Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (197K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (9)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Perrin, L. C.
Right arrow Articles by Phillips, D. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Perrin, L. C.
Right arrow Articles by Phillips, D. R.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Nucleic Acids Research Pages 1781-1787  

Barminomycin forms GC-specific adducts and virtual interstrand crosslinks with DNA
Introduction
Materials And Methods
   DNA substrates
   Transcription
   Electrophoresis and autoradiography
   Crosslinking assay
Results
   Transcriptional blockages
   Stability of adducts at 37°C
   Drug concentration and time dependence of adduct formation
   DNA interstrand crosslinking
   Thermal stability of interstrand crosslink
   Rate of crosslinking
Discussion
   Sequence specificity
   Stability of the barminomycin-DNA lesion
   Structure of lesion
   Conclusions
Acknowledgement
References

Barminomycin forms GC-specific adducts and virtual interstrand crosslinks with DNA

Barminomycin forms GC-specific adducts and virtual interstrand crosslinks with DNA

Leah C. Perrin, Carleen Cullinane, Ken-ichi Kimura1 and Don R. Phillips*

Department of Biochemistry, La Trobe University, Bundoora, Victoria 3083, Australia and 1Research Institute of Life Sciences, Snow Brand Milk Products Co. Ltd, Ishibashi-machi, Shimotsuga-gun, Tochigi 329-05, Japan

Received as resubmission February 22, 1999; Revised and Accepted February 26, 1999

ABSTRACT

The sequence specificity of the binding of barminomycin (SN-07 chromophore) to DNA was investigated using an in vitro transcription assay. It was found that this compound formed blockages to transcription, and these blocks were highly selective for 5[prime]-GC sequences. The half-lives of the first seven transcriptional blockages at 37°C were 14-130 min, plus one site >>200 min, with widely varying levels of essentially permanent blockages at each site (0-100%; average of 40%), indicative of considerable dependence on flanking sequences of adducts stability at individual GC sites. Barminomycin was also shown to form DNA virtual (i.e. functional) interstrand crosslinks. Such crosslinks were also relatively heat stable, with 40% of the DNA remaining crosslinked after heating at 90°C for 5 min. The barminomycin-DNA adducts and crosslinks appear to be essentially identical to those formed between adriamycin and DNA. Whereas adriamycin requires prior activation with formaldehyde in order to form adducts and crosslinks, barminomycin behaves in all respects as if it is a pre-activated form of adriamycin.

INTRODUCTION

The active macromolecular antitumour antibiotic (SN-07) was isolated from Actinomadura roseoviolacea var. miuraensis nov. var in 1985 (1). This macromolecular antibiotic has been shown to be mutagenic in both prokaryotic and eukaryotic cells and does not require metabolic activation (2). The active chromophore (denoted SN-07 chromophore) was isolated from the DNA-chromophore complex and was shown to be barminomycin (Fig. 1, part 4) (3-5). It exists primarily in the carbinolamine form (Fig. 1, part 4) but can slowly isomerise to the imine structure (Fig. 1, part 5) (6).

Barminomycin has dramatic biological properties, as revealed by inhibition of growth of L1210 cells at an IC50 of 0.007 ng/ml (7). It was also shown to be cytotoxic against HeLa cells (IC50 = 0.72 ng/ml, after 24 h exposure) (6). It has also shown potent activity against P388 leukemia cells, being 1000-fold more cytotoxic than adriamycin (IC50 = ~0.01 ng/ml for barminomycin, compared with 13 ng/ml for adriamycin) (8). Since the drug has shown such intense biological activity it is of considerable interest to establish its mechanism of action.

Daunomycin (daunorubicin) and adriamycin (doxorubicin) (Fig. 1) are two well-known anthracyclines which are used extensively for the treatment of a wide range of tumours (9). Elucidation of the mechanism of action of these agents has been the subject of intense investigation, with the expectation that this will lead to more active derivatives (10-14). It is known that these compounds localise in the nucleus where they intercalate with DNA, leading to impairment of nuclear processes such as replication and transcription. Recently it has been found that reaction between adriamycin and formaldehyde leads to a form of the drug (Fig. 1, part 3), which could explain some of the observed activity of Adriamycin (15). This compound has the potential to not only intercalate with DNA but also to form a covalent bond (N-C-N aminal linkage) between the 3[prime]-amino of adriamycin and the N2 of guanine. The result of these two interactions with DNA leads to a drug adduct which forms primarily at 5[prime]-GC sequences and has sufficient stability (16) to act as a ‘virtual crosslink’ (12-16). These interstrand crosslinks have now been detected in tumour cells treated with adriamycin (C.Cullinane, S.M.Cutts, C.Panousis and D.R.Phillips, unpublished results).


Figure 1. Structure of relevant anthracyclines. Daunomycin (1), adriamycin (2), simplified monomeric representation of the formaldehyde-activated form of adriamycin (3), carbinolamine form of barminomycin (4) and the imine form of barminomycin (5).


Examination of the structure of barminomycin reveals that it is similar to the activated form of adriamycin in that it contains an intercalating aglycone chromophore with a side-chain containing a functional group capable of forming a covalent bond with DNA. Barminomycin exhibits preferential binding for 5[prime]-GC sequences, as shown by a restriction enzyme digestion assay (17). This work also indicated an interaction between the drug and the 2-amino group of guanine. Further work involving DNase I footprinting experiments showed that the drug exhibited some selectivity for 5[prime]-GC binding sites (18).

In the present work we show that when the drug is bound to DNA it inhibits the process of RNA synthesis in a model in vitro transcription system. Analysis of the truncated transcripts revealed the precise site of drug binding, and that the drug lesion at these sites is that of a stable adduct. Furthermore, an assay to probe for interstrand crosslinks was employed and revealed that the DNA-barminomycin adducts were interstrand crosslinks.

MATERIALS AND METHODS

Barminomycin (SN-07 chromophore) was isolated and characterised as described previously (3). A stock drug solution (1 mM) was prepared in methanol and stored in the dark at 4°C. Urea was purchased from ICN Biomedicals, TEMED and ribonuclease inhibitor were from Promega, acrylamide and bisacrylamide from Bio-Rad. Nucleotides (including 3[prime]-O-methylnucleotides and the primer GpA), Escherichia coli RNA polymerase (nuclease free), BSA (nuclease free) and Klenow fragment of E.coli DNA polymerase I were purchased from Pharmacia. [[alpha]-32P]UTP (specific activity, 3000 Ci/mmol) and [[alpha]-32P]dCTP (specific activity, 3000 Ci/mmol) were obtained from Amersham while restriction enzymes (PvuII and HindIII) were from New England Biolabs. The ‘Biotrap’ and appropriate membranes were obtained from Schleicher and Schuell. Heparin was from Sigma. The Nensorb 20 columns were purchased from NEN Research Products, whilst calf thymus DNA was purchased from Worthington Biochemical Corporation. All other reagents were of analytical grade and all solutions were prepared using water purified through a Milli-Q four-stage system (Millipore, MA).

DNA substrates

Isolation of pCC1 and subsequent restriction digestion with PvuII and HindIII to yield a 512 bp fragment containing the lac UV5 promoter was as previously described (19).

Transcription

Transcription reaction conditions employed have previously been described (20-22). Briefly, the DNA fragment containing the lac UV5 promoter was incubated with E.coli RNA polymerase at 37°C in a transcription buffer (40 mM Tris, pH 8.0, 100 mM KCl, 3 mM MgCl2, 100 µM EDTA, 10 mM DTT, 160 µg/ml BSA, 1 U/µl RNase inhibitor) to form an open ternary complex. Non-specifically bound polymerase was removed by the addition of heparin to a final concentration of 400 µg/ml. A labelled initiated complex was formed in the presence of 200 µM GpA, 5 µM ATP and GTP, and [[alpha]-32P]UTP. The drug was then added and after a designated incubation time, the transcript was elongated by adjusting the conditions to 2 mM of all four nucleotides and 400 mM KCl. Reactions were terminated by the addition of an equal volume of loading dye (10 M urea, 10% sucrose, 40 mM EDTA, 0.1% xylene cyanol, 0.1% bromophenol blue). The RNA was then denatured by heating the samples at 90°C for 5 min.

Electrophoresis and autoradiography

The transcripts were subjected to electrophoresis through a 12% denaturing polyacrylamide gel. The gel was subsequently fixed (10% methanol, 10% acetic acid) and dried using standard procedures, then exposed to a phosphor screen overnight. Analysis and quantitation was performed using a 400B PhosphorImager and ImageQuant software (Molecular Dynamics, CA).

Crosslinking assay

The DNA (pCC1; 3496 bp) was linearised with HindIII and then radioactively labelled using the Klenow fragment of E.coli DNA polymerase I and [[alpha]-32P]dCTP. The end-labelled DNA was then purified by chromatography through a Nensorb 20 column. Finally, the DNA was dissolved in a TE solution (10 mM Tris, pH 8.0, 1 mM EDTA) containing 470 µM bp sonicated calf thymus DNA.

The end-labelled DNA (25 µM bp) was reacted with increasing concentrations of the drug in transcription buffer at 37°C. Reaction times exceeding 1 h were performed in the dark. The reactions were terminated by the addition of equal volume of a loading dye (90% formamide, 10 mM EDTA, 0.1% bromophenol blue, 0.1% xylene cyanol) and denatured by heating (90°C, 5 min) prior to loading onto a 0.8% agarose gel. The samples were subjected to electrophoresis in TAE buffer at 40 V for 16 h to separate double-stranded DNA from the denatured single-stranded DNA. The gel was then dried (80°C, in vacuo) prior to exposure to a phosphor screen. The gel was analysed and quantitated using a 400B PhosphorImager and ImageQuant software (Molecular Dynamics, CA).

RESULTS

Transcriptional blockages

The sequence specificity of DNA binding by barminomycin was determined using an in vitro transcription assay. The inititated complex was incubated with 40 µM drug for times up to 5 h prior to elongation of the transcripts. The resolved transcripts are shown in Figure 2. Elongation of the control template resulted in high levels of full-length transcript within 1 min. In contrast, drug induced blockages were clearly evident within the first minute of the reaction (lane 8) and these blockages continued to increase with time, as indicated by the progressive loss of full-length transcript. The mole fraction of blocked transcript at the first drug binding site encountered by the RNA polymerase (truncated RNA of 37 nt) was quantitated for each of the incubation times and is shown in Figure 3A. Drug binding at this site continued to increase as a function of reaction time up to 5 h where the drug occupancy reached ~50%.


Figure 2. Sequence dependence of barminomycin adducts. Transcription was initiated at the -1 position from the lac UV5 promoter of the 512 bp fragment. The initiated complex was then incubated for 1, 5, 10 and 30 min and 1, 3 and 5 h in the absence (lanes 1-7) and the presence (lanes 8-14) of barminomycin (40 µM) before elongation of the transcripts for 10 min. Lane I shows the initiated complex (i.e. nascent RNA prior to commencement of elongation), while C and G represent sequencing lanes obtained using 3-O-methyl CTP and 3-O-methyl GTP, respectively, during the elongation process.


Figure 3. Rate of reaction and sequence dependence of barminomycin adducts. (A) Rate of reaction of barminomycin with DNA. Lanes 8-14 of Figure 2 were quantitated and the mole-fraction of blocked transcripts at the first block site (37 on Fig. 2) was calculated and expressed as a function of incubation time of the initiated transcription complex with barminomycin (40 µM). (B) Sequence-dependence and relative occupancy of barminomycin adducts. The mole-fraction of each of the barminomycin-induced blocked transcripts was quantitated from the 30 min lane (lane 11) of Figure 2 (40 µM barminomycin) and is shown as drug occupancy (%) at the blockage site corresponding to the sequence and transcript length of the non-template strand. The numbering represents the length of the RNA transcript beginning from the G of GpA in each chain. The dashed line defines high occupancy sites as above an occupancy of 5%.

The sequence specificity of the blockages induced by barmino-mycin-DNA adduct formation (Fig. 2, lane 11) and the relative drug occupancy at each site is shown in Figure 3B. All seven of the major blockage sites (drug occupancy >5%) resolved for RNA lengths up to 140 nt were at the G of 5[prime]-GpC sequences. The drug occupancy was not quantitated to higher lengths of the truncated transcripts because of the significantly reduced fraction of E.coli RNA polymerase reaching these sites (23).

Stability of adducts at 37°C

Previous work with adriamycin has shown that adriamycin-DNA adducts are not conventional stable covalent adducts but have some reversibility and this is reflected by a gradual loss of transcriptional blockages with time (24). It was therefore important to examine whether the barminomycin-DNA adducts exhibited a similar lability. In order to assess the stability of the barminomycin-DNA blockages, adducts were formed with the initiated transcription complex, and elongation then allowed to continue for up to 3 h. As seen in Figure 4A, the stability of barminomycin induced blockages was heterogeneous with several sites remaining unchanged while the intensity of others decreased as a function of elongation time. The decay of the first seven well resolved major sites was quantitiated (Fig. 4B) and showed a loss of blockages with time, consistent with dissociation of barminomycin from these sites (23).


Figure 4. Stability of barminomycin adducts at 37°C. (A) Barminomycin (5 µM) was reacted with the initiated DNA for 5 min prior to elongation of the RNA for 1, 5, 10, 20 or 30 min or 1 or 3 h (lanes 8-14). Lanes 1-7 represent the corresponding lanes for elongation of the drug-free initiated DNA. Lane I shows the initiated complex prior to elongation of the control solution, while C and G represent sequencing lanes obtained using 3-O-methyl CTP and 3-O-methyl GTP, respectively. (B) The decay of the major transcriptional blockages from (A) is shown as the relative occupancy at individual sites over the 3 h time period: [closed square], 37; [open square], 82; [open triangle], 97; +, 110; [closed circle], 117; [open circle], 125; [closed triangle], 129. The insert shows the first order fit to the decay of these blockage sites.

The decay over the first 60 min was fully described by simple first order dissociation kinetics, yielding half-lives of 14-130 min, plus one site which was essentially irreversible, with a half-life >>200 min (Table 1). However, more importantly, a significant fraction of the blockage remained even after an elongation time of 3 h, consistent with drug providing an essentially permanent lesion at these sites. This essentially permanent drug occupancy was quantitated as a percentage of the blockage remaining at 3 h compared to that at 1 min, and reveals a wide range of responses (0-100%; average of 40%) at individual sites (Table 1).

Drug concentration and time dependence of adduct formation

Since the extent of barminomycin adduct formation was expected to exhibit conventional bimolecular reaction kinetics, it was of some significance to establish conditions which yielded high levels of adducts. The initiated transcription complex was therefore reacted with varying concentrations of the drug, at two extremes of reaction time (5 min and 4 h). The number of adducts formed under each condition was quantitated, and is expressed in Figure 5 as the percentage of transcripts not reaching full length (i.e. the summed mole-fraction of all blocked transcripts in the lane, expressed as a percentage of the total amount of RNA in the lane). For short reaction times, a relatively high drug concentration of 12 µM was required to ensure that half of the DNA contained at least one adduct, whereas for a 4 h exposure to drug, only ~2 µM of drug was necessary (Fig. 5). These results therefore show that at long drug reaction times (>>4 h) significant adduct formation can be expected to occur at sub-micromolar drug concentrations.


Figure 5. Concentration and time dependence of adduct formation. The initiated DNA was reacted with increasing concentration (0-60 µM) barminomycin for 5 min followed by elongation for 5 min ([open square]) or incubation of 4 h followed by elongation for 10 min ([closed square]). The total blocked transcripts were quantitated and are shown as the percentage of total RNA in the lane as a function of the barminomycin incubation time.

Table 1. Stability of adducts at individual drug sites
Site Sequence of non-template stranda Half-life (min)b % Residual blockagec
37 CAGCTA 50 ± 3 40
82 CAGCTT 130 ± 20 80
97 AAGCTG 14 ± 1 0
110 ACGCCG >>200 ~100
117 ACGCAT 40 ± 6 25
125 TGGCCG 35 ± 4 20
129 CGGCAT 90 ± 20 20
aThe underlined base shows the position at which transcription blockage was observed.
bThe half-lives of the decay of the major transcriptional blockages were determined from a first-order kinetic analysis of the relative mole-fraction of drug-occupied sites. The error has been calculated from the standard error in the rate constant.
cThe % residual blockage was determined as the percentage of blockage at 180 min compared to that at 1 min.

DNA interstrand crosslinking

Since the transcriptional blockages showed that a stable DNA adduct was formed by barminomycin, and similar adducts formed by adriamycin have been shown to function as interstrand crosslinks, it was likely that barminomycin adducts might also exhibit interstrand crosslinking properties. The assay employed to assess the crosslinking potential of barminomycin was based on that described by Hartley et al. (25) and used previously to detect adriamycin crosslink formation (12). End-labelled DNA was reacted with barminomycin (5 µM) for times up to 2 h prior to the DNA being denatured and subsequently subjected to agarose gel electrophoresis to separate single-stranded and crosslinked DNA. Since it was known that the macromolecular SN-07 complex can be destabilised with a loss of 35% of the bound drug following heating of the complex at 100°C for 5 min (4), denaturation of the DNA was performed at a lower temperature, and this was accomplished by heating at 90°C for 5 min in the presence of 45% formamide.


Figure 6. Formation of DNA interstrand crosslinks induced by barminomycin. (A) End-labelled pCC1 (25 µM bp) was incubated in the presence (+) of barminomycin (5 µM) for 1, 5, 10, 20, 30, 45, 60, 90 or 120 min (lanes 7-15) and similarly in the absence (-) for 1, 30, 60 or 120 min (lanes 3-6). The samples, excluding lanes 1 and 2, were then denatured in a 45% formamide loading buffer at 90°C for 5 min, and the all samples were subjected to electrophoresis on a 0.8% agarose gel. DS represents double-stranded DNA, whereas SS represents single-stranded DNA. (B) Quantitation of lanes 7-15 of the gel shown in (A). The % double-stranded crosslinked DNA in each lane is shown as a function of the incubation time with barminomycin.


As seen in Figure 6A, DNA incubated in the absence of drug was completely denatured and migrated at single-stranded DNA molecular weight. A time-dependent increase in DNA resistant to strand separation was observed following treatment of DNA with barminomycin, consistent with the formation of interstrand crosslinks. Crosslinking was detected in the drug-treated samples within 20 min, and after 2 h, 65% of the drug-treated DNA contained at least one crosslink. Quantitation of crosslink formation revealed 50% of DNA was crosslinked in 1 h (Fig. 6B). Removal of unreacted drug by phenol extraction was shown to have no effect on the level of crosslinks detected.

Thermal stability of interstrand crosslink

Since the barminomycin-DNA complex was known to exhibit some thermal lability, this prompted an examination of the effect of denaturation temperature on the amount of residual crosslinked DNA following the exposure of barminomycin-treated DNA (in 45% formamide) to increasing temperatures for 5 min. The results shown in Figure 7 demonstrate that at 70°C the drug-treated DNA was ~95% crosslinked, whereas non-treated samples were completely denatured under the same conditions. A more extensive analysis of the stability of the crosslink revealed that exposure of the drug-treated template to temperatures of >70°C for 5 min resulted in some loss of the crosslinks with only 40% of the crosslinks surviving exposure to 90°C for 5 min (Fig. 7A).


Figure 7. Stability of barminomycin-DNA interstrand crosslinks. (A) End-labelled pCC1 (25 µM bp) was incubated with barminomycin (5 µM) for 120 min. The samples were then denatured in a 45% formamide loading buffer at varying temperatures (62-90°C) for 5 min and then all samples were subjected to electrophoresis on a 0.8% agarose gel. The percentage of double-stranded DNA was quantitated and is expressed as a function of the temperature of denaturation (5 min) for drug-treated ([closed square]) and control ([open square]) samples. (B) DNA was crosslinked with barminomycin as described in (A). The DNA was then denatured at 835C ([closed square]) or 90°C ([open square]) for up to 20 min. The percentage of double-stranded DNA was quantitated and expressed as a function of time of denaturation.

The stability of the crosslinks was also examined at high crosslinking levels to assess the effect of time of exposure to high temperatures (83 or 90°C). There was a near linear loss of crosslinks with time of exposure to 90°C, and this was substantially delayed at 83°C (Fig. 7B).

Rate of crosslinking

The rate of DNA interstrand crosslink formation by barminomycin was examined using three drug concentrations (1, 2.5 and 5 µM), and substantial crosslinking was observed at all drug concentrations (Fig. 8). The higher drug concentrations led to a faster rate of crosslinking and a greater extent of crosslinking. However, at each concentration there appeared to be an upper limit to the absolute level of crosslinking, indicating that the crosslinking reaction is essentially complete within ~10 h, but may require longer for extremely low levels of drug.


Figure 8. Rate of formation of interstrand crosslinks by barminomycin. End-labelled pCC1 (25 µM bp) was incubated in the presence of barminomycin ([closed square], 5 µM; [open circle], 2.5 µM; and [open square], 1 µM) for times up to 24 h. The samples were then denatured in a 45% formamide loading buffer at 85°C for 5 min, and subjected to electrophoresis on a 0.8% agarose gel. The extent of crosslinking was quantitated and is shown as a function of reaction time.


DISCUSSION

Sequence specificity

The transcription assay revealed that barminomycin avidly binds to DNA and required only micromolar levels to reduce the full-length transcript to negligible levels when reacted with the initiated complex for 4 h. The binding specificity of the drug at high occupancy sites was exclusively at 5[prime]-GC sequences. The 5[prime]-GC specificity for DNA binding by barminomycin is similar to that observed previously with other alkylating anthracylines, including adriamycin and cyanomorpholinoadriamycin (26,27). However, in contrast to adriamycin, which showed some binding at alternate sequences including isolated guanine and GG sequences, and cyanomorpholinoadriamycin which showed additional binding at GG sites, barminomycin bound exclusively at GC sequences with no other clear sequence preference. This strict sequence requirement reflects a probable difference in stereochemistry and reactivity of barminomycin (compared to adriamycin), most likely brought about by the inclusion of the carbinolamine functionality at the 6[prime][prime] position of the compound. By comparison, the carbinolamine functionality which forms the covalent bond in adriamycin induced DNA crosslinks, arises after activation of adriamycin by formaldehyde (13-16).

Stability of the barminomycin-DNA lesion

The transcription assay revealed limited decay of the adducts causing the transcriptional blockage over a 3 h time period. The half-life of the decay at the first blockage site (truncated transcript 37 nt long) was 50 min, compared to a half-life of 168 min for adriamycin at the same site and under the same conditions (28). While this indicates that the barminomycin adduct is less stable than adriamycin, this is not the case since 40% of the barminomycin at this site behaved as long-lived lesions. Overall, the barmino-mycin-DNA adducts have a wide range of half-lives (~0.1 to >>3 h) and this, together with the wide range of essentially permanent lesions at individual sites (0-100%), suggests that the DNA region flanking the site of adduct formation plays a significant role in the overall stability of these adducts. The high level of essentially irreversible lesions is consistent with the known properties of SN-07, which is readily isolated from cells, in contrast to the known lability of cellular adriamycin-DNA adducts (16,17,29).

Analysis of the stability of the lesion in terms of interstrand crosslinking revealed that the DNA interstrand crosslink formed by barminomycin in vitro is relatively heat stable (40% remaining after exposure to 90°C for 5 min), similar to that observed with the macromolecular antibiotic SN-07 (40% remaining after exposure to 100°C for 5 min) (5). The stability of the barminomycin crosslink towards heat is significantly greater than that observed for adriamycin crosslinks which have an overall melting temperature of 67°C when exposed to heat for 10 min (24).

Structure of lesion

Previous studies in which a similar drug, carminomycin III (which resembles barminomycin except it does not contain the carbinolamine linkage; 28) was bound to calf thymus DNA and subsequently heated at 100°C for 5 min, have shown that all of the drug chromophore was released from the DNA (5). Since this compound does not contain the carbinolamine moiety it appears that it is unable to form covalent adducts with DNA. The presence and reactivity of the carbinolamine moiety of barminomycin is therefore expected to be important in determining its biological properties.

The apparent similarity between the barminomycin-DNA adduct and the adriamycin-DNA adduct, together with the greater overall stability of this adduct compared to the adriamycin lesion, and the known requirement for the carbinolamine group for DNA crosslink formation, suggest that barminomycin forms an adduct with guanine involving a heat labile aminal linkage (i.e. a linkage from the N-2 of guanine to the nitrogen of the amino sugar of barminomycin), identical to that of the adriamycin-DNA lesion. It should be noted, however, that the high degree of essentially irreversible barminomycin adducts was not observed with adriamycin and is consistent with the general observation that barminomycin-DNA adducts are more stable than adriamycin-DNA adducts. The additional stability of the barminomycin adduct presumably arises from additional H-bonds formed between the carbinolamine and DNA, compared to adriamycin lesion.

Conclusions

The collective information obtained from the transcriptional and crosslinking assays indicates that barminomycin acts on DNA by the rapid formation of adducts which are highly selective for 5[prime]-GC sequences of DNA. These adducts stabilise the DNA sufficiently such that they behave as a functional interstrand DNA crosslink. It is most likely that such a crosslink involves simultaneous intercalation of the aglycone chromophore and covalent bond formation between the carbinolamine at the 6[prime][prime] position of barminomycin and the N-2 of guanine of DNA. Unlike adriamycin, which has the ability to form crosslinks after activation by formaldehyde, barminomycin forms these stable lesions without any such prior activation as it is already in an essentially activated form.

ACKNOWLEDGEMENT

This work was carried out with support from the Australian Research Committee.

REFERENCES

1. Kikuchi,Y., Niwano,M., Yajima,N., Nakamura,G. and Miyata,N. (1985) J. Antibiot., 38, 1670-1676. MEDLINE Abstract

2. Yajima,N., Ishida,S., Miyata,N., Kishi,T. and Kawanishi,G. (1989) Mutat. Res., 210, 165-172. MEDLINE Abstract

3. Kimura,K., Nakayama,S., Koyama,T., Shimada,S., Kawaguchi,N., Miyata,N., Takeshita,Y. and Kawanishi,G. (1987) J. Antibiot., 40, 1353-1355. MEDLINE Abstract

4. Kimura,K., Nakayama,S., Miyata,N., Takeshita,Y. and Kawanishi,G. (1988) J. Antibiot., 41, 411-414. MEDLINE Abstract

5. Kimura,K., Takahashi,H., Nakayama,S., Miyata,N. and Kawanishi,G. (1989) Agric. Biol. Chem., 53, 1797-1803.

6. Kimura,K., Nakayama,S., Miyata,N. and Kawanishi,G. (1989) J. Antibiot., 42, 127-131. MEDLINE Abstract

7. Kimura,K., Takahashi,H., Takaoka,H., Miyata,N. and Kawanishi,G. (1990) Agric. Biol. Chem., 54, 1645-1650.

8. Uchida,T., Imoto,M., Takahashi,Y., Odagawa,A., Sawa,T., Tatsuta,K., Naganawa,H., Hamada,M., Takeuchi,T. and Umezawa,H. (1988) J. Antibiot., 41, 404-408. MEDLINE Abstract

9. De Vita,V.T., Hellman,S. and Rosenberg,S.A. (1989) Cancer: Principles and Practice of Oncology, 3rd Ed. Lippincot, Philadelphia, PA.

10. Arcamone,F. (1981) Doxorubicin Anticancer Antibiotics. Academic Press, New York, NY.

11. Weiss,R.B. (1992) Semin. Oncol., 19, 670-786. MEDLINE Abstract

12. Cullinane,C., van Rosmalen,A. and Phillips,D.R. (1994) Biochemistry, 33, 4632-4638. MEDLINE Abstract

13. Taatjes,D.J., Gaudiano,G. and Koch,T.H. (1997) Chem. Res. Toxicol., 10, 953-961. MEDLINE Abstract

14. Phillips,D.R. and Cullinane,C. (1999) In Creighton,T.E. (ed.), Encyclopedia of Molecular Biology. John Wiley and Sons, New York.

15. Taatjes,D.J., Gaudiano,G., Resing,K. and Koch,T.H. (1996) J. Med. Chem., 39, 4135-4138. MEDLINE Abstract

16. Zeman,S.M., Phillips,D.R. and Crothers,D.M. (1998) Proc. Natl Acad. Sci. USA, 95, 11561-11565. MEDLINE Abstract

17. Kimura,K., Nakayama,S., Miyata,N. and Kawanishi,G. (1989)Agric. Biol. Chem., 53, 1805-1810.

18. Kimura,K., Morinaga,T., Miyata,N. and Kawanishi,G. (1989) J. Antibiot., 42, 1838-1843. MEDLINE Abstract

19. Cullinane,C. and Phillips,D.R. (1993) Nucleic Acids Res., 21, 1857-1862. MEDLINE Abstract

20. Skorobogaty,A., White,R.J., Phillips,D.R. and Reiss,J.A. (1988)Drug Des. Deliv., 3, 125-152. MEDLINE Abstract

21. White,R.J. and Phillips,D.R. (1989) Biochemistry, 28, 6259-6269. MEDLINE Abstract

22. Phillips,D.R., White,R.J., Trist,H., Cullinane,C., Dean,D. and Crothers,C.M. (1990) Anticancer Drug Des., 5, 21-29. MEDLINE Abstract

23. Phillips,D.R., Moate,P.J. and Boston,R.C. (1994) Anticancer Drug Des., 9, 200-219.

24. van Rosmalen,A., Cullinane,C., Cutts,S.M. and Phillips,D.R. (1995) Nucleic Acids Res., 23, 42-50. MEDLINE Abstract

25. Hartley,J.A., Berardini,M.D. and Souhami,R.L. (1991) Anal. Biochem., 193, 131-134. MEDLINE Abstract

26. Cullinane,C. and Phillips,D.R. (1992) Biochemistry, 31, 9513-9519. MEDLINE Abstract

27. Cullinane,C. and Phillips,D.R. (1990) Biochemistry, 29, 5638-5646. MEDLINE Abstract

28. Cullinane,C. and Phillips,D.R. (1993) Nucleic Acids Res., 21, 1857-1862. MEDLINE Abstract

29. Kimura,K., Koyamam,T., Nakayama,S., Tamura,K., Miyata,N. and Kawanishi,G. (1988) J. Antibiot., 41, 1918-1921. MEDLINE Abstract

*To whom correspondence should be addressed. Tel: +61 3 94792182; Fax: +61 3 94792467; Email: d.phillips@latrobe.edu.au

This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 26 Mar 1999
Copyright©Oxford University Press, 1999.
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Cancer Res.Home page
S. M. Cutts, A. Rephaeli, A. Nudelman, I. Hmelnitsky, and D. R. Phillips
Molecular Basis for the Synergistic Interaction of Adriamycin with the Formaldehyde-releasing Prodrug Pivaloyloxymethyl Butyrate (AN-9)
Cancer Res., November 1, 2001; 61(22): 8194 - 8202.
[Abstract] [Full Text] [PDF]

Home page
 Nucleic Acids ResHome page
B. S. Parker, S. M. Cutts, C. Cullinane, and D. R. Phillips
Formaldehyde activation of mitoxantrone yields CpG and CpA specific DNA adducts
Nucleic Acids Res., February 15, 2000; 28(4): 982 - 990.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (197K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (9)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Perrin, L. C.
Right arrow Articles by Phillips, D. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Perrin, L. C.
Right arrow Articles by Phillips, D. R.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?