Objective. The imbalance between matrix metalloproteinases (MMPs) 1, 3, and 9 and their specific inhibitor, tissue inhibitor of metalloproteinases 1 (TIMP-1), is a critical step in cartilage injury and angiogenesis in arthritis. To explore the therapeutic potential of TIMP-1 gene transfer in erosive arthritis, the effects of an adenoviral vector (Ad-TIMP-1) were assessed in DBA/1 mice with collagen-induced arthritis (CIA).
Methods. DBA/1 mice with CIA received an intravenous injection of replication-deficient adenovirus containing the human TIMP-1 gene or a control LacZ gene on day 28 postimmunization. The efficiency of gene transfer was determined by serum TIMP-1 detection, measurements of paw swelling, as well as radiologic and histologic examination of the paws.
Results. A single administration of Ad-TIMP-1 resulted in detectable serum levels of the exogenous protein for at least 13 days. The incidence and onset of arthritis were not statistically modified after human TIMP-1 gene transfer in DBA/1 mice compared with control mice. However, the severity of inflammation was statistically significantly increased in Ad-TIMP-1treated mice and a similar trend was observed in the histologic and radiologic scores. With regard to the mechanisms of the worsened effect in the Ad-TIMP-1-treated mice, we observed 1) higher serum levels of anti-type II collagen IgG2a, 2) a significant increase in endogenous soluble tumor necrosis factor receptor I (TNFRI) in sera, and 3) increased labeling of mouse tumor necrosis factor ␣ and TNFRI within arthritic joints.
Conclusion. These findings show that overexpression of TIMP-1 does not prevent osteochondral injury in a mouse model of arthritis. Since MMPs have overlapping properties in terms of their roles in extracellular matrix degradation, angiogenesis, and shedding of cell surface adhesion molecules, cytokines, and cytokine receptors, the paradoxical results obtained suggest that TIMP-1 is probably not the main inhibitor to target.