Since most chemical carcinogens react with DNA and are mutagenic, interactions with DNA
have been viewed as the most important reactions of these agents with cellular macromolecules.
Reaction of chemical carcinogens with DNA is the simplest mechanism that explains
the induction of a heritable change in a cell leading to malignant transformation; thus many
investigators view this as the most plausible mechanism for initiation of carcinogenesis. Representative
agents from virtually all classes of chemical carcinogens have been shown to affect
DNA in some way, and a number of distinct biochemical-reaction products have been identified
after treatment of cells in vivo or in culture with carcinogenic agents. The principal reaction products of the nitrosamines and similar alkylating agents with DNA areN-7andO6 guanine derivatives.
However, the extent of O6 alkylation of DNA guanine residues correlates better with mutagenic and carcinogenic activity than the quantitatively greater N-7 alkylation of guanine residues . Reactions also occur with other DNA bases, and these may be important in subsequent mutagenic or carcinogenic events. Aflatoxin forms adducts of guanine at the N-7 position after metabolic activation. The principal reaction product of AAF with cellular DNA is the C-8 position ofguanine,justasit is NA.Othercarcinogenic
aromatic amines, such as N-methyl-4- aminoazobenzene, also produce C-8 substituted guanine residues as their major nucleic acid reaction product (adduct). Polycyclic aromatic hydrocarbons, after activation, also react with DNA and RNA, forming adducts involving the 2-amino group of guanine, but other reaction
products derived from guanine, adenine, and cytosine have been observed as well.7 The potential biological consequences of DNA base–adduct formation by chemical carcinogens are several. In some cases, it may stabilize an intercalation reaction in which the flat planar rings of a polycyclic hydrocarbon are inserted between the stacked bases of doublehelical DNA and distort the helix, leading to a
frame-shift mutation during DNA replication past the point of the intercalation.19 Alkylated bases in DNA can mispair with the wrong base during DNA replication—for example, O6 methylguanine pairs with thymine instead of cytosine during DNA replication, leading to a base transition (i.e., GC?AT) type of mutation during the next round of DNA replication.20 Many of the base adducts formed by carcinogens
involve modifications of N-3 or N-7 positions on purines that induce an instability in the glucosidic bond between the purine base and deoxyribose, resulting in loss of the base and creation ofan apurinic site inDNA.21 This ‘‘open’’ apurinic site can then be filled by any base, but most commonly by adenine, during subsequent DNAreplication.
This substitution can result in a base transition (purine–pyrimidine base change, but in the same orientation, e.g., GC?AT) or a base transversion (inverted purine–pyrimidine orientation, e.g., GC?TA). Finally, interaction with some carcinogens has been shown to favor a conformational transition of DNA from itsusual double-helical B form to a Z-DNA form. This could alter the transcribability of certain
genes, since B?Z conformational transitions are thought to be involved in regulating chromatin
structure. Another interesting point is that interaction of chemical carcinogens with DNA or chromatin
does not appear to be a random process. For example, when the ultimate carcinogen of benzo[
a]pyrene, that is, its diol epoxide metabolite, is reacted with cloned chicken b-globin DNA, it
preferentially binds in a 300–base pair sequence immediately 50 to the RNA cap site.22 Since this
region is thought to contain sequences involved in regulating gene transcription, its alteration by
a chemical carcinogen could change the function of genes downstream from the regulatory
sequences. Moreover, treatment of the large polythene chromosomes of Chironomus with the
ultimate carcinogen benzo(a)pyrene diol epoxide in vitro or administration of the parent unmetabolized compound in vivo to Chironomus larvae demonstrates that the carcinogen binds preferentially to areas most active in gene transcription.23 DNA in transcribing regions associated with the nuclear matrix also appears to be a preferential target for carcinogen binding. Taken together, these data indicate that the
specificity of carcinogen binding is determined to some extent by the base sequence of DNA, its
location within the nucleus (e.g., association with nuclear matrix), and the structure of chromatin,
with active, ‘‘open’’ sites being favored.
