Skip to main page content
U.S. flag
See More

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2011 Sep;9(9):e1001162.
doi: 10.1371/journal.pbio.1001162. Epub 2011 Sep 27.

Mesenchymal transition and dissemination of cancer cells is driven by myeloid-derived suppressor cells infiltrating the primary tumor

Benjamin Toh  1 , Xiaojie Wang, Jo Keeble, Wen Jing Sim, Karen Khoo, Wing-Cheong Wong, Masashi Kato, Armelle Prevost-Blondel, Jean-Paul Thiery, Jean-Pierre Abastado
Affiliations

Mesenchymal transition and dissemination of cancer cells is driven by myeloid-derived suppressor cells infiltrating the primary tumor

Benjamin Toh et al. PLoS Biol. 2011 Sep.

Abstract

In order to metastasize, cancer cells need to acquire a motile phenotype. Previously, development of this phenotype was thought to rely on the acquisition of selected, random mutations and thus would occur late in cancer progression. However, recent studies show that cancer cells disseminate early, implying the existence of a different, faster route to the metastatic motile phenotype. Using a spontaneous murine model of melanoma, we show that a subset of bone marrow-derived immune cells (myeloid-derived suppressor cells or MDSC) preferentially infiltrates the primary tumor and actively promotes cancer cell dissemination by inducing epithelial-mesenchymal transition (EMT). CXCL5 is the main chemokine attracting MDSC to the primary tumor. In vitro assay using purified MDSC showed that TGF-β, EGF, and HGF signaling pathways are all used by MDSC to induce EMT in cancer cells. These findings explain how cancer cells acquire a motile phenotype so early and provide a mechanistic explanation for the long recognized link between inflammation and cancer progression.

PubMed Disclaimer

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Figure 1
Figure 1. CXCR2 ligands attract PMN-MDSC to the primary tumor.
(A) Quantification of the subsets of immune cells infiltrating primary tumors and cutaneous metastases. Data were obtained by flow cytometry from 20 cutaneous and 13 eye tumors. Numbers show the frequencies of each subset among CD45+ cells. Gating strategy for immune cell subsets is illustrated in Figure S1D. (B) Preferential accumulation of PMN-MDSC (Gr1hi F4/80−) in primary tumors. Composition of the primary tumor and a cutaneous cheek tumor from the same mouse. Cells are gated on CD45+CD11c−CD11b+DAPI− tumor cells. (C) Frequency of PMN-MDSC in primary (n = 13) and cutaneous (n = 20) tumors. Bars represent mean ± SEM. ***p value <0.0001, two-tailed unpaired Student’s t test. (D) Repertoire of chemokines and cytokines differentially expressed in primary tumors and metastases. Fold changes represent the log2 ratio of gene expression measured by qRT-PCR in primary tumors (n = 11) and in the metastatic tumors (n = 11). p values were calculated by two-tailed paired Student’s t test. (E) Relative expression of CXCL1 (CX1), CXCL2 (CX2), and CXCL5 (CX5) in 11 primary tumors and 11 metastases measured by qRT-PCR. *p<0.05, ***p<0.001. (F–H) CXCL1 (F), CXCL2 (G), and CXCL5 (H) induce PMN-MDSC chemotaxis in vitro, but have no effect on monocytes or T, B, and NK lymphocytes. x-axis represents the amount of chemokine added in ng/ml. Percentage migration was calculated as (number of migrated cells) ×100/(total cells added per well). Data are from three independent experiments carried out in duplicates. Bars represent mean ± SEM *p<0.05, **p<0.01, ***p<0.001, two-tailed t test. (I) Inhibition of CXCR2 reduces PMN-MDSC migration in vitro. PMN-MDSC were treated 1 h before and during the migration assay with CXCR2 inhibitors—Inh1 (SB265610) and Inh2 (SB225002). Percentage migration was calculated as (number of migrated cells) ×100/(total cells added per well). Data are from two independent experiments carried out in duplicates. Bars represent mean ± SEM *p<0.05, **p<0.01, two-tailed t test. (J) CXCR2 is required for PMN-MDSC accumulation into the primary tumor. Equal numbers of bone marrow cells from Rosa mT/mG reporter mice expressing tdTomato (used as fluorescently tagged wild-type [wt] cells) and Il8rb-KO mice expressing GFP were adoptively transferred into tumor-bearing RETAAD mice. The graph shows the contribution of each genotype to donor-derived PMN-MDSC present in the primary tumor and spleen 18 h after transfer. Data are from four mice. Bars represent mean ± SEM *p<0.05, **p<0.01, two-tailed t test.
Figure 2
Figure 2. PMN-MDSC favor tumor cell proliferation in the primary tumor.
(A) Five-week-old mice were depleted of PMN-MDSC by bi-weekly injection of anti-Ly6G antibody (NIMP-R14) (depletion scheme B). At 20 wk of age, mice were euthanized and the size of the primary tumors was measured. Results are from a total of 22 tumors derived from four depleted mice (Dp) and seven littermates injected with a control antibody (Ct). (B–D) One-week-old mice were depleted of PMN-MDSC by bi-weekly injection of anti-Ly6G antibody (depletion scheme A). At 7 wk of age, mice were culled and eye tumors were analyzed by two-color immunohistochemistry using S100B- (brown) and Ki67- (blue) specific antibodies. (B) Comparison of the mitotic indices measured in 172 nodules taken from eight depleted mice (Dp) and 130 nodules from seven control littermates (Ct). Panels C and D show representative examples of tumors from control and depleted mice. Bars in panels A and B represent mean ± SEM. *p value <0.05, two-tailed unpaired Student’s t test.
Figure 3
Figure 3. PMN-MDSC favor tumor cell proliferation in vitro.
(A) Melan-ret cells (5×103 cells per well) were cultured with increasing concentrations of PMN-MDSC purified from tumor-bearing RETAAD mice (12-wk-old) for 48 h. Proliferation of tumor cells was assessed by [3H]-thymidine incorporation. PM, PMN-MDSC only; T:PM, ratio of tumor cells to PMN-MDSC; T, tumor cells only. (B) Irradiated PMN-MDSC but not macrophages induce Melan-ret tumor cell proliferation. Macrophages and PMN-MDSC were irradiated at 2,000 rads before co-culture with tumor cells for 48 h. iPM, Irradiated PMN-MDSC; iMF, Irradiated macrophages; T, tumor cells. (C) PMN-MDSC-induced proliferation does not require direct contact. PMN-MDSC and tumor cells were plated in the top and bottom wells of transwell inserts (pore size=0.4 µm; Millipore), respectively. PMN-MDSC cultured in the upper chamber were able to induce proliferation of Melan-ret tumors cells in the bottom chamber after 48 h. Data are from four independent experiments carried out in duplicates. Bars represent the average fold change relative to tumor cell only ± SEM. *p value <0.05, two-tailed paired Student’s t test.
Figure 4
Figure 4. PMN-MDSC favor multinodular growth of the primary tumor.
(A and B) Multinodular eye tumors from control (A) and PMN-MDSC depleted (B) mice stained with anti-S100B antibody (depletion scheme A). (C) Number of nodules per mouse. Results are from 11 depleted (depletion scheme A) mice (Dp) and 10 littermates injected with control antibody (Ct). *p value <0.05, two-tailed paired t test.
Figure 5
Figure 5. PMN-MDSC stimulate tumor cell dissemination and metastatic outgrowth.
(A) Dissemination of tumor cells to the draining mandibular lymph nodes (from 12 control and 20 depleted mice) and lungs (from 11 control and 11 depleted mice) was assessed by measuring ectopic expression of Dct and Mitf by qRT-PCR. The graph shows the relative gene expression normalized to GAPDH in mice injected with the control (Ct) or the anti-Ly6G antibody (Dp) (depletion scheme A). Bars represent mean ± SEM. *p<0.05, two-tailed t test. (B) Decreased number of cutaneous tumors in mice depleted from PMN-MDSC. Data are from 11 control and 11 PMN-MDSC-depleted mice (depletion scheme A). Bars represent mean ± SEM. *p<0.05, two-tailed Wilcoxon matched-pairs test. (C) Depletion of PMN-MDSC does not affect the size of cutaneous tumors. The depletion scheme A was used, but similar results were obtained with depletion scheme B (Figure S4). Bars represent mean ± SEM. ns, non-significant. (D) Delayed depletion of PMN-MDSC has no effect on the incidence of cutaneous metastases. Depletion scheme B was used. Data are from four control (Ct) and six PMN-MDSC-depleted (Dp) mice. Bars represent mean ± SEM. ns, non-significant. (E) Disseminated melanoma cells found in the lungs of 7-wk-old mice are mostly individual cells. Red, S100B; blue, CD45.
Figure 6
Figure 6. PMN-MDSC induce EMT in vitro and in vivo.
(A) PMN-MDSC (1×104 cells per well) were purified from 12-wk-old RETAAD mice and co-cultured with NBT-II cells for 24 h. HGF (5 ng/ml) was added as a positive control for induction of EMT. Green, Actin (upper panel) or E-Cadherin (lower panel); red, H2b (nuclear stain). Images are representative of three independent experiments. (B–D) Individual colonies of NBT-II cells were tracked for 6 h (during hours 12–18) out of the 22 h captured. Tracks of 59 cells from the control and 32 cells co-cultured with PMN-MDSC (1×104 cells per well) were analyzed. Panel B illustrates the individual tracks of all the cells in the colony. Panels C and D show an increase in distance traveled, speed, and displacement of NBT-II cells when co-cultured with PMN-MDSC. PM, PMN-MDSC; Ct, Control. Bars represent mean ± SEM. ***p<0.0001, two-tailed t test. (E) Primary RETAAD tumors (top panel) or cutaneous metastases (bottom panel) were dissociated and co-cultured for 24 h with (black line) or without (red line) PMN-MDSC purified from the same mouse. Data are gated on CD45− live cells. Filled histograms: isotype control. (F) Co-culture for 24 h with PMN-MDSC (as described in Figure 6E) induced a decrease in the number of tumor cells expressing E-Cadherin. Data are from five primary tumors and five cutaneous metastases in three independent experiments. Bars represent the mean decrease in the number of live CD45− E-Cadherin+ cells ± SEM *p<0.05, **p<0.01, two-tailed t test. (G) PMN-MDSC decrease CDH1 gene expression in vitro. Human melanoma cell line, 888mel, was co-cultured with PMN-MDSC purified from a tumor-bearing RETAAD mouse. Cells were washed and harvested after 24 h. Data are from three independent experiments. Bars represent mean ± SEM *p<0.05. (H–I) Staining for S100A4 in primary eye tumor nodules. Panel H is a representative staining of primary tumors for S100A4. Red, S100A4; blue, Heamatoxylin (Nuclear stain). Panel I shows a decrease in the percentage of S100A4+ cells in the tumors depleted from PMN-MDSC (depletion scheme A). Dp, PMN-MDSC depleted mice; Ct, Control mice. Bars represent mean ± SEM of 176 control and 87 PMN-MDSC depleted tumor nodules. ***p<0.0001, two-tailed t test. (J) Decreased percentage of vimentin+ cells in tumors depleted from PMN-MDSC (depletion scheme A). Dp, PMN-MDSC depleted mice; Ct, Control mice. Bars represent mean ± SEM of 14 pairs of primary tumors. *p<0.05, two-tailed paired t test. (K) The majority of S100A4+ cells express melanoma specific antigens. Eye tumors were co-stained for melanoma antigens (HMB45 and MelanA/MART-1) and S100A4. White arrows indicate S100A4+ melanoma cells exhibiting a fusiform morphology. Note the different morphologies of the double-stained mesenchymal-like cells and single-stained ones. Blue, DAPI (nucleus); green, HMB45/DT101/BC199 (melanoma); red, S100A4. (L) Decreased percentage of S100A4+ melanoma cells in tumors depleted from PMN-MDSC (depletion scheme A). Double positive cells stained for HMB45/MART-1 and S100A4 were quantified from three depleted (Dp) and three control (Ct) mice.
Figure 7
Figure 7. Inducers of EMT.
(A) Expression of Hgf, Egf, and Tgfb1 genes in RETAAD tumors analyzed by qRT-PCR. Hgf and Tgfb1 are preferentially expressed in PMN-MDSC, while Egf is preferentially expressed in tumor cells. Data are from four individual experiments using sorted fractions of PMN-MDSC and melanoma cells. Bars represent mean ± SD. (B) NBT-II cells (100 cells per well) were plated for 4 d to allow for colony growth and were pre-treated for 24 h with inhibitors before the addition of PMN-MDSC. After co-culture with PMN-MDSCs in the presence of inhibitors, cells were fixed and stained with anti-rat desmoplakin. Green, Desmoplakin; red, H2b (nuclear stain). PD153035 – EGFR inhibitor, JNJ38877605 – c-met (HGFR) inhibitor, and SB525334 – TGF-βR1 inhibitor. Images are representative of three independent experiments.
See this image and copyright information in PMC

References

    1. Jemal A, Siegel R, Ward E, Hao Y, Xu J, et al. Cancer statistics, 2008. CA Cancer J Clin. 2008;58:71–96. - PubMed
    1. Nguyen D. X, Bos P. D, Massague J. Metastasis: from dissemination to organ-specific colonization. Nat Rev Cancer. 2009;9:274–284. - PubMed
    1. Fearon E. R, Vogelstein B. A genetic model for colorectal tumorigenesis. Cell. 1990;61:759–767. - PubMed
    1. Bernards R, Weinberg R. A. A progression puzzle. Nature. 2002;418:823. - PubMed
    1. Eyles J, Puaux A. L, Wang X, Toh B, Prakash C, et al. Tumor cells disseminate early, but immunosurveillance limits metastatic outgrowth, in a mouse model of melanoma. J Clin Invest. 2010;120:2030–2039. - PMC - PubMed

Publication types

MeSH terms

Substances