Recruited monocytic myeloid-derived suppressor cells promote the arrest of tumor cells in the premetastatic niche through an IL-1β-mediated increase in E-selectin expression

International Journal of CancerVolume 140, Issue 6 p. 1370-1383 Tumor Immunology and MicroenvironmentFree Access Recruited monocytic myeloid-derived suppressor cells promote the arrest of tumor cells in the premetastatic niche through an IL-1β-mediated increase in E-selectin expression Huifang Shi, Huifang Shi The Key Laboratory of Molecular Epigenetics of MOE, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin, ChinaSearch for more papers by this authorJuechao Zhang, Juechao Zhang The Key Laboratory of Molecular Epigenetics of MOE, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin, ChinaSearch for more papers by this authorXiaoqing Han, Xiaoqing Han The Key Laboratory of Molecular Epigenetics of MOE, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin, ChinaSearch for more papers by this authorHuihan Li, Huihan Li The Key Laboratory of Molecular Epigenetics of MOE, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin, ChinaSearch for more papers by this authorMingshu Xie, Mingshu Xie The Key Laboratory of Molecular Epigenetics of MOE, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin, ChinaSearch for more papers by this authorYingying Sun, Yingying Sun The Key Laboratory of Molecular Epigenetics of MOE, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin, ChinaSearch for more papers by this authorWenguang Liu, Corresponding Author Wenguang Liu liuwg788@nenu.edu.cn The Key Laboratory of Molecular Epigenetics of MOE, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin, ChinaCorrespondence to: Xianlu Zeng, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin 130024, China, Tel.: 86-0431-85099317, E-mail: zengx779@nenu.edu.cn or Wenguang Liu, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin 130024, China, Tel.: 86-0431-85098837, E-mail: liuwg788@nenu.edu.cnSearch for more papers by this authorXueqing Ba, Xueqing Ba The Key Laboratory of Molecular Epigenetics of MOE, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin, ChinaSearch for more papers by this authorXianlu Zeng, Corresponding Author Xianlu Zeng zengx779@nenu.edu.cn The Key Laboratory of Molecular Epigenetics of MOE, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin, ChinaCorrespondence to: Xianlu Zeng, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin 130024, China, Tel.: 86-0431-85099317, E-mail: zengx779@nenu.edu.cn or Wenguang Liu, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin 130024, China, Tel.: 86-0431-85098837, E-mail: liuwg788@nenu.edu.cnSearch for more papers by this author Huifang Shi, Huifang Shi The Key Laboratory of Molecular Epigenetics of MOE, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin, ChinaSearch for more papers by this authorJuechao Zhang, Juechao Zhang The Key Laboratory of Molecular Epigenetics of MOE, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin, ChinaSearch for more papers by this authorXiaoqing Han, Xiaoqing Han The Key Laboratory of Molecular Epigenetics of MOE, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin, ChinaSearch for more papers by this authorHuihan Li, Huihan Li The Key Laboratory of Molecular Epigenetics of MOE, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin, ChinaSearch for more papers by this authorMingshu Xie, Mingshu Xie The Key Laboratory of Molecular Epigenetics of MOE, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin, ChinaSearch for more papers by this authorYingying Sun, Yingying Sun The Key Laboratory of Molecular Epigenetics of MOE, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin, ChinaSearch for more papers by this authorWenguang Liu, Corresponding Author Wenguang Liu liuwg788@nenu.edu.cn The Key Laboratory of Molecular Epigenetics of MOE, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin, ChinaCorrespondence to: Xianlu Zeng, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin 130024, China, Tel.: 86-0431-85099317, E-mail: zengx779@nenu.edu.cn or Wenguang Liu, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin 130024, China, Tel.: 86-0431-85098837, E-mail: liuwg788@nenu.edu.cnSearch for more papers by this authorXueqing Ba, Xueqing Ba The Key Laboratory of Molecular Epigenetics of MOE, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin, ChinaSearch for more papers by this authorXianlu Zeng, Corresponding Author Xianlu Zeng zengx779@nenu.edu.cn The Key Laboratory of Molecular Epigenetics of MOE, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin, ChinaCorrespondence to: Xianlu Zeng, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin 130024, China, Tel.: 86-0431-85099317, E-mail: zengx779@nenu.edu.cn or Wenguang Liu, Institute of Genetics and Cytology, School of Life Sciences, Northeast Normal University, Changchun, Jilin 130024, China, Tel.: 86-0431-85098837, E-mail: liuwg788@nenu.edu.cnSearch for more papers by this author First published: 25 November 2016 https://doi.org/10.1002/ijc.30538Citations: 55AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinkedInRedditWechat Abstract The tumor premetastatic niche initiated by primary tumors is constructed by multiple molecular factors and cellular components and provides permissive condition that allows circulating tumor cells to successfully metastasize. Myeloid-derived suppressor cells (MDSCs), a population of immature cells in pathological conditions, play a critical role in the formation of the premetastatic niche. However, few researches are focused on the function of monocytic MDSCs (mo-MDSCs), a subtype of MDSCs, in the construction of the niche. Here, we show that the number of mo-MDSCs is significantly increased in the premetastatic lungs of tumor-bearing mice, thus promoting tumor cell arrest and metastasis. Before the arrival of tumor cells, the lung-recruited mo-MDSCs produced IL-1β, thereby increasing E-selectin expression and promoting tumor cell arrest on endothelial cells. Depletion of mo-MDSCs in the premetastatic lungs decreased IL-1β production, resulting in reduced E-selectin expression. In addition, compared with alveolar macrophages and interstitial macrophages, mo-MDSCs were the major source of IL-1β expression in the premetastatic lungs. Cytokine array analyses and transwell experiments revealed that CCL12 recruits mo-MDSCs to premetastatic lungs. CCL12 knockdown in tumor-bearing mice significantly decreased mo-MDSC infiltration into the premetastatic lungs, leading to reduced E-selectin expression. Overall, the permissive conditions produced by the infiltrated mo-MDSCs correlated with increased tumor cell arrest and metastasis. These results reveal a novel role of mo-MDSCs in constructing the premetastatic niche. Thus, inhibition of mo-MDSCs infiltration may change the premetastatic niche to normal condition and attenuate tumor metastasis. Abstract What's new? During premetastasis, the tissue microenvironment of distant metastatic target organs is altered to permit habitation for circulating tumor cells. Premetastatic niche formation is influenced by granulocytic myeloid-derived suppressor cells (G-MDSCs), but according to this study, the other major MDSC subtype, monocytic MDSCs (mo-MDSCs), hitherto unknown factors in premetastatic niche construction, are also critical. In tumor-bearing mice, mo-MDSCs were greatly increased in premetastatic lung tissue, where they produced IL-1β, thereby promoting E-selectin expression and subsequent tumor cell adhesion to the vascular endothelium. Lung infiltration by mo-MDSCs was significantly decreased via knockdown of chemokine (C-C motif) ligand 12 (CCL12). Abbreviations CCL2 chemokine (C-C motif) ligand 2 CCL12 chemokine (C-C motif) ligand 12 CXCL10 chemokine (C-X-C motif) ligand 10 IL-1β interleukin-1 beta MDSCs myeloid-derived suppressor cells MMP matrix metalloproteinase TCM tumor-conditioned medium TNF-α tumor necrosis factor-alpha Tumor metastasis is a complicated process, including tumor cell dissemination from primary foci, infiltration into the vascular system, transit through the circulation, capture in capillary beds, and colonization and formation of metastatic nodules in distant target organs.1, 2 According to recent studies, tumor metastasis can be separated into three phases: premetastasis, micrometastasis, and macrometastasis.3 Among them, the construction of premetastatic microenvironment is the most essential step for successful tumor metastasis. Premetastasis is defined as the environment of the distant metastatic target organs, which is rebuilt before the arrival of the tumor cells from the primary foci. This environment is beneficial for tumor cell arrest, survival and proliferation.4 Construction of the premetastatic microenvironment requires multiple molecular and cellular elements, including primary tumor-secreted growth factors, bone marrow-derived hematopoietic progenitor cells, modified extracellular matrix, disrupted vasculature and the expression of novel signaling molecules and cytokines by stromal cells and endothelial cells.5, 7, 8 One well-acknowledged concept is that tumor premetastasis occurs in a stepwise fashion.4 In order to accomplish a distant engraftment, primary tumor cells secrete growth factors, including vascular endothelial growth factor A (VEGFA), transforming growth factor-β (TGF-β), placental growth factor (PlGF) and granulocyte colony stimulating factor (G-CSF).5, 9 In response to these factors, the expression of inflammatory chemokines and cytokines in lungs is upregulated, thus leading to a recruitment of bone marrow-derived hematopoietic progenitor cells to the premetastatic niche.10, 11 These newly recruited myeloid cells, together with stromal cells and endothelial cells in the tissue parenchyma, modulate the local microenvironment by secreting chemokines, inflammatory factors and matrix-degrading enzymes, thus providing a platform for tumor cell metastasis.12-14 During this process, myeloid-derived suppressor cells (MDSCs), a significant component of the recruited bone marrow-derived hematopoietic cells, play predominant roles in initiating the premetastatic niche and promoting tumor premetastasis. MDSCs are a heterogeneous population of immature cells which are expanded in pathological conditions, such as cancer, acute or chronic infections, and trauma, as well as during the bone marrow transplantation process.15 These cells consist of myeloid progenitor cells and immature myeloid cells, and characterized by the strong immunosuppressive properties. In mice, the phenotype of MDSCs is commonly defined as CD11b+ Gr1+. In human, MDSCs are identified as CD11b+ CD33+ HLA-DR−.15 In pathological conditions, the activation of MDSCs results in the upregulation of arginase 1, inducible nitric oxide synthase (iNOS) and reactive oxygen species (ROS), thus leading to the suppression of T-cell responses.16, 17 Beyond their immunosuppressive functions, studies have also reported that MDSCs are recruited in the premetastatic niche and metastatic niche and have significant roles in promoting tumor metastasis.18 In response to tumor-secreted factors, such as VEGFA, GM-CSF and G-CSF, MDSCs may further differentiate and participate in promoting tumor neovascularization.19, 20 MDSCs consist of two subtypes: granulocytic MDSCs (G-MDSCs) and monocytic MDSCs (mo-MDSCs). In mice, G-MDSCs have a CD11b + Ly6ClowLy6G+ phenotype, and mo-MDSCs are CD11b+ Ly6C+ Ly6G−.21 While in humans, the G-MDSCs have a CD11b+ CD33+ HLA-DR−CD15+ phenotype22 and the mo-MDSCs have a CD11b+ CD33+ HLA-DR-CD14+ phenotype.23 Numerous studies have reported that the recruited G-MDSCs secrete proteases such as MMP2 and MMP9, which facilitate tumor cell invasion by degrading the extracellular matrix.24, 25 The G-MDSCs have also been shown to regulate angiogenesis in tumors and to facilitate tumor metastasis by promoting vascular remodeling9 and decreasing immune surveillance26 in the premetastatic niche. Although emerging data have shown the significant roles of the recruited G-MDSCs in the construction of the premetastatic niche, mo-MDSCs have not yet been identified as having potential roles in this niche, and their functions and molecular mechanism after recruitment still remain to be elucidated. In this study, we report that mo-MDSCs are recruited into the lungs of mice with melanoma before the arrival of tumor cells, and the recruited mo-MDSCs promote E-selectin expression in the endothelial cells by secreting IL-1β, thus leading to increased distant metastasis. Our results reveal that the recruited mo-MDSCs have important roles in rebuilding the microenvironment of premetastatic lungs. Material and Methods Mice and cell lines C57BL/6 mice were purchased from the Animal Experimental Center of Jilin University (Changchun, China). Mice were housed in a pathogen-free environment, and all animal procedures performed were approved by the Animal Care Committee of the Northeast Normal University, China. Experimental animals were used at 8 to 10 weeks of age. B16F10 cells and HEK-293T cells were purchased from the American Type Culture Collection (ATCC). The MS1 cells were a gift from Dr. Zheng (Institute of Genetics and Cytology, Northeast Normal University). All cells were cultured in Dulbecco's Modified Eagle's Medium (DMEM) supplemented with 10% heat-inactivated fetal bovine serum (HyClone) and 1% penicillin/streptomycin (Sigma). Reagents and antibodies The reagents and antibodies are listed in the Supporting Information for Material and Methods. Single cell preparation and flow cytometry analysis Flow cytometry was performed by using a FACSCantoII (BD Biosciences), and data were analyzed with FACSDiva software (BD Biosciences) and Flow Jo 7.6.1 software. Detailed information was described in the Supporting Information for Material and Methods. Tumor cell arrest and metastasis assay B16F10 cells were first stained with PKH26, according to the manufacturer's instructions, and then 2 × 105 PKH26-stained B16F10 cells were i.v. injected into the untreated normal mice, tumor-bearing mice and mo-MDSCs adoptive transferred tumor-bearing mice. The lungs of the treated mice were collected and perfused with PBS to remove the circulating tumor cells in the blood vessels 3 h after the tumor cells were injected. Frozen sections were generated, and the PKH26-B16F10 cells were detected under a fluorescence microscope (Nikon Eclipse 80i). To detect metastasis, 2 × 105 B16F10 cells were i.v. injected, and the mice were sacrificed to analyze the number of metastatic foci 2 weeks after the tumor cells were injected. Cell isolation mo-MDSCs and G-MDSCs were sorted with a Myeloid-Derived Suppressor Cell Isolation Kit using AutoMACS sorter (Miltenyi Biotec) according to the manufacturer's instructions. Detailed information was described in the Supporting Information for Material and Methods. MDSCs depletion and mo-MDSCs replenishment in vivo MDSC depletion was carried out as described by Vincent et al.27 with minor modification. The MDSCs depletion assay in the 3-week tumor-bearing mice was performed in two steps. First, 50 μg/mouse of an anti-Gr1 antibody (RB6-8C5), which recognizes and depletes the cells with both Ly6G+ and Ly6C+ antigens, was i.v. injected into the mice, and most of the G-MDSCs were depleted, as assessed by flow cytometry 24 h later. Two days later, another 300 μg/mouse of the anti-Gr1 antibody (RB6-8C5) was i.v. injected into the mice to deplete both G-MDSCs and mo-MDSCs. To simulate mo-MDSCs depletion, 107 isolated G-MDSCs were replenished by i.v. injection after G-MDSCs and mo-MDSCs depletion. The lungs of the treated mice were collected 24 h postinjection. To replenish the mo-MDSCs, 2 × 106 isolated mo-MDSCs from the blood of tumor-bearing mice were i.v. injected into tumor-bearing mice. RT-PCR and real-time PCR analysis Experiments were carried out as described in the Supporting Information for Material and Methods. Fluorescence IHC staining and hematoxylin and eosin (H&E) staining Experiments were carried out as described in the Supporting Information for Material and Methods. Construction of the lentiviral expression plasmid and transfection A detailed description of procedures is presented in the Supporting Information for Material and Methods. Co-culture system MS1 cells were plated in 6 cm plates at a density of 106 cells/plate on the first day. On the second day, the mo-MDSCs and G-MDSCs were isolated from the blood of tumor-bearing mice, and the alveolar macrophages were harvested from the alveolar lavage fluid. Then the collected cells were added to the plates in the presence or absence of TCM (tumor-conditioned medium). Two hours later, the supernatants were discarded, and the MS1 cells were collected for western blot analysis. In a separate experiment, the co-culture system was administrated with IL-1Ra at a concentration of 25 μg/mL. Static adhesion assay Experiments were carried out as described in the Supporting Information for Material and Methods. Cytokine array for lung tissue For the cytokine array, the lung tissues were lysed with lysis buffer and processed according to the manufacturer's instructions (R&D Systems). In vitro migration assay The migration of mo-MDSCs was evaluated using a transwell chamber with 5 μm pores (Corning). Blood was collected, and the erythrocytes were eliminated using hypotonic lysis buffer. The remaining cells were separated into two layers using Ficoll (Sigma-Aldrich). The cells in the middle layer (5 × 104 cells) were seeded in the upper wells, and chemokines, such as CCL12, CCL2, CXCL10 or IL-1β, were placed in the lower wells. The plates were incubated at 37°C for 3 h with 5% CO2. The cells that had migrated to the lower wells were collected and stained with antibodies to examine the absolute number of transmigrated mo-MDSCs using flow cytometry. Statistical analysis The data were analyzed using Student's t-test and presented as the means ± SEM. Significance was determined as p < 0.05. Results Mo-MDSCs are recruited to the premetastatic niche in tumor-bearing mice It has been reported that the number of circulating MDSCs is increased in the tumor-bearing mice,28 and the G-MDSC subset is recruited into the premetastatic niche.26 To examine whether the mo-MDSCs is recruited in the premetastatic niche, a tumor premetastatic model was established by subcutaneously inoculating 3 × 106 B16F10-GFP cells into the flanks of C57BL/6 mice. The B16F10 cells have been reported to specifically metastasize to the lungs.29 Then, we explored the time course over which the metastatic B16F10 cells arrived at the lungs. No band was found until the 5th week from our RT-PCR results, indicating that the tumor cells did not arrive at the lungs until 5 weeks after subcutaneous inoculation (Fig. 1a a). This result was confirmed by the histological staining, which showed that the B16F10-GFP cells began to metastasize to the lungs in the 5th week (Fig. 1a b). Then, we examined the expression of cytokines in lungs, which have been reported to be hallmarks of premetastatic niche formation.4, 30 As shown in Figure 1b; a, a significant increase in S100A8, S100A9 and IL-1β was observed in the lungs from the 2nd week to the 4th week (Fig. 1b; a). IFN-γ expression in CD8 + T cells in the lungs, which was the indicative of immune response, was significantly decreased from the 2nd week (Fig. 1b; b, Supporting Information 1). Both of these results indicate that the inflammatory and immunosuppressive premetastatic niche begin to form from the 2nd week. Therefore, the period of premetastatic niche formation is defined as from the 2nd week to the 4th week in the present study. Figure 1Open in figure viewerPowerPoint mo-MDSCs are recruited to the lungs of tumor-bearing mice prior to tumor cells. (a) GFP-positive B16F10 cells were analyzed in the total lungs after subcutaneous inoculation with B16F10-GFP cells at different time points by RT-PCR; B16F10 cells, B16F10-GFP cells and a lung lobe containing B16F10-GFP cells (0 cell, 100 cells, 1,000 cells) as a control (a) and metastatic B16F10-GFP cells were detected in lung frozen sections (scale bar, 10 µm) (b); n ≥ 5. (b) The expression levels of the S100A8, S100A9 and IL-1β genes in the whole lungs of the untreated normal mice and tumor-bearing mice were analyzed by real-time PCR at different time points (a) and the percentage of IFN-γ-positive CD8+T cells in the lungs of the normal mice and tumor-bearing mice was quantified (b); n = 3. (c) H&E staining of lung frozen sections from the untreated normal mice and tumor-bearing mice was conducted (upper) and the total number of cells per field were quantified (lower). Shown is one of the three experiments performed. (d) Phenotype of mo-MDSCs (CD11b+Ly6C+) and G-MDSCs (CD11b+Ly6G+). IgG controls were included in all experiments. (e) Percentages of mo-MDSCs (upper) and G-MDSCs (lower) out of CD45-positive cell were quantified in the premetastatic lungs at different time points after subcutaneous inoculation with B16F10 cells; n ≥ 3. At least three independent experiments were performed. *p < 0.05; **p < 0.01, ***p < 0.0001. [Color figure can be viewed at wileyonlinelibrary.com] The premetastatic niche is a reconstructed environment that is ready for tumor cells to metastasize. Therefore, we next examined the structural alterations in the premetastatic lungs. A clear increase in the cellular density and a reduction in the alveolar space were observed in the lungs of the tumor-bearing mice compared with the untreated normal mice, as observed by H&E staining and counting the exact number of cells per field (Fig. 1c). These changes indicated that during the formation of premetastatic niche, a large number of cells were recruited to the lungs. To investigate whether the increased number of cells includes mo-MDSCs, the proportion of mo-MDSCs out of CD45-positive cells was examined in the lungs of tumor-bearing mice at different time points. A significant increase in the percentage of mo-MDSCs was observed compared with normal mice, and the increase started at the 2nd week and reached the maximum in the 3rd week. The proportion of G-MDSCs was also examined, and the results showed that the percentage continued to increase from the 2nd week (Figs. 1d and 1e). Although the proportion of mo-MDSCs decreased in the 4th week, the absolute number of the cells did not decrease (Supporting Information 2). Together, these data suggest that mo-MDSCs and G-MDSCs are recruited to the lungs before the arrival of tumor cells. Furthermore, there was no detectable increase in the numbers of mo-MDSCs and G-MDSCs in the hearts, kidneys and livers of the tumor-bearing mice compared with the normal mice (Supporting Information 3), thus further confirming that the lung is a tropism of B16F10 cells that have a priority to metastasize.26 mo-MDSC-promoted tumor cell arrest is associated with an increase of E-selectin expression Because an increased number of mo-MDSCs were observed in the premetastatic lungs, we hypothesized that the recruited mo-MDSCs may be associated with premetastatic niche construction and tumor cell metastasis. To precisely monitor the arrest and outgrowth of tumor cells following the entry of tumor cells into the bloodstream, we used an experimental lung metastasis model by intravenously injecting B16F10 cells to mice. Results showed that a significant increased number of metastatic foci were detected in the tumor-bearing mice compared with the normal mice 2 weeks after injection (Fig. 2a). To test if the recruited mo-MDSCs are responsible for the increased tumor metastasis, we isolated the mo-MDSCs from blood of the tumor-bearing mice, and examined their purity by FSC (≥90%). The isolated mo-MDSCs were intravenously injected into the subcutaneous tumor-bearing mice, and an increased number of mo-MDSCs was detected in the premetastatic lungs 24 h later (Supporting Information 4). In parallel experiments, 24 h after mo-MDSCs transplantation, the mice were intravenously injected with B16F10 cells. The number of metastatic foci was evaluated in lungs of the treated mice. A significant increase of metastatic foci was observed in the mo-MDSCs-transferred tumor-bearing mice compared with the untreated tumor-bearing mice (Fig. 2a). Because successful distant metastasis requires the arrest of surviving tumor cells in blood vessels,31 we also counted the number of arrested tumor cells in the lungs 3 h after the mice were intravenously injected with PKH26-stained B16F10 cells. A dramatic increase of tumor cell arrest was observed in the lungs of the mo-MDSCs-transferred tumor-bearing mice compared with the tumor-bearing mice and untreated normal mice (Fig. 2b). Together, these results indicate that the recruited mo-MDSCs in the premetastatic niche promote tumor cell arrest and metastasis. Figure 2Open in figure viewerPowerPoint mo-MDSCs promote E-selectin-dependent tumor cell arrest and metastasis. (a) The number of metastatic foci was detected in the lungs from normal, 2-week tumor-bearing and mo-MDSCs adoptive transferred-2-week tumor-bearing mice 2 weeks after the i.v. injection of B16F10 cells. The transferred mo-MDSCs were isolated from the blood of tumor-bearing mice; n ≥ 5. (b) The number of arrested B16F10 cells in the lungs of normal, 2-week tumor-bearing and mo-MDSCs adoptive transferred-2-week tumor-bearing mice 3 h was quantified after the i.v. injection of B16F10 cells. The B16F10 cells were prestained with PKH26 (scale bar, 400 µm); n ≥ 6. (c) E-selectin expression in the lungs of normal or tumor-bearing mice was obtained from analyzing the extracted protein from the harvested whole lungs by western blotting (a) and from immunohistochemistry staining of the lung frozen sections (scale bar, 100 µm) (b) 2 to 4 weeks after the subcutaneous inoculation; n ≥ 3. (d) E-selectin expression in the lungs was assessed by western blotting (a) and the number of arrested B16F10 cells was counted (b) in the lungs of shCtrl normal mice, shCtrl 3-week tumor-bearing mice and shE-selectin 3-week tumor-bearing mice (scale bar, 400 µm); n ≥ 3. The data are representative of more than three independent experiments. *p < 0.05; **p < 0.01, ***p < 0.0001. [Color figure can be viewed at wileyonlinelibrary.com] The arrest of tumor cells requires an interaction between adhesion molecules on the tumor cells and receptors on the vascular endothelium.32 The known adhesion receptors on endothelial cells that play functions in the adhesion of tumor cells are E-selectin and P-selectin.33, 34 Therefore, the E-selectin and P-selectin expression levels in the lungs of the subcutaneous tumor-bearing mice and untreated normal mice were determined. The protein was extracted from the whole lungs, and results showed that E-selectin expression was dramatically increased in the premetastatic niche from the 2nd week and reached a maximum in the 3rd week (Fig. 2c; a). This result was further confirmed by immunohistochemistry (Fig. 2c; b). P-selectin expression was also examined, but no significant change was detected (data not shown). Furthermore, the expressions of ICAM-1, VCAM-1 and PECAM-1, which participate in the later stage of extravasation cascade of tumor cells, were examined, and data showed that the expression of PECAM-1 was increased during the premetastatic phase, while ICAM-1 and VCAM-1 did not change (Supporting Information 5). To test the function of E-selectin in the premetastatic lungs, lentivirus-based RNA interference against E-selectin was intrapleurally injected into 2-week tumor-bearing mice three times in 1 week according to a previous report.35 Then mice were intravenously injected with PKH26-stained B16F10 cells. Data showed that when E-selectin expression was effectively knocked down in the lungs of tumor-bearing mice (Fig. 2d; a), the number of arrested tumor cells was significantly decreased compared with shCtrl tumor-bearing mice (Fig. 2d; b). Furthermore, no significant change in E-selectin expression was detected in the hearts, kidneys and livers of the tumor-bearing mice compared with the normal mice (Supporting Information 6).