Keywords

Cell Migration; Glioblastoma Mutiforme; Human Umbilical Cord Matrix-Derived Mesenchymal Cells; MRI.

Introduction

GBM is a primary malignant brain tumor, which is the most common and aggressive type of brain tumor in humans. Despite improvement in therapy techniques, patients have a short life expectancy (12-15 months) [1-3]. This poor outcome is mainly related to the inability in delivering therapeutic agents to the tumor [4]. It is a great challenge to 1) identify the new therapeutic strategies and agents that could penetrate well through the blood-brain barrier, 2) reach to the tumor mass, and yet cause minimal harm to normal cells [5]. In recent decades, new strategies such as gene therapy have been developed, but gene therapy is confounded by the inefficiencies of viral and adenoviral vectors [6, 7].

Another novel approach is the use of stem cells [8]. Recent evidences suggest that stem cells are capable of migrating to the vicinity of tumor mass [5, 9]. The tropism of stem cells towards the cancerous cells has been reported in many types of cancers such as brain, breast, lung, and ovary tumors [8, 10-14]. Hence, stem cells are great options in the therapy methods of tumors because of their inherent tendency to migrate to the site of injury, ischemia, and tumor [15, 16]. Aboody et al [8] demonstrated the tropism of the neural stem cells into glioma. When NSCs were implanted intracranially, they migrated through the normal parenchyma and the primary tumor bed toward tumor mass and tumor microsatellites. This feature of NSCs led to study the migration behavior of other stem cells in order to develop tumor targeting and therapy strategy by stem cells. Researchers have shown that mesenchymal stem cells are capable of migrating to the tumor cells. Moreover, MSCs are well suited in cell therapy procedures because they are easily accessible and can expand in culture [17-19]. In addition, they are capable of differentiating into neurons, astrocytes, and other types of neuronal lineages [20-22].

Human umbilical cord matrix is an extraordinary source of stem cells for clinical applications [23]. Furthermore, hUCMs are easily obtained from delivered offspring, and have the plasticity for differentiation into adipocytes, chondrocytes, osteocytes, and also neuronal lineages [24, 25]. Tropism of hUCMs has been reported in animal models of ischemia and injury [16, 26]. However, no report has shown the migratory capacity of these cells toward the tumor mass. The aim of this study was to evaluate the migration tendency of hUCMs to tumor mass and their effects on tumor progression in an animal model of brain cancer. To assess this, Magnetic Resonance Imaging, H&E, and immunohistochemistry studies have been applied in this study.

Methods and Materials

Collection and Culture of hUCMs

Human umbilical cord was obtained from consented healthy donor. It was delivered to the laboratory in sterile PBS and %1 Penicillin/ Streptomycin. The procedure for harvesting hUCMs from umbilical cord was is described previously [27]. Briefly, the matrix was cut into 2-3mm pieces and the pieces were put onto 3 cm petri dishes and culture medium (DMEM, %10 FBS, %1 Penicillin/ Streptomycin, 0.0025μg/ml Amphotripcin B) was added. The harvested cells were detached by Trypsin/EDTA and were cultured in 25 cm2 flasks. The hUCMs from the 3rd passages were used for experiments.

hUCM characterization

Flowcytometry: The method for hUCM cell markers analysis was obtained from Seyedi et al., [28]. Briefly, CD34, CD73, and CD90 markers were evaluated by a FACS Calibur flow cytometer with Cell Quest software (BD Bioscience). Data were analyzed using WinMDI2.9 software.

Adipogenic and Osteogenic differentiation: Adipogenic and osteogenic protocols for differentiation of hUCMs have been described elsewhere [29]. Briefly, 2-3 ×104 viable hUCMs were seeded on clean, sterile coverslips and fed by adipogenic and osteogenic inducing medium for 21 days. Control cells were cultured in convention culture medium without the inducing agents. After 21 days, the Oil-red and Alizarin red staining were used for detection of fat droplets and Calcium-phosphate precipitates in the induced cells.

U87 Cells labeling

U87 cell line was purchased from the Pasteur Institute, Tehran, Iran. The cells were cultured in DMEMF12 with 10% FBS and incubated in a humidified atmosphere at 37°C with %5 CO2 in the air. At confluence of %80-90, they were detached from the substratum; counted, and used for the experiments or cryopreserved with 10% DMSO for further use. The PKH26 (red dye) was used to label the U87 cells. The cells were labeled according to the manufacture's protocol. For this reason, the cells were trypsinized; centrifuged, counted, and about 107 viable cells were exposed to PKH26 as recommended by the manufacturer. PKH26 positive cells were examined for their fluorescence emission by an inverted microscope equipped with fluorescence apparatus (Olympus, Japan).

Animals

Eighteen male Sprague-Dawley rats weighing 220- 250 g were used for the experiments. The rats were housed in the animal room, with 12/12 h day/night cycle, free access to rodents chew and drinking water. They were randomly divided into Tumor, hUCM groups of 7rats, and PBS group of 4 rats.

Implantation of U87-PKH+26 cells

Animals were anesthetized intraperitoneally using a mixture of 100 mg/kg ketamine and 10 mg/kg xylazine. Head of the animals got shaved and fixed in a stereotaxic frame. The scalp was incised in the midline of the skull. A hole with the coordinate; anterior = 1mm, lateral = 2mm of the Bergma was drilled in the right half of the skull. The Hamilton syringe was fixed on the arm of the stereotaxic frame. For tumor and hUCM groups, the needle was inserted to the depth of 3-4 mm, and 10 μl of cell suspension (106 cells) was injected slowly (1μl/1min). After 5 minutes, the needle was withdrawn gently. The hole was covered with bone wax and the incision was sutured with 0-4 suture. The same procedure was carried out for PBS group, except that the cells were removed.

Determination of tumor by MRI

MRI was performed on the animals 10 days after tumor inoculation. The animals were anesthetized by intraperitoneal injection of ketamine and xylazine. The rats were positioned head-up in the knee coil of 1.5-Tesla unite (AVAN TO Siemens imaging system, Germany), and T1-weighted spin-echo images (TR= 400 ms, TE= 19 ms, slice thickness=2 mm), T2-weighted spin-echo images (TR= 2000 ms, TE= 72 ms, slice thickness=2 mm) were obtained.

hUCM transplantation

hUCM cells were injected into the heart of the hUCM group animals, 10 days after implantation of tumor. To visualize transplanted cells, PKH67 (green dye) was used for labeling as recommended by manufacturer. The animals were anesthetized with ketamine/xylazine; the skin of the thorax was shaved, and a 1cm oblique incision was made in the left half; 2 cm lateral to the midline. The pectoralis major and external intercostal muscles were cut in the fifth intercostal space. Afterward, 2×106 hUCMPKH67+ cells suspended in 100 μL PBS was injected into the left ventricular by insulin syringe. Injections were done manually over 3 minutes, and the incision was sutured with 0-4 suture. 17 days after implantation of tumor, the rats were anesthetized, and were perfused with 4% paraformaldehyde .The brains were post-fixed by 10% buffered formalin for 24 h and were immersed in 30% sucrose in PBS for 48 h.

H & E and Immunohistochemistry Staining

For Hematoxylin-Eosin and immunohistochemical staining, formalin-fixed, paraffin-embedded 3 μm tissue sections were mounted on Plus slides. Following deparaffinization and rehydration, numbers of sections were incubated with primary antibodies as follows; monoclonal mouse anti-human Ki-67; 1:100 (Dako, Carpinteria, CA), monoclonal mouse anti-human Nestin; 1:100 (Santa Cruz Biotechnology), PECAM-1 monoclonal mouse anti-rat CD31; 1: 100 (a Bio-Rad), monoclonal mouse anti- human antibody GFAP; 1:100 (Santa Cruz Biotechnology) all diluted in PBS. Slides were incubated with biotinylated secondary antibodies against rat and rabbit followed by incubation with the streptavidin and biotinylated peroxidase complex (Dako REAL™ EnVision™ Detection System, Peroxidase/DAB, Rabbit/Mouse). Sections were counterstained with hematoxylin and mounted by a coverslip. The proportion of Ki-67, GFAP, CD-31, and Nestin positive cells was calculated as the percentage of the stained cells in 200 cells. Some of the sections were stained with H&E for histologic evaluation of tumors.

To detect PKH+cells in the brain tissue, the brains (n=3) was immediately frozen in Tissue-Tek OCT embedding gels. Cryostatgenerated 8-10 μm sections were mounted onto Superfrost Plus slides (Fisher) and examined under a fluorescence microscope (IX 71, Olympus, Japan) equipped with a digital camera.

Statistical Analysis

Data are expressed as mean ± SEM. Distribution of data was analyzed by the Kolmogorov-Smirnov test. Statistical differences were analyzed with the use of one way ANOVA test with IBM SPSS Statistics version 21. A P value < 0.05 was considered to detect a statistically significant difference.

Results

hUCM characterization

Flowcytometric analysis revealed that hUCM cells were positive for the typical mesenchymal markers CD73 and CD90, but negative for the hematopoietic marker CD34 (Figure 1 A). The induced hUCMs were capable of differentiating into osteocytes (Figure 1 B) and adipocytes (Figure 1 C) lineages.

×

Figure1:

Close

Figure1. A-Flow cytometry analysis of hUCM cells. Red histograms display the isotype control-stained cells, and the black histograms show the antibody stained. B- Oil red staining exhibited lipid droplets in the cytoplasm of the induced cells. CAlizarin red staining, displayed Ca2+ aggregation in the extracellular matrix of induced cells.

MRI

MRI was performed in groups to assess cerebral injury and tumor development. It, also, performed on the 10th day of experiments using T1 and T2-weighted MRI without contrast enhancement.

On T2-W, images a hyperintense spot was detected in the white matter of the right frontal lobe without obvious damage. On coronal T1-W images, the spot was iso- to hypointense in the tumor groups. Any sign of cerebral injury like edema, cyst formation and hemorrhage were observed in the animals of the tumor and PBS groups (Figure 2 A and B).

×

Figure2:

Close

Figure 2. A, B- The magnetic resonance images of both groups.

Histopathology

To assess whether transplanted hUCM cells could migrate towards the tumor mass frozen serial sections were examined with a fluorescence microscope. Tumor area was clearly seen in the slides (Figure 3 A-B). hUCM-PKH67+ cells were detectable inside and vicinity the tumor mass (Figure 3 C-D). Normal parenchyma far from the tumor was devoid of any detectable fluorescence signals.

H& E staining in U87 inoculated animals revealed that glioma masses were stained strongly with hematoxylinin compared with the normal parenchyma. The presence of neoplastic characteristics such as mitotic activity, migration of tumor cells, neovascularization, necrotic clusters, pseudopalisading cells, and hemorrhage was observed in tumor and hUCM groups (Figure 4 A-B). Interestingly, hemorrhage was very intense in hUCM treated animals compared with the tumor group (Figure 4 B). In PBS group, infiltration of immune system cells, and dead blood cells were observed in the injection area (Figure 4 C). For neo-vascularization determination, 10 H& E sections were investigated randomly, and new vessels were counted at 200 magnification. No significant difference was found between the hUCM and tumor groups in angiogenesis. While, both groups had statistically significant difference with the PBS group (Figure 4 D-F and 5 A).

Immunohistology

GFAP: GFAP positive cells were present in all groups (Figure 6 A-C). Interestingly, GFAP expression was lower in hUCM group, and depicted a significant difference with other groups. The rate of GFAP positive cells was significantly lower in the PBS group than in the tumor group (Figure 5-B).

Ki-67: The proportion of Ki-67 positive cells in the tumor group was significantly higher than other groups (P< 0.01, Figure 5-B). In addition, hUCM group had significantly higher Ki-67 positive cells than the PBS group. The increase of the Ki-67 positive cells revealed an increase in mitotic activity in tumor mass (Figure 6 D-F).

Nestin: The rate of Nestin positive cells was 9.75% in the tumor group. That it was significantly higher than other groups (Figure 5-B). There was also no significant difference between the PBS and hUCM group (Figure 6 G-I).

CD-31: The tissues of the tumor group exhibited more than %5 CD31 immunostained cells (Figure 5-B). It was also detectable in hUCM group, but no statistical difference was observed between the groups (Figure 6 J-K).

×

Figure3:

Close

Figure 3. Frozen sections of tumoral tissues were shown in the groups of tumor implantation A, B– Frozen sections were seen with green filter for tracking PKH-26 positive tumor cells. Frozen sections were counterstained with hoechst. Merge of images for detection of areas of tumor ×200. C. Frozen sections were seen with blue filter for tracking PKH-67 positive hUM cells. Frozen sections were counterstained with hoechst. Merge of images for detection of PKH-67 positive hUM cells ×200. D. Merge of the frozen sections of hUMC group that were seen with green and blue filter. E. Frozen sections were seen with blue filter for tracking PKH-67 positive hUM cells in the vicinity of tumor mass. Frozen sections were counterstained with hoechst. Merge of images for detection of PKH-67 positive hUM cells in the vicinity of tumor mass ×200.

×

Figure4:

Close

Figure 4. H&E sections of groups. A- The area of tumor in the tumor group (black arrow) × 100. B- The area of tumor in vthe hUMC group with hemorrhage (black arrow) × 100. C- Injection site of PBS (black arrow) ×100. D- Angiogenesis in tumor mass in the tumor group (black arrow) × 200. E-Angiogenesis in tumor mass in the hUMC group (black arrow) × 200. F- Angiogenesis in injection site of PBS in the PBS group (black arrow) × 200.

×

Figure5:

Close

Figure 5. A- Angiogenesis analysis between groups. B- Immunohistology analysis of Nestin, CD-31, Ki-67, and GFAP in groups. Data are expressed as mean ± SEM. *P < 0.05, **< 0.01.

×

Figure6:

Close

Figure 6. Immunostaining of groups. A- GFAP positive tumor cell in the tumor area in the tumor group (black arrows) × 200. B- GFAP positive tumor cell in the tumor area in the hUMC group (black arrows) × 100. C- GFAP positive cells in the PBS group (black arrows) × 200. D- Ki-67 positive tumor cell in the tumor area in the tumor group (black arrows) ×320. E- Ki-67 positive tumor cell in the tumor area in the hUMC group (black arrows) × 200. F- Ki-67 positive tumor cell in were present in the site of injection in the PBS group (black arrow) × 200. G- Nestin positive tumor cell in the tumor area in the tumor group (black arrows) × 320. H- Nestin positive tumor cell in the tumor area in the hUMC group (black arrows) ×320. I- Nestin in PBS group; a few Nestin positive were detected in the injection area (black arrows) × 200. J- CD-31 positive cells in tumor area in the tumor group (black arrows) ×200. K- CD-31 positive cells in tumor area in the hUMC group (black arrows) × 200.

Discussion

Human umbilical cord matrix-derived mesenchymal cells (hUCM) have been recognized for their capacity to differentiation into a neural phenotype in vitro[23]. Transplanted hUCMs showed remarkable ability to migrate toward pathological areas such as ischemia and injury [16,26,30]. Recently, evidences have shown the ability of hUCMs in wound healing after spinal cord injury [31]. We could show the potential of hUCMs migration toward the tumor area and their positive role in tumor regression. The surface makes analysis of hUCMs and their ability to differentiate into adipogenic and osteogenic lineages showed their mesenchymal origin and potential, which was consistent with the previous studies [28,29]. MRI was employed for determination of tumor in the brain of animals. Tumor growth was monitored by 1.5 Tesla MRI system, and the tumor mass was observed 10days after tumorimplantation. Obtained data were similar to the other reported data by using 1.5 T MRI systems [32].

Specifically, after intravascular injection of hUCMs, the migration of hUCMs in tumor-bearing rats was investigated. The results showed the weak presence of hUCMs in the tumor area. Their data are in agreement with the data presented by other researchers, who found low tropism of human mesenchymal stem cell in the tumor area after cell transplantation via tail vein injection [8,33]. Our primary results (data not shown) indicated that via tail vein injection was not a suitable approach for migration of stem cells towards tumor mass, thus, we switched to another method of cell inoculation, which uses intracardiac route. We found that hUCMs specifically localized in the tumor mass, but not in the normal brain. Thus, our results suggested homing of hUCMs is not merely the result of cerebral blood flow patterns, rather is due to the inherent tendency of hUCMs for the tumor. Reports indicated that the tropism of stem cells toward tumor may be mediated by tumor microenvironment, because the tumor’s milieu expresses EGF, PDGF, FGF, and SDF-1a which EGF, PDGF, and SDF-1 enhanced migration of stem cells toward tumor [34-36].

Analysis of histological slides of the brain revealed a similar histopathological characteristic of human glioblastoma such as polymorphism, mitotic activity, invasion, necrosis, hemorrhage, and neoangiogenesis [3] in tumor and hUCM groups. In this study, Ki-67 antigen expression indicated that the proliferation of U87 cells was decreased in the hUCM group, which suggested the role of hUCM in the inhibition of tumor progression. Moreover, the rate of neoangiogenesis and the expression rate of CD31 were decreased in the hUCM group in comparison with the Tumor group, however no statistical difference was found between them. Thus, it can be concluded that the presence of hUCMs in the tumor mass had negative effects on tumor growth and led to inhibit of progression of glioma. Some reports confirmed the role of stem cells in the inhibition of tumor progression and angiogenesis [33,37,38]. While, some researchers reported that the presence of mesenchymal stem cells in tumor area led to further growth and increased rate of angiogenesis in the tumor [39].

Recently, the expression rate of Nestin has been used as a marker to determine GBM prognosis in cancerous patients. Nestin is usually expressed in the primary brain tumor as well as some areas in the brain such as subventricular zone [40]. Our data showed Nestin expression in tumor mass and periphery in tumor and hUCM groups. In addition, we observed a decreased expression of GFAP in the treated rats. However, the expression rate of GFAP in glioma-bearing rats was very intense in the borders of the tumor as reported earlier [41]. Interestingly, the decrease in GFAP expression was observed in hUCM group, which might be because of the low synthesis of GFAP after migration of hUCM toward glioma. Our IHC staining data indicated the positive role of hUCMs in the inhibition of tumor progression and angiogenesis. Nevertheless, more investigations about 1) the mechanisms of hUCMs migration to the tumor site, 2) the mechanisms of hUCMs in the inhibition of tumor progression and 3) the final fate of hUCMs after the migration toward tumor mass is needed.

Conclusion

The present study suggests that the human umbilical cord matrixderived stem cells have the ability of migration toward glioma mass. In fact, this is the first demonstration that human umbilical cord matrix-derived stem cells can penetrate through the bloodbrain barrier, and reach to the human glioma. This is important because the hUMCs may be a suitable cellular vehicle for delivery of therapeutic agents to sites of tumor types in human. In addition; our work suggests that the intracardiac injection was a suitable approach for transplantation of stem cells in brain cancer in comparison with intravascular injection.