Secretion of the powerful angiogenic factor MFG-E8 by pericytes can bypass the therapeutic effects of anti-VEGF therapy, but the mechanisms by which MFG-E8 acts are not fully understood. In this study, we investigated how this factor acts to promote the growth of melanomas that express it. We found that mouse bone marrow–derived mesenchymal stromal cells (MSC) expressed a substantial amount of MFG-E8. To assess its expression from this cell type, we implanted melanoma cells and MSC derived from wild type (WT) or MFG-E8 deficient [knockout (KO)] into mice and monitored tumor growth. Tumor growth and M2 macrophages were each attenuated in subjects coimplanted with KO-MSC compared with WT-MSC. In both xenograft tumors and clinical specimens of melanoma, we found that MFG-E8 expression was heightened near blood vessels where MSC could be found. Through in vitro assays, we confirmed that WT-MSC–conditioned medium was more potent at inducing M2 macrophage polarization, compared with KO-MSC–conditioned medium. VEGF and ET-1 expression in KO-MSC was significantly lower than in WT-MSC, correlating in vivo with reduced tumor growth and numbers of pericytes and M2 macrophages within tumors. Overall, our results suggested that MFG-E8 acts at two levels, by increasing VEGF and ET-1 expression in MSC and by enhancing M2 polarization of macrophages, to increase tumor angiogenesis. Cancer Res; 76(14); 4283–92. ©2016 AACR.

Mesenchymal stromal cells (MSC) are bone marrow–derived nonhematopoietic pluripotent progenitor cells with the capacity to differentiate into various cell types, including chondrocytes, adipocytes, and osteocytes (1–3). Recent evidence indicates that pericytes and MSC are similar cells that are located external to the vasculature and are involved in angiogenesis, repair, and tissue maintenance (4–6). It has been reported that malignant tumor cells, such as malignant glioma and melanoma, can recruit MSC from surrounding tissue or the circulation and stimulate the growth of MSC by the secretion of soluble factors, including platelet-derived growth factor (PDGF; refs. 7–9). These tumor-associated MSCs secrete growth factors or cytokines, resulting in the promotion of angiogenesis (7–9). In addition, several studies have reported that MSC can differentiate into fibroblasts, myofibroblasts, or pericyte-like cells and enhance angiogenesis, resulting in tumor progression and metastasis in vivo (10–14). MSCs also have an immunosuppressive function and may help tumor escape from immune surveillance (11). Tumor-resident and injected MSCs have been demonstrated to promote the recruitment of tumor-associated macrophages (TAM; refs. 15, 16). These findings have led to an increased interest in understanding the role of MSC in tumor growth and the potential of MSC to serve as therapeutic targets in melanoma.

Milk fat globule EGF factor 8 (MFG-E8) is a secreted glycoprotein and consists of two EGF-like domains and two discoidin-like domains with sequence homology to the blood coagulation factors V and VIII (17, 18). The second EGF-like domain contains an RGD motif that binds to integrin αvβ3/5 (18, 19). The carboxy-terminal domains of MFG-E8 can bind to negatively charged phospholipids (20), resulting in the opsonization of apoptotic cells for uptake by phagocytes (19, 21). In addition, interactions between MFG-E8 and integrin αv have been implicated in the enhancement of angiogenesis in mice (22–24).

Many studies have indicated that MFG-E8 enhances tumor cell survival, invasion, and angiogenesis and contributes to local immune suppression (23, 25–29). In a murine melanoma model, MFG-E8 enhanced the tumorigenicity and metastatic capacity through Akt and Twist-dependent pathways (26). In addition, MFG-E8 produced by TAM in melanomas activated STAT3 and sonic hedgehog pathways in cancer stem cells (27). Furthermore, systemic MFG-E8 blockade using an anti-MFG-E8 antibody cooperates with cytotoxic chemotherapy, molecular targeted therapy, and radiation to induce the destruction of murine tumors, including melanoma (28).

In prior studies, we demonstrated that pericytes and/or pericyte precursors were important sources of MFG-E8 in melanoma tumors, and pericytes/pericyte precursor-derived MFG-E8 enhanced angiogenesis in melanoma tumors (29). MFG-E8 associated with integrin αv and PDGF receptor β (PDGFRβ) on the surface of pericytes/pericyte precursors after PDGF treatment, and the formation of this complex inhibited degradation of PDGFRβ, resulting in the enhancement of PDGFRβ signaling (30). These findings suggest that MFG-E8–expressing perivascular cells regulate angiogenesis and tumor growth in melanoma by potentiating PDGFR signaling in perivascular cells. Considering that pericytes and MSC are similar perivascular cells, and that tumor-recruited MSCs secrete proangiogenic factors, we hypothesized that MSC might express a substantial amount of MFG-E8 and that MFG-E8 might regulate the functions of MSC, including angiogenesis in melanoma tumors. To test this hypothesis, we analyzed the expression of MFG-E8 in bone marrow–derived MSC (BM-MSC) and compared the effects of MSC from MFG-E8 wild-type (WT) and knockout (KO) mice on melanoma tumor growth and vascularity.

Cell culture

The mouse melanoma cell line (B16-F10) and the mouse monocyte/macrophage cell line (RAW 264.7) were obtained directly from the ATCC between 2011 and 2015, and frozen aliquots of cells were prepared upon receipt. Cells were used within 6 months after receipt, and a link to the method of authentication is provided here: http://www.atcc.org/support/faqs/eae27/Authenticating%20cell%20lines-249.aspx.

Mice

MFG-E8 KO C57BL/6 mice were generated and genotyped as described previously (24, 29, 31). MFG-E8 KO mice were generated by interbreeding homozygous animals carrying the targeted MFG-E8 allele. Interbreeding homozygous C57BL/6 mice and C57BL/6-Tg (CAG-EGFP) mice were purchased from Japan SLC. Eight- to 12-week-old mice were used for all experiments.

Isolation and characterization of BM-MSC

Bone marrow cell suspensions were obtained from MFG-E8 WT/KO C57BL/6 female mice and cultured. Magnetic-activated cell sorting (Miltenyi Biotec) was performed to remove CD11b+ cells. For examination of surface expression of MSC markers, BM-MSCs were incubated consecutively with Alexa 488-conjugated anti-human Sca-1, CD105, CD44, CD45, CD11b Ab, or isotype control Ab (BioLegend) before flow cytometric analysis with a FACSCalibur instrument (BD Biosciences).

MSC differentiation assay

To analyze the differentiation potential of MSC, we used Mouse Mesenchymal Stem Cell Functional Identification Kit (R&D Systems) according to the manufacturer's instructions.

RNA extraction and real-time RT-PCR

Total RNA was isolated using RNeasy Mini Kits (Qiagen) and was subjected to reverse transcription with a SuperScript III First-Strand Synthesis System for RT-PCR (Invitrogen). Quantitative RT-PCR was performed via the TaqMan system (Applied Biosystems) using a 7300 Real-Time PCR machine (Applied Biosystems).

To inhibit MFG-E8 production, 5 × 105 MSC were transfected with 10 nmol/L siRNA using HiPerFect Transfection Reagent (Qiagen). At 24 hours after siRNA transfection, cells were incubated under hypoxic conditions (1% O2, 5% CO2, and 94% N2) for 24 hours, and then, RNA was extracted.

Melanoma xenograft model

B16F10 melanoma cells (2 × 105 cells) and MFG-E8 WT/KO-MSC (2 × 105 cells) were implanted subcutaneously into the flanks of MFG-E8 WT/KO mice. MSC was incubated under hypoxic conditions for 24 hours prior to implantation. Tumor sizes (width × length; mm2) were determined with calipers every 2 or 3 days. To analyze the localization of implanted MSC in melanomas, MSCs derived from bone marrow of GFP transgenic mice were incubated under hypoxic conditions for 24 hours, and then GFP-MSCs were implanted with B16F10 cells into MFG-E8 KO mice. MSC was labeled with CM-DiI (Molecular Probes) as described previously (32). CM-DiI–labeled MSCs were implanted with B16F10 cells into mice. To analyze the effect of macrophages depletion on tumor growth, 200 μL clodronate liposomes (Clophosome-A, FormuMax) or placebo control liposomes (FormuMax) was injected intraperitoneally at day 0, and then 100 μL clodronate liposomes or control was injected twice per week.

Immunofluorescence staining

Tumors (100 mm2) were excised from flank skin and fixed in 4% paraformaldehyde and 30% sucrose/H2O. After blocking, frozen sections (4 μm thick) were stained with the antibody of interest, followed by the secondary antibody conjugated with Alexa Fluor 488 or 568. Human melanoma tissues were used after surgical resection in the patients treated at the Department of Dermatology, of Gunma University Hospital (Maebashi, Japan). The study was approved by the Institutional Review Board of the Gunma University.

Macrophage differentiation assay

MFG-E8 WT/KO-MSCs (5 × 105 cells) were incubated for 24 hours under hypoxic conditions, and then conditioned media were collected. RAW 264.7 cells (1 × 106 cells) were incubated in 1 mL of MFG-E8 WT/KO-MSC–conditioned medium. After 48 hours of incubation, RNA was extracted. To assess the effect of recombinant MFG-E8 (rMFG-E8) on macrophage differentiation, RAW 264.7 cells were incubated with or without rMFG-E8 (500 ng/mL; R&D Systems) for 48 hours, and then RNA was extracted.

Bone marrow chimeric mice and tumor implantation

C57BL/6 MFG-E8 WT/KO-BM cells were collected from the femurs of mice by aspiration and flushing. Recipient C57BL/6 mice were irradiated with 12 Gy and then 5 × 106 MFG-E8 WT/KO-BM were injected intravenously. Eight weeks after bone marrow cell injection, B16F10 melanoma cells were implanted subcutaneously into the flanks of mice.

Statistical analysis

P values were calculated using the Student t test (two sided) or by analysis of one-way ANOVA followed by Bonferroni posttest as appropriate. Error bars represent SEs of the mean, and numbers of experiments (n) are as indicated.

Mice BM-MSC produced a substantial amount of MFG-E8

To examine the expression of MFG-E8 in MSC, mRNA and protein levels of MFG-E8 in mouse bone marrow cells, BM-MSC, 10T1/2 pericyte-like cells, and B16 melanoma cells were compared. We found that BM-MSC produced much more MFG-E8 mRNA and protein as compared with bone marrow cells, 10T1/2 cells, and B16 melanoma cells (Fig. 1A). Next, the surface markers of MFG-E8 WT- and KO-MSC were compared. Typical MSC markers, such as Sca-1, CD105, and CD44, were expressed in more than 80% of MSC, and negative markers, CD45 and CD11b, were not expressed (Fig. 1B). There were no differences between the surface marker profiles of WT- and KO-MSC (Fig. 1B). Next, the differentiation potential of MFG-E8 WT- and KO-MSC was compared. Adipogenic differentiation assessed by staining with Oil-red-O in MFG-E8 KO-MSC was significantly enhanced (Fig. 1C). The adipogenic marker FABP4 mRNA expression in KO-MSC was also markedly enhanced compared with that in WT-MSC (Fig. 1C). In contrast, osteogenic differentiation in KO-MSC was significantly suppressed (Fig. 1D). There was no obvious difference in the chondrogenic differentiation capacity of WT- and KO-MSC (Fig. 1E). These results suggest that MSC expressed a substantial amount of MFG-E8, and MFG-E8 might suppress adipogenic differentiation but promote osteogenic differentiation of MSC.

MSCs were localized around CD31+ endothelial cell and expressed pericytes marker NG2 and MFG-E8 in melanoma tumors

Next, we examined the localization of MSC and the expression of MFG-E8 in MSC in melanoma tumors coimplanted with MSC. B16 melanoma cells and GFP-MSC were coinjected into MFG-E8 KO mice. Recent studies have shown that hypoxia preconditioning improved the viability and tissue repair capacity of MSC after transplantation, leading to increased angiogenesis and the improvement of myocardial and brain ischemic model (33, 34). We found that hypoxic treatment significantly enhanced cell proliferation and MFG-E8 mRNA expression in MSC (Supplementary Fig. S1A and S1B). Therefore, MSCs were incubated under hypoxic conditions for 24 hours prior to implantation in this study. GFP-MSCs were localized around CD31+ endothelial cells (EC; Fig. 2A) and several GFP-MSCs expressed pericytes marker NG2 (Fig. 2B). MFG-E8 staining was weakly positive in melanoma cells and strongly positive in GFP-MSC and around the vessel-like structures in melanoma tumors (Fig. 2C). These results suggest that MSC might localize to the perivascular area, and MFG-E8 derived from MSC might be involved in MSC-induced tumor angiogenesis and tumor growth.

To examine the distribution of MFG-E8 WT- and KO-MSC in B16 tumor, MFG-E8 WT- and KO-MSC were labeled with CM-DiI and coimplanted with B16 melanoma cells into mice. CM-DiI–labeled WT- and KO-MSC were localized around CD31+ EC (Fig. 2D). There was no obvious difference in the distribution of WT- and KO-MSC in B16 tumors. In addition, several CM-DiI–labeled WT- and KO-MSC expressed NG2 (Fig. 2E). These results suggest that MFG-E8 in MSC may not be associated with the distribution of MSC in melanoma tumors.

MFG-E8 in MSC accelerated the growth of melanoma tumors

It has been reported that MSCs promote tumor growth through enhanced angiogenesis and immune modulation (10–14). In contrast, few studies have reported that MSCs inhibit tumor growth (35). Therefore, we initially examined the effect of MSC on melanoma tumor growth and found that MSC enhanced the growth of melanoma tumors compared with that of melanoma cells alone (Fig. 2F; B16 alone vs. B16 + WT- or KO-MSC). Next, both melanoma cells and MFG-E8 WT/KO-MSC were implanted subcutaneously into mice. We found that the enhancement of tumor growth by coimplantation of KO-MSC was less than that caused by coimplantation of WT-MSC (Fig. 2F; B16 + WT-MSC vs. B16 + KO-MSC). The enhancement of survival in mice coimplanted with melanoma cells and KO-MSC was also significantly less than that in mice coimplanted with WT-MSC (Fig. 2G). These results indicate that MFG-E8 in MSC might contribute to MSC-induced melanoma tumor growth.

MFG-E8 in MSC accelerated the number of pericytes in melanoma tumors

Next, we examined the vascularity in tumors and found that the numbers of NG2+ pericytes in melanoma tumors coimplanted with MFG-E8 WT-MSC were significantly higher than those in tumors coimplanted with KO-MSC (Fig. 3A and B). The numbers of CD31+ EC in melanoma tumors coimplanted with WT-MSC tended to be higher than those in tumors coimplanted with KO-MSC; however, this difference was not significant (Fig. 3A and B). The level of pericyte coverage, which was assessed by the pericytes/EC ratio in tumors coimplanted with WT-MSC, tended to be higher than those in tumors coimplanted with KO-MSC (Fig. 3B). These results suggest that MFG-E8 in MSC might accelerate angiogenesis, especially the number of pericytes in melanoma tumors.

MFG-E8 in MSC induced tumor-associated M2 macrophages

We next assessed the numbers of infiltrated TAM in melanoma tumors coimplanted with MFG-E8 WT- or KO-MSC. The numbers of CD68+, CD206+ M2 macrophages in melanoma tumors coimplanted with KO-MSC were significantly reduced compared with those in tumors coimplanted with WT-MSC (Fig. 4A), suggesting that MFG-E8 in MSC might be involved in the increased infiltration of TAM into melanoma tumors.

We next examined the effect of the depletion of macrophages on MFG-E8 WT-MSC-induced melanoma tumor growth. The number of CD68+ macrophages in melanoma tumor in mice treated with clodronate liposomes were significantly inhibited compared with that in mice treated with control liposomes (Supplementary Fig. S2A). We found that clodronate liposomes treatment inhibited tumor growth of B16 melanoma alone and coimplanted with MFG-E8 WT/KO-MSC (Fig. 4B and C; control + B16 alone, control + B16 + WT-MSC or KO-MSC vs. clodronate + B16 alone, clodronate + B16 + WT-MSC or KO-MSC). In addition, the enhancement of tumor growth and survival in mice by coimplantation of WT-MSC was partially reduced by clodronate liposomes treatment, but not completely attenuated (Fig. 4B and C and Supplementary Fig. S2B). These results suggest that TAM might partially contribute to the WT-MSC–induced tumor growth.

It has been reported that tumor-resident and injected MSC increased the infiltration of TAM and tumor growth (15, 16). In addition, M2 markers, arginase-1, CD206, and Ym1 in macrophages, including RAW264.7 cells were increased by culturing with MSC-conditioned medium (36–39). Therefore, we examined whether MFG-E8 plays a role in MSC-induced M2 macrophage polarization. The expression of iNOS, a M1 macrophage marker, in macrophages cultured with WT- or KO-MSC–conditioned medium were unchanged. Culturing RAW 264.7 cells with WT-MSC–conditioned medium increased mRNA and protein levels of arginase-1, CD206, and Ym1 expressions in macrophages; however, KO-MSC–conditioned medium did not (Fig. 4D and E). Next, to assess the direct effect of MFG-E8 on M2 macrophage polarization, rMFG-E8 was added to the medium of cultured macrophages. The addition of rMFG-E8 in the culture medium of RAW 264.7 macrophages did not affect the expression of M1 and M2 markers in macrophages (Fig. 4F). These results suggest that MFG-E8 in MSC might regulate MSC-induced M2 macrophage polarization, in conjunction with factors other than MFG-E8 that are also secreted from MSC.

Depletion of MFG-E8 decreased the production of ET-1 and VEGF in MSC

The effect of MFG-E8 on the cell proliferation of MSC was analyzed by an in vitro assay. The proliferation of MFG-E8 KO-MSC tended to be decreased in comparison with that in WT-MSC; however, the difference was not significant (Fig. 5A). We next examined the role of MFG-E8 in the production of angiogenic factors in MSC in vitro. The levels of VEGF and ET-1 mRNA in KO-MSC were significantly suppressed (Fig. 5B and C). The mRNA levels of angiopoietin in MFG-E8 WT/KO-MSC were unchanged (Fig. 5D). We also examined the role of MFG-E8 in MSC using MFG-E8 siRNA. MFG-E8 siRNA inhibited mRNA levels of MFG-E8 by 60% (Fig. 5E). VEGF and ET-1 mRNA levels in MFG-E8 siRNA-treated MSC were reduced compared with those in control siRNA-treated MSC (Fig. 5F and G). The mRNA levels of angiopoietin in control or MFG-E8 siRNA-treated MSC were unchanged (Fig. 5H). These results suggest that MFG-E8 might regulate the production of angiogenic factors, such as VEGF and ET-1, in MSC.

Contribution of MFG-E8+ bone marrow–derived cells to melanoma tumor growth, the number of pericytes, and tumor-associated M2 macrophages

We examined the contribution of bone marrow–derived MFG-E8+ cells to melanoma tumor growth using bone marrow chimeric mice generated from MFG-E8 WT and KO mice. Melanoma tumor growth in MFG-E8 KO-BM–transplanted mice was modestly inhibited beginning 12 days after inoculation compared with that in WT-BM–transplanted mice (Fig. 6A). The number of NG2+ pericytes in tumors in MFG-E8 KO-BM–transplanted mice was significantly decreased (Fig. 6B). In addition, the number of CD68+, CD206+ M2 macrophages in tumors was significantly reduced in KO-BM–transplanted mice (Fig. 6C). These results suggest that bone marrow–derived MFG-E8+ cells might contribute to the number of pericytes and TAM and melanoma tumor growth.

Perivascular expression of MFG-E8 in human melanoma tumors

Finally, we examined the distribution of MFG-E8 in human melanoma tumors. We found that MFG-E8 staining in human melanoma was primarily observed around blood vessels, especially in αSMA+ pericytes (Fig. 7A and B). MFG-E8 staining around the vessels was more frequently observed inside of melanoma tumors compared with outside of the tumors (Fig. 7A, bottom). These results suggest that the perivascular expression of MFG-E8 in human melanoma tumors might be associated with tumor angiogenesis and tumor growth.

We previously demonstrated that PDGFRβ+ pericytes/pericyte precursors are a predominant source of MFG-E8 in melanoma tumors (29). In this study, we demonstrated that BM-MSCs produce a substantial amount of MFG-E8. Several reports have proposed that the perivascular zone is the MSC niche in vivo and perivascular MSCs contribute to angiogenesis (4–6), suggesting that MSC-derived MFG-E8 might regulate the functions of MSC and contribute to tumor angiogenesis and growth.

We demonstrated that adipogenic differentiation in MFG-E8 KO-MSC was markedly enhanced, and osteogenic differentiation in KO-MSC was significantly suppressed. With respect to MFG-E8 and osteogenesis, Abe and colleagues reported that chronic periodontal bone loss occurred more frequently in MFG-E8 KO mice compared with that in WT mice (40). However, the mechanisms regulating MSC differentiation by MFG-E8 remain unknown.

We determined that GFP-MSC localized in the perivascular area and expressed NG2 and MFG-E8 in melanoma tumors. We also found that the number of pericytes and the level of pericyte coverage in melanomas coimplanted with MSC were higher than those in melanomas without MSC coimplantation. Interestingly, the increased number of pericytes by coimplantation with MSC was reduced in melanomas coimplanted with MFG-E8 KO-MSC as compared with MFG-E8 WT-MSC. These results suggest that MSCs might localize around the vessels and act as pericytes in melanoma tumors, and MFG-E8 in MSC might be involved in these mechanisms.

Regarding TAM and MFG-E8, MFG-E8 induced efferocytosis of apoptotic prostate cancer cells and promoted M2 phenotype polarization in macrophages (41), suggesting that MFG-E8 might regulate the functions of TAM. We demonstrated that the enhancement of melanoma tumor growth by coimplantation of WT-MSC was partially suppressed by macrophages depletion in vivo, suggesting that MFG-E8 in MSC might enhance tumor growth via the regulation of TAM. Furthermore, we determined that MFG-E8 KO-MSC reduced M2 macrophage infiltration in vivo and that MFG-E8 KO-MSC reduced MSC-induced M2 macrophage polarization in vitro. These results suggest that MFG-E8 in MSC might positively regulate TAM in melanoma tumors. MFG-E8 WT-MSC–conditioned medium induced M2 macrophage polarization; however, rMFG-E8 did not, suggesting that secreted factors other than MFG-E8 might also be required for M2 macrophage polarization. We further examined the protein levels of IL10 and IL4 in the conditioned medium of WT- and KO-MSC incubated under hypoxic conditions for 24, 48, and 72 hours by ELISA. However, IL10 and IL4 secretions were not detected in the supernatant of WT- and KO-MSC (data not shown), suggesting that IL10 and IL4 might not be associated with WT-MSC conditioned medium–induced M2 polarization of macrophages.

ET-1/ET receptor signaling has been shown to play a role in the growth and progression of melanoma (42, 43). In addition, ET-1 contributes to angiogenesis, extracellular matrix degeneration, and macrophage chemoattraction (44). According to these results, MFG-E8 might increase the expression of ET-1 and VEGF in MSC, resulting in enhancement of angiogenesis, macrophage infiltration, and tumor growth in melanoma.

Using bone marrow chimeric mice, we identified that bone marrow–derived MFG-E8+ cells contribute to increased numbers of pericytes, M2 macrophage infiltration, and melanoma tumor growth. Although bone marrow–derived MFG-E8+ cells include various types of cells, such as leukocytes, macrophages, and endothelial progenitor cells, BM-MSCs express a greater amount of MFG-E8 than other bone marrow–derived cells; therefore, we speculate that MFG-E8 in MSC may regulate the migration of MSC from bone marrow to tumors. Our previous results that MFG-E8 enhanced PDGF:PDGFRβ signaling and the migration of 10T1/2 MSC cells (29, 30) are consistent with this speculation. However, the current results using a tumor model in bone marrow chimeric mice do not prove that MFG-E8 regulates MSC migration in vivo, and further investigation is required.

Collectively, we propose the following model to clarify the role of MFG-E8 in MSC-induced angiogenesis, M2 macrophage polarization, and melanoma tumor growth (Fig. 7C). BM-MSC may migrate and localize around blood vessels in melanoma tumors. MSCs produce substantial amounts of MFG-E8, and MFG-E8 production is additionally induced by hypoxic conditions in tumors. MFG-E8 in MSC might increase the expression of VEGF and ET-1 in MSC and enhance M2 macrophage polarization, leading to the enhancement of angiogenesis and tumor growth. As previously reported, MFG-E8 also acts on pericytes and EC by potentiating the stimulatory effects of PDGF and VEGF, respectively, leading to enhanced angiogenesis (22, 30). This regulation of MSC by MFG-E8 might provide new insight into the pathogenesis of tumor growth in melanoma.

In human melanoma, MFG-E8 was predominantly expressed around blood vessels, especially in pericytes, suggesting that our concept may be involved in the pathogenesis of human melanoma progression. However, additional studies regarding the role of MSCs in human melanoma are warranted.

No potential conflicts of interest were disclosed.

Conception and design: S. Motegi

Development of methodology: S. Motegi

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): K. Yamada, A. Uchiyama, A. Uehara, B. Perera, S. Ogino, Y. Yokoyama, Y. Takeuchi, S. Motegi

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): K. Yamada, A. Uchiyama, A. Uehara, B. Perera, S. Ogino, Y. Yokoyama, Y. Takeuchi, O. Ishikawa, S. Motegi

Writing, review, and/or revision of the manuscript: K. Yamada, A. Uchiyama, M.C. Udey, O. Ishikawa, S. Motegi

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): S. Motegi

Study supervision: O. Ishikawa, S. Motegi

This work was financially supported by Japanese Dermatological Association research grant (Shiseido donation), Takeda Science Foundation, research grant 2013, and JSPS KAKENHI, grant number 24791135 and 26461654 (S. Motegi), and the Intramural Program of NIH, Center for Cancer Research, NCI (M.C. Udey).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Pittenger
MF
,
Mackay
AM
,
Beck
SC
,
Jaiswal
RK
,
Douglas
R
,
Mosca
JD
, et al
Multilineage potential of adult human mesenchymal stem cells
.
Science
1999
;
284
:
143
7
.
2.
Orlic
D
,
Kajstura
J
,
Chimenti
S
,
Jakoniuk
I
,
Anderson
SM
,
Li
B
, et al
Bone marrow cells regenerate infarcted myocardium
.
Nature
2001
;
410
:
701
5
.
3.
Sasaki
M
,
Abe
R
,
Fujita
Y
,
Ando
S
,
Inokuma
D
,
Shimizu
H
. 
Mesenchymal stem cells are recruited into wounded skin and contribute to wound repair by transdifferentiation into multiple skin cell type
.
J Immunol
2008
;
180
:
2581
7
.
4.
Göritz
C
,
Dias
DO
,
Tomilin
N
,
Barbacid
M
,
Shupliakov
O
,
Frisén
J
. 
A pericyte origin of spinal cord scar tissue
.
Science
2011
;
333
:
238
42
.
5.
Crisan
M
,
Yap
S
,
Casteilla
L
,
Chen
CW
,
Corselli
M
,
Park
TS
, et al
A perivascular origin for mesenchymal stem cells in multiple human organs
.
Cell Stem Cell
2008
;
3
:
301
13
.
6.
da Silva Meirelles
L
,
Caplan
AI
,
Nardi
NB
. 
In search of the in vivo identity of mesenchymal stem cells
.
Stem Cells
2008
;
26
:
2287
99
.
7.
Campagnoli
C
,
Roberts
IA
,
Kumar
S
,
Bennett
PR
,
Bellantuono
I
,
Fisk
NM
. 
Identification of mesenchymal stem/progenitor cells in human first-trimester fetal blood, liver, and bone marrow
.
Blood
2001
;
98
:
2396
402
.
8.
Karp
JM
,
Leng Teo
GS
. 
Mesenchymal stem cell homing: the devil is in the details
.
Cell Stem Cell
2009
;
4
:
206
16
.
9.
Hata
N
,
Shinojima
N
,
Gumin
J
,
Yong
R
,
Marini
F
,
Andreeff
M
, et al
Platelet-derived growth factor BB mediates the tropism of human mesenchymal stem cells for malignant gliomas
.
Neurosurgery
2010
;
66
:
144
56
.
10.
Suzuki
K
,
Sun
R
,
Origuchi
M
,
Kanehira
M
,
Takahata
T
,
Itoh
J
, et al
Mesenchymal stromal cells promote tumor growth through the enhancement of neovascularization
.
Mol Med
2011
;
17
:
579
87
.
11.
Han
Z
,
Tian
Z
,
Lv
G
,
Zhang
L
,
Jiang
G
,
Sun
K
, et al
Immunosuppressive effect of bone marrow-derived mesenchymal stem cells in inflammatory microenvironment favours the growth of B16 melanoma cells
.
J Cell Mol Med
2011
;
15
:
2343
52
.
12.
Karnoub
AE
,
Dash
AB
,
Vo
AP
,
Sullivan
A
,
Brooks
MW
,
Bell
GW
, et al
Mesenchymal stem cells within tumour stroma promote breast cancer metastasis
.
Nature
2007
;
449
:
557
63
.
13.
Bexell
D
,
Gunnarsson
S
,
Tormin
A
,
Darabi
A
,
Gisselsson
D
,
Roybon
L
, et al
Bone marrow multipotent mesenchymal stroma cells act as pericyte-like migratory vehicles in experimental gliomas
.
Mol Ther
2009
;
17
:
183
90
.
14.
Traktuev
DO
,
Prater
DN
,
Merfeld-Clauss
S
,
Sanjeevaiah
AR
,
Saadatzadeh
MR
,
Murphy
M
, et al
Robust functional vascular network formation in vivo by cooperation of adipose progenitor and endothelial cells
.
Circ Res
2009
;
104
:
1410
20
.
15.
Jia
XH
,
Du
Y
,
Mao
D
,
Wang
ZL
,
He
ZQ
,
Qiu
JD
, et al
Zoledronic acid prevents the tumor-promoting effects of mesenchymal stem cells via MCP-1 dependent recruitment of macrophages
.
Oncotarget
2015
;
6
:
26018
28
.
16.
Ren
G
,
Zhao
X
,
Wang
Y
,
Zhang
X
,
Chen
X
,
Xu
C
, et al
CCR2-dependent recruitment of macrophages by tumor-educated mesenchymal stromal cells promotes tumor development and is mimicked by TNFalpha
.
Cell Stem Cell
2012
;
11
:
812
24
.
17.
Stubbs
JD
,
Lekutis
C
,
Singer
KL
,
Bui
A
,
Yuzuki
D
,
Srinivasan
U
, et al
cDNA cloning of a mouse mammary epithelial cell surface protein reveals the existence of epidermal growth factor-like domains linked to factor VIII-like sequences
.
Proc Natl Acad Sci U S A
1990
;
87
:
8417
21
.
18.
Taylor
MR
,
Couto
JR
,
Scallan
CD
,
Ceriani
RL
,
Peterson
JA
. 
Lactadherin (formerly BA46), a membrane-associated glycoprotein expressed in human milk and breast carcinomas, promotes Arg-Gly-Asp (RGD)-dependent cell adhesion
.
DNA Cell Biol
1997
;
16
:
861
9
.
19.
Hanayama
R
,
Tanaka
M
,
Miwa
K
,
Shinohara
A
,
Iwamatsu
A
,
Nagata
S
. 
Identification of a factor that links apoptotic cells to phagocytes
.
Nature
2002
;
417
:
182
7
.
20.
Shi
J
,
Heegaard
CW
,
Rasmussen
JT
,
Gilbert
GE
. 
Lactadherin binds selectively to membranes containing phosphatidyl-L-serine and increased curvature
.
Biochim Biophys Acta
2004
;
1667
:
82
90
21.
Asano
K
,
Miwa
M
,
Miwa
K
,
Hanayama
R
,
Nagase
H
,
Nagata
S
, et al
Masking of phosphatidylserine inhibits apoptotic cell engulfment and induces autoantibody production in mice
.
J Exp Med
2004
;
200
:
459
67
.
22.
Silvestre
JS
,
Théry
C
,
Hamard
G
,
Boddaert
J
,
Aguilar
B
,
Delcayre
A
, et al
Lactoadherin promotes VEGF-dependent neovascularization
.
Nat Med
2005
;
11
:
499
506
.
23.
Neutzner
M
,
Lopez
T
,
Feng
X
,
Bergmann-Leitner
ES
,
Leitner
WW
,
Udey
MC
. 
MFG-E8/lactadherin promotes tumor growth in an angiogenesis-dependent transgenic mouse model of multistage carcinogenesis
.
Cancer Res
2007
;
67
:
6777
85
.
24.
Uchiyama
A
,
Yamada
K
,
Ogino
S
,
Yokoyama
Y
,
Takeuchi
Y
,
Udey
MC
, et al
MFG-E8 regulates angiogenesis in cutaneous wound healing
.
Am J Pathol
2014
;
184
:
1981
90
.
25.
Oba
J
,
Moroi
Y
,
Nakahara
T
,
Abe
T
,
Hagihara
A
,
Furue
M
. 
Expression of milk fat globule epidermal growth factor-VIII may be an indicator of poor prognosis in malignant melanoma
.
Br J Dermatol
2011
;
165
:
506
12
.
26.
Jinushi
M
,
Nakazaki
Y
,
Carrasco
DR
,
Draganov
D
,
Souders
N
,
Johnson
M
, et al
Milk fat globule EGF-8 promotes melanoma progression through coordinated Akt and twist signaling in the tumor microenvironment
.
Cancer Res
2008
;
68
:
8889
98
.
27.
Jinushi
M
,
Chiba
S
,
Yoshiyama
H
,
Masutomi
K
,
Kinoshita
I
,
Dosaka-Akita
H
, et al
Tumor-associated macrophages regulate tumorigenicity and anticancer drug responses of cancer stem/initiating cells
.
Proc Natl Acad Sci U S A
2011
;
108
:
12425
30
.
28.
Jinushi
M
,
Sato
M
,
Kanamoto
A
,
Itoh
A
,
Nagai
S
,
Koyasu
S
, et al
Milk fat globule epidermal growth factor-8 blockade triggers tumor destruction through coordinated cell-autonomous and immune-mediated mechanisms
.
J Exp Med
2009
;
206
:
1317
26
.
29.
Motegi
S
,
Leitner
WW
,
Lu
M
,
Tada
Y
,
Sárdy
M
,
Wu
C
, et al
Pericyte-derived MFG-E8 regulates pathologic angiogenesis
.
Arterioscler Thromb Vasc Biol
2011
;
31
:
2024
34
.
30.
Motegi
S
,
Garfield
S
,
Feng
X
,
Sárdy
M
,
Udey
MC
. 
Potentiation of platelet-derived growth factor receptor-β signaling mediated by integrin-associated MFG-E8
.
Arterioscler Thromb Vasc Biol
2011
;
31
:
2653
64
.
31.
Uchiyama
A
,
Yamada
K
,
Perera
B
,
Ogino
S
,
Yokoyama
Y
,
Takeuchi
Y
, et al
Protective effect of MFG-E8 after cutaneous ischemia-reperfusion injury
.
J Invest Dermatol
2015
;
135
:
1157
65
.
32.
Weir
C
,
Morel-Kopp
MC
,
Gill
A
,
Tinworth
K
,
Ladd
L
,
Hunyor
SN
, et al
Mesenchymal stem cells: isolation, characterisation and in vivo fluorescent dye tracking
.
Heart Lung Circ
2008
;
17
:
395
403
.
33.
Hu
X
,
Yu
SP
,
Fraser
JL
,
Lu
Z
,
Ogle
ME
,
Wang
JA
, et al
Transplantation of hypoxia-preconditioned mesenchymal stem cells improves infarcted heart function via enhanced survival of implanted cells and angiogenesis
.
J Thorac Cardiovasc Surg
2008
;
135
:
799
808
.
34.
Theus
MH
,
Wei
L
,
Cui
L
,
Francis
K
,
Hu
X
,
Keogh
C
, et al
In vitro hypoxic preconditioning of embryonic stem cells as a strategy of promoting cell survival and functional benefits after transplantation into the ischemic rat brain
.
Exp Neurol
2008
;
210
:
656
70
.
35.
Khakoo
AY
,
Pati
S
,
Anderson
SA
,
Reid
W
,
Elshal
MF
,
Rovira
II
, et al
Human mesenchymal stem cells exert potent antitumorigenic effects in a model of Kaposi's sarcoma
.
J Exp Med
2006
;
203
:
1235
47
.
36.
Cho
DI
,
Kim
MR
,
Jeong
HY
,
Jeong
HC
,
Jeong
MH
,
Yoon
SH
, et al
Mesenchymal stem cells reciprocally regulate the M1/M2 balance in mouse bone marrow-derived macrophages
.
Exp Mol Med
2014
;
46
:
e70
.
37.
Geng
Y
,
Zhang
L
,
Fu
B
,
Zhang
J
,
Hong
Q
,
Hu
J
, et al
Mesenchymal stem cells ameliorate rhabdomyolysis-induced acute kidney injury via the activation of M2 macrophages
.
Stem Cell Res Ther
2014
;
5
:
80
.
38.
Gao
S
,
Mao
F
,
Zhang
B
,
Zhang
L
,
Zhang
X
,
Wang
M
, et al
Mouse bone marrow-derived mesenchymal stem cells induce macrophage M2 polarization through the nuclear factor-κB and signal transducer and activator of transcription 3 pathways
.
Exp Biol Med
2014
;
239
:
366
75
.
39.
Zanier
ER
,
Pischiutta
F
,
Riganti
L
,
Marchesi
F
,
Turola
E
,
Fumagalli
S
, et al
Bone marrow mesenchymal stromal cells drive protective M2 microglia polarization after brain trauma
.
Neurotherapeutics
2014
;
11
:
679
95
.
40.
Abe
T
,
Shin
J
,
Hosur
K
,
Udey
MC
,
Chavakis
T
,
Hajishengallis
G
. 
Regulation of osteoclast homeostasis and inflammatory bone loss by MFG-E8
.
J Immunol
2014
;
193
:
1383
91
.
41.
Soki
FN
,
Koh
AJ
,
Jones
JD
,
Kim
YW
,
Dai
J
,
Keller
ET
, et al
Polarization of prostate cancer-associated macrophages is induced by milk fat globule-EGF factor 8 (MFG-E8)-mediated efferocytosis
.
J Biol Chem
2014
;
289
:
24560
72
.
42.
Spinella
F
,
Rosanò
L
,
Di Castro
V
,
Decandia
S
,
Nicotra
MR
,
Natali
PG
, et al
Endothelin-1 and endothelin-3 promote invasive behavior via hypoxia-inducible factor-1alpha in human melanoma cells
.
Cancer Res
2007
;
67
:
1725
34
.
43.
Spinella
F
,
Caprara
V
,
Di Castro
V
,
Rosanò
L
,
Cianfrocca
R
,
Natali
PG
, et al
Endothelin-1 induces the transactivation of vascular endothelial growth factor receptor-3 and modulates cell migration and vasculogenic mimicry in melanoma cells
.
J Mol Med
2013
;
91
:
395
405
.
44.
Grimshaw
MJ
,
Wilson
JL
,
Balkwill
FR
. 
Endothelin-2 is a macrophage chemoattractant: implications for macrophage distribution in tumors
.
Eur J Immunol
2002
;
32
:
2393
400
.