Macrophage infiltration predicts a poor prognosis for human ewing sarcoma

Abstract

Ewing sarcoma-primitive neuroectodermal tumor (EWS) is associated with the most unfavorable prognosis of all primary musculoskeletal tumors. The objective of the present study was to investigate whether tumor-associated macrophages (TAMs) affect the development of EWS. TAMs were isolated from mouse xenografts using CD11b magnetic beads and examined for their cytokine expression and osteoclastic differentiation. To evaluate the role of TAMs in xenograft formation, liposome-encapsulated clodronate was used to deplete TAMs in mice. Macrophage infiltration and tumor microvascular density were histologically evaluated in 41 patients with EWS, and association with prognosis was examined using Kaplan-Meier survival analysis. In mouse EWS xenografts, TAMs expressed higher concentrations of cytokines including interleukin-6, keratinocyte-derived chemokine, and monocyte chemotactic protein-1. TAMs were more capable than normal monocytes of differentiating into tartrate-resistant acid phosphatase-positive giant cells. Depleting macrophages using liposome-encapsulated clodronate significantly inhibited development of EWS xenografts. In human EWS samples, higher levels of CD68-positive macrophages were associated with poorer overall survival. In addition, enhanced vascularity, increase in the amount of C-reactive protein, and higher white blood cell counts were also associated with poor prognosis and macrophage infiltration. TAMs seem to enhance the progression of EWS by stimulating both angiogenesis and osteoclastogenesis. Further investigation of the behavior of TAMs may lead to development of biologically targeted therapies for EWS.

Copyright © 2011. Published by Elsevier Inc.

Figures

Figure 1

Identification and isolation of TAMs from mouse EWS xenograft tumors. A: Immunohistochemical staining for CD99 and F4/80 in mouse EWS xenografts. Nude mice were subcutaneously inoculated with RD-ES or TC-71 EWS cells. EWS tumors were characterized by using H&E staining and CD99 immunostaining. Representative images of EWS xenografts infiltrated by F4/80-positive macrophages are shown (arrow). Scale bar= 20 μm. B: Surface marker expression on dissociated xenograft cells. After mincing, xenograft cells were dissociated using collagenase and DNase and were subjected to flow cytometric analysis. C: Surface marker expression on isolated CD11b+ cells. After isolating cells using anti-CD11b beads, cells from EWS tumor xenografts (TAMs) or liver (control macrophages) were subjected to flow cytometric analysis.

Figure 2

Cytokine expression by TAMs in mouse EWS xenograft tumors. A: Chemokine expression by isolated CD11b+ cells. TAMs or control macrophages were incubated in serum-free DMEM for 72 hours, and the conditioned medium was examined using a Luminex multiplex assay system. Results are given as mean ± SD. *P < 0.05. B: Effect of conditioned medium on monocytic migration was examined using the Transwell system. Monocytic RAW264.7 cells were added to the upper well, and conditioned medium from TAMs or control macrophages was placed in the lower well. After 4 hours of incubation, the cells that had migrated to the bottom surface were stained and counted. Results are given as mean ± SD. *P < 0.05. **P < 0.01.

Figure 3

Osteoclastic differentiation of TAMs in EWS. A: Induction of osteoclastic differentiation. TAMs or control macrophages were incubated with sRANKL and M-CSF for 4 days. Osteoclastic differentiation was visualized using TRAP staining (left). TRAP-positive multinucleated giant cells were counted (right). Results are given as mean ± SD. *P < 0.05. B: RT-PCR was performed to detect the expression of RANKL and M-CSF mRNA in six EWS cell lines. C: Quantitative RT-PCR was performed to detect osteoclastic differentiation in CD11b+ cells. All expression levels were normalized on the basis of expression of GAPDH. Data show the relative expression in TAMs (gray bars) compared with control macrophages (white bars). Results are given as mean ± SD. *P < 0.05. D: TRAP staining of EWS xenografts that developed from RD-ES or TC-71 cells. Sections were counterstained with diluted methyl green solution. Scale bars: 20 μm (A); 50 μm (D).

Figure 4

EWS cell lines stimulate monocyte migration via VEGF signaling. A: Migration of monocytic cells was examined using the Transwell system. The lower wells were filled with serum-free medium, RD-ES cells, or TC-71 cells, and RAW264.7 cell migration to the bottom surface of the Transwell was assessed. Results are given as mean ± SD. **P < 0.01. B: The Luminex multiplex assay system was used to screen for chemotactic factors produced by EWS cells. C: Inhibitory effects of VEGFR-TKI on migration of RAW264.7 cells. RAW264.7 cells were co-cultured with RD-ES cells, and their migration to the bottom surface of the Transwell in the presence of VEGFR-TKI was assessed. Results are given as mean ± SD. **P < 0.01. D: Quantification of VEGF secretion by EWS cells. RD-ES or TC-71 cells were stimulated with conditioned medium (CM) from TAMs for 48 hours, and the VEGF concentration in conditioned medium from EWS cells was examined using a human VEGF enzyme-linked immunosorbent assay kit. Serum-free DMEM was used as negative control. Results are given as mean ± SD. *P < 0.05.

Figure 5

Effects of macrophage depletion in an EWS xenograft model. A: Left, Cl2MDP-Lip (white squares) or PBS-Lip (black circles) was administered i.v. in nude mice 1 day before inoculation with RD-ES cells. Mice received 200 μL liposomes through the tail vein every 3 days. Length and width of the tumors were measured for 3 weeks after inoculation. Right, Dot plot for tumor volumes at 20 days after inoculation. Tumor volumes in the group treated with Cl2MDP-Lip were significantly lower than those in the PBS-Lip group. Five mice were used in each group. Results are given as mean (dotted lines) ± SD (straight lines). *P < 0.05. B: Tumors were excised and weighed at 3 weeks after inoculation. C: Immunohistochemical staining of macrophages and the tumor vasculature in EWS xenografts. Infiltrating macrophages (arrow) were visualized using anti-F4/80 antibodies (left). Tumor vasculature (asterisk) was visualized using anti-CD31 antibodies (right). Scale bars = 20 μm. D: Mean number of F4/80-positive macrophages and CD31-positive vessels in six random field profiles were used for subsequent statistical analyses (Mann-Whitney U-test). Results are given as mean ± SD. **P < 0.01.

Figure 6

Immunohistochemical staining of human EWS sections. Representative staining of macrophages, tumor vasculature, and MIB1 in EWS samples. Paraffin sections were immunohistochemically stained using anti-CD68, anti-CD31, and anti-MIB1 antibodies, and were visualized using the diaminobenzidene substrate system. Counterstaining was then performed using diluted hematoxylin. Compared with cases with lower macrophage infiltration [≤30 CD68 cells/HPF; case 4 (continuously disease-free) and case 6 (no evidence of disease)] (A), prominent tumor microvasculature and MIB1 expression were evident in cases with higher macrophage infiltration (>30 CD68 cells/HPF; cases 36 and 37 died of disease) (B). Scale bars: 20 μm (A and B).

Figure 7

Association between macrophage infiltration and poor prognosis in EWS. A–F: Kaplan-Meier survival curves for all patients based on CD68-positive macrophage infiltration (low, ≤30 CD68 cells/HPF; high, >30 CD68 cells/HPF) (A), microvascular density (low, ≤10/HPF; high, >10/HPF) (B), serum CRP concentration (low, ≤ 0.2 mg/dL; high, >0.2 mg/dL) (C), WBC counts (low, ≤6800 cells/μL; high, >6800 cells/μL) (D), MIB1 expression (low MIB1 index, <40; high MIB1 index, ≥40) (E), and tumor size (small,<8 cm; large, ≥8) (F). Log-rank tests were performed to determine statistical significance, with P < 0.05 defined as significant. *P < 0.05. **P < 0.01.

Figure 8

Model for TAM-mediated modulation of EWS microenvironment. TAMs have important roles as modulators of inflammation, angiogenesis, and osteoclastogenesis during EWS development. TAMs accumulation is mediated by VEGF secretion from EWS, and is further enhanced by various cytokines and chemokines released from the TAMs themselves, resulting in an inflammatory reaction in EWS. TAMs stimulate tumor angiogenesis by enhancing VEGF production from tumor cells, resulting in a poorer prognosis. Enhanced osteoclastogenesis induced by TAMs enhances bone tumor progression and may affect the prognosis in EWS.