Our study reveals a non-canonical role for CCL2 in modulating non-macrophage, myeloid-derived suppressor cells (MDSCs) and shaping a tumor-permissive microenvironment during colon cancer development. CCL2, leading to recruitment of tumor-promoting myeloid cells into the tumor during carcinogenesis. qRT-PCR analysis of 43 healthy and CRC patients (nine or ten subjects per CRC stage) suggested that CCL2 transcripts were increased in CRC versus normal tissues (Wolf et al., 2012). Based on these data, we sought to confirm that CCL2 protein levels were increased in human colon ACA. Using human tissue microarrays, which included both normal colon tissue (n = 29) and ACA (n = 119), we found that CCL2 levels significantly increased in the ACA samples (Figure 3A). Figure 3 CCL2 Levels Increase in Human Sporadic CRC and CCL2 Enhances Tumor MDSC Accumulation during Colonic Adenocarcinoma Growth To examine whether cancer-cell-produced CCL2 affects accumulation and function of MDSC populations in colonic ACAs, we employed the Colon-26 colonic ACA transplantation model (Ohana et al., 2003), as ACAs do not develop in and shRNA for GFP (shControl) as a control. We verified knockdown by CCL2 protein determinations from supernatants of the shCCL2 Colon-26 cell lines and selected one stable cell line (shCCL2) (Figure S3A). We measured tumor volume and size and counted the number of intratumoral MDSCs at day 14 after subcutaneous injection of shControl and shCCL2 stable cell lines. This time point was selected to comply with humane endpoints governing tumor size. shControl tumors were considerably larger (nearly 9-fold) than shCCL2 tumors in volume and size (Figure 3B) (p < 0.0001); however, there was no significant difference in in vitro proliferation between shControl and shCCL2 Colon-26 cells (Figure S3B). As expected, we observed higher intratumoral CCL2 levels in shControl tumor as compared to shCCL2 tumor (Figure S3C). We examined CD11b+Gr-1+ MDSC accumulation in shControl and shCCL2 tumors using immunofluorescence microscopy (Figure 3C). Next, we characterized the intratumoral myeloid cell populations using flow cytometry (Figures 3D and 3E) as in Figure 2. MDSC numbers increased over 4-fold (p < 0.0001) and Mo-MDSC and PMN-MDSC subpopulations accumulated over 2-fold in the shControl tumors compared to IL8 those from shCCL2 tumors (p < 0.05 and p < 0.001, respectively) when normalized by tumor weight (Figure 3F). To confirm if CCL2 was driving increased tumor growth and MDSC accumulation, we performed add-back experiments wherein we intratumorally injected recombinant CCL2 or PBS into shCCL2 tumor-bearing mice at day 5 after injection of shCCL2 Colon-26 cells and examined tumor volume and the number of MDSCs at day 14 after injection. Tumor volume was significantly increased in shCCL2 tumor-bearing mice injected with recombinant CCL2 (p < 0.001) (Figure 3G), as were intratumoral MDSC and PMN-MDSC 326914-06-1 numbers (p < 0.05) (Figure 3H). To address if CCL2 affects other myeloid cells in the tumor microenvironment, we examined tumor-promoting macrophages, including TAMs (CD11b+Gr-1?F4/80+) and M2-like TAMs (CD11b+Gr-1?F4/80+MMR+). Tumor-promoting macrophages significantly increased 326914-06-1 in the shControl tumors but were significantly fewer than the MDSC numbers (Figure 3I). There were no statistically significant differences in tumor-associated neutrophils (CD11b+Ly6G+) between shControl and shCCL2 tumor-bearing mice (Figure S3D) (Fridlender et al., 2009). To determine if MDSCs or TAMs contribute to tumor growth, we sorted splenic MDSCs and TAMs from shControl tumor-bearing mice at day 14 and intratumorally injected the cells into shCCL2 tumor-bearing mice at day 5. shCCL2 tumor-bearing mice injected with MDSCs from shControl tumors showed significantly increased tumor growth 326914-06-1 as compared with mice receiving TAMs or PBS (Number 3J). These results indicate that CCL2 runs MDSC build up in the tumor microenvironment and support that MDSCs contribute to improved tumor growth. CCL2 Modulates PMN-MDSC Suppression of Capital t Cells by Increasing PMN-MDSC ROS and Enhancing PMN-MDSC-Mediated Decreases in Capital t Cell Receptor Chain MDSCs suppress Capital t cell service (Gabrilovich and Nagaraj, 2009). While there are some similarities among MDSCs (both PMN-MDSCs and Mo-MDSCs), neutrophils, and inflammatory monocytes, a principal difference is definitely that PMN-MDSCs and Mo-MDSCs suppress Capital t cell expansion but neutrophils and inflammatory monocytes do not (Damuzzo et al., 2015; Gabrilovich et al., 2012). Also, while neutrophils from naive tumor-free mice and intratumoral PMN-MDSCs display related Ly6C and Ly6G staining, their CCR2.