The inhalational form of anthrax, a disease caused by infection with Bacillus anthracis, is often fatal and is a serious threat for use in bioterror attacks. The major virulence factors are encoded by the genes for protective antigen (PA), lethal factor (LF), and edema factor. Lethal toxin, comprising PA and LF, is responsible for systemic fatal pathophysiology associated with the B. anthracis infection. Currently available methods of protecting the general civilian population against anthrax are limited. A vaccine based on recombinant PA is being developed and anti-PA (|[alpha]|PA) monoclonal antibodies are being evaluated for induction of passive immunity against B. anthracis. We have previously demonstrated that an adenovirus vector encoding |[alpha]|PA single-chain antibody provided a survival advantage from anthrax lethal toxin challenge (Kasuya et al. Mol. Ther.2005, 11: 237|[ndash]|244). In the present study, we hypothesized that an adeno-associated virus type rh10 vector encoding an |[alpha]|PA (heavy and light chain) full-length antibody would provide a robust, persistent immunotherapy against anthrax toxin. We evaluated the expression of full-length mouse monoclonal |[alpha]|PA from three different constructs encoding |[alpha]|PA light chain (LC) and heavy chain cDNAs, including: (1) expression from a single CMV promoter with a poliovirus internal ribosome entry site (IRES) between the subunits [CMV(LC)IRES(HC)]; (2) expression of each subunit from its own CMV promoter [CMV(HC)CMV(LC)]; and (3) expression from one CMV promoter with a self cleaving peptide 2A between subunits [CMV(HC)2A(LC)]. These constructs were packaged into AAVrh10 vectors and evaluated for the expression of |[alpha]|PA in vitro following infection of 293 cells. At 72 hr post-infection, the |[alpha]|PA antibody titers (PA-specific ELISA) in the infected cell medium were 100 |[plusmn]| 3 for AAVrh10CMV(LC)IRES(HC), 60 |[plusmn]| 2 for AAVrh10CMV(HC)CMV(LC), and 600 |[plusmn]| 20 for A AVrh10CMV(HC)2A(LC), with the titer generated by AAVrh10CMV(HC)2A(LC) significantly higher than the other groups (p<0.001). All vectors were evaluated for the expression of |[alpha]|PA in vivo following intravenous administration of 1011 genome copies to male C57BL/6 mice. After 2 wk, serum titers of anti-PA (ELISA) were: AAVrh10CMV(LC)IRES(HC) 20 |[plusmn]| 2, AAVrh10CMV(HC)CMV(LC) 200 |[plusmn]| 40, and AAVrh10CMV(HC)2A(LC) 5300 |[plusmn]| 400 [AAVrh10CMV(HC)2A(LC) p<0.001 compared to other groups]. The AAVrh10CMV(HC)2A(LC) vector was administered (1011 genome copies) to male C57BL/6 mice via intrapleural or intravenous routes, and serum titers of |[alpha]|PA measured at various times. Mice administered the vector via the intrapleural route had serum |[alpha]|PA titers of 2400 |[plusmn]| 400, 6000 |[plusmn]| 1300, and 9700 |[plusmn]| 1200 at 1, 2, and 4 wk, respectively. Intravenous administration resulted in serum |[alpha]|PA titers of 2200 |[plusmn]| 100, 5400 |[plusmn]| 450, and 6700 |[plusmn]| 240 at 1, 2, and 4 wk, respectively. These data indicate that passive immunization with an AAVrh10CMV(HC)2A(LC) vector may be a convenient and highly effective approach for rapidly and persistently protecting susceptible populations against anthrax.