Induction of a step in carcinogenesis that is normally associated with mutagenesis by nonmutagenic concentrations of 5-azacytidine

The controversial denouement of vertebrate DNA methylation research

The study of the biological role of DNA methylation in vertebrates has involved considerable controversy. Research in this area has proceeded well despite the complexity of the subject and the difficulties in establishing biological roles, some of which are summarized in this review. Now there is justifiably much more interest in DNA methylation than previously, and many more laboratories are engaged in this research. The results of numerous studies indicate that some tissue-specific differences in vertebrate DNA methylation help maintain patterns of gene expression or are involved in fine-tuning or establishing expression patterns. Therefore, vertebrate DNA methylation cannot just be assigned a role in silencing transposable elements and foreign DNA sequences, as has been suggested. DNA methylation is clearly implicated in modulating X chromosome inactivation and in establishing genetic imprinting. Also, hypermethylation of CpG-rich promoters of tumor suppressor genes in cancer has a critical role in downregulating expression of these genes and thus participating in carcinogenesis. The complex nature of DNA methylation patterns extends to carcinogenesis because global DNA hypomethylation is found in the same cancers displaying hypermethylation elsewhere in the genome. A wide variety of cancers display both DNA hypomethylation and hypermethylation, and either of these types of changes can be significantly associated with tumor progression. These findings and the independence of cancer-linked DNA hypomethylation from cancer-linked hypermethylation strongly implicate DNA hypomethylation, as well as hypermethylation, in promoting carcinogenesis. Furthermore, various DNA demethylation methodologies have been shown to increase the formation of certain types of cancers in animals, and paradoxically, DNA hypermethylation can cause carcinogenesis in other model systems. Therefore, there is a need for caution in the current use of demethylating agents as anti-cancer drugs. Nonetheless, DNA demethylation therapy clearly may be very useful in cases where better alternatives do not exist.

DNA hypomethylation, cancer, the immunodeficiency, centromeric region instability, facial anomalies syndrome and chromosomal rearrangements

Inadequate attention has been paid to the frequent and often extensive cancer-associated DNA hypomethylation. This hypomethylation usually includes undermethylation of certain DNA repeats in constitutive heterochromatin, although it is not limited to such sequences. Many cancers display an overall deficiency in the levels of genomic 5-methylcytosine compared to a variety of normal postnatal somatic tissues. The immunodeficiency, centromeric region instability, facial anomalies (ICF) syndrome, a rare recessive DNA methyltransferase deficiency disease, results in a small decrease in the extent of global genomic methylation. In ICF, DNA hypomethylation is targeted to the satellite DNA in juxtacentromeric (centromere-adjacent) heterochromatin of chromosomes 1 and 16 (1qh and 16qh), which are prone to rearrangements in ICF lymphoid cells. Also, 1qh and 16qh DNA sequences frequently are hypomethylated in human cancers and rearrangements in their vicinity are overrepresented in cancers. These often lead to chromosome arm imbalances and gene dosage imbalances that could participate in carcinogenesis. Studies of ICF cells suggest that hypomethylation in the normally highly methylated 1qh and 16qh regions predisposes to heterochromatin decondensation in these regions, which in turn leads to elevated levels of rearrangements. Studies of ICF cells also suggest that some of these rearrangements, namely multiradial chromosomes with multiple arms joined in the pericentromeric region, may be unstable intermediates in formation of more stable pericentromeric rearrangements in cancer. Microarray gene expression analysis on ICF and normal lymphoblastoid cell lines suggests that this hypomethylation also may affect gene expression elsewhere in the genome.

5-Azacytidine-induced 6-thioguanine resistance at the gpt locus in AS52 cells: cellular response

Treatment of AS52 cells with 5-azacytidine resulted in an induction of 6-thioguanine-resistant [6TG] colonies, which reached a maximum by an expression time of 9 days. Dose responses for both cytotoxicity and mutation induction were determined following treatment with 5-azacytidine. At 20 microM treatment, 5-azacytidine exposure resulted in about 50% survival. Mutant frequency reached a maximum of 10 microM. At concentrations between 10 and 20 microM, 5-azacytidine was a potent mutagen but did not exhibit a dose response. Although many compounds both induce cell death and affect the growth rate of cells, 5-azacytidine specifically induced cell death and did not affect the doubling time of the surviving treated cell population.

An in situ protocol for measuring the expression of chemically-induced mutations in mammalian cells

The generation of expression curves and the evaluation of mutagenic responses of mammalian cells using standard mutagenesis assays can be inaccurate because mutant and wild-type cells are usually mixed during the expression phase. If some mutant progenitors or mutants grow more slowly than the wild-type cells during the expression period, there will be a decrease in the mutant to wild-type ratio with time and the mutant fraction will not accurately represent the number of mutational events that occurred. The mutant fraction may also inaccurately assess the number of mutations if these mutations are expressed over a number of generations during the time before selection. We previously showed that recovery of L5178Y mouse cell mutants is not complete when mutations are allowed to express in suspension because slowly growing mutants and/or mutant progenitors are diluted out during this time (Rudd et al., 1990). In order to more accurately quantitate the mutagenic response of the cells, we developed an in situ procedure which segregates and immobilizes cells during expression. Because of this immobilization, slowly growing mutant progenitors and mutants expressed at different times will have an equal probability of being scored as mutants. Thus, one mutation leads to one mutant colony and the measurement of the mutagenic response of the cells to the chemical accurately reflects the mutational events that occurred. We plated L5178Y tk+/- mouse cells in semisolid medium immediately after treatment. As the cells grew and formed microcolonies, the selective agent TFT was added as an overlay at specified times, permitting only TFTr cells to survive. In this procedure, each mutation was captured as an individual colony; consequently, the measured mutation fraction accurately reflected the mutational events that occurred at the selected locus. In addition, the induced mutant colonies arising in the agar are the result of independent mutational events. We previously described the in situ protocol for L5178Y cells and showed that the spontaneous mutation rate measured was 50-fold greater than when the cells expressed the phenotype in suspension (Rudd et al., 1990). From this we concluded that the slow growth phenotype was expressed before TFT resistance. In the present paper, we evaluate the effect of chemical treatment on the mutation fraction as a function of the time to TFT addition. Using the in situ protocol, we generated expression curves for three nucleotide analogs, 5-azacytidine, TFT and AraC. The numbers of TFTr colonies produced at various times after treatment indicated that chemically-treated cultures had higher mutation fractions than the solvent controls.(ABSTRACT TRUNCATED AT 400 WORDS)

5-Azacytidine and RNA secondary structure increase the retrovirus mutation rate

A broad spectrum of mutations occurs at a high rate during a single round of retrovirus replication (V.K. Pathak and H. M. Temin, Proc. Natl. Acad. Sci. USA 87:6019-6023, 1990). We have now determined that this high rate of spontaneous mutation can be further increased by 5-azacytidine (AZC) treatment or by regions of potential RNA secondary structure. We found a 13-fold increase in the mutation rate after AZC treatment of retrovirus-producing cells and target cells. The AZC-induced substitutions were located at the same target sites as previously identified spontaneous substitutions. The concordance of the AZC-induced and spontaneous substitutions indicates the presence of reverse transcription "pause sites," where the growing point is error prone. An analysis of nucleotides that neighbored substitutions revealed that transversions occur primarily by transient template misalignment, whereas transitions occur primarily by misincorporation. We also introduced a 34-bp potential stem-loop structure as an in-frame insertion within a lacZ alpha gene that was inserted in the long terminal repeat (LTR) U3 region and determined whether this potential secondary structure increased the rate of retrovirus mutations. We found a threefold increase in the retrovirus mutation rate. Fifty-seven of 96 mutations were deletions associated with the potential stem-loop. We also determined that these deletion mutations occurred primarily during minus-strand DNA synthesis by comparing the frequencies of mutations in recovered provirus plasmids containing both LTRs and in provirus plasmids containing only one LTR.

Sudden outbreak of a leukemia-like lesion in female CBA mice after repeated injections of 5-azacytidine

A total of 11 i.p. injections of 1 mg/kg 5-azacytidine given within 21 weeks induced a sudden outbreak of a leukemia-like disease in female CBA mice about 2 weeks after the last injection. The outbreak of disease was highly synchronous; 68% of animals who had survived the treatment period succumbed within a period of 5 days, 2/3 of them on a single day. The lesion had a characteristic form commonly presenting as a result of hemorrhagic effusion in the thoracic and/or abdominal cavities, associated with diffuse infiltration of lymphatic and fatty tissue by lymphoblasts. It appears that a critical combination of factors was responsible for the unexpected appearance of the lesion. Female BALB/c mice treated identically were not affected. The factors indicating that the lesion was induced by an epigenetic mechanism are discussed.

Effects of cytidine analogs on methylation of DNA and retrovirus induction

Several cytidine analogs with known mutagenic capability were tested for their effects on DNA methylation and on induction of endogenous murine retrovirus. For each of the compounds tested it was found that DNA methylation was inhibited at the same concentrations that were required to induce virus expression. With each compound it was observed that increased dose levels produced an increase in the ability to inhibit methylation and an increase in the ability to induce virus.

Clonal analysis of the malignant properties of B16 melanoma cells treated with the DNA hypomethylating agent 5-azacytidine

The role of DNA methylation in the generation of tumor cell variants with altered growth behavior has been investigated. Cultures of the clonally heterogeneous B16 melanoma cell line and a clonal population (B16-CL) derived from it were treated with the DNA hypomethylating agent, 5-azacytidine (5-Aza-CR). The tumorigenic and metastatic properties of (sub)clones isolated from these cultures before and after drug treatment were assayed by injection via multiple routes into syngeneic C57BL/6 mice using a range of cell doses. The rate of tumor growth was monitored following intrafootpad (i.f.p.) injection and the tumor incidence was calculated from the frequency of tumor formation at i.f.p. and supraclavicular subcutaneous (s.c.) sites. Formation of both spontaneous (i.f.p., s.c. inoculations) and experimental (intravenous (i.v.) inoculation) metastatic potential was also investigated. The most consistent effect of 5-Aza-CR was the introduction of heterogeneity with respect to the tumorigenic phenotype. The effect of 5-Aza-CR treatment on metastatic behavior was variable. The majority of tumor cell variants that arose following 5-Aza-CR treatment displayed decreased malignant potential and reduced DNA methylation levels relative to untreated control cells, but the correlation was not absolute. The decreases in DNA methylation levels induced by 5-Aza-CR were unstable and began to rebound within 1 week of drug treatment. The results of the current study indicate that although 5-Aza-CR can introduce significant shifts in the malignant properties of treated cells, the direction and magnitude of the induced alterations are not predictable and are influenced by a variety of experimental parameters including the starting tumor cell population, route of tumor cell inoculation, and the drug treatment protocol. In addition, because DNA methylation levels can rebound rapidly (days) it is difficult to correlate changes in this parameter with the observed alterations in malignancy, which can only be assessed in long-term biological assays (weeks).

The majority of independently transformed BHK cell clones share a single functional lesion which determines anchorage independence and influences tumorigenicity

Expression of the anchorage-independent transformed phenotype in BHK 21/13 cells generally behaves as a recessive trait. When chemically induced and spontaneously arising transformants are fused to the nontransformed parent line, transformation is initially suppressed, reappearing after extended growth of the hybrids. In this paper, complementation for the expression of anchorage independence was sought among a large group of such transformants, all independently derived from BHK 21/13 cells. Tumorigenicity studies on selected hybrids and parental lines indicated that the in vitro trait of anchorage independence is an accurate indicator of in vivo neoplasia for these cells. Seventeen of the 18 clones tested did not complement one or more of three tester strains. This result indicates that anchorage independence arose in these clones as a result of lesions in the same genetic function and suggests that the final step in the progressive changes of carcinogenesis may frequently be restricted to lesions at a single locus.