Effect of Human Mesenchymal Stem Cells on the Growth of HepG2 and Hela Cells

Abstract

Human mesenchymal stem cells (hMSCs) accumulate at carcinomas and have a great impact on cancer cell’s behavior. Here we demonstrated that hMSCs could display both the promotional and inhibitive effects on growth of HepG2 and Hela cells by using the conditioned media, indirect co-culture, and cell-to-cell co-culture. Cell growth was increased following the addition of lower proportion of hMSCs while decreased by treatment of higher proportion of hMSCs. We also established a novel noninvasive label way by using internalizing quantum dots (i-QDs) for study of cell-cell contact in the co-culture, which was effective and sensitive for both tracking and distinguishing different cells population without the disturbance of cells. Furthermore, we investigated the role of hMSCs in regulation of cell growth and showed that mitogen-activated protein kinase (MAPK) and phosphatidylinositol 3-kinase (PI3K) signaling pathways were involved in hMSC-mediated cell inhibition and proliferation. Our findings suggested that hMSCs regulated cancer cell function by providing a suitable environment, and the discovery from the study would provide some clues for development of effective strategy for hMSC-based cancer therapies.

Introduction

MSCs have been defined as a major type of non-haematopoietic cells that display a potential to self-renew and to undergo multilineage differentiation due to the multi-potent capacity. Indeed, they are shown to differentiate into bone, connective tissue, cartilage, muscle, myocardium, and neuron under appropriate conditions (Caplan, 2007; Dai et al., 2005; Jiang et al., 2002). MSCs are a subject of of great interest for tissue restoration because they are known to be recruited to site of inflammation and damaged tissue, contributing to injured tissue repairing and regeneration. For example, MSCs are shown to be called to sites of injury or chronic inflammation like bone defects and sclerosis, where they replace damaged tissue and format fibrous scars (Gregory, 2005). These conditions are typically accompanied by the release of specific intervening molecules by hMSCs, or by cell interaction between MSCs and damage tissues. It has been shown that MSCs can produce a vast array of cytokines and extracellular molecules and express receptors and/or counterreceptors both for cell–cell and cell–matrix interactions (Son et al., 2006; Muller et al., 2001). More recently, MSCs have become a topic of great focus in relation to the cancer. Growing evidence suggests that MSCs home to not just injured tissues but tumors (Yang et al., 2008), where they have an important role in affecting the behavior of tumor cells. Furthermore, tumor cells may secrete factors that can activate signaling pathways that facilitate MSCs targeting to the tumors (Ringden and LeBlanc, 2005). Thus, the ease of MSCs isolation and the fact that they can home to wound or cancer sites make them a potential platform for cell-based targeting delivery.

Despite the potential use of MSCs as a therapeutic tool for regenerative or anticancer drug delivery, several reports raise the concern of their effect on tumor cells growth. It is clear that the MSCs microenvironment plays an important role in tumor growth. However, whether MSCs exert an inhibitory or promotional effect on tumor cells remains a controversy, hampering both MSCs basic research and clinical therapy. Several studies suggest the role of MSCs in tumor growth and progression (Houghton et al., 2004; Mishra et al., 2009; Xu et al., 2009; Sasser et al., 2007; Zhu et al., 2006; Karnoub et al., 2007). For example, it has been shown that the development of gastric carcinoma is associated with MSCs, which is mediated by the interaction between tumor cells and MSCs (Houghton et al., 2004). In another example, co-culture or co-injection of MSCs with osteosarcoma cells could promote proliferation of osteosarcoma cells and lead to an enhanced tumor growth of mice. Moreover, MSC-conditioned media also stimulate tumor cells proliferation (Xu et al., 2009), suggesting contact-independent interactions between MSCs and cancer cells. The activity of cell growth promotion might closely mimics that of wound healing and scar development that render tumor a ‘wound that never heals’ (Bissell and Radisky, 2001). In contrast to studies that MSCs can favor tumor cell growth, however, there is mounting evidence showing that MSCs have tumor-inhibitory effects. For example, MSCs could exert potent anticancer effect in a model of Kaposi’s sarcoma, and the mechanisms involved might be explained by downregulating Akt activity in tumor cells (Khakoo, 2006). Similarly, MSCs are implicated in the inhibition of tumors growth in other types such as hepatoma, lymphoma, and insulinoma cells (Zhu et al., 2009; Qiao et al., 2008; Yuan et al., 2006; Stagg, 2008, Omuro et al., 2003). The controversy that MSCs exhibit pro- or anti-proliferative effects on growth of tumor cells might be interpreted by the complexity of MSC source, the malignant cell type being studied, and the interaction fashion between MSCs and the tumor cells. In addition, the cell number of MSCs used might be associated with different effects on the tumor growth. Indeed, it has been reported that tumor growth is not affected by the co-injection of the same amounts of MSCs, but is increased in presence of 10-fold more MSCs (Hasebe et al., 2001). Therefore, it is better to understand the effect of tumor growth by MSCs before they are used as candidates in cell/gene therapies. In addition, the mechanisms underlying the interactions between MSCs and tumor cells still remain to be demonstrated.

In the present study, we investigated the effect of hMSCs derived from the bone marrow on growth of HepG2 and Hela cells by establishing conditioned media, indirect interaction (by a transwell), and cell-to-cell co-culture. Additionally, we developed a differential labeling using i-QDs for distinguishing cells population in direct contact co-culture system. Furthermore, we assessed the role of hMSCs in exhibiting promotional or inhibitive effect on cancer cells by examining the involvement of MAPK and PI3-K signaling pathways, with the hope of better understanding the mechanism of the interaction between hMSCs and cancer cells for development of improved therapy in cancer.

Materials and Methods

Reagents and chemicals

Dulbecco’s modified Eagle’s medium (DMEM) was obtained from Wako (Wako Pure Chemical Industrial LTd., Japan). Fetal Bovine Serum (FBS), trypsin, RnaseA, and propidium iodide were from Sigma-Aldrich (Louis, MO, USA). Antibiotic–antimycotic was from Gibco (Grand Island, NY, USA). Polyclonal antibodies against ERK1, phosphor-ERK1/2 (Thr202/Tyr204), phosphor-Akt (Ser473), and Cyclin D were purchased from GenScript (Biology CRO, NJ, USA). Horseradish peroxidase (HRP)-linked antibody IgG was from BioVision Resarch (CA, USA). Monoclonal antibody against GAPDH was purchased from Cell Signaling Technology (Cell Signaling Technology Inc., Swiss). Antibody conjugation kit and Quant-iTTMPicoGreen ds DNA assay Kit were from Invit-rogen (Carlsbad, CA, USA). Transcriptor First Strand cDNA Synthesis Kit, FastStart DNA Master SYBR Green I Kit, and Rneasy Mini Kit were from Roche Diagnostics (GmbH, Mannhem, Germany). Amersham ECL Plus western blotting reagents, hybond-C nitrocellulose membrane, trypan blue, and agarose were from GE Healthcare (Bio-Sciences AB, Sweden). PrimeSTAR HS (Premix) agent was from Takara (Takara Biotech., Shiga, Japan). Other reagents were of analytical grade. All buffers are prepared with Milli-Q deionized water.

HMSCs (passage 3, derived from the bone marrow) were purchased from Lonza Walkersville (Walkersville, MD, USA). Human hepatoblastoma cell (HepG2) and cervical carcinoma cell (Hela) line were from Heath Protection Agency of Culture Collection (ECACC).

Cell culture, conditioned media treatment, and CCK-8 assay

HMSCs were cultured with standard medium consisting of 10% (v/v) FBS and 1% antibiotic-antimicotic (Ohyabu et al., 2009). Cells were incubated at 37°C in a humidified atmosphere of 5% CO₂. For preparing conditioned media, hMSCs were grown to 70–80% confluence and the conditioned media were collected by filtering. Fresh medium was mixed with conditioned media to form various percentages (ranging from 0 to 100%). The conditioned media were stored at −80°C until use.

HepG2 or Hela cells were seeded onto 96-well plates (100 mm²) and incubated for 24 h. Then cells were exposed to a series of percentage of the conditioned media for 3 and 5 days. Cells treated with fresh medium were used as the control. The effect of conditioned media on the cell viability was evaluated by the CCK-8 assay. Briefly, after the treatment, 10 μL of CCK-8 reagent was added to each microculture well, and plates were incubated for 2 h at 37°C, after which absorbance at 490 nm was measured using a microplate reader (Molecular Devices, Japan). Results were expressed as the percentage of the control at completion of each incubation period.

Indirected co-culture by a transwell system

HepG2 or Hela cells were cultured in 12 well-plates and hMSCs were seeded in a filter (0.4 μm pore size, ThermoFisher Scientific, Roskilde, Denmark) for up to 24 h. Then the filters were inserted in the wells and co-incubated for 3 and 5 days. The ratio of number of hMSCs vs cancer cells added was 1:10, 1:1, 2:1, respectively. Cancer cells without hMSCs for culture were used as the control. Following the treatment, the effect of the co-incubation on the growth of cancer cells was assessed by picogreen assay. Briefly, cells were extracted with lysis buffer (100 mM Tris-HCl, PH 7.5, 5 mM MgCl₂, 0.2% Triton-X100), ultrasonicated for 4×5 s on ice, and collected at 6000 g for 10 min at 4°C. After the addition of equal volume of picogreen solution to the extraction, the fluorescent absorbance (ìex 480 nm, ìem 520 max nm) was measured using a spectrofluorometer (MolDev, Sunnyvale, USA). The results were expressed as the percentage relative to the control.

Trypan blue assay and microscope observation

In parallel with the CCK-8 assay, trypan blue assay was also used to detect the effect of conditioned media on cancer cell viability using a Coulter counter (Invitrogen, Carlsbad, USA). Microscopic observation of cell morphology was examined with an inverse microscope (Carl Zeiss, NY, USA). The images were acquired with a cool digital camera (Coolsnap QH, USA) and analyzed with the MetaMorph software package.

Cell cycle analysis

Synchronized cells were incubated with various percentage of conditioned media as described above, and cell cycle status was measured by using propidium iodide staining as described (Fu et al., 2009). Briefly, cells harvested were washed, fixed using 70% ice-cold ethanol for 20 minutes at −20°C, and then resuspended in the staining buffer (50 μg/ml) in the presence of RNase A (50 g/ ml) for 30 min at 37°C in the dark. Cells were re-suspended in PBS and the cell cycle analysis was performed with a flow cytometer (Becton Dickinson, CA, USA). To assess cell phase distribution, FlowJo software (Ashland, Oregon USA) was used to calculate the percentages of cells in G0/G1, S, and G2/M phase. Quantitative PI was according to the equation PI=(G2/M+S)÷(G0/G1+S+G2/M)×100% (Plas et al., 2001).

Cell–cell contacting co-culture between hMSCs and cancer cells

For labeling of hMSCs or cancer cells, mortalin antibody-based differential i-QDs were used. i-QDs were prepared by conjugation of Qdot 655 or Qdot 565 with an internalizing anti-mortalin antibody, using an antibody conjugation kit according to the manufacturer’s instruction. The labeling of hMSCs with i-QDs (Qdot 655) was conducted as described previously (Ohyabu et al., 2009). For generating i-Qdot labeled cancer cells, QD-antibody conjugates (Qdot 565) were added to cell culture in time- and dose-depended fashion to ensure the optimal staining of cells. The labeling efficiency was examined based on fluorescence image detection or flow cytometry analysis. For fluorescent microscope observation, i-QD-labeled cells were rinsed twice with PBS and fixed with a 4% paraformaldehyde in PBS for 1 h at 4°C. Cell labeling was examined with a microscopic system (Carl Zeiss, NY, USA) equipped with a specific Qdot filter set (Omega Optical, Brattleboro, VT). Labeling efficiency was also analyzed with a flow cytometry (BD BioSciences, CA, USA). Cells at different time intervals were detached by digestion, suspended in PBS containing 2% FCS, and subjected to flow cytometric investigation. QD fluorescence labeling was on the basis of blue laser excitation, and change in the measurements (Laser Octagon 488, Lp Mirror 565, BP Filter 695/40) was examined.

For co-culture of hMSCs with cancer cells, QD-labeled hMSCs and cancer cells were mixed with a different ratio (the number of hMSCs vs cancer cells was 1:10, 1:1, 2:1, respectively). The mixtures were co-cultured by plating the cells on coverslips (Iwaki, ASAHI Glass Co. LTD., Japan) or 6 well-plates (Falcon, NJ, USA). Microscopy analysis was performed by mounting the glass slides followed by examination with a microscope equipped with differential Qdot filters. Fluorescent images were acquired using an attached AxioCamMRmmonochrome charge-coupled device camera (Carl Zeiss, NY, USA). Distinguishing between cells populations was performed with fluorescence-activated cell sorting (FACS) (BDFACSAria Cell Sort, Becton Dickinson, NJ, USA). Both hMSCs (Qdot 655-labeled, Lp Mirror 655) and cancer cells (Qdot565-labeled, Lp Mirror 556) were recognized based on differential fluorescent labeling distribution.

Western blotting analysis

The western blotting was performed as described previously (Long et al., 2011). Briefly, cell lysates (prepared in 10 mM Tris-HCl pH 7.5, 150 mM NaCl, 10% glycerol, and 1% Triton X-100 containing 1 mM proteinase inhibitor PMSF) were separated by SDS-PAGE using a PROTEAN electrophoresis system (Bio-Rad Laboratories, Inc., USA), and transferred onto nitrocellulose membrane using a Semi-Dry Transfer Blot (Bio-Rad Laboratories, USA). The membranes were then incubated with the primary antibodies, followed by probing with HRP-conjugated IgG secondary antibody. GAPDH was used as sample loading protein. The blots were visualized using ECL-Plus according to the manufacturer’s instruction, and scanned by a Model 300A densitometer (Molecular Dynamics, CA, USA).

Real-time quantitative RT-PCR analysis

For total RNA isolation, cells were extracted using the RNeasy Mini kit following the the manufacturer instruction. To ensure the RNA samples free of contaminating genomic DNA, all samples were treated with DNase on column using a DNA-free kit. RNA quality was determined by the measurement of absorbance at 260 and 280 nm. The concentration of the RNA was calculated from absorption measurements at 260 nm with a Nanodrop ND-1000 spectrophotometer (Scrum Inc., Japan). All buffers were free of RNase.

Reverse transcription was performed using Transcriptor First Strand cDNA Synthesis Kit and PCR amplification was performed using LightCycler FastStart DNA Master SYBR Green I kit according to the manufacturer’s protocols. The real-time PCR was run in a Light-Cycler instrument (Roche Diagnostics, GmbH, Mannheim, Germany). Primer sets for PCR were the following: AKT (Forward primer: 5′-gcagcacgtgtacgagaaga-3′, Reverse primer: 5′-ggtgtcagtctccgacgtg-3′), ERK1 (Forward primer: 5′-tacaccaacctcgtacatcg-3′, Reverse primer: 5′-catgtctgaagcgcagtaagatt-3′), ERK2 (Forward primer: 5′-tcacacagggttcctgacaga-3′, Reverse primer: 5′-tgcagcctacagaccaaatatca-3′), and GAPDH (Forward primer: 5′-cga-catcaagaaggtggtga-3′, Reverse primer: 5′-ccagcatcgaaggtagagga-3′), produced by Custom Primer Greiner Bio-One Co., Japan. GAPDH was used as an internal standard. Cycling program was pre-incubation at 95°C for 10 min, followed by 10s at 98°C, 5 s at 60°C, and 15 s at 72°C for 40 cycles. The expression of different genes was first normalized to GAPDH and then relative to the control. The abundance difference was expressed as fold change.

Statistical analysis

The results were expressed as the arithmetic mean±standard deviation (SD) for at least 3 repeated individual experiments for each sample group. Statistical differences between the values were determined by Student’s t test, with a value of P<0.05 being considered statistically significant.

Results

Culture of cancer cells with hMSC-conditioned media

To determine whether the cultured hMSCs release soluble factors that can affect cancer cells growth, we cultured HepG2 or Hela cells in hMSCs-conditioned media. Cells were exposed to different percentage of conditioned media for 3 and 5 days and subjected to CCK-8 assay. We found that conditioned media derived from hMSCs could inhibit or promote cell growth relative to control cells (P<0.05, Student’s t test, Fig. 1). Conditioned media added to cancer cells had effects ranging from 19.8±1.1% and 12.4±0.5% stimulation to 30.4±1.5% and 22.6±1.8% inhibition for HepG2 or Hela cells, respectively. The increase in cell growth was observed at lower percentages of the conditioned media while higher percentages of the conditioned media led to the suppression in cell growth. The results were generally in agreement with those obtained by Trypan Blue assay, in with the conditioned media exerted a stimulatory or inhibitive effect on cell growth (data not shown). This suggested that the growth of cancer cells were regulated by hMSCs-conditioned media as a result of soluble factors derived from hMSCs in the culture.

Fig. 1

Effect of the conditioned media on growth of cancer cells (A: HepG2, B: Hela) as detected by CCK-8 assay. Conditioned media were prepared by culturing hMSCs to 70–80% confluence, and then conditioned media were collected and mixed with fresh medium to form a series of percentages. Cells were exposed to various percentages of the conditioned media for 3 and 5 days. Cells treated with fresh medium were used as the control. Shown on x-axis represented the hMSCs percentage obtained from pre-established conditioned medium, and shown on negative and positive y-axis represented relative change of cell growth promotion and inhibition, respectively.Data shown in the bars were the mean±the standard deviation of at least three independent experiments (x±SD, *P<0.05, **P<0.01, with significant difference compared with the control using independent two-sample Student’s t test).

We further investigated the effect of the conditioned media on cell cycle progression, as shown in Fig. 2. We showed that quantitative PI (Quantitative PI was according to the equation PI=(S+G2/M)/(G0/G1+S+G2/M) (Plas et al., 2001) determined by flow cytometry analysis) increased in cells treated with lower percentages of hMSC-conditioned media, while it decreased at higher percentages of the conditioned media with an apoptotic peak occurring (Fig. 2A and B, left). Furthermore, the increased or decreased growth for cancer cells was associated with G0/G1-phase entry during cell cycle progression (Fig. 2A and B, right). This suggested that the proliferative or suppressive effect on cell growth by the conditioned media was shown to participate in cell cycle regulation, giving rise to an increased or decreased G0/G1 phase rate and, hence, to decreased or increased DNA synthesis.

Fig. 2

Effects of the conditioned media on cell cycle progression in cancer cells (A: HepG2, B: Hela). Synchronized cells were incubated with different percentage of the conditioned media and cell cycle status was measured by flow cytometry. Upper-panel for A and B: Typical image and the analysis of the cell cycle distributions. Lower-left panel: Quantification of the PI (PI=(S+G2/M)/(G0/G1+S+G2/M) of cells in cell cycle. Lower-right panel: Percentage of cells in G0/G1 phase was indicated. The plots shown were representative of three independent experiments.

Indirect cell co-culture between hMSCs and cancer cells using a transwell

To examine the co-inductive effect of indirect cell-cell contact on cancer cell growth, we established a transwell system, in which the two cell populations were cultured on their own by a separated permeable membrane, which precluded direct cell-cell contact but allowed communication via soluble factors. Different number of hMSCs proportional to cancer cells was added and the co-incubation was continued for 3 or 5 days. Fig. 3 showed the effect on cell growth following the indirect co-culture. We found that cancer cells treated with lower number of hMSCs showed an increase in the growth whereas high number of hMSCs suppressed cell proliferation (P<0.01, Student’s t test). The inhibitory or stimulative effect was from 48.4±2.2% and 36.2±1.9% stimulation to 38.4±1.5% and 32.2±1.4% inhibition for HepG2 and Hela cells, respectively, when fewer or higher number of hMSCs (proportionable to cancer cells) was added. This suggested that the indirect co-culture could result in increased or decreased cell growth. Moreover, inhibitive or profiliterative induction by the presence of hMSCs in the non-contact co-culture was higher than that obtained from the conditioned media, suggesting that the ability to inhibit or stimulate cell proliferation resulted from the additional cooperative contribution between the two cell types in a transwell system that provided the cancer cells whith a suppressive/proliferative advantage.

Fig. 3

Effect of hMSCs on cell proliferation by non-contact co-culture (A: HepG2, B: Hela). HMSCs pre-cultured were inserted into the well (seeded with Hepg2 or Hela cells) and co-incubation was for 3 and 5 days. Different ratio of hMSCs vs cancer cells was added, and cancer cells not receiving hMSCs for co-culture were used as the control. Shown on x-axis represented the ratio of hMSCs vs cancer cells by a indirect transwell co-cuture, and shown on negative and positive y-axis represented relative change of cell growth proliferation and inhibition, respectively. Differences between values were tested using the independent sample Student’s t test (compared with the control, x±SD, n=3, *P<0.05, **P<0.01).

We observed that the no-contact co-culture caused hMSCs to undergo morphological change under microscope observation, as shown in Supplementary Fig. 1 (upper panel). HMSC cells propagated in indirect co-culture with cancer cells tended to grow in a distinct more-rounded, flat shape of cells, as compared with cells in that from a culture without cancer cells, which grew as more a spindle, fibroblast-shaped population with the antennae from cells more evident. This suggested that the interactive induction play a role in hMSCS’s morphologic induction. Moreover, we noted that when cancer cells cultured above were detached and re-seeded in hMSCs-free media with the same number, the cell proliferation displayed enhanced or reduced potential (lower panel). This suggested that the cancer cells can be ‘educated’ by hMSCs to obtain an increased or decreased growth ability during co-culture by an active non-contact communication.

Co-culture of hMSCs cells with cancer cells

Cell-to-cell contact is also believed to be crucial in affecting the behavior of cancer cells, since MSCs could interact with cancer cells both by direct contact via membrane receptors and by ligation of growth factors, cytokines, and extracellular matrix molecules (Muller et al., 2001). To characterize cell-to-cell co-culture mediating the interaction between hMSCs and cancer cells, we established a co-culture assay, in which the differential i-QD labeled hMSCs and cancer cells were co-cultured in the same dish, and then cell populations in culture system were distinguished from each other using fluorescent microscope and FACS. The use of the nanoparticle-based QD strategy as an efficient and sensitive tool for nontoxic, genetically noninvasive, and functionally inert cells labeling for long-time image observation has been reported (Babic et al., 2008; Edgar et al., 2006; Alivisatos et al., 2005). To ensure the labeling efficiency, cells were incubated with QD-antibody conjugates in concentration- and time-depended manners. As shown in Supplementary Fig. S2, cells took up nearly 100% of i-QDs after optimization detected by flow cytometry. The presence of fluorescent labeling was also verified by fluorescent microscopy detection, and time course of cell labeling in response to QD-antibody conjugates addition was shown in Fig. 4.

Fig. 4

Time course of cell labeling in response to i-QD addition. Cells were incubated with i-QD conjugates for different time intervals and the labeling efficiency was examined with fluorescence microscope. Upper panel: a–e corresponded to the overlaid/fluorescent images of different time periods i.e. 0, 2, 4, 6, and 8 h, respectively. Magnification 40X. Lower panel: cut view mode.

To determine effect of hMSCs on cancer cells behavior by co-culture, we admixed iQD-labeled hMSCs to cancer cells with different ratio for cell-cell co-culture. The representative profile for the differential i-QD expressing population of cells was shown following the co-culture (Supplementary Fig. S3). To distinguish hMSCs or cancer cells in the mixture at specific time points, each cell population was isolated by FACS based on differential labeling. We showed that hMSCs were completely devoid of HepG2 cells after sorting (Supplementary Fig. S4). Also the tracking of labeled cells was performed with fluorescent microscope when two cell types were co-incubated in the same dish. We observed that the co-culture caused the two types of cells to move up to each other for cell-to-cell interaction (Fig. 5). The cells exhibited a strong staining for overlapped fluorescence after cell-to-cell contact (Red: Q655-labeled hMSCs, Blue: Q565-labeled HepG2 cells, Purple: overlapped cells), an observation that was not seen when the cells were co-cultured on day 0, indicating an inductive interaction for motility and intercellular talk between hMSCs and cancer cells in co-culture system.

Fig. 5

Fluorescence images of differential i-QD-labeled cancer cells and hMSCs. Differential i-QD labeled HepG2 cells and hMSCs were mixed for co-culture for different time points, and fluorescent labels were observed by fluorescent microscope. HepG2 cells were detected in blue (Q565-labeled), and hMSCs detected in red (Q655-labeled). The purple on the panel represented an overlapping labeling. Upper-left panel: co-culture for day 0, and upper-right panel: co-culture for day 2. Magnification 40X. Lower panel: magnified images showing the area of the arrowhead indicated on panel.

Immunoblotting analysis

To investigate the ability of hMSCs to affect cancer cells’ functionality and possible molecular mechanism of the interaction between hMSCs and cancer cells, we conducted westernblotting analysis to test whether the mitogen-activated protein kinases (MAPKs) and phosphoinositol-3 kinase (PI3-K)/protein kinase B (Akt) transduction pathways are involved in government of cell growth by hMSCs, since MAPKs or PI3-K/Akt signal pathway mediates diverse biological functions, such as cell growth, survival, and apoptosis (Wada and Penninger, 2004; Chang et al., 2003). Cells were treated with hMSCs-conditioned media or exposed to hMSCs for co-culture with indicated percentages. As shown in Fig. 6, the levels of cyclin D, ERK1 (extra-cellular signal regulated kinase), phosphorylated ERK1 (p-ERK1), and phosphorylated Akt were elevated correspondingly by the treatment with lower percentages of the conditioned media, and all substantially decreased with higher percentages of conditioned media induction (Fig. 6A). Of note, the higher or lower levels of the proteins were also observed when lower or higher hMSCs were added to the culture (Fig. 6B). MAPKs or PI3K/Akt pathway is a cellular signaling mediated by diverse molecules, such as growth factors, cytokines, and other physiologic stimuli, which phosphorylates the substrates and regulates several cellular processes, including cell growth, differentiation, and apoptosis. It is also known that they have a role in the transcription, translation, cell cycle regulation, cell motility, and cancer malignance, and the dysregulation has often been implicated in tumor progression (Cuní et al., 2004; Itoh et al., 2002). The results that proteins typical of MAPK/ERK1 and PI3K/Akt pathways were regulated by the increase or decrease in expressions, suggesting the possible involvement of the pathways in regulating cell growth mediated by hMSCs.

Fig. 6

Western blot analysis of HepG2 cells exposed to conditioned media or hMSC cells. HepG2 cells were treated with different percentage of conditioned media or different ratio of hMSCs, and expressional levels of cyclin D, p-Akt, ERK1, and p-ERK1 were detected by western blotting assay. A: Cells treated with the conditioned media. a–f indicated the percentage of conditioned media added: control, 20, 40, 50, 80, and 100%, respectively. B: Cells after exposure to hMSCs. Control (left), 1:10, and 2:1 (Ratio of hMSCs vs cancer cells, middle and right), respectively. Total 100 μg proteins for each sample were loaded. GAPDH was used as the loading protein.

Real-time quantitative RT-PCR analysis

We performed real-time RT-PCR quantitative assay on cancer cells to identify the involvement MAPKs/ERK1 or PI3-K/Akt signaling pathway in hMSCs-induced cell proliferation or suppression by detecting the genes expression. Table I showed the expression levels of the selected genes following the exposure to conditioned media or hMSCs addition. Clearly, the expressions of genes were increased when lower percentages of the conditioned media was added while decreased at higher dosings of conditioned media. For example, HepG2 cells treated with 20% of conditioned media showed an increase in genes expression of AKT, ERK1 as well as ERK2 by 1.74±0.21, 1.75±0.19, and 2.87±0.34 fold, respectively, whereas 100% of conditioned media led to 2.77±0.18, 3.85±0.25, and 6.81±0.83 fold reduction in genes expression (Student’s t test, statistical significance beyond P<0.05). Additionally, higher or lower levels of the genes expression were also observed following the addition of hMSCs to the culture in HepG2 cells. The result based on the observation by employing the approach using real-time PCR assay was consistent with the westernblot analysis and suggested that hMSCs could activate MAPK and PI3-K/Akt cascade pathways for regulating cell proliferation and apoptosis.

Table I Induction of the selected genes in cancer cells by hMSCs determined by real-time PCR

Conditioned medium (%)	Genes			Coculture Ratios of hMSCs/cancer cells	Genes
	AKT	ERK1	ERK2		AKT	ERK1	ERK2
	HepG2					HepG2
20	1.74±0.21	1.75±0.19	2.87±0.34	1:10	3.68±0.13	3.29±0.04	2.38±0.10
80	−1.78±0.20	−2.64±0.31	−3.80±0.26	1:1	−1.78±0.19	1.27±0.11	−1.58±0.13
100	−2.77±0.18	−3.85±0.25	−6.81±0.83	2:1	−3.73±0.09	−3.94±0.11	−3.43±0.14
Hela				Hela
2	1.07±0.23	1.87±0.19	2.26±0.31	1:10	1.97±0.12	2.76±0.19	2.28±0.10
10	1.51±0.17	2.37±0.18	3.82±0.28	1:1	1.29±0.18	1.42±0.21	1.24±0.12
80	−2.96±0.21	−4.97±0.43	−5.28±0.62	2:1	−2.99±0.08	−2.46±0.15	−5.88±0.10

Real-time quantitative PCR were performed in the LightCycler System. Expressions of selected genes were normalized to house-keeping gene (GAPDH) and relative to the control. Data are expressed as mean fold change, with the standard deviation from at least 3 measurements. ‘+’ and ‘−’ indicated up- or downregulation in treated group versus the control.

Discussion

MSCs have received increased attention for their ability to interact with cancer cells and make a impact on cancer cells’ behavior. However, their potential to exhibit promotional or inhibitive properties by providing the cancer cells with a microenvironment remains controversial. Some reports suggested that they exhibit a potent promotional effect by providing growth signals, while others suggested they have potential role in inhibition of the proliferation of cancer cells. Additionally, the exact mechanism of their action in the regulation of cancer cells has not been well elucidated. In the present work, we investigated the effect of hMSC on growth of HepG2 and Hela cells by establishing conditioned media, indirect interaction, and cell-cell contact co-culture. We demonstrated the ability of hMSCs to inhibit or promote the growth of cancer cell in three experimental systems. Cell proliferation was increased when being treated with lower proportion of hMSCs while decreased with higher proportion of hMSCs. It is known that hMSCs secrete various growth factors, inflammatory cytokines, and chemokines, and express receptors and/or counterreceptors both for cell-cell and cell-matrix intervention (Muller et al., 2001; Yang et al., 2008). Therefore, it is likely that hMSCs co-culture with cancer cells could be acting as a ‘tumor promoter’ when fewer number of hMSCs or lower percentage of hMSCs-conditioned media were used in the culture, since MSCs secrete several growth factors or cytokines like platelet-derived growth factor (PDGF), epidermal growth factor (EGF), and fibroblast growth factor (FGF) etc, which promote the cell growth compared to that cultured in absence of hMSCs incubation. Moreover, the cooperative induction between hMSCs and cancer cells might be an additional contribution responsible for pro-proliferative effect on cell growth. On the other hand, high proportion of hMSCs could clearly exhibit a cell suppressive effect instead of proliferation stimulation that acts to decrease the proliferation. This suggested that bioactive molecules with inhibitory activities were predominantly after the addition of higher amounts of hMSCs. It was reported that tumor growth can acquire a variety of components through surrounding stromal such as MSCs with both pro- and anti-proliferative activities, contributing to mediating tumor cell growth through the interactions (Bissell and Radisky, 2001). It has been shown that low amounts of MSCs display stimulatory effect on T cell proliferation while high number of MSCs suppress T cells proliferation (LeBlanc et al., 2003). Our data demonstrated hMSCs have stimulatory or inhibitive effect on cancer cell growth. This distinctive property of hMSCs is probably related to the competition of production of a combination of cytokines with various effects in culture that display inhibitive or stimulational activities in cells.

Also we investigated the effect of hMSC on cell proliferation by adding conditioned medium of hMSCs to cancer cells cultured for 3 and 5 days (Fig. 1 and Fig. 3). We found that growth-promoting effect was marked on day 3 than on day 5, while growth-suppressing effect was higher following the prolonged exposure of cancer cells to either conditioned medium or hMSC cells. The observed time-dependent response of cancer cells to hMSCs in the culture suggested that hMSCs had a proliferative or inhibitory effect on cell growth. It is known that hMSCs secrete various cytokines, chemokines, extracellular matrix molecules and express receptors/counterreceptors to modulate the environment either through cell–cell or cell–matrix interactions. For example, Akt is activated by a diverse growth factors and other stimulus in a PI3K-dependent manner, and the activation of Akt results in activation of a wide range of their adaptors/receptors including metabolic enzymes, transcription factors, and cell cycle regulators, thus play an important role in cell growth, apoptosis, and cell-cycle progression (Vorotnikov, 2011). Therefore, it is likely that MSCs co-culture with cancer cells act to promote cell growth by establishing cell–cell or cell–matrix contacts with increased levels of stimulatory signals factors earlier, while long-lasting co-culture reduced receptors by occupying corresponding growth receptors necessary for proliferation stimulation, leading to a conversion from a stimulation to suppression of cell proliferation. Our findings demonstrate here that hMSCs have a stimulatory or suppressive effect on cell growth, depending on dose- and time-response. Data from this study might provide some clues for the antitumor activity with optimal proportion of MSCs to avoid untoward effects of administered MSCs.

We showed that non-contact co-culture in transwell caused a significantly higher stimulatory or inhibitory effect on cell proliferation than that obtained from conditioned medium. The cooperative induction of two cell populations in the transwell might play an important role in the differential regulation of cell growth. However, there might be several other possibilities, such as the difference in local concentration, or stability of soluble factors that provided cancer cells with a suppressive/proliferative advantage. For example, the sustained delivery of the various cytokines in co-cuture by a transwell system resulted in supplying the surrounding cancer cells with increased levels of these factors, which might in turn give the cells a proliferative or inhibitory induction. The inductive intercellular communication and continuous activation of hMSC’s function by the interaction between the hMSCs and cancer cells might also play a role in the regulation of cell growth. Thus, the differential effect of hMSCs on cell growth is probably controlled by external factors in the microenvironment, including cell–cell and cell–ECM adhesion, soluble cytokines, matrix molecules, and growth factors availability. Further functional investigation of the cellular and molecular events will enable the improved understanding of the mechanism of the action between cancer cells and hMSCs for development of effective therapy in the treatment of cancers.

In fact, we demonstrated here that the growth of cancer cells was affected by either adding the pre-established conditioned medium or cultured in transwell system through indirect cell–cell co-culture. The inhibitive or profiliterative induction of hMSCs was in dose- and time-dependent manners, indicating that hMSCs had the ability to stimulate or suppress cell proliferation in vitro. To investigate further whether hMSCs have the ability to promote or inhibit cell growth in case of co-culture system, we used established stable iQDs to label cells prior to co-culture, and measurement of differential iQDs-labelled cell populations at a specific time point in situ was determined by the FACS. Analysis of proportion of cells in co-culture showed that co-culture of HepG2 cells with hMSCs increased the percentage of cell population from 38.5±1.8% to 59.7±2.6% when lower hMSCs/cancer cells rate was used (The rate is 1/5, and 1/10, respectively. Relative percentage obtained from three replicate experiments by data normalization, as represented by mean SD). In contrast, the addition of higher proportion of hMSCs led to a suppression ranging from 20.2±1.2% to 38.0±1.6% on day 3 compared to that cultured in parallel without hMSCs, as shown in Supplementary Fig. 5. The screen for the levels of cell expressing population in the mixture showed that enhanced or decreased percentage of cancer cell population was apparently in presence of more or less hMSCs. This clearly indicated that cell proliferation could be regulated by hMSCs as a result of interaction between hMSCs and cancer cells. Consistent with these results, as revealed by western immunoblotting and RT-PCR analysis, hMSCs treatment caused a marked higher or lower levels of key molecules, such as cyclin D, Akt, ERK1, and phosphorylated Akt that were involved in MAPKs or PI3K/Akt pathway. Furthermore, the cell stimulatory or inhibitive effect revealed by quantitative PI analysis showed that the differential effect on cell proliferation is correlated with an increase or decrease in G0/G1 ratio and thus DNA synthesis. Collectively, these findings suggested that hMSCs may have a prominent role in regulation of cell proliferation as a result of the communication between hMSCs and cancer cells, including cell–cell and cell–ECM adhesion, cytokines, and growth factors and receptors. However, further study is still needed to understand the role of hMSCs-mediated cell growth for developing attracive cell-based cancer therapies in these cell types.

It is known that ERK1/MAPK and PI3K signaling cascades are involved in the cell growth, apoptosis, differentiation, and oncogenesis, taking into account that over-expression of the components of the pathways are observed in many types of cancers. Moreover, they are generally considered as necessities for cell cycle progression regulation (Cuní et al., 2004). For example, as an important component of signaling molecules, Akt (also known as protein kinase B), is a key enzyme and identified as one of the downstream targets of PI3K (Staal et al., 1977). It is known that Akt is implicated in the regulating tumor growth, and also plays a role in cell cycle regulation by participating in getting across cell cycle arrest in G1 and G2 phases (Thakkar et al., 2001; Graff et al., 2000). In the current study, we examined the molecular events in the interaction between hMSCs and cancer cells by performing western-blotting and real-time PCR assay. We found that cancer cells exposed to lower proportion of conditioned media or hMSCs showed an increase in the expression of key signaling proteins or genes involved in ERK1/MAPK or PI3K pathway. In contrast to this, cancer cells cultured with conditioned media or hMSCs in a lower proportion resulted in a decreased proteins/genes expression. Moreover, cell cycle analysis demonstrated that hMSCs-mediated cell growth gave rise to G0/G1-phase transition to S phase. This suggested that the activation of PI3K/Akt and ERK1/MAPK pathways might play an important role in regulating cell growth by hMSCs. As shown in the study of Khakoo et al. (2006), hMSCs exerted a potent suppressive effect on Kaposi’s sarcoma through specific inhibition of the protein kinase Akt activation within tumor cells. This finding was consistent with our results and suggested an important role for Akt-signaling in cancer growth regulation. In contrast, we identified an elevated level of Akt expression upon culture of cancer cells with a lower proportion of hMSCs. This suggested that hMSCs exerted promotional effect on cancer cell growth through the upregulation of Akt or other proteins/genes involved the signal transduction pathways. As previously reported, a diverse of growth signals such as EGF, PDGF family, or other extracellular stimuli, via membrane receptors, play an important role in mediating cell growth through the activation of PI3K dependent activation of AKT and highly conserved MAPK/ERK1 cascades (Shelton et al., 2004; Ciardiello and Tortora, 2001). Our findings suggested that the proliferative or suppressive effect by hMSCs participate in activation of the PI3K/Akt and ERK1/MAPK signalling pathways for modulating cell growth. The altered patterns of expression levels of genes/ proteins and the ability of hMSCs to exert a suppressive or proliferative effect on cancer cell growth might provide targeting strategy for development of MSCs-based cancer therapies.

Finally, we established stable and nondisruptive cell labelings by iQD-based strategy, offering an efficient and noninvasive way to label cells for subsequent monitoring and enrichment of labeling cells. Recently, QDs can serve as a long-term and sensitive live cell imaging tool for study of stem cell growth, nanomedicine development, bacteria detection, and immunoassay (Babic et al., 2008; Edgar et al., 2006; Alivisatos et al., 2005). The previous experiments suggest QDs do not interfere with cell physiology (Arya et al., 2005; Hoshino et al., 2005), thus provide a nontoxic and genetically noninvasive labeling strategy for studying of cell function, development, disease, and therapeutics without the disturbance of cell. Previously, we reported the application of iQD-labeled MSCs for improved long-term tracking for regenerative therapeutics (Ohyabu et al., 2009). In this study, we described the use of differential i-QDs for detection of cell’s behavior in the co-culture system. We demonstrated that i-QDs led to an effective and sensitive cell labeling (nearly 100% took up by detecting the stained cells). Moreover, tracking observation of cells by fluorescent image showed an inductive interaction for motility by cell response to each other upon co-culture of cancer cells with hMSCs. Furthermore, at the different time interval, cancer cells or hMSCs can be distinguished from each other based on FACS selection for differential enrichment in the cancer cell-MSCs co-culture. Thus, i-QD-based labeling we applied here provide a good tool for analysis of the interaction between hMSCs and cancer cells by both tracking and distinguishing cells without the disturbance of cells. Further investigation of role of hMSCs-mediated cell growth would enable deeper insight the mechanism that underlie cancer cell malignance for development of cancer therapeutics.

In summary, we explored the effect of hMSCs isolated from the bone marrow on growth of HepG2 and Hela growth, by using pre-established conditioned media, trans-well culture, and cell-to-cell co-culture. We demonstrated that hMSCs displayed a growth-inhibitory or promotional effect on cell growth. Low proportion of hMSCs stimulated cell proliferation whereas high proportion of hMSCs led to a decrease in cell growth, which was associated with ERK1/ MAPK and PI3K pathways. As hMSCs play an important role in influencing the behaviour of cancer cells, findings from our results might provide some new insights into the mechanisms by which MSCs exhibit the pro- or anti-proliferative impact on cancer cells. Further deciphering the role of hMSCs in cancer cell determination will allow for exploration of improved strategies for MSC-based cancer therapies.

Conflicts of interest

None declared.

Acknowledgments

This work was supported by grants-in-aid for scientific research from the Japan Society for the Promotion of Science. We here thank Dr. R. Gao for generously providing the technical support.

References

• Alivisatos, A.P., Gu, W., and Larabell, C. 2005. Quantum dots as cellular probes. Annu. Rev. Biomed. Eng., 7: 55–76.
• Arya, H., Kaul, Z., Wadhwa, R., Taira, K., Hirano, T., and Kaul, S.C. 2005. Quantum dots in bio-imaging: Revolution by the small. Biochem. Biophys. Res. Commun., 329: 1173–1177.
• Babic, M., Horák, D., Trchová, M., Jendelová, P., Glogarová, K., Lesný, P., Herynek, V., Hájek, M., and Syková, E. 2008. (PolyL-lysine)-modified iron oxide nanoparticles for stem cell labelling. Bioconjug. Chem., 19: 740–750.
• Bissell, M.J. and Radisky, D. 2001. Putting tumours in context. Nature Rev. Cancer, 1: 46–54.
• Caplan, A.I. 2007. Adult mesenchymal stem cells for tissue engineering versus regenerative medicine. J. Cell. Physiol., 213: 341–347.
• Chang, F., Lee, J.T., Navolanic, P.M., Steelman, L.S., Shelton, J.G., Blalock, W.L., Franklin, R.A., and McCubrey, J.A. 2003. Involvement of PI3K/Akt pathway in cell cycle progression, apoptosis, and neoplastic transformation, a target for cancer chemotherapy. Leukemia, 17: 590–603.
• Ciardiello, F. and Tortora, G. 2001. A novel approach in the treatment of cancer: targeting the epidermal growth factor receptor. Clin. Cancer Res., 7: 2958–2970.
• Cuní, S., Pérez-Aciego, P., Pérez-Chacón, G., Vargas, J.A., Sánchez, A., Martín-Saavedra, F.M., Ballester, S., García-Marco, J., Jordá, J., and Durántez, A. 2004. A sustained activation of PI3K/NF-kappaB pathway is critical for the survival of chronic lymphocytic leukemia B cells. Leukemia, 18: 1391–1400.
• Dai, W., Hale, S., Martin, B.J., Kuang, J.Q., Dow, J.S., Wold, L.E., and Kloner, R.A. 2005. Allogeneic mesenchymal stem cell transplantation in postinfarcted Rat myocardium, short- and long-term effects. Circulation, 112: 214–223.
• Edgar, R., McKinstry, M., Hwang, J., Oppenheim, A.B., Fekete, R.A., Giulian, G., Merril, C., Nagashima, K., and Adhya, S. 2006. High–sensitivity bacterial detection using biotin–tagged phage and quantum-dot nanocomplexes. Proc. Nat. Acad. Sci. USA, 103: 4841–4845.
• Fu, L.J., Li, J.X., Yang, X.G., and Wang, K. 2009. Gadolinium-promoted cell cycle progression with enhanced S-phase entry via activation of both ERK and PI3K signaling pathways in NIH 3T3 cells. J. Biol. Inorg. Chem., 14: 219–227.
• Graff, J.R., Konicek, B.W., McNulty, A.M., Wang, Z., Houck, K., Allen, S., Paul, J.D., Hbaiu, A., Goode, R.G., Sandusky, G.E., Vessella, R.L., and Neubauer, B.L. 2000. Increased AKT activity contributes to prostate cancer progression by dramatically accelerating prostate tumor growth and diminishing p27Kip1 expression. J. Biol. Chem., 275: 24500–24505.
• Gregory, C.A. 2005. Non-hematopoietic bone marrow stem cells, molecular control of expansion and differentiation. Exp. Cell Res., 306: 330–335.
• Hasebe, T., Sasaki, S., Suqitoh, M., Ono, M., Saitoh, N., and Ochiai, A. 2001. Highly proliferative intratumoral fibroblasts and a high proliferative microvessel index are significant predictors of tumor metastasis in T3 ulcerative-type colorectal cancer. Hum. Pathol., 32: 401–409.
• Hoshino, A., Manabe, N., Fujioka, K., Suzuki, K., Yasuhara, M., and Yamamoto, K. 2005. Use of fluorescent quantum dot bioconjugates for cellular imaging of immune cells, cell organelle labeling, and nanomedicine: Surface modification regulates biological function, including cytotoxicity. J. Artif. Organs, 10: 149–157.
• Houghton, J., Stoicov, C., Nomura, S., Rogers, A.B., Carlson, J., Li, H., Cai, X., Fox, J.G., Goldenring, J.R., Wang, T.C., Houghton, J., Stoicov, C., Nomura, S., Rogers, A.B., Carlson, J., Li, H., Cai, X., Fox, J.G., Goldenring, J.R., Wang, T.C., Houghton, J., Stoicov, C., Nomura, S., Rogers, A.B., Carlson, J., Li, H., Cai, X., Fox, J.G., Goldenring, J.R., Wang, T.C., Houghton, J., Stoicov, C., Nomura, S., Rogers, A.B., Carlson, J., Li, H., Cai, X., Fox, J.G., Goldenring, J.R., Wang, T.C., Houghton, J., Stoicov, C., Nomura, S., Rogers, A.B., Carlson, J., Li, H., Cai, X., Fox, J.G., Goldenring, J.R., Wang, T.C., Houghton, J., Stoicov, C., Nomura, S., Rogers, A.B., Carlson, J., Li, H., Cai, X., Fox, J.G., Goldenring, J.R., Wang, T.C., Houghton, J., Stoicov, C., Nomura, S., Rogers, A.B., Carlson, J., Li, H., Cai, X., Fox, J.G., Goldenring, J.R., Wang, T.C., Houghton, J., Stoicov, C., Nomura, S., Rogers, A.B., Carlson, J., Li, H., Cai, X., Fox, J.G., Goldenring, J.R., and Wang, T.C. 2004. Gastric cancer originating from bone marrow-derived cells. Science, 306: 1568–1571.
• Itoh, N., Semba, S., Ito, M., Takeda, H., Kawata, S., Yamakawa, M. 2002. Phosphorylation of Akt/PKB is required for suppression of cancer cell apoptosis and tumor progression in human colorectal carcinoma. Cancer, 94: 3127–3134.
• Jiang, Y., Jahagirda, B.N., and Reinhardt, R.L. 2002. Pluripotency of mesenchymal stem cells derived from adult marrow. Nature, 418: 41–49.
• Karnoub, A.E., Dash, A.B., Vo, A.P., Sullivan, A., Brooks, M.W., Bell, G.W., Richardson, A.L., Polyak, K., Tubo, R., and Weinberg, R.A. 2007. Mesenchymal stem cells within tumour stroma promote breast cancer metastasis. Nature, 449: 557–563.
• Khakoo, A.Y., Pati, S., Anderson, S.A., Reid, W., Elshal, M.F., Rovira, I.I., Nguyen, A.T., Malide, D., Combs, C.A., Hall, G., Zhang, J., Raffeld, M., Rogers, T.B., Stetler-Stevenson, W., Frank, J.A., Reitz, M., and Finkel, T. 2006. Human mesenchymal stem cells exert potent anti-tumorigenic effects in a model of Kaposi’s sarcoma. J. Exp. Med., 203: 1235–1247.
• LeBlanc, K., Tammik, L., Sundberg, B., Haynesworth, S.E., and Ringden, O. 2003. Mesenchymal stem cells inhibit and stimulate mixed lymphocyte cultures and mitogenic responses independently of the major histocompatibility complex. Scand. J. Immunol., 57: 11–20.
• Long, X.H., Zhang, J., Zhang, Y., Yao, J., Cai, Z.W., and Yang, P.Y. 2011. Nano-LC-MS/MS based proteomics of hepatocellular carcinoma cells compared to Chang liver cells and tanshinone IIA induction. Mol. Bio-Syst., 7: 1728–1741.
• Mishra, P.J., Glod, J.W., and Banerjee, D. 2009. Mesenchymal stem cells, flip side of the coin. Cancer Res., 69: 1255–1258.
• Muller, A., Homey, B., Hortensia, S., Ge, N., Catron, D., Buchanan, M.E., McClanahan, M.E., Murphy, E., Yuan, W., Wagner, S.N., Barrera, J.L., Mohar, A., Verastegui, E., and Zlotnik, A. 2001. Involvement of chemokine receptors in breast cancer metastasis, Nature, 410: 50–56.
• Ohyabu, Y., Kaul, Z., Yoshioka, T., Inoue, K., Sakai, S., Mishima, H., Uemura, T., Kaul, S.C., and Wadhwa, R. 2009. Stable and nondisruptive in vitro/in vivo labeling of mesenchymal stem cells by internalizing quantum dots. Hum. Gene Ther., 20: 219–226.
• Omuro, Y., Matsumoto, G., Sasaki, T., Tanaka, Y., Maeda, Y., Sakamaki, H., Hiruma, K., Tsuruta, K., and Takahashi, T. 2003. Regression of an unresectable pancreatic tumor following nonmyeloablative allogeneic peripheral-blood stem-cell transplantation. Bone Marrow Transplant., 31: 943–945.
• Plas, E., Carroll, V.A., Jilch, R., Simak, R., Mihaly, J., Melchior, S., Thüroff, J.W., Binder, B.R., and Pflüger, H. 2001. Variations of components of the plasminogen activation system with the cell cycle in benign prostate tissue and prostate cancer. Cytometry, 46: 184–189.
• Qiao, L., Xu, Z., Zhao, T., Zhao, Z., Shi, M., Zhao, R.C., Ye, L., and Zhang, X. 2008. Suppression of tumorigenesis by human mesenchymal stem cells in a hepatoma model. Cell Res., 18: 500–507.
• Ringden, O. and LeBlanc, K. 2005. Allogeneic hematopoietic stem cell transplantation, state of the art and new perspectives. APMIS, 113: 813–830.
• Sasser, A.K., Mundy, B.L., Smith, K.M., Studebaker, A.W., Axel, A.E., Haidet, A.M., Fernandez, S.A., and Hall, B.M. 2007. Human bone marrow stromal cells enhance breast cancer cell growth rates in a cell line-dependent manner when evaluated in 3D tumor environments. Cancer Lett., 254: 255–264.
• Shelton, J.G., Steelmam, L.S., White, E.R., and McCubrey, J.A. 2004. Synergy between PI3K/Akt and Raf/MEK/ERK pathways in IGF-1R mediated cell cycle progression and prevention of apoptosis in hematopoietic cells. Cell Cycle, 3: 372–379.
• Son, B.R., Marquez-Curtis, L.A., Kucia, M., Wysoczynski, M., Turner, A.R., Ratajczak, M.Z., and Janowska-Wieczorek, A. 2006. Migration of bone marrow and cord blood mesenchymal stem cells in vitro is regulated by stromal-derived factor-1-CXCR4 and hepatocyte growth factor-c-met axes and involves matrix metalloproteinases. Stem Cells, 24: 1254–1264.
• Staal, S.P., Hartley, J.W., and Rowe, W.P. 1977. Isolation of transforming murine leukemia viruses from mice with a high incidence of spontaneous lymphoma. Proc. Natl. Acad. Sci. USA, 74: 3065–3067.
• Stagg, J. 2008. Mesenchymal Stem Cells in Cancer Stem Cell. Rev & Rep, 4: 119–124.
• Thakkar, H., Chen, X., and Tyan, F. 2001. Pro-survival function of Akt/ protein kinase B in prostate cancer cells Relationship with TRAIL resistance. J. Biol. Chem., 276: 38361–38369.
• Vorotnikov, A.V. 2011. Chemotaxis: Movement, Direction, Control. Biochem., 76: 1528–1555.
• Wada, T. and Penninger, J.M. 2004. Mitogen-activated protein kinases in apoptosis regulation. Oncogene, 23: 2838–2849.
• Xu, W.T., Bian, Z.Y., Fan, Q.M., Li, G., and Tang, T.T. 2009. Human mesenchymal stem cells (hMSCs) target osteosarcoma and promote its growth and pulmonary metastasis. Cancer Lett., 281: 32–41
• Yang, S.H., Wu, C.C., Shih, T.T., Sun, Y.H., and Lin, F.H. 2008. In vitro study on interaction between human nucleus pulposus cells and mesenchymal stem cells through paracrine stimulation. Spine, 33: 1951–1957.
• Yuan, X., Hu, J., Belladonna, M.L., Black, K.L., and Yu, J.S. 2006. Interleukin-23-expressing bone marrow-derived neural stem-like cells exhibit antitumor activity against intracranial glioma. Cancer Res., 66: 2630–2638.
• Zhu, W., Xu, W., Jiang, R., Qian, H., Chen, M., Hu, J., Cao, W., Han, C., and Chen, Y. 2006. Mesenchymal stem cells derived from bone marrow favor tumor cell growth in vivo. Exp. Mol. Pathol., 80: 267–274.
• Zhu, Y., Sun, Z., and Han, Q. 2009. Human mesenchymal stem cells inhibit cancer cell proliferation by secreting DKK-1. Leukemia, 23: 925–933.