A Novel Role of Coagulation Proteases on Viral-Based Gene Transfer Efficacy

• Published in: Blood Vol. 104; no. 11; p. 691
• Main Authors: Schuettrumpf, Joerg, Zou, Jianxiang, Tai, Shin Jen, Schlachterman, Alexander, Tian, Kian, Edmonson, Shyrie, Liu, Jianhua, Andrade-Gordon, Patricia, High, Katherine, Arruda, Valder
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 16-11-2004
• ISSN: 0006-4971; 1528-0020
• DOI: 10.1182/blood.V104.11.691.691

Coagulation proteases are crucial for hemostasis and have also been implicated in inflammatory responses, blood vessel formation, and tumor cell metastasis. Cellular responses triggered by proteases are mediated by protease-activated receptors (PAR). Adeno-associated virus (AAV)-2 vectors hold promise for the treatment of several diseases and were already tested in Phase I studies for hemophilia B following intramuscular or hepatic artery deliveries. Previously, we determined an unexpected inhibitory effect (60–70% downregulation) on AAV-2 and adenovirus mediated gene transfer by thrombin- or FXa inhibitors. These results were independent of mouse strain, transgene product, or vector promoter, and gene expression by vectors of alternate serotypes AAV-5 or -8, which do not share cellular receptors with AAV-2, were not affected by any drug. Here we present in vivo evidence of a novel role of coagulation proteases and PARs in modulating gene transfer by viral vectors. We tested AAV-2 gene transfer efficacy in (a) animal models for proteases deficiency [FX and FIX deficient animals], (b) PAR-1 or PAR-2 deficient mice, (c) and following in vivo activation of PARs. FX knockout mice with residual activity of only 1–3% of normal (n=9) were injected with AAV-2-human(h)FIX vector and compared to littermates with FX levels of 50% (n=4). FIX expression levels were 2-fold lower among FX-deficient mice compared to controls (p<0.03). The second model, FIX deficient mice, received AAV expressing α1-antitrypsin (AAT-1). Severe hemophilia B models due to large-gene deletion (n=5) or missense mutation (R180T) in the FIX gene (n=3, <1% FIX) were compared to littermate controls with normal FIX levels (n=6). The results showed that AAT-1 levels among hemophilia B mice were 2-fold lower than in controls (24 vs 48 ng/ml, p<0.05, respectively). Because PAR activation by thrombin enhances αVβ5 (co-receptor for AAV-2 and adenovirus)-dependent cellular function (JBC 276:10952) we hypothesized that PAR modulates AAV-2 gene transfer. Homozygous (−/−) or heterozygous deficient (+/−) PAR-1 (n=24) or PAR-2 (n=25) mice received AAV-2-hF.IX and were compared to littermate controls (+/+). FIX levels among PAR-1 controls (1.9 μg/ml) were comparable to levels obtained among heterozygotes but higher than in homozygotes (1.1 μg/ml, p<0.02). Similarly, PAR-2 deficient mice presented 2-fold lower FIX levels than controls (0.7 vs 1.3 μg/ml, p<0.02) whereas heterozygous mice presented intermediate levels. To further confirm the role of PARs in AAV-2 gene transfer we activated PARs prior to AAV-2 injection. C57BL/6 mice received specific peptide agonists at doses ranging from 10 to 60 μM/kg (n=4 per dose and per peptide) and were compared to controls receiving scramble peptide. FIX levels increased 1.5 to 5-fold in a dose-dependent manner and the activation of PAR-1 and -2 simultaneously was superior to single peptide. Gene copy monitoring revealed low vector uptake by livers of PAR knockout mice while activation of PARs increased uptake. In conclusion, these data demonstrated a novel in vivo role of coagulation proteases and PARs on viral vectors (AAV-2 and adenovirus)-mediated gene expression and provide an alternative target to modulate gene therapy strategies.