Avidin as a model for charge driven transport into cartilage: relevance to osteoarthritis drug delivery

Abstract

Purpose: There is an unmet need for effective ways to locally and safely introduce DMOADs into cartilage, to minimize systemic adverse effects and to increase drug specificity for targets in the tissue. Several studies have focused on drug-encapsulating polymeric particles whose effectiveness depends on their ability to enter the dense cartilage ECM and to be retained over time. Our recent studies using solutes having a wide size range (1-15nm diameter) revealed that nanoparticles with diameter <10nm could penetrate into and diffuse throughout cartilage while larger solutes became trapped within the superficial zone. We now confirmed this size limitation using NeutrAvidin, a globular protein (diam∼7nm, 60kDa, electrically neutral). While nanoparticle transport is size and shape dependent, surface charge can greatly affect their binding within ECM. Therefore, we compared the binding and transport properties of NeutrAvidin to Avidin, a highly positively charged protein of similar MW and size. Methods: Cartilage disks (6mm diam, 1mm thick) were harvested from the femoropatellar grooves of 1-2 week old bovine calf knee joints. Transport and binding properties of FITC-labeled NeutrAvidin (pI∼7) and Avidin (pI∼10.5) were measured. A transparent PMMA transport chamber (Fig 1F) allowed diffusion of solutes from only one side (superficial zone) of the disk. Half-cylindrical disks were placed within slots in the middle of the chambers. The chamber side facing the superficial zone was filled with ∼45μl solution of solute in 1X-PBS; the other side was filled with 1X PBS. The chamber was placed on a low-speed rocker inside a 37-deg incubator. At selected times, samples were removed from the chamber and gently rinsed. A full-thickness slice was cut from the center of the sample and imaged using a confocal microscope at 10X magnification. For desorption studies, the solute solution was removed from the chamber and replaced with 1X or 10X PBS. Results: Figs 1A, B, C along with their fluorescence intensity-vs-depth graphs show increasing penetration of NeutrAvidin at 1, 2 and 4 days from the superficial zone (denoted by arrows). Fluorescence intensity is plotted vs distance from the left edge of the images. Figs. 1D, E show explants after 24h desorption of NeutrAvidin in 1X and 10X PBS respectively, suggesting that most NeutrAvidin diffused out similarly by 24h. In contrast, Avidin completely penetrated and diffused throughout the cartilage within 24h (Fig 2A). While some Avidin diffused out of the explants after 24h in 1XPBS (Fig 2B), significantly more Avidin diffused out in 10X PBS. Explants within the chamber (Fig. 1F) show visual evidence of significantly higher uptake of Avidin compared to NeutrAvidin in 24h (i.e., cartilage fluorescence staining). Conclusions: Our results suggest that the transport and binding of Avidin into bovine cartilage is driven by strong electrostatic interactions. Despite their similar sizes, Avidin penetrated through the full thickness (1mm) of cartilage in 24h, while it took 4 days for NeutrAvidin to penetrate half the thickness. A 400-times greater uptake was measured for Avidin compared to NeutrAvidin in 24h. NeutrAvidin diffused out in both 1X and 10X PBS while Avidin remained bound in 1X PBS but diffused out in 10X PBS. We hypothesize that Avidin binds to negatively charged GAG chains in the cartilage: both Avidin and NeutrAvidin exhibited identical absorption and desorption properties in GAG-depleted cartilage explants (data not shown), further confirming the effects of electrostatic interactions. Using principles of Donnan equilibrium, we calculated a net charge of approx. +15 on Avidin. This work suggests that polymeric particles mimicking the structure of Avidin (i.e., ∼7nm diameter with high positive charge) might facilitate higher and faster uptake into cartilage from the surrounding synovial fluid, and also bind within the tissue, thereby providing sustained local drug delivery.