J.K. Venkatesan

PEO-PPO-PEO Carriers for rAAV-Mediated Transduction of Human Articular Chondrocytes in Vitro and in a Human Osteochondral Defect Model

Gene therapy is an attractive strategy for the durable treatment of human osteoarthritis (OA), a gradual, irreversible joint disease. Gene carriers based on the small human adeno-associated virus (AAV) exhibit major efficacy in modifying damaged human articular cartilage in situ over extended periods of time. Yet, clinical application of recombinant AAV (rAAV) vectors remains complicated by the presence of neutralizing antibodies against viral capsid elements in a majority of patients. The goal of this study was to evaluate the feasibility of delivering rAAV vectors to human OA chondrocytes in vitro and in an experimental model of osteochondral defect via polymeric micelles to protect gene transfer from experimental neutralization. Interaction of rAAV with micelles of linear (poloxamer PF68) or X-shaped (poloxamine T908) poly(ethylene oxide) (PEO) and poly(propylene oxide) (PPO) copolymers (PEO-PPO-PEO micelles) was characterized by means of isothermal titration calorimetry. Micelle encapsulation allowed an increase in both the stability and bioactivity of rAAV vectors and promoted higher levels of safe transgene (lacZ) expression both in vitro and in experimental osteochondral defects compared with that of free vector treatment without detrimental effects on the biological activity of the cells or their phenotype. Remarkably, protection against antibody neutralization was also afforded when delivering rAAV via PEO-PPO-PEO micelles in all systems evaluated, especially when using T908. Altogether, these findings show the potential of PEO-PPO-PEO micelles as effective tools to improve current gene-based treatments for human OA.

Effective Remodelling of Human Osteoarthritic Cartilage by sox9 Gene Transfer and Overexpression upon Delivery of rAAV Vectors in Polymeric Micelles

Recombinant adeno-associated virus (rAAV) vectors are well suited carriers to provide durable treatments for human osteoarthritis (OA). Controlled release of rAAV from polymeric micelles was already shown to increase both the stability and bioactivity of the vectors while overcoming barriers, precluding effective gene transfer. In the present study, we examined the convenience of delivering rAAV vectors via poly(ethylene oxide) (PEO) and poly(propylene oxide) (PPO) polymeric (PEO-PPO-PEO) micelles to transfer and overexpress the transcription factor SOX9 in monolayers of human OA chondrocytes and in experimentally created human osteochondral defects. Human osteoarthritic (OA) chondrocytes and human osteochondral defect models were produced using human OA cartilage obtained from patients subjected to total knee arthroplasty. Samples were genetically modified by adding a rAAV-FLAG-h sox9 vector in its free form or via polymeric micelles for 10 days relative to control conditions (unmodified cells). The effects of sox9 overexpression in human OA cartilage samples were monitored by biochemical, histological, and immunohistochemical analyses. Delivery of rAAV-FLAG-h sox9 via polymeric micelles enhanced the levels of sox9 expression compared with free vector administration, resulting in increased proteoglycan deposition and in a stimulated cell proliferation index in OA chondrocytes. Moreover, higher production of type II collagen and decreased hypertrophic events were noted in osteochondral defect cultures when compared with control conditions. Controlled therapeutic rAAV sox9 gene delivery using PEO-PPO-PEO micelles is a promising, efficient tool to promote the remodelling of human OA cartilage.

Effects of single rAAV-mediated TGF-beta overexpression over time in human osteoarthritic articular cartilage

Purpose: rAAV vectors are promising tools to directly apply candidate genes in osteoarthritic (OA) cartilage. Here, we analyzed the effects of TGF-b overexpression via rAAV upon the structure of human OA cartilage in situ using one single application over a prolonged period of time. Methods: The vectors were packaged, purified, and titrated using standard protocols. Human normal articular cartilage was obtained from unaffected areas in knee joints removed during tumor surgery (n 1⁄4 10) and human OA cartilage from joints undergoing total knee arthroplasty (n 1⁄4 12) (Mankin score 7-9). Explants and chondrocytes were prepared using standard protocols. Cells and explants were transduced with 40 ml vectors for up to 30 days. The DNA contents were assayed using Hoechst 33258, the proteoglycan contents by binding to DMMB, and the type-II and type-X collagen contents by ELISA. Paraffinembedded sections were stained with safranin O/H&E and to detect TGF-b, type-II and type-X collagen. TGF-b was also monitored by ELISA. Proliferation was evaluated by immunolabeling following BrdU incorporation. Morphometric measurements were performed at 3 standardized sites along the explant surface by image analysis. Each condition was performed in duplicate in 3 independent experiments. The t-test and Mann-Whitney Rank Sum Test were employed with P 0.05 considered statistically significant. Results: TGF-b production was sustained in rAAV-hTGF-b chondrocytes compared with rAAV-lacZ (184.2 vs.11.3 pg/ml/24 h in normal cells, 219.4 vs. 10.6 pg/ml/24 h in OA cells, up to 21-fold difference, always P 0.001). The % of cells positive for BrdU uptake (Fig. 1) and the contents of DNA, proteoglycans, and type-II collagen were always higher in the rAAV-hTGF-b normal and OA cells compared with rAAVlacZ (up to 12-fold, always P 0.001). The type-X collagen contents significantly decreased in OA cells with rAAV-hTGF-b (1.6-fold, P 0.001). TGF-b production was also sustained in rAAV-hTGF-b explants compared with rAAV-lacZ (724.5 vs. 92.3 pg/ml/24 h in normal cartilage, 987.7 vs. 83.4 pg/ml/24 h in OA cartilage, up to 12-fold difference, always P 0.001) (Fig. 2). The cell densities (Fig. 3), % of cells positive for BrdU uptake (Fig. 4), DNA contents, proteoglycan contents and matrix staining intensity (Fig. 3), and type-II contents and immunoreactivity (insets of Fig. 3) were always higher in the rAAV-hTGFb normal and OA explants compared with rAAV-lacZ (up to 23-fold, always P 0.001). The type-X collagen contents and immunoreactivity significantly decreased in OA cartilage with rAAV-hTGF-b (4fold, P 0.001) (Fig. 5). Conclusions: TGF-b can be overexpressed in human normal and OA chondrocytes via rAAV in vitro and in situ, leading to enhanced cell proliferation and matrix synthesis and to OA cartilage remodelling with reduced hypertrophic type-X collagen expression. Administration of the vector in vivo, probably combined with other favorable candidates, will allow to determine the beneficial effects of the approach on reconstructing OA cartilage to avoid osteophyte formation. Fig. 1. Immunocytochemical detection of BrdU in transduced human normal and OA chondrocytes (day 30). Magnification x10.