Mesenchymal stem cell therapy of hepatocellular carcinoma in rats: Detection of cell homing and tumor mass by magnetic resonance imaging using iron oxide nanoparticles

,


Introduction
Hepatocellular carcinoma (HCC) is a widespread, treatment-resistant malignancy, culminating in various molecular etiologies. 1 It is a significant worldwide health problem with more than half a million cases diagnosed annually. 2Current data indicate that during transformation to cancer, 2 main pathogenic mechanisms occur: cirrhosis after tissue damage and mutations in oncogenes or tumor suppressor genes.Both mechanisms are followed by changes in important intracellular signals.Targeting these signaling pathways may have a therapeutic perspective, as it may help to reverse, delay or prevent carcinogenesis. 1 Mesenchymal stem cells (MSCs) are being investigated as possible anti-cancer agents indicating a potential to be used to treat different types of malignant tumors.In a review by Shah, it was reported that stem cell therapy holds a remarkable promise for the treatment of human disease. 3SCs possess inherent migratory properties toward tumor cells.This may enable their use as a means for delivering effective, targeted therapy to tumor cells.They could be programmed to express anti-proliferative, pro-apoptotic, anti-angiogenic agents, targeting different cancer types.
Qiao and coworkers showed that human mesenchymal stem cells (hMSCs) can localize tumor sites, preventing the proliferation of tumor cells. 4Others reported that hMSCs, in high proportions, inhibited the proliferation of HepG2 human liver cancer cell lines through mitogen-activated protein kinase (MAPK) and phosphatidylinositol 3-kinase (PI3K) signaling pathways. 5The stem cells seem to create signals leading to the inhibition of cell proliferation and stimulation of differentiation. 6Tumor cells are capable of secreting factors that can attract hMSCs to the tumor site. 7MSCs were proven to regenerate damaged livers.
Currently, super paramagnetic iron oxide nanoparticles (SPION) are safely used for cell tracking. 8In vivo tracking and visualization of the transplanted stem cells is an essential means for monitoring their tissue localization and may confirm their therapeutic potential. 9his study aimed to confirm the role played by MSCs in the amelioration of hepatic pathology in the rat model of cirrhosis/HCC, through MRI tracking of cells labeled with iron oxide nanoparticles.Also, to elucidate the molecular signals responsible for the transformation of liver cells to cancer.The combination of cell labeling and MRI tracking in a time dependent manner with close follow-up of the lesion size and histologic changes provide a novel, more objective assessment of the role of MSCs in disease control.

Material and methods
The approval of the biomedical research ethics committee of the Faculty of Medicine, King Abdulaziz University (KAU), Jeddah, was obtained prior to the study.All the institution guidelines for conducting animal research were followed throughout this work.

Preparation of bone marrow derived MSCs
The method used was a modification of that of Abdel Aziz et al. 10 Bone marrow was collected from the long bones of male rats, and washed with Dulbecco's Modified Eagle's medium (DMEM, GIBCO/BRL) supplemented with 10% fetal bovine serum (GIBCO/BRL) and 1% penicillin-streptomycin (GIBCO/BRL).Unattached cells were removed daily, and adherent cells were kept in the incubator at 37°C in 5% humidified CO 2 for 10-15 days as primary culture.When the cells reached 80-90% confluence, they were washed twice with phosphate buffer saline (PBS), and trypsinized with 0.25% trypsin in 1 mM EDTA (GIBCO/BRL) for 5 min at 37°C.Cells were then centrifuged and resuspended in 50 cm 2 culture flasks (falcon), and incubated (first passage cultures).MSCs in culture were characterized by their adhesiveness and spindle shape and by flow cytometry.

Flow cytometry analysis
MSCs were characterized using cell surface markers to confirm their identity by being CD29 + and CD45 - using specific monoclonal antibodies (Alexa Fluor 488 anti-mouse/rat CD29 antibody, cat# 102212 and PE Antirat CD45 antibody, cat# 202207, BioLegend, California, USA). 5 × 10 5 cells were mixed with 10 μL of the fluorescent-labeled monoclonal antibodies, then incubated in the dark at 2-8°C for 30 min.They were washed twice with phosphate buffered saline (PBS), and the pellet was re-suspended in PBS and analyzed immediately on FACS Aria III (Becton Dickinson, New Jersey, USA).

Labeling of MSCs with iron oxide nanoparticles
Fluid MAG-nanoparticles (Chemicell, Berlin, Germany) are ferrofluid, 200-nm-diameter particles composed of a magnetite core coated by hydrophilic polymers protective against aggregation. 11To improve the ability of these particles to be endocytosed by MSCs, poly-l-lysine was utilized as a prospective vehicle for iron oxide nanoparticle transport into cells. 12SCs were counted, then seeded in micro-well plates at the density of 10 5 cells/well in a culture medium.They were then incubated with medium containing 25 µg Fe/mL and 0.75 mg/mL Poly-l-lysine for 24 h.
To confirm MSCs labeling with iron oxide nanoparticles by transmission electron microscopy, cells were trypsinized, washed 3 times with PBS, centrifuged, and fixed in reagent (glutaraldehyde).The viability of unlabelled MSCs and those incubated with SPION was analyzed by Micro culture Tetrazolium Assay (MTT, Sigma Chemical Company, St Louis, USA).

Animal model for liver cirrhosis and hepatocellular carcinoma
Female albino rats, bred in the animal facilities of King Fahd Center for Medical Research (King Abdulaziz University, Jeddah) were included in the study.They ranged in age between 16 and 24 weeks.Their body weight was 120-150 g.
HCC was induced by a single intra-peritoneal injection of Diethylnitrosamine (DENA, Sigma, USA), containing 200 mg/kg body weight. 13iver cirrhosis was induced by subcutaneous injection of 3 mL/kg body weight of carbon tetrachloride (CCl 4 ) in castor oil. 13Injection was carried out 3 times in the 1 st week, then twice per week for 12 weeks.Animals were divided equally into 4 groups (10 animals each).The 1 st group is a control group of 10 normal rats that received castor oil subcutaneously.Five of them received labeled MSCs after 12 weeks to serve as a control for the radiology study.Cells were injected into the rattail vein (3 × 10 6 cells in PBS).Cells were injected twice with 1-week interval between injections.The same cell count and injection protocol was followed in all groups.The 2 nd group (no treatment group) included rats, which received MSCs vehicle (PBS) only after induction of HCC by DENA and CCl 4. The 3 rd group included rats, which received MSCs after induction of HCC by DENA and CCl 4 (treatment group) .The 4 th group included rats, which received iron oxide nanoparticles-labeled MSCs after induction of HCC by DENA and CCl 4. Rats from this group were included in the radiological study only to follow the disease progression by MRI.

Histological examination
Liver tissue sections were fixed in buffered formalin, paraffin-embedded, then cut at 5-µm sections and stained with hematoxylin and eosin.Liver tissue was analyzed for the HCC induction and progression of treatment with MSCs.

Magnetic resonance imaging
Localization and tracking of the migration of stem cells to the site of the lesions was observed using iron nanoparticles.MRI was used to track the injected cells in a time dependent manner.
The scans were performed on a Biospec 94/30 USR research MR scanner utilizing a T2 Turbo RARE (rapid acquisition with relaxation enhancement). 14,15T2 weighted MR sequences are best suited to evaluate the liver in cases of cirrhosis and hepatocellular carcinoma. 16The base component of nanoparticles is iron oxide, which being para-magnetic, causes shortening of the T2 signal. 17,18o separate time course MR imaging were employed.Axial T2 Turbo RARE images were performed on rats with nanoparticle labeled MSCs (Fig. 5) on days 5, 7, 9 and 13 after the 1 st dose; and days 2 and 6 after the 2 nd dose of the nanoparticle labeled stem cells.Additionally, in the rats with nanoparticles, Axial T2 RARE scans were performed at time intervals of 30 min after injection of the nano-labeled stem cells up to 90 min.Care was taken to ensure that the sequence parameters were kept standard in all scans (TR -110 ms, TE -27 ms, FA -180 degrees, NEX 4).

Biochemical investigations
Serum activities of enzymes aspartate aminotransferase (AST) and alanine aminotransferase (ALT) were measured by the Flex ® reagent cartridges using the Dimension Vista ® system analyzer (Siemens Healthcare Diagnostics Incorporated, Newark, USA).Serum level of rat α-fetoprotein was measured by enzyme linked immunosorbent assay (ELISA) using a kit purchased from USCN Life Science (Wuhan, China).

Ribonucleic acid (RNA) isolation
Total RNA was isolated from liver tissue with the MagNA Pure Compact (Roche) System according to the manufacture's protocol for RNA purification (Cat# 04802993001).

Reverse transcription polymerase chain reaction (RT-PCR) and real time PCR assay
In brief, complementary deoxyribonucleic acid (cDNA) was generated from total RNA using High-Capacity cDNA Reverse Transcription (Applied Biosystem, and Cat # 4375222).RT-PCR was performed in duplicate on AB StepOne Plus System using TaqMan ® Universal PCR Master Mix.The total reaction volume was 20 µL.Expression patterns were evaluated for cyclin D1, beta-catenin and survivin, using TaqMan gene expression assays (Applied Biosystems; Rn00432360, Rn00584431, and Rn00574012, respectively).Real Time PCR was then performed.Cycling condition included: initiation step at 95°C for 2 min, denaturation at 95°C for 1 min, annealing at 60°C for 30 s, extension at 72°C for 30 s, followed by 40 cycles.Relative gene expression data was analyzed using a mathematical equation.Hypoxanthine phosphoribosyltransferase 1 (Hprt1, AB, Rn01527840) was used as endogenous control for normalization.

Statistical analysis
Statistical analysis for all studied parameters was carried out using the Statistical Package for Social Sciences v. 16 (SPSS, Inc.Chicago, USA).Significance level was set at <0.05.Univariate analysis of data was carried out by applying measures of central tendency to describe quantitative variables, mean and standard deviation (SD).A one-way analysis of variance (ANOVA) was used to test for difference in the readings between the control, no treatment and treatment groups.When significance was found, it was followed by a post hoc test to determine significant inter-group difference.

Results
Isolated rat bone marrow MSCs were identified by their fusiform shape.Flow cytometry confirmed their identity by being CD29 + and CD45 -(Fig.1).
Labeling of MSCs with iron oxide nanoparticles was successful as detected by transmission electron microscope (Fig. 2).Viability of MSCs was not adversely affected by iron as proven by viability assay (Fig. 3).

Histology findings
The investigators succeeded in developing the desired pathology of cirrhosis (Fig. 4. C, D) and hepatocellular carcinoma (Fig. 4. E, F) in the rat model.This is in comparison to the normal liver shown in Fig. 4 (A, B).The liver restored its original histologic architecture after treatment with MSCs (Fig. 4. G, H), which shows apparently normal looking liver trabeculae and sinusoids with no evidence of preneoplastic or neoplastic cellular changes.

Magnetic resonance imaging findings
A sequential fall in the signal intensity of the liver was detected in the scan performed at 30 min intervals (Fig. 5).Given the paramagnetic properties of nanoparticles, there is a fall in signal associated with SPION-labeled MSCs reaching the liver.This fall in signal intensity was noted for a further 48 hours after which the signals increased The cells were CD29 + (A) and CD 45 -(B).MSCs were stained using Alexa Fluor 488 anti-mouse/rat CD29 antibody (A.Positive) and PE Anti-rat CD45 antibody (B.Negative).These molecules are surface markers for the characterization of MSCs.
(Fig. 6).This is most likely due to a combination of the nanoparticles excretion and the regeneration of liver parenchyma due the effect of the infused MSCs.A representative nodule was chosen in each of the scans, and its dimensions were measured over time using a vendor provided calipers tool with measurements in millimeters.The lesion showed a significant reduction in dimensions (Fig. 7).

Biochemical findings
Rats treated with MSCs showed a significant decrease in serum level of α-fetoprotein, and serum activities of ALT and AST after MSCs therapy, when compared to the diseased untreated group (Table 1).The same effect was observed with both labeled and unlabeled cells.However, AST activity and the level of α-fetoprotein remained significantly higher in the MSCs treated group when compared to the control group.

Molecular findings
Gene expression of survivin, cyclin D1 and beta-catenin were compared before and after MSCs therapy (Fig. 8).A decrease in the gene expression of cyclin D1 was observed with stem cell therapy.This decrease was not statistically significant when compared to the control and the no treatment groups.Gene expression of beta-catenin was significantly lower in the treatment group compared to the control and the no treatment groups.p-value (ANOVA ) 0.000* 0.000* 0.000* p-value significant at <0.05; *p-value significant between all tested groups; a p-value significant between control and the untreated group; b -p-value significant between untreated group and the treatment group; c -p-value significant between control and treatment group. .Pictures E and F are presenting hepatocellular carcinoma; there are areas of invasive cells assuming a trabecular pattern (E), whereas in picture D there are pleomorphic hepatocytes with hyperchromatic nuclei, prominent large nucleoli, scattered mitotic figures (arrow) and high N/C ratios (HE ×100 and ×400).Following treatment with MSCs (G and H), the liver shows apparently normal looking liver trabeculae and sinusoids with no evidence of preneoplastic or neoplastic cellular changes (HE ×200 -×400).

Discussion
This work aimed at investigating the therapeutic potential of MSCs on HCC in the rat model of liver cirrhosis/HCC.The work intended to document and correlate radiologic evidence with both histological and molecular evidence of the reported healing effect of MSCs.
MSCs were isolated from the bone marrow of male rats and propagated in culture.MSCs were identified by their morphology and flow cytometry.These results are in agreement with those of Abdel Aziz et al., who developed a rat model of cirrhosis and HCC. 13 Scientists isolated MSCs from the bone marrow of rats and were able to use them to treat both liver cirrhosis and HCC.
The MSCs were labeled with iron oxide nanoparticles, which did not affect their viability or proliferative ability.These results are in agreement with those of Wang et al., 2009. 14They found that growing Adipose tissue derived MSCs (ASCs) in SPION-containing medium did not adversely affect the survival rate (97-99%).The labeled cells continued to function normally and to express cell factors, and were able to trans-differentiate.The study concluded that SPION did not affect viability, trans-differentiation potential or cell-factor secretion of ASCs.The results also agree with those of Zhao et al., who transplanted SPION labeled MSCs via the rat tail vein and the portal vein into fibrotic rat livers, then visualized them by MRI. 19n this work, MSCs injection improved hepatic pathology as depicted by the diminishing cirrhosis and the signs of recovery in the affected liver.These findings agree with those of Abdel Aziz et al. and Zhao et al. 13,19 The biochemical, molecular and radiologic findings confirmed the recovery of the injured liver.The mean serum level of α-fetoprotein showed a significant decline in the animals that received MSCs therapy compared  to the no treatment group.Serum activities of ALT and AST significantly declined following MSCs therapy.These results are in agreement with those of Zhao et al. 19 Time course MRI was adopted to confirm the homing of MSCs to the injured liver, and to document their therapeutic effects by correlating the radiologic findings to the hepatic pathology.Figure 5 follows the labeled MSCs into the injured liver by MRI.There was an initial decline in signal intensity, coinciding with MSCs injection and continuing for 24-48 h.This was followed by a gradual increase in signal intensity till the animals were sacrificed.The decline in signal intensity could be explained by the homing of iron-labeled cells into the liver in the first 24-48 h following IV injection.The increase in intensity thereafter may be explained by the improvement in liver pathology associated with MSCs injection, as well as the clearance of iron from the injected cells, possibly through extrusion and biliary excretion or engulfment by hepatic Kupffer cells.Figure 7 is focused on measuring the size of a representative focal lesion in the liver and the progressive decline in its diameter associated with stem cell therapy.This confirms and correlates with the amelioration of hepatic pathology.
Bos et al. used magnetic resonance imaging (MRI) for tracking SPION-labeled MSCs. 20These labeled cells were transfused into the renal arteries and portal veins of rats.SPION labeling had no hazardous effect on the viability and differentiation of the injected cells.The liver showed diffuse granular appearance associated with stem cell infusion.The study concluded that MRI is important for monitoring of SPION-labeled MSCs in vivo, in the liver.The results also agree with those of Zhao et al., who used MRI to confirm that SPION-labeled MSCs colonized the liver. 19They concluded that in vivo MRI of SPIONlabeled MSCs can be used to trace real-time liver healing during clinical treatment following hepatectomy.However, they did not use the MSCs to treat cirrhosis or HCC.
Bayo et al. investigated the possible relation between MSCs, autocrine motility factor (AMF) and HCC. 15 The study investigated the role played by AMF in mediating the migration of MSCs, of various sources, towards human HCC.The AMF produced by HCC was found to induce the migration of different MSCs in vitro, and to enhance their metalloproteinase 2 activities.Stimulation of MSCs with recombinant AMF also induced the in vitro adhesion to endothelial cells and increased the attraction of MSCs towards HCC tumor cells.They concluded that AMF plays a role in MSCs recruitment towards HCC.Gene expression of survivin, cyclin D1 and beta-catenin were compared with and without MSCs therapy.Decrease in gene expression of cyclin D1 was observed with stem cell therapy, though it was not statistically significant when compared to both the control and the no treatment groups.However, the expression of beta-catenin was statistically significant when compared to both the control and the no treatment groups.Beta-catenin activation has been linked to the development of HCC. 21Therefore, in this work, the decline in its expression observed with MSCs treatment might be one of the mechanisms through which MSCs exert their potential anti-tumor effect.The results of cyclin D1 and beta-catenin are in agreement with those of Abdel Aziz et al., who showed that MSCs decreased the level of gene expression of several carcinogenic genes in liver tissues after HCC induction in rats. 13They concluded that the administration of MSCs in experimental HCC models has tumor suppressive effects as evidenced by the decline in expression of Wnt signaling target genes with subsequent improvement of liver function.
The results of this work may indicate that MSCs could be labeled with nanoparticles and traced in normal liver, cirrhotic liver, and liver with HCC in animals.Labeling of cells did not affect adversely cell viability and/or proliferation.MRI is an important monitoring tool for the homing and localization of MSCs in the liver.MSCs infusion in animal models of liver cirrhosis and/or carcinoma may prove to be useful in limiting the cirrhotic and neoplastic changes.It is recommended to repeat this experiment using larger animal models, followed by its application in clinical trials for verification of its therapeutic potential.

Fig. 5 .Fig. 6 .Fig. 7 .
Fig. 5. Sequential fall in the signal intensities at scan intervals of 30 min, denoting the paramagnetic effect of the nanoparticles

Fig. 8 .
Fig. 8. Gene expression of survivin, cyclin D1 and beta-catenin with and without MSCs treatment

Table 1 .
Comparison of mean levels of serum activity of alanine aminotransferase (ALT), aspartate aminotransferase (AST) and serum level of α-fetoprotein in the studied groups