Although it has been demonstrated that the adenovirus IVa2 protein binds to
the packaging domains on the viral chromosome and interacts with the viral
L1 52/55-kDa protein, which is required for viral DNA packaging, there has
been no direct evidence demonstrating that the IVa2 protein is involved in
DNA packaging. To understand in greater detail the DNA packaging mechanism
s of adenovirus, we have asked whether DNA packaging is serotype or subgrou
p specific. We found that Ad7 (subgroup B), Ad12 (subgroup A), and Ad17 (su
bgroup D) cannot complement the defect of an Ad5 (subgroup C) mutant, pm800
1, which does not package its DNA due to a mutation in the L1 52/55-kDa gen
e. This indicates that the DNA packaging systems of different serotypes can
not interact productively with Ad5 DNA. Based on this, a chimeric virus con
taining the Ad7 genome except for the inverted terminal repeats and packagi
ng sequence from Ad5 was constructed. This chimeric virus replicates its DN
A and synthesizes Ad7 proteins, but it cannot package its DNA in 293 cells
or 293 cells expressing the Ad5 L1 52/55-kDa protein. However, this chimeri
c virus packages its DNA in 293 cells expressing the Ad5 IVa2 protein. Thes
e results indicate that the IVa2 protein plays a role in viral DNA packagin
g and that its function is serotype specific. Since this chimeric vit us ca
nnot package its own DNA, but produces all the components for packaging Ad7
DNA, it may be a more suitable helper virus for the growth of Ad7 gutted v
ectors for gene transfer.