Importance of minor-groove contacts for recognition of DNA by the binding domain of Hin recombinase

Incorporation of the DNA-cleaving moiety EDTA.Fe at discrete amino acid residues along a DNA-binding protein allows the positions of these residues relative to DNA bases, and hence the organization of the folded protein, to be mapped by high-resolution gel electrophoresis. A 52-residue protein, based on the sequence-specific DNA-binding domain of Hin recombinase (139-190), with EDTA at the NH2 terminus cleaves DNA at Hin recombination sites. The cleavage data for EDTA-Hin(139-190) reveal that the NH2 terminus of Hin(139-190) is bound in the minor groove of DNA near the symmetry axis of Hin-binding sites [Sluka, J. P., Horvath, S. J., Bruist, M. F., Simon, M. I., & Dervan, P. B. (1987) Science 238, 1129]. Six proteins, varying in length from 49 to 60 residues and corresponding to the DNA-binding domain of Hin recombinase, were synthesized by solid-phase methods: Hin(142-190), Hin(141-190), Hin(140-190), Hin(139-190), Hin(135-190), and Hin(131-190) were prepared with and without EDTA at the NH2 termini in order to test the relative importance of the residues Gly139-Arg140-Pro141-Arg142, located near the minor groove, for sequence-specific recognition at five imperfectly conserved 12-base-pair binding sites. Footprinting and affinity cleaving reveal that deletion of Gly139 results in a protein with affinity and specificity similar to those of Hin(139-190) but that deletion of Gly139-Arg140 affords a protein with altered affinities and sequence specificities for the five binding sites. It appears that Arg140 in the DNA-binding domain of Hin is important for recognition of the 5'-AAA-3' sequence in the minor groove of DNA. Our results indicate modular DNA and protein interactions with two adjacent DNA sites (major and minor grooves, respectively) bound on the same face of the helix by two separate parts of the protein.     

The effects of ethidium bromide and Mg++ ion on strand exchange in the Hin-mediated inversion reaction

The biochemical reaction of a site-specific recombinase such as Hin invertase or gammadelta resolvase starts with binding of the recombinase to its recombination site and cleavage of the DNA in the center of the site. This is followed by strand exchange and finally ligation of the ends of the recombined strands. Previous biochemical studies have shown that Hin invertase and gammadelta resolvase cannot proceed beyond DNA cleavage in the absence of Mg++ ion, indicating that these recombinases require Mg++ ion in the strand exchange process. We have observed that the intercalating agent, ethidium bromide (2 microM), does not interfere with DNA cleavage, but slows strand exchange in a concentration-dependent manner. Levels of Mg++ ion below 5 mM also slow strand exchange substantially. We infer that random intercalation of ethidium bromide inhibits unwinding of the double helix at the recombination site in the negatively supercoiled DNA and propose that Mg+ may be required for Hin to deform the secondary structure of B-DNA prior to strand exchange.     

Cleavage of DNA with methidiumpropyl-EDTA-iron(II): reaction conditions and product analyses

The synthesis of methidiumpropyl-EDTA (MPE) is described. The binding affinities of MPE, MPE.Ni(II), and MPE.Mg(II) to calf thymus DNA are 2.4 X 10(4) M-1, 1.5 X 10(5) M-1, and 1.2 X 10(5) M-1, respectively, in 50 mM NaCl, pH 7.4. The binding site size is two base pairs. MPE.Mg(II) unwinds PM2 DNA 11 +/- 3 degrees per bound molecule. MPE.Fe(II) in the presence of O2 efficiently cleaves DNA and with low sequence specificity. Reducing agents significantly enhance the efficiency of the cleavage reaction in the order sodium ascorbate greater than dithiothreitol greater than NADPH. At concentrations of 0.1-0.01 microM in MPE.Fe(II) and 10 microM in DNA base pairs, optimum ascorbate and dithiothreitol concentrations for DNA cleavage are 1-5 mM. Efficient cleavage of DNA (10 microM in base pairs) with MPE.Fe(II) (0.1-0.01 microM) occurs over a pH range of 7-10 with the optimum at 7.4 (Tris-HCl buffer). The optimum cleavage time is 3.5 h (22 degrees C). DNA cleavage is efficient in a Na+ ion concentration range of 5 mM to 1 M, with the optimum at 5 mM NaCl. The number of single-strand scissions on supercoiled DNA per MPE.Fe(II) under optimum conditions is 1.4. Metals such as Co(II), Mg(II), Ni(II), and Zn(II) inhibit strand scission by MPE. The released products from DNA cleavage by MPE.Fe(II) are the four nucleotide bases. The DNA termini at the cleavage site are 5'-phosphate and roughly equal proportions of 3'-phosphate and 3'-(phosphoglycolic acid). The products are consistent with the oxidative degradation of the deoxyribose ring of the DNA backbone, most likely by hydroxy radical.     

Sequence-specific interaction of the Salmonella Hin recombinase in both major and minor grooves of DNA.

The Hin recombinase of Salmonella catalyzes a site-specific recombination event which leads to flagellar phase variation. Starting with a fully symmetrical recombination site, hixC, a set of 40 recombination sites which vary by pairs of single base substitutions was constructed. This set was incorporated into the Salmonella-specific bacteriophage P22 based challenge phage selection and used to define the DNA sequence determinants for the binding of Hin to DNA in vivo. The critical sequence-specific contacts between a Hin monomer and a 13 bp hix half-site are at two T:A base pairs in the major groove of the DNA which are separated by one base pair, and two consecutive A:T contacts in the minor groove. The base substitutions in the major groove recognition portion which were defective in binding Hin still retained residual binding capability in vivo, while the base pair substitutions affecting the minor groove recognition region lost all in vivo binding. Using in vitro binding assays, Hin was found to bind to hix symmetrical sites with A:T base pairs or I:C base pairs in the minor groove recognition sequences, but not to G:C base pairs. In separate in vitro binding assays, Hin was equally defective in binding to either a G:C or a I:C contact in a major groove recognition sequence. Results from in vitro binding assays to hix sites in which 3-deazaadenine was substituted for adenine are consistent with Hin making a specific contact to either the N3 of adenine or O2 of thymine in the minor groove within the hix recombination site on each symmetric half-site. These results taken with the results of previous studies on the DNA binding domain of Hin suggest a sequence-specific minor groove DNA binding motif.     

The effects of symmetrical recombination site hixC on Hin recombinase function.

An artificial recombination site hixC composed of two identical half-sites that bind the Hin recombinase served as a better operator in vivo than the wild type site hixL (Hughes, K. T., Youderian, P., and Simon, M. I (1988) Genes & Dev. 2, 937-948). In vitro binding assays such as gel retardation assay and methylation protection assay demonstrated that Hin binds to hixC as tightly as it binds to hixL, even when the sites are located in negatively supercoiled plasmids. However, hixC served as a poor recombination site when it was subjected to the standard inversion assay in vitro. hixC showed a 16-fold slower inversion rate than the wild type. A series of biochemical assays designed to probe different stages of the Hin-mediated inversion reaction, demonstrated that Hin dimers bound to hixC have difficulty in forming paired hix site intermediates. KMnO4 and S1 nuclease assays detected an anomalous structure of the center of hixC only when the site was in negatively supercoiled plasmids. Mutational analysis in the central region of hixC and assays of paired hix site formation with topoisomers of the hixC substrate plasmid suggested that Hin is not able to pair hixC sites because of the presence of the anomalous structure in the center of the site. The structure does not behave like a DNA "cruciform" since Hin dimers still bind efficiently to the site. It is thought to consist of a short denatured "bubble" encompassing 2 base pairs. During the study of mutations in the center of hixC, it was found that Hin is not able to cleave DNA if a guanine residue is one of the two central nucleotides close to the cleavage site. Furthermore, Hin acts in a concerted fashion and cannot cleave any DNA strand if one of the four strands in the inversion intermediate is not cleavable.     

Formation of modified cleavage termini from the reaction of chromium(V) with DNA

Reaction of a 25 bp oligonucleotide with the high valent chromium complex, bis(2-ethyl-2-hydroxybutyrato)oxochromate(V) (Cr(V)-EHBA) produced both Frank- and alkali-labile strand breaks that were sequence-neutral. Frank strand break formation was found to be O2-dependent while formation of alkali-labile strand breaks were O2-independent. Reaction of Cr(V)-EHBA with the 5'-32P-labeled oligomer under oxygenated conditions formed the modified 3'-terminus, 3'-phosphoglycolate, as well as the 3'-phosphate terminus. Formation of the 3'-phosphoglycolate termini, and the O2 dependence of the reactions were consistent with a mechanism involving abstraction of the C4' hydrogen atom from the deoxyribose moiety of DNA. Identical reactions using the 3'-32P-labeled oligomer yielded only 5'-phosphate termini as assigned by co-migration with Maxam-Gilbert markers. Analogous cleavage profiles and modified termini were observed for the reaction of Cr(V)-EHBA and DNA in the presence of hydrogen peroxide. With the addition of hydrogen peroxide, the DNA cleavage reactions were O2-independent and the level of DNA cleavage was enhanced over that observed with Cr(V)-EHBA alone. These findings suggest an oxidation mechanism whereby a reductive intermediate of the carcinogen chromate, Cr(V), can cause DNA damage that mimics oxygen radical DNA damaging pathways.     

New complex of post-activated neocarzinostatin chromophore with DNA: bulge DNA binding from the minor groove

Neocarzinostatin (NCS-chrom), a natural enediyne antitumor antibiotic, undergoes either thiol-dependent or thiol-independent activation, resulting in distinctly different DNA cleavage patterns. Structures of two different post-activated NCS-chrom complexes with DNA have been reported, revealing strikingly different binding modes that can be directly related to the specificity of DNA chain cleavage caused by NCS-chrom. The third structure described herein is based on recent studies demonstrating that glutathione (GSH) activated NCS-chrom efficiently cleaves DNA at specific single-base sites in sequences containing a putative single-base bulge. In this structure, the GSH post-activated NCS-chrom (NCSi-glu) binds to a decamer DNA, d(GCCAGAGAGC), from the minor groove. This binding triggers a conformational switch in DNA from a loose duplex in the free form to a single-strand, tightly folded hairpin containing a bulge adenosine embedded between a three base pair stem. The naphthoate aromatic moiety of NCSi-glu intercalates into a GG step flanked by the bulge site, and its substituent groups, the 2-N-methylfucosamine carbohydrate ring and the tetrahydroindacene, form a complementary minor groove binding surface, mostly interacting with the GCC strand in the duplex stem of DNA. The bulge site is stabilized by the interactions involving NCSi-glu naphthoate and GSH tripeptide. The positioning of NCSi-glu is such that only single-chain cleavage via hydrogen abstraction at the 5'-position of the third base C (which is opposite to the putative bulge base) in GCC is possible, explaining the observed single-base cleavage specificity. The reported structure of the NCSi-glu-bulge DNA complex reveals a third binding mode of the antibiotic and represents a new family of minor groove bulge DNA recognition structures. We predict analogue structures of NCSi-R (R = glu or other substituent groups) may be versatile probes for detecting the existence of various structures of nucleic acids. The NMR structure of this complex, in combination with the previously reported NCSi-gb-bulge DNA complex, offers models for specific recognition of DNA bulges of various sizes through binding to either the minor or the major groove and for single-chain cleavage of bulge DNA sequences.     

Hin4II, a new prototype restriction endonuclease from Haemophilus influenzae RFL4: discovery, cloning and expression in Escherichia coli

The genes encoding restriction-modification system of unknown specificity Hin4II from Haemophilus influenzae RFL4 were cloned in Escherichia coli and sequenced. The Hin4II system comprises three tandemly arranged genes coding for m6A DNA methyltransferase, m5C DNA methyltransferase and restriction endonuclease, respectively. Restriction endonuclease was expressed in E. coli and purified to apparent homogeneity. The DNA recognition sequence and cleavage positions were determined. R.Hin4II recognizes the novel non-palindromic sequence 5'-CCTTC-3' and cleaves the DNA 6 and 5 nt downstream in the top and bottom strand, respectively. The new prototype restriction endonuclease Hin4II was classified as a potential candidate of HNH nuclease family after comparison against SMART database. An amino acid sequence motif 297H-X14-N-X8-H of Hin4II was proposed as forming a putative catalytic center.     

A role for the metal binding domain in determining the DNA sequence selectivity of Fe-bleomycin.

Previous studies of Fe-bleomycin-mediated DNA cleavage have established that the bithiazole moiety + C-terminal substituent of bleomycin are required for DNA binding, while the metal binding domain is responsible for O2 activation. Although recent studies have indicated that the metal binding domain also participates in DNA unwinding, and in determining the sequence and strand selectivity of DNA cleavage, no study has defined the structural domain that bears primary responsibility for the observed pattern of bleomycin-mediated DNA degradation. Presently, by the use of four synthetic analogs of bleomycin demethyl A2 having the functional domains connected by rigid spacers of varying lengths, the source of DNA cleavage specificity has been determined. When the four analogs cleaved 242- and 127-base pair 5'-32P-end-labeled DNA restriction fragments containing isolated Fe-bleomycin cleavage sites, all four produced cleavage at the same preferred sites. Because the (oligo)glycine spacers altered the distance between the domains by as much as 14 A, the identical cleavage patterns argue that the primary determinant of sequence specificity for these analogs is the metal binding domain.     

Interactions between eukaryotic DNA topoisomerase I and a specific binding sequence

The interaction between eukaryotic DNA topoisomerase I and a high affinity binding sequence was investigated. Quantitative footprint analysis demonstrated that the substrate preference results from strong specific binding of topoisomerase I to the sequence. The specificity was conferred by a tight noncovalent association between the enzyme and its target DNA, whereas the transient formation of a covalently bound enzyme.nicked DNA intermediate contributed insignificantly to the overall affinity. Topoisomerase I protected both strands over a 20-base pair region in which the cleavage site was centrally located. DNA modification interference analysis revealed a 16-base pair interference region on the scissile strand. Essential bases were confined to the 5' side of the cleavage site. The 6-base pair interference region observed on the complementary strand did not contain essential bases.     

Role of arginine-43 and arginine-69 of the Hin recombinase catalytic domain in the binding of Hin to the hix DNA recombination sites

The Hin recombinase mediates the site-specific inversion of a segment of the Salmonella chromosome between two flanking 26 bp hix DNA recombination sites. Mutations in two amino acid residues, R43 and R69 of the catalytic domain of the Hin recombinase, were identified that can compensate for loss of binding resulting from elimination of certain major and minor groove contacts within the hix recombination sites. With one exception, the R43 and R69 mutants were also able to bind a hix sequence with an additional 4 bp added to the centre of the site, unlike wild-type Hin. Purified Hin mutants R43H and R69C had both partial cleavage and inversion activities in vitro while mutants R43L, R43C, R69S, and R69P had no detectable cleavage and inversion activities. These data support a model in which the catalytic domain plays a role in DNA-binding specificity, and suggest that the arginine residues at positions 43 and 69 function to position the Hin recombinase on the DNA for a step in the recombination reaction which occurs either at and/or prior to DNA cleavage.     

Site-specific binding constants for actinomycin D on DNA determined from footprinting studies.

We report site-specific binding constants for the intercalating anticancer drug actinomycin D (Act-D), binding to a 139-base-pair restriction fragment from pBR 322 DNA. The binding constants are derived from analysis of footprinting experiments, in which the radiolabeled 139-mer is cleaved using DNase I, the cleavage products undergo gel electrophoresis, and, from the gel autoradiogram, spot intensities, proportional to amounts of cleaved fragments, are measured. A bound drug prevents DNase I from cleaving at approximately 7 bonds, leading to decreased amounts of corresponding fragments. With the radiolabel on the 3' end of the noncoding strand (A-label), we measured relative amounts of 54 cleavage products at 25 Act-D concentrations. For cleavage of the 139-mer with the label on the 3' end of the coding strand (G-label), relative amounts of 43 cleavage products at 11 Act-D concentrations were measured. These measurements give information about approximately 120 base pairs of the restriction fragment (approximately 12 turns of the DNA helix); in this region, 14 strong and weak Act-D binding sites were identified. The model used to interpret the footprinting plots is derived in detail. Binding constants for 14 sites on the fragment are obtained simultaneously. It is important to take into account the effect of drug binding at its various sites on the local concentration of probe elsewhere. It is also necessary to include in the model weak as well as strong Act-D sites on the carrier DNA which is present, since the carrier DNA controls the free-drug concentration. As expected, the strongest sites are those with the sequence (all sequences are 5'----3') GC, with TGCT having the highest binding constant, 6.4 x 10(6) M-1. Sites having the sequence GC preceded by G are weak binding sites, having binding constants approximately 1 order of magnitude lower than those of the strong sites. Also, the non-GC-containing sequences CCG and CCC bind Act-D with a binding constant comparable to those of the weak GGC sites. The analysis may reveal drug-induced structural changes on the DNA, which are discussed in terms of the mechanism of Act-D binding.     

Unusual reactivity in a commercial chromium supplement compared to baseline DNA cleavage with synthetic chromium complexes

Commercially available chromium supplements were tested for their DNA cleavage ability compared with synthetic chromium(III) complexes, including chromium(III) tris-picolinate [Cr(pic)3], basic chromium acetate [Cr3O(OAc)6]+, model complexes, and recently patented Cr-complexes for use in supplements or therapy. Four different supplements (P1-P4) were tested for their DNA cleaving activity in the presence and the absence of H2O2, dithiothreitol (DTT) or ascorbate. One supplement, P1, showed nicking of DNA in the absence of oxidant or reductant at 120 microM metal concentration. Different lot numbers of P1 were also tested for DNA cleavage activity with similar results. Commercial supplements containing Cr(pic)3 nicked DNA at 120 microM metal concentrations in the presence of 5 mM ascorbate or with excess hydrogen peroxide, analogous to reactions with synthetic Cr(pic)3 reported elsewhere. Another chromium (non-Cr(pic)3) supplement, P2, behaves in a comparable manner to simple Cr(III) salts in the DNA nicking assay. Chromium(III) malonate [Cr(mal)2] and chromium(III) acetate [Cr(OAc)] can nick DNA in the presence of ascorbate or hydrogen peroxide, respectively, only at higher metal concentrations. The Cr(III) complexes of histidine, succinate or N-acetyl-L-glutamate do not nick DNA to a significant degree.     

Product analyses in DNA strand scission by antitumor antibiotic elsamicin A.

Elsamicin A is an antitumor antibiotic with fascinating chemical structure and a good candidate for pharmaceutical development. Molecular mechanism of DNA backbone cleavage mediated by Fe(II)-elsamicin A has been examined. Product analysis using DNA sequencing gels and HPLC reveals the production of damaged DNA fragments bearing 3'-/5'-phosphate and 3'-phosphoglycolate termini associated with formation of free base. In addition, hydrazine-trapping experiments indicate that C-4' hydroxylated abasic sites are formed concomitant with DNA degradation by Fe(II)-elsamicin A. The results lead to the conclusion that the hydroxyl radical formed in Fe(II)-elsamicin A plus dithiothreitol system oxidizes the deoxyribose moiety via hydrogen abstraction predominantly at the C-4' carbon of the deoxyribose backbone and ultimately produces strand breakage of DNA.     

Footprint Analysis of the RAG Protein Recombination Signal Sequence Complex for V(D)J Type Recombination

We have studied the interaction between recombination signal sequences (RSSs) and protein products of the truncated forms of recombination-activating genes (RAG) by gel mobility shift, DNase I footprinting, and methylation interference assays. Methylation interference with dimethyl sulfate demonstrated that binding was blocked by methylation in the nonamer at the second-position G residue in the bottom strand and at the sixth- and seventh-position A residues in the top strand. DNase I footprinting experiments demonstrated that RAG1 alone, or even a RAG1 homeodomain peptide, gave footprint patterns very similar to those obtained with the RAG1-RAG2 complex. In the heptamer, partial methylation interference was observed at the sixth-position A residue in the bottom strand. In DNase I footprinting, the heptamer region was weakly protected in the bottom strand by RAG1. The effects of RSS mutations on RAG binding were evaluated by DNA footprinting. Comparison of the RAG-RSS footprint data with the published Hin model confirmed the notion that sequence-specific RSS-RAG interaction takes place primarily between the Hin domain of the RAG1 protein and adjacent major and minor grooves of the nonamer DNA.     

Ku antigen interacts with abasic sites

One of the most abundant lesions in DNA is the abasic (AP) sites arising spontaneously or as an intermediate in base excision repair. Certain proteins participating in the processing of these lesions form a Schiff base with the deoxyribose of the AP site. This intermediate can be stabilized by NaBH(4) treatment. By this method, DNA duplexes with AP sites were used to trap proteins in cell extracts. In HeLa cell extract, along with a prevalent trap product with an apparent molecular mass of 95 kDa, less intensive low-molecular-weight products were observed. The major one was identified as the p80-subunit of Ku antigen (Ku). Ku antigen, a DNA binding component of DNA-dependent protein kinase (DNA-PK), participates in double-stranded break repair and is responsible for the resistance of cells to ionizing radiation. The specificity of Ku interaction with AP sites was proven by more efficient competition of DNA duplexes with an analogue of abasic site than non-AP DNA. Ku80 was cross-linked to AP DNAs with different efficiencies depending on the size and position of strand interruptions opposite to AP sites. Ku antigen as a part of DNA-PK was shown to inhibit AP site cleavage by apurinic/apyrimidinic endonuclease 1.     

Molecular determinants of site-specific inhibition of human DNA topoisomerase I by fagaronine and ethoxidine. Relation to DNA binding

DNA topoisomerase (top) I inhibition activity of the natural alkaloid fagaronine (NSC157995) and its new synthetic derivative ethoxidine (12-ethoxy-benzo[c]phenanthridine) has been correlated with their molecular interactions and sequence specificity within the DNA complexes. Flow linear dichroism shows that ethoxidine exhibits the same inhibition of DNA relaxation as fagaronine at the 10-fold lower concentration. The patterns of DNA cleavage by top I show linear enhancement of CPT-dependent sites at the 0.016-50 microM concentrations of fagaronine, whereas ethoxidine suppress both top I-specific and CPT-dependent sites. Suppression of top I-mediated cleavage by ethoxidine is found to be specific for the sites, including strand cut between A and T. Fagaronine and ethoxidine are DNA major groove intercalators. Ethoxidine intercalates DNA in A-T sequences and its 12-ethoxy-moiety (absent in fagaronine) extends into the DNA minor groove. These findings may explain specificity of suppression by ethoxidine of the strong top I cleavage sites with the A(+1), T(-1) immediately adjacent to the strand cut. Fagaronine does not show any sequence specificity of DNA intercalation, but its highly electronegative oxygen of hydroxy group (absent in ethoxidine) is shown to be an acceptor of the hydrogen bond with the NH(2) group of G base of DNA. Ability of fagaronine to stabilize top I-mediated ternary complex is proposed to be determined by interaction of its hydroxy group with the guanine at position (+1) of the DNA cleavage site and of quaternary nitrogen interaction with top I. The model proposed provides a guidance for screening new top I-targeted drugs in terms of identification of molecular determinants responsible for their top I inhibition effects.     

Characterization of the binding specificity of two anticruciform DNA monoclonal antibodies

Two monoclonal antibodies (2D3 and 4B4) have been raised against a stable cruciform DNA structure containing the 27-base pair palindrome of the SV40 origin of replication on one strand and an unrelated 26-base pair palindrome on the complementary strand (pRGM 21 x pRGM 29) and have been shown to recognize conformational determinants specific to cruciform DNA structures (Frappier, L., Price, G.B., Martin, R. G., and Zannis-Hadjopoulos, M. (1987) J. Mol. Biol. 193, 751-758). To define the region(s) of the cruciform that is recognized by these antibodies, we examined the ability of 2D3 and 4B4 to protect the single-stranded tips of the loops or the four-way junctions at the base of the stem of stable cruciform molecules against cleavage by mung bean nuclease or T7 endonuclease 3, respectively. Both antibodies were found to protect two of the four elbow-like structures at the base of the cruciform from T7 endonuclease 3 cleavage, but not the tips of the cruciform arms from mung bean nuclease cleavage. Also, predigestion of the cruciform with mung bean nuclease did not affect the binding of either antibody. In addition, 2D3 bound to a cruciform and a T-shaped structure involving the palindromic sequence at the cloning site of pUC7, which is completely unrelated in sequence to the palindrome of pRGM 21 x pRGM 29, and protected the base of these stem-loop structures against cleavage by T4 endonuclease VII. These results indicate that 2D3 and 4B4 bind at or near the base of the cruciform molecules and that, at least for 2D3, binding is independent of DNA sequence.     

A role of oxidative DNA sugar damage in mutagenesis by neocarzinostatin and bleomycin

The anti-tumor antibiotics neocarzinostatin and bleomycin specifically oxidize deoxyribose in DNA at the C-5' and C-4' positions, respectively. The resulting DNA lesions include strand breaks and apyrimidinic sites. Both agents are broad specificity mutagens, inducing, in various systems, base substitutions, frameshifts and deletions. Sequencing studies in bacterial systems have suggested that the base substitutions may result primarily from replicative bypass of the oxidized apyrimidinic sites.