A novel, clinically expedient, AAV-based gene therapy for bone healing - Abstract This project will develop a novel gene therapy for bone healing based upon new insights from our laboratory on the osteogenic properties of transgenic bone morphogenetic protein-2 (BMP-2), its modulation by inflammatory mediators, especially interleukin-1 (IL-1), and the biology of the healing of large osseous segmental defects. A validated rat model will be used in which a 5 mm critical size, segmental defect is surgically created in the femur. Both female and male animals will be used. BMP-2 is a highly osteogenic morphogen. Recombinant, human (rh) BMP-2 is used clinically in bone healing but its potency is weak, necessitating the use of milligram amounts of rhBMP-2. These are expensive and produce a variety of side effects, some of them serious. Prior work by our group has shown that transgenic BMP-2 delivered by an adenovirus vector is 2-3 log orders more effective than rhBMP-2 in healing rat segmental defects but that its effectiveness is limited by an inflammatory response to the adenovirus and, in particular, the local synthesis of IL-1 which reduces the effectiveness of BMP-2. By inhibiting chondrogenesis, IL-1 inhibits endochondral ossification and drives osteogenesis down the intramembraneous pathway. We have shown that, in this model, the intramembraneous route produces bone of inferior quality. In the proposed research we will develop an effective, affordable, clinically expedient gene therapy for healing critical size, osseous defects based on the use of adeno-associated virus (AAV), a less inflammatory vector, to deliver BMP-2 to the defect site. Because our laboratory focuses on research translation, we evaluated different serotypes of AAV to identify a serotype that transduced both human and rat mesenchymal stromal cells (MSCs). Finding none, we engineered AAV8 to include a RGD sequence to enable integrin binding and confirmed that this novel serotype transduced both human and rat MSCS, and transduced cells within the femoral defect in vivo. We thus propose to use AAV8-RGD in this project. In Specific Aim 1, vector encoding a green fluorescent protein-luciferase (GFP-luc) marker gene will be implanted into the defect on the same collagen sponge that is used clinically to deliver rhBMP-2. The level and duration of transgene expression will be measured by an in vitro imaging system (IVIS), enabling us to identify a dose that efficiently transduces cells within the defect with transgene expression persisting for at least 1-3 weeks which our prior data suggest is optimal for bone healing in this model. In the second part of this aim, AAV8-RGD will be engineered to encode either BMP-2 alone or BMP-2 and IL-1 receptor antagonist (IL-1Ra) in a bicistronic vector. The two constructs will be compared in their ability to heal the defect and an optimal dose identified. In Specific Aim 2, the quality of the regenerate will be assessed by mechanical testing and micro-computed tomography. Histology will determine whether the new bone form intramembraneously or by endochondral ossification. ELISA and RT-PCR will be used to measure expression of IL-1a, IL-1b and human and rat BMP-2. In Specific Aim 3, single cell RNA sequencing and in situ hybridization (RNAScope) will be used to identify transduced cells and locate them within the defect.
