Targeting of CXCR3 improves anti-myeloma efficacy of adoptively transferred activated natural killer cells

Abstract

Background: The peculiar multiple myeloma microenvironment, characterized by up-regulated levels of several inflammatory chemokines, including the CXCR3 receptor ligands CXCL9 and CXCL10, limits NK cell positioning into the bone marrow by interfering with CXCR4 function. It is still unclear if the consequent reduced influx of transferred cells into the tumor represents a potential limiting factor for the success of NK cell-based adoptive therapy. We hypothesize that inhibition of CXCR3 function on NK cells will result in increased tumor clearance, due to higher NK cell bone marrow infiltration. Methods: Since different activation protocols differently affect expression and function of homing receptors, we analyzed the bone marrow homing properties and anti-tumor efficacy of NK cells stimulated in vitro with two independent protocols. NK cells were purified from wild-type or Cxcr3 -/- mice and incubated with IL-15 alone or with a combination of IL-12, IL-15, IL-18 (IL-12/15/18). Alternatively, CXCR3 function was neutralized in vivo using a specific blocking antibody. NK cell functional behavior and tumor growth were analyzed in bone marrow samples by FACS analysis. Results: Both activation protocols promoted degranulation and IFN-γproduction by donor NK cells infiltrating the bone marrow of tumor-bearing mice, although IL-15 promoted a faster but more transient acquisition of functional capacities. In addition, IL-15-activated cells accumulated more in the bone marrow in a short time but showed lower persistence in vivo. Targeting of CXCR3 increased the bone marrow homing capacity of IL-15 but not IL12/15/18 activated NK cells. This effect correlated with a superior and durable myeloma clearance capacity of transferred cells in vivo. Conclusions: Our results demonstrate that in vitro activation affects NK cell anti-myeloma activity in vivo by regulating their BM infiltration. Furthermore, we provided direct evidence that CXCR3 restrains NK cell anti-tumor capacity in vivo according to the activation protocol used, and that the effects of NK cell-based adoptive immunotherapy for multiple myeloma can be improved by increasing their bone marrow homing through CXCR3 inhibition.