Osteoarthritis (OA) affects over 40 million people annually. We evaluated interleukin-1 receptor antagonist (IL-1ra) gene transfer in an equine model based on IL-1ra protein therapy which inhibits inflammation through blocking IL-1. Using the self-complementary adeno-associated virus (scAAV)IL-1ra equine gene as a starting construct, we optimized the transgene cassette by analyzing promoters (cytomegalovirus (CMV) versus chicken β-actin hybrid (CBh)), coding sequences (optimized versus unoptimized), vector capsid (serotype 2 versus chimeric capsid), and biological activity in vitro. AAV serotypes 2 and 2.5 CMV scAAVoptIL-1ra were tested in equine joints.
We evaluated two doses of scAAVIL-1ra, scAAVGFP, and saline. We developed a novel endoscopy procedure and confirmed vector-derived transgene expression (GFP) in chondrocytes 6 months post-injection. AAVIL-1ra therapeutic protein levels were 200–800 ng/ml of synovial fluid over 23 and 186 days, respectively. No evidence of intra-articular toxicity was detected and no vector genomes were found in contralateral joints based on GFP fluorescence microscopy and quantitative PCR. Finally, we assayed vector-derived IL-1ra activity based on functional assays which supported anti-inflammatory activity of our protein. These studies represent the first large animal intra-articular gene transfer approach with a therapeutic gene using scAAV and demonstrate high levels of protein production over extended time supporting further clinical investigation using scAAV gene therapy for OA.
Osteoarthritis (OA) is a progressive and debilitating joint disease for which there is no cure. It affects over 40 million people annually and is responsible for tremendous financial burdens on health systems especially in societies with growing populations of elderly and obese individuals.1 Furthermore, the disease negatively affects younger, active individuals that participate in high demand sports and suffer from post-traumatic arthritis or overuse syndrome.2 The proinflammatory cytokine primarily responsible for many of the pathological features of OA is interleukin-1β (IL-1β).3 IL-1β has been found to be upregulated in synovial tissues and cartilage and a strong correlation has been consistently found with levels of IL-1β and osteoarthritic changes such as cartilage degeneration and presence of inflammation.1 Initially, treatment for OA consisted of nonsteroidal anti-inflammatory drugs and cyclooxygenase-2 inhibitors, which reduced symptoms of pain and inflammation, caused by IL-1β but did not deter progression of OA.4,5 In addition, with prolonged use, these drugs can be associated with gastrointestinal disorders and cardiovascular ischemic events that can sometimes even lead to the demise of the patient.6
In the last two decades, researchers have sought to explore pharmaceuticals that can not only control symptoms of OA but also slow the progression and even prevent or stop degeneration of the joint thus bypassing the need of inevitable joint replacement. Although drugs such as hyaluronan and glucosamine/chondroitin sulfate have been somewhat disease modifying, they have not alleviated symptoms as effectively as originally forecasted.7,8,9,10,11 A logical target for blocking effects of IL-1β is the molecule IL-1 receptor antagonist (IL-1ra). This molecule is the natural inhibitor of IL-1β and competes with IL-1β for occupancy of the IL-1 cell surface receptors but cannot initiate cellular signals when bound to these receptors.3 Studies have revealed that the IL-1ra concentration is low in inflamed joints and a level of tenfold to 1,000-fold excess of IL-1ra over IL-1β is required to effectively block all of the available IL-1 receptors enough to inhibit joint degeneration.12,13,14 Autologous-conditioned serum, harvested from patient’s blood, is a biological treatment that has high levels of constitutive IL-1ra concentrations and has been met with varying degrees of success in both humans and horses suffering from OA.15,16 One reason may be that levels of IL-1ra protein are not high enough and not sustained for periods required to inhibit ongoing inflammation.16
Gene transfer using viral vectors to initiate therapeutic levels of IL-1ra or growth factors in joints is a promising approach and “proof of concept” in delivering therapeutic genes to joints by direct in vivo injection.17,18,19,20,21 Various viral vectors have been utilized for intra-articular gene therapy including adenovirus, retrovirus, lentivirus and adeno-associated virus (AAV).17,19,22,23 Adenoviral vectors have resulted in significant elevations of protein (IL-1ra, insulin-like growth factor-I) when injected intra-articularly; however, levels only remain elevated for 14–21 days and adenoviral vectors themselves can cause significant inflammation due to their immunogenic stimulation.17,18,24 Retroviral vectors have also significantly increased protein levels, however; these vectors do not efficiently transduce nondividing cells and therefore are less suited for joint tissues where cell turnover is low.22 Lentiviral vectors based on integration into the chromosome, offers the potential of long-term expression for OA; however, when injected into the knee joints of immunocompetent rats, the results demonstrate a sharp decrease in protein expression at ~20 days and therefore may not be the vector of choice for long-term protein production for direct intra-articular injections.23 AAV has been studied and validated to result in significant elevations in protein expression.19,25 Kay et al. compared single-stranded AAV to self-complementary AAV (scAAV) and reported a 25-fold greater transgene expression level for scAAV confirming that second strand DNA synthesis can be a major impediment to transduction efficiency in the joint.19 Furthermore, in Kay et al. scAAV not only resulted in dramatically increased transduction efficiency when compared with the single-stranded AAV but also did not reveal a difference in transduction between normal and inflamed articular environments suggesting that scAAV may be an appropriate vector in OA. Suitability of scAAV vectors intra-articularly was further confirmed in equine joints when transduction efficiency of AdGFP, rAAV, and scAAV were compared over a period of 8 weeks.26 Goodrich et al. compared scAAV serotypes in synoviocytes and chondrocytes in vitro and revealed the importance of serotype affinity in that significant differences existed in transduction efficiencies in both synoviocytes and chondrocytes which are the main cell types in joints.27 Serotype 6 was better in chondrocytes and serotype 3 in synoviocytes whereas serotype 2 was best in both chondrocytes and synoviocytes.27 This was also revealed by Sun et al. where various serotypes of scAAVFIX resulted in differing factor IX levels in the joints of mice.28
In addition to varying serotype efficiencies intra-articularly, various promoters may also change protein production.29 Few studies have been performed to investigate optimal use of promoters to enhance transgene expression intra-articularly but existing data suggest different promoters may play an important role in maximizing protein expression and variability exists with different promoters.30,31 Furthermore, optimization of transgene cassette often enhances therapeutic output of protein through favoring codon usage to exploit amino acyl-transfer RNA molecules that are most abundant in mammalian cells.29
Regardless of promising results seen in vitro, protein expression in cell culture may not always represent protein expression in vivo due to unforeseen immunity, transduction efficiency, and species differences. Since gene therapeutic vectors targeted for OA should produce protein levels for long periods of time (over 4 months), the objective of this study was to optimize variables of our scAAV vector such as promoter and transgene, first in vitro and then, using our optimized promoter and transgene, compare our two best scAAV serotypes in vivo to explore the application of scAAV gene therapy in an equine model.11,18,32 We hypothesized that we could optimize promoter and transgene in vitro and test this vector in vivo to determine efficacy of transgene production in the equine joint. This is the first study to report significant protein elevations for up to 6 months using scAAV in large animal (equine) joints.