Platelet-like nanoparticles: mimicking shape, flexibility, and surface biology of platelets to target vascular injuries

Abstract

Targeted delivery of therapeutic and imaging agents in the vascular compartment represents a significant hurdle in using nanomedicine for treating hemorrhage, thrombosis, and atherosclerosis. While several types of nanoparticles have been developed to meet this goal, their utility is limited by poor circulation, limited margination, and minimal targeting. Platelets have an innate ability to marginate to the vascular wall and specifically interact with vascular injury sites. These platelet functions are mediated by their shape, flexibility, and complex surface interactions. Inspired by this, we report the design and evaluation of nanoparticles that exhibit platelet-like functions including vascular injury site-directed margination, site-specific adhesion, and amplification of injury site-specific aggregation. Our nanoparticles mimic four key attributes of platelets, (i) discoidal morphology, (ii) mechanical flexibility, (iii) biophysically and biochemically mediated aggregation, and (iv) heteromultivalent presentation of ligands that mediate adhesion to both von Willebrand Factor and collagen, as well as specific clustering to activated platelets. Platelet-like nanoparticles (PLNs) exhibit enhanced surface-binding compared to spherical and rigid discoidal counterparts and site-selective adhesive and platelet-aggregatory properties under physiological flow conditions in vitro. In vivo studies in a mouse model demonstrated that PLNs accumulate at the wound site and induce ∼65% reduction in bleeding time, effectively mimicking and improving the hemostatic functions of natural platelets. We show that both the biochemical and biophysical design parameters of PLNs are essential in mimicking platelets and their hemostatic functions. PLNs offer a nanoscale technology that integrates platelet-mimetic biophysical and biochemical properties for potential applications in injectable synthetic hemostats and vascularly targeted payload delivery.

Keywords: hemostat; layer-by-layer; nanotechnology; platelets; synthetic cells; vascular targeting.

Figure 1

Platelet interactions in hemostasis and corresponding platelet-inspired design of PLN technology. Schematic showing normal hemostatic interactions that inspire PLN design.

Figure 2

Synthesis and characterization of PLNs. (a) Schematic showing the LbL synthesis of PLNs. Note that the schematic shows only two bilayers of PAH/BSA, whereas four bilayers of PAH/BSA were used in this study. (b) Scanning electron micrographs (SEM) of (i) sacrificial 200 nm spherical polystyrene templates, (ii) (PAH/BSA)₄-coated polystyrene templates, and (iii) final PLNs. Scale bars = 200 nm. (c) Complementary coatings of poly(allylamine hydrochrloride)–AlexaFluor 594 (odd layers) and bovine serum albumin–AlexaFluor 488 (even layers) with confocal imaging of final PLNs. (d) FTIR spectra of PS templates, (PAH/BSA)₄ coated templates, and PLNs.

Figure 3

Particle binding. (a) Percentage of surface area coverage of OVA-coated spherical particles of 200 nm (white), 1 μm (light gray), and 2 μm (hatched) diameter to anti-OVA coated microfluidic channel (inset) under flow. (b) Percentage of surface area coverage of 200 nm OVA-coated spheres (cross-hatched), discs stretched from 200 nm spheres (gray) and 200 nm PLNs (black) to anti-OVA coated microfluidic channels (inset) under flow (see Figure 2a (Supporting Information) for device dimensions). Representative image scale bars = 20 μm. At least 10 images for each condition were used for analysis. *Denotes statistical difference (P < 0.05) from all other groups.

Figure 4

In vitro binding of PLNs at various shear stresses. (a) PLNs modified with CBP + VBP show specific adhesion to VWF–collagen-coated surfaces (filled squares), PLNs without peptide modification show minimal adhesion to VWF–collagen-coated surfaces (circles), and PLNs modified with CBP + VBP show minimal adhesion to nontargeted control BSA surfaces (filled triangles) in a parallel plate flow chamber (PPFC) at various shear values. Representative images are shown for the end of the experiment (45 min) for each condition. At least 10 images at for each condition at each time were used for analysis. Scale bars = 50 μm. (b) PLNs modified with FMP aggregate with natural platelets (filled squares), PLNs without peptide modification show minimal aggregation with natural platelets (circles), and PLNs modified with FMP show minimal aggregation with nontargeted control BSA surfaces (filled triangles) in a PPFC at various shear values. Representative images are shown for the end of the experiment (45 min) for each condition. At least 10 images at for each condition at each time were used for analysis. Scale bars = 50 μm. All images were taken at the same magnification. *Denotes statistical difference (P < 0.05) from all other groups at each time point.

Figure 5

In vitro PLN wound adhesion and platelet aggregation in simulated wound environment. (a) Experimental setup for flowing PLNs and activated platelets together over VWF–collagen-coated surfaces. PLNs first encounter a BSA region so as to limit nonspecific binding at the collagen–VWF-coated surface. (b) PLNs are able to adhere to VWF–collagen-coated surfaces while aggregating with natural platelets in a PPFC better than PLNs with no peptides, FMP-only, or CBP + VBP-only. Representative images are shown for the end of the experiment (30 min) for each condition. Scale bar = 50 μm. All images were taken at the same magnification. (c) Pearson’s colocalization coefficient values for: (i) activated platelets + PLNs without peptides (white bars), (ii) activated platelets + PLNs with CBP + VBP modification (hatched bars), (iii) activated platelets + PLNs with FMP modification (gray bars), and (iv) activated platelets + PLNs with CBP + VBP + FMP modification (black bars). At least 10 images for each condition were used for analysis. *Denotes statistical difference (P < 0.05) from all other groups.

Figure 6

In vivo hemostatic effect, biodistribution, and imaging of PLNs in mouse tail transection model. (a) PLNs reduce bleeding times in a tail amputation model in mice. PLNs functionalized with CBP + VBP + FMP peptides reduce bleeding times by 65% compared to control (no injections). (b) Plain PLNs or PLNs functionalized with CBP + VBP + FMP organ distribution at 1 h or 55 min after tail amputation. (c) Brightfield and fluorescent images of tail section clot confirming fluorescently labeled CBP + VBP + FMP PLNs interacting with tail section clot. Scale bar = 100 μm. *Denotes statistical difference (P < 0.05) from saline and PLNs without peptide controls. **Denotes statistical difference (P < 0.05) between PLN formulations in specific tissue.