Splenic macrophage phagocytosis of intravenously infused mesenchymal stromal cells attenuates tumor localization

Abstract

Mesenchymal stromal cells (MSCs) possess remarkable tumor tropism, making them ideal vehicles to deliver tumor-targeted therapeutic agents; however, their value in clinical medicine has yet to be realized. A barrier to clinical utilization is that only a small fraction of infused MSCs ultimately localize to the tumor. In an effort to overcome this obstacle, we sought to enhance MSC trafficking by focusing on the factors that govern MSC arrival within the tumor microenvironment. Our findings show that MSC chemoattraction is only present in select tumors, including osteosarcoma, and that the chemotactic potency among similar tumors varies substantially. Using an osteosarcoma xenograft model, we show that human MSCs traffic to the tumor within several hours of infusion. After arrival, MSCs are observed to localize in clusters near blood vessels and MSC-associated bioluminescence signal intensity is increased, suggesting that the seeded cells expand after engraftment. However, our studies reveal that a significant portion of MSCs are eliminated en route by splenic macrophage phagocytosis, effectively limiting the number of cells available for tumor engraftment. To increase MSC survival, we transiently depleted macrophages with liposomal clodronate, which resulted in increased tumor localization without substantial reduction in tumor-associated macrophages. Our data suggest that transient macrophage depletion will significantly increase the number of MSCs in the spleen and thus improve MSC localization within a tumor, theoretically increasing the effective dose of an anti-cancer agent. This strategy may subsequently improve the clinical efficacy of MSCs as vehicles for the tumor-directed delivery of therapeutic agents.

Keywords: cancer cell therapy; lentiviral transduction; mesenchymal stromal cells (MSCs); phagocytosis; splenic macrophage; stem cell transplantation; tumor homing.

Figure 1.

Characterization of human MSCs labeled with ffLuc and ZsGreen. (A) Flow cytometric analysis showed that ffLuc MSCs were positive for CD90, CD73 and CD105 and negative for hematopoietic cell markers CD45, CD34, CD3, CD14, CD19 and HLA-II. (B) The ffLuc MSCs differentiated into adipocytes (Oil Red O staining) and osteocytes (Alizarin Red S staining) in vitro (magnification ×20). (C) MSCs were transfected with ffLuc and ZsGreen reporter gene by lentiviral transduction. Transduction efficiency was analyzed by flow cytometry. (D) The ffLuc MSCs uniformly expressed ZsGreen within the cytosol. (E, F) In vitro analysis showed a linear correlation between the cell number and bioluminescence imaging signal of ffLuc MSCs and rLuc OS17, respectively. (G) The rLuc-expressing OS17 cells and ffLuc-expressing MSCs displayed specific reactivity with coelentrazine and D-luciferin, respectively.

Figure 2.

MSC migration assays toward various cancer cell CMs. The ffLuc MSCs significantly migrated toward CM from CHP-134, OS17, OHS, OS-25, A673 and U87MG cell lines in transwell migration assay. Representative of 2 independent experiments performed in triplicate. **P< 0.01, ***P< 0.001, ****P< 0.0001. RFU, relative fluorescence unit.

Figure 3.

Biodistribution of ffLuc MSCs in OS17 tumor-bearing nude mice. (A) MSCs were injected intravenously and imaged 1 day, 3 days, 5 days, 7 days, 10 days, 14 days and 17 days after injection. Initial localization of MSCs in lung area and eventual localization in tumor area were demonstrated. (B) Bioluminescence signal intensity was quantified over the tumor regions. (C) Immunohistochemistry was performed with MSC marker anti-human CD90 antibody and Hematoxylin on tumor sections to detect MSCs. Representative images of tumor regions were taken 17 days after MSC injection of MSC- and PBS-injected mice (magnification ×20). Data were analyzed using a linear mixed-effects model and post hoc pairwise comparisons. Tukey-adjusted Pvalues were used to determine significance. *P< 0.05, **P< 0.01, ****P< 0.0001. PBS, phosphate-buffered saline.

Figure 4.

Biodistribution of MSCs in nude and NSG tumor models. (A) The ffLuc MSCs were injected intravenously into nude and NSG mice and imaged on day 1, day 2, day 5, day 8, day 10 and day 14 post-injection using Xenogen IVIS. (B) Bioluminescence signal was quantified in the tumors of nude and NSG groups. (C) Tumor sizes were measured with bioluminescence imaging using coelentrazine substrate in rLuc OS17 tumor-bearing nude and NSG animals. (D) Correlation between tumor sizes and MSC migration was quantified in both animal models. Data were analyzed using a linear mixed-effects model and post hoc pairwise comparisons. Tukey-adjusted Pvalues were used to determine significance. ***P< 0.001, ****P< 0.0001. IVIS, in vivo imaging system.

Figure 5.

Effects of NK cells and B lymphocytes on MSC engraftment. (A) NK cells were depleted with ASGM1 antibody in spleen. (B) NK cell depletion did not have any impact on splenic macrophage levels. (C) MSC migration was analyzed on day 1, day 2, day 5, day 8 and day 11 by bioluminescence imaging and compared with NK cell non-depleted and control groups. Bioluminescence signal was quantified and analyzed in the tumor regions. (D) MSC migration was analyzed in SCID and nude mouse models. On day 1, day 3, day 5, day 7, day 10, day 12 and day 14 after MSC injection, animals were imaged using Xenogen IVIS, and MSC signal was analyzed in tumor regions, comparing each experimental cohort with their own strain-specific controls. Data were analyzed using a linear mixed-effects model and post hoc pairwise comparisons. Tukey-adjusted Pvalues were used to determine significance. *P< 0.05, **P< 0.01, ****P< 0.0001. ASGM1, anti-asialo GM1; IVIS, in vivo imaging system.

Figure 6.

Effects of macrophage depletion on MSC engraftment and survival. (A, B) Flow cytometric analysis of macrophage depletion in spleen with liposomal clodronate. Percentage of F4/80+CD11b+ macrophages after liposomal clodronate administration in lung (C) and tumor (D) tissues. (E) After macrophage depletion, ffLuc MSCs were injected intravenously; 3 h after MSC injection, MSC migration to the tumor area was analyzed by bioluminescence imaging. (F) MSC localization in the spleen area was compared with the non-depleted group 5 h and 48 h after MSC injection. (G) Initial localization of MSCs in lung area and eventual localization in tumor area were imaged on day 1, day 2, day 5, day 8, day 11 and day 14 post-injection in macrophage-depleted and non-depleted control mice. Bioluminescence signal was quantified and analyzed in the tumor regions of animal groups. (H) On day 16, MSC engraftment in tumor regions of macrophage-depleted nude mice was confirmed by immunohistochemistry. Tumor sections harvested from macrophage-depleted nude mice exposed to ffLuc MSCs or negative control (PBS) were stained with human anti-CD90 antibody to show MSC incorporation within tumor sections. Data were analyzed using a linear mixed-effects model and post hoc pairwise comparisons. Tukey-adjusted P values were used to determine significance. *P < 0.05, ***P < 0.001, ****P < 0.0001. mac, macrophage; PBS, phosphate-buffered saline.

Figure 7.

Analysis of spleen macrophages after MSC injection. (A) After intravenous injection of Claret-labeled MSCs, spleen cells were isolated and labeled with F4/80 and CD11b antibodies. Representative example of flow cytometric analysis at 3 h and 24 h in macrophage spleen cell population. After gating on macrophages (F4/80+CD11b+), Claret+ cells were identified according to the expression of Claret on this macrophage population. (B) Following isolation of spleen cells, cells were cultured for 3 days and stained with F4/80 antibody. Claret staining was found inside the macrophages. Smaller panels represent enlarged views of dotted rectangles. DAPI (blue) stained nucleus and anti-F4/80 antibody (red) stained macrophages. (C) MSCs were labeled with DiI membrane stain (red) and administered intravenously 4 h before animals were killed. Spleen cell suspension was stained with F4/80 antibody (green) and DAPI (blue), cytospinned and imaged with confocal microscopy. The image was digitally rotated along the x- or y-axis to demonstrate that the DiI signal (engulfed MSCs) was within the macrophage (magnification ×63). DAPI, 4’,6-diamidino-2-phenylindole; fmo, fluorescence minus one.