in HFD- and STZ-vehicle mice, but was decreased to control levels by MSC and MSC-CM therapies (Figs. 3D and 4D). Expression of Ccr2 (C-C chemokine receptor 2), a receptor for monocyte chemotactic protein (MCP)−1, in GFP-positive BMDCs in the liver was significantly increased in HFD- and STZ-vehicle mice, compared to controls, in which Ccr2 had been reversed to an almost normal level by MSC and MSC-CM therapies (Figs. 3E and 4E). Furthermore, expression of Ccr2 in circulating mononuclear cells in peripheral blood of HFD- and STZ-vehicle mice was significantly increased, but MSC and MSC-CM therapies again reversed this increase (Figs. 3F and 4F).

MSC and MSC-CM Therapies Suppressed Proinflammatory Cytokine/Chemokine Expression in Liver of HFD- and STZ-Diabetic Mice

Inflammatory molecules, such as TNFα, MCP-1, and Toll-like receptor 4 (TLR4), were densely stained in both hepatocytes and BMDCs of HFD- and STZ-vehicle mice, and the staining was remarkably decreased by MSC and MSC-CM therapies (Figs. 5A and 6A). Fatty-acid-binding protein 4 (FABP4) was expressed in both sinusoidal endothelial cells (SECs) and hepatocytes of HFD-vehicle mice, and the staining intensity in hepatocytes was decreased by MSC and MSC-CM therapies (Fig. 5A). The receptor for advanced glycation endproducts was also densely stained in hepatocytes of STZ-vehicle mice, but staining intensity was decreased by MSC and MSC-CM therapies (Fig. 6A).

image

Figure 5. Anti-inflammatory, proliferative, and antiapoptotic effects of rMSCs and rMSC-CM therapies in liver of HFD-diabetic mice. (A) Immunofluorescence staining (red) of TNFα, MCP-1, TLR4, and FABP4 in the liver. Bar, 50 µm. Protein levels of (B) JNK1/3, phosphorylation of JNK (p-JNK), p38-MAPK, phosphorylation of p38-MAPK (p-p38), (C) Erk1/2, p-Erk, Akt, and p-Akt in liver of mice in each group. (D) Photomicrographs of representative sections of the TUNEL reaction. The apoptotic index is quantified in right panel (mean value of 20 panels per group). (E) Protein levels of Bax, Bcl2, caspase-3, and cleaved caspase-3 in liver of mice in each group. Relative amounts of protein are normalized to an internal control, β-actin. Intensity is shown as an arbitrary unit (n = 5, each group). Data are expressed as mean ± SE values *P < 0.05; **P < 0.01.

Download figure to PowerPoint

image

Figure 6. Anti-inflammatory and proliferative effects and regenerative action of rMSCs and rMSC-CM therapies in liver of STZ-diabetic mice. (A) Immunofluorescence staining (red) of TNFα, MCP-1, TLR4, and receptor for advanced glycation endproducts (RAGE) in the liver. Bar, 50 µm. Protein levels of (B) JNK1/3, p-JNK, p38, p-p38, (C) Erk1/2, p-Erk, Akt, p-Akt, (D) HNF-4, and C/EBPα in liver of mice in each group. Relative amounts of protein are normalized to an internal control, β-actin. Intensity is shown as an arbitrary unit (n = 5, each group). (E) mRNA expression of Trop2 in liver of each group (n = 5, each group). Relative amounts of mRNA are normalized to an internal control, β-actin. Data are expressed as mean ± SE values. *P < 0.05; **P < 0.01. mRNA, messenger RNA.

Download figure to PowerPoint

MSC and MSC-CM Therapies Regulated Inflammation, Proliferation, and Survival Signaling of Hepatocytes in HFD- and STZ-Diabetic Mice

Phosphorylation of c-Jun amino-terminal kinases (JNKs) and p38 mitogen-activated protein kinase (MAPK) was activated in liver of HFD- and STZ-vehicle mice (Figs. 5B, 6B), whereas activation of extracellular signal-regulated kinase (ERK)1/2 and protein kinase B (Akt) was suppressed in the liver (Figs. 5C and 6C). MSC and MSC-CM therapies reversed these alterations in both HFD- and STZ-diabetic mice.

MSC and MSC-CM Therapies Suppressed Apoptosis of Hepatocytes in HFD-Diabetic Mice and Enhanced Regeneration of Damaged Hepatocytes in STZ-Diabetic Mice

MSC and MSC-CM therapies markedly decreased terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL)-positive cells in liver of HFD-diabetic mice (Fig. 5D). Mitochondria control cell fate by apoptosis-related molecules, such as Bax and B-cell lymphoma 2 (Bcl2), and the subsequent activation of caspase-3. The results showed that expression of Bax and cleaved caspase-3 was significantly increased in liver of HFD-vehicle mice. In addition, Bcl2 expression was significantly decreased in liver of HFD-vehicle mice. These alterations were reversed by MSC and MSC-CM therapies (Fig. 5E). On the other hand, expression of hepatocyte nuclear factor 4 (HNF-4) was stimulated, and the transcription factor, CCAAT/enhancer-binding protein alpha (C/EBPα), which was down-regulated in STZ-vehicle mice, was notably enhanced or recovered by MSC and MSC-CM therapies (Fig. 6D). Expression of trophoblast cell-surface antigen 2 (also known as Trop2), which was remarkably decreased in STZ-vehicle mice, was recovered by MSC and MSC-CM therapies (Fig. 6E).

MSC and MSC-CM Therapies Prevented Lipogenesis, Fibrosis, Glucogenesis, and IR in Liver of HFD-Diabetic Mice

In HFD-vehicle mice, a number of large lipid droplets were precipitated in hepatocytes, but were decreased in both size and number in HFD-MSC and HFD-MSC-CM mice (Fig. 7A, top). Examination of Azan-stained sections revealed periportal fibrosis in HFD-vehicle mice (Fig. 7A, middle), which was suppressed by MSC and MSC-CM therapies. Expression of transforming growth factor beta 1 (TGF-β1), a marker of stellate cells, was increased in both resident cells and BMDCs in liver of HFD-vehicle mice, but was recovered by MSC and MSC-CM therapies (Fig. 7A, bottom).

image

Figure 7. Effects of rMSC and rMSC-CM therapies on lipogenesis, fibrosis, glucogenesis, and IR of liver in HFD-diabetic mice. (A) Oil Red O staining (top), Azan staining (middle), and TGF-β1 staining (bottom) in liver 8 weeks after initial treatment of MSC and MSC-CM. Lipid droplets are stained red and fibrosis are stained blue. Bar, 50 µm. Protein levels of (B) Irs2, abnormal phosphorylation of Irs2 serine723 (p-Irs2), (C) SREBP-1, (D) FoxO1, and phosphorylated FoxO1 (p-FoxO1) in liver of each group. Relative amounts of protein are normalized to an internal control, β-actin. Intensity is shown as an arbitrary unit (n = 5, each group). (E) Immunofluorescence staining (red) of resistin and GLUT2 in liver of each group. Bar, 50 µm. Data are expressed as mean ± SE values. *P < 0.05; **P < 0.01.

Download figure to PowerPoint

Phosphorylation of insulin receptor substrate 2 (p-Irs2) serine and sterol response element-binding protein 1c (SREBP-1c) levels was notably up-regulated in HFD-vehicle mice (Fig. 7B,C) and was completely normalized by MSC and MSC-CM therapies. Beneficent transcriptional activity by forkhead box O1 (FoxO1) was suppressed in liver of HFD-vehicle mice and normalized by MSC and MSC-CM therapies (Fig. 7D). On the other hand, neither total cholesterol nor triglyceride levels were altered by MSC and MSC-CM therapies (Supporting Fig. 2A).

Resistin was densely stained in liver of HFD-vehicle mice and was reduced by MSC and MSC-CM therapies (Fig. 7E, top). Glucose transporter 2 (GLUT2) staining in the liver was reduced in liver of HFD-vehicle mice, but was recovered by MSC and MSC-CM therapies (Fig. 7E, bottom).

We also examined effects of MSC and MSC-CM therapies on serum adiponectin, insulin, and homeostasis model assessment-estimated IR (HOMA-IR) as major factors of IR. Serum adiponectin level was decreased in HFD-vehicle mice and normalized by MSC and MSC-CM therapies (Supporting Fig. 2B). The HOMA-IR level in the liver was significantly increased in HFD-vehicle mice and significantly reduced by MSC and MSC-CM therapies (Supporting Fig. 2B).

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References
  8. Supporting Information

We investigated the mechanism of action of MSC therapy for different pathological conditions in liver of HFD- and STZ-induced diabetic mice. Because only a limited number of donor MSCs were observed in the liver, we considered that the effects of the administered MSCs on damaged hepatocytes was a result of the humoral factors released from the MSCs. MSC-CM is a cocktail of various trophic factor secreted from MSCs. The present results show that the curative effects of the MSC and MSC-CM therapies on damaged hepatocytes in diabetic mice were similar: Both therapies ameliorated liver dysfunction and IR.

In previous studies reporting on effects of MSCs on hepatocyte damage, blood-glucose levels were recovered by pancreatic β-cell regeneration,[6, 7] suggesting that amelioration of hyperglycemia improves liver dysfunction. The present study revealed that even though hyperglycemia did not recover, diabetes-induced liver dysfunction was remarkably reversed by the MSC and MSC-CM therapies. The most interesting point was that hepatocyte regeneration occurred in the STZ-diabetic liver under persistent hyperglycemic conditions.

The present results showed a significant increase in BMDC infiltration into liver of HFD- and STZ-induced diabetic mice, which was completely reversed by the MSC and MSC-CM therapies. Macrophages are classified as classically activated macrophages (M1 macrophages) or as alternatively activated macrophages (M2 macrophages).[14] In this study, BMDCs in liver of diabetic mice displayed a molecular profile similar to that of proinflammatory M1 macrophages by expressing Il-6 and Cd11c; however, they changed into M2 macrophages expressing Fizz1 and Mrc1 after application of the MSC and MSC-CM therapies. Further changes, including Ccr2 expression, were also demonstrated in BMDCs in the liver and in mononuclear cells in the peripheral circulation. Expression of MCP-1 was stimulated in liver of diabetic mice, but was remarkably decreased by both therapies, suggesting that accelerated infiltration of BMDCs into the liver induced by MCP-1/CCR2 interactions was blocked by these treatments. ICAM-1 is induced in HSECs by cytokine or chemokine signals and enhances recruitment of inflammatory cells from vessels.[15] Here, notable expression of ICAM-1 in SECs contributed to excessive infiltration of BMDCs and was also suppressed by the MSC and MSC-CM therapies.

The present findings showed that both the MSC and MSC-CM therapies ameliorated inflammatory change and apoptotic reactions in hepatocytes caused by HFD- and STZ-induced diabetes. Hyperlipidemia and hyperglycemia cause an increase in circulating endotoxins, which are ligands of TLR4.[16] In addition, hyperlipidemia and hyperglycemia induce transcription of proinflammatory cytokines, such as TNFα and MCP-1, and adipokines, such as FABP4, by nuclear translocation of NF-κB (nuclear factor kappa light-chain enhancer of activated B cells), which results in hepatic injury and IR.[17, 18] FABP4 is known as a major adipokine that induces IR and is highly expressed in both adipose tissue and endothelial cells.[19, 20] The liver is an important site for the clearance and catabolism of circulating advanced glycation endproducts.[21] Interaction of advanced glycation endproducts and their receptor induces NF-κB transcriptional activation and cytokine production by JNK and p38-MAPK activation.[22] On the other hand, activation of ERK1/2 and Akt induces hepatocyte proliferation, survival, and regeneration. These signaling pathways also regulate apoptosis of parenchymal cells by mitochondrial factors, such as Bax and Bcl-2, and caspase 3.[23] Therefore, in the present study, the inflammatory changes and apoptotic reactions in hepatocytes might have been the result of the following dual mechanism induced by the MSC and MSC-CM therapies: first through reduction of TLR4 expression and sequential inactivation of the p38-MAPK and JNK pathways and second through the direct effect of trophic factors derived from MSCs enhancing the ERK1/2 and Akt pathways.

Our MSC and MSC-CM therapies stimulated hepatocyte regeneration in STZ-diabetic mice. HNF-4 and C/EBPα are transcriptional factors that play a crucial role in the differentiation process of liver stem cells for fully functional mature hepatic cells.[24] Trop2 is not only a novel marker for oval cells, but also a regulator of cell-cell adhesion and cell-cycle progression through phosphorylation of ERK1/2.[25] The results showed that the MSC and MSC-CM therapies induced expression of HNF-4 and recovery of C/EBPα and Trop2 in the STZ-diabetic liver, which might have activated progenitor cells for proliferation and functional recovery of hepatocytes. In contrast, none of the above effects were observed in HFD-diabetic mice, suggesting that the severe fatty degeneration of hepatocytes surpassed the regenerative potential (data not shown).

MSC and MSC-CM therapies not only reduced lipid accumulation in hepatocytes, but also prevented liver fibrosis. As previously reported, stellate cells in the liver are associated with liver fibrosis,[26] and activated stellate cells express TGF-β1.[27] Expression of TGF-β1 was up-regulated in HFD-vehicle mice and reversed by the MSC and MSC-CM therapies. Because almost 50% of BMDCs expressed TGF-β1, it was expected that liver fibrosis in HFD mice was induced by pathological BMDCs that had differentiated into stellate cells in the liver.

The MSC and MSC-CM therapies reversed hyperinsulinemia and down-regulated HOMA-IR in HFD-diabetic mice, suggesting that both therapies ameliorated insulin resistance. SREBP-1c and FoxO1 are master transacting factors that mediate lipogenesis, glucogenesis, lipid metabolism, and insulin sensitivity.[28, 29] Resistin, an adipose-derived hormone, stimulates synthesis and secretion of proinflammatory cytokines, resulting in enhanced liver gluconeogenesis and inflammation.[30] GLUT2 is expressed in hepatocytes and regulates glucose uptake and glucose metabolism.[31] Adiponectin and its receptor are key components in IR that facilitate the translocation of GLUT2 through adenosine-monophosphate–activated protein kinase activation.[32] Here, MSC and MSC-CM therapies ameliorated IR by regulating expression of HOMA-IR, adiponectin, resistin, and proinflammatory cytokines.

Taken together, we demonstrated that MSC and MSC-CM therapies are powerful tools for repairing diabetes-induced hepatocyte damage in HFD- and STZ-diabetic mice and sequential IR in HFD-diabetic mice (Fig. 8). In STZ-diabetic mice, neither therapy induced β-cell regeneration immediately after MSC therapy; instead, they induced hepatocyte regeneration, despite having been conducted under hyperglycemic conditions. Effects of MSC cell therapy might be the result of MSC-derived trophic factors. We speculate that the mechanism of hepatocyte damage repair in diabetes may involve various factors that regulate immune reaction, lipogenesis, glucogenesis, and regeneration of hepatocytes (Fig. 8). MSC therapy with such a pleiotropic action is an effective treatment for diabetic liver damage and may open the door for a novel remedy.

image

Figure 8. Proposed mechanisms for therapeutic effects of MSCs and MSC-CM for type 2 and 1 diabetic liver damage. Effective mechanisms contribute to common, distinctive pathogenesis in liver of type 2 and 1 diabetes.

Download figure to PowerPoint

Acknowledgment

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References
  8. Supporting Information

The authors thank J. Yanagawa, K. Kamiya, and Y. Hayakawa, research assistants in the Second Department of Anatomy, and K. Fujii, research assistant in the First Department of Internal Medicine, for their technical assistance.

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References
  8. Supporting Information
  • 1
    Brunt EM, Janney CG, Di Bisceglie AM, Neuschwander-Tetri BA, Bacon BR. Nonalcoholic steatohepatitis: a proposal for grading and staging the histological lesions. Am J Gastroenterol 1999;94:24672474.
    Direct Link:
  • 2
    West J, Brousil J, Gazis A, Jackson L, Mansell P, Bennett A, Aithal GP. Elevated serum alanine transaminase in patients with type 1 or type 2 diabetes mellitus. QJM 2006;99:871876.
  • 3
    Kojima H, Fujimiya M, Matsumura K, Nakahara T, Hara M, Chan L. Extrapancreatic insulin-producing cells in multiple organs in diabetes. Proc Natl Acad Sci U S A 2004;101:24582463.
  • 4
    Fujimiya M, Kojima H, Ichinose M, Arai R, Kimura H, Kashiwagi A, Chan L. Fusion of proinsulin-producing bone marrow-derived cells with hepatocytes in diabetes. Proc Natl Acad Sci U S A 2007;104:40304035.
  • 5
    Terashima T, Kojima H, Fujimiya M, Matsumura K, Oi J, Hara M, et al. The fusion of bone-marrow-derived proinsulin-expressing cells with nerve cells underlies diabetic neuropathy. Proc Natl Acad Sci U S A 2005;102:1252512530.
  • 6
    Bell GI, Broughton HC, Levac KD, Allan DA, Xenocostas A, Hess DA. Transplanted human bone marrow progenitor subtypes stimulate endogenous islet regeneration and revascularization. Stem Cells Dev 2012;21:97109.
  • 7
    Anzalone R, Lo Iacono M, Loria T, Di Stefano A, Giannuzzi P, Farina F, La Rocca G. Wharton's jelly mesenchymal stem cells as candidates for beta cells regeneration: extending the differentiative and immunomodulatory benefits of adult mesenchymal stem cells for the treatment of type 1 diabetes. Stem Cell Rev 2011;7:342363.
  • 8
    Ezquer M, Ezquer F, Ricca M, Allers C, Conget P. Intravenous administration of multipotent stromal cells prevents the onset of non-alcoholic steatohepatitis in obese mice with metabolic syndrome. J Hepatol 2011;55:11121120.
  • 9
    Si Y, Zhao Y, Hao H, Liu J, Guo Y, Mu Y, et al. Infusion of mesenchymal stem cells ameliorates hyperglycemia in type 2 diabetic rats: identification of a novel role in improving insulin sensitivity. Diabetes 2012;61:16161625.
  • 10
    Prockop DJ, Oh JY. Mesenchymal stem/stromal cells (MSCs): role as guardians of inflammation. Mol Ther 2012;20:1420.
  • 11
    Kuroda Y, Kitada M, Wakao S, Dezawa M. Bone marrow mesenchymal cells: how do they contribute to tissue repair and are they really stem cells? Arch Immunol Ther Exp (Warsz) 2011;59:369378.
  • 12
    van Poll D, Parekkadan B, Cho CH, Berthiaume F, Nahmias Y, Tilles AW, Yarmush ML. Mesenchymal stem cell-derived molecules directly modulate hepatocellular death and regeneration in vitro and in vivo. Hepatology 2008;47:16341643.
  • 13
    Javazon EH, Colter DC, Schwarz EJ, Prockop DJ. Rat marrow stromal cells are more sensitive to plating density and expand more rapidly from single-cell-derived colonies than human marrow stromal cells. Stem Cells 2001;19:219225.
  • 14
    Gordon S. Alternative activation of macrophages. Nat Rev Immunol 2003;3:2335.
  • 15
    Arii S, Imamura M. Physiological role of sinusoidal endothelial cells and Kupffer cells and their implication in the pathogenesis of liver injury. J Hepatobiliary Pancreat Surg 2000;7:4048.
  • 16
    Shi H, Kokoeva MV, Inouye K, Tzameli I, Yin H, Flier JS. TLR4 links innate immunity and fatty acid-induced insulin resistance. J Clin Invest 2006;116:30153025.
  • 17
    Tilg H. The role of cytokines in non-alcoholic fatty liver disease. Dig Dis 2010;28:179185.
  • 18
    Solomon SS, Odunusi O, Carrigan D, Majumdar G, Kakoola D, Lenchik NI, Gerling IC. TNF-alpha inhibits insulin action in liver and adipose tissue: A model of metabolic syndrome. Horm Metab Res 2010;42:115121.
  • 19
    Queipo-Ortuno MI, Escote X, Ceperuelo-Mallafre V, Garrido-Sanchez L, Miranda M, Clemente-Postigo M, et al. FABP4 dynamics in obesity: discrepancies in adipose tissue and liver expression regarding circulating plasma levels. PLoS One 2012;7:e48605.
  • 20
    Maeda K, Cao H, Kono K, Gorgun CZ, Furuhashi M, Uysal KT, et al. Adipocyte/macrophage fatty acid binding proteins control integrated metabolic responses in obesity and diabetes. Cell Metab 2005;1:107119.
  • 21
    Sebekova K, Kupcova V, Schinzel R, Heidland A. Markedly elevated levels of plasma advanced glycation end products in patients with liver cirrhosis—amelioration by liver transplantation. J Hepatol 2002;36:6671.
  • 22
    Yeh CH, Sturgis L, Haidacher J, Zhang XN, Sherwood SJ, Bjercke RJ, et al. Requirement for p38 and p44/p42 mitogen-activated protein kinases in RAGE-mediated nuclear factor-kappaB transcriptional activation and cytokine secretion. Diabetes 2001;50:14951504.
  • 23
    Pradelli LA, Beneteau M, Ricci JE. Mitochondrial control of caspase-dependent and -independent cell death. Cell Mol Life Sci 2010;67:15891597.
  • 24
    Nagaki M, Moriwaki H. Transcription factor HNF and hepatocyte differentiation. Hepatol Res 2008;38:961969.
  • 25
    Okabe M, Tsukahara Y, Tanaka M, Suzuki K, Saito S, Kamiya Y, et al. Potential hepatic stem cells reside in EpCAM+ cells of normal and injured mouse liver. Development 2009;136:19511960.
  • 26
    Sakaida I, Nagatomi A, Hironaka K, Uchida K, Okita K. Quantitative analysis of liver fibrosis and stellate cell changes in patients with chronic hepatitis C after interferon therapy. Am J Gastroenterol 1999;94:489496.
  • 27
    Fujimiya T, Liu J, Kojima H, Shirafuji S, Kimura H, Fujimiya M. Pathological roles of bone marrow-derived stellate cells in a mouse model of alcohol-induced fatty liver. Am J Physiol Gastrointest Liver Physiol 2009;297:G451G460.
  • 28
    Ide T, Shimano H, Yahagi N, Matsuzaka T, Nakakuki M, Yamamoto T, et al. SREBPs suppress IRS-2-mediated insulin signalling in the liver. Nat Cell Biol 2004;6:351357.
  • 29
    Matsumoto M, Han S, Kitamura T, Accili D. Dual role of transcription factor FoxO1 in controlling hepatic insulin sensitivity and lipid metabolism. J Clin Invest 2006;116:24642472.
  • 30
    Muse ED, Obici S, Bhanot S, Monia BP, McKay RA, Rajala MW, et al. Role of resistin in diet-induced hepatic insulin resistance. J Clin Invest 2004;114:232239.
  • 31
    Burcelin R, Dolci W, Thorens B. Glucose sensing by the hepatoportal sensor is GLUT2-dependent: in vivo analysis in GLUT2-null mice. Diabetes 2000;49:16431648.
  • 32
    Kadowaki T, Yamauchi T, Kubota N, Hara K, Ueki K, Tobe K. Adiponectin and adiponectin receptors in insulin resistance, diabetes, and the metabolic syndrome. J Clin Invest 2006;116:17841792.

Supporting Information

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References
  8. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
hep26975-sup-0001-suppinfo.pdf664KSupplementary Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.