The fate of nanoscale drug delivery systems for cartilage is influenced by osteoarthritis progression and tissue targeting strategy

Abstract

Purpose: Advancing therapeutics for osteoarthritis (OA) is a pressing clinical need. Many emerging therapies act on chondrocytes, but drug delivery to cartilage has prohibitively low efficiency due to its dense, avascular extracellular matrix. Accordingly, there is significant interest in engineering targeted nanoscale delivery vehicles to facilitate retention in cartilage for localized drug release, thereby improving therapeutic impact. Here, a polymeric nanoparticle (NP) system was developed with amine functionalization for passive cartilage targeting via electrostatic interactions, and conjugation of a type 2 collagen (col2) binding peptide for active targeting. We have shown previously that cationic NPs indeed exhibit greater particle retention in healthy cartilage, likely due to electrostatic interactions with glycosaminoglycans (GAGs), but this observation is diminished in diseased tissue as GAGs are depleted. It is conceivable that the targeting approach may need to be tuned for specific diseases states or clinical applications; therefore, the purpose of this study was to directly compare targeting methods at different stages of cartilage degeneration. We hypothesize that active targeting to the col2 matrix will provide superior tissue retention relative to passive targeting via electrostatic interactions in diseased cartilage, as the tissue becomes less anionic and the porosity increases. Methods: NPs were synthesized with a poly(lactide-co-glycolide) core and polymers on the surface: polyvinyl alcohol for stabilization and/or polyallylamine (PAA) for amine decoration and surface charge modification. For passive cartilage targeting, PAA was increased to yield a cationic NP, which is designed to target the anionic cartilage matrix. For untargeted NP controls, polyallylamine concentration was reduced until the NPs were neutrally charged. For active targeting, NPs were functionalized with col2 binding peptides via conjugation. AlexaFluor dyes were conjugated to the NPs for fluorescence labelling. NP accumulation (loading) within bovine cartilage was assessed after exposing NP suspensions to the articular surface of fresh and enzymatically digested (“OA” mimic) cartilage explants for 1hr. NP release from the explants was assessed after incubation in phosphate buffered saline (PBS) at 37°C for 24hrs. Quantification of the NP loading into and release from cartilage was determined by fluorescence measurements of the tissue homogenate and surrounding PBS media, respectively, and calculated from standard curves. In vivo distribution was performed in healthy and collagenase-induced OA knees by injecting 20uL of NPs intra-articularly. Biodistribution was assessed 48hrs post injection by isolating the major knee tissues and imaging via an In Vivo Imaging System (IVIS). Results: All NP formulations were statistically the same size by dynamic light scattering (Fig 1A). Passive and active NPs were positively charged while untargeted NPs were neutral (Fig 1B). When loading NPs into cartilage, significant improvements were observed in OA tissue relative to healthy tissue in untargeted and active targeted NPs, likely because of increased pore size and access to the col2 network (Fig 1C). Passive targeted NPs demonstrated similar loading in healthy and OA tissue, likely because the increased porosity was counteracted by the weakened electrostatic gradient associated with GAG loss. After 24 hours, most of the untargeted NPs were released from the cartilage in both healthy and OA conditions (Fig 1D). In contrast, the majority of targeted NPs were retained in healthy and OA cartilage, with active targeting resulting in greater retention than passive targeting in OA tissue (Fig 1D). After intra-articular injection, whole joint distribution of both actively and passively targeted NPs was impacted by disease state, with greater cartilage accumulation in OA relative to healthy cartilage (Fig 2). Conclusions: In vitro data suggest that the tissue condition is the factor driving NP loading into cartilage, while NP retention within the cartilage is more heavily influenced by NP targeting strategy. This trend regarding NP loading was confirmed in vivo, with considerable differences in joint tissue biodistribution between healthy and OA joints. Future work will assess the NP-cartilage interactions at longer time scales for improved clinical relevance, as well as the interplay between NP retention, release of emerging OA drugs from the NP system, and therapeutic impact in vitro and in vivo. Overall, these studies highlight the importance of evaluating NP systems across degrees of disease progression. Accordingly, patient stage of disease may inform vehicle design for optimal delivery of emerging OA therapeutics.View Large Image Figure ViewerDownload Hi-res image Download (PPT)