Tumor-Derived Factors Differentially Affect the Recruitment and Plasticity of Neutrophils

Abstract

Neutrophils play a key role in cancer biology. In contrast to circulating normal-density neutrophils (NDN), the amount of low-density neutrophils (LDN) significantly increases with tumor progression. The correlation between these neutrophil subpopulations and intratumoral neutrophils (TANs) is still under debate. Using 4T1 (breast) and AB12 (mesothelioma) tumor models, we aimed to elucidate the source of TANs and to assess the mechanisms driving neutrophils’ plasticity in cancer. Both NDN and LDN were found to migrate in response to CXCL1 and CXCL2 exposure, and co-infiltrate the tumor site ex vivo and in vivo, although LDN migration into the tumor was higher than NDN. Tumor-derived factors and chemokines, particularly CXCL1, were found to drive neutrophil phenotypical plasticity, inducing NDN to transition towards a low-density state (LD-NDN). LD-NDN appeared to differ from NDN by displaying a phenotypical profile similar to LDN in terms of nuclear morphology, surface receptor markers, decreased phagocytic abilities, and increased ROS production. Interestingly, all three subpopulations displayed comparable cytotoxic abilities towards tumor cells. Our data suggest that TANs originate from both LDN and NDN, and that a portion of LDN derives from NDN undergoing phenotypical changes. NDN plasticity resulted in a change in surface marker expression and functional activity, gaining characteristics of LDN.

The Bilateral Interplay between Cancer Immunotherapies and Neutrophils’ Phenotypes and Sub-Populations

“…Injection of isolated NDN to recipient LKRM tumor-bearing mice resulted in decreased tumor growth and a significant response, whereas injection of LDN had no effect on tumor growth ( Figure 4 ). Both NDN and LDN adoptive transfers induced a tendency to increase the percentage of TAN in the tumor ( Figure 4 ), supporting our recent findings demonstrating the ability of both NDN and LDN to infiltrate tumors in vivo [ 47 ].…”

“…We could not find, however, clear changes induced by blocking PD-L1 in vitro in the functions of LDN and TAN we tested. Since we have recently shown that NDN can change in vivo to both LDN and TAN [ 47 ], it is possible that the changes found in the NDN functions are eventually seen in the other neutrophils’ subpopulations in vivo, affecting the results of ICI. In that context, it is important to remember that high levels of PD-L1 in the tumor is a well-known biomarker for the efficacy of ICI [ 48 , 49 ].…”

“…Additionally, up-regulation of PD-L1 expression on neutrophils has been negatively correlated with patients’ survival [ 27 ]. We and others have shown that cytokines such as TGF-β can drive a phenotypical modulation of TAN from N1 (anti-tumor) to N2 (pro-tumor) [ 21 , 22 , 23 ], as well as modulating NDN to become LDN [ 16 , 17 , 47 ]. In parallel, it has been demonstrated that the addition of an anti-TGF-β agent increases the efficacy of ICI treatments in animal models of lung and breast cancer [ 29 , 30 ], although they did not assess the specific changes in neutrophils following this treatment.…”

Chemokines as Regulators of Neutrophils: Focus on Tumors, Therapeutic Targeting, and Immunotherapy

“…The role of CXCR2 and its ligands in neutrophil activation and maturation in the tumor context is still discussed. CXCL1 induced neutrophil transition from NDN to a lowdensity state (LD-NDN) with a phenotype similar to that of LDN, which could promote tumor development [141]. In addition, combined inhibition of CXCR2 and SHP2 in nonsmall cell lung cancer (NSCLC) models, selectively targeted a population of TANs with an immunosuppressive phenotype [142].…”

CXCR2 Mediates Distinct Neutrophil Behavior in Brain Metastatic Breast Tumor

“…Therefore, based on the available information of neu phil involvement in breast tumor progression, it is reasonable to hypothesize that the trophil infiltration and activation within a primary TIME could, at least in part, med the metastatic cascade of the brain-tropic breast tumor cells (schematically shown in ure 1). However, a more fundamental question that remains to be answered is: how would the neutrophil activity and plasticity within a primary TIME be regulated differently [50], depending on the existence of intrinsic differences in the inflammatory signals expressed, by the pro-brain metastatic subtypes present within the parental breast cancer (PBC) [51]?…”