Functional interaction of the adenovirus IVa2 protein with adenovirus type 5 packaging sequences

Abstract

Adenovirus type 5 (Ad5) DNA packaging is initiated in a polar fashion from the left end of the genome. The packaging process is dependent on the cis-acting packaging domain located between nucleotides 230 and 380. Seven AT-rich repeats that direct packaging have been identified within this domain. A1, A2, A5, and A6 are the most important repeats functionally and share a bipartite sequence motif. Several lines of evidence suggest that there is a limiting trans-acting factor(s) that plays a role in packaging. Both cellular and viral proteins that interact with adenovirus packaging elements in vitro have been identified. In this study, we characterized a group of recombinant viruses that carry site-specific point mutations within a minimal packaging domain. The mutants were analyzed for growth properties in vivo and for the ability to bind cellular and viral proteins in vitro. Our results are consistent with a requirement of the viral IVa2 protein for DNA packaging via a direct interaction with packaging sequences. Our results also indicate that higher-order IVa2-containing complexes that form on adjacent packaging repeats in vitro are the complexes required for the packaging activity of these sites in vivo. Chromatin immunoprecipitation was used to study proteins that bind directly to the packaging sequences. These results demonstrate site-specific interaction of the viral IVa2 and L1 52/55K proteins with the Ad5 packaging domain in vivo. These results confirm and extend those previously reported and provide a framework on which to model the adenovirus assembly process.

FIG. 1.

Ad5 packaging sequences. (A) Schematic representation of the left end of the Ad5 genome. Nucleotide coordinates, relative to the left terminus, are indicated, and the inverted terminal repeat (ITR) is represented by a gray box. A-repeats 1 to 7 are represented by arrows between nt 230 and 380. The E1A transcription initiation site is located at nt 499. The packaging repeat consensus motif is shown under the diagram. (B) Sequences surrounding A-repeats 1 and 2 and site-specific point mutants. A-repeats 1 and 2 are indicated by bold lines above the sequence. Site-specific point mutations in these A-repeats are shown with arrows; the nucleotide(s) changed and the mutant designation are shown.

FIG. 2.

Single-step growth curve analysis of mutant viruses. (A) N52.E6 cells were infected with the different viruses indicated and subsequently washed. Total cellular lysates were prepared at 4, 24, 48, and 72 h postinfection. The yield of infectious virus was measured by plaque assays in N52.E6 cells. WT, wild type. (B) N52.E6 cells were coinfected with PM2 and wild-type Ad5, and packaging efficiency was determined. The replication ratio is the level of PM2 viral DNA divided by the level of total viral DNA in coinfected N52.E6 cells. The packaging ratio is the level of packaged PM2 DNA divided by the level of total viral DNA packaged. “% Packaging efficiency” is the percentage of PM2 packaged compared to that of the coinfecting wild-type virus, corrected for replication. Results of two independent packaging assays of PM2 are shown. Exp, experiment.

FIG. 3.

Binding of IVa2 to wild-type (WT) and mutant minimal packaging domains. (A) The top line shows the sequence of the wild-type A1-A2 region; packaging consensus sequences are boldfaced and underlined. Point mutations introduced into A-repeat 1 or 2 are shown below, with the mutated nucleotides boldfaced. (B) Gel mobility shift assays were performed using Ad-infected cell nuclear extracts and a dimeric A1+2-WT (lanes 1, 2, 7, and 11), PM2 (lanes 3 and 4), PM12 (lanes 5 and 6), PM3 (lane 8), PM-CG (lane 9), PM13 (lane 10), PM3/CG-2 (lane 12), PM3-1 (lane 13), PM3-2 (lane 14), PM-CG-1 (lane 15), or PM-CG-2 (lane 16) probe. Several binding activities specific for IVa2 are evident with the wild-type probe and are indicated by arrows (lane 1, complexes 1, 2, and 3). An anti-IVa2 monoclonal antibody (mAb) was added to the binding reaction mixtures in lanes 2, 4, and 6. (C) Schematic depiction of the basis for the formation of IVa2 complexes 1, 2, and 3 on A-repeats 1 and 2. IVa2 complex 1 forms over the CG motif of A-repeat 1. Complex 2 additionally contains an infection-specific protein(s) bound to the TTTG motif of A-repeat 2, and complex 3 likely contains an additional IVa2 protein bound to the CG motif of A-repeat 2.

FIG. 4.

CHIP analysis of IVa2 and L1 52/55K protein binding to Ad genomes containing or lacking the packaging domain in vivo. (A) Schematic depiction of the parent virus containing a synthetic packaging domain (dimer of A-repeats 5 to 7 in place of nt 194 to 814 [26] flanked by loxP sites [designated Ad5ψ-loxP]). Infection of N52.E6-Cre cells results in the recombination (floxing) of the packaging sequences from the viral genome. Primer pair 7+8 is located outside of the floxed interval and is common to both viral genomes, while primer pair 5+6 is specific to the unfloxed genome and was used to determine the efficiency of recombination in viral DNAs present in the starting chromatin preparation. ITR, inverted terminal repeat. (B) CHIP analysis using chromatin isolated from N52.E6-Cre cells infected with Ad5-194/814+A5-7² or Ad5-194/814+A5-7²+loxP (Ad5ψ-loxP), anti-IVa2 and anti-L1 52/55K sera, and Q-PCR with a primer pair (7+8) corresponding to the left-end packaging region. The concentration of DNA recovered (copies per nanogram) was determined by extrapolation of the crossing point from a standard curve for each specific primer pair. The data were evaluated and are presented as described for Table 1.

FIG. 5.

CHIP analysis of IVa2 and L1 52/55K protein binding to Ad genomes carrying left-end or right-end packaging domains in vivo. (A) Schematic depiction of the parent viruses that carry left-end (dl309) or right-end (in340-Δ11) packaging domains. Primer pair 1+2 is located within the packaging region and is common to both viral genomes. Primer pair 9+10 is specific to the left end of the genome outside the packaging domain, and primer pair 3+4 is common to both viruses and specific to the right end of the genome, at a location to the left of where the packaging domain is inserted into virus in340-Δ11. ITR, inverted terminal repeat. (B) CHIP analysis using chromatin isolated from N52.E6 cells infected with dl309 or in340-Δ11, anti-IVa2 and anti-L1 52/55K sera, and Q-PCR with primer pairs 1+2, 9+10, and 3+4. The data were evaluated and are presented as described for Table 1.

FIG. 6.

CDP binding to A-repeats does not correlate with packaging efficiency. Gel mobility shift assays were performed by using the CDP present in fractionated HeLa cell nuclear extracts and a dimeric A1/2-WT (lane 1), LS (lane 2), CG (lane 3), PM2 (lane 4), PM3 (lane 5), PM12 (lane 6), or PM13 (lane 7) probe. Arrow indicates CDP binding activity.