Base analogs are derivatives of the normal DNA bases, which may mimic to varying extents the properties of the normal bases. As such, they have the ability to interfere with normal nucleotide metabolism and exert a variety of toxic and mutagenic effects. One example of their application as a toxic compound is usage as antiviral or antitumor agents. Our main interest in base analogs centers around their properties as mutagens, and we use them as probes for studying the various ways by which cells either make mutations or try to avoid them, using the bacterium E. coli as a model system. Specifically, we use purine analogs such as 2-aminopurine (AP), 6-hydroxylaminopurine (HAP), and 2-amino-6-hydroxylaminopurine (AHAP), and the pyrimidine analog BN4-hydroxycytosine to investigate (i) the mechanisms by which these analogs are converted to mutationally active forms (e.g., a modified dNTP) and (ii) the protective mechanisms that cells use to avoid or diminish analog-induced mutagenesis. The basic approach in our studies is a genetic one, in which analog-induced toxicity or mutagenesis is studied in a variety of E. coli genetic mutants. These mutants are (i) affected in established pathways of DNA replication, repair, or nucleotide metabolism (to delineate the role of these systems) or (ii) newly isolated mutants generated on the basis of their altered response to these agents (to discover novel pathways for analog toxicity or mutagenesis). The most striking discovery has been the identification of a hitherto unknown detoxification pathway for N-hydroxylated bases, which requires the Molybdenum Cofactor (MoCo). In a search for the responsible MoCo-dependent activity, we have systematically deleted all the known and putative molybdoenzymes from E. coli. No base-analog sensitivity was associated with any of these mutants (including one lacking all combined activities), indicating that the base-analog detoxifying activity must result from an as yet unidentified group (family) of molybdoenzymes. We also have discovered that the novel activity does not require the MGD (Molybdopterin Guanine Dinucleotide) form of MoCo, which is commonly used by the bacterial enzymes. Instead, the MPT (molybdopterin) form of MoCo, commonly used in eukaryotic systems, is sufficient. Searches for novel HAP-sensitive mutants of E. coli identified the ycbX and yiiM open reading frames as determinants for base-analog resistance. Based on these and other findings, we have proposed that the YcbX and YiiM proteins represent MPT-containing enzymes that are capable of detoxifying HAP, presumably by reduction to adenine. Consistent this this hypothesis, we have been able to show that cell-free extracts of E. coli are indeed capable of converting HAP to adenine. This reaction does not take place in strains lacking MoCo or ycbX yiiM deficient strains. We have also recently identified additional components of the electron transfer pathway. In case of the YcbX protein, electrons are transferred from NADPH to HAP via CysJ flavin reductase. Evidence has been obtained that YcbX and CysJ form a physical complex. In the case of YiiM protein, electrons are transferred via Fre flavin reductase. One additional focus is the precise mechanism by which MoCo-deficient cells are actively killed by exposure to HAP, for which experiments have been initiated.
