Expression of Viral DNA in Adenovirus Type 12-Transformed Cells, in Tumor Cells, and in Revertants

Abstract

The expression as cytoplasmic RNA of integrated human adenovirus type 12 (Ad12) DNA in transformed and tumor cell lines and in revertants was investigated. The transformed and tumor cells contained multiple copies of the viral genome, 3 to 22 copies per cell in different cell lines. The integrated Ad12 DNA molecules persisted intact or nearly intact and in most cases colinear with the virion DNA. In the revertant cell lines, which were derived from cell line T637 (22 copies of Ad12 DNA per cell), all of the Ad12 DNA molecules were lost (line F10) or only one copy and a fraction of a second copy persisted (line TR12). The size classes and map locations of Ad12-specific cytoplasmic RNAs in three Ad12-transformed hamster cell lines (T637, HA12/7, and A2497-3), in two revertant lines (F10 and TR12), in one Ad12-induced hamster (CLAC3), and in one rat brain tumor line (RBT12/3) were determined. Cytoplasmic RNA from uninfected B3 hamster cells and from human KB cells productively infected with Ad12 served as controls. In the latter control experiments, the RNA was isolated early or late postinfection. With respect to the amounts of Ad12-specific RNAs detected in cytoplasmic RNA from various Ad12-transformed or Ad12-induced tumor cell lines, we could not establish any correlations to the number of Ad12 genome copies integrated into the cellular DNAs. Thus, the expression of the integrated viral genomes in these lines was regulated by mechanisms more complicated than simple gene dosage effects. Using cloned fragments of Ad12 DNA as hybridization probes, we analyzed the cytoplasmic RNAs from the cell lines mentioned by electrophoresis on agarose gels, blotting, and DNA-RNA hybridization. For each transformed and tumor cell line, except for the revertants, several size classes of Ad12-specific cytoplasmic RNA were detected for the early E1, E2, and E4 regions of Ad12 DNA. Some of these size classes were similar but not identical to those observed in cytoplasmic RNA isolated early from human KB cells productively infected with Ad12. Only cell lines A2497-3, T637, and RBT12/3 contained several size classes of cytoplasmic RNA homologous to the E3 region of Ad12 DNA. Weak homologies to the E1 region of Ad12 DNA were also detected in the revertant lines F10 and TR12. Late regions of Ad12 DNA were expressed as cytoplasmic RNA in cell lines CLAC3 and RBT12/3. Weak homologies were detected between certain segments of the Ad12 genome (the Eco RI-B, -C, and -D fragments) and the cytoplasmic RNA from uninfected hamster cells. These homologies had no apparent counterpart at the level of DNA, perhaps because these homologies could be detected only due to an overrepresentation of RNA sequences. In preliminary experiments, we failed to detect the expression as cytoplasmic RNA of the so-called virus-associated RNA in transformed and tumor cell lines. Virus-associated RNA represents a population of low-molecular-weight RNAs that map at around 30 fractional length units on the viral genome.

The unmethylated state of the promoter/leader and 5′-regions of integrated adenovirus genes correlates with gene expression.

“…In cell lines T637, HA12/7, and A2497-3 the early regions El, E2a, and E4, and in lines T637 and A2497-3 the early region E3 as well, are expressed as mRNA. Region E2b and the late regions of Adl2 DNA are not expressed (Ortin et al, 1976;Schirm and Doerfler, 1981;Esche and Siegmann, 1982). Thus, the inverse correlation between gene expression and DNA methylation described previously is upheld by the present, much more refined, mapping results.…”

“…We have now precisely mapped many of the methylated and the unmethylated 5'-CCGG-3' (HpaII/MspI) sites in the Adl2 genomes integrated in the three Ad12-transformed hamster cell lines T637, HA12/7, and A2497-3 (Sutter et al, 1978;Doerfler, 1979, 1980;Stabel et al, 1980). With few exceptions, the early viral genes are expressed in these lines and the late genes are inactive (Ortin et al, 1976;Schirm and Doerfler, 1981;Esche and Siegmann, 1982). We find that the promoter/leader and 5'-regions of early Adl2 genes carry unmethylated HpaII sites when these genes are expressed.…”

“…As a control, Adl2 virion DNA, which was essentially unmethylated (Gunthert et al, 1976), was treated in the same way and was co-electrophoresed in many of the experiments. Finally, the results on methylation ( Figure 5) were correlated with the expression of specific viral DNA fragments in the transformed cell lines investigated (Ortin et al, 1976;Schirm and Doerfler, 1981). Mapping of methylated HpaII sites in integrated viral DNA in Adl2-transformed lines T637, HA 12/7, and A2497-3…”

“…Expression of Adl2-specific RNA from early regions El and E3 in cell lines A2497-3, T637, and HA12/7. Cytoplasmic RNA from the Adl2-transformed cell lines as indicated, from B3 hamster cells and from KB cells productively infected with Adl2 (8 h post-infection) was isolated, fractionated by electrophoresis on 1 %o formamide agarose slab gels and transferred to nitrocellulose filters as described earlier (Schirm and Doerfler, 1981). Adl2-specific RNA sequences were visualized by hybridization to the cloned EcoRI-C (a) or the PstI-H (b) fragments derived from the El or E3 regions of Adl2 DNA, respectively (c.f.…”

“…The E3 region is not expressed as mRNA in cell line HA12/7 (Ortin et al, 1976;Schirm and Doerfler, 1981) and in vitro translation experiments using RNA selected on Ad12 DNA fragments corresponding to the E3 region have not revealed Adl2-specific products (Esche and Siegmann, 1982). By using the cloned PstI-H or PstI-G fragment of Adl2 DNA (data obtained with the latter fragment not shown) as hybridization probe with cytoplasmic RNA from HA12/7 cells which was electrophoresed and transferred to nitrocellulose filters (Alwine et al, 1977;Schirm and Doerfler, 1981), we have confirmed the absence of E3-specific cytoplasmic RNA in extracts of HA12/7 cells (Figure 6b), whereas the El region is expressed in all three transformed cell lines (Figure 6a). We thus conclude that an inverse correlation exists between the levels of DNA methylation at the promoter/leader and the 5'-regions of specific groups of integrated Adl2 genes and the extent to which the regions regulated by these promoters are expressed as mRNA.…”