Dependence of tumor-forming capacities of cells transformed by recombinants between adenovirus types 5 and 12 on expression of early region 1

Infection and transformation of mouse cells by human adenovirus type 2

The susceptibilities of C57Bl/6J mouse embryo fibroblasts (MEF) and baby mouse kidney (BMK) cells to infection or transformation by adenovirus type 2 have been compared to those of rat embryo fibroblasts (REF). Both MEF and BMK cells were some 10-fold less permissive to replication of the virus than were REF cells, even though similar fractions of all three cell types, a maximum of 50-60%, produced viral tumor and structural antigens. This observation suggests that a very late step in adenovirus production, such as assembly or maturation, occurs much less efficiently in mouse cells than it does in rat cells. No significant differences in the frequencies of transformation, as assayed by the appearance of foci of morphologically transformed cells, were observed following transfection of adenovirus type 2 or type 5 DNA into the three cell types. However, it proved extremely difficult to establish permanent lines of adenovirus-transformed mouse cells: only 2 of more than 100 attempts were successful, compared to a success rate of close to 100% with adenovirus type 2-transformed REF or SV40-transformed MEF or BMK cells. The two lines of type 2 adenovirus-transformed MEF that were established have been shown to retain and express viral genetic information.

Construction and characterization of an adenovirus type 5/adenovirus type 12 recombinant virus

We have constructed an adenovirus type 5 (Ad5) recombinant virus in which the early region 1b (E1b) of the nononcogenic Ad5 is replaced by the E1b region of the highly oncogenic Ad12. Analysis of cells lytically infected with the recombinant virus showed that both the Ad5 E1a genes and the Ad12 E1b genes are faithfully expressed. The recombinant virus replicates efficiently in human embryonic kidney cells and in HeLa cells, indicating that the Ad12 E1b region can fully replace the Ad5 E1b region in lytic infection. Inoculation of the Ad5/Ad12 hybrid virus into newborn hamsters did not result in development of tumors. This shows that the E1b region of Ad12, previously shown to be responsible for the high oncogenic potential of Ad12-transformed cells in nude mice is not capable of converting the nononcogenic Ad5 into an oncogenic virus.

Isolation of a nondefective recombinant between adenovirus type 5 and early region 1A of adenovirus type 12

A nondefective recombinant between adenovirus type 5 (Ad5) and type 12 (Ad12), rc-1 (Ad5 dl312, carrying the Ad12 E1A gene), was isolated from hamster cell foci transformed by a defective recombinant, rcB-1 (dl312, carrying the Ad12 E1 gene). The recombinant rc-1 grew in human embryo kidney and KB cells in the absence of helper and synthesized Ad12 T antigen g, the product of the E1A gene. The genome of rc-1 has a deletion between 79.9 and 82.5 map units of Ad5 dl312 DNA with an insertion of 0.1 to 5.5 map units of Ad12 DNA at the deletion site. The mRNAs of Ad12 E1A were transcribed from the Ad12 E1A promoter, and unusual RNAs were abundantly transcribed from the Ad5 E3 promoter on the opposite strand. The frequency of cell transformation with rc-1 was lower than those with Ad5 and Ad12 wild types.

Expression of adenovirus type 12 early region 1 in KB cells transformed by recombinants containing the gene

The adenovirus type 12 (Ad12) early region 1 (E1) gene was introduced into KB cells by using a dominant selection vector, pSV2-gpt, and over 80 Gpt+ KB cell clones were established. Three types of recombinant DNAs (gAE1A, gARC, and gABA) were constructed. They contained the AccI-H, EcoRI-C, and BamHI-A fragments, respectively, of Ad12 DNA in pSV2-gpt. Five of 50 (10%) gABA-transformed cell clones, 12 of 18 (67%) gAE1A-transformed cell clones, and 10 of 18 (56%) gARC-transformed cell clones complemented the growth of Ad5 dl312 (deletion in E1A) and were designated as Gpt+ Ad+ cell clones. In these cell clones at their early passages, recombinant genome sequences were detected in cellular DNA and were expressed. T antigen g (the E1A gene product) was detected by immunofluorescence. The Gpt+ Ad+ cell clones supported the growth of Ad5 deletion mutants in parallel with the expression of Ad12 E1A or E1A plus E1B genes. After infection of Gpt+ Ad+ cell clones with Ad5 dl312, the early genes of dl312 were efficiently transcribed, indicating the expression of the pre-early function of the Ad12 E1A gene. Two clones each from gAE1A-,gARC-, and gABA-transformed cells were subcultured for a long period to determine the stability of the transfecting DNAs. Subculture in a nonselective medium resulted in cells which lost the transfecting DNAs. Subculture in a selective medium resulted in the selection of cells which maintained the gpt gene expression but lost the Ad12 gene expression. These results indicate that the transfecting DNA is present in an unstable state in KB cells.

Adenovirus-induced alterations of the cell growth cycle: a requirement for expression of E1A but not of E1B

Mutants dl312, dl314, hr1, and hr3 with mutations in region E1A of adenovirus type 5 were defective for the induction of cell cycle abnormalities detectable by flow cytometry, cell DNA replication, thymidine kinase production, and chromosome aberrations and did not synthesize the viral DNA-binding protein (E2A) in rat cells. dl311, a leaky E1A mutant, induced cell cycle effects at high multiplicity in only one of three experiments, and synthesized the DNA-binding protein. hr7 (E1B) gave a wild-type response in all tests. dl313 was also positive in all tests, although it induced fewer polyploid cells than did wild-type virus, probably because of the leftward extension of the dl313 E1B deletion into E1A. sub315 and sub316, with mutations which also span the E1A-E1B border, synthesized DNA-binding protein, but caused no cell cycle alterations detectable by flow cytometry in rat or mouse cells. Although the participation of other viral early regions cannot be completely excluded, our results suggest that alteration of cell cycle progression is a direct effect of E1A unrelated to its control of other viral early regions, and may be the function of E1A in transformation.