Tel.: +81-82-257-5191, Fax: +81-82-257-5194
Cancer Cell Biology
Mesenchymal stem cells enhance growth and metastasis of colon cancer
Article first published online: 6 MAY 2010
Copyright © 2010 UICC
International Journal of Cancer
Volume 127, Issue 10, pages 2323–2333, 15 November 2010
How to Cite
Shinagawa, K., Kitadai, Y., Tanaka, M., Sumida, T., Kodama, M., Higashi, Y., Tanaka, S., Yasui, W. and Chayama, K. (2010), Mesenchymal stem cells enhance growth and metastasis of colon cancer. Int. J. Cancer, 127: 2323–2333. doi: 10.1002/ijc.25440
- Issue published online: 6 MAY 2010
- Article first published online: 6 MAY 2010
- Manuscript Accepted: 28 APR 2010
- Manuscript Received: 3 JAN 2010
- Ministry of Education, Culture, Science, Sports and Technology of Japan (Grants-in-Aid for Cancer Research)
- Ministry of Health, Labor and Welfare of Japan
- mesenchymal stem cells;
- carcinoma-associated fibroblasts;
- orthotopic colon cancer model;
- tumor microenvironment
Recently, mesenchymal stem cells (MSCs) were reported to migrate to tumor stroma as well as injured tissue. We examined the role of human MSCs in tumor stroma using an orthotopic nude mice model of KM12SM colon cancer. In in vivo experiments, systemically injected MSCs migrated to the stroma of orthotopic colon tumors and metastatic liver tumors. Orthotopic transplantation of KM12SM cells mixed with MSCs resulted in greater tumor weight than did transplantation of KM12SM cells alone. The survival rate was significantly lower in the mixed-cell group, and liver metastasis was seen only in this group. Moreover, tumors resulting from transplantation of mixed cells had a significantly higher proliferating cell nuclear antigen labeling index, significantly greater microvessel area and significantly lower apoptotic index. Splenic injection of KM12SM cells mixed with MSCs, in comparison to splenic injection of KM12SM cells alone, resulted in a significantly greater number of liver metastases. MSCs incorporated into the stroma of primary and metastatic tumors expressed α-smooth muscle actin and platelet-derived growth factor receptor-β as carcinoma-associated fibroblast (CAF) markers. In in vitro experiments, KM12SM cells recruited MSCs, and MSCs stimulated migration and invasion of tumor cells through the release of soluble factors. Collectively, MSCs migrate and differentiate into CAFs in tumor stroma, and they promote growth and metastasis of colon cancer by enhancing angiogenesis, migration and invasion and by inhibiting apoptosis of tumor cells.
Mesenchymal stem cells (MSCs) are characterized by their ability to self-renew and differentiate into tissues of mesodermal origin, including bone, cartilage and adipose and connective tissues. Thus, they contribute to tissue regeneration.1 MSCs are recruited from bone marrow to inflamed or damaged tissues by local endocrine signals, resulting in the formation of fibrous scars.2, 3 Tumor tissue contains abundant growth factors, cytokines and matrix-remodeling proteins, explaining why tumors are likened to wounds that never heal.4 MSCs are reported to migrate to tumor sites as well as sites of injury and to incorporate into tumor stroma, but the effects of interactions between MSCs and tumor cells and the mechanisms underlying these effects remain unclear. Recent coinjection experiments revealed that MSCs promote tumor growth and metastasis.5–13 Reports suggest that MSCs are involved in tumor invasion and angiogenesis,5–7, 14 immunosuppression8, 9 and inhibition of apoptosis.11 Mishra et al. reported that MSCs can differentiate into carcinoma-associated fibroblast (CAF)-like cells by prolonged exposure to tumor-conditioned medium and that these cells promote tumor growth.12, 13 Theirs was the first report to show the relations between cancer cells, MSCs and CAFs in detail. Because accumulating evidence suggests that CAFs indeed promote the growth of tumors,15–20 the hypothesis that CAFs originate from MSCs may have interesting clinical implications.
In most coinjection studies concerning the effect of MSCs on tumors, subcutaneous ectopic transplantation models were used, but these models are considered insufficient for examining tumor–stroma interactions.21 The influence of the organ microenvironment on the biology of tumor cells has been recognized since Paget's “seed and soil” hypothesis, which suggests that multiple interactions between tumor cells and specific organs determine whether metastasis will occur.22 Organ-specific factors can influence the growth, vascularization, invasion and metastasis of human neoplasms.23–25 Thus, we examined tumor–MSC interactions in the tumor microenvironment using a mouse orthotopic transplantation model of human colon cancer.
We found that circulating MSCs migrate not only to the orthotopic colon tumor but also to the metastatic liver tumors. In addition, we showed by coinjection study that MSCs promote the growth and metastasis of colon cancer and that MSCs incorporated into the tumor stroma express CAF markers. This is the first reported study to show the significance of tumor–MSC interactions in the growth and metastasis of colon cancer.
Material and Methods
Isolation and culture of human MSCs
MSCs were obtained from the iliac crest and plated in a dish with Dulbecco's Modified Eagle's Medium (DMEM) supplemented with 10% fetal bovine serum (FBS), L-glutamine and a penicillin–streptomycin mixture according to a protocol approved by the Ethics Committee of Hiroshima University Graduate School of Medicine, as described previously.26 Nonadherent cells were removed after 72 hr, and adherent cells were detached from the plates and subcultured every 4–5 days in fresh medium supplemented with 1 ng/ml fibroblast growth factor-2.27 Aliquots from passages 3–5 were frozen in liquid nitrogen for future use.
Characterization of human MSCs in vitro
In culture medium, MSCs formed a monolayer of adherent cells and looked like long spindle-shaped fibroblastic cells. The capacity for chondrogenic, adipogenic and osteogenic differentiation was confirmed with the use of a Human Mesenchymal Stem Cell Functional Identification Kit (R&D Systems, Minneapolis, MN). Cell surface antigens on these cells were analyzed by fluorescence-activated cell sorting, and we confirmed that the cells were positive for CD29, CD44, CD73, CD90, CD105, CD166 and MHC-DR, but negative for CD14, CD34 and Flk-1, as described previously.26
Colon cancer cell line and culture conditions
The human colon cancer cell line KM12SM28, 29 was kindly gifted by Dr. Isaiah J. Fidler (University of Texas). Cells were maintained in DMEM supplemented with 10% FBS, L-glutamine and a penicillin–streptomycin solution. The cultures were maintained for no longer than 12 weeks after recovery from frozen stock.
Animals and transplantation of tumor cells
Female athymic BALB/c nude mice were obtained from Charles River Japan (Tokyo, Japan). The mice were maintained under specific pathogen-free conditions and used when 8 weeks old. Study was carried out after permission was granted by the Committee on Animal Experimentation of Hiroshima University.
To produce cecal tumors, KM12SM cells in 50 μl of Hanks' balanced salt solution (HBSS) were injected into the cecal wall of nude mice under a dissecting microscope as described previously.28 To produce experimental liver metastases, the cells were injected into the spleen of nude mice as described previously.29
Assessing migration of MSCs in vivo
To determine whether circulating MSCs have the ability to migrate to the orthotopic colon tumor, 1.0 × 106 KM12SM cells were transplanted into the cecal wall of three mice on day 0. Three weeks after tumor cell transplantation (on day 21), each mouse underwent injection of 1.0 × 106 PKH26 (Sigma)-labeled MSCs (in 200 μl of HBSS) into the tail vein. One week after this injection (on day 28), the mice were killed and necropsied.
To determine whether circulating MSCs also have the ability to migrate to the metastatic liver tumor, 0.5 × 106 KM12SM cells were transplanted into the spleen of three mice on day 0. One week after tumor cell transplantation (on day 7), each mouse underwent injection of 1.0 × 106 PKH26-labeled MSCs (in 200 μl of HBSS) into the tail vein. Three weeks after this injection (on day 28), the mice were killed.
Tumors were removed and stored in OCT Compound (Sakura Finetek Japan, Tokyo, Japan) and then snap-frozen in liquid nitrogen and stored at −80°C until tissue processing. Sections of PKH26-labeled tissues were analyzed by means of fluorescence confocal microscopy.
Examining the effect of MSCs on tumor growth in orthotopic colon tumors
To examine the effect of MSCs on tumor growth at the orthotopic site, coinjection studies were carried out. Mice were divided into three groups and underwent injection of (a) KM12SM cells alone (0.5 × 106, n = 24), (b) KM12SM cells mixed with MSCs [KM12SM:MSCs (0.5 × 106:1.0 × 106, ratio of 1:2, n = 28)], or (c) MSCs alone (1.0 × 106, n = 10).
Six weeks after intracecal transplantation of these cells, surviving mice were killed and necropsied [(a) n = 21, (b) n = 12, (c) n = 10]. Tumor weights, incidences of liver metastasis, and survival rates were evaluated. One part of the tumor tissue from each mouse was fixed in formalin-free IHC Zinc Fixative (PharMingen, San Diego, CA) and embedded in paraffin, and the other part was embedded in OCT Compound, rapidly frozen in liquid nitrogen and stored at −80°C.
Examining differentiation of MSCs commingled with tumor cells in orthotopic colon tumors
To evaluate whether commingled MSCs can differentiate into CAF-like cells, KM12SM cells mixed with PKH26-labeled MSCs (0.5 × 106:1.0 × 106, ratio of 1:2) were injected into the cecal wall of 3 mice. Three weeks after intracecal transplantation, tumors were excised. Excised tumors were analyzed by means of fluorescence confocal microscopy to detect PKH26-labeled MSCs after immunofluorescence staining.
Assessing the effect of MSCs on liver metastasis and differentiation of MSCs in a liver metastasis model
To highlight the effect of MSCs on liver metastasis, we developed a model of liver metastasis by injecting tumor cells into mice spleen as described previously.28, 29 Mice were divided into two groups: (a) those injected with KM12SM cells alone (0.5 × 106, n = 10), and (b) those injected with KM12SM cells along with PKH26-labeled MSCs [KM12SM:PKH26-labeled MSCs (0.5 × 106:1.0 × 106, ratio of 1:2, n = 10)].
Four weeks after intrasplenic transplantation, tumor nodules on the liver surface were counted macroscopically. To examine whether MSCs can migrate from the primary site to the metastatic site and differentiate, metastatic tumors were analyzed by means of fluorescence confocal microscopy to detect PKH26-labeled MSCs after immunofluorescence staining.
The primary antibodies used were rabbit anti-platelet–derived growth factor receptor (PDGFR)-β (Santa Cruz Biotechnology, Santa Cruz, CA), rabbit anti-α-smooth muscle actin (SMA) (Abcam, Cambridge, UK), mouse anti-desmin (Molecular Probes, Eugene, OR), rabbit anti-fibroblast activation protein (FAP) (Abcam, Cambridge, UK), rabbit anti-fibroblast specific protein (FSP) (Abcam, Cambridge, UK), mouse anti- proliferating cell nuclear antigen (PCNA) (Dako, Glostrup, Denmark) and rat anti-mouse CD31 (BD Pharmingen, San Diego, CA). Biotinylated rabbit anti-rat IgG (Dako) and biotinylated goat anti-mouse IgG (Dako) were used as secondary antibodies. The fluorescent secondary antibody was Alexa Fluor® 488-labeled goat anti-rabbit IgG (Invitrogen, Carlsbad, CA).
Immunohistochemical determination of PCNA, apoptotic cells and microvessel area (MVA)
Paraffin-embedded tissues cut into 4-μm sections and frozen tissues cut into 8-μm sections were used for immunohistochemical identification of PCNA and CD31, respectively. Immunohistochemistry was performed as described previously.30 The PCNA labeling index (PCNA-LI) was expressed as the ratio of positively stained tumor cells to the total tumor cells, given as a percentage for each case. Ten random areas without necrosis in a section were selected by light microscopy; at least 2,000 cells were counted under 400× magnification. Apoptotic cells in frozen sections of KM12SM tumors were detected by terminal deoxynucleotide transferase-mediated dUTP-biotin nick end labeling (TUNEL method) with the ApopTag Plus Peroxidase In Situ Apoptosis Detection Kit (Chemicon, Temecula, CA) according to the manufacturer's instructions. The apoptotic index (AI) was expressed as the ratio of positively stained tumor cells and apoptotic bodies to all tumor cells, given as a percentage for each case. Twenty random areas without necrosis in a section were selected by light microscopy; at least 5,000 cells were counted under 400× magnification. Angiogenic activity was evaluated according to the areas of microvessels stained with anti-mouse CD31 antibody. For quantification of the MVA, ten random fields at 100× magnification were captured for each tumor, and the outline of each microvessel including a lumen was manually traced. The area was then calculated with the use of NIH ImageJ software.
Frozen specimens cut into 8-μm sections or cells cultured on slide glass were fixed for 15 min in 4% paraformaldehyde in phosphate-buffered saline (PBS). The slides were blocked briefly in protein blocking solution, incubated overnight at 4°C with the Fab fragment of anti-mouse IgG to block endogenous immunoglobulins if necessary, and incubated overnight at 4°C with anti-α-SMA (1:400), anti-FAP (1:100), anti-FSP (1:100), anti-desmin (1:400) or anti-PDGFR-β (1:400). The slides were washed with PBS and then incubated for 1 hr at room temperature with Alexa Fluor® 488-labeled secondary antibody (1:600). Nuclear counterstain with 4′,6-diamidino-2-phenylindole (DAPI) was applied for 10 min, and mounting medium was placed on each specimen with a glass coverslip. α-SMA-, FAP-, FSP-, desmin- or PDGFR-β-positive cells were identified by green fluorescence, whereas PKH26 on MSCs was identified by red fluorescence. Colocalization of PKH26 and α-SMA, FAP, FSP, desmin or PDGFR-β was detected by yellow staining.
Confocal fluorescence images were captured with a 20× or 40× objective lens on a Zeiss LSM 510 laser scanning microscopy system (Carl Zeiss, Thornwood, NY) equipped with a motorized Axioplan microscope, argon laser (458/477/488/514 nm, 30 mW), HeNe laser (543 nm, 1 mW), HeNe laser (633 nm, 5 mW), LSM 510 control and image acquisition software and appropriate filters (Chroma Technology Corp., Brattleboro, VT). Confocal images were exported to Adobe Photoshop software, and image montages were prepared.
Collection of conditioned medium
To detect any paracrine effects of MSCs on KM12SM cells, conditioned medium from MSCs was collected and used for subsequent study. In detail, 1.0 × 106 MSCs were seeded in a 100-mm plate, and 48 hr later, 20 ml of DMEM with 0.5% FBS was added for 24-hr incubation. The medium was then collected, sterile filtered, aliquoted and stored at −20°C until use. DMEM with 0.5% FBS was used as control medium.
The proliferative effect of MSC-conditioned medium on KM12SM cells was analyzed. KM12SM (4.0 × 104) cells were seeded in a 12-well plate, and the next day, control medium or MSC-conditioned medium was added (day 0). The medium was changed on days 2 and 4, and the number of cells was counted on days 2, 4 and 7 (n = 6 in each group).
The migratory ability of KM12SM cells was assayed with the use of a 12-well microchamber plate with uncoated inserts (12-μm pore size, Corning Costar, Tokyo, Japan). Either 1.0 × 105 MSCs in DMEM with 0.5% FBS or the medium alone was plated into the lower chambers. After 24 hr of incubation at 37°C, upper chambers containing 1.0 × 105 KM12SM cells in DMEM with 0.5% FBS were set into the lower chambers. Three wells were used for each experiment. After 48 hr of incubation at 37°C, inserts were fixed with 10% buffered formalin solution and stained with hematoxylin. The cells on the upper surface of the membranes were removed with cotton swabs. The number of migrating cells was counted in three random fields per filter at 200× magnification.
The migratory ability of MSCs was also assayed by means of a 24-well microchamber plate with uncoated inserts (8-μm pore size, Corning Costar). Either 2.0 × 104 KM12SM in DMEM with 0.5% FBS or medium alone was plated into the lower chambers. After 4 hr of incubation at 37°C, upper chambers containing 2.0 × 104 MSCs in DMEM with 0.5% FBS were set into the lower chambers. Three wells were used for each experiment. After 16 hr of incubation at 37°C, inserts were fixed with 10% buffered formalin solution and stained with hematoxylin. The number of migrating cells was determined as described above.
The invasive ability of KM12SM cells was assayed by using a 24-well microchamber plate (12 μm pore size, Corning Costar) according to a previously described method,31 but with minor modifications. In brief, the upper surface of each membrane was coated with Matrigel (240 μg per filter, BD Biosciences, San Jose, CA). KM12SM (1.0 × 105) cells were seeded in the upper chamber with serum free medium, and MSC-conditioned medium or control medium was added to the lower chambers. Three wells were used for each experiment. After 46 hr of incubation at 37°C, inserts were treated as in the migration assay. The number of invading cells was counted in three random fields per filter at 100× magnification.
Survival curves were drawn by the Kaplan and Meier method, and the log rank test was used to analyze differences in survival rates. Student's or Welch's t-test or the Wilcoxon test was used to analyze differences in other variables, as appropriate. Data are expressed as mean ± standard error (SE). Probability values of <0.05 were considered significant. All statistical analyses were performed with JMP software (SAS Institute, Cary, NC).
Migration of MSCs to the stroma of orthotopic primary tumors and to metastatic liver tumors
After injection of PKH-labeled MSCs into the tail veins of tumor-bearing mice, MSCs were detected specifically in the tumor stroma (Fig. 1a) at the primary site. MSCs were also detected in the stroma of metastatic liver tumors (Fig. 1b). In contrast, MSCs were not detected in the noncancerous tissues, such as normal colonic mucosa and liver (Figs. 1c and 1d).
Promotion of the growth of orthotopic colon tumors by MSCs
At 6 weeks after transplantation of KM12SM cells alone, KM12SM cells mixed with MSCs, or MSCs alone into the cecal wall of nude mice, the weight of tumors resulting from mixed cells was significantly greater than the weight of tumors resulting from tumor cells alone (0.47 ± 0.10 vs. 1.03 ± 0.38 g, p < 0.05; Fig. 2a). The survival rate was significantly lower in the mixed-cell group than in the group that received KM12SM cells alone (p < 0.001; Fig. 2b). MSCs alone did not have the ability to generate any tumor and did not affect the survival of mice. Surviving mice were killed on day 42. Liver metastasis was not seen macroscopically in the 21 mice that received tumor cells alone, but liver metastasis was seen macroscopically in two of the 12 mice in the mixed-cell group (Fig. 2c). To clarify the mechanisms underlying the growth-promoting effect, we examined proliferation, apoptosis and angiogenesis in primary tumors by immunohistochemistry. The PCNA-LI, AI and MVA were compared between the tumors resulting from transplantation of KM12SM cells alone and those resulting from transplantation of KM12SM cells mixed with MSCs. The PCNA-LI was significantly higher (42.0 ± 5.3 vs. 62.9 ± 4.6 %, p < 0.05; Fig. 3a) and the AI was significantly lower in the mixed-cell group (7.0 ± 0.6 vs. 3.4 ± 0.3 %, p < 0.001; Fig. 3b). The MVA was significantly greater in the tumors from the mixed-cell group (13,719 ± 2,154 vs. 24,594 ± 2,230 μm2, p < 0.01; Fig. 3c).
Differentiation of MSCs in orthotopic colon tumors
Three weeks after transplantation of KM12SM cells mixed with PKH26-labeled MSCs, commingled MSCs were incorporated into the tumor stroma and expressed α-SMA, PDGFR-β, desmin, FSP and FAP as CAF markers (Fig. 4a), but they did not express human CD31 (data not shown). The morphology and distribution of MSCs within tumor stroma were similar to those of CAFs. Before the introduction of MSCs into mice, we examined the expression of α-SMA and PDGFR-β in MSCs during in vitro propagation. MSCs expressed PDGFR-β at a low level but not α-SMA (Fig. 4b).
Increase in the number of liver metastases by injection of KM12SM cells mixed with MSCs
Transplantation of KM12SM cells mixed with MSCs into the spleen of nude mice resulted in a significantly greater number of liver metastases at 4 weeks than that resulting from transplantation of KM12SM cells alone (2.6 ± 4.3 vs. 15.1 ± 6.0, p < 0.01; Fig. 5a). In addition, commingled MSCs migrated to the stroma of metastatic liver tumors and expressed α-SMA, PDGFR-β, desmin, FAP and FSP as CAF markers (Fig. 5b).
Attraction between KM12SM cells and MSCs in vitro
To investigate whether KM12SM cells have the ability to attract MSCs in vitro, migration assay was performed. We found that more human MSCs migrated toward the KM12SM cell culture than toward the medium without KM12SM cells (30.9 ± 3.5 vs. 24.2 ± 3.6 cells/field, p < 0.05; Fig. 6a).
We then examined the effects of MSCs on migration and invasion of KM12SM cells. The ability of KM12SM cells to migrate toward the MSC culture was significantly greater than the ability of these cells to migrate toward the medium without MSCs (114.9 ± 16.1 vs. 170.2 ± 18.5 cells/field, p < 0.05; Fig. 6b). Moreover, the invasive ability of KM12SM cells was significantly greater toward MSC-conditioned medium than toward control medium (33.1 ± 7.4 vs. 59.0 ± 7.9 cells/field, p < 0.05; Fig. 6c).
The effect of the soluble factors secreted from MSCs on the proliferation of KM12SM cells was also examined. KM12SM cells were exposed to control medium or MSC-conditioned medium and then counted on days 2, 4 and 7. No significant difference was observed in the number of KM12SM cells between the two groups at any time point (Fig. 6d).
It has been reported that MSCs migrate to a variety of tumors, such as melanomas,32, 33 gliomas34, 35 and colon,14, 36 pancreatic37, 38 and breast cancers.10, 33, 39, 40 Studies have implicated molecules such as CXCL12 (SDF-1)/CXCR4,36 CCL2 (MCP-1)/CCR239 and PDGF37 in the tumor-homing ability of MSCs. In the present study, we showed that circulating MSCs have the ability to migrate not only to the stroma of orthotopic colon tumors but also to metastatic lesions. In in vitro experiments, we observed that tumor cells recruit MSCs through the release of soluble factors. Although the exact molecular mechanisms that govern MSC migration are not fully characterized, this migratory ability points to MSCs as attractive candidates for delivery vehicles of antitumor agents.32–35 However, several coinjection experiments have revealed that MSCs promote tumor growth and metastasis,5–13 which would present a serious obstacle to using MSCs as delivery vehicles for anti-cancer therapy. Thus, the precise effects of MSCs on tumor growth and progression and the mechanism underlying these effects should be elucidated.
We previously reported that expression of metastasis-related genes by cancer cells and stromal cells is higher in orthotopic (cecal wall) KM12SM human colon carcinomas than in ectopic (subcutaneous) tumors.15, 16, 25 The stromal reaction and metastatic potential are increased in the orthotopic microenvironment. Therefore, orthotopic models should be used to study tumor–stroma or tumor–MSC interaction.21, 23, 41, 42 In our study, orthotopic transplantation of tumor cells mixed with MSCs into the cecal wall, in comparison to orthotopic transplantation of tumor cells alone, resulted in significantly greater tumor weight. In primary tumors, PCNA-LI (proliferation) and MVA (angiogenesis) were significantly increased, and the AI (apoptosis) was significantly lower in the tumors from the mixed-cell group. More intriguingly, macroscopic liver metastasis appeared only in the mixed-cell group, and the survival rate was significantly lower in this group. MSCs promoted migration and invasion of KM12SM cells in vitro. Because MSCs enhance proliferation, angiogenesis, migration and invasion and inhibit apoptosis of tumor cells, they may promote tumor growth and metastasis, which are followed by decreased survival. These findings are consistent with previously reported findings.6, 7, 11
In addition, we found commingled MSCs were functionally incorporated into the stroma of orthotopic colon tumors, where they expressed α-SMA, PDGFR-β, desmin, FAP and FSP as CAF markers. These results appear to corroborate the in vitro findings of Mishra et al., who showed that MSCs exposed to tumor-conditioned medium for 30 days can act as CAF-like cells.12, 13 CAFs are known to promote tumor growth and metastasis, and phenotypes of CAFs are distinct from those of normal fibroblasts.43, 44 CAFs may arise from fibroblasts residing in local tissues,18 periadventitial cells including pericytes and vascular smooth muscle cells,45 endothelial cells46 or bone marrow–derived cells including various stem cells.47 MSCs differentiate by nature into tissues of mesodermal origin, so it is reasonable to suppose that CAFs derive from MSCs.
To more specifically determine the effect of MSCs on metastasis in vivo, we used an experimental liver metastasis model, and we found that MSCs enhanced the metastatic potential of KM12SM cells. The number of liver metastases was increased by coimplantation with MSCs; thus, MSCs may affect colonization step and maintain cancer cells. MSCs commingled with tumor cells at the primary tumor site (spleen) were able to travel to the stroma of metastatic foci in the liver, where they expressed CAF markers. These findings indicate that MSCs support tumor metastasis not only at the primary site but also at metastatic sites.
Karnoub et al. reported that CCL5 secreted from MSCs acts on the CCR5 of breast cancer cells only at the primary site and temporarily promotes lung metastasis without influencing primary tumor growth and neoangiogenesis10; in fact, MSCs did not exist at the sites of lung metastasis. They speculated that the enhanced metastatic ability was due to enhanced extravasation and colonization of cancer cells at the metastatic site. They also showed that MSCs promoted lung metastasis of all breast cancer cell lines used in their study, but the growth kinetics of MSC-containing tumors at the primary sites differ between each breast cancer cell line. Thus, differences in experimental conditions, such as the kinds of cell lines and the sites (ectopic or orthotopic) of transplantation, may influence growth and angiogenesis of the primary tumors.
Our data indicate that MSCs may directly encourage tumor cells at the primary site to metastasize to the liver and that metastasized tumor cells may in turn recruit MSCs to metastatic sites where the MSCs differentiate into supporting CAF-like cells. Understanding the molecular mechanisms of interaction between tumor cells and MSCs could lead to establishment of new therapies for targeting tumor stroma at both primary and metastatic sites. Orthotopic transplantation models should be used in in vivo studies of tumor–MSC interactions for the design of anticancer therapies.
The authors thank Ms. Megumi Wakisaka, Mr. Shinichi Norimura, and Ms. Emiko Hisamoto for their excellent technical assistance. This work was carried out with the kind cooperation of the Analysis Center of Life Science, Hiroshima University (Hiroshima, Japan).