Engineered cell homing

Abstract

One of the greatest challenges in cell therapy is to minimally invasively deliver a large quantity of viable cells to a tissue of interest with high engraftment efficiency. Low and inefficient homing of systemically delivered mesenchymal stem cells (MSCs), for example, is thought to be a major limitation of existing MSC-based therapeutic approaches, caused predominantly by inadequate expression of cell surface adhesion receptors. Using a platform approach that preserves the MSC phenotype and does not require genetic manipulation, we modified the surface of MSCs with a nanometer-scale polymer construct containing sialyl Lewis(x) (sLe(x)) that is found on the surface of leukocytes and mediates cell rolling within inflamed tissue. The sLe(x) engineered MSCs exhibited a robust rolling response on inflamed endothelium in vivo and homed to inflamed tissue with higher efficiency compared with native MSCs. The modular approach described herein offers a simple method to potentially target any cell type to specific tissues via the circulation.

Figure 1

Cell surface engineered MSCs display enhanced rolling intactions in vitro. (A) Conjugation of sLe^x on the surface of the MSCs through covalent biotinylation and a streptavidin-biotin bridge. (B) Velocity of sLe^x-modified cells compared with PBS-treated cells and glucose-modified cells at 0.36 dyne/cm² on P-selectin–coated substrates. (C) Number of interacting sLe^x-modified cells compared with PBS-treated cells and glucose-modified cells per unit area at 0.36 dyne/cm² on P-selectin–coated substrate over 10 seconds with 0.45 mm² area. (D) Velocity of sLe^x-modified MSCs, HL60, and free stream velocity (theoretically calculated from flow chamber geometry and fluid flow rate) at increasing shear stress.

Figure 2

In vivo rolling of surface engineered MSCs. (A) In vivo confocal video images of sLe^x-MSCs with velocity 250 μm/s (green) were taken at 30 frames/s within the inflamed ear vessel (blue) after injection of MSCs. (B) In vivo confocal video images of unmodified MSCs with velocity 1100 μm/s (red) were taken at 30 frames/s within the inflamed ear vessel (blue) after injection of MSCs. (A-B) The vessel diameter is approximately 60 μm, and the critical velocity is 571 μm/s. (C) Representative image of sLe^x-MSCs interacting with inflamed endothelium in approximately 60 μm vessel resulting in velocity of 100 μm/s (103 frames stacked with 30 fps; ie, sLe^x-MSC remains in the field of view for > 3.3 seconds). (D) Representative image of unmodified MSCs interacting with inflamed endothelium in approximately 60 μm vessel resulting in velocity of 750 μm/s (27 frames stacked with 30 fps; ie, unmodified MSC remains in the field of view for < 0.9 seconds). Bar represents 50 μm.

Figure 3

In vivo rolling velocity of surface engineering MSCs and selectin expression. (A) Velocity of sLe^x-modified MSCs and unmodified MSCs on inflamed endothelium within a vessel of 47 μm diameter where the critical velocity (V_crit) is 191 μm/s. (B) Representative distribution of velocity showing 75% of sLe^x-MSCs and 25% of unmodified MSCs are below the critical velocity. Cells traveling below the critical velocity decelerate on the vessel wall through receptor/ligand-mediated adhesive interactions. (C) In vivo confocal images of anti–P-selectin–labeled postcapillary venules 24 hours after LPS stimulation. Top: Two venules represent anti–P-selectin–Cy5 labeling in LPS-treated ear and saline-treated ear. Bottom: Two venules represent IgG1, λ-Cy5 labeling in LPS-treated ear and saline-treated ear. Bar represents 500 μm. (D) In vivo confocal images of anti–E-selectin–labeled postcapillary venules 24 hours after LPS stimulation. Top venule represents anti–E-selectin–Cy3 labeling in LPS-treated ear, and bottom venule represents anti–E-selectin–Cy3 labeling in saline-treated ear. Bar represents 500 μm *P < .05.

Figure 4

Targeted homing of surface engineering MSCs to inflamed tissue. (A) Percentage increase in sLe^x-modified MSCs compared with unmodified MSCs that homed to the inflamed and saline ear (noninflamed) 24 hours after systemic infusion. The average number of sLe^x-modified MSCs (per field of view) that homed to the inflamed ear was 48 compared with 31 unmodified MSCs, whereas the average number of sLe^x-modified MSCs (per field of view) that homed to the saline ear (noninflamed) was 31 compared with 29 unmodified MSCs. (B) Representative image of MSC localization in the inflamed ear at 24 hours after injection of DiD-labeled sLe^x-MSCs (red) and DiR-labeled unmodified MSCs (green). Most cells extravasated though the vessel walls (visualized by FITC-dextran, blue). No differences in extravasation efficiency were observed, thus indicating that the enhanced homing was because of an engineered rolling response through MSC surface functionalization with sLe^x *P < .05.