Macrophages attenuate survival of intravenously administered human mesenchymal stromal cells by phagocytosis

Abstract

Human mesenchymal stromal cells (MSCs) are increasing used in clinical trials for a diverse array of applications. Whether allogenic or autologous cells are used, unqualified efficacy has not been shown, suggesting the innate immune system may be impairing MSC activity. For example, MSCs inherently traffic to osteosarcoma (OS) rendering them potentially effective delivery vehicles of therapeutics; however, such strategies have not emerged, possibly due to inadequate tumor localization. We sought to enhance MSC trafficking to OS studying xenograft models of human osteosarcoma in nude and NOD-scid-IL2rγ^null (NSG) mice. After IV infusion, firefly luciferase-expressing MSCs trafficked to the tumor in both models. Interestingly, we observed 3.5-fold greater bioluminescent signal per unit volume from the OS in NSG compared to nude mice, suggesting macrophages in nude mice may be clearing the MSCs. To test our hypothesis, we infused MSCs into tumor-bearing nude mice after clodronate-mediated macrophage depletion and in non-depleted controls. The macrophage-depleted nude mice showed 3.7-fold more MSCs tumor localization on day 11 compared with controls. To evaluate a role of macrophages, 4 hours after claret-labeled MSCs were infused into nude mice. Flow cytometric analysis of a single cell suspension of lung and spleen revealed a claret⁺ human CD90⁺ fraction, consistent with human MSCs, and a claret⁺ F4/80⁺ human CD90⁻ fraction suggestive of macrophages that have phagocytosed MSCs. Fluorescence microscopy confirmed claret+ cells within macrophages. Since clodronate depletes splenic, but not pulmonary, macrophages the splenic macrophages likely have the greatest impact which can be mitigated by transient macrophage depletion.