Activation of the major late promoter in adenovirus transformed cells by 5-azacytidine

Demethylation in the 5'-flanking region of mouse cellular retinoic acid binding protein-I gene is associated with its high level of expression in mouse embryos and facilitates its induction by retinoic acid in P19 embryonal carcinoma cells

The mouse cellular retinoic acid binding protein-I (CRABP-I) gene is specifically up-regulated by retinoic acid (RA) in P19 mouse embryonal carcinoma cells, and its expression in animals is spatially and temporally restricted to RA-sensitive tissues during embryonic development. This study demonstrates that, in adult mouse tissues and P19 cells where the expression of CRABP-I is detected at the basal level, the 5'-flanking region of the CRABP-I gene is hypermethylated at the C residues of all the Hpa II sites. Conversely, in mouse embryos during early stages of development when the expression of CRABP-I gene is detected at a much higher level, this region is demethylated at these Hpa II sites. In P19, enhancement on the RA-induced up-regulation of CRABP-I can be observed in cells treated with 5-azacytidine (5-AzaC) in conjunction with RA, where partial demethylation in the 5'-flanking region of CRABP-I gene is observed. Nuclear run-on experiments indicate that increased message levels of CRABP-I in P19 cells can be accounted for, at least partially, by increases in its transcription rates. The induction of retinoic acid receptor (RAR) beta by RA can also be enhanced by 5-AzaC, but to a much lesser degree. In contrast, all the Hpa II sites in the structural gene portion, at least in the first two exons, are fully demethylated at the C residues.

Establishment of a human cell line for the detection of demethylating agents

To detect xenobiotics affecting the DNA methylation pattern we have established a human clonal cell line (A4/4) derived from the 293 epithelial cells. This cell line is stable after more than 1 year in culture and we report here its characteristics. The A4/4 cells carry the Escherichia coli lacI gene and the symmetric lacO sequence upstream of the beta-galactosidase coding gene (lacZ). Both sequences have been independently integrated into the genomic DNA. The lacZ gene is under the control of the metal-inducible mouse metallothionein-I (mMT-I) promoter. Vector integration followed by cell ageing in culture gave rise to the methylation of the 5'CpG3' sites in the mMT-I promoter and the 5' part of the lacZ gene. The reactivation of the lacZ gene by 5-azacytidine (5-Aza-CR) treatment allowed a tremendous lacZ expression which is correlated with demethylation in the mMT-I promoter and its neighboring regions. This enhanced transcription is easily quantified by measuring the beta-galactosidase activity in cell extracts. 5-Aza-CR, the specific isopropyl beta-D-thiogalactoside inducer, and heavy metal ions added together trigger an increase of the beta-galactosidase activity up to 600-fold over the basal level. These cells can be used for a rapid assessment of the demethylating potential of environmental chemicals.

Genetic characterization of adenovirus transformed cell revertants resistant to methylglyoxal bis(guanylhydrazone): evidence for the involvement of three genetic loci

Five new independent rat somatic cell mutants resistant to the antileukemic drug methylglyoxal bis(guanylhydrazone) (MGBG) were isolated after mutagen treatment. The mutants were 7- to 10-fold more resistant to MGBG than were the parental wild-type cells. When the MGBG-resistant (MGR) mutants were exposed to the drug in the presence of Tween-80, a nonionic detergent, they became as sensitive (MGS) to MGBG as the wild-type cells, indicating that they were probably permeability mutants. Genetic analysis of hybrids between MGR mutants and wild-type cells showed that MGR and the nontransformed alleles to be recessive to the MGS (wild-type) and transformed phenotype, respectively. Complementation analysis of the seven mutants revealed three functional genetic units or loci responsible not only for the MGR phenotype but also for tumorigenicity as determined in nude mice. Only the MGS hybrids produced tumors in the nude mice, whereas the MGR hybrids and mutants did not. Our results suggest the existence of cellular membrane components that are responsible both for cellular tumorigenicity and resistance to MGBG.