Group B coxsackieviruses (CVB) trigger human myocarditis, while human adenovirus type 2 (Ad2) is implicated as an agent of this disease. demonstrated production of the expected capsid protein. Viral proteins were detected earlier and in approximately fourfold greater amounts in CVB3-PL2-Ad2L1-infected HeLa cells Pluripotin than in CVB3/0-infected cells. Cleavage of the CVB3-PL2-Ad2L1 polyprotein by 2Apro was slowed, accompanied by an accumulation of the fusion 1D-L1 loop protein. Reverse transcription-PCR sequence analysis of CVB3-PL2-Ad2L1 RNA demonstrated that the Ad2 hexon polypeptide coding sequence was maintained in the chimeric viral genome through at least 10 passages in HeLa cells. Mice inoculated with CVB3-PL2-Ad2L1 demonstrated a brief viremia with no replication detectable in the heart but prolonged replication of virus in the pancreas in the absence of pathologic changes in either organ. CVB3-PL2-Ad2L1 induced binding and neutralizing anti-Ad2 antibodies, in addition to antibodies against CVB3 in mice. CVB3-PL2-Ad2L1 was used to challenge mice previously inoculated with CVB3/0 and with preexisting anti-CVB3 neutralizing-antibody titers; anti-Ad2 neutralizing and binding antibodies were induced in these mice at higher levels than in mice without anti-CVB3 immunity. The data demonstrate that a CVB vector can stably express an antigenic polypeptide of Ad2 from within the CVB open reading frame that results in the induction of protective immune responses against both viruses. The six serotypes of the group B coxsackieviruses (CVB1 to CVB6) are human enteroviruses (family ). Coxsackievirus capsids are 29-nm-diameter icosahedral structures with the typical enterovirus canyon-like depressions surrounding the fivefold axes (46, 47), which, by analogy to polioviruses and rhinoviruses, are binding sites for the cell membrane receptor human coxsackievirus adenovirus (Ad) receptor (HCAR) (9, 12, 60). The CVB genome encodes four capsid Gja8 and seven nonstructural proteins, including two proteases, within a single open reading frame (ORF) with a coding capacity of 2,185 amino acids (16, 37, 63). While diverse viruses can cause human myocarditis (11), the CVB serotypes are those most commonly implicated. CVB have been isolated from hearts of pediatric patients with myocarditis (20, 21, 35, 66), enteroviral RNA has been detected by reverse transcription-PCR (RT-PCR) and in situ hybridization in approximately 20 to 25% of heart samples from patients with myocarditis or Pluripotin dilated cardiomyopathy (reviewed in references 5 and 41), and murine models of experimental CVB-induced myocarditis exist that recapitulate many aspects of the human disease counterpart (reviewed in references 27, 33, 34, and 67). More recently, human Ad DNA has been detected in hearts of patients with myocarditis (40), with subsequent sequence analysis of the amplimers from diseased hearts shown to be consistent with infections by Ad2 (53). There are no commercially available vaccines against either CVB or Ad. The small enterovirus genome is limited in terms of being able to add foreign sequences to it for successful expression. The crystallographic solution of poliovirus and rhinovirus capsid structures led to an early focus on expressing small peptides in external loops of capsid proteins (3, 18, 25, 29), but such inserts were limited to short peptides and chimeric viral genomes were generally unstable. Expression of foreign sequences within the enterovirus ORF, utilizing the ability of viral proteases to cleave both in and in to process the foreign polypeptide, were found to alleviate in part the size limitation of clonable inserts. Two sites within the enterovirus genome have been identified as useful, i.e., those immediately upstream of the start of translation and at the junction of the Pluripotin sequences encoding the capsid protein P-1D (1D) and the protease P-2A (2Apro). Although success was originally reported at the site upstream of the enteroviral ORF (68), insertions at this site tend to delete readily; expression of simian immunodeficiency virus (SIV) polypeptides suggested that this insertion site was less preferable, on the basis of stability and chimeric virus replication rate, than that between the 1D and the 2Apro (59). Once deletion of the inserted coding sequence Pluripotin from the enterovirus genome has occurred, the selection of the more efficiently replicating deleted genome (parental or near parental) quasispecies to become the dominant population can be rapid (48). Expression vectors that utilize the 1D/2Apro cloning site rely on 2Apro to cleave autocatalytically between its amino terminus and the carboxyl terminus of the inserted polypeptide and then to cleave the enterovirus capsid protein 1D from the foreign polypeptide sequence in haplotype) with similar results,.