BACKGROUND AND AIMS Thrombomodulin (TM), through its lectin-like domain (TMD1), sequesters proinflammatory high-mobility group box 1 (HMGB1) to prevent it from engaging the receptor for advanced glycation end product (RAGE) that sustains inflammation and tissue damage. Our previous study demonstrated that short-term treatment with recombinant TM containing all the extracellular domains (i.e., rTMD123) inhibits HMGB1-RAGE signaling and confers protection against CaCl2-induced AAA formation. In this study, we attempted to further optimize TM domains, as a potential therapeutic agent for AAA, using the recombinant adeno-associated virus (AAV) vector. METHODS The therapeutic effects of recombinant TMD1 (rTMD1) and recombinant AAV vectors carrying the lectin-like domain of TM (rAAV-TMD1) were evaluated in the CaCl2-induced AAA model and angiotensin II-infused AAA model, respectively. RESULTS In the CaCl2-induced model, treatment with rTMD1 suppressed the tissue levels of HMGB1 and RAGE, macrophage accumulation, elastin destruction and AAA formation, and the effects were comparable to a mole-equivalent dosage of rTMD123. In the angiotensin II-infused model, a single intravenous injection of rAAV-TMD1 (1011 genome copies), which resulted in a persistently high serum level of TMD1 for at least 12 weeks, effectively attenuated AAA formation with suppression of HMGB1 and RAGE levels and inhibition of proinflammatory cytokine production, macrophage accumulation, matrix metalloproteinase activities and oxidative stress in the aortic wall. CONCLUSIONS These findings corroborate the therapeutic potential of the TM lectin-like domain in AAA. The attenuation of angiotensin II-infused AAA by one-time delivery of rAAV-TMD1 provides a proof-of-concept validation of its application as potential gene therapy for aneurysm development.