Selection-dominant and nonaccessible epitopes on cell-surface receptors revealed by cell-panning with a large phage antibody library.

To generate antibodies to defined cell-surface antigens, we used a large phage antibody fragment library to select on cell transfectants expressing one of three chosen receptors. First, in vitro panning procedures and phage antibody screening ELISAs were developed using whole live cells stably expressing the antigen of interest. When these methodologies were applied to Chinese hamster ovary (CHO) cells expressing one of the receptors for a neuropeptide, somatostatin, using either direct cell panning or a strategy of depletion or ligand-directed elution, many different pan-CHO-cell binders were selected, but none was receptor specific. However, when using direct panning on CHO-cells expressing the human membrane protein CD36, an extraordinary high frequency of antigen-specific phage antibodies was found. Panning on myoblasts expressing the rat homologue of CD36 revealed a similar selection dominance for anti-(CD36). Binding of all selected 20 different anti-(CD36) phage was surprisingly inhibited by one anti-(CD36) mAb CLB-IVC7, which recognizes a functional epitope that is also immunodominant in vivo. Similar inhibition was found for seven anti-(rat) CD36 that cross-reacted with human CD36. Our results show that, although cells can be used as antigen carriers to select and screen phage antibodies, the nature of the antigen target has a profound effect on the outcome of the selection.

A sensitive immunoassay for rat fatty acid translocase (CD36) using phage antibodies selected on cell transfectants: abundant presence of fatty acid translocase/CD36 in cardiac and red skeletal muscle and up-regulation in diabetes.

The rat membrane protein fatty acid translocase (FAT), which shows sequence similarity to human CD36 (a membrane protein supposedly involved in a variety of membrane processes), is implicated in the transport of long-chain fatty acids across cellular membranes. To set up an immunoassay for quantification of FAT in different tissues, we isolated a series of anti-FAT antibodies by panning a large naive phage antibody library on FAT-transfected H9c2 cells. All seven different phage antibody fragments isolated reacted specifically with FAT, and most likely recognize the same or closely located immunodominant sites on FAT, as a competitive monoclonal antibody (mAb) (CLB-IV7) completely blocked the binding of all these phage antibodies to cells. A sandwich ELISA was set up using mAb 131. 4 (directed against purified CD36 from human platelets) as capture antibody and phage antibodies and anti-phage sera as detector. With this ELISA (sensitivity 0.05 microgram/ml), the FAT content in isolated cardiomyocytes was found to be comparable with that of total heart ( approximately 3 mg/g of protein), while liver tissue and endothelial cells were below the detection limit (<0.1 mg of FAT/g of protein). During rat heart development, protein levels of FAT rose from 1.7+/-0.7 mg/g of protein on the day before birth to 3.6+/-0.4 mg/g of protein on day 70. Comparing control with streptozotocin-induced diabetic rats, a statistically significant (P<0.05) 2-4-fold increase of FAT was seen in heart (from 4.2+/-2.3 to 11.0+/-5.7 mg/g of protein), soleus (from 0.6+/-0.1 to 1.4+/-0.5 mg/g of protein) and extensor digitorum longus (EDL) muscle (from 0.3+/-0.1 to 1. 2+/-0.8 mg/g of protein). In addition, the FAT contents of each of these muscles were found to be of similar magnitude to the contents of cytoplasmic heart-type fatty-acid-binding protein in both diabetic rats and controls, supporting the suggested roles of these two proteins in cellular fatty acid metabolism. This is the first time phage display technology has been succesfully applied for direct selection, from a large naive antibody library, of antibodies that recognize selected membrane proteins in their natural context.

Somatostatin displayed on filamentous phage as a receptor-specific agonist.

1. In search of methods to identify bio-active ligands specific for G protein-coupled receptors with seven transmembrane spanning regions, we have developed a filamentous phage-based selection and functional screening method. 2. First, methods for panning peptide phage on cells were established, using the hormone somatostatin as a model. Somatostatin was displayed on the surface of filamentous phage by cloning into phage(mid) vectors and fusion to either pIII or pVIII viral coat proteins. Peptide displaying phage bound to a polyclonal anti-somatostatin serum, and, more importantly, to several somatostatin receptor subtypes (Sst) expressed on transfected CHO-K1 cells, in a pattern which was dependent on the used display method. Binding was competed with somatostatin, with an IC50 in the nanomolar range. The phage were specifically enriched by panning on cells, establishing conditions for cell selections of phage libraries. 3. Binding of somatostatin displaying phage to sst2 on a reporter cell line, in which binding of natural ligand reduces secretion of alkaline phosphatase (via a cyclic AMP responsive element sensitive promoter), proved that the phage particles act as receptor-specific agonists. Less than 100 phage particles per cell were required for this activity, which is approximately 1000 fold less than soluble somatostatin, suggesting that phage binding interferes with normal receptor desensitization and/or recycling. 4. The combination of biopanning of phage libraries on cells with functional screening of phage particles for receptor triggering activity, may be used to select novel, bio-active ligands from phage libraries of random peptides, antibody fragments, or libraries based on the natural receptor ligand.

Induction of oxidative DNA damage and early lesions in rat gastro-intestinal epithelium in relation to prostaglandin H synthase-mediated metabolism of butylated hydroxyanisole.

The effect of metabolic activation of the food additive 3-tert-butyl-4-hydroxyanisole (BHA) by prostaglandin H synthase on the gastro-intestinal cell proliferation was determined by studies of the nature and the time dependency of early lesions in the forestomach, glandular stomach and colon/rectum of rats given BHA with and without co-administration of acetylsalicyclic acid (ASA: an inhibitor of prostaglandin H synthase), in combination with the formation of oxidative DNA damage in the epithelial cells of glandular stomach and colon/rectum as well as in the liver. BHA appeared to be a strong inducer of oxidative DNA damage in the epithelial cells of the glandular stomach, increasing the level of 7-hydro-8-oxo-2'deoxyguanosine (8-oxodG) with increasing duration of BHA administration. Similar observations were made in colorectal DNA although levels of oxidative DNA damage tend to be smaller. In liver DNA, BHA appeared to be capable of increasing background 8-oxodG levels only after 14 days of treatment. This relatively slow response may be related to very low prostaglandin H synthase activity of liver cells. The severity of hyperplasia and inflammation in both forestomach and glandular stomach appeared to increase gradually with continued BHA administration. The hyperplasia induced by BHA was paralleled by inflammatory changes. In colorectal tissue, however, no tissue abnormalities were observed. This indicates that oxidative DNA damage induced by BHA is not a consequence of early lesions in gastro-intestinal epithelium, but might be the initial step in the stimulation of gastro-intestinal cell proliferation which, as shown previously, also occurs in colon epithelium. Co-administration of the prostaglandin H synthase inhibitor ASA resulted in a significant decrease of both epithelial oxidative DNA damage and the incidence of lesions, which indicates that this enzyme system is involved in the enhancement of cellular proliferation induced by BHA. Co-oxidation by prostaglandin H synthase of the BHA-metabolite tert-butylhydroquinone into tert-butylquinone, yielding active oxygen species, might therefore be responsible for the carcinogenic effects of this food antioxidant.

The formation of one-G deletions as a consequence of single-oxygen-induced DNA damage.

Single-stranded M13mp10 DNA containing a 144-bp mutational target sequence in the lacZ alpha gene was treated with singlet oxygen (1O2) generated by thermodissociation of the endoperoxide of 3,3'-(1,4-naphthalene-1,4-diyl)dipropionate (NDPO2). After transfection to non-SOS-induced E. coli cells, 32 mutants preselected for a mutation in the 144-bp target were collected and analyzed by DNA sequencing. One-G deletions represented the predominant type of mutation accounting for 50% of the mutations analyzed. The remaining part appeared to consist of base substitutions, i.e. G-->T transversions (34%), C-->T transitions (12.5%) and one T-->C transition (3%). Sixty percent of the mutations were found in two major mutational hotspots. We conclude that the predominant one-G deletions are due to a guanine reaction product which might be specific for 1O2.

Mutational specificity of oxidative DNA damage.

In this paper we describe our studies on the mutagenic consequences of oxidative DNA damage introduced by radiation-induced OH radicals (.OH) and by exposure to singlet oxygen (1O2), released by thermo-dissociation of the endoperoxide 3,3'-(1,4-naphthalidene) dipropionate (NDPO2). We have made use of M13mp10 bacteriophage and pUC18 plasmid DNA, containing a 144 base pair (bp) insert in the lacZ alpha gene. This 144 bp insert was used as a mutational target sequence. When dilute aqueous solutions of double-stranded (ds) M13mp10 (plus 144 bp insert) were gamma-irradiated in the presence of oxygen (O2; 100% .OH) or nitrous oxide (N2O; 90% .OH, 10% .H), very specific mutation spectra were found. Mainly bp substitutions were observed, of which C/G to G/C transversions are the predominant type. Moreover, the mutations are for the most part concentrated into two mutational hot spots: a minor and major one. Differences between the oxic (O2) and anoxic (N2O) mutation spectra could also be observed. Under N2O-1 bp deletions were detected, which are absent in the presence of O2, and in the anoxic spectrum more C/G to A/T transversions are present. To investigate whether these differences were due to the small amount of H radicals, which are formed under N2O, ds M13mp10 (plus 144 bp insert) was exposed to gamma-rays in phosphate buffer under nitrogen (55% .H, 45% .OH). Under these conditions a remarkable shift was observed from C/G-->G/C to C/G-->A/T transversions, while the mutations were far more scattered along the 144 bp sequence and no -1 bp deletions were detected. These results strongly suggest that H radicals do not cause -1 bp deletions, but may be responsible for the observed C/G to A/T transversions. The kind of bp substitution not only appeared to be dependent on the type of the water radicals, but also appeared to be strongly influenced by the replicon in which the target sequence is incorporated. When an oxygenated solution of pUC18 plasmid DNA (plus 144 bp insert) is irradiated, mainly C/G to A/T transversions were found at the same major hot spot instead of C/G to G/C transversions when the 144 bp sequence is part of M13mp10 DNA. Finally, in agreement with the observation that 1O2 reacts preferentially with guanine in DNA, a guanine is involved in most of the mutations scored after exposure of single-stranded (ss) M13mp10 DNA to NDPO2-generated 1O2.(ABSTRACT TRUNCATED AT 400 WORDS)

Singlet oxygen-induced DNA damage: product analysis, studies of biological consequences and characterization of mutations.

The DNA lesions induced by free 1O2 and the biological and mutagenic consequences of 1O2-induced DNA damage have been studied. Using anion exchange HPLC, reverse-phase HPLC with electrochemical detection and 32P-postlabelling methods, we have shown that 1O2 reacts with 2'-deoxyguanine 3'-monophosphate (dGp) but not with any other dNp. Reaction with dGp yields a large number of products; one minor product was identified as 7-hydro-8-oxo-2'-deoxyguanosine 3'-monophosphate (8-oxo-dGp), and a second tentatively as a formamidopyridine derivative of dGp. 8-Oxo-dGp was also found after reaction of 1O2 with single-stranded (ss) DNA, double-stranded (ds) DNA or an oligonucleotide (16-mer) having one G. With the oligonucleotide we found a second unidentified reaction product. With ss DNA, 8-oxo-dG was a much more prominent product than in the reaction of 1O2 with free dGp and the yield was about eight-fold higher than with ds DNA. This agrees with our finding that ss M13 DNA is at least 100-fold more sensitive than ds M13 DNA to biological inactivation by 1O2. The inactivation of ss M13 DNA must be largely due to 1O2-induced lesions other than 8-oxo-dG. In agreement with the observed preferential reaction of 1O2 with dG, most of the mutations induced by 1O2 in ss or ds M13mp10 DNA occurred at a G or G/C basepair, respectively. A preference for G(C) to T(A) transversions was observed for which 8-oxo-dG might have been responsible. In ss DNA, a significant number of mutations are characterized by the fact that a G is deleted.

Interaction of singlet oxygen with DNA and biological consequences.

To study the interaction of singlet oxygen (1O2) with DNA and the biological consequences of 1O2-induced DNA damage, we used the thermodissociable endoperoxide of 3,3'-(1,4 naphthalidene) dipropionate (NDPO2) as a generator of free 1O2 in reactions with (1) 2'-deoxynucleoside 3'-monophosphates (dNps), (2) an oligonucleotide (16-mer) having one deoxyguanine (dG), (3) native and denaturated rat kidney DNA and (4) single-stranded (ss) and double-stranded (ds) bacteriophage M13mp10 DNA. Using both anion exchange and reversed phase HPLC and 32P-postlabeling analyses, it was found that exposure of the various dNps to chemically generated 1O2 led to a detectable reaction with dGp and not with dAp, dCp, d5mCp or Tp. The reaction with dGp led to degradation of this nucleotide and the formation of a large number of reaction products, one of which could be identified as 7-hydro-8-oxo-2'-deoxyguanosine 3'-monophosphate (8-oxo-dGp). A second product could tentatively be identified as a formamido pyrimidine derivative of dGp (Fapy-dGp). When ss DNA, ds DNA or the oligonucleotide were exposed to 1O2, the formation of 8-oxo-dG could also be demonstrated. With the oligonucleotide, we found a so far unidentified reaction product. Under the same reaction conditions the yield of 8-oxo-dG was about 8-fold higher in ss DNA than in ds DNA. In ss DNA 8-oxo-dG seemed to be a more prominent product than in the case of reaction of 1O2 with free dGp. Reaction of 1O2 with ss or ds M13mp10 DNA led to biological inactivation of these DNAs, ss DNA being at least 100-fold more sensitive than ds DNA. It could be concluded that inactivation of the ss DNA must be largely due to 1O2-induced DNA lesions other than 8-oxo-dG. In agreement with the observed preferential reaction of 1O2 with dG most of the so far sequenced mutations, induced by 1O2 in a 144 bp mutation target sequence inserted in the lacZ alpha gene of ss or ds M13mp10 DNA, occurred at a G or G/C base pair respectively. A preference for G(C) to T(A) transversions can be observed for which 8-oxo-dG might have been responsible. In ss DNA a significant number of the mutations are characterized by the fact that a G is deleted.

Detection of 8-hydroxyguanine in small amounts of DNA by 32P postlabeling.

A method for the sensitive detection of 8-hydroxyguanine residues in small amounts of DNA (0.2-2 micrograms) was developed. It comprises (i) the enzymatic hydrolysis of DNA to 2'-deoxyribonucleotide 3'-monophosphates, (ii) degradation of the bulk amount of normal purine and pyrimidine deoxyribonucleotides in the DNA digest by treatment with trifluoroacetic acid and hydrazine, respectively, under conditions retaining the structure of d(8-OH-G)p necessary for 5' phosphorylation by T4 polynucleotide kinase (PNK), (iii) 5' phosphorylation of d(8-OH-G)p by T4 PNK-catalyzed transfer of 32P from [gamma-32P]ATP, and (iv) 2D thin-layer chromatography on polyethyleneimine-cellulose sheets to purify and resolve 32P-postlabeled d(8-OH-G)p. Model experiments with mixtures composed of synthesized d(8-OH-G)p and DNA hydrolysate indicate that it is possible to detect one 8-hydroxyguanine residue out of 2 x 10(6) normal bases starting with 1 microgram DNA. The methodology, which allows for a further decrease of this detection limit, might be very useful for the sensitive detection of DNA damage induced by activated oxygen species in small amounts of DNA. We demonstrate the formation of 8-OH-G in DNA in vitro by low doses of 60Co gamma-rays.

A novel DNA nucleotide in Trypanosoma brucei only present in the mammalian phase of the life-cycle.

The existence of an unusual form of DNA modification in the bloodstream form of the African trypanosome Trypanosoma brucei has been inferred from partial resistance to cleavage of nuclear DNA with PstI and PvuII (Bernards et al, 1984; Pays et al, 1984). This putative modification is correlated with the shut-off of telomeric Variant-specific Surface Glycoprotein (VSG) gene expression sites (ESs). The modification only affects inactive VSG genes with a telomeric location, and it is absent in procyclic (insect form) trypanosomes in which no VSG is made at all. Previous attempts to detect unusual nucleosides in T.brucei DNA were unsuccessful, but we now report the detection of two unusual nucleotides, called pdJ and pdV, in T.brucei DNA, using the 32P-postlabeling technique. Nucleotide pdV was present in both bloodstream form and procyclic T.brucei DNA and co-migrated in two different two-dimensional thin layer chromatography (2D-TLC) systems with hydroxymethyldeoxyuridine 5'-monophosphate (pHOMedU). In contrast, nucleotide pdJ was exclusively present in bloodstream form trypanosomal DNA. Levels of pdJ were higher in DNA enriched for telomeric sequences than in total genomic DNA and pdJ was also detected in other Kinetoplastida species exhibiting antigenic variation. Postlabeling and 2D-TLC analyses showed base J to be different from the known eukaryotic unusual DNA bases 5-methylcytosine, N6-methyladenine and hydroxymethyluracil, and also from (glucosylated) hydroxymethylcytosine, uracil, alpha-putrescinylthymine, 5-dihydroxypentyluracil and N6-carbamoylmethyladenine. We conclude that pdJ is a novel eukaryotic DNA nucleotide and that it is probably responsible for the partial resistance to cleavage by PvuII and PstI of inactive telomeric VSG genes. It may therefore be involved in the regulation of ES activity in bloodstream form trypanosomes.

Polycyclic aromatic hydrocarbon-DNA adducts in lung tissue from lung cancer patients.

In an attempt to probe for polycyclic aromatic hydrocarbon (PAH)-DNA adducts in human subjects resulting from smoking (or other chronic environmental exposure), lung tissue and lung tumours were obtained from patients hospitalized for lung cancer. DNA was isolated from the tissue samples and examined both in an ELISA using a polyclonal antibody against (+/-)trans-7,8-dihydroxy-anti-9,10-epoxy-7,8,9,10- tetrahydrobenzo[a]pyrene (BPDE)-DNA as well as by the nuclease P1-mediated modification of the 32P-post-labelling technique. The ELISA results showed BPDE-DNA antigenicity in lung DNA from 6 out of 21 patients, and adduct levels ranged from 2 to 134 adducts per 10(8) nucleotides. For all 21 patients, the autoradiographs of chromatograms of 32P-postlabelled digests of DNA from non-tumorous lung tissue showed a strong diagonal radioactive zone (DRZ). This DRZ was generally absent in tumorous tissue. DNA samples that were positive in the ELISA contained a dominant spot within the DRZ that co-chromatographed with the major BPDE-DNA adduct (BPDE-dG). The quantities of the BPDE-dG spots ranged from 2.1 to 42 adducts in 10(9) nucleotides. These values were lower than the levels found in the ELISA but correlated well with the ELISA results (Kendall W = 0.97; P = 0.00). The levels of the DRZ adducts ranged from 1.9 to 34 adducts in 10(8) nucleotides. Correlations between smoking and DNA adduct levels were poor because of the small number of current smokers (n = 13). However, smokers of filter cigarettes had significantly lower DNA adduct levels compared with smokers of cigarettes without a filter (P = 0.02 by Fischer's exact test).

Use of shuttle vectors to study the molecular processing of defined carcinogen-induced DNA damage: mutagenicity of single O4-ethylthymine adducts in HeLa cells.

We developed a simian virus 40 based shuttle vector system to study the molecular consequences of distinct carcinogen-induced DNA lesions in human cells. To establish the mutagenicity of O4-ethylthymine adducts, oligonucleotides carrying a single O4-ethylthymine adduct at a unique position were ligated into the vector molecules. Following replication in HeLa cells on average 23% of the progeny molecules carried a mutation in the region of modification. The vast majority of these mutations represented single T----C transitions at the position of the modified base, most probably as a consequence of mispairing of the O4-ethylthymine residues during replication. To a minor extent the O4-ethylthymine adduct may also induce T----A transversions or double point mutations. The in vivo mutation frequency of the adduct was found to be comparable to that of a C-A mismatch at the same position, but was lower than that expected from in vitro experiments with adducted DNA templates and purified DNA polymerases.

Synthesis of heterocyclic N-acetoxyarylamines and their reactivity with DNA.

2-Acetoxyamino-5-phenylpyridine and 2-acetoxyamino-3-methyl-5-phenylpyridine, being proposed ultimate carcinogens of the heterocyclic aromatic amines 2-amino-5-phenylpyridine (APP) and 2-amino-3-methyl-5-phenylpyridine (AMPP), respectively, were synthesized, crystallized and characterized. Using the 32P-postlabelling technique, we show that the total amount of adducts found in DNA after reaction with these N-acetoxyarylamines is at least 30- and 450-fold higher than in DNA reacted with equimolar amounts of the proposed proximate carcinogens 2-hydroxyamino-5-phenylpyridine and 2-hydroxyamino-3-methyl-5-phenylpyridine, respectively. These results support a postulated activation mechanism, in which N-acetoxyarylamines are the ultimate reactive species responsible for DNA modification by carcinogenic aromatic amines in vivo. The possibility to obtain the reactive 0-acetyl derivatives of APP and AMPP in crystalline form makes them unique model compounds for studies on the interaction of ultimate carcinogens of aromatic amines with DNA.

An optical study of the conformation of the aminofluorene-DNA complex.

Calf thymus DNA was modified with 2-aminofluorene (AF) to different extents by treatment with N-hydroxy-2-aminofluorene. The AF-modified DNAs together with free AF, the AF-modified guanine (Gua-C8-AF) and the AF-modified deoxyguanosine (dGuo-C8-AF) were subsequently studied by u.v. absorbance, linear dichroism and fluorescence spectroscopy. The emission and absorption properties of double-stranded DNA-AF and single-stranded DNA-AF closely resemble those of dGuo-C8-AF. The emission spectra of these three compounds show a broad, red-shifted emission, characteristic for exciplex formation. The linear dichroism and circular dichroism spectra of double-stranded DNA-AF show that the AF moiety forms a well-defined, regular structure. The dichroic ratio in the 310-340 nm region is constant, which indicates the presence of only one type of adduct. The long-wavelength transition moment of this adduct makes an angle of 72-74 degrees with the DNA helix axis. The binding of AF to double-stranded is DNA is accompanied by a destabilization of the DNA helix structure, a strong quenching of the AF emission quantum yield, intense AF circular dichroism and an apparent immobilization of the dGuo-C8-AF complex. In single-stranded DNA-AF, the AF conformation appears more random, although the interactions between AF and the surrounding bases persist. The strong interactions between AF and the surrounding bases which dominate the optical properties of the studied complexes, the significant destabilization of the DNA double helix after modification with AF, and the relatively small angle between AF and the base planes support a model in which the adduct is inserted into the DNA helix.

The biological activity of single-stranded phi X174 DNA, modified by N-hydroxy-2-aminofluorene, is inhibited by guanine imidazole ring-opening of the major, non-lethal aminofluorene-DNA adduct.

The major aminofluorene-DNA derivative, found in the liver of rats after administration of the hepatocarcinogen N-acetyl-2-aminofluorene and identified as N-(deoxyguanosin-8-yl)-2-aminofluorene (dGuo-C8-AF), was introduced in different amounts in single-stranded phi X174 DNA by reacting the DNA with tritium labeled N-hydroxy-2-aminofluorene. The modified DNA was subsequently incubated in 0.1 M NaOH at 37 degrees C for increasing periods of time to convert the dGuo-C8-AF residues into their guanine imidazole ring-opened forms. The degree of conversion was determined by measuring the amount of residual N-(guanin-8-yl)-2-aminofluorene in trifluoroacetic acid hydrolyzates of the alkali-treated DNA by h.p.l.c. In addition, the effect of ring opening on the biological activity of the DNA was monitored by transfecting the DNA to Escherichia coli wild-type spheroplasts. The results indicate that the major aminofluorene-DNA adduct formed initially, which contributes little to inactivation, becomes lethal when its guanine imidazole ring is opened.

By-pass of the major aminofluorene-DNA adduct during in vivo replication of single- and double-stranded phi X174 DNA treated with N-hydroxy-2-aminofluorene.

To examine the effects of aminofluorene-DNA adduct formation on the biological activity of DNA, single-stranded (ss) phi X174 DNA and phi X174 replicative form (RF) DNA were modified to different extents with 3H-labeled N-hydroxy-2-aminofluorene and subsequently transfected to Escherichia coli spheroplasts with different repair capabilities. When the fraction of active ss phi X174 DNA molecules was measured as a function of the mean number of adducts per molecule, exponential survival curves were obtained from which it could be deduced that in wild-type, uvrA- and recA- cells at least 86%, and in uvrC- cells at least 82% of the introduced adducts do not cause inactivation. In the case of RF DNA the survival curves are non-exponential, but they nevertheless show that an exceptionally high number of adducts per RF molecule must be introduced to destroy its biological activity. On average 52 adducts per RF molecule were needed to reduce the survival to 37%, irrespective of whether wild-type, uvrA- or recA- cells were used. On the other hand, the survival of the uvrC- cells was considerably lower, but even in these cells a majority of the adducts is not lethal. By h.p.l.c. analysis of the modified DNA after hydrolysis with trifluoroacetic acid, 81 and 84% of the adducts in ss- and RF DNA, respectively, could be identified as N-(guanin-8-yl)-2-aminofluorene. The results strongly indicate that this type of major modification product is very frequently by-passed during replication of both single- and double-stranded DNA. The results together with the data obtained by sucrose gradient analysis both before and after an alkali treatment and those obtained by h.p.l.c. analysis suggest that inactivation of ssDNA is mainly due to minor modifications such as secondary lesions consisting of chain breaks and alkali-labile sites together with unidentified interaction products.

Effects of adduct formation on the biological activity of single- and double-stranded øX174 DNA, modified by N-acetoxy-N-acetyl-2-aminofluorene.

In order to establish a good quantitative relationship between the number of acetylaminofluorene adducts and the extent of inactivation of DNA, single-stranded (ss) øX174 DNA and øX174 RF DNA were modified to various extents with 3H-labelled N-acetoxy-N-acetyl-2-aminofluorene (N-AcO-AAF) and subsequently transfected to Escherichia coli spheroplasts having different repair capabilities. Exponential survival curves were obtained. In the case of ssDNA about one adduct per molecule appears to be lethal. On the other hand only 1 out of 10.2 adducts is found to inactivate RF DNA if tested on wild-type E. coli. However, when assayed on strains deficient in excision repair 1 out of 2.3 adducts leads to inactivation of RF DNA. RecA-dependent postreplication repair only has little influence on these figures. Product analysis of the modified DNAs shows that in RF DNA at least 76% of the interaction products is N-(deoxyguanosin-8-yl)-N-acetyl-2-aminofluorene (dGuo-C8-AAF) and at least 6% and at most 12% is 3-(deoxyguanosin-N2-yl)-N-acetyl-2-aminofluorene (dGuo-N2-AAF). In ssDNA only dGuo-C8-AAF is formed. No apurinic sites could be detected in the modified DNAs. From these results it can be concluded that in RF DNA most of the dGuo-C8-AAF is removed by excision repair. The remaining damage, consisting probably both of dGuo-N2-AAF and unexcised dGuo-C8-AAF, inactivates RF DNA. Inactivation can be explained by a model which shows that only damage in the minus strand of RF DNA inhibits replication and/or transcription.