Liposomal Nanocarriers Designed for Sub-Endothelial Matrix Targeting under Vascular Flow Conditions

Abstract

Vascular interventions result in the disruption of the tunica intima and the exposure of sub-endothelial matrix proteins. Nanoparticles designed to bind to these exposed matrices could provide targeted drug delivery systems aimed at inhibiting dysfunctional vascular remodeling and improving intervention outcomes. Here, we present the progress in the development of targeted liposomal nanocarriers designed for preferential collagen IV binding under simulated static vascular flow conditions. PEGylated liposomes (PLPs), previously established as effective delivery systems in vascular cells types, served as non-targeting controls. Collagen-targeting liposomes (CT-PLPs) were formed by conjugating established collagen-binding peptides to modified lipid heads via click chemistry (CTL), and inserting them at varying mol% either at the time of PLP assembly or via micellar transfer. All groups included fluorescently labeled lipid species for imaging and quantification. Liposomes were exposed to collagen IV matrices statically or via hemodynamic flow, and binding was measured via fluorometric analyses. CT-PLPs formed with 5 mol% CTL at the time of assembly demonstrated the highest binding affinity to collagen IV under static conditions, while maintaining a nanoparticle characterization profile of ~50 nm size and a homogeneity polydispersity index (PDI) of ~0.2 favorable for clinical translation. When liposomes were exposed to collagen matrices within a pressurized flow system, empirically defined CT-PLPs demonstrated significant binding at shear stresses mimetic of physiological through pathological conditions in both the venous and arterial architectures. Furthermore, when human saphenous vein explants were perfused with liposomes within a closed bioreactor system, CT-PLPs demonstrated significant ex vivo binding to diseased vascular tissue. Ongoing studies aim to further develop CT-PLPs for controlled targeting in a rodent model of vascular injury. The CT-PLP nanocarriers established here show promise as the framework for a spatially controlled delivery platform for future application in targeted vascular therapeutics.

Keywords: hemodynamic flow; liposomes; targeted drug delivery; vascular shear stress; vascular therapeutics.

Figure 1

Spectra and structures of lipid/peptide constituents used for collagen-targeting peptide-modified lipid (CTL) conjugation. The mass spectrometry confirmed peptide conjugation, where peaks of aggregate molecular weight were demonstrated in the reaction product spectra, compared to peaks in the collagen-targeting peptide (CTP) spectra and pre-clicked DSPE-PEG-DBCO spectra alone.

Figure 2

Schematic representation of collagen-targeting liposome (CT-PLP) and PEGylated liposome (PLP) controls. PLPs were assembled with bulk lipid DOPC and Chol at a mole ratio of 7:3 plus 10 mol% DSPE-PEG and 0.1 mol% rhodamine-DOPE for fluorescent labeling. CT-PLPs were formed likewise with the addition of CTP-conjugated CTL.

Figure 3

Schematic representation of the FCS3 parallel-plate flow chamber setup used for the simulated hemodynamic flow.

Figure 4

Schematic representation of the bioreactor setup used for human vessel explant perfusion.

Figure 5

CTL incorporation increased liposomal collagen binding in a dose-dependent manner. The binding affinity of CT-PLPs assembled with 2.5 and 5 mol% CTL was significantly higher than in all other groups. Binding was detected by the fluorometry of Rho-labeled lipids, and data of the groups were presented as mean % lipid bound compared to respective total lipid serial dilution curves.

Figure 6

Pre-insertion was determined to be the optimal modification strategy for CT-PLP assembly. (A) When 5 mol% CTL was incorporated via pre-insertion (PreCTL), liposome binding was significantly increased over CTL incorporation via post-insertion (PostCTL) and over PLP controls at all time points. Binding was detected by the fluorometry of Rho-labeled lipids, and data in the groups were presented as mean % lipid bound compared to the respective total lipid serial dilution curves at each time point. (B) Representative fluorescent microscopy images showing Rho-labeled liposomes bound to collagen IV matrices after 24 h static incubation.

Figure 7

CT-PLPs bound to collagen IV matrices under a continuous flow. (A) At hemodynamic conditions, simulated at venous and arterial physiological and pathological sheer stress, CT-PLPs assembled with 5 mol% CTL via pre-insertion demonstrated increased binding over time. (B) After 45 min of a continuous flow, CT-PLP binding was significantly increased over PLP control binding at all simulated venous and arterial physiological and pathological sheer stresses. Binding was detected by the fluorescent microscopy of Rho-labeled lipids, and data were presented as the mean fluorescent intensity normalized to the background fluorescent intensity. (C) Representative fluorescent microscopy images showing Rho-labeled CT-PLP and PLP liposomes bound to collagen IV matrices after 45 min of a continuous flow at the simulated venous and arterial physiological and pathological sheer stresses.

Figure 8

CT-PLP binding was significantly increased over PLP controls in human lower extremity vessel explants under ex vivo vascular perfusion. (A) Representative near-infrared (NIR) images showing Cy7-labeled liposomes bound to saphenous vein segments after 30 min continuous ex vivo perfusion. (B) After 30 min of a continuous flow, CT-PLP binding was significantly increased over PLP controls. Binding was detected using in vivo imaging system software to quantify fluorescent intensity per vessel area, and data were presented as the mean fluorescent intensity normalized to the fluorescent intensity of the background of non-perfused control vessels.