SEARCH

SEARCH BY CITATION

Keywords:

  • Mesenchymal stromal cells;
  • Multiple myeloma;
  • Microenvironment;
  • Chemokine

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Multiple myeloma (MM) is a malignancy of terminally differentiated plasma cells that are predominantly localized in the bone marrow (BM). Mesenchymal stromal cells (MSCs) give rise to most BM stromal cells that interact with MM cells. However, the direct involvement of MSCs in the pathophysiology of MM has not been well addressed. In this study, in vitro and in vivo migration assays revealed that MSCs have tropism toward MM cells, and CCL25 was identified as a major MM cell-produced chemoattractant for MSCs. By coculture experiments, we found that MSCs favor the proliferation of stroma-dependent MM cells through soluble factors and cell to cell contact, which was confirmed by intrafemoral coengraftment experiments. We also demonstrated that MSCs protected MM cells against spontaneous and Bortezomib-induced apoptosis. The tumor-promoting effect of MSCs correlated with their capacity to enhance AKT and ERK activities in MM cells, accompanied with increased expression of CyclinD2, CDK4, and Bcl-XL and decreased cleaved caspase-3 and poly(ADP-ribose) polymerase expression. In turn, MM cells upregulated interleukin-6 (IL-6), IL-10, insulin growth factor-1, vascular endothelial growth factor, and dickkopf homolog 1 expression in MSCs. Finally, infusion of in vitro-expanded murine MSCs in 5T33MM mice resulted in a significantly shorter survival. MSC infusion is a promising way to support hematopoietic recovery and to control graft versus host disease in patients after allogeneic hematopoietic stem cell transplantation. However, our data suggest that MSC-based cytotherapy has a potential risk for MM disease progression or relapse and should be considered with caution in MM patients. STEM CELLS 2012; 30:266–279.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Mesenchymal stromal cells (MSCs) are self-renewing and multipotent progenitors that can differentiate into a variety of cell types, such as adipocytes, osteoblasts, and chondrocytes. Several other terms for MSCs have been used interchangeably, including mesenchymal stem cells, marrow stromal cells, and multipotent stromal cells but at this moment none is used uniformly. Bone marrow (BM) is the most common source of MSCs but these stem cells have also been isolated from various other sources, such as placenta, amniotic fluid, cord blood, retina, inner ear, gastric epithelium, tendons, synovial membrane, hair follicle, teeth, fetal liver, and adipose tissue [1–8]. In addition to their multilineage differentiation potential, MSCs also possess immunomodulary properties and have the capacity to home to injury/tumor sites as well as to produce a variety of cytokines/chemokines. All these features offer these stem cells the potential to be used for various preclinical and clinical applications, including support of hematopoietic stem cell engraftment, tissue engineering, gene therapy, and control of immune reactions in graft versus host disease, organ transplantation, and autoimmune diseases [9–14].

Recently, evidence was provided that MSCs have preferential tropism for tumor sites [15–18]. However, the exact effect of MSCs on tumor growth and development is still in debate [19]. Khakoo et al. [20] reported in experimental models of Kaposi's sarcoma (KS) that the coinjection of human MSCs (hMSCs) with KS cells inhibited primary tumor growth and they identified Akt inhibition in the KS cells through direct contact with MSCs as a possible mechanism for inhibition of tumor progression. An MSC-mediated tumor inhibition effect has also been observed in experimental models of lymphoma, melanoma, and hepatoma [21–23]. Meanwhile, many studies have demonstrated the capacity of MSCs to initiate and/or to support survival, progression, and metastasis of certain tumor types. Karnoub et al. [24] demonstrated that MSCs could increase the metastasis rate of breast cancer cells through secretion of CCL5 by MSCs. Several other groups have also reported that MSCs injected together with tumor cells can favor tumor growth in vivo, as observed for osteosarcoma, large-cell lung cancer, and colon cancer [25–27]. While a lot of studies indicate that nonmodified MSCs have no significant influence on the progression of tumors [28, 29], there are two reports showing that MSCs exhibit opposite effects on the same tumor model in vitro and in vivo [30, 31], indicating that effects of MSCs on cancer cell growth are difficult to interpret. Klopp et al. [19] recently reviewed the differences in the methodology of these reported studies and proposed that the time of MSCs introduction into tumors might be a critical factor for the contradicting results. Other possible explanations for these conflicting outcomes might include the ratio of MSC numbers to cancer cell numbers that were used in the different studies as well as the nature of the tumor type itself [32].

Multiple myeloma (MM) is a malignant plasma cell disorder characterized by an accumulation of monoclonal terminally differentiated plasma cells in the BM and the presence of a monoclonal immunoglobulin fraction in the blood and/or urine [33]. The involvement of the BM microenvironment in the pathophysiology of the MM disease is nowadays well-documented. The crosstalk between BM stromal cells and MM cells supports the proliferation, survival, migration, and drug resistance of MM cells as well as osteoclastogenesis and angiogenesis [34, 35]. Recent insight into the functional importance of the BM stroma and its interaction with MM cells lead to the identification of many new molecular targets and derived treatment regimens [36]. Although MSCs are the precursors of BM stromal cells, their direct involvement in the progression of MM is not clearly defined. In this study, we demonstrated that MM cells induce in vitro as well as in vivo chemotaxis of MSCs. Through chemokine profile analysis, we identified CCL25 as a major MM cell-produced chemokine that is involved in MSC chemotaxis. By conducting coculture experiments in vitro as well as in vivo experiments, we also studied the effect of MSCs on MM cell growth and apoptosis. Despite the fact that MSCs constitute only a small population in the BM cell population, even after exogenous transplantation, our findings indicate that the supportive role of MSC in MM cell growth cannot be ignored.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

5T33MM Murine Model

The 5T33MM mice originated spontaneously from elderly C57Bl/KaLwRij mice, and this MM model is maintained by i.v. transfer of diseased BM cells into young syngeneic C57Bl/KaLwRij mice (Harlan CPB, Horst, The Netherlands, www.harlan.com) at 6–8 weeks of age [37, 38]. The development of myeloma was assessed by the level of monoclonal antibody present in the serum (paraprotein) by protein electrophoresis. Mice were housed and maintained following the conditions approved by the Ethical Committee for Animal Experiments, Vrije Universiteit Brussels (license no. LA1230281).

Myeloma Cell Lines

The murine 5T33MMvv and 5T33MMvt MM cells were used in this study. The 5T33MMvv cells grow in vitro stroma dependently with a survival of only a few days. The 5T33MMvt is a clonally identical but in vitro stroma-independent growing variant of the 5T33MMvv cells. Stroma-independent human MM cell lines RPMI8226, LP-1, Karpas, U266, MM5.2, MM S1 and one stroma-dependent human MM cell line MM5.1 were also used in this study. All cell lines were kept in culture as described [39, 40].

Primary Human MM Cells

BM samples of MM patients were collected for routine diagnostic purposes after obtaining informed consent. Each MM patient was diagnosed and staged according to the criteria of Durie and Salmon [41]. The study was approved by the local ethical committee. Primary MM cells were immunomagnetically separated using the magnetic cell sorting system (MACS; Miltenyi Biotech, Leiden, The Netherlands, www.miltenyibiotec.com) with CD138 microbeads according to the manufacturer's instruction. CD138 (+) cells were recovered, and viability was assessed with trypan blue staining. MACS purification revealed a pure primary MM cell population (>98% plasma cells) as determined by May-Grünwald Giemsa-stained cytospin preparations.

Primary Culture of MSCs

hMSCs

Human BM samples, aspirated from the sternum, were obtained from healthy donors with informed consent. hMSCs were isolated and cultured according to the methods described previously [42]. hMSCs were used at passages 3–5 in this study.

Murine MSCs

Murine MSCs (mMSCs) are more difficult to culture due to the low mMSC numbers in BM and a considerable level of contaminated hematopoietic cells in the long-term in vitro culture [43]. We established a modified isolation and culture method based on the combination of mechanical crushing and collagenase digestion at the moment of harvest, followed by an immunodepletion step using microbeads coated with CD11b, CD45, and CD34 antibodies [44]. mMSCs were used at passages 3–5 for all experiments.

Flow Cytometry Analysis

A two-step staining method was used for immunophenotyping as described [42]. The following antibodies were used: CD14, CD34, CD105, Sca-1, CD45, CD90, vascular cell adhesion molecule 1 (VCAM-1), anti-mouse CCR9 (all purchased from eBioscience, San Diego, CA), CD73, integrin α4 (both from BD Biosciences, San Diego, CA, www.bdbiosciences.com), integrin β1 (Acris Antibodies, Herford, Germany, www.acris-antibodies.com), and anti-human CCR9 (R&D, Minneapolis, MN, www.rndsystems.com). Cells were analyzed with a FACSCanto flow cytometer (Becton Dickinson, San Jose, CA, www.bdbiosciences.com). WinMDI 2.8 software was used to create the overlap histograms.

Preparation of Conditioned Medium

A total of 2.5 × 105 MSCs were cultured for 2 days in 5 ml serum-free RPMI-1640 medium and culture supernatant was harvested as a source of conditioned medium (CM). CM was centrifuged at 2000 rpm to remove cell debris and frozen at −20°C until use.

For the collection of MM cell CM, MM cell lines or primary human MM cells (CD138 positive population) were incubated in serum-free RPMI-1640 medium for, respectively, 72 and 24 hours, at a fixed cell concentration of 1 × 106 per milliliter.

In Vitro Migration Assays

The migratory ability of MSCs was determined as described [42]. MM cell CM was added as chemoattractant source. Serum-free medium and 10% fetal calf serum containing medium were used as negative and positive controls, respectively. Migration experiments were also performed using hMSCs with knockdown of CCR9 via RNA interference.

In Vivo Migration Assays

5T33MM mice were diseased as described above. After 2 weeks of inoculation, 5T33MM mice and naive mice were injected i.v. with DiI (1,1′dioctadecyl-3,3,3′,3′tetramethylindocarbocyanine perchlorate) (Molecular Probes, Invitrogen, Merelbeke, Belgium, www.invitrogen.com)–labeled 3 × 105 mMSCs suspended in 200 μl of 0.9% NaCl (n = 20 per group). At days 1, 3, 5, and 7 post-MSC injection, five 5T33MM mice and five naive mice were sacrificed. Lung, heart, liver, spleen, kidney, and tibia were removed, and 5-μm paraffin sections were made. After deparaffin treatment, DiI-labeled mMSCs were observed and counted under fluorescent microscope OLYMPUS IX81. To quantify MSCs distribution among different organs, MSCs were counted in eight random fields on each section. Five uncontinuous sections were scored to calculate the mean value. Adjacent sections were analyzed by hematoxylin and eosin staining. Blood was also taken to determine serum paraprotein concentration at the indicated time.

Chemokine Expression in MM Cells

To detect the presence of chemokines potentially mediating MSCs migration, CM from the human MM cell line RPMI8226 was collected as described above and tested by DiscoveryMAP v1.0 analysis (Rules Based Medicine, Austin, TX, www.rulesbasedmedicine.com) using antibodies targeting 22 different chemokines.

CCR9 Knockdown by RNA Interference

To knockdown CCR9 expression, hMSCs were transfected with FlexiTube GeneSolution for CCR9 (GS10803, Qiagen, Hilden, Germany, www.qiagen.com). The FlexiTube GeneSolution for CCR9 provides four nonoverlapping CCR9 RNAi duplexes for this gene to obtain high knockdown efficiency. These duplexes were transfected into hMSC with Lipofectamine RNAiMAX (Invitrogen) according to the manufacturer's protocol. AllStars Negative Control small interfering RNA (siRNA; SI03650318, Qiagen) was used as negative control. The efficiency of CCR9 knockdown was evaluated by real time polymerase chain reaction (PCR) and fluorescence-activated cell sorting (FACS).

The knockdown for CCR9 expression in mMSCs was performed using the same strategy but cells were transfected with mouse CCR9 RNAi duplexes (GS12769, Qiagen).

In Vitro Proliferation Assay

A total of 5 × 104 MM cells were plated in 96-well plates for 24 hours by direct coculturing with increasing numbers of MSCs (previously irradiated at 1,500 rad) in serum-free RPMI-1640 medium or by MSC CM. The proliferation of MM cells was then determined by thymidine incorporation assay, described as previously [45]. Results are expressed as the relative DNA synthesis compared with MM cells alone.

In Vivo Proliferation Assay

Mice were anesthetized, and a small incision was made over the right knee to gain access to the kneecap. A 27-gauge needle was used to drill a hole in the femur, and 20 μl of RPMI-1640 containing 5 × 106 freshly isolated 5T33MMvv cells alone or in the presence of 0.5 × 106 mMSCs was slowly injected into the cavity with a 26-gauge microliter syringe (n = 5 per group). After 2 weeks, femur with injection in situ, opposite femur and spleen were removed, and the number of 5T33MMvv cells in these organs was assessed by immunostaining with anti-idiotype antibodies. The development of the anti-idiotype antibody against 5T33MM cells, called 3H2, was described previously [38].

Assessment of Apoptosis

Apoptosis of MM cells was induced by nutritional starvation or Bortezomib activity (Janssen Pharmaceutica N.V., J&J PRD). MM cells were washed twice with phosphate-buffered saline (PBS) and stained with 3 μl of 7-amino-actinomycin D (7-AAD; BD Pharmingen, Franklin Lakes, NJ, www.bdbiosciences.com) and 4 μl of Annexin V–fluorescein isothiocyanate in 100 μl of binding buffer for 15 minutes in the dark at room temperature. Then, cells were resuspended in 400 μl of binding buffer and immediately analyzed using a FACSCanto flow cytometer. The cells undergoing early apoptosis are in the lower right quadrant being Annexin V positive and 7-AAD negative; late apoptotic or dead cells are in the upper right quadrant being 7-AAD positive and Annexin V positive; live cells are in the lower left quadrant being negative for both fluorescent probes.

Immunofluorescence Staining

To assess MM cell proliferation and apoptosis in situ after intrafemoral injection, immunofluorescence staining using antiproliferation cell nuclear antigen (PCNA) antibody or TUNEL (terminal deoxynucleotidyl transferase dUTP nick end labeling) assay were performed. The femurs were removed and fixed in decalcification solution. All fixed samples were embedded in paraffin and 5-μm sections were cut. After being deparaffinized and rehydrated, slides were incubated in citrate buffer (pH 6.0), and antigen retrieval was performed in a microwave two times for 5 minutes. After cooling down at room temperature, slides were washed twice for 5 minutes with PBS and incubated with 10% normal goat serum for 30 minutes at room temperature. To determine MM cell proliferation, anti-PCNA antibody (sc-7907, 1/100, rabbit-anti-mouse, Santa Cruz Biotechnology, Santa Cruz, CA, www.scbt.com) in PBS were added, and the slides were incubated at 4°C overnight. The next day, after three washes in PBS for 5 minutes, secondary antibody Alexa Fluor 488 conjugated goat-anti-rabbit IgG (1/250, Invitrogen) was added and allowed to incubate for 1 hour in the dark at room temperature. After three washes in PBS for 5 minutes, the slides were counterstained, mounted with SlowFade Gold antifade reagent with 4′,6-diamidino-2-phenylindole (DAPI) (Invitrogen), and left for 10 minutes in the dark at room temperature before examination by fluorescence microscopy (Zeiss Axioplan 2, Carl Zeiss, Gottingen, Germany, www.zeiss.com). For testing MM cell apoptosis in situ, a TUNEL assay was performed according to the manufacturer's protocol. In brief, after antigen retrieval, the slides were washed in PBS three times and incubated with 50 μl of TUNEL reaction mixture (11684817910, In situ Cell Death Detection Kit, Roche, Mannheim, Germany, www.roche-applied-science.com) for 1 hour at 37°C. After three washes in PBS for 5 minutes, the slides were also counterstained by DAPI (Invitrogen) and evaluated by fluorescence microscopy. For both PCNA and TUNEL staining, adjacent sections were analyzed to determine MM cells location. The proliferation index or apoptosis index was quantified by the average percentage of PCNA(+)/DAPI(+) or TUNEL(+)/DAPI(+) per field of ×200 using Image-Pro Plus 6.0 software. Three random fields with more than 90% MM cells were counted per femur. n = 5 mice per group.

Real Time PCR

Total RNA was isolated using Trizol (Invitrogen) and RNeasy Mini Kit (Qiagen, Germany), following the manufacturer's instructions. The concentration and purity of RNA was determined by Quant-iT RNA BR Assay kit (Invitrogen) with Qubit fluorometer (Invitrogen). cDNA was synthesized using the Thermoscript reverse-transcription PCR system (Invitrogen) with random hexamers as primers. Quantitative real-time PCR analysis was done using the iCycler (Bio-Rad Laboratories, Hercules, CA, www.bio-rad.com) using the SYBR GreenER qPCR SuperMix for iCycler (Invitrogen) according to manufacturer's instructions. The primer sequences used are listed in Table 1. Transcript levels were normalized to the housekeeping gene β-actin and analyzed by the relative quantification 2−ΔΔCt method.

Table 1. Primers for quantitative real time PCR
inline image

Western Blot Analysis

mMSCs and 5T33MMvv cells were cocultured for 24 hours at a ratio of 1:10. 5T33MMvv cells were then harvested by gently pipetting to separate them from adherent mMSCs. Preparation of whole cell lysates and immunoblotting were performed as previously described [45], using the following antibodies: Cylin D2, CDK4, Bcl2, Bcl-XL, Caspase-3, poly(ADP-ribose) polymerase (PARP), AKT, Phospho-Akt, p44/42 mitogen-activated protein kinase (MAPK). Phospho-p44/42 MAPK, and β-actin (all from Cell Signaling, Danvers, MA, www.cellsignal.com).

In Vivo mMSCs Infusion Study

After inoculation of the 5T33MMvv cells (1 × 105 cells per mice) on day 0, 5T33MM diseased mice (n = 10) were treated with mMSCs (2 × 105 cells in 200 μl of 0.9% NaCl) or vehicle (0.9% NaCl) intravenously on days 6, 10, and 14. To determine the safety of mMSCs, naive mice (n = 10) received the same amount of mMSCs according to the same schedule. Additional naive mice (n = 10) without treatment were included as negative controls. The survival time of each mouse was determined based on the occurrence of morbidity, namely hind limb paralysis [46]. In parallel, an additional in vivo mMSCs infusion study was performed. The difference was that the mice were all killed when the first mouse showed the sign of morbidity. Blood samples were obtained to determine serum paraprotein concentrations.

Statistical Analysis

Statistical analysis was done using GraphPad Prism 5 software. All data represent the mean ± SD, and results were analyzed using the Mann–Whitney U test. Survival curves were plotted using the Kaplan-Meier method. p < .05 was considered statistically significant. All experiments were repeated in at least triplicates.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

MSCs Can Migrate Toward Myeloma Cells

Prior to their use in experiments, the true nature of the in vitro-expanded MSCs was confirmed by morphological analysis, immunostaining with specific surface markers, and their differentiation ability toward adipocytes, osteoblasts, and chondrocytes in specific induction media (Supporting Information Fig. 1).

The directed migration of MSCs in response to MM cells was investigated using an in vitro transwell system. CM of murine MM cells and human MM cells were used as source of chemoattractants, while serum-free medium and 10% fetal calf serum were used as negative and positive controls, respectively. We observed a significant MSC migration in response to both murine and human MM cells, although hMSC migration toward MM cells was higher than mMSC migration (Fig. 1A).

thumbnail image

Figure 1. MM cells stimulate migration of MSCs in vitro and in vivo. (A): Quantification of MSC migration in vitro in response to factors secreted by MM cells (5T33MMvt and 5TMMvv MM cells for mMSCs, RPMI8226, and LP-1 MM cells for humans MSCs). Data are the mean ± SD of results from three independent experiments. (B): Schematic chart of in vivo MSCs migration assay. mMSCs were labeled with the cell tracker DiI beforehand and injected i.v. into 5T33MM mice on 4th day. Naïve mice were used as control. On the indicated days post-mMSC administration, mice were sacrificed, organs were harvested, and paraffin sections were made. (C): DiI-labeled mMSCs were visualized under fluorescence microscope. Representative photographs are shown for lung, heart, liver, spleen, kidney, and tibia of 5T33MM mice on 7th day after mMSCs injection. Scale bar = 50 μm. (D): In vivo mMSC distribution in these organs was compared between naive and 5T33MM mice on day 7 post-mMSC administration. *, p < .05. (E): Quantification of the number of MSC that migrated to MM-infected organs (spleen and tibia) post-mMSC administration is consistent with the MM disease progression as determined by M-compound level. *, p < .05; **, p < .01; ***, p < .001 compared to mMSCs number on day 1 after i.v. injection. Abbreviations: DiI, 1,1′dioctadecyl-3,3,3′,3′tetramethylindocarbocyanine perchlorate; FCS, fetal calf serum; mMSC, murine mesenchymal stromal cells; MM, multiple myeloma.

Download figure to PowerPoint

MSC migration toward MM cells was investigated in vivo using the 5T33MM mouse model. To track the migration of transplanted MSC, MSCs were labeled with the cell tracker dye DiI prior to in vivo administration. After 2 weeks of the administration of 5T33MMvv cells, MM and naive mice were injected i.v. with DiI-labeled mMSC. On the 1st, 3rd, 5th, and 7th day, several organs (heart, lungs, liver, spleen, kidney, and tibia) were harvested for histological analysis, and blood samples were taken to measure serum paraprotein concentration (Fig. 1B). DiI-labeled mMSCs were visualized under the fluorescent microscope (Fig. 1C). Microscopic analysis provided evidence that 7 days after MSCs administration there were many mMSCs detectable in lungs and liver which are “barrier” organs. A decreased MSC migration to lungs and an increased migration to liver were observed over time (Supporting Information Fig. 2). However, more mMSCs were detected in spleen and tibia in 5T33MM mice compared with naive mice (Fig. 1D). Spleen and BM are two major MM cell invaded organs in 5T33MM mice [38]. Moreover, on the indicated harvest days, we observed that the number of migrated mMSCs was gradually increased in spleen and tibia of 5T33MM mice, which was in accordance with the increasing level of M compound, a marker representing disease progression (Fig. 1E). Although MSCs do not exclusively home to MM sites, these data demonstrate that MSCs can preferentially home to MM sites.

MM Cell-Produced CCL25 Is a Potent Chemokine for Attraction of MSCs

Chemotaxis involves both release of migratory signals through chemokines and expression of specific receptors on the migrating effector cells. Having found that MSCs could migrate in response to MM cells, we started to identify MM cell-secreted chemokines that could induce recruitment of MSCs. The CM of RPMI8226 MM cells was used for DiscoveryMAP v1.0 analysis, which allows simultaneous detection of a large range of chemokines and cytokines. It was found that RPMI8226 cells could secrete a variety of inflammatory chemokines, including eotaxin-3, IP-10, MIG, RANTES, and TECK, among which CCL25 (TECK) showed the highest concentration (Fig. 2A). To identify whether CCL25 is widely expressed in MM cells, 6 other human MM cell lines as well as primary MM cells from 14 patients were tested by semiquantitative PCR. It was found that CCL25 was detected in 11 out of 14 patients and 5 out of 6 MM cell lines (Fig. 2B).

thumbnail image

Figure 2. Involvement of CCL25 in MM cell-induced migration of mMSCs. (A): Chemokine analysis of serum-free conditioned medium from the MM cell line RPMI8226. Results are expressed as concentration of secreted chemokines (pg/ml). (B): CCL25 mRNA expression in isolated bone marrow (BM) plasma cells from 14 patient samples (MM1 to MM14) and 7 human MM cell lines. β-Actin mRNA expression was used as positive control. (C): One representative FACS analysis of hMSCs after CCR9 knockdown from three independent experiments. (D): CCR9 knocked-down hMSCs exhibited a decreased in vitro migration ability toward primary human MM cells and three MM cell lines. Data are the mean ± SD of results from three experiments. *, p < .05 compared to hMSCs transfected with control siRNA. (E): One representative fluorescence-activated cell sorting (FACS) analysis of mMSCs after CCR9 knockdown from three independent experiments. (F): After 7 days of i.v. injection into 5T33MM mice, CCR9 knocked-down mMSCs exhibited a significantly decreased in vivo migration toward the MM-invaded spleen, as well as a decreased migration toward the tibia. n = 5 per group; *, p < .05 compared to mMSCs transfected with control siRNA. Abbreviations: DiI, 1,1′dioctadecyl-3,3,3′,3′tetramethylindocarbocyanine perchlorate; hMSCs, human mesenchymal stromal cells; mMSCs, murine mesenchymal stromal cells; MM, multiple myeloma; PE, phycoerythrin; siRNA, small interfering RNA.

Download figure to PowerPoint

Furthermore, to examine whether MM cell-secreted CCL25 is functionally involved in MSC migration, the CCL25 unique receptor, CCR9, was knocked-down in hMSCs by RNA interference. FACS analysis showed the efficiency of CCR9 knockdown (Fig. 2C). hMSCs with CCR9 knockdown exhibited a significant decrease in migration (32%–43%) toward MM cells using an in vitro transwell system (Fig. 2D). In addition, we confirmed that 5T33MM murine MM cells were positive for CCL25 (Supporting Information Fig. 3) and knocked down CCR9 expression in mMSCs by RNA interference (Fig. 2E). Then, an in vivo migration assay was performed using CCR9 knockdown or control siRNA transfected mMSCs into 5T33MM mice. After 7 days of injection, we could observe that there were significantly less mMSCs in the spleen of 5T33MM mice with CCR9 knockdown mMSCs injection. There were less mMSCs in the tibia as well, although this was not significant (Fig. 2F).

MSCs Stimulate Myeloma Cell Proliferation In Vitro and In Vivo

To test the effect of MSCs on the proliferation of MM cells, we first cocultured the human MM cell lines MM5.1 (stroma-dependent) and RPMI8226 (stroma-independent) as well as the murine MM cell lines 5T33MMvv (stroma-dependent) and 5T33MMvt (stroma-independent), with hMSCs or mMSCs, respectively, at different ratios for 24 hours in serum-free medium. MSCs exhibited a dose-dependent effect on the proliferation of human and murine stroma-dependent MM cells but had no apparent effect on stroma-independent MM cells (Fig. 3A). To analyze whether soluble factors from MSCs are involved in the stimulation of MM proliferation, MSC CM was used to culture stroma-dependent MM cells, MM5.1 and 5T33MMvv. A stimulatory effect was also detected but less pronounced when compared with the effect observed in the cultures with direct cell–cell contact (Fig. 3B).

thumbnail image

Figure 3. MSCs favor stroma-dependent MM cell growth in vitro and in vivo. (A): The proliferation of MM cells was assessed after coculture with pre-irradiated MSCs. Human MSCs and mMSCs contribute to the proliferation of the stroma-dependent MM cell line MM5.1 and 5T33MMvv, in a dosage-dependent manner but have no remarkable effect on the stroma-independent MM cell line RPMI8226 and 5T33MMvt. *, p < .05; **, p < .01; ***, p < .001 compared to tumor cells alone. (B): Stroma-dependent MM cells were cultured with MSCs (MM cells/MSCs ratio = 10:1) or in the presence of MSCs conditioned medium. A growth promoting effect was detected in both conditions but less pronounced with MSC conditioned medium as compared to the direct cell–cell contact condition. (C): Schematic chart of the in vivo proliferation assay. 5T33MMvv cells were injected with or without mMSCs into the right femur. After 2 weeks, cell suspensions from bone marrow of two hind legs and the spleen were evaluated for the presence of MM cells using idiotype-antibody FACS analysis and in situ immunostaining. (D): The level of MM cells in both femurs and spleen was monitored by anti-idiotype FACS analysis at 2 weeks after injection. n = 5 per group. *, p < .05 compared to tumor cell injection alone. (E): 5T33MMvv cells (5 × 106) were injected intrafemorally with or without mMSCs (5 × 105). After 1 week, more MM cells were in a proliferative state as shown by positive PCNA expression in the femur where mMSCs were coinjected compared to tumor cell injection alone. The data represented the percentage of PCNA (+) cells in all DAPI (+) cells per field. n = 5 mice per group. *, p < .05 compared to tumor cell injection alone. Abbreviations: DAPI, 4′,6-diamidino-2-phenylindole; FACS, fluorescence-activated cell sorting; mMSCs, murine mesenchymal stromal cells; MM, multiple myeloma; PCNA, proliferation cell nuclear antigen.

Download figure to PowerPoint

Next, we investigated whether MSCs could favor MM cell growth in vivo. Naive mice received 5 × 106 5T33MMvv cells with or without 5 × 105 mMSCs by intrafemoral injection in the right leg. After 2 weeks, cell suspensions from the BM of two hind legs and the spleen were evaluated for the presence of MM cells using idiotype-antibody FACS analysis (Fig. 3C). In the right femur where MM cells and mMSCs were coinjected, 29.1% MM cells were detected, while only 12.4% MM cells were detected in the femur without mMSCs injection. After injection of MM cells into the right femur, some MM cells could be detected in the opposite leg and in the spleen as well. By anti-idiotype FACS analysis, we found a higher proportion of MM cells in the opposite leg and spleen of mice that were coinjected with MSCs when compared with mice injected with 5T33MM cells alone, however, without significant difference (Fig. 3D). To confirm the in vivo data that MSC can promote proliferation of MM cells, we injected 5T33MMvv cells intrafemorally into naive mice together with or without mMSCs intrafemorally. After 7 days, by an immunofluorescence staining of the proliferation antigen PCNA, we observed that there were significantly more MM cells positive for PCNA expression in the femur where mMSCs were coinjected (proliferation index: tumor alone vs. MSC coinjection is 12.4 ± 2.8% and 19.8 ± 5.3%, respectively, p < .05; Fig. 3E, Supporting Information Fig. 4).

MSCs Protect MM Cells Against Apoptosis In Vitro and In Vivo

Human and murine MM cells were cultured in serum-free medium for 24 or 48 hours to induce spontaneous apoptosis. In the presence of MSCs, the proportion of apoptotic cells was dramatically decreased, in a dosage-dependent manner (Fig. 4A). Chemoresistance and relapse is the key feature of the clinical course in MM, and it is well-known that the BM microenvironment protects MM cells against chemotherapy. Here we examined whether MSCs could protect MM cells against chemotherapy-induced apoptosis. Bortezomib is a clinically available proteasome inhibitor that is currently among the most potent chemotherapeutic drugs used in the treatment of MM. When the human MM cell line RPMI8226 and the murine MM cell line 5T33MMvt were cultured in complete medium with 5 nM Bortezomib for 48 hours, approximately 70% apoptotic cells (early + late) were observed. However, in the presence of MSCs, the proportion of apoptotic cells significantly decreased to 35% and 53%, respectively. Similar findings were observed for the stroma-dependent human MM cell line MM 5.1 and the murine MM cell line 5T33MMvv (Fig. 4B). After separation from hMSCs, we observed less than 2% hMSCs contamination in the human MM fraction, excluding significant influence on the apoptotic analyses of the MM cells (Supporting Information Fig. 5). A comparable finding was also observed in the coculture of mMSCs and murine MM cells.

thumbnail image

Figure 4. MSCs protect MM cells against apoptosis in vitro and in vivo. (A): MSCs were cocultured in serum-free medium with stroma-dependent MM cells (5T33MMvv and MM5.1) for 24 hours or stroma-independent MM cells (5T33MMvt and RPMI8226) for 48 hours, at a ratio of 1:100 or 1:10 (MSCs/MM cells). As measured by AnnexinV/7-AAD staining, MSCs could protect both stroma-dependent and stroma-independent MM cells against spontaneous apoptosis induced by serum starvation in a dose-dependent manner. (B): In complete growth medium, MSCs had no apparent effect on apoptosis of stroma-independent MM cell 5T33MMvt and RPMI8226 but could protect against Bortezomib-induced apoptosis at a ratio of 1:10 (MSCs/MM cells). Similarly, MSCs can also protect Bortezomib-induced apoptosis of stroma-dependent MM cells 5T33MMvv and MM 5.1. Representative fluorescence-activated cell sorting profiles are shown. *, p < .05; **, p < .01; ***, p < .001. (C): 5T33MMvv cells (5 × 106) were injected intrafemorally with or without mMSCs (5 × 105). After 1 week, there were less MM cells undergoing apoptosis in the femur where mMSCs were coinjected, compared to tumor cell injection alone, although this was not significant. TUNEL immunofluorescence staining was performed, and the slides were counterstained by DAPI. n = 5 mice per group. (D): After 5T33MMvv injection into naive mice intrafemorally, five doses of Bortezomib at 0.6 mg/kg i.p. every other day were given, leading to MM cell apoptosis as measured by TUNEL staining. However, with the coinjection of mMSCs, significantly less MM cells underwent apoptosis. The data represented the percentage of TUNEL (+) cells in all DAPI (+) cells per field. n = 5 mice per group. *, p < .05 compared to tumor cell injection alone. Abbreviations: 7-AAD, 7-amino-actinomycin D; DAPI, 4′,6-diamidino-2-phenylindole; hMSCs, human mesenchymal stromal cells; mMSCs, murine mesenchymal stromal cells; MM, multiple myeloma; TUNEL, terminal deoxynucleotidyl transferase dUTP nick end labeling.

Download figure to PowerPoint

To confirm in vivo that MSC could protect against spontaneous apoptosis of MM cells, 5T33MMvv cells were intrafemorally injected into naive mice together with or without mMSCs. By the TUNEL apoptotic assay, less MM cells were found to undergo apoptosis in mMSC coinjected femurs, although no significance was observed (apoptosis index: tumor alone vs. MSC coinjection is 8.9 ± 2.8% and 6.2 ± 2.3%, respectively; Fig. 4C, Supporting Information Fig. 6A). In addition, to explore whether MSC could protect MM cells against Bortezomib-induced apoptosis in vivo, mice were given Bortezomib after intrafemoral injection of 5T33MMvv cells with or without mMSCs, and the TUNEL apoptotic assay was performed. We found significantly less apoptotic MM cells in the mMSC coinjected femur indicating that MSC could protect Bortezomib-induced MM apoptosis in vivo (apoptosis index: tumor alone vs. MSC coinjection is 81.4 ± 10.0% and 63.7 ± 13.3%, respectively, p < .05; Fig. 4D, Supporting Information Fig. 6B).

Crosstalk Between MM Cells and MSCs

Previous studies have shown that a regulatory network of cytokines and growth factors exist between BM stromal cells and MM cells within the MM microenvironment [47–49]. In this study, we tested the effect of 5T33MMvt cells on the expression of MM specific growth and survival factors in mMSCs. After transwell coculture with MM cells for 48 hours, interleukin-6 (IL-6), insulin growth factor-1 (IGF-1), vascular endothelial growth factor (VEGF), IL-10, and dickkopf homolog 1 (DKK1) expression in mMSCs were all significantly upregulated (Fig. 5A). To determine how MSCs could favor the growth of stroma-dependent MM cells and protect MM cells against apoptosis, we further examined the molecular changes of MM cells in the presence of MSCs. After coculturing mMSCs and murine MM cells 5T33MMvv for 24 hours, MM cells showed enhanced expression of pAKT and pERK. Positive cell cycle protein Cyclin D2 and CDK4 were both upregulated. In addition, the antiapoptotic protein Bcl-XL, but not Bcl-2, was increased in MM cells, accompanied with the downregulation of the proapoptotic proteins, cleaved Caspase-3 and PARP (Fig. 5B).

thumbnail image

Figure 5. Crosstalk between MM cells and MSCs. (A): MM cells alter cytokine expression in MSCs. After coculture with 5T33MMvt cells in a transwell system for 48 hours, IL-6, IGF-1, VEGF, IL-10, and DKK1 expression were all upregulated in mMSCs as measured by real time quantitative PCR. (B): MSC-induced molecular changes in MM cells. By Western blot analysis, 5T33MMvv MM cells, cocultured with mMSCs for 24 hours, exhibited upregulated expression of the cell cycle proteins cyclin D2 and CDK4, and the antiapoptotic protein Bcl-XL, as well as decreased expression of cleaved Caspase-3 and PARP. Phosphorylated AKT and phosphorylated ERK1/2 were also increased. One of three independent experiments is shown. (C): Kaplan Meier survival curve. Mice were injected i.v. with mMSCs or the vehicle on the indicated days (arrows) after inoculation with 5T33MM cells (n = 10 per group). On average, the onset of morbidity in the vehicle group was 22 days, and the mMSCs infusion group lived 4 days shorter. *, p < .05; **, p < .01. (D): In a parallel mMSC transplantation experiment, all mice were sacrificed and serum paraprotein was analyzed when the first mice showed signs of morbidity. 5T33MM mice with mMSC infusion had higher paraprotein levels than the mice without mMSCs infusion. *, p < .05. n = 10 per group. Abbreviations: DKK1, dickkopf homolog 1; IGF-1, insulin growth factor-1; IL-6, interleukin-6; mMSCs, murine mesenchymal stromal cells; MM, multiple myeloma; PARP, poly(ADP-ribose) polymerase; PCR, polymerase chain reaction; VEGF, vascular endothelial growth factor.

Download figure to PowerPoint

MSCs Infusion Decreases the Survival of 5T33MM Diseased Mice

As MSCs were able to migrate toward MM sites and favor MM cell growth, we examined whether infusion of MSCs could affect MM disease progression and survival in vivo. 5T33MM mice were intraveneously injected with mMSCs on days 6, 10, and 14 after tumor cell inoculation. MSCs infusion led to a significant decrease in survival, to a mean of 18 days in the mMSCs treated MM mice compared to 22 days in vehicle-treated MM mice (n = 10 per group; p = .022; Fig. 5C). No particular side effects of mMSCs infusion were observed.

To confirm that MSCs accelerate the development of MM in 5T33MM mice, we performed an additional MSC infusion experiment. 5T33MM mice were injected with mMSC or vehicle at the same time as shown above. When the first mouse showed signs of morbidity, all mice were killed, and serum paraprotein, the indicator for MM disease progression, was measured (n = 10 per group). As shown in Figure 5D, the MM mice with MSC infusion had significantly higher paraprotein levels compared to the mice with vehicle treatment, indicating that MSC infusion did accelerate MM disease progression.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

A basic feature of MM cells concerns their reciprocal interaction with surrounding stromal cells which constitutes the so-called “MM tumor microenvironment.” Cellular components of the MM tumor microenvironment include osteoclasts, endothelial cells, macrophages, fibroblasts, and others. These cells play a very essential role in supporting the proliferation, survival, chemoresistance, and migration of MM cells. Here, we demonstrate that MSCs, the precursors of most BM stromal cells, can be recruited by MM cells and have a direct impact on tumor expansion and disease progression (Fig. 6).

thumbnail image

Figure 6. Schematic model of MSCs effects on MM disease. After systemic infusion, in vitro-expanded MSCs are able to migrate toward MM sites, attracted by MM cell-secreted CCL25 and/or other chemokines. Once MSCs localize in MM sites, by means of direct cell–cell contact and/or a paracrine mechanism, they can favor MM cell growth, protect MM cells against apoptosis, and induce drug resistance. Hence, MM disease progression and relapse can be accelerated. Abbreviations: MSC, mesenchymal stromal cell; MM, multiple myeloma.

Download figure to PowerPoint

Many studies in tumor models indicated that MSCs, delivered intravenously or intraarterialy, have preferential tropism toward tumor sites, primarily as a result of tumor cell-secreted paracrine factors and/or damaged tissue-produced inflammatory mediators, and therefore some reports have shown the use of MSCs as vehicles for cancer gene therapy [50–53]. To increase the therapeutic efficacy of MSCs in these cases, some intervention measures, including irradiation, hypoxic preconditioning, and genetic modification, have been used to enhance MSCs migratory capacity [54–56].

Using an in vitro transwell system, we found in this study that both mMSCs and hMSCs can migrate in response to MM cells. In the in vivo 5TMM mouse model, we demonstrated that fluorochrome-labeled mMSCs have the ability to home into MM cell invaded sites after systemic administration. Although a major part of mMSCs were found 7 days postadministration in lungs and liver, there were remarkably more mMSCs detectable in spleen and tibia of 5TMM diseased mice as compared to naive mice. Moreover, the number of migratory mMSC in spleen and tibia in 5TMM diseased mice was consistent with MM disease progression as determined by the serum paraprotein level. To elucidate the mechanism of MSC recruitment by MM cells, a chemokine array of MM cells was performed. Among all the chemokines tested, CCL25, also called TECK, was found to be secreted by RPMI8226 MM cells at the highest level. CCL25, a CC chemokine expressed predominantly in thymus and epithelium of the small intestine, mediates chemotaxis of T cells through its counter receptor CCR9 [57]. In patients with inflammatory bowel disease, it was found that there were increased numbers of CCR9 (+) lymphocytes circulating in the peripheral blood [58]. In some tumor models, such as melanoma, prostate cancer, breast cancer, and ovarian cancer, the expression and activation of CCR9 affect cancer cell migration and invasion, which can modulate cancer cells metastasis to distal sites that express CCL25 [59–62]. Previous studies have demonstrated that CCL25 can function as a chemoattractant for MSCs and periosteal progenitor cells in vitro [63, 64]. However, the role of tumor cell-produced CCL25 in the attraction of other cell types has not been reported so far. In this study, PCR analysis revealed that CCL25 was expressed widely in primary human MM cells and MM cell lines. A previous study has shown that both human and murine BM-derived MSCs express CCR9 [65]. Subsequently, we confirmed this expression and used siRNA to downregulate CCR9 expression on hMSCs. We observed that hMSCs transfected with CCR9 siRNA exhibited a profoundly lower migratory ability in response to CM of MM cells in vitro. In addition, CCR9 knocked-down mMSCs exhibited significantly lower migration ability toward MM-invaded spleen of 5T33MM mice in vivo. Both in vitro and in vivo loss of function studies indicate that CCL25 is involved in MM cell-mediated chemotaxis of MSCs. It is worthy of being noticed that MSCs express chemokine receptors also for other chemokines produced by MM cells. These include CCR1, CCR3, CCR5, and CXCR3, which can interact with MM cell-produced CXCL9 (MIG), CXCL10 (IP-10), CCL5 (RANTES), and CCL26 (Eotaxin-3) [65]. Some of these chemokines/chemokine receptors might play a role in MSCs recruitment by MM as well, but this needs to be further explored.

Some studies have shown enhancement of tumor growth by MSCs, potentially through immunomodulatory and/or proangiogenic properties of these stem cells. However, other studies have demonstrated that inhibition of tumor growth and extended survival could not show any apparent effect [20–31]. Which effect MSCs might have on the tumor growth and disease progression in MM is currently unclear. In this study, we found through in vitro coculture assays that MSCs favor the proliferation of stroma-dependent MM cells by secreting soluble factors and cell–cell contact. This growth-promoting effect was also confirmed by coengraftment experiments in vivo. After 2 weeks of coengraftment of MSCs and MM cells intrafemorally, there are more than twice MM cells detectable in the in situ femur compared to MM cells injection alone. Additional PCNA immunofluorescence staining on the in situ femur confirmed that MM cells have a higher proliferation ability when MSCs are coinjected intrafemorally

Besides favoring MM cell growth, MSCs were also found to protect MM cells against apoptosis, induced by serum starvation or by the potent anti-MM drug Bortezomib, as shown by in vitro Annexin V/7-AAD FACS analysis and by in vivo TUNEL immunofluorescence staining. Recent evidence showed that survivin was involved in the antiapoptotic effect of MSCs on myeloma cells [66]. In the coculture system, we observed that MM cells tended to grow adherently on MSCs, and direct contact could influence MM cells growth and their differentiation phenotype [67]. It has been shown previously that very late antigen 4, or integrin α4/β1 complex, is one of the most important surface molecules for MM cell homing, adhesion, growth, survival, and resistance to chemotherapy, mediated by its ligand VCAM-1 on BM stromal cells [68–70]. However, blocking VCAM-1 on mMSCs could neither decrease MM cell adhesion significantly nor impair the supporting effect on proliferation of 5T33MMvv myeloma cells (data not shown), indicating that other adhesion molecules appear to be involved in the interaction of MSCs and MM cells.

The functional crosstalk between MM cells and BM stromal cells has been emphasized previously [47–49]. It has been demonstrated that MSCs can produce a variety of cytokines and growth factors, such as IL6, IL-10, hepatocyte growth factor, stem cell factor, VEGF, IGF-1, and tumor necrosis factor α, all of which have evidence to favor MM cell tumorigenesis [71]. We found in this study that after indirect coculture with MM cells in vitro, MSCs could express increased levels of IL-6, IGF-1, VEGF, IL-10, and DKK1. There is evidence showing that MM patient-derived MSCs (MM-MSCs) have the ability to express higher levels of cytokines and growth factors, and they have a distinct transcriptional pattern, resulting in impaired immunomodulary effect and elevated angiogenic activity compared with their normal counterparts [72–77]. Our study suggests that at least some of these abnormalities of MM-MSCs are “inducible” by the interaction with MM cells. Meanwhile, MSCs, in turn, can favor MM cell growth and protect MM cells against spontaneous and drug induced apoptosis. It is believed that PI3K/AKT and MAPK signaling pathways are important in antiapoptosis and proliferation of MM cells [78, 79]. After coculture with MSCs, MM cells exhibit an upregulation of phosphorylated AKT and ERK activity, accompanied with increased cyclinD2, CDK4, and Bcl-XL, and decreased cleaved Caspase-3 and PARP.

A very recent report indicates that placenta-derived adherent cells, which are mesenchymal-like stem cells isolated from postpartum human placenta, effectively suppress bone destruction and tumor growth in an in vivo severe combined immunodeficiency–rab mouse myeloma model although these stem cells significantly support MM cell growth in vitro [80]. We speculate that some factors like differences in cell source (placenta vs. BM) and the nature of the in vivo MM model used can explain the discrepancy between these published findings and the data presented in our study.

Given recent findings that infusion of MSCs is a promising way to support hematopoietic recovery and control of graft versus host disease after allogeneic hematopoietic stem cell transplantation, we have to consider that this type of cell therapy might be not be safe in MM. Recently, a randomized clinical trial indicated that infusion of MSCs results in an increased frequency of disease relapse in patients with hematological malignancies (acute myeloid leukemia, chronic myelogenous leukemia, acute lymphoblastic leukemia, myelodysplastic syndrome, and non-Hodgkin lymphoma) [81]. Although MM patients were not included in this clinical trial, the decrease in survival we observed in this study using an in vivo MM mouse model indicates that systemic infusion of MSCs might contribute to MM patient disease progression. Infused MSCs might migrate to the patients' BM and favor the growth of residual MM cells. As MSCs are also immunosuppressive, both effects (growth stimulation and immune suppression) might probably lead to MM disease relapse. We therefore suggest that experimental MSC-based cell therapies in MM patients should be considered with extreme caution.

CONCLUSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

We demonstrated that MM cells could induce in vitro as well as in vivo chemotaxis of MSCs through CCL25/CCR9 axis, and MSCs play a supportive role in MM cell growth. Therefore, our data suggest that MSC-based cytotherapy might have a potential risk for favoring MM disease progression or relapse in MM patients.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

We thank Nicole Arras, Wim Renmans, Angelo Willems, Carin Seynaeve, Prof. Herman Tournaye, and Dr. Ning Liang for their expert technical assistance, and Prof. Zhou Qinghua for his administrative support. Our research work is supported by grants from the FWO-Vlaanderen, Vlaamse Liga tegen Kanker (Stichting Emmanuel Van der Shueren), the Vrije Universiteit Brussel (HOA), and the Scientific Foundation Willy Gepts (WFWG) UZ Brussel. Xu is supported by CSC-VUB scholarship.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

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

FilenameFormatSizeDescription
STEM_787_sm_supplfigure1.pdf484KSupplemental Figure 1. Characterization of human and murine MSCs. (A) hMSCs at passage 2 showed typical fibroblast like morphology, and were positive for CD90, CD166, CD73 and negative for CD45. When plated in corresponding differentiation media, hMSCs were capable of differentiating into adipocytes, osteoblasts and chondrocytes, evaluated respectively by Oil Red O staining (×200), Alizarin Red S staining (×100) and collagen type II staining (×100). (B) Using our optimized protocol, mMSCs already appeared morphologically homogeneous at passage 2. They were positive for CD105, CD90, Sca-1, lacked expression of CD45, and also exhibited tri-lineages differentiation potentials. Representative pictures of five independent experiments are shown. Scale bar=100μm. Plots show isotype control IgG staining profile (red) versus specific Ab staining profile (white)
STEM_787_sm_supplfigure2.pdf131KSupplemental Figure 2. Migration of mMSCs towards lung and liver over time. After inoculation of mMSCs i.v. in naive and 5T33MM mice, the cell number in the lungs gradually decreased, while it increased in the liver.
STEM_787_sm_supplfigure3.tif2110KSupplemental Figure 3. Murine 5T33MMvv and 5T33MMvt cells were both positive for CCL25 by PCR analysis.
STEM_787_sm_supplfigure4.pdf300KSupplemental Figure 4. MSCs favor the proliferation of MM cells in vivo. 5T33MMvv cells (5×106) were injected intrafemorally with or without mMSCs (5×105). One week later, more MM cells were in a proliferative state as shown by positive PCNA expression in the femur where mMSCs were co-injected compared to tumor cell injection alone. The slides were counterstained by DAPI. One representative photo is shown. The scale bar=100μm.
STEM_787_sm_supplfigure5.pdf282KSupplemental Figure 5. Minimally detectable MSCs were present in the MM cell population after separation from cell-cell contact coculture. (A) More than 99% hMSCs are positive for CD90. (B) RPMI8226 cells are negative for CD90 (middle). After 48h of cell-cell contact coculture (MSCs: MM=1:10), MM cells were collected by gently pipetting to separate them from adherent hMSCs, and used to test apoptosis and indicated protein expression. In the collected MM cell fraction, there was minimal contamination of detectable MSCs (right).
STEM_787_sm_supplfigure6.pdf624KSupplemental Figure 6. MSCs protect MM cells against apoptosis in vivo. A. MSCs protected MM cells against spontaneous apoptosis in vivo. 5T33MMvv cells (5×106) were injected intrafemorally with or without mMSCs (5×105). One week later, there were less MM cells undergoing apoptosis in the femur where mMSCs were co-injected, compared to tumor cell injection alone, although this was not significant. TUNEL immunofluorescence staining was performed, and the slides were counterstained by DAPI. One representative photo is shown. The scale bar=100μm. B. MSCs protected MM cells against Bortezomib induced apoptosis in vivo. After 5T33MMvv injection into naive mice intrafemorally, five doses of Bortezomib at 0.6mg/kg i.p. every other day were given, leading to MM cell apoptosis as measured by TUNEL staining. However, with the co-injection of mMSCs, significantly less MM cells underwent apoptosis. One representative photo is shown. The scale bar=100μm.

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