Chemotherapy may facilitate colorectal cancer metastasis through vessel formation by upregulating G-CSF

Chemotherapy may facilitate colorectal cancer metastasis through vessel formation by upregulating G-CSF | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Chemotherapy may facilitate colorectal cancer metastasis through vessel formation by upregulating G-CSF Yueping zhan, Kai Fang, Yuqian wang, Mengting Liang, Ruiying Wu, and 11 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-1464660/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background Chemotherapy remains the mainstay of treatment for various cancers. However recent studies suggested that it may also induce local or systemic changes that promote the dissemination and proliferation of cancer cells leading to successful metastasis. Understanding this process is therefore instrumental to help us identify patient population that may or may not be benefit from chemotherapy, and develop innovative approaches to prevent metastasis. Methods In this study, we investigated the mechanism of chemotherapy causing the metastasis of colorectal cancer, and the roles of G-CSF induced by chemotherapy in the vessel formation. Eight-week-old C57/BL6 mice were used as animal models in this study to investigate the lung metastasis by the chemotherapy. Conditional endothelial cell STAT3 −/− (Signal Transducer and Activator of Transcription, STAT) knockout mice (STAT3 flox/flox ; Tek-Cre mice), were used to investigate the function of STAT3 on lung metastasis under the chemotherapy. Bone marrow and plasma was selected to detect the mobilization of endothelial progenitor cells (EPCs) and MDSCs. To trace BM-derived cells, we prepared chimeric mouse transplanted with BM cells from green fluorescent protein (GFP) transgenic mouse. In vitro, experiments were performed in human umbilical vein endothelial cells (HUVEC) to analyse the function of G-CSF on angiogenesis. Detected the G-CSF content in the plasma of the patients who received an adjuvant chemotherapy with the XELOX (oxaliplatin plus capecitabine) regimen. Results Our study showed that oxaliplatin chemotherapy could increase the expression of G-CSF to promote lung metastasis. First, G-CSF/STAT3 signaling facilitated lung metastasis by enhancing vascular adhesion rather than diminishing the blood vessel density. G-CSF also promoted Endothelial Progenitor Cells (EPCs) mobilization that devoted to vasculogenesis, a critical step for vessel density in metastatic sites; moreover, chemotherapy augmented the mobilization of MDSCs from the bone marrow by G-CSF. In consistent with these, anti-G-CSF suppressed the formation of a functional vasculature and induced tumoral immunosuppression, resulting in an anti-metastasis effect during chemotherapy. Furthermore, at human level, we observed high level of G-CSF, either at baseline or after receiving adjuvant XELOX chemotherapy, correlated with poor overall and recurrence-free survival. Conclusions These results demonstrate that certain chemotherapy could paradoxically result in worse outcome due to increased expression of G-CSF. Our findings provide mechanistic insight into cautious use of G-CSF and potential utility of anti-G-CSF in personalized cancer therapy. Vessel formation metastasis G-CSF chemotherapy Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Background Cancer metastasis is a multi-step complicated process involving the spreading, localization and adaptation of cancer cells to the new environment in remote organs [ 1 , 2 ]. Although chemotherapy is one of the main effective treatments, recent studies have shown that it may also induce tumoral or host changes, such as EMT [ 3 ], host-initiated proinflammatory responses [ 4 ], and recruitment of BMDCs/MSCs to secondary sites [ 5 ], which could paradoxically facilitate the proliferation and dissemination of cancer cells, hence metastasis. Deciphering such observation is therefore of great clinical value. The formation of new blood vessels is essential for the development and progression of cancer as it provides nutrition and oxygen and removes metabolic waste products [ 6 ]. Vasculogenesis, one of the important mechanisms of neovascularization, is initiated by the recruitment of EPCs [ 7 ], which belong to a subtype of stem cell with high proliferative and differentiation potential and are mainly derived from bone marrow (BM) and peripheral blood (PB) progenitor cells [ 8 ]. Circulating EPCs can be recruited to avascular tumor sites to promote vasculogenesis via de novo blood vessel formation and production of proangiogenic cytokines [ 9 ]. Shaked et al. found that certain chemotherapy drugs could induce acute EPC mobilization from bone marrow [ 10 ]. Other evidence indicates that circulating EPCs and cytokines negative correlate to progression-free and overall survival in patients after chemotherapy [ 11 ]. In addition, pediatric patients with solid tumors exhibited elevated levels of circulating endothelial cells and EPCs after chemotherapy [ 12 ]. Furthermore, multiple cytokines and enzymes, such as G-CSF [ 13 ], GM-CSF [ 14 ] and MMP9 [ 15 ], may directly or indirectly promote the mobilization of EPCs. Granulocyte-colony stimulating factor (G-CSF) is a glycoprotein mainly produced by fibroblasts, endothelial cells, bone marrow stromal cells, monocytes and macrophages [ 16 ]. G-CSF is well known to promote the survival, proliferation, and differentiation of neutrophil precursors by interacting with the G-CSF receptor (G-CSFR) [ 17 ], which activates downstream JAK/STAT3 pathway [ 18 ]. Amplified JAK/STAT3 signaling could lead to increased adhesivity of the endothelial surface [ 19 ]. Because of its capacity of promoting the mobilization, proliferation and differentiation of hematopoietic progenitor cells, G-CSF is widely used in clinical practice to treat chemotherapy induced neutropenia [ 20 ]. However, recent preclinical studies suggested that exogenous G-CSF administration as well as other amplified neutrophil-associated pathological conditions may promote the metastasis of solid tumors [ 21 – 23 ], therefore raised clinical concern that needs scientific explanation. In this study, we investigated the mechanism by which chemotherapy could promote colon cancer lung metastasis via elevated G-CSF and its role in tumor vasculature during chemotherapy. We found that G-CSF/STAT3 signaling enhanced vascular adhesion properties and EPC mobilization devoted to vessel density in metastasis sites. In addition, chemotherapy improved the mobilization of MDSCs from bone marrow. Anti-G-CSF given at the right time may suppress the formation of a functional vasculature and induce anti-tumor immune function, and prevent metastasis during chemotherapy. Methods Cell culture Human umbilical vein endothelial cells (HUVECs, #8000, ScienCell, USA) and mouse EPCs (CP-M140, Procell, China) were cultured in endothelial cell medium containing 10% FBS and 1% penicillin/streptomycin. HUVECs within the first three passages were used for experiments. CT26-Luc, MC38-Luc and Lewis-Luc cells were cultured in RPMI-1640 containing 10% FBS and 1% penicillin/streptomycin. B16F1 cells were cultured in DMEM containing 10% FBS and 1% penicillin/streptomycin. All cells were cultured at 37°C in a humidified incubator with 5% CO 2 . Tumors and Animal Models Eight- to twelve-week-old C57BL/6 or BALB/c mice (Songlian Experimental Animal Institute, Songjiang District, Shanghai) were treated with chemotherapy drugs. The mice were fed with a standard chow diet ad libitum and housed in a clean grade room at 21 ± 1 °C and 60 ± 5% humidity under a 12 h light/12 h dark cycle with free access to water and food in a specific pathogen-free environment. MC38-Luc cells (2×10 6 ) were implanted into immunocompetent C57BL/6 mice, C57BL/6 mice which previously irradiated and then transplanted with green fluorescent protein + (GFP + )-bone marrow cells [24], or STAT3 flox/flox and STAT3 flox/flox Tek-cre C57BL/6 mice (Shanghai Model Organisms, Chania) by tail vein injection. CT26-Luc cells and Lewis cells (2×10 6 ) were tail vein injected into BALB/c mice. B16F1 melanoma cells (2×10 6 ) were tail vein injected into C57BL/6 mice. In all in vivo studies mice were randomly grouped (n=4–6/group). Tumor lung metastasis progression was monitored using an Xenogen IVIS system (PerkinElmer IVIS Lumina Ⅲ-Living imaging). Mice were anesthetized using 2.5% isoflurane and administered D-luciferin (150 mg/kg; Yeasen) via intraperitoneal (i.p.) injection 10 min before the imaging. All protocols were approved by the Institutional Animal Care and Use Committee of Putuo Hospital, Shanghai University of Traditional Chinese Medicine, P. R. China. All animal studies were conducted in accordance with the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals. Flow cytometry analysis of EPCs and MDSCs BM cells were stained with BV421-conjugated anti-CD34 (562208, BD) and anti-CD133-PE (130–117–784, MACS), anti-VEGFR2-APC antibodies (560070, BD) and anti-CD133-PE to detect EPCs, anti-CD11b-FITC (557396, BD) and anti-Ly-6G-BV421 (562737, BD) to detect MDSCs. Blood samples were collected using EDTA tubes. Red blood cells were lysed, and EPCs were studied. Labeled cells were detected using a flow cytometer (C6, BD) and analyzed by CellQuest and FlowJo software. ELISA Levels of mouse or human G-CSF (H-GCSF, EK0360, Boster; M-GCSF, EK0361, Boster), VEGF (H-VEGF, EK0539, Boster; M-VEGF, EK0541, Boster) and SDF-1α (H-SDF 1A, SEKH-0310, Solarbio) were assessed by ELISA (Biolegend) according to the manufacturers’ instructions. Cytokine antibody array Serum samples were collected and frozen at -80°C for subsequent cytokine analysis using the R&D Mouse Cytokine Array Panel A (ARY006) according to the manufacturer’s instructions. Tube formation assay. HUVECs were treated with G-CSF for 1 hour before tube formation. HUVECs (3*10 4 ) were seeded into a 96-well plate coated with Matrigel (BD, #356234, USA). Tube formation was observed at 4 h. Three random fields were acquired by a microscope (Leica, DMI3000B, Germany). The number of Nodes and Branches points were quantified by ImageJ (Nodes/Branches: defined as individual junctions/branching points). Cell migration assay In the cell migration assay, HUVECs (3*10 4 ) in serum-free medium were seeded in a small chamber (3422, Corning) inserted into a 24-well plate. 10% FBS-containing medium was added to the plate located below the chamber to serve as chemical attractant. After 24 h, the migrating cells located on the lower surface of the chamber were stained with 0.1% crystal violet and counted. Adhesion assay HUVECs were pretreated with G-CSF. Then, HCT116-GFP cells were incubated with the treated HUVECs for 2 hours. The cells were washed with PBS and photographed under a fluorescence microscope. Quantitative RT–PCR Total RNA was extracted from cultured HUVECs with TRIzol (Invitrogen). The concentration of total RNA was quantified by measuring the absorbance at 260 nm. FAST qPCR with a TAKARA SYBR kit (RR420A, TAKARA) was used to perform qRT–PCR, and an Applied Biosystems PCR System (Fisher Scientific) was used to collect and analyze the data. The 2 −ΔΔCt method was used to determine the relative expression level of each gene. The primer sequences were as follows: VEGF, F 5’-TTGCTGCTCTACCTCCAC-3’ and R 5’-AATGCTTTCTCCGCTCTG-3’; E-selectin, F 5’-ATGTTCAAGCCTGGCAGTTCCG-3’ and R 5’-GCAGAGCCATTGAGCGTCCATC-3’; P-selectin, F 5’-CGCTCTGGACCAACCCTGTTTC-3’ and R 5’-CTCCTGGCTTCTGTGGCTTGTG-3’; ICAM, F 5’-TGCAAGAAGATAGCCAACCAAT-3’ and R 5’-GTACACGGTGAGGAAGGTTTTA-3’; VCAM, 5’-F CAGGCTGGAGATAGACTTACTG-3’ and R 5’-CCTCAATGACAGGAGTAAAGGT-3’; Western Blot Analysis Cells were lysed with cell lysis solution (9803s, CST) containing protease (B14001, Bimake) and phosphatase inhibitors (B15001, Bimake), and a BCA protein assay kit (P0011, Beyotime) was used for protein quantification. Equivalent amounts of cell lysates were separated by 10% SDS–PAGE, and the fractionated proteins were transferred to PVDF membranes (cat. no. IPVH00010, Millipore). After blocking with 5% skim milk, the membranes were incubated with primary antibodies against p-STAT3 (9145S, CST), STAT3 (9139S, CST), SELE (BBA16, R&D), SELP (sc-19672, Santa Cruz), ICAM (ab2213, Abcam) and CD31 (ab28364, Abcam). Then, the cells were incubated with a goat anti-rabbit or anti-mouse IgG-HRP secondary antibody (R, ab6721; M, ab6789; Abcam). Actin was used as the loading control (ab6276，Abcam). Protein expression was assessed with enhanced chemiluminescent substrate (WBKLS0500, Millipore, USA). Histopathological assay The tissue was fixed in 10% formalin, embedded in paraffin, and then sectioned (5 mm thick). The histopathology assay procedure was completed by conventional hematoxylin-eosin (H&E) staining according to standard techniques. Immunofluorescence HUVECs were fixed with methanol for 15 minutes and blocked with 5% bovine serum albumin (BSA) for 30 min at room temperature. The cells were incubated overnight at 4°C with primary antibodies (against p-STAT3, 9145S, CST; STAT3, 9139S, CST; SELE, BBA16, R&D; SELP, sc-19672, Santa Cruz; ICAM, ab2213, Abcam) and then incubated with secondary antibodies for 1h at room temperature. Each step was followed by washing 3 times for 5 minutes in PBS. The prepared specimens were counterstained with DAPI (Beyotime, Shanghai, China) for 5 min and photographed with a Zeiss LSM880 confocal microscope. The tissue specimens were fixed with cold methanol for 10 minutes and blocked with 5% BSA for 1h. Primary antibodies (Extended Data Table 1) were applied in blocking buffer, incubated overnight at 4 °C, and then the cells were incubated with Alexa Fluor® 488 (ab150113, Abcam) and Alexa Fluor® 647-labeled (ab150075, Abcam) secondary antibodies for 2h at room temperature. Nuclei were stained with DAPI. Images were taken by a Zeiss LSM880 confocal microscopy. Immunohistochemistry (IHC) The tissue was fixed in 10% formalin, embedded in paraffin, and then sectioned (5 mm thick). Paraffin sections were deparaffinized in xylene and rehydrated in a series of graded alcohols. The antigen was retrieved in 0.01 M sodium citrate buffer. The sections were incubated with 3% H 2 O 2 for 10 min and then blocked in 5% BSA for 30 minutes at room temperature. The sections were incubated overnight at 4 °C with CD31 (77699, CST), CD4 (ab183685, abcam) and CD8 (ab4055, abcam) primary antibodies. The appropriate secondary antibody was used for 1 h at room temperature. An EnVision (K4007, Dako) signal enhancement system was used to develop the bound antibodies. The sections were stained with Harris hematoxylin, dehydrated and fixed. Images were captured by a microscope (Leica, Wetzlar, Germany). Images were processed using ImageJ to quantify CD31-positive areas. CD31-positive areas. Patient samples. The serum and tumor samples were collected from 39 patients diagnosed with colorectal cancer who received neoadjuvant chemotherapy followed by surgical resection at Putuo Hospital, Shanghai University of Traditional Chinese Medicine, China, from January 2015 to December 2016. Written informed consent was obtained from the patients, in accordance with the institutional guidelines, before sample collection, and the study was approved by the Committees for the Ethical Review of Research at the Putuo Hospital, Shanghai University of Traditional Chinese Medicine, China. Informed consent for the use of samples for research purposes was obtained from the participants. Statistical Analysis Statistical analyses were performed using GraphPad Prism 8 and SPSS. For all experiments, three technical replicates were analyzed by counting a minimum of 3 fields/ technical replicate. Error bars in the experiments indicate standard deviation (s.d.) The legends illustrate the number of events and independent experiments, as well as the information related to the statistical details and methods. Statistical significance calculated using one-way ANOVA with Dunnett’s test. P values of clinical data were determined by two-tailed Pearson correlation coefficient test. P values <0.05 were considered statistically significant. Results Chemotherapy could promote metastasis by affecting vascular properties . Accumulating evidence suggested that chemotherapy could promote series of changes that facilitate metastasis [1,2]. Here we used a lung metastasis model to evaluate the effect of chemotherapeutic drugs commonly used to treat CRC patients. As schematized in Extended Data Fig. 1A, we injected CT-26-Luc cancer cells into BALB/c mice via the tail vein, followed by saline control, CPT11, OXA or 5-FU treatment and analysis of metastatic lesions. Mice received OXA had most metastatic sites in the lung (Extended data Fig. 1B, C). We found microvascular density increased dramatically in the metastatic lesions. In addition, STAT3 was markedly activated in the vessels after treatment with OXA MTD (maximal tolerating dose) (Extended data Fig. 1D). These data indicated that certain chemotherapeutic agent could activate STAT3 in vascular endothelial cells, which is consistent with its role in promoting angiogenesis, a critical step to facilitate metastasis. To study the role of endothelial cells in lung metastasis, we generated mice with conditionally inactivated STAT3 in endothelial cells (STAT3 flox/flox Tek-cre). Compared to wild-type mice (STAT3 flox/flox ) (Fig. 1A), the weight of the mice in all groups was not affected by chemotherapy (Fig. 1B). As shown in Fig. 1C and 1D, STAT3 flox/flox Tek-cre mice treated with chemotherapy had smaller metastatic areas than their KO counterparts (KO-OXA). Although deletion of STAT3 in ECs did not affect the formation of blood vessels in the metastatic lesions (Fig. 1E, 1F), it did reduce the vascular adhesion molecule Selectin E (SELE) (Fig. 1G). These data suggested that STAT3 KO in ECs reduced OXA-induced metastasis by affecting vascular adhesion properties, not necessarily angiogenesis. We therefore need to explore alternative mechanism of new vessel formation. Oxaliplatin promoted metastasis by increasing immunosuppression. Myeloid cells have also been implicated in contributing to angiogenesis and to the generation of an immunosuppressive tumor microenvironment [25]. Consistently, we also found an enrichment of myeloid-derived suppressor cells (MDSCs, which are known to be very potent suppressors of cytotoxic T-cell immunity) and scarce CD4 + and CD8 + T cells in the metastatic lesions after chemotherapy. Interestingly, STAT3 KO did not influence the infiltration of MDSCs or CD4 + and CD8 + T cells compared to that of the WT group regardless whether or not the mice received OXA chemotherapy (Extended data Fig. 2). These data indicate that although OXA promoted an immunosuppressive phenotype, it was not through the activation of STAT3 in ECs. Oxaliplatin promoted metastasis by affecting vasculogenesis. Vasculogenesis, a key paradigm of neovascularization, refers to the de novo formation of blood vessels derived from stem/progenitor cells [7]. Several studies suggested that EPC recruitment could contribute to tumor neovascularization and targeting EPCs might be an attractive antiangiogenic treatment [8]. We further explored the reason for the upregulation of vessel density by chemotherapy. We focused on EPCs. As expected, EPC markers (CD34 + and VEGFR2 + ) showed higher expression in the OXA group than in the Ctrl group (Fig. 1 I, J, K). To track EPC homing and retention in tumor metastasis sites, experiments were performed using GFP + bone marrow cells obtained from C57BL/UBI/GFP mice that were transplanted into C57BL/6 mice that had been lethally irradiated (Fig. 2A). Four weeks later, the mice were used as recipients for an injection of OXA followed three days later by a tail vein injection of MC-38-Luc cells. Fourteen days later, their lungs were obtained for the evaluation of GFP+ bone marrow cell colonization and incorporation into the tumor vessels, as shown in Fig. 2B. We found an increase in bone marrow-derived GFP + cells in the lung metastasis sites after treatment with OXA. CD31 + and GFP + cells represented EPC-modulated vasculogenesis. The colonization of EPCs in the metastasis site was higher in the OXA group than in the control group (Fig. 2C). To study whether chemotherapy can induce the mobilization of EPCs from bone marrow to plasma, we first monitored the levels of EPCs for up to 72 hours after chemotherapy drug injection in C57BL/6 mice (Fig. 2D) or BALB/c mice (Extended Fig. 3A). The proportion of EPC from bone marrow decreased with time, and proportion from the plasma increased (Fig. 2E; Extended Fig. 3B, 3C). Next, we tested the plasma contents of stromal cell-derived factor-1α (SDF-1α) and VEGF, which are secreted by EPCs, and found that their level increased along with the mobilization of EPCs to the plasma (Fig. 2F, G). Myeloid cells have also been implicated in contributing to an angiogenic-to-vasculogenic switch [26]. Subsequently, the FACS results showed that OXA chemotherapy significantly promoted the mobilization of MDSCs (Fig. 2H; Extended Fig. 3D). These data indicated that chemotherapy promoted the formation of blood vessels by mobilizing EPCs. In addition, it mobilized MDSCs to contribute to tumor vasculogenesis and immunosuppression. Chemotherapy promoted colorectal cancer metastasis through vessel formation by upregulating G-CSF. Cytokines are extracellular signaling molecules that may facilitate the mobilization of EPCs and/or affect vascular properties. To further assess this possibility, we obtained plasma samples from C57/Bl6 mice treated with OXA. We found multiple cytokines including G-CSF, IP-10, KC and JE were upregulated following OXA treatment. G-CSF, which is known to promote the mobilization of stem/progenitor cells, had highest fold of change based on immunoblot (Fig. 3A). This was further confirmed by ELISA (Fig. 3B). Although KC, JE, and IP-10 have previously been reported to be associated with angiogenesis, there are very few reports about the effects of KC, JE, and IP-10 on EPC mobilization. Next, we found that G-CSF was upregulated to a high level at 2 h following chemotherapy and this decreased over the next 6 hours (Fig. 3C). To further evaluate the proangiogenic roles of G-CSF, we performed G-CSF on HUVECs. G-CSF promoted HUVEC tube formation (Fig. 3D) and adhesion (Fig. 3E) along with STAT3 phosphorylation (Fig. 4F, 4I). However, it did not affect the proliferation or migration of HUVECs (Extended Fig. 4A, 4B). In addition, we observed increased level of cell adhesion molecules including SELE, SELP and ICAM by G-CSF (Fig. 4G, 4H, 4J-L). Collectively, these data suggest that G-CSF could activate p-STAT3 and promote vascular adhesion following OXA treatment. To examine the effect of G-CSF on the vessel properties of EPCs, we performed G-CSF on mouse primary EPCs. Our results showed that G-CSF promoted EPCs tube formation (Fig. 3M) and migration (Fig. N-O). In summary, G-CSF affected not only vascular adhesion but also vasculogenesis. Anti-G-CSF reduced chemotherapy-induced lung metastasis by inhibiting blood vessel formation Based on the above experimental results, we questioned whether anti-G-CSF antibodies could be used to prevent and/or inhibit metastasis induced by chemotherapy. To assess this, C57BL/6 mice were treated with OXA and then injected with anti-G-CSF two hours later. After three days, we injected MC-38-Luc cancer cells into their tail vein (Fig. 4A). Similarly, the CT26 BALB/c metastasis model was used and is shown in Extended Data Fig. 5A. As expected, anti-G-CSF reduced lung metastasis in the mice (Fig. 4B-D; Extended data Fig. 5B-D). In addition, the lung tissue vessel density was significantly decreased (Fig. 4E; Extended data Fig. 5E), and the activation of STAT3 in ECs was markedly reduced by anti-G-CSF (Fig. 4F; Extended data Fig. 5F). Anti-G-CSF also reduced SELE expression in the vasculature (Fig. 4G; Extended data Fig. 5G). We further investigated whether cells from bone marrow homing to metastasis sites following chemotherapy could be prevented by anti-G-CSF. A decrease of MDSCs was indeed found in the metastasis sites (Fig. 4H; Extended data Fig. 5H), and CD4 + and CD8 + T cells (Extended data Fig. 6A, 6B) were increased. Next, we found that the expression of the EPC markers CD34 and VEGFR2 in metastatic tumors was lower in the anti-G-CSF group (Fig. 4I-J; Extended data Fig. 5I-J). To track EPC homing and retention in tumor metastasis sites, GFP + trans mice were used as recipients for an injection of OXA and then injected with anti-G-CSF two hours later, as shown in Fig. 5A. The EPCs from the bone marrow (CD31 + GFP + cells) were reduced by anti-G-CSF (Fig. 5B). To study the impact of anti-G-CSF on the chemotherapy-induced mobilization of EPCs and MDSCs, C57BL/6 mice were treated with OXA and then injected with anti-G-CSF two hours later. As shown in Fig. 5C, bone marrow and plasma were collected and measured 72 hours after chemotherapy. Anti-G-CSF abolished the upregulated mobilization of EPCs from bone marrow to plasma (Fig. 5D, 5E) induced by chemotherapy. Anti-G-CSF also significantly reduced the mobilization of MDSCs (Fig. 5F). Similarly, the expression of SDF-1α and VEGF was reversed by anti-G-CSF. (Fig. 5G, H). These data indicated that anti-G-CSF reduced oxaliplatin-induced lung metastasis by inhibiting vasculogenesis and immunosuppression. G-CSF is positively associated with disease recurrence and metastasis in CRC patients To evaluate the contribution of G-CSF in clinical setting, we examined G-CSF level in plasma obtained from CRC patients who underwent surgical resection followed by adjuvant XELOX chemotherapy regimen containing OXA. We collected patients’ plasma before and after 1 course of XELOX regimen. First, we correlated G-CSF before surgical resection to recurrence and metastasis. Survival analysis indicated that patients with high G-CSF level in their plasma had significantly shorter overall survival and recurrence-free survival (Fig. 6A-B). In addition, higher G-CSF correlated with more advanced nodal and remote metastasis (Fig. 6C-D). Analyses of blood vessels (CD31; Figure 6E) and EPC-produced cytokines (SDF-1α and VEGF; Fig. 6F) also demonstrated that G-CSF level positively correlated with increased tumor vasculature. Next, we correlated G-CSF level of post one cycle of XELOX with CRC recurrence and metastasis. In consistent with the in vivo experiments, G-CSF level increased in majority of patients after receiving XELOX (Fig. 6G). More importantly, increased G-CSF level after chemotherapy was closely associated with poorer overall survival (Fig. 6H) and recurrence-free survival (Fig. 6I). Similarly, SDF-1α and VEGF levels increased significantly after XELOX treatment (Fig. 6 J-I). Of note, SDF-1α and VEGF levels were significantly higher in patients with their G-CSF increased after chemotherapy (Fig. 6L), and more than 80% of those patients had increased SDF1 and VEGF (Fig. 6K). SDF-1α and VEGF showed a significant positive correlation with the production of G-CSF, and the correlation coefficient increased after chemotherapy (Fig. 6M-N). These results suggested that the number of EPCs in the plasma positively correlated with G-CSF level following chemotherapy. Taken together, these data indicate that elevated G-CSF in the serum following chemotherapy likely enhance blood vessel formation and could be used as a prognostic biomarker to predict a poor outcome. Tracking G-CSF level after chemotherapy and optimizing appropriate intervention will have clinical value to benefit patients from chemotherapy while simultaneously avoid increased risk of metastasis. Discussion Chemotherapy remains the mainstay treatment for locally advanced/metastatic cancers. In addition, it has proven clinical value in various neoadjuvant and adjuvant settings. However, recent studies have also demonstrated that chemotherapy can induce local and systemic pro-tumorigenic/metastatic changes, raising the concern of increased risk of metastasis facilitated by certain chemotherapy regimens [ 28 , 29 ]. Identifying the underlying mechanisms and validating such observation in the clinical setting are therefore critically important to uncouple the therapeutic benefit and undesirable risk of chemotherapy for our cancer patients. In this study, we have illustrated that at least certain chemotherapy drug such as OXA could increase the level of circulating G-CSF, resulting in enhanced lung metastasis of colorectal cancer (as well as lung cancer and melanoma, Extended data Fig. 7). Such elevated level of G-CSF enhanced vascular adhesion via G-CSF/STAT3 signaling, and augmented EPC mobilization to induce vasculogenesis, both are critical to directly facilitate metastasis. In addition, OXA facilitated the mobilization of MDSCs from bone marrow to promote immunosuppressive phenotype. We also showed anti-G-CSF could reverse above-mentioned process and successfully block metastasis promoted by OXA chemotherapy. More importantly, in our CRC patients, the baseline and post-chemotherapy G-CSF levels positively correlated to worse clinical outcome. The formation of new blood vessels is an essential component of malignant tumor development and progression [ 6 ]. It has become increasingly evident that this process utilizes various mechanisms, including angiogenesis [ 30 ], vasculogenesis [ 31 ], vasculogenic mimicry [ 32 ] and vessel co-option [ 33 ], etc. We found that chemotherapy drug OXA activated STAT3 in a G-CSF dependent manner. STAT3 is a transcription factor that regulates a variety of cellular events. Many studies have suggested that the activation of STAT3 promotes tumor angiogenesis [ 34 – 36 ]. STAT3 knockout in ECs mitigated the ability of chemotherapy to exacerbate metastasis. However, deletion of STAT3 in ECs did not affect the formation of blood vessels in metastatic tumors or affect vascular adhesion molecules. G-CSF promoted HUVEC tube formation and adhesion but not proliferation and migration by activating STAT3. These results indicated that STAT3 KD in ECs reduced OXA-induced metastasis by affecting vascular adhesion properties, not angiogenesis. Chemotherapy-induced sprouting vessels are dependent on multiple mechanisms that work together. We need to further explore the mechanism of new vessel formation. Vasculogenesis refers to the de novo formation of blood vessels derived from stem/progenitor cells [ 37 ]. Yuval Shaked et al. [ 10 ] showed that paclitaxel can rapidly induce bone marrow-derived EPC mobilization and subsequent tumor homing, whereas gemcitabine does not. To unearth the mechanism of new vessel formation induced oxaliplatin chemotherapy, we focused on EPC mobilization. Our results showed that EPC mobilization to vasculogenesis induced by OXA is critical for lung metastasis. In addition, EPCs produce a variety of proangiogenic cytokines and growth factors, such as SDF-1α and VEGF. SDF-1 is a key chemokine that regulates the transport of hematopoietic stem cells between the bone marrow and peripheral circulation and it attracts EPCs to ischemic areas. VEGF promotes the proliferation and migration of pre-existing ECs, contributing to angiogenesis [ 38 – 40 ]. In this work, we found that the expression of SDF-1α and VEGF was upregulated by OXA chemotherapy. Circulating EPCs derived from bone marrow (BM) might differentiate into mature endothelial cells [ 41 ]. We found that G-CSF could promote primary EPC differentiation into mature endothelial cells. MDSCs are a type of immunosuppressive cell population, and their response to tumor antigens presented as peptides on their surface contributes to antitumor escape. They may negatively affect immune responses by regulating the function and proliferation of CD4 + and CD8 + cells [ 42 – 43 ]. Myeloid cells have also been implicated in contributing to an angiogenic-to-vasculogenic switch [ 26 ]. We found that MDSCs rapidly mobilized from the bone marrow in response to OXA chemotherapy and that the infiltration of MDSCs into the lung tissue was improved. STAT3 KO affected vascular adhesion properties but did not affect angiogenesis or the infiltration of MDSCs. These data suggest that OXA-induced MDSC mobilization also contributed to metastasis by immunosuppression and vasculogenesis. Bone marrow-derived stem/progenitor cells could be strongly mobilized into circulation by G-CSF [ 44 , 45 ]. Exogenous G-CSF administration could exacerbate solid tumor metastasis [ 46 ]. In our work, we found that G-CSF was upregulated by OXA chemotherapy. We used anti-G-CSF to abolish the mobilization of EPCs from bone marrow to plasma and lung metastases triggered by chemotherapy. The lung tissue vessel density was significantly decreased. Similarly, the expression of SDF-1α and VEGF was reversed by anti-G-CSF. Anti-G-CSF also changed the vascular adhesion ability by reducing SELE expression in the vasculature via STAT3. In addition, anti-G-CSF may repress tumoral immunosuppression to trigger antitumor immune responses by preventing MDSC mobilization. These data demonstrated that anti-G-CSF could be used as an adjuvant treatment for metastasis caused by chemotherapy. In the clinical study, we found that G-CSF was significantly upregulated by the XELOX regimen. Survival analysis indicated that patients with upregulated G-CSF in their serum had significantly worse survival outcomes and higher recurrence and metastasis rates. We also investigated the expression of endogenous G-CSF in patients with colon cancer. We found that patients with high expression of G-CSF in their serum had significantly worse survival outcomes and higher recurrence and metastasis rates. Patients in the high G-CSF expression group had more blood vessels in their tumor tissue. Consistent with our findings, in human tumors, there are many reports of extremely aggressive G-CSF in multiple tumor types, demonstrating the important and modulatory effect of G-CSF in tumor progression and metastasis [ 47 – 49 ]. As G-CSF is commonly used to prevent/treat serious neutropenia during cancer chemotherapy [ 50 – 51 ], our study did provide mechanistic evidence to caution the use of G-CSF and balance its benefits and potential risks. This brings important questions regarding the timing, duration and circumstances under which we shall consider using G-CSF. Although an absolute neutrophil count equal or less than 500/mm 3 is generally considered an indication to use G-CSF in clinical practice, more studies are needed to optimize the duration of use, along with dose reduction of chemotherapy as the later could also up-regulate G-CSF based on our current study. Interestingly, Peishan Li et al. [ 52 ] also found that administration of exogenous G-CSF promoted metastasis in host mice with intact NK cells but achieved opposite antimetastatic effect in those mice lacking functional NK cells, suggesting host immune status may also need to be considered for a personalized treatment decision. In summary, our study has illustrated that at least certain chemotherapeutic agent such as oxaliplatin could facilitate cancer metastasis via G-CSF induced neovascularization. This cautions routine use of G-CSF during cancer chemotherapy and quests optimal chemotherapy regimen and use of anti-G-CSF therapy that warrants further exploration. Abbreviations EPCs Endothelial Progenitor Cells BM bone marrow PB peripheral blood G-CSF Granulocyte-colony stimulating factor G-CSFR G-CSF receptor HUVECs Human umbilical vein endothelial cells H&E hematoxylin-eosin BSA bovine serum albumin MDSCs myeloid-derived suppressor cells SELE Selectin E OXA Oxaliplatin. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Availability of data and materials The datasets during and/or analyzed during the current study are available from the corresponding authors upon reasonable request. Competing interests The authors declare that they have no conflict of interest. Funding This project was sponsored by the Science and Technology Innovation Project of Putuo District Health System (ptkwws201905, ptkwws202004), the Natural Science Foundation of Shanghai (20ZR1450500), Shanghai Key Medical Specialty Construction Project (NO. ZK2019B18), Clinical Specialized Disease Construction Project of Shanghai Putuo District Municipal Health Comission (NO. 2019tszb01), the National Key Research and Development Program of China (grant number 2019YFC1316000), as well as the Scientific Research Project of Putuo Central Hospital (2020362A). Authors' contributions K.X., X.L., and P.Y., conceived and directed the project. K.X. designed the experiments. Y.Z., K.F., Y.W., M.L., R.W., J.X., and Z.Y. carried out the experiments. J.W., C.W., R.Z. J.X., T.C., and J.W. were responsible for collecting tissue specimen. K.X., Y.Z., and K.F. conducted the data analysis and interpreted the results. Y.Z., K.X., JZ, K.F., Y.W., and P.Y. wrote, criticized and edited the paper. All authors read and approved the final manuscript. Acknowledgements Special thanks to Dr. Teng Chen and Dr. Jianhua Xu for the help during the study. References Liu Y, Cao X. Organotropic metastasis: role of tumor exosomes. Cell Res 2016;26:149–50 Doglioni G, Parik S, Fendt SM. Interactions in the (Pre)metastatic Niche Support Metastasis Formation. Front Oncol 2019;9:219 Yang JM, Yang GY, Medina DJ, Vassil AD, Liao J, Hait WN. Treatment of multidrug resistant (MDR1) murine leukemia with P-glycoprotein substrates accelerates the course of the disease. Biochem Biophys Res Commun 1999;266:167–73 Byrd-Leifer CA, Block EF, Takeda K, Akira S, Ding A. The role of MyD88 and TLR4 in the LPS-mimetic activity of Taxol. Eur J Immunol 2001; 31:2448–57 Kaplan RN, Rafii S, Lyden D. Preparing the "soil": the premetastatic niche. Cancer Res 2006;66:11089–93 Shaked Y. The pro-tumorigenic host response to cancer therapies. Nat Rev Cancer 2019;19:667–85 de la Puente P, Muz B, Azab F, Azab AK. Cell trafficking of endothelial progenitor cells in tumor progression. Clin Cancer Res 2013;19:3360–8 Asahara T, Murohara T, Sullivan A, Silver M, van der Zee R, et al. Isolation of putative progenitor endothelial cells for angiogenesis. Science 1997;275:964–7 Fadini GP, Losordo D, Dimmeler S. Critical reevaluation of endothelial progenitor cell phenotypes for therapeutic and diagnostic use. Circ Res 2012;110:624–37 Shaked Y, Henke E, Roodhart JM, Mancuso P, Langenberg MH, et al. Rapid chemotherapy-induced acute endothelial progenitor cell mobilization: implications for antiangiogenic drugs as chemosensitizing agents. Cancer Cell 2008;14:263–73 Roodhart JM, Langenberg MH, Vermaat JS, Lolkema MP, Baars A, et al. Late release of circulating endothelial cells and endothelial progenitor cells after chemotherapy predicts response and survival in cancer patients. Neoplasia 2010;12:87–94 Sayed HAR, Hetta HF, Shafik EA, Zahran AM. Flow cytometric analysis of circulating endothelial cells and endothelial progenitor cells in pediatric solid tumors: prognostic impact on treatment response and survival. Cancer Immunol Immunother 2021;70:755–61 Yeo B, Redfern AD, Mouchemore KA, Hamilton JA, Anderson RL. The dark side of granulocyte-colony stimulating factor: a supportive therapy with potential to promote tumour progression. Clin Exp Metastasis 2018; 35:255–67 Wang K, Dai X, He J, Yan X, Yang C, et al. Endothelial Overexpression of Metallothionein Prevents Diabetes-Induced Impairment in Ischemia Angiogenesis Through Preservation of HIF-1α/SDF-1/VEGF Signaling in Endothelial Progenitor Cells. Diabetes 2020;69:1779–92 Liu Y, Xu J, Gu R, Li Z, Wang K, et al. Circulating exosomal miR-144-3p inhibits the mobilization of endothelial progenitor cells post myocardial infarction via regulating the MMP9 pathway. Aging (Albany NY) 2020;12:16294–303 Cetean S, Căinap C, Constantin AM, Căinap S, Gherman A, et al. The importance of the granulocyte-colony stimulating factor in oncology. Clujul Med 2015;88:468–72 Morris KT, Khan H, Ahmad A, Weston LL, Nofchissey RA, et al. G-CSF and G-CSFR are highly expressed in human gastric and colon cancers and promote carcinoma cell proliferation and migration. Br J Cancer 2014;110:1211–20 Sacks D, Baxter B, Campbell BCV, Carpenter JS, Cognard C, et al. Multisociety Consensus Quality Improvement Revised Consensus Statement for Endovascular Therapy of Acute Ischemic Stroke. Int J Stroke 2018;13:612–32 Martino N, Ramos RB, Lu S, Leyden K, Tomaszek L, et al. Endothelial SOCS3 maintains homeostasis and promotes survival in endotoxemic mice. JCI Insight 2021;6 Demetri GD, Griffin JD. Granulocyte colony-stimulating factor and its receptor. Blood 1991;78:2791–808 Wculek SK, Malanchi I. Neutrophils support lung colonization of metastasis-initiating breast cancer cells. Nature 2015;528:413–7 El Rayes T, Catena R, Lee S, Stawowczyk M, Joshi N, et al. Lung inflammation promotes metastasis through neutrophil protease-mediated degradation of Tsp-1. Proc Natl Acad Sci U S A 2015;112:16000–5 Coffelt SB, Kersten K, Doornebal CW, Weiden J, Vrijland K, et al. IL-17-producing γδ T cells and neutrophils conspire to promote breast cancer metastasis. Nature 2015;522:345–8 Shaked Y, Ciarrocchi A, Franco M, Lee CR, Man S, et al. Therapy-induced acute recruitment of circulating endothelial progenitor cells to tumors. Science 2006;313:1785–7 Yang L, DeBusk LM, Fukuda K, Fingleton B, Green-Jarvis B, et al. Expansion of myeloid immune suppressor Gr + CD11b + cells in tumor-bearing host directly promotes tumor angiogenesis. Cancer Cell 2004;6:409–21 Greenfield JP, Cobb WS, Lyden D. Resisting arrest: a switch from angiogenesis to vasculogenesis in recurrent malignant gliomas. J Clin Invest 2010;120:663–7 Valastyan S, Weinberg RA. Tumor metastasis: molecular insights and evolving paradigms. Cell 2011;147:275–92 Gingis-Velitski S, Loven D, Benayoun L, Munster M, Bril R, et al. Host response to short-term, single-agent chemotherapy induces matrix metalloproteinase-9 expression and accelerates metastasis in mice. Cancer Res 2011;71:6986–96 Daenen LG, Roodhart JM, van Amersfoort M, Dehnad M, Roessingh W, et al. 2011. Chemotherapy enhances metastasis formation via VEGFR-1-expressing endothelial cells. Cancer Res 71:6976–85 Veith AP, Henderson K, Spencer A, Sligar AD, Baker AB. Therapeutic strategies for enhancing angiogenesis in wound healing. Adv Drug Deliv Rev. 2019;146: 97–125. Wang W, Zhang Y, Hui H, Tong W, Wei Z, et al. The effect of endothelial progenitor cell transplantation on neointimal hyperplasia and reendothelialisation after balloon catheter injury in rat carotid arteries. Stem Cell Res Ther 2021;12:99 Young MR, Newby M, Wepsic HT. Hematopoiesis and suppressor bone marrow cells in mice bearing large metastatic Lewis lung carcinoma tumors. Cancer Res 1987;47:100–5 Ray P, Arora M, Poe SL, Ray A. Lung myeloid-derived suppressor cells and regulation of inflammation. Immunol Res 2011;50:153–8 Wei D, Le X, Zheng L, Wang L, Frey JA, et al. Stat3 activation regulates the expression of vascular endothelial growth factor and human pancreatic cancer angiogenesis and metastasis. Oncogene 2003;22:319–29 Niu G, Wright KL, Huang M, Song L, Haura E, et al. Constitutive Stat3 activity up-regulates VEGF expression and tumor angiogenesis. Oncogene 2002;21:2000–8 Kujawski M, Kortylewski M, Lee H, Herrmann A, Kay H, Yu H. Stat3 mediates myeloid cell-dependent tumor angiogenesis in mice. J Clin Invest 2008;118:3367–77 Fujita Y, Kawamoto A. Stem cell-based peripheral vascular regeneration. Adv Drug Deliv Rev 2017;120:25–40 Ii M, Nishimura H, Iwakura A, Wecker A, Eaton E, et al. Endothelial progenitor cells are rapidly recruited to myocardium and mediate protective effect of ischemic preconditioning via "imported" nitric oxide synthase activity. Circulation 2005;111:1114–20 Jujo K, Ii M, Losordo DW. Endothelial progenitor cells in neovascularization of infarcted myocardium. J Mol Cell Cardiol 2008;45:530–44 Miyamoto Y, Suyama T, Yashita T, Akimaru H, Kurata H. Bone marrow subpopulations contain distinct types of endothelial progenitor cells and angiogenic cytokine-producing cells. J Mol Cell Cardiol 2007;43:627–35 Wang W, Zhang Y, Hui H, Tong W, Wei Z, et al. The effect of endothelial progenitor cell transplantation on neointimal hyperplasia and reendothelialisation after balloon catheter injury in rat carotid arteries. Stem Cell Res Ther 2021;12:99 Lamsfus Calle C, Fendel R, Singh A, Richie TL, Hoffman SL, et al. Expansion of Functional Myeloid-Derived Suppressor Cells in Controlled Human Malaria Infection. Front Immunol 2021;12:625712 Tanaka R, Masuda H, Kato S, Imagawa K, Kanabuchi K, et al. Autologous G-CSF-mobilized peripheral blood CD34 + cell therapy for diabetic patients with chronic nonhealing ulcer. Cell Transplant 2014;23:167–79 Hannoush EJ, Sifri ZC, Elhassan IO, Mohr AM, Alzate WD, et al. Impact of enhanced mobilization of bone marrow derived cells to site of injury. J Trauma 2011;71:283–9; discussion 9–91 Kowanetz M, Wu X, Lee J, Tan M, Hagenbeek T, et al. Granulocyte-colony stimulating factor promotes lung metastasis through mobilization of Ly6G + Ly6C + granulocytes. Proc Natl Acad Sci U S A 2010;107:21248–55 Jardin F, Vasse M, Debled M, Dominique S, Courville P, et al. Intense paraneoplastic neutrophilic leukemoid reaction related to a G-CSF-secreting lung sarcoma. Am J Hematol 2005;80:243–5 Kawano M, Mabuchi S, Matsumoto Y, Sasano T, Takahashi R, et al. The significance of G-CSF expression and myeloid-derived suppressor cells in the chemoresistance of uterine cervical cancer. Sci Rep 2015;5:18217 Braun B, Lange M, Oeckler R, Mueller MM. Expression of G-CSF and GM-CSF in human meningiomas correlates with increased tumor proliferation and vascularization. J Neurooncol 2004;68:131–40 Schniewind B, Christgen M, Hauschild A, Kurdow R, Kalthoff H, Klomp HJ. Paraneoplastic leukemoid reaction and rapid progression in a patient with malignant melanoma: establishment of KT293, a novel G-CSF-secreting melanoma cell line. Cancer Biol Ther 2005;4:23–7 Kuderer NM, Dale DC, Crawford J, Cosler LE, Lyman GH. Mortality, morbidity, and cost associated with febrile neutropenia in adult cancer patients. Cancer 2006;106:2258–66 Mehta HM, Malandra M, Corey SJ. G-CSF and GM-CSF in Neutropenia. J Immunol 2015;195:1341–9 Li P, Lu M, Shi J, Hua L, Gong Z, et al. Dual roles of neutrophils in metastatic colonization are governed by the host NK cell status. Nat Commun 2020;11:4387 Additional Declarations No competing interests reported. Supplementary Files Extendeddata.pdf figure7Graphicalabstract.tif Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-1464660","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":92239670,"identity":"d6af2fad-a165-406d-a19c-75b5d25aca44","order_by":0,"name":"Yueping zhan","email":"","orcid":"","institution":"Shanghai University of Traditional Chinese Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Yueping","middleName":"","lastName":"zhan","suffix":""},{"id":92239671,"identity":"f8354f72-fe76-48ff-9212-c1dcea8cb4c3","order_by":1,"name":"Kai Fang","email":"","orcid":"","institution":"Shanghai University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Kai","middleName":"","lastName":"Fang","suffix":""},{"id":92239672,"identity":"d2273516-6261-4612-a839-517c9994ed4d","order_by":2,"name":"Yuqian wang","email":"","orcid":"","institution":"Shanghai University of Traditional Chinese Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Yuqian","middleName":"","lastName":"wang","suffix":""},{"id":92239673,"identity":"e0d97722-56cc-4e2f-b892-dd5708b6dc86","order_by":3,"name":"Mengting Liang","email":"","orcid":"","institution":"Shanghai University of Traditional Chinese Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Mengting","middleName":"","lastName":"Liang","suffix":""},{"id":92239674,"identity":"85ea3519-a225-468c-b779-77db3f81ac1a","order_by":4,"name":"Ruiying Wu","email":"","orcid":"","institution":"Shanghai University of Traditional Chinese Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Ruiying","middleName":"","lastName":"Wu","suffix":""},{"id":92239675,"identity":"91c2e372-f6f9-4123-affb-0cf8522aa20e","order_by":5,"name":"Jian Xu","email":"","orcid":"","institution":"Shanghai University of Traditional Chinese Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Jian","middleName":"","lastName":"Xu","suffix":""},{"id":92239676,"identity":"8b35f524-da89-44c2-8d2d-8195849c906d","order_by":6,"name":"Zeting Yuan","email":"","orcid":"","institution":"Shanghai University of Traditional Chinese Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Zeting","middleName":"","lastName":"Yuan","suffix":""},{"id":92239677,"identity":"12006458-553b-49e4-b2fb-2b4192f79652","order_by":7,"name":"Chengqi Wu","email":"","orcid":"","institution":"Shanghai University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Chengqi","middleName":"","lastName":"Wu","suffix":""},{"id":92239678,"identity":"65445c03-5f04-4ced-a8a8-177a1df6978a","order_by":8,"name":"Ruiqiu Zhu","email":"","orcid":"","institution":"Shanghai University of Traditional Chinese Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Ruiqiu","middleName":"","lastName":"Zhu","suffix":""},{"id":92239679,"identity":"566b8f97-9295-4322-bbbf-a97fd029c6b3","order_by":9,"name":"Jianhua Xu","email":"","orcid":"","institution":"Shanghai University of Traditional Chinese Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Jianhua","middleName":"","lastName":"Xu","suffix":""},{"id":92239680,"identity":"ffd2ec56-fe4b-4cfb-a33d-6dc3f2a5dadd","order_by":10,"name":"Teng Chen","email":"","orcid":"","institution":"Shanghai University of Traditional Chinese Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Teng","middleName":"","lastName":"Chen","suffix":""},{"id":92239681,"identity":"1dd7e954-b82a-467a-b538-5efbf6ba3177","order_by":11,"name":"Jie Wang","email":"","orcid":"","institution":"Shanghai University of Traditional Chinese Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Jie","middleName":"","lastName":"Wang","suffix":""},{"id":92239682,"identity":"09398f1b-0adf-4015-9ba3-f0fa59e56dc6","order_by":12,"name":"Jun Zhang","email":"","orcid":"","institution":"University of Kansas Medical Center","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Jun","middleName":"","lastName":"Zhang","suffix":""},{"id":92239683,"identity":"d7d277e6-6f9d-4268-a39c-14c65e487ae1","order_by":13,"name":"Peihao Yin","email":"","orcid":"","institution":"Shanghai University of Traditional Chinese Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Peihao","middleName":"","lastName":"Yin","suffix":""},{"id":92239684,"identity":"aa535cc8-bc21-408c-8303-7fcb43c40a20","order_by":14,"name":"Xin Liang","email":"","orcid":"","institution":"East China University of Science and Technology","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Xin","middleName":"","lastName":"Liang","suffix":""},{"id":92239685,"identity":"bb8b9262-8f38-44bc-a236-8799131525e4","order_by":15,"name":"Ke Xu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAr0lEQVRIiWNgGAWjYBACCRDxgeEAiDIgXgvjDJK1MPOQpEWyvcdM2ubPncQG9uZtEgw1dwhrkeY5liad2/YssYHnWJkEw7FnhLXISSQfk85tOJzYIJFjJsHYcJgILfIP26Qt/gC1yL8hUou0BPMxaQY2kC08RGqR7ElLtuxtO2zcxpNWbJFwjAgtEsfPGN748eewbD/74Y03PtQQoQUO2EBEAgkaRsEoGAWjYBTgAQCjpzcOLJBOrwAAAABJRU5ErkJggg==","orcid":"","institution":"Shanghai University","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Ke","middleName":"","lastName":"Xu","suffix":""}],"badges":[],"createdAt":"2022-03-18 07:59:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-1464660/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-1464660/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":19633561,"identity":"fe4855ee-1387-41ba-b32a-a4e03a531e71","added_by":"auto","created_at":"2022-03-25 22:07:41","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1830057,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSTAT3 KD in ECs reduces chemotherapeutic-induced metastasis by affect the vascular adhesion property.\u003c/strong\u003e A, Study design for B–H (n = 5 for Stat3\u003csup\u003eflox/flox\u003c/sup\u003e mice injected with PBS or OXA; n = 5 for Stat3\u003csup\u003eflox/flox\u003c/sup\u003e TEK-cre mice injected with PBS and OXA, respectively, three days later MC38-luc cells were injected by tail vein injection). B, The mice weight of each group. C, Luciferase-based bioluminescence imaging on indicated mice in A. For each group, five mice were used for quantification. D, Representative pictures and quantitative results of hematoxylin and eosin (H\u0026amp;E) staining of lung sections from indicated mice in A. Left panel: H\u0026amp;E images; right panel: the quantitative data for lung metastatic. E, The p-STAT3\u003csup\u003e+ \u003c/sup\u003eof CD31\u003csup\u003e+\u003c/sup\u003e cells were assayed by immunofluorescence (Left). The percent of p-STAT3\u003csup\u003e+\u003c/sup\u003e area was averaged from 10 random area (2 random area from each mice, n=5). Scale bars, 30 μm. F, Images of CD31 in tumor metastasis site (2 random area from each mice, n=5). Scale bars, 30 μm. G, H, The expression of SELE and SELP detected by immunofluorescence (2 random area from each mice, n=5). Scale bars, 30 μm. I, A, Study design for J–K (n = 5 for C57BL/6 mice injected OXA, or blank controls, three days later MC38-luc cells were injected by tail vein injection. J, K, The expression of SELE and SELP detected by immunofluorescence (2 random area from each mice, n=5). Scale bars, 30 μm. . Statistical significance calculated using one-way ANOVA with Dunnett’s test. Data represented as mean ± s.d.\u003c/p\u003e","description":"","filename":"figure1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-1464660/v1/f9ecb5f1405c53ef716658a8.jpg"},{"id":19634607,"identity":"80e92fe6-c63a-4c69-82a1-0861a5e702cc","added_by":"auto","created_at":"2022-03-25 22:17:41","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1713825,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMobilization EPCs and MDSCs to mediate vasculogenesis and immunosuppression after chemotherapy. \u003c/strong\u003eA, B, Study design for C (n = 5 for transplanted with green fluorescent protein\u003csup\u003e+\u003c/sup\u003e (GFP\u003csup\u003e+\u003c/sup\u003e)-bone marrow cells C57BL/6 mice injected with PBS and OXA, three days later MC38-luc cells were injected by tail vein injection\u003cstrong\u003e. C, \u003c/strong\u003eThe CD31\u003csup\u003e+\u003c/sup\u003e of GFP\u003csup\u003e+\u003c/sup\u003e cells were assayed by immunofluorescence (Left). The number of EPCs was averaged from 10 random area (2 random area from each mice, n=5). Scale bars, 30 μm. D, Study design for E-H, n=3. E, The mobilization of EPCs from bone marrow to plasma detected by FACS. F, G, Concentration of circulating SDF-1α and VEGF. Data representative of Three independent experiments. H, The mobilization of MDSCs from bone marrow to plasma detected by FACS. Statistical significance calculated using one-way ANOVA with Dunnett’s test. Data represented as mean ± s.d.\u003c/p\u003e","description":"","filename":"figure2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-1464660/v1/1fda9072ff21b42e5d0aeedb.jpg"},{"id":19634215,"identity":"8895be6b-b97b-4e15-8da8-87bba699f91c","added_by":"auto","created_at":"2022-03-25 22:12:41","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1674735,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eG-CSF exacerbates colorectal cancer metastasis through vessel formation. \u003c/strong\u003eA, The key cytokine after chemotherapy detected by Mouse Cytokine Array Panel. Data representative of Three independent experiments. B, C, The expression of G-CSF detected by ELISA. Data representative of Three independent experiments. D, In vitro assay for vascular mimicry of HUVECs in treatment of G-CSF (Left). Tube formation quantification (The number of nodes and brunches points) was analyzed under G-CSF treatment. E, The adhesion ability of HUVECs. Data representative of Three independent experiments. F-H. The proteins and mRNA detected by WB and RT-PCR. I-L, p-STAT3, SELE, ICAM and SELP were assayed by immunofluorescence. M, In vitro assay for the vessel formation of EPCs in treatment of G-CSF (Left). Tube formation quantification (brunches points) was analyzed under G-CSF treatment. N, O, The migration of EPCs evaluated by Wound Healing and Transwell test. Data representative of Three independent experiments. Statistical significance calculated using one-way ANOVA with Dunnett’s test. Data represented as mean ± s.d.\u003c/p\u003e","description":"","filename":"figure3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-1464660/v1/f2c22753f6a2f7e43c5943ce.jpg"},{"id":19633562,"identity":"770f5287-64ed-4374-82a8-52b48718d52b","added_by":"auto","created_at":"2022-03-25 22:07:41","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1968098,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAnti-G-CSF reduced the chemotherapy-induced lung metastasis by modulate vascular adhesion. \u003c/strong\u003eA, Study design for B–J (n = 5 for C57BL/6 mice injected OXA, two hours later i.p. injected with anti G-CSF or IgG; three days later MC38-luc cells were injected by tail vein injection). B, Luciferase-based bioluminescence imaging on indicated mice in A. For each group, five mice were used for quantification. C, D, The metastasis of lung imaging on indicated mice in A. E, Representative pictures and quantitative results of hematoxylin and eosin (H\u0026amp;E) staining of lung sections from indicated mice in A. Left panel: H\u0026amp;E images; right panel: the quantitative data for lung metastatic. F, The p-STAT3\u003csup\u003e+ \u003c/sup\u003eof CD31\u003csup\u003e+\u003c/sup\u003e cells were assayed by immunofluorescence (Left). The percent of p-STAT3\u003csup\u003e+\u003c/sup\u003e area was averaged from 10 random area (2 random area from each mice, n=5). Scale bars, 30 μm. F, Images of CD31 in tumor metastasis site (2 random area from each mice, n=5). Scale bars, 30 μm. G, H, The expression of SELE, MDSCs, CD34 and VEGFR2 detected by immunofluorescence (2 random area from each mice, n=5). Scale bars, 30 μm. Statistical significance calculated using one-way ANOVA with Dunnett’s test. Data represented as mean ± s.d.\u003c/p\u003e","description":"","filename":"figure4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-1464660/v1/6bfa7ecb3b175335feaaff63.jpg"},{"id":19633564,"identity":"27ea8abb-0d15-484a-bcff-9677ccb70b60","added_by":"auto","created_at":"2022-03-25 22:07:41","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1758139,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAnti-G-CSF reduced the mobilization of EPCs and MDSCs to mediate vasculogenesis and immunosuppression after chemotherapy. \u003c/strong\u003eA, Study design for C (n = 5 for transplanted with green fluorescent protein\u003csup\u003e+\u003c/sup\u003e (GFP\u003csup\u003e+\u003c/sup\u003e)-bone marrow cells C57BL/6 mice injected OXA, two hours later i.p. injected with anti G-CSF or IgG; three days later MC38-luc cells were injected by tail vein injection\u003cstrong\u003e. \u003c/strong\u003eB\u003cstrong\u003e, \u003c/strong\u003eThe CD31\u003csup\u003e+\u003c/sup\u003e of GFP\u003csup\u003e+\u003c/sup\u003e cells were assayed by immunofluorescence (Left). The number of EPCs was averaged from 10 random area (2 random area from each mice, n=5). Scale bars, 30 μm. C, Study design for D-H, n=3. D, E, The mobilization of EPCs from bone marrow to plasma detected by FACS. F, The mobilization of MDSCs from bone marrow to plasma detected by FACS.\u0026nbsp;G, H, Concentration of circulating SDF-1α and VEGF. Data representative of Three independent experiments. Statistical significance calculated using one-way ANOVA with Dunnett’s test. Data represented as mean ± s.d.\u003c/p\u003e","description":"","filename":"figure5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-1464660/v1/15f40f87c25438e4c19791b4.jpg"},{"id":19633566,"identity":"99db71f5-2969-4edc-b652-b0b5546ecd3c","added_by":"auto","created_at":"2022-03-25 22:07:41","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1703481,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eG-CSF is associated with recurrence and metastasis in CRC patients. \u003c/strong\u003eA, Kaplan-Meier analysis of overall survival in patients with variable G-CSF expression, according to the data from the selected CRC plasma samples (p = 0.023). B, Kaplan–Meier analyses of the correlations between G-CSF expression and Recurrence-free-Survival. (p = 0.0027) C, D, Pathological tumor-node-metastasis (TNM) stage, and proximal and distal edges analysis. E, The expression of CD31 were detected in tumor tissues by IHC. F, The concentration of SDF-1α and VEGF in the G-CSF high-expression group. Statistical significance calculated using one-way ANOVA with Dunnett’s test. G, The expression of G-CSF after chemotherapy. H, I, Kaplan-Meier analysis of overall survival and recurrence-free survical in patients with chemotherapy induced G-CSF expression, according to the data from the selected CRC plasma samples (p = 0.033). I, K, The concentration of SDF-1α and VEGF in the G-CSF up regulated group by chemotherapy. Statistical significance calculated using one-way ANOVA with Dunnett’s test. L, The percentage of up-regulation SDF-1α and VEGF in the G-CSF up regulated group by chemotherapy. Statistical significance calculated using one-way ANOVA with Dunnett’s test. M, N, Graphs depict correlation between the G-CSF expression and SDF-1α or VEGF in CRC samples before and after neoadjuvant chemotherapy; P values were determined by two-tailed Pearson correlation coefficient test (n= 39 patients).\u003c/p\u003e","description":"","filename":"fIGURE6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-1464660/v1/7b8819c75257df44723f4266.jpg"},{"id":19634611,"identity":"cd7af749-e93b-4171-be47-23f57532c72d","added_by":"auto","created_at":"2022-03-25 22:17:44","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1765979,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-1464660/v1/ef81ee5f-9de3-4c04-a905-10670518b555.pdf"},{"id":19634217,"identity":"74c97416-5ab0-40e8-a684-073edc11ce71","added_by":"auto","created_at":"2022-03-25 22:12:41","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":2961205,"visible":true,"origin":"","legend":"","description":"","filename":"Extendeddata.pdf","url":"https://assets-eu.researchsquare.com/files/rs-1464660/v1/09e959f030d2e31a0f95e8d1.pdf"},{"id":19633568,"identity":"b70993b0-b8e2-4da8-ac63-fa7030fbe70b","added_by":"auto","created_at":"2022-03-25 22:07:42","extension":"tif","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":16771012,"visible":true,"origin":"","legend":"","description":"","filename":"figure7Graphicalabstract.tif","url":"https://assets-eu.researchsquare.com/files/rs-1464660/v1/bab15ff071ccf6b97447db8e.tif"}],"financialInterests":"No competing interests reported.","formattedTitle":"Chemotherapy may facilitate colorectal cancer metastasis through vessel formation by upregulating G-CSF","fulltext":[{"header":"Background","content":"\u003cp\u003eCancer metastasis is a multi-step complicated process involving the spreading, localization and adaptation of cancer cells to the new environment in remote organs [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Although chemotherapy is one of the main effective treatments, recent studies have shown that it may also induce tumoral or host changes, such as EMT [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e], host-initiated proinflammatory responses [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e], and recruitment of BMDCs/MSCs to secondary sites [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e], which could paradoxically facilitate the proliferation and dissemination of cancer cells, hence metastasis. Deciphering such observation is therefore of great clinical value.\u003c/p\u003e \u003cp\u003eThe formation of new blood vessels is essential for the development and progression of cancer as it provides nutrition and oxygen and removes metabolic waste products [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Vasculogenesis, one of the important mechanisms of neovascularization, is initiated by the recruitment of EPCs [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e], which belong to a subtype of stem cell with high proliferative and differentiation potential and are mainly derived from bone marrow (BM) and peripheral blood (PB) progenitor cells [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Circulating EPCs can be recruited to avascular tumor sites to promote vasculogenesis via de novo blood vessel formation and production of proangiogenic cytokines [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eShaked et al. found that certain chemotherapy drugs could induce acute EPC mobilization from bone marrow [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Other evidence indicates that circulating EPCs and cytokines negative correlate to progression-free and overall survival in patients after chemotherapy [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. In addition, pediatric patients with solid tumors exhibited elevated levels of circulating endothelial cells and EPCs after chemotherapy [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. Furthermore, multiple cytokines and enzymes, such as G-CSF [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e], GM-CSF [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e] and MMP9 [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e], may directly or indirectly promote the mobilization of EPCs.\u003c/p\u003e \u003cp\u003eGranulocyte-colony stimulating factor (G-CSF) is a glycoprotein mainly produced by fibroblasts, endothelial cells, bone marrow stromal cells, monocytes and macrophages [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. G-CSF is well known to promote the survival, proliferation, and differentiation of neutrophil precursors by interacting with the G-CSF receptor (G-CSFR) [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e], which activates downstream JAK/STAT3 pathway [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Amplified JAK/STAT3 signaling could lead to increased adhesivity of the endothelial surface [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. Because of its capacity of promoting the mobilization, proliferation and differentiation of hematopoietic progenitor cells, G-CSF is widely used in clinical practice to treat chemotherapy induced neutropenia [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. However, recent preclinical studies suggested that exogenous G-CSF administration as well as other amplified neutrophil-associated pathological conditions may promote the metastasis of solid tumors [\u003cspan additionalcitationids=\"CR22\" citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e], therefore raised clinical concern that needs scientific explanation.\u003c/p\u003e \u003cp\u003eIn this study, we investigated the mechanism by which chemotherapy could promote colon cancer lung metastasis via elevated G-CSF and its role in tumor vasculature during chemotherapy. We found that G-CSF/STAT3 signaling enhanced vascular adhesion properties and EPC mobilization devoted to vessel density in metastasis sites. In addition, chemotherapy improved the mobilization of MDSCs from bone marrow. Anti-G-CSF given at the right time may suppress the formation of a functional vasculature and induce anti-tumor immune function, and prevent metastasis during chemotherapy.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eCell culture\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHuman umbilical vein endothelial cells (HUVECs, #8000,\u0026nbsp;ScienCell, USA) and mouse EPCs (CP-M140, Procell, China) were cultured in endothelial cell medium containing 10% FBS and 1% penicillin/streptomycin. HUVECs within the first three passages were used for experiments. CT26-Luc, MC38-Luc and Lewis-Luc cells were cultured in RPMI-1640 containing 10% FBS and 1% penicillin/streptomycin.\u0026nbsp;B16F1\u0026nbsp;cells were cultured in DMEM containing 10% FBS and 1% penicillin/streptomycin. All cells were cultured at 37°C in a humidified incubator with\u0026nbsp;5%\u0026nbsp;CO\u003csub\u003e2\u003c/sub\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTumors and Animal Models\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEight- to twelve-week-old C57BL/6 or BALB/c mice (Songlian Experimental Animal Institute, Songjiang District, Shanghai) were treated with chemotherapy drugs.\u0026nbsp;The mice were fed with\u0026nbsp;a\u0026nbsp;standard chow diet ad libitum and housed in a clean grade room at 21 ±\u0026nbsp;1 °C\u0026nbsp;and 60 ± 5% humidity under a 12 h light/12 h dark cycle with free access to water and food in a specific pathogen-free environment.\u0026nbsp;MC38-Luc cells (2×10\u003csup\u003e6\u003c/sup\u003e) were implanted into immunocompetent C57BL/6\u0026nbsp;mice, C57BL/6 mice which previously irradiated and then transplanted with green fluorescent protein\u003csup\u003e+\u003c/sup\u003e (GFP\u003csup\u003e+\u003c/sup\u003e)-bone marrow cells\u0026nbsp;[24],\u0026nbsp;or\u0026nbsp;STAT3\u003csup\u003eflox/flox\u003c/sup\u003e and STAT3\u003csup\u003eflox/flox\u003c/sup\u003e Tek-cre\u0026nbsp;C57BL/6 mice (Shanghai Model Organisms, Chania) by tail vein injection. CT26-Luc cells and Lewis cells (2×10\u003csup\u003e6\u003c/sup\u003e) were tail vein injected into BALB/c mice. B16F1 melanoma cells (2×10\u003csup\u003e6\u003c/sup\u003e) were tail vein injected into C57BL/6 mice. In all in vivo studies mice were randomly grouped (n=4–6/group).\u0026nbsp;Tumor lung metastasis progression was monitored\u0026nbsp;using an Xenogen IVIS system (PerkinElmer IVIS\u0026nbsp;Lumina\u0026nbsp;Ⅲ-Living imaging). Mice were anesthetized using 2.5% isoflurane and administered D-luciferin (150 mg/kg; Yeasen) via intraperitoneal (i.p.) injection 10 min before the imaging. All protocols were approved by the Institutional Animal Care and Use Committee of Putuo Hospital, Shanghai University of Traditional Chinese Medicine, P. R. China. All animal studies were conducted in accordance with the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFlow cytometry analysis of EPCs and MDSCs\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBM cells were stained with BV421-conjugated anti-CD34 (562208, BD) and anti-CD133-PE (130–117–784, MACS), anti-VEGFR2-APC antibodies (560070, BD) and anti-CD133-PE to detect EPCs, anti-CD11b-FITC (557396, BD) and anti-Ly-6G-BV421\u0026nbsp;(562737, BD) to detect MDSCs. Blood samples were collected using EDTA tubes. Red blood cells were lysed, and\u0026nbsp;EPCs were studied. Labeled cells were detected using a flow cytometer (C6, BD) and\u0026nbsp;analyzed\u0026nbsp;by CellQuest and FlowJo\u0026nbsp;software.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eELISA\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eLevels of mouse or human G-CSF\u0026nbsp;(H-GCSF,\u0026nbsp;EK0360,\u0026nbsp;Boster;\u0026nbsp;M-GCSF,\u0026nbsp;EK0361,\u0026nbsp;Boster), VEGF (H-VEGF,\u0026nbsp;EK0539,\u0026nbsp;Boster;\u0026nbsp;M-VEGF,\u0026nbsp;EK0541,\u0026nbsp;Boster) and SDF-1α (H-SDF 1A,\u0026nbsp;SEKH-0310,\u0026nbsp;Solarbio) were assessed by ELISA (Biolegend)\u0026nbsp;according to the manufacturers’ instructions.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCytokine antibody array\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSerum samples\u0026nbsp;were collected and frozen at -80°C\u0026nbsp;for subsequent cytokine analysis using the\u0026nbsp;R\u0026amp;D\u0026nbsp;Mouse Cytokine Array Panel A (ARY006) according to the manufacturer’s instructions.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTube formation assay.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHUVECs were treated with G-CSF for 1 hour before tube formation. HUVECs (3*10\u003csup\u003e4\u003c/sup\u003e) were seeded into a 96-well plate coated with Matrigel (BD, #356234, USA). Tube formation was observed at 4 h. Three random fields were acquired by a microscope (Leica, DMI3000B, Germany). The number of Nodes and\u0026nbsp;Branches\u0026nbsp;points were quantified by\u0026nbsp;ImageJ (Nodes/Branches: defined as individual junctions/branching points).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCell migration assay\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn the cell migration assay, HUVECs (3*10\u003csup\u003e4\u003c/sup\u003e) in serum-free medium\u0026nbsp;were\u0026nbsp;seeded\u0026nbsp;in\u0026nbsp;a small chamber (3422, Corning) inserted into\u0026nbsp;a\u0026nbsp;24-well plate. 10% FBS-containing medium\u0026nbsp;was\u0026nbsp;added to the plate located below\u0026nbsp;the\u0026nbsp;chamber to serve as chemical attractant. After 24 h, the migrating cells located on the lower surface of the chamber were stained with 0.1% crystal violet and counted.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAdhesion assay\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHUVECs were pretreated with G-CSF. Then,\u0026nbsp;HCT116-GFP cells\u0026nbsp;were incubated with\u0026nbsp;the treated\u0026nbsp;HUVECs\u0026nbsp;for 2 hours.\u0026nbsp;The cells were washed\u0026nbsp;with PBS and photographed under a fluorescence microscope.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eQuantitative RT–PCR\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTotal RNA was extracted from cultured HUVECs with TRIzol (Invitrogen). The concentration of total RNA was quantified by measuring the absorbance at 260 nm. FAST qPCR with\u0026nbsp;a\u0026nbsp;TAKARA SYBR kit (RR420A, TAKARA) was used to perform qRT–PCR,\u0026nbsp;and an Applied Biosystems PCR\u0026nbsp;System\u0026nbsp;(Fisher Scientific) was used to collect and analyze the data.\u0026nbsp;The\u0026nbsp;2\u003csup\u003e−ΔΔCt\u003c/sup\u003e method\u0026nbsp;was used\u0026nbsp;to determine the relative expression level of each gene.\u003c/p\u003e\n\u003cp\u003eThe primer sequences were as follows:\u003c/p\u003e\n\u003cp\u003eVEGF, F 5’-TTGCTGCTCTACCTCCAC-3’ and R 5’-AATGCTTTCTCCGCTCTG-3’;\u003c/p\u003e\n\u003cp\u003eE-selectin, F 5’-ATGTTCAAGCCTGGCAGTTCCG-3’ and R 5’-GCAGAGCCATTGAGCGTCCATC-3’;\u003c/p\u003e\n\u003cp\u003eP-selectin, F 5’-CGCTCTGGACCAACCCTGTTTC-3’ and R 5’-CTCCTGGCTTCTGTGGCTTGTG-3’;\u003c/p\u003e\n\u003cp\u003eICAM, F 5’-TGCAAGAAGATAGCCAACCAAT-3’ and R 5’-GTACACGGTGAGGAAGGTTTTA-3’;\u003c/p\u003e\n\u003cp\u003eVCAM, 5’-F CAGGCTGGAGATAGACTTACTG-3’ and R 5’-CCTCAATGACAGGAGTAAAGGT-3’;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eWestern Blot Analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCells were lysed with\u0026nbsp;cell\u0026nbsp;lysis solution (9803s, CST) containing\u0026nbsp;protease\u0026nbsp;(B14001, Bimake) and\u0026nbsp;phosphatase inhibitors\u0026nbsp;(B15001, Bimake),\u0026nbsp;and\u0026nbsp;a\u0026nbsp;BCA\u0026nbsp;protein assay\u0026nbsp;kit (P0011, Beyotime) was used for protein quantification. Equivalent amounts of cell lysates were separated by 10% SDS–PAGE,\u0026nbsp;and the fractionated proteins were transferred to PVDF membranes (cat. no. IPVH00010, Millipore). After blocking with 5% skim milk, the membranes were incubated with primary antibodies\u0026nbsp;against\u0026nbsp;p-STAT3 (9145S, CST), STAT3\u0026nbsp;(9139S, CST), SELE (BBA16, R\u0026amp;D), SELP (sc-19672, Santa Cruz), ICAM (ab2213, Abcam) and CD31 (ab28364, Abcam). Then, the cells were\u0026nbsp;incubated with a goat anti-rabbit or anti-mouse\u0026nbsp;IgG-HRP secondary antibody (R, ab6721; M, ab6789; Abcam). Actin was used as the loading control (ab6276，Abcam). Protein expression was assessed with enhanced chemiluminescent substrate (WBKLS0500, Millipore, USA).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eHistopathological assay\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe tissue was fixed in 10% formalin, embedded in paraffin, and then sectioned (5 mm thick). The histopathology assay procedure\u0026nbsp;was\u0026nbsp;completed by conventional hematoxylin-eosin (H\u0026amp;E) staining according to standard techniques.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eImmunofluorescence\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHUVECs were fixed with methanol for 15 minutes and blocked with 5% bovine serum albumin (BSA) for 30 min at room temperature. The cells were incubated overnight at 4°C with primary antibodies (against p-STAT3, 9145S, CST; STAT3, 9139S, CST; SELE, BBA16, R\u0026amp;D; SELP, sc-19672, Santa Cruz; ICAM, ab2213, Abcam) and then incubated with secondary antibodies for 1h at room temperature. Each step was followed by washing 3 times for 5 minutes in PBS. The prepared specimens were counterstained with DAPI (Beyotime, Shanghai, China) for 5 min and\u0026nbsp;photographed\u0026nbsp;with\u0026nbsp;a\u0026nbsp;Zeiss LSM880 confocal\u0026nbsp;microscope.\u003c/p\u003e\n\u003cp\u003eThe tissue specimens were\u0026nbsp;fixed with cold methanol for 10 minutes and blocked with 5% BSA for\u0026nbsp;1h. Primary antibodies (Extended\u0026nbsp;Data Table 1) were applied in blocking buffer,\u0026nbsp;incubated overnight at 4 °C, and then the cells were incubated with Alexa Fluor® 488 (ab150113, Abcam) and Alexa Fluor® 647-labeled (ab150075, Abcam) secondary antibodies for\u0026nbsp;2h\u0026nbsp;at room temperature. Nuclei were stained with DAPI. Images were taken by a Zeiss LSM880 confocal microscopy.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eImmunohistochemistry (IHC)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe tissue was fixed in 10% formalin, embedded in paraffin, and then sectioned (5 mm thick).\u0026nbsp;Paraffin sections were deparaffinized in xylene and rehydrated in a series of graded alcohols. The antigen was retrieved in 0.01 M sodium citrate buffer. The sections were incubated with 3% H\u003csub\u003e2\u003c/sub\u003eO\u003csub\u003e2\u003c/sub\u003e for 10 min and then blocked in 5% BSA for 30 minutes at room temperature. The sections were incubated overnight at 4 °C with CD31 (77699, CST), CD4 (ab183685, abcam) and CD8 (ab4055, abcam) primary\u0026nbsp;antibodies. The\u0026nbsp;appropriate secondary antibody\u0026nbsp;was used\u0026nbsp;for 1 h at room temperature.\u0026nbsp;An\u0026nbsp;EnVision (K4007, Dako) signal enhancement system was used to develop the bound antibodies. The sections were stained with Harris hematoxylin, dehydrated and fixed. Images were captured by a microscope (Leica, Wetzlar, Germany). Images were processed using ImageJ to quantify CD31-positive areas.\u003c/p\u003e\n\u003cp\u003eCD31-positive areas.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePatient samples.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe serum and tumor samples were collected from 39 patients diagnosed with colorectal cancer\u0026nbsp;who received\u0026nbsp;neoadjuvant chemotherapy followed by surgical resection at Putuo Hospital, Shanghai University of Traditional Chinese Medicine, China, from January 2015 to December 2016. Written informed consent was obtained from the patients, in accordance with the institutional guidelines, before sample collection, and the study was approved by the Committees for the Ethical Review of Research at the Putuo Hospital, Shanghai University of Traditional Chinese Medicine, China. Informed consent for the use of\u0026nbsp;samples\u0026nbsp;for research purposes was obtained from the participants.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistical Analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eStatistical analyses were performed using GraphPad Prism 8 and SPSS. For all experiments, three technical replicates were analyzed by counting a minimum of 3 fields/ technical replicate. Error bars in the experiments indicate standard deviation (s.d.) The legends illustrate the number of events and independent experiments, as well as the information related to the statistical details and methods. Statistical significance calculated using one-way ANOVA with Dunnett’s test. P values of clinical data were determined by two-tailed Pearson correlation coefficient test. P values \u0026lt;0.05 were considered statistically significant.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eChemotherapy could\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003epromote\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;metastasis by\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eaffecting\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003evascular\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eproperties\u003c/strong\u003e\u003cstrong\u003e.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAccumulating evidence suggested that chemotherapy could promote series of\u0026nbsp;changes\u0026nbsp;that facilitate metastasis [1,2]. Here we used a lung metastasis model to evaluate the effect of chemotherapeutic\u0026nbsp;drugs\u0026nbsp;commonly used to treat CRC patients. As schematized in\u0026nbsp;Extended Data\u0026nbsp;Fig. 1A, we injected CT-26-Luc cancer cells into BALB/c mice\u0026nbsp;via\u0026nbsp;the tail vein, followed by saline control, CPT11, OXA or 5-FU treatment and analysis of metastatic lesions. Mice received OXA had most\u0026nbsp;metastatic\u0026nbsp;sites in the lung\u0026nbsp;(Extended data Fig. 1B, C). We found microvascular density increased dramatically in the metastatic lesions.\u0026nbsp;In addition, STAT3 was markedly activated in the\u0026nbsp;vessels after treatment\u0026nbsp;with OXA MTD (maximal tolerating dose) (Extended data Fig. 1D). These data indicated that certain chemotherapeutic agent could\u0026nbsp;activate\u0026nbsp;STAT3 in vascular endothelial cells, which is consistent with its role in promoting angiogenesis,\u0026nbsp;a critical step to facilitate metastasis.\u003c/p\u003e\n\u003cp\u003eTo study\u0026nbsp;the\u0026nbsp;role\u0026nbsp;of\u0026nbsp;endothelial cells in lung metastasis, we generated mice with conditionally inactivated STAT3 in endothelial cells (STAT3\u003csup\u003eflox/flox\u003c/sup\u003e Tek-cre). Compared to wild-type mice (STAT3\u003csup\u003eflox/flox\u003c/sup\u003e) (Fig. 1A), the weight of the mice in all groups was not\u0026nbsp;affected\u0026nbsp;by chemotherapy (Fig. 1B). As shown in Fig. 1C and 1D, STAT3\u003csup\u003eflox/flox\u003c/sup\u003e Tek-cre mice\u0026nbsp;treated\u0026nbsp;with chemotherapy had smaller metastatic\u0026nbsp;areas\u0026nbsp;than\u0026nbsp;their\u0026nbsp;KO counterparts (KO-OXA). Although deletion of STAT3 in ECs did not affect\u0026nbsp;the\u0026nbsp;formation of blood vessels in the\u0026nbsp;metastatic\u0026nbsp;lesions (Fig. 1E, 1F), it did\u0026nbsp;reduce\u0026nbsp;the vascular adhesion\u0026nbsp;molecule\u0026nbsp;Selectin E (SELE) (Fig. 1G).\u0026nbsp;These data suggested that STAT3 KO in ECs reduced OXA-induced metastasis by\u0026nbsp;affecting\u0026nbsp;vascular adhesion\u0026nbsp;properties, not necessarily angiogenesis. We therefore need\u0026nbsp;to\u0026nbsp;explore alternative mechanism of new vessel formation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eOxaliplatin\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003epromoted\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;metastasis by\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eincreasing\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;immunosuppression.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMyeloid cells have also been implicated in contributing to angiogenesis\u0026nbsp;and\u0026nbsp;to the generation of an immunosuppressive tumor microenvironment [25]. Consistently, we also\u0026nbsp;found an enrichment of myeloid-derived suppressor cells (MDSCs, which are known to be very potent suppressors of cytotoxic T-cell immunity) and scarce CD4\u003csup\u003e+\u003c/sup\u003e and CD8\u003csup\u003e+\u003c/sup\u003e T cells in the metastatic lesions after chemotherapy.\u0026nbsp;Interestingly,\u0026nbsp;STAT3\u0026nbsp;KO\u0026nbsp;did not\u0026nbsp;influence the infiltration of\u0026nbsp;MDSCs or\u0026nbsp;CD4\u003csup\u003e+\u003c/sup\u003e and CD8\u003csup\u003e+\u003c/sup\u003e T cells\u0026nbsp;compared\u0026nbsp;to\u0026nbsp;that of\u0026nbsp;the WT group\u0026nbsp;regardless whether or not\u0026nbsp;the mice received OXA chemotherapy (Extended data Fig. 2). These data indicate that although OXA\u0026nbsp;promoted an\u0026nbsp;immunosuppressive phenotype, it was not through the activation of STAT3 in ECs.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eOxaliplatin\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;promoted\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;metastasis by\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eaffecting\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003evasculogenesis.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eVasculogenesis, a key paradigm of neovascularization, refers to the de novo formation of blood vessels derived from stem/progenitor cells\u0026nbsp;[7].\u0026nbsp;Several studies\u0026nbsp;suggested that\u0026nbsp;EPC\u0026nbsp;recruitment could contribute to tumor neovascularization and targeting EPCs might be an attractive\u0026nbsp;antiangiogenic\u0026nbsp;treatment [8].\u0026nbsp;We\u0026nbsp;further\u0026nbsp;explored\u0026nbsp;the reason\u0026nbsp;for\u0026nbsp;the upregulation of vessel density by chemotherapy. We focused on EPCs.\u0026nbsp;As\u0026nbsp;expected,\u0026nbsp;EPC\u0026nbsp;markers (CD34\u003csup\u003e+\u003c/sup\u003e and VEGFR2\u003csup\u003e+\u003c/sup\u003e) showed higher expression in the OXA group than in the Ctrl group (Fig. 1 I, J, K).\u003c/p\u003e\n\u003cp\u003eTo track\u0026nbsp;EPC\u0026nbsp;homing and retention in tumor metastasis sites, experiments were performed using GFP\u003csup\u003e+\u003c/sup\u003e bone marrow cells obtained from C57BL/UBI/GFP mice\u0026nbsp;that\u0026nbsp;were transplanted into C57BL/6 mice that had been lethally irradiated (Fig. 2A). Four weeks later, the mice were used as recipients for an injection of OXA followed three days later\u0026nbsp;by a\u0026nbsp;tail vein injection\u0026nbsp;of\u0026nbsp;MC-38-Luc cells.\u0026nbsp;Fourteen\u0026nbsp;days later, their lungs were obtained for the evaluation of GFP+ bone marrow cell colonization and incorporation into the tumor vessels, as shown in Fig. 2B. We found\u0026nbsp;an increase in\u0026nbsp;bone marrow-derived GFP\u003csup\u003e+\u003c/sup\u003e cells in the lung metastasis sites after treatment with OXA. CD31\u003csup\u003e+\u003c/sup\u003e and GFP\u003csup\u003e+\u003c/sup\u003e cells represented\u0026nbsp;EPC-modulated\u0026nbsp;vasculogenesis. The colonization of EPCs in the metastasis site was higher in the OXA group than\u0026nbsp;in\u0026nbsp;the control\u0026nbsp;group\u0026nbsp;(Fig. 2C).\u003c/p\u003e\n\u003cp\u003eTo study whether chemotherapy can induce\u0026nbsp;the\u0026nbsp;mobilization of EPCs from bone marrow to plasma, we first monitored\u0026nbsp;the\u0026nbsp;levels of EPCs for up to 72 hours after chemotherapy drug injection in C57BL/6 mice (Fig. 2D) or BALB/c mice (Extended Fig. 3A). The proportion of\u0026nbsp;EPC\u0026nbsp;from bone marrow decreased with time, and proportion from the plasma increased (Fig. 2E; Extended Fig. 3B, 3C).\u0026nbsp;Next, we\u0026nbsp;tested\u0026nbsp;the plasma\u0026nbsp;contents\u0026nbsp;of stromal cell-derived factor-1\u0026alpha; (SDF-1\u0026alpha;) and VEGF,\u0026nbsp;which\u0026nbsp;are\u0026nbsp;secreted by EPCs,\u0026nbsp;and found that their level\u0026nbsp;increased along with the mobilization of\u0026nbsp;EPCs\u0026nbsp;to the plasma\u0026nbsp;(Fig. 2F, G). Myeloid cells have also been implicated in contributing to an angiogenic-to-vasculogenic switch\u0026nbsp;[26]. Subsequently, the FACS results showed that\u0026nbsp;OXA chemotherapy significantly promoted the mobilization of MDSCs (Fig. 2H; Extended Fig. 3D). These data indicated that chemotherapy promoted\u0026nbsp;the\u0026nbsp;formation of blood\u0026nbsp;vessels\u0026nbsp;by\u0026nbsp;mobilizing\u0026nbsp;EPCs. In addition, it mobilized MDSCs to contribute to tumor vasculogenesis and immunosuppression.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eChemotherapy promoted colorectal cancer metastasis through vessel formation by upregulating G-CSF.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCytokines are extracellular\u0026nbsp;signaling\u0026nbsp;molecules that may facilitate the mobilization of EPCs and/or affect vascular properties. To further assess this possibility, we obtained plasma samples from C57/Bl6 mice treated with OXA. We found multiple cytokines including G-CSF, IP-10, KC and JE were upregulated following OXA treatment. G-CSF, which\u0026nbsp;is\u0026nbsp;known\u0026nbsp;to promote the\u0026nbsp;mobilization of stem/progenitor cells, had highest fold of change based on immunoblot (Fig. 3A). This was further confirmed by ELISA (Fig. 3B). Although KC, JE, and IP-10 have previously been reported to be associated with angiogenesis, there are very few reports about the effects of KC, JE, and IP-10 on\u0026nbsp;EPC\u0026nbsp;mobilization. Next, we found\u0026nbsp;that\u0026nbsp;G-CSF\u0026nbsp;was upregulated to a\u0026nbsp;high level at\u0026nbsp;2\u0026nbsp;h\u0026nbsp;following chemotherapy and this decreased over the next 6 hours\u0026nbsp;(Fig. 3C). To further\u0026nbsp;evaluate the proangiogenic\u0026nbsp;roles of G-CSF, we performed G-CSF\u0026nbsp;on\u0026nbsp;HUVECs.\u0026nbsp;G-CSF promoted\u0026nbsp;HUVEC\u0026nbsp;tube formation (Fig. 3D) and adhesion (Fig. 3E)\u0026nbsp;along with\u0026nbsp;STAT3 phosphorylation (Fig. 4F, 4I). However,\u0026nbsp;it\u0026nbsp;did not\u0026nbsp;affect the proliferation or migration of HUVECs (Extended Fig. 4A, 4B). In addition, we observed increased level of cell adhesion molecules including SELE, SELP and ICAM by G-CSF\u0026nbsp;(Fig. 4G, 4H, 4J-L). Collectively, these data suggest that G-CSF could activate p-STAT3 and promote vascular adhesion\u0026nbsp;following OXA treatment.\u003c/p\u003e\n\u003cp\u003eTo examine the effect of G-CSF on the vessel properties of\u0026nbsp;EPCs, we performed G-CSF\u0026nbsp;on\u0026nbsp;mouse primary EPCs. Our results showed that\u0026nbsp;G-CSF promoted\u0026nbsp;EPCs\u0026nbsp;tube formation (Fig. 3M) and\u0026nbsp;migration (Fig. N-O). In summary, G-CSF affected not only vascular adhesion but also vasculogenesis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAnti-G-CSF reduced chemotherapy-induced lung metastasis by\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003einhibiting\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;blood vessel formation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBased on the above experimental results, we questioned whether anti-G-CSF antibodies could be used to prevent and/or inhibit metastasis induced by chemotherapy. To assess this, C57BL/6 mice were treated with OXA\u0026nbsp;and\u0026nbsp;then injected\u0026nbsp;with\u0026nbsp;anti-G-CSF two hours later. After three days, we injected MC-38-Luc cancer cells\u0026nbsp;into\u0026nbsp;their tail vein (Fig. 4A). Similarly,\u0026nbsp;the\u0026nbsp;CT26 BALB/c metastasis model\u0026nbsp;was used and is shown in\u0026nbsp;Extended\u0026nbsp;Data\u0026nbsp;Fig. 5A. As expected, anti-G-CSF\u0026nbsp;reduced\u0026nbsp;lung metastasis in the mice (Fig. 4B-D; Extended data Fig. 5B-D). In addition, the lung tissue vessel density was\u0026nbsp;significantly\u0026nbsp;decreased\u0026nbsp;(Fig. 4E; Extended data Fig. 5E), and the activation of STAT3 in ECs was markedly reduced by anti-G-CSF (Fig. 4F; Extended data Fig. 5F). Anti-G-CSF\u0026nbsp;also\u0026nbsp;reduced SELE expression in the\u0026nbsp;vasculature\u0026nbsp;(Fig. 4G; Extended data Fig. 5G).\u003c/p\u003e\n\u003cp\u003eWe further investigated whether cells from bone marrow homing to metastasis sites following chemotherapy could be prevented by anti-G-CSF. A decrease\u0026nbsp;of\u0026nbsp;MDSCs was indeed found in the metastasis sites (Fig. 4H; Extended data Fig. 5H), and CD4\u003csup\u003e+\u003c/sup\u003e and CD8\u003csup\u003e+\u003c/sup\u003e T cells (Extended data Fig. 6A, 6B)\u0026nbsp;were\u0026nbsp;increased. Next, we found\u0026nbsp;that\u0026nbsp;the expression of\u0026nbsp;the EPC markers\u0026nbsp;CD34 and VEGFR2 in\u0026nbsp;metastatic tumors was\u0026nbsp;lower in the anti-G-CSF group (Fig. 4I-J; Extended data Fig. 5I-J). To track\u0026nbsp;EPC\u0026nbsp;homing and retention in tumor metastasis sites, GFP\u003csup\u003e+\u003c/sup\u003e trans\u0026nbsp;mice\u0026nbsp;were used as recipients for an injection of OXA\u0026nbsp;and\u0026nbsp;then injected\u0026nbsp;with\u0026nbsp;anti-G-CSF two hours later, as shown in Fig. 5A. The EPCs from the bone marrow (CD31\u003csup\u003e+\u003c/sup\u003eGFP\u003csup\u003e+\u003c/sup\u003e cells) were reduced by anti-G-CSF (Fig. 5B).\u003c/p\u003e\n\u003cp\u003eTo study the impact of anti-G-CSF on the chemotherapy-induced mobilization of EPCs and MDSCs, C57BL/6 mice were treated with OXA\u0026nbsp;and\u0026nbsp;then injected\u0026nbsp;with\u0026nbsp;anti-G-CSF two hours later. As\u0026nbsp;shown\u0026nbsp;in Fig. 5C, bone marrow and plasma\u0026nbsp;were collected and measured\u0026nbsp;72 hours after chemotherapy. Anti-G-CSF abolished the\u0026nbsp;upregulated\u0026nbsp;mobilization of EPCs from bone marrow to plasma (Fig. 5D, 5E) induced by chemotherapy. Anti-G-CSF\u0026nbsp;also\u0026nbsp;significantly reduced the mobilization of MDSCs (Fig. 5F). Similarly, the expression of SDF-1\u0026alpha; and VEGF\u0026nbsp;was\u0026nbsp;reversed by anti-G-CSF. (Fig. 5G, H). These data indicated that anti-G-CSF reduced oxaliplatin-induced lung metastasis by\u0026nbsp;inhibiting\u0026nbsp;vasculogenesis and immunosuppression.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eG-CSF is positively associated with disease recurrence and metastasis in CRC patients\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo evaluate the contribution of G-CSF in clinical setting, we examined G-CSF level in plasma obtained from CRC patients who underwent surgical resection followed by adjuvant XELOX chemotherapy regimen containing OXA. We collected patients\u0026rsquo; plasma before and after 1 course of XELOX regimen. First, we correlated G-CSF before surgical resection to recurrence and metastasis. Survival analysis indicated that patients with high G-CSF level in their plasma had significantly shorter overall survival and recurrence-free survival (Fig. 6A-B). In addition, higher G-CSF correlated with more advanced nodal and remote metastasis (Fig. 6C-D). Analyses of blood vessels (CD31; Figure 6E) and EPC-produced cytokines (SDF-1\u0026alpha; and VEGF; Fig. 6F) also demonstrated that G-CSF level positively correlated with increased tumor vasculature.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eNext, we correlated G-CSF level of post one cycle of XELOX with CRC recurrence and metastasis. In\u0026nbsp;consistent with the in vivo experiments, G-CSF level increased in majority of patients after receiving XELOX (Fig. 6G). More importantly, increased G-CSF level after chemotherapy was closely associated with poorer overall survival (Fig. 6H) and recurrence-free survival (Fig. 6I). Similarly, SDF-1\u0026alpha; and VEGF levels increased significantly after XELOX treatment (Fig. 6 J-I).\u0026nbsp;Of note,\u0026nbsp;SDF-1\u0026alpha; and VEGF levels were significantly higher in patients with their G-CSF increased after chemotherapy (Fig. 6L), and more than 80% of those patients had increased SDF1 and VEGF (Fig. 6K). SDF-1\u0026alpha; and VEGF showed a significant positive correlation with the production of G-CSF, and the correlation coefficient increased after chemotherapy (Fig. 6M-N). These results suggested that\u0026nbsp;the number of EPCs in the plasma positively correlated with G-CSF level following chemotherapy.\u003c/p\u003e\n\u003cp\u003eTaken together, these data indicate that elevated G-CSF in the serum following chemotherapy likely enhance blood vessel formation and could be used as a prognostic biomarker to predict a poor outcome. Tracking G-CSF level after chemotherapy and optimizing appropriate intervention will have clinical value to benefit patients from chemotherapy while simultaneously avoid increased risk of metastasis.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eChemotherapy remains the mainstay treatment for locally advanced/metastatic cancers. In addition, it has proven clinical value in various neoadjuvant and adjuvant settings. However, recent studies have also demonstrated that chemotherapy can induce local and systemic pro-tumorigenic/metastatic changes, raising the concern of increased risk of metastasis facilitated by certain chemotherapy regimens [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. Identifying the underlying mechanisms and validating such observation in the clinical setting are therefore critically important to uncouple the therapeutic benefit and undesirable risk of chemotherapy for our cancer patients.\u003c/p\u003e \u003cp\u003eIn this study, we have illustrated that at least certain chemotherapy drug such as OXA could increase the level of circulating G-CSF, resulting in enhanced lung metastasis of colorectal cancer (as well as lung cancer and melanoma, Extended data Fig.\u0026nbsp;7). Such elevated level of G-CSF enhanced vascular adhesion via G-CSF/STAT3 signaling, and augmented EPC mobilization to induce vasculogenesis, both are critical to directly facilitate metastasis. In addition, OXA facilitated the mobilization of MDSCs from bone marrow to promote immunosuppressive phenotype. We also showed anti-G-CSF could reverse above-mentioned process and successfully block metastasis promoted by OXA chemotherapy. More importantly, in our CRC patients, the baseline and post-chemotherapy G-CSF levels positively correlated to worse clinical outcome.\u003c/p\u003e \u003cp\u003eThe formation of new blood vessels is an essential component of malignant tumor development and progression [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. It has become increasingly evident that this process utilizes various mechanisms, including angiogenesis [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e], vasculogenesis [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e], vasculogenic mimicry [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e] and vessel co-option [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e], etc. We found that chemotherapy drug OXA activated STAT3 in a G-CSF dependent manner. STAT3 is a transcription factor that regulates a variety of cellular events. Many studies have suggested that the activation of STAT3 promotes tumor angiogenesis [\u003cspan additionalcitationids=\"CR35\" citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. STAT3 knockout in ECs mitigated the ability of chemotherapy to exacerbate metastasis. However, deletion of STAT3 in ECs did not affect the formation of blood vessels in metastatic tumors or affect vascular adhesion molecules. G-CSF promoted HUVEC tube formation and adhesion but not proliferation and migration by activating STAT3. These results indicated that STAT3 KD in ECs reduced OXA-induced metastasis by affecting vascular adhesion properties, not angiogenesis. Chemotherapy-induced sprouting vessels are dependent on multiple mechanisms that work together. We need to further explore the mechanism of new vessel formation.\u003c/p\u003e \u003cp\u003eVasculogenesis refers to the de novo formation of blood vessels derived from stem/progenitor cells [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. Yuval Shaked et al. [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e] showed that paclitaxel can rapidly induce bone marrow-derived EPC mobilization and subsequent tumor homing, whereas gemcitabine does not. To unearth the mechanism of new vessel formation induced oxaliplatin chemotherapy, we focused on EPC mobilization. Our results showed that EPC mobilization to vasculogenesis induced by OXA is critical for lung metastasis. In addition, EPCs produce a variety of proangiogenic cytokines and growth factors, such as SDF-1α and VEGF. SDF-1 is a key chemokine that regulates the transport of hematopoietic stem cells between the bone marrow and peripheral circulation and it attracts EPCs to ischemic areas. VEGF promotes the proliferation and migration of pre-existing ECs, contributing to angiogenesis [\u003cspan additionalcitationids=\"CR39\" citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e]. In this work, we found that the expression of SDF-1α and VEGF was upregulated by OXA chemotherapy. Circulating EPCs derived from bone marrow (BM) might differentiate into mature endothelial cells [\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]. We found that G-CSF could promote primary EPC differentiation into mature endothelial cells.\u003c/p\u003e \u003cp\u003eMDSCs are a type of immunosuppressive cell population, and their response to tumor antigens presented as peptides on their surface contributes to antitumor escape. They may negatively affect immune responses by regulating the function and proliferation of CD4\u0026thinsp;+\u0026thinsp;and CD8\u0026thinsp;+\u0026thinsp;cells [\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e]. Myeloid cells have also been implicated in contributing to an angiogenic-to-vasculogenic switch [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. We found that MDSCs rapidly mobilized from the bone marrow in response to OXA chemotherapy and that the infiltration of MDSCs into the lung tissue was improved. STAT3 KO affected vascular adhesion properties but did not affect angiogenesis or the infiltration of MDSCs. These data suggest that OXA-induced MDSC mobilization also contributed to metastasis by immunosuppression and vasculogenesis.\u003c/p\u003e \u003cp\u003eBone marrow-derived stem/progenitor cells could be strongly mobilized into circulation by G-CSF [\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. Exogenous G-CSF administration could exacerbate solid tumor metastasis [\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e]. In our work, we found that G-CSF was upregulated by OXA chemotherapy. We used anti-G-CSF to abolish the mobilization of EPCs from bone marrow to plasma and lung metastases triggered by chemotherapy. The lung tissue vessel density was significantly decreased. Similarly, the expression of SDF-1α and VEGF was reversed by anti-G-CSF. Anti-G-CSF also changed the vascular adhesion ability by reducing SELE expression in the vasculature via STAT3. In addition, anti-G-CSF may repress tumoral immunosuppression to trigger antitumor immune responses by preventing MDSC mobilization. These data demonstrated that anti-G-CSF could be used as an adjuvant treatment for metastasis caused by chemotherapy.\u003c/p\u003e \u003cp\u003eIn the clinical study, we found that G-CSF was significantly upregulated by the XELOX regimen. Survival analysis indicated that patients with upregulated G-CSF in their serum had significantly worse survival outcomes and higher recurrence and metastasis rates. We also investigated the expression of endogenous G-CSF in patients with colon cancer. We found that patients with high expression of G-CSF in their serum had significantly worse survival outcomes and higher recurrence and metastasis rates. Patients in the high G-CSF expression group had more blood vessels in their tumor tissue. Consistent with our findings, in human tumors, there are many reports of extremely aggressive G-CSF in multiple tumor types, demonstrating the important and modulatory effect of G-CSF in tumor progression and metastasis [\u003cspan additionalcitationids=\"CR48\" citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eAs G-CSF is commonly used to prevent/treat serious neutropenia during cancer chemotherapy [\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e], our study did provide mechanistic evidence to caution the use of G-CSF and balance its benefits and potential risks. This brings important questions regarding the timing, duration and circumstances under which we shall consider using G-CSF. Although an absolute neutrophil count equal or less than 500/mm\u003csup\u003e3\u003c/sup\u003e is generally considered an indication to use G-CSF in clinical practice, more studies are needed to optimize the duration of use, along with dose reduction of chemotherapy as the later could also up-regulate G-CSF based on our current study. Interestingly, Peishan Li et al. [\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e] also found that administration of exogenous G-CSF promoted metastasis in host mice with intact NK cells but achieved opposite antimetastatic effect in those mice lacking functional NK cells, suggesting host immune status may also need to be considered for a personalized treatment decision.\u003c/p\u003e \u003cp\u003eIn summary, our study has illustrated that at least certain chemotherapeutic agent such as oxaliplatin could facilitate cancer metastasis via G-CSF induced neovascularization. This cautions routine use of G-CSF during cancer chemotherapy and quests optimal chemotherapy regimen and use of anti-G-CSF therapy that warrants further exploration.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eEPCs\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eEndothelial Progenitor Cells\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eBM\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003ebone marrow\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003ePB\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eperipheral blood\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eG-CSF\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eGranulocyte-colony stimulating factor\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eG-CSFR\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eG-CSF receptor\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eHUVECs\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eHuman umbilical vein endothelial cells\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eH\u0026amp;E\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003ehematoxylin-eosin\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eBSA\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003ebovine serum albumin\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eMDSCs\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003emyeloid-derived suppressor cells\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eSELE\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eSelectin E\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eOXA\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eOxaliplatin.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets during and/or analyzed during the current study are available from the corresponding authors upon reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no conflict of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis project was sponsored by the Science and Technology Innovation Project of Putuo District Health System (ptkwws201905, ptkwws202004),\u0026nbsp;the Natural Science Foundation of Shanghai (20ZR1450500), Shanghai Key Medical Specialty Construction Project (NO. ZK2019B18),\u0026nbsp;Clinical Specialized Disease Construction Project of Shanghai Putuo District Municipal Health Comission (NO. 2019tszb01), the National Key Research and Development Program of China (grant number 2019YFC1316000), as well as the\u0026nbsp;Scientific Research Project of Putuo Central Hospital (2020362A).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eK.X., X.L., and P.Y., conceived and directed the project. K.X. designed the experiments. Y.Z., K.F., Y.W., M.L., R.W., J.X., and Z.Y. carried out the experiments.\u0026nbsp;J.W., C.W., R.Z. J.X., T.C., and J.W. were responsible for collecting tissue specimen. K.X., Y.Z., and K.F. conducted the data analysis and interpreted the results. Y.Z., K.X., JZ, K.F., Y.W., and P.Y. wrote, criticized and edited the paper. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSpecial thanks to Dr. Teng Chen and Dr. Jianhua Xu for the help during the study.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eLiu Y, Cao X. Organotropic metastasis: role of tumor exosomes. Cell Res 2016;26:149\u0026ndash;50\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDoglioni G, Parik S, Fendt SM. Interactions in the (Pre)metastatic Niche Support Metastasis Formation. Front Oncol 2019;9:219\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang JM, Yang GY, Medina DJ, Vassil AD, Liao J, Hait WN. Treatment of multidrug resistant (MDR1) murine leukemia with P-glycoprotein substrates accelerates the course of the disease. Biochem Biophys Res Commun 1999;266:167\u0026ndash;73\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eByrd-Leifer CA, Block EF, Takeda K, Akira S, Ding A. The role of MyD88 and TLR4 in the LPS-mimetic activity of Taxol. Eur J Immunol 2001; 31:2448\u0026ndash;57\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKaplan RN, Rafii S, Lyden D. Preparing the \"soil\": the premetastatic niche. Cancer Res 2006;66:11089\u0026ndash;93\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShaked Y. The pro-tumorigenic host response to cancer therapies. Nat Rev Cancer 2019;19:667\u0026ndash;85\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ede la Puente P, Muz B, Azab F, Azab AK. Cell trafficking of endothelial progenitor cells in tumor progression. Clin Cancer Res 2013;19:3360\u0026ndash;8\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAsahara T, Murohara T, Sullivan A, Silver M, van der Zee R, et al. Isolation of putative progenitor endothelial cells for angiogenesis. Science 1997;275:964\u0026ndash;7\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFadini GP, Losordo D, Dimmeler S. Critical reevaluation of endothelial progenitor cell phenotypes for therapeutic and diagnostic use. Circ Res 2012;110:624\u0026ndash;37\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShaked Y, Henke E, Roodhart JM, Mancuso P, Langenberg MH, et al. Rapid chemotherapy-induced acute endothelial progenitor cell mobilization: implications for antiangiogenic drugs as chemosensitizing agents. Cancer Cell 2008;14:263\u0026ndash;73\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRoodhart JM, Langenberg MH, Vermaat JS, Lolkema MP, Baars A, et al. Late release of circulating endothelial cells and endothelial progenitor cells after chemotherapy predicts response and survival in cancer patients. Neoplasia 2010;12:87\u0026ndash;94\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSayed HAR, Hetta HF, Shafik EA, Zahran AM. Flow cytometric analysis of circulating endothelial cells and endothelial progenitor cells in pediatric solid tumors: prognostic impact on treatment response and survival. Cancer Immunol Immunother 2021;70:755\u0026ndash;61\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYeo B, Redfern AD, Mouchemore KA, Hamilton JA, Anderson RL. The dark side of granulocyte-colony stimulating factor: a supportive therapy with potential to promote tumour progression. Clin Exp Metastasis 2018; 35:255\u0026ndash;67\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang K, Dai X, He J, Yan X, Yang C, et al. Endothelial Overexpression of Metallothionein Prevents Diabetes-Induced Impairment in Ischemia Angiogenesis Through Preservation of HIF-1α/SDF-1/VEGF Signaling in Endothelial Progenitor Cells. Diabetes 2020;69:1779\u0026ndash;92\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu Y, Xu J, Gu R, Li Z, Wang K, et al. Circulating exosomal miR-144-3p inhibits the mobilization of endothelial progenitor cells post myocardial infarction via regulating the MMP9 pathway. Aging (Albany NY) 2020;12:16294\u0026ndash;303\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCetean S, Căinap C, Constantin AM, Căinap S, Gherman A, et al. The importance of the granulocyte-colony stimulating factor in oncology. Clujul Med 2015;88:468\u0026ndash;72\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorris KT, Khan H, Ahmad A, Weston LL, Nofchissey RA, et al. G-CSF and G-CSFR are highly expressed in human gastric and colon cancers and promote carcinoma cell proliferation and migration. Br J Cancer 2014;110:1211\u0026ndash;20\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSacks D, Baxter B, Campbell BCV, Carpenter JS, Cognard C, et al. Multisociety Consensus Quality Improvement Revised Consensus Statement for Endovascular Therapy of Acute Ischemic Stroke. Int J Stroke 2018;13:612\u0026ndash;32\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMartino N, Ramos RB, Lu S, Leyden K, Tomaszek L, et al. Endothelial SOCS3 maintains homeostasis and promotes survival in endotoxemic mice. JCI Insight 2021;6\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDemetri GD, Griffin JD. Granulocyte colony-stimulating factor and its receptor. Blood 1991;78:2791\u0026ndash;808\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWculek SK, Malanchi I. Neutrophils support lung colonization of metastasis-initiating breast cancer cells. Nature 2015;528:413\u0026ndash;7\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEl Rayes T, Catena R, Lee S, Stawowczyk M, Joshi N, et al. Lung inflammation promotes metastasis through neutrophil protease-mediated degradation of Tsp-1. Proc Natl Acad Sci U S A 2015;112:16000\u0026ndash;5\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCoffelt SB, Kersten K, Doornebal CW, Weiden J, Vrijland K, et al. IL-17-producing γδ T cells and neutrophils conspire to promote breast cancer metastasis. Nature 2015;522:345\u0026ndash;8\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShaked Y, Ciarrocchi A, Franco M, Lee CR, Man S, et al. Therapy-induced acute recruitment of circulating endothelial progenitor cells to tumors. Science 2006;313:1785\u0026ndash;7\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang L, DeBusk LM, Fukuda K, Fingleton B, Green-Jarvis B, et al. Expansion of myeloid immune suppressor Gr + CD11b + cells in tumor-bearing host directly promotes tumor angiogenesis. Cancer Cell 2004;6:409\u0026ndash;21\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGreenfield JP, Cobb WS, Lyden D. Resisting arrest: a switch from angiogenesis to vasculogenesis in recurrent malignant gliomas. J Clin Invest 2010;120:663\u0026ndash;7\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eValastyan S, Weinberg RA. Tumor metastasis: molecular insights and evolving paradigms. Cell 2011;147:275\u0026ndash;92\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGingis-Velitski S, Loven D, Benayoun L, Munster M, Bril R, et al. Host response to short-term, single-agent chemotherapy induces matrix metalloproteinase-9 expression and accelerates metastasis in mice. Cancer Res 2011;71:6986\u0026ndash;96\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDaenen LG, Roodhart JM, van Amersfoort M, Dehnad M, Roessingh W, et al. 2011. Chemotherapy enhances metastasis formation via VEGFR-1-expressing endothelial cells. Cancer Res 71:6976\u0026ndash;85\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVeith AP, Henderson K, Spencer A, Sligar AD, Baker AB. Therapeutic strategies for enhancing angiogenesis in wound healing. Adv Drug Deliv Rev. 2019;146: 97\u0026ndash;125.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang W, Zhang Y, Hui H, Tong W, Wei Z, et al. The effect of endothelial progenitor cell transplantation on neointimal hyperplasia and reendothelialisation after balloon catheter injury in rat carotid arteries. Stem Cell Res Ther 2021;12:99\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYoung MR, Newby M, Wepsic HT. Hematopoiesis and suppressor bone marrow cells in mice bearing large metastatic Lewis lung carcinoma tumors. Cancer Res 1987;47:100\u0026ndash;5\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRay P, Arora M, Poe SL, Ray A. Lung myeloid-derived suppressor cells and regulation of inflammation. Immunol Res 2011;50:153\u0026ndash;8\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWei D, Le X, Zheng L, Wang L, Frey JA, et al. Stat3 activation regulates the expression of vascular endothelial growth factor and human pancreatic cancer angiogenesis and metastasis. Oncogene 2003;22:319\u0026ndash;29\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNiu G, Wright KL, Huang M, Song L, Haura E, et al. Constitutive Stat3 activity up-regulates VEGF expression and tumor angiogenesis. Oncogene 2002;21:2000\u0026ndash;8\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKujawski M, Kortylewski M, Lee H, Herrmann A, Kay H, Yu H. Stat3 mediates myeloid cell-dependent tumor angiogenesis in mice. J Clin Invest 2008;118:3367\u0026ndash;77\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFujita Y, Kawamoto A. Stem cell-based peripheral vascular regeneration. Adv Drug Deliv Rev 2017;120:25\u0026ndash;40\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIi M, Nishimura H, Iwakura A, Wecker A, Eaton E, et al. Endothelial progenitor cells are rapidly recruited to myocardium and mediate protective effect of ischemic preconditioning via \"imported\" nitric oxide synthase activity. Circulation 2005;111:1114\u0026ndash;20\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJujo K, Ii M, Losordo DW. Endothelial progenitor cells in neovascularization of infarcted myocardium. J Mol Cell Cardiol 2008;45:530\u0026ndash;44\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMiyamoto Y, Suyama T, Yashita T, Akimaru H, Kurata H. Bone marrow subpopulations contain distinct types of endothelial progenitor cells and angiogenic cytokine-producing cells. J Mol Cell Cardiol 2007;43:627\u0026ndash;35\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang W, Zhang Y, Hui H, Tong W, Wei Z, et al. The effect of endothelial progenitor cell transplantation on neointimal hyperplasia and reendothelialisation after balloon catheter injury in rat carotid arteries. Stem Cell Res Ther 2021;12:99\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLamsfus Calle C, Fendel R, Singh A, Richie TL, Hoffman SL, et al. Expansion of Functional Myeloid-Derived Suppressor Cells in Controlled Human Malaria Infection. Front Immunol 2021;12:625712\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTanaka R, Masuda H, Kato S, Imagawa K, Kanabuchi K, et al. Autologous G-CSF-mobilized peripheral blood CD34 + cell therapy for diabetic patients with chronic nonhealing ulcer. Cell Transplant 2014;23:167\u0026ndash;79\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHannoush EJ, Sifri ZC, Elhassan IO, Mohr AM, Alzate WD, et al. Impact of enhanced mobilization of bone marrow derived cells to site of injury. J Trauma 2011;71:283\u0026ndash;9; discussion 9\u0026ndash;91\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKowanetz M, Wu X, Lee J, Tan M, Hagenbeek T, et al. Granulocyte-colony stimulating factor promotes lung metastasis through mobilization of Ly6G + Ly6C + granulocytes. Proc Natl Acad Sci U S A 2010;107:21248\u0026ndash;55\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJardin F, Vasse M, Debled M, Dominique S, Courville P, et al. Intense paraneoplastic neutrophilic leukemoid reaction related to a G-CSF-secreting lung sarcoma. Am J Hematol 2005;80:243\u0026ndash;5\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKawano M, Mabuchi S, Matsumoto Y, Sasano T, Takahashi R, et al. The significance of G-CSF expression and myeloid-derived suppressor cells in the chemoresistance of uterine cervical cancer. Sci Rep 2015;5:18217\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBraun B, Lange M, Oeckler R, Mueller MM. Expression of G-CSF and GM-CSF in human meningiomas correlates with increased tumor proliferation and vascularization. J Neurooncol 2004;68:131\u0026ndash;40\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchniewind B, Christgen M, Hauschild A, Kurdow R, Kalthoff H, Klomp HJ. Paraneoplastic leukemoid reaction and rapid progression in a patient with malignant melanoma: establishment of KT293, a novel G-CSF-secreting melanoma cell line. Cancer Biol Ther 2005;4:23\u0026ndash;7\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKuderer NM, Dale DC, Crawford J, Cosler LE, Lyman GH. Mortality, morbidity, and cost associated with febrile neutropenia in adult cancer patients. Cancer 2006;106:2258\u0026ndash;66\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMehta HM, Malandra M, Corey SJ. G-CSF and GM-CSF in Neutropenia. J Immunol 2015;195:1341\u0026ndash;9\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi P, Lu M, Shi J, Hua L, Gong Z, et al. Dual roles of neutrophils in metastatic colonization are governed by the host NK cell status. Nat Commun 2020;11:4387\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Vessel formation, metastasis, G-CSF, chemotherapy","lastPublishedDoi":"10.21203/rs.3.rs-1464660/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-1464660/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eChemotherapy remains the mainstay of treatment for various cancers. However recent studies suggested that it may also induce local or systemic changes that promote the dissemination and proliferation of cancer cells leading to successful metastasis. Understanding this process is therefore instrumental to help us identify patient population that may or may not be benefit from chemotherapy, and develop innovative approaches to prevent metastasis.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eIn this study, we investigated the mechanism of chemotherapy causing the metastasis of colorectal cancer, and the roles of G-CSF induced by chemotherapy in the vessel formation. Eight-week-old C57/BL6 mice were used as animal models in this study to investigate the lung metastasis by the chemotherapy. Conditional endothelial cell STAT3\u003csup\u003e\u0026minus;/\u0026minus;\u003c/sup\u003e (Signal Transducer and Activator of Transcription, STAT) knockout mice (STAT3\u003csup\u003eflox/flox\u003c/sup\u003e; Tek-Cre mice), were used to investigate the function of STAT3 on lung metastasis under the chemotherapy. Bone marrow and plasma was selected to detect the mobilization of endothelial progenitor cells (EPCs) and MDSCs. To trace BM-derived cells, we prepared chimeric mouse transplanted with BM cells from green fluorescent protein (GFP) transgenic mouse. In vitro, experiments were performed in human umbilical vein endothelial cells (HUVEC) to analyse the function of G-CSF on angiogenesis. Detected the G-CSF content in the plasma of the patients who received an adjuvant chemotherapy with the XELOX (oxaliplatin plus capecitabine) regimen.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eOur study showed that oxaliplatin chemotherapy could increase the expression of G-CSF to promote lung metastasis. First, G-CSF/STAT3 signaling facilitated lung metastasis by enhancing vascular adhesion rather than diminishing the blood vessel density. G-CSF also promoted Endothelial Progenitor Cells (EPCs) mobilization that devoted to vasculogenesis, a critical step for vessel density in metastatic sites; moreover, chemotherapy augmented the mobilization of MDSCs from the bone marrow by G-CSF. In consistent with these, anti-G-CSF suppressed the formation of a functional vasculature and induced tumoral immunosuppression, resulting in an anti-metastasis effect during chemotherapy. Furthermore, at human level, we observed high level of G-CSF, either at baseline or after receiving adjuvant XELOX chemotherapy, correlated with poor overall and recurrence-free survival.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eThese results demonstrate that certain chemotherapy could paradoxically result in worse outcome due to increased expression of G-CSF. Our findings provide mechanistic insight into cautious use of G-CSF and potential utility of anti-G-CSF in personalized cancer therapy.\u003c/p\u003e","manuscriptTitle":"Chemotherapy may facilitate colorectal cancer metastasis through vessel formation by upregulating G-CSF","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2022-03-25 22:07:39","doi":"10.21203/rs.3.rs-1464660/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"13f249f9-4b69-488c-afd9-11aa8b9590b3","owner":[],"postedDate":"March 25th, 2022","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2022-03-25T22:07:41+00:00","versionOfRecord":[],"versionCreatedAt":"2022-03-25 22:07:39","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-1464660","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-1464660","identity":"rs-1464660","version":["v1"]},"buildId":"7rjqhiLT3MXkJMwkYKINL","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}