Effects of DMA Superhelical Changes Induced by Ethidium … · DMA-degrading Activity of Two Antitumor Antibiotics, Bleomycin and Phleomycin1 Cheng-Hsiung Huang,2-3 Christopher K

[CANCER RESEARCH 43, 2849-2856, June 1983]0008-5472/83/0043-0000$02.00

Effects of DMA Superhelical Changes Induced by Ethidium Bromide on theDMA-degrading Activity of Two Antitumor Antibiotics, Bleomycinand Phleomycin1

Cheng-Hsiung Huang,2-3 Christopher K. Mirabelli,3 Seymour Mong,3 and Stanley T. Crooke3

Department of Pharmacology, Baylor College of Medicine, Texas Medical Center, Houston, Texas 77030

ABSTRACT

The effects of changes in the conformational state of DNA onthe single-strand and double-strand breakage activity of two

antitumor antibiotics, bleomycin (BLM) A2 and phleomycin Di,have been studied by the gel electrophoretic analysis of the drug-

degraded PM2 phage superhelical DNA pretreated with an intercalating agent, ethidium bromide (EB). Both the single-strandand double-strand breakage activities of BLM A2 increased as

the negatively superhelical turns of native PM2 DNA were gradually removed by intercalation with increasing EB concentrations.The activities peaked when DNA was completely relaxed andgradually decreased as the higher concentrations of EB twistedDNA into the positively superhelical form. The decrease in breakage activity was not due to any inhibitory effect of EB at higherconcentrations, since treatment of the relaxed Form Io DNA with

low EB concentrations also reduced the activity. In contrast toBLM A2, phleomycin D! responded minimally to DNA conformational changes, which suggested further that the two drugs mayreact with DNA differently. The differential responses of BLM A2activity towards different DNA conformational states may havebiological implications, since DNA in cells may exist in differentconformational states relating to various gene functions. Thecurrent study may serve as a model for studying combinedeffects of intercalative and nonintercalative antitumor antibioticswhich are used frequently in combination treatments of cancer.

INTRODUCTION

BLMs4 are a family of glycopeptide antitumor antibiotics used

in clinical treatment of certain human cancers (6). The antitumoractivity of BLMs is thought to be related to their ability to induceSS and DS DNA breaks (15, 33, 48, 50). The NH2-terminal half-molecule of BLM may complex with Fe(ll) and oxygen (8, 27, 46)and may produce free radicals which cause DNA strand breaks.The COOH-terminal half-molecule may interact with specific DNA

sequences (9, 18, 23, 25, 26, 31, 49), in a specific manner (16,18, 26), such that free radicals may cause either SS or DS (orboth) DNA breaks.

We have reported previously the importance of the NH2-terminal BLM half-molecule portion, which presumably coordinates with Fe(ll), in the BLM-DNA interaction and the degradationof DNA (17-19). Recently, a similar concept has also been

1This work was supported in part by a grant from the Bristol Laboratories andby Grant Ã‡A10893-P12 from the National Cancer Institute.

2To whom requests for reprints should be addressed, at Smith Kline & FrenchLaboratories, P. O. Box 7929 (J-100), Philadelphia, Pa. 19101.

3 Present address: Smith Kline & French Laboratories, P. O. Box 7929, Phila

delphia, Pa. 19101.4The abbreviations used are: BLM, bleomycin; SS, single strand; DS, double

strand; Fe(ll), ferrous iron; EB, ethidium bromide; PLM, phleomycin.Received December 18,1981 ; accepted March 4,1983.

proposed by Povirk ef al. (42). Although the COOH-terminal BLMhalf-molecule portion has been suggested to interact with DNA,the DNA-degradative activity of several BLM analogues was not

correlated with their interaction with DNA (18, 23) but wascorrelated with the Fe(ll)-induced quenching effect on the BLMfluorescence (17). Structural modifications on the NH2-terminalBLM half-molecule portion affected not only the BLM interaction

with DNA but also the extent of production of either SS or DSDNA breaks (19). These observations led us (17, 18) to suggestthe importance of a proper conformation of the Fe(ll)-BLM com

plex in the action of BLM on DNA. Similar emphasis has alsobeen suggested by Sugiura et al. (47) and Oppenheimer ef al.(37). It is possible that the overall conformational state of thecomplex of BLM, Fe(ll), and DNA is also important in the degra-

dative activity.The overall conformational arrangement of the drug/DNA/Fe(ll)

complex may also be affected by the conformational state of theDNA, i.e., at the level of either the superstructure (superhelix) orthe double helix if a superhelical DNA is used. Thus, we havestudied here the effects of the conformational changes of thesuperhelical PM2 DNA induced by a DNA intercalator, EB, onthe DNA-degradative activity of BLM drugs such as BLM A2 and

PLM Di. The observation that BLM A2 but not PLM DTproducedDS breaks, in conjunction with other observations, has led us tosuggest that the 2 drugs may differ in the mode of interactionwith DNA (19). The reduction of a double bond in the coplanarbithiazole moiety (for structure, see Ref. 17) renders the intercalation of PLM D, with DNA more difficult. Thus, the EB-induced

conformational changes in DNA can serve as a system forprobing the differences between these 2 drugs more extensively.

Furthermore, we considered it important to study the effectsof DNA conformation on the degradative activity of BLMs, sinceDNA in cells may exist in different conformational forms whichare organized into various types of higher-ordered genomic

structures such as nucleosomes and superhelical or solenoidalarrays of polynucleosomes (12,16, 35,44, 54) and nucleoids (5,28). A portion of this study has been presented at a symposium(7).

MATERIALS AND METHODS

Materials. BLM A2 and PLM D, were obtained from Bristol Laboratories, Syracuse, N. Y. Covalently closed, circular, superhelical Form IPM2 phage DNA was isolated as described previously (19). The DNApreparations used contained at least 85% Form I superhelical DNA. Therelaxed, circular, duplex Form Io PM2 DNA was prepared by treatment

of Form I DNA with calf thymus topoisomerase (32). EB, Tris, borate,disodium EDTA, and /3-mercaptoethanol were obtained from SigmaChemical Co., St. Louis, Mo. Agarose-ME was purchased from Miles

Laboratories, Elkhart, Ind.Assays of DNA Breakage Activity of BLM and PLM by Gel Electro-

JUNE 1983 2849

Research. on October 9, 2020. © 1983 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

C-H. Huang et al.

phoresis. Details of the assay procedures have been described previously (19, 30). A reaction mixture (final volume, 0.15 to 0.20 ml) containing 50 ITIMborate buffer (pH 9.5), 66 mw NaCI, 110 to 140 ÃŸMPM2 FormI or Form Io DNA preparations, and various amounts of EB, if added,

was incubated for 5 min at room temperature. After incubation, 25 mwÃŸ-mercaptoethanol and a given amount of BLM A2 or PLM Di, from 17to 30 nw, were added to start the DNA-degradative activity. After 30 min

at room temperature, the degradative activity was terminated with theaddition of an equal volume of a mixture containing 56% glycerol (v/v),50 rriM EDTA, and 0.5% bromophenol blue (w/v). Aliquots containing 0.8to 1.5 Â¿jgDNA were layered onto a 0.9% agarose slab gel and wereelectrophoresed in a horizontal slab gel apparatus for 6 to 8 hr at roomtemperature with a 40 mw Tris-HCI buffer containing 5 ITIM sodium

acetate and 1 ITIMEDTA, pH 7.8. After electrophoresis, gels were stainedwith EB (0.5 iig/ml) in the electrophoresis buffer for at least 2 hr. Thestained gels were then excited with a transilluminator (Ultra-Violet Products, Inc.) and photographed with a Polaroid CU-5 Land Camera

equipped with a No. 8 Kodak Wratten gelation filter (Eastman KodakCo., Rochester, N. Y.) and Type 665 Land Films. The negative film ofgel was used for determination of SS and DS DNA breakage activity bydensitometry.

Quantitation of SS and DS DNA Breaks by Densitometric Scanningsof Negative Films of Gels. The negative films of the EB-stained gelpatterns of the drug-treated PM2 DNA were scanned with a recording

Transdyne General Densitometer equipped with an automatic computingintegrator. As illustrated in Chart 1, the production of the nicked, relaxed,duplex Form II DNA and the linear Form III DNA from the covalentlyclosed, superhelical Form I was considered to be the result of SS andDS breaks, respectively (19, 25, 26, 30). The relative stainability of theForms I, II, and III of DNA in the agarose gels was tested experimentallywith known amounts of respective DNA preparations. In agreement withprevious reports (25, 26), it was found that at the same amount of DNA,the EB fluorescence intensity of Form I DNA was only 70% of that ofForm II or Form III DNA; thus, this factor was used to normalize allobservations. In agreement with the report of Prunell ef a/. (43), at theDNA concentration range used in the present study, the densitometricreadings were linearly proportional to the DNA concentrations.

The stainability of DNA in agarose gel was not significantly affectedby the pretreatment of DNA with EB since, during the electrophoresis ofthe gel, most of the EB bound to DNA was dissociated from DNA. Thishas been varified by quantitative measurements of gel patterns of DNApretreated with various amounts of EB (data not shown) and comparisonto DNA not previously treated with EB. To retain EB bound to DNA, thegels and the electrophoresis buffer must contain an equivalent amountof EB(11).

Preparation of Topoisomerase-relaxed pBR322 Plasmid DNA and

Analysis of Degradation by BLM A2. Purified pBR322 plasmid DNA (30)was treated with various amounts of calf thymus topoisomerase (0 to

PM2 DNA SYSTEM

+EB

ssb

dsb

+EB

dsb

Chart 1. PM2 DNA system used for studies of the effects of EB-induced DNAconformational changes on the DNA-degradative activity of BLM A2 and PLM D,.I', native, negatively superhelical DNA; f, completely relaxed DNA with virtually no

superhelical turns; //, nicked DNA; ///, linear DNA; ssi>, SS break; dsb, DS break(see text for explanation).

0.2 nQ per Mg of DNA) at 37Â°for 1 hr in a buffer of 50 HIM Tris-HCI (pH

8), 0.2 M NaCI, and 5 ITIMEDTA. The preparations were then subjectedto EB-CsCI gradient to remove the small amount (<30%) of broken DNA

(Forms II and III) resulting from the topoisomerase treatment. As performed routinely in the purification of Form I DNA from either PM2 phage(19) or pBR322 plasmid (30, 32), the EB in the EB-DNA complex fromthe EB-CsCI gradient was removed by extracting 5 times with NaCI-

saturated isopropyl alcohol and 5 times with ether. Ether was thenremoved by bubbling air through solution. Form I DNA was then precipitated with 70% ethanol. The purified preparations contained 88 to 92%unbroken DNA as assayed from EB-agarose gel which separates all of

the unbroken DNA molecules from Form II and Form III DNA. The DNApreparations were treated with BLM A2 (20 nM) in borate buffer (pH 9.5)at 37Â°for 30 min and analyzed by agarose gels with or without EB.

RESULTSDNA Breakage Activity of BLM A2 with EB-treated PM2

DNA. Chart 1 illustrates the EB/PM2 DNA system used forstudying the effects of DNA conformational changes on the DNA-

degradative activity of BLM A2. The addition of low concentrations of EB causes the unwinding of the double-helix structures

and the reduction of the number of the negatively supercoiledturns of the native Form r (or I) PM2 DNA. Addition of more EBeventually removes all the negatively supercoiled turns of DNAto form the relaxed, double-stranded, circular Form IoDNA. Uponfurther addition of EB, the Form IoDNA supercoils in the opposite

direction to become positively supercoiled Form r DNA. Thistype of superhelical change is a relatively specific effect ofintercalative agents on covalently closed, circular DNA (2).

The system used for studying the SS and DS breakageactivities of BLM A2 and PLM DT has been described previously(19, 25, 26, 30). The introduction of a SS break (nick) convertsForm r, Io, or r DNA to the nicked, relaxed, circular Form IIDNA. The introduction of a DS break converts Form r, Io, r, or

II DNA to a linear Form III DNA. These forms of PM2 DNA differin mobility in agarose gels. In a 0.9% gel, Forms r and r havethe fastest mobility, followed by Form III, then by Form Io, andthen by Form II DNA which comigrates with Form Io. Thus, the

production of Form III DNA results primarily from the DS breaks.Fig. 1 shows the agarose gel electrophoretic pattern of the

PM2 DNA products after DNA was pretreated with increasingconcentrations of EB and then degraded with a fixed concentration of BLM A2 (20.7 nw). Lane a shows the untreated, super-

helical Form I DNA preparation which primarily contained Form IDNA and a trace amount of Form II DNA. When treated withBLM A2 in the absence of EB (Lane b), both the nicked Form II(upper band) and the linear Form III (middle band) were producedat the expense of Form I DNA (lower band). Lanes c to m showthe BLM A2-degraded DNA products when DNA was pretreated

with increasing concentrations of EB. These results show that,as the EB concentration was increased, the production of bothForm II and Form III DNA increased and then decreased afterreaching a maximum at an EB/DNA ratio of between 0.10 (Lanef) and 0.11 (Laneg).

The gel pattern of PM2 form I DNA after treatment with EB atEB/DNA equal to ratios used in Fig. 1, but without addition ofBLM A2, was also obtained (not shown). No production of eitherForm II or Form III DNA from Form I DNA was detected. Withthe increasing concentration of EB, the mobility of the Form IDNA band was slightly reduced and was then restored as moreEB was added, in a manner similar to that shown by the residualForm I DNA band after treatment with EB and BLM A2 (Fig. 1).

2850 CANCER RESEARCH VOL. 43



Effects of DNA Conformation on BLM and PLM Activities

abcdefghiJklm

Fig. 1. Agarose gel electrophoretic pattern of DMA products after treatment of PM2 Form I DMA (11.5 fiM) with a fixed amount of BLM A2 (20.7 nw). PM2 Form I DMAwas pretreated with increasing concentrations of EB. Gel mobility, top io bottom; fast-moving band, Form I (or I") DMA; slow-moving band, Form II DMA; intermediate

band, Form III DNA; Lane a, untreated Form I DNA; Lanes b to m, BLM As-degraded DNA which was pretreated with EB at EB/DNA concentration ratios of 0 (Lane b),0.02 (Lane c), 0.04 (Lane d), 0.07 (Lane e), 0.10 (Lane I), 0.1 1 (Lane g), 0.13 (Lane n), 0.15 (Lane /), 0.17 (Lane /), 0.19 (Lane k), 0.22 (Lane /), and 0.26 (Lane m).

60

Â«

Â«DNA

20

! 1 I

A

v0-*

0.1 0.2CEB3/CDNA]

in

Chart 2. Percentage distribution of DNA conformational forms after treatmentof Form I DNA (115.8 ^M) with 20.7 nM BLM A2. DNA was pretreated with increasingconcentrations of EB. Percentage distribution was measured from the gel electrophoretic pattern shown in Fig. 1. O, Form I DNA; A, Form II DNA; G, Form III DNA.

Intercalation of EB induces superhelical relaxation and thusshould change the gel electrophoretic mobility of superhelicalDNA. However, the mobility changes are usually not obtained inslab gels because of the dissociation of bound EB from DNAduring electrophoresis. To observe these changes, disc gelshave to be used, and each of the gels has to contain equivalentamounts of EB throughout the electrophoresis to retain thebound EB (11).

The gel pattern shown in Fig. 1 was analyzed quantitatively,and the results were shown in Chart 2, in which the percentagedistribution of each DNA conformational form was plotted againstEB/DNA ratios. The untreated DNA preparation contained 90%Form I DNA and 10% Form II DNA. Degradation of DNA withBLM A2 in the absence of EB pretreatment resulted in 61% Form

I, 33% Form II, and 6% Form III DNA. The maximal productionof Form II DNA (54%) occurred at an EB/DNA ratio of between0.10 and 0.1 1, whereas that of Form III (15%) occurred at a ratioof 0.10. These results show an approximately 2-fold increase in

the production of both Form II and Form III DNA as compared tothat in the absence of EB treatment, and the SS/DS DNA breakratio did not change significantly. At higher ratios, the productionof Forms II and III decreased gradually. At a ratio of 0.26, therewere 39% Form II DNA and 7% Form III DNA.

Degradation of Form Io DNA with BLM A2. To exclude the

possibility that EB at concentrations higher than a ratio of 0.10inhibited the DNA-degradative activity of BLMs and thus contrib

uted to the decline in the activity of BLM A2 shown in Chart 2,we have studied (Fig. 2) the degradation by BLM A2 of the EB-pretreated PM2 Form Io DNA which was produced by treatment

of PM2 Form I DNA with calf thymus topoisomerase (32). SinceForm Io DNA produced by topoisomerase is virtually totally

relaxed, the addition of a small amount of EB (EB/DNA < 0.1)turns Form Io to r DNA. Lane a shows that untreated Form Io

DNA which, due to the presence of a high concentration of EB(0.5 Mg/ml) Â¡nthe gel electrophoretic system, has become positively superhelical and thus migrated to a position equivalent tothat of a native superhelical DNA such as Form I DNA. In a gelwithout EB, Form Io DNA would comigrate with Form II DNA. InLane b, the Form Io DNA was degraded with BLM A2 in theabsence of EB pretreatment. The Form Io DNA was degradedalmost completely and produced Form II (slow-moving band) and

Form III DNA. Lanes c to j show the pattern of degraded DNAproducts from BLM A2 treatment of Form IoDNA which has been

pretreated with increasing amounts of EB (from an EB/DNA ratioof 0.022 to 0.20). It is clear that when PM2 Form Io DNA was

pretreated with increasing concentrations of EB, the productionof both Form II and Form III DNA decreased, and the retentionof the undegraded Form Io increased.

The quantitative relationship of the changes among 3 DNAforms described in Fig. 2 is shown in Chart 3. The untreatedPM2 DNA preparation shown in Fig. 2, Lane a, contains 85%Form Io DNA, 15% Form II DNA, and virtually no Form III DNA.The degradation by BLM A2 of PM2 Form Io DNA which was notpretreated with EB resulted in 7% DNA in Form Io, 59% DNA in

JUNE 1983 2851



C-H. Huang et al.

Form II, and 34% DNA in Form III species. Treatment of DMAwith a low concentration of EB, e.g., an EB/DNA ratio of 0.022,caused a significant inhibition of DNA-degradative activity of BLMA2 and resulted in 28% DMA in Form Io,52% in Form II, and 20%in Form III species. Pretreatment of Form Io DNA with increasing

concentrations of EB increased the extent of inhibition on thedegradative activity of BLM A2. As a result, the percentage ofDNA remaining in the undegraded Form Io increased, whereas

that of Forms II and III decreased significantly. Thus, the inhibitionby even low concentrations of EB on the BLM A2-induceddegradation of Form Io DNA contrasts with the stimulatory effect

of low EB concentrations (EB/DNA < 0.1) on the degradation ofthe negatively superhelical Form I DNA (Chart 2). The observationis consistent with the interpretation of results shown in Fig. 1 inwhich the increase in positive superhelical turns of DNA reducedthe degradative activity of BLM A2.

Effects of Topoisomerase-induced Superhelical Relaxation

on Degradative Activity of BLM A2. To substantiate our conclusion that decreasing negative superhelical turns increases thedegradative activity of BLM A2, we have prepared circularpBR322 DNA preparations which were relaxed to different extents by the treatments of increasing amounts of calf thymustopoisomerase. All preparations were repurified through EB-CsCI

gradient to remove the small amount (<30%) of Forms II and IIIproduced as a result of topoisomerase treatments. These preparations are shown in Fig. 3. Each of the topoisomerase-treated

preparations shows a series of DNA bands differing by onesuperhelical turn. Treatment with more topoisomerase resultedin more reduction of the gel mobility of DNA due to moresuperhelical relaxation. These preparations were then treatedwith the same amount of BLM A2 and analyzed with an EB gel(not shown), which converted all of the unbroken DNA moleculesinto the positively superhelical form and thus separated thesefrom Form II and III DNA molecules produced as a result of BLMA2 treatment. The results are shown in Table 1, which shows anincrease in the production of Form II DNA with treatments ofincreasing amounts of topoisomerase. These results are consistent with those obtained from the EB-treated DNA prepara

tions.Degradation of EB-treated Form I DNA with PLM D,. Fig. 4

shows the gel pattern, and Chart 4 shows the quantitativeanalysis of the DNA products after the EB-treated Form I DNA

was degraded with a fixed amount of PLM DL The resultsconfirmed our previous observation (19) that, under the conditions used, PLM D, caused primarily SS DNA breaks, resulting

60 A-

20 -

0.04 0.08 0.12 0.16 0.20

CEB]/CDNA]

Chart 3. Percentage distribution of DNA forms after treatment of PM2 Form IoDNA with BLM A2. DNA was pretreated with increasing concentrations of EB.Results were measured from the gel electrophoretic pattern shown in Fig. 2. O,undegraded Form Io DNA; A, Form II DNA; D. Form III DNA.

Table 1

Degradative activity of BLM A2 on pBR322 Form I DNA partially relaxed byincreasing amounts of calf thymus topoisomerase

Analysis was described in "Materials and Methods."

% of DNA distribution after BLM A2 treatment"

Topoisomerase3 ( Form Ie Form II Form III

01.02.55.057524943404447523445

3 Extents of relaxation by topoisomerase were shown in Fig. 3.6 DNA preparations which were not treated with BLM A2 contained 90% Form

I, 10% Form II, and 0% Form III DNA.0 Form I DNA identified in EB gels derives from unbroken Form Io DNA converted

to Form T by the presence of EB in the gels.

60

20

~

O 0.1 0.2 0.3CEB3/CDNA3

Chart 4. Percentage distribution of DNA forms after treatment of Form I DNAwith PLM Di in the presence of increasing EB concentrations. Data were measuredfrom the gel electrophoretic pattern shown in Fig. 4. O, Form I DNA; A, Form IIDNA; D, Form III DNA.

in the production of Form II DNA. In contrast with the observations for BLM A2 activity (Chart 2), the production of Form II DNAby PLM D! was not affected significantly by the pretreatment ofDNA with increasing amounts of EB.

Degradation of EB-treated Form Io DNA with PLM D,. Chart

5 shows the quantitative analysis of the production of forms IIand III DNA from EB-pretreated Form Io DNA which was pro

duced by topoisomerase. The data were obtained from the scansof EB-containing gel patterns of the PLM D,-degraded DNA

products (not shown). The results show clearly that pretreatmentwith EB had little effect on the extent of Form IoDNA degradation

by PLM D,.

DISCUSSION

Results in this study show that both the SS and the DSbreakage activities of the BLM A2 responded to the alterationsin the superhelicity of PM2 DNA. Relaxed DNA form was significantly more sensitive to degradation by BLM A2 than was eitherthe negative or the positive superhelical DNA. When Form I DNAwas treated with EB, the sensitivity to degradation by BLM A2increased until an EB/DNA ratio of 0.10 to 0.11 was used, at





40

20

0 0.04 0.08 0.12 0.16 0.20

CEBI/CUNA:Chart 5. Percentage distribution of DMA forms after-treatment of PM2 Form Io

DMA (125.5 MM)with a fixed amount of PLM D, (21 nM). Before treatment of PLMD,, DNA was incubated with increasing concentrations of EB. For gel electropho-resis, both gel and electrophoretic buffer contained EB (0.5 vg/ml). O, undegradedForm Io DNA; A, Form II DNA; D, Form III DNA.

which point Form I DNA was maximally relaxed (i.e., Form Io

DNA). That the decrease in sensitivity to degradation by BLM A2observed at EB/DNA ratios >0.11 was due to the introductionof the positive superhelicity in the DNA rather than to theinhibitory effect of higher concentrations of EB on the degrada-tive activity is demonstrated by the following observations: (a)The EB/DNA ratio of 0.10 (Chart 2) required to induce themaximal degradative activity of BLM A2 approximates the ratioof 0.091 we have reported previously (20) for EB to completelyremove the superhelical turns of PM2 DNA; and (b) the inductionof the relaxed Form Io DNA by EB at EB/DNA ratios <0.10 into

positive superhelical form resulted in a reduction of the degradative activity of BLM A2. In agreement with these interpretations, we have also observed that BLM A2 was more activetowards the more relaxed pBR322 DNA prepared by topo-

isomerase treatment.The reason for the increase in DNA-degradative activity of

BLM towards the superhelically relaxed DNA is unclear to us atpresent. The relaxation in the DNA tertiary structure may havefacilitated the interaction of BLM A2 with specific sites of DNAmolecules, thus increasing the rate of degradation. It is alsopossible that the relaxation of the DNA superhelical structuremay open new sites in DNA molecules for degradation by BLMA2. Whether the site-sequence specificities of the BLM A2-in-

duced DNA breaks are altered by the changes in DNA superhelicity is currently under study.

The phenomenon is not unique to EB, since we have observeda similar effect induced by a bifunctional intercalative agent,BBM-928A (20, 21). The maximal increase in the BLM A2 degradative activity was observed at a concentration of BBM-928A

which induced a maximal superhelical relaxation. Thus, the conformation of the DNA clearly affected the activity of BLM A2.

In genetic structures of eukaryotic systems, DNA may assumespecific conformational states such as superhelical or solenoidalforms found in higher-ordered structures such as nucleosomes

and their polynucleosomal arrays (12,14,16, 35, 38,44, 54) andnucleoids (26, 28). Nucleoids were suggested to contain super-

helical DNA with topological constraint similar to that of circularDNA. Unfolding or relaxation of nucleosomes by EB has beenwell studied (39, 55). These superstructures can provide a wayof controlling gene activities such as transportation, repair, and

replication (1). Changes of these structures have been proposedto be associated with certain gene functions and regulations. Forexample, transcriptionally active genomic structures may bepartially unfolded or may be different from transcriptionally inactive structures (13, 34,40, 53). A decrease in number of topological turns was associated with differentiation in nucleoids ofFriend erythroleukemia cells (28). Relaxation of supercoiling wasrelated to initiation and elongation of DNA synthesis (29). Super-helix-altering enzymes such as topoisomerases (including gyr-

ase) have been suggested to play an important role in genefunction (4, 22, 24, 34, 36, 51). Thus, the differences in thedegradative activity of BLM A2, depending on the conformationalstates of genomic DNA, may have different biological consequences.

In addition, the combined effects of an intercalating agent, EB,and a DNA-degrading agent, BLM A2, may be of interest when

considering the use of combination chemotherapy involving BLMand DNA-intercalative antitumor antibiotics. Several groups of

intercalative antibiotics or synthetic compounds with antitumoractivities have been reported, and some of them have been usedin clinical treatment of certain cancers. These include the mon-

ofunctional agents such as anthracyclines (10) and the bifunctional agents such as the diacridines (3), pyridocarbazole dimer(45), echinomycin (52), and BBM-928A (20, 21).

In contrast with BLM A2, the activity of PLM DT was notaffected significantly by the EB-induced DNA superhelix confor

mational changes. The reason for this difference is unclear. PLMDT is actually 7,8-dihydrobleomycin B2, and one of the double

bonds in one of the 2 coplanar thiazole rings is reduced byhydrogÃ©nationwhich renders the intercalation with DNA difficult.The different responses between BLM A2 and PLM D, may haveresulted from the thiazole ring modification, since BLM B2 behaved in a manner similar to that of BLM A2 (data not shown).Several differences between BLM and PLM in the interactionwith DNA have been summarized (19). For example, both BLMA2 and BLM B2 produced SS and DS breaks in a similar manner,whereas PLM D, produced essentially only SS breaks (19).Recent studies of Povirk et al. (42) also suggested differencesbetween BLM and PLM in their interaction with DNA. Thedestruction of the coplanar bithiazole rings could affect thespatial arrangement of the 2 rings and of the COOH-terminal

amine relative to the rest of the drug molecule. Consequently,the intercalation of bithiazole with DNA was prevented, and theinteraction of COOH-terminal amine with DNA was altered. Thus,it is possible that the overall conformational arrangement of DNA-

drug complex is different between BLM A2 and PLM Di, and asa result these 2 drugs responded differently to changes in DNAsuperhelicity.

Another possible difference in the spatial arrangement of theDNA-drug complex between the 2 drugs could be that, while

interacting with DNA, BLM A2 may form dimers through stackingof the coplanar bithiazole moieties, whereas PLM D, may fail todo so. The activity of the dimeric BLM A2 molecules, the existence of which has been suggested by Lloyd ef al. (26) to accountfor the production of SS as well as DS breaks, may be moresensitive to the superhelical changes, whereas the activity of themonomeric PLM D^ which produced mainly SS breaks (19), maybe less sensitive.

Our observation that the relaxed PM2 DNA was more sensitiveto the BLM A2 degradative activity differs from those reported

JUNE 1983 2853



C-H. Huang et al.

by Lloyd ef al. (26) and Povirk ef al. (41). Using a gel electropho-

retic system, Lloyd ef al. (26) found that Form I PM2 DMA wasslightly more sensitive than Form Io DMA to the clinical prepara

tion of bleomycins. By studying the sedimentational properties,Povirk ef al. (41) reported the same preference to argue in favorof the presence of the intercalation of the bithiazole moiety withDMA. Currently, we have no clear explanation for this discrepancy. However, we want to emphasize that our conclusionswere derived from studies of the effects on the BLM A2 activitywith systematic changes in superhelicity of the same DMA preparation. In our hands, we observed some variations in the DNA-

degradative activity of BLM A2 depending upon the Form I DMApreparations. We have observed that the concentration of BLMA2 to cause breakage of 50% of DNA molecules could vary from15 to 32 pM in different Form I DNA preparations (19). We alsoobserved variations in the degradative activity of BLM A2 ondifferent preparations of purified Form Io DNA, depending very

much on the amounts of the accompanying Form II DNA in theForm Io preparations. This type of variation may complicate anydirect comparison between Form I and Form Io DNA prepara

tions. Nevertheless, we have approached this problem by usingpreparations of pBR322 DNA treated with varying amounts oftopoisomerase and purified by EB/CsCI gradient centrifugation.These studies confirm the observations made using EB treatment.

Furthermore, the present system using EB to induce super-

helical changes encompasses the changes in negative as wellas positive superhelicities of DNA. EB-free circular DNA prepa

rations with different degrees of negative superhelicity can beprepared by the treatment of topoisomerase in a manner similarto that for our Form Io DNA, but in the presence of varying

amounts of EB, which is removed subsequently. However, sucha procedure cannot produce positively superhelical DNA forms.In our study, the presence of EB during the digestion of DNA byBLM may lead to the possibility that EB at low concentrationsmay directly stimulate and, at high concentrations, inhibit thedegradative activity of BLM (Fig. 1; Chart 3), disregarding thesuperhelical state of DNA. However, the results shown in Fig. 2and Chart 3 indicated that whether EB stimulated or inhibitedthe BLM A2 activity depended on the decrease or increase insuperhelicity of DNA rather than on the changes of the EBconcentration. The senseness of superhelicity seemed to beirrelevant. Thus, we think that the present system provides asimple system for studies of the effects of DNA superhelicalchanges and is confirmed by the studies using topoisomerase.

ACKNOWLEDGMENTS

The authors wish to thank Professor Harris Busch for his encouragement andDr. Archie W. Prestayko for his advice and suggestions. We also thank MelÃ³meJan for excellent technical assistance and Mary Safrit for excellent typographicalassistance.

REFERENCES

1. Axel, R., Cedar, H.. and Felsenfeld, G. Chromatin template activity andchromatin structure. Cold Spring Harbor Symp. Quant. Biol., 38: 773-783,1973.

2. Bauer, W. R. Structure and reactions of closed duplex DNA. Annu. Rev.Biophys. Bioeng., 7: 287-313, 1978.

3. Canellakis, E. S., Shaw, Y. H., Hanners, W. E., and Schwartz, R. A. Diacridines:bifunctional intercalators. Biochim. Biophys. Acta, 481: 277-289, 1976.

4. Champoux, J. J. Proteins that affect DNA conformation. Annu. Rev. Biochem.,47:449-479. 1978.

5. Cook, P. R., and Brazell, I. A. Characterization of nuclear structures containingsuperhelical DNA. J. Cell Sci., 22: 303-324, 1976.

6. Crooke, S. T., and Bradner, W. T. Bleomycin, a review. J. Med. (Basel), 7:333-428,1976.

7. Crooke, S. T., DuVemay, V. H., Huang, C. H., Mirabelli, C., and Prestayko,A. W. In: Molecular pharmacology of several antitumor antibiotics, pp. MEI-023. 179th Annual Meeting of the American Chemical Society, March 23 to28, 1980, Houston, Texas.

8. Dabrowiak, J. C., Greenaway, F. T., Longo. W. E., van Husen, M., and Crooke,S. T. A spectroscopic investigation of the metal binding site of bleomycins A2.Biochim. Biophys. Acta, 577: 517-526,1978.

9. D'AndrÃ©a,A. D., and Haseltine, W. A. Sequence specific cleavage of DNA by

the anti-tumor antibiotics neocarzinostatin and bleomycin. Proc. Nati. Acad.Sei. U. S. A., 75: 3608-3612, 1978.

10. DiMarco, A., Arcamone, F., and Zunino, F. Daunomycin and adnamycin andstructural analogues: biological activity and mechanism of action. In: AntibioticsIll-Mechanism of Action of Antimicrobial and Antitumor Agents, pp. 101-128.Berlin: Springer-Verlag, 1975.

11. Espejo, R. T., and Lebowitz, J. A simple electrophoretic method for thedetermination of superhelix density of closed circular DNAs and for observationof their superhelix density heterogeneity. Anal. Biochem., 72: 95-103,1976.

12. Finch, J. T., and Klug, A. Solenoidal model for superstructure in chromatin.Proc. Nati. Acad. Sei. U. S. A., 73: 1897-1901, 1976.

13. Garel, A., and Axel, R. Selective digestion of transcriptionally active ovalbumingenes from oviduct nuclei. Proc. Nati. Acad. Sei. U. S. A., 73: 3966-3970,

1977.14. Germond, J. E., Hirt, B., Oudet, P., Gross-Bellard, M., and ChambÃ³n, P.

Folding of the DNA double helix in chromatin-like structures from simian virus40. Proc. Nati. Acad. Sci. U. S. A., 72: 1843-1847,1975.

15. Haidle, C. W. Fragmentation of DNA by bleomycin. Mol. Pharmacol., 7: 645-652,1971.

16. Hewish, D. R., and Burgoyne, L. A. Chromatin sub-structure. The digestion ofchromatin DNA at regularly spaced sites by a nuclear deoxyribonuclease.Biochem. Biophys. Res. Commun., 52: 504-510,1973.

17. Huang, C. H., Galvan, L., and Crooke, S. T. Quenching of fluorescence ofbleomycins by ferrous ion and its correlation with DNA-breakage activity.Biochemistry, 78: 2880-2887, 1979.

18. Huang, C. H., Galvan, L., and Crooke, S. T. Interactions of bleomycin analogueswith deoxyribonucleic acid and metal ions studied by fluorescence quenching.Biochemistry, 79: 1761-1767, 1980.

19. Huang, C. H., Mirabelli, C. K., Jan, Y., and Crooke, S. T. Single-strand anddouble-strand deoxyribonucleic acid breaks produced by several bleomycinanalogues. Biochemistry, 20: 233-238, 1981.

20. Huang, C. H., Mong, S., and Crooke, S. T. Interactions of a new antitumorantibiotic BBM-928A with deoxyribonucleic acid. Bifunctional intercalative binding studied by fluorometry and viscometry. Biochemistry, 79: 5537-5542,1980.

21. Huang, C. H., Prestayko, A. W.. and Crooke, S. T. Bifunctional intercalation ofantitumor antibiotics BBM-928A and echinomycin with deoxyribonucleic acid.Effects of intercalation on deoxyribonucleic acid degradative activity of bleomycin and phleomycin. Biochemistry, 27: 3704-3710, 1982.

22. Javahehan, K., Liu, L. F., and Wang, J. C. Nonhistone proteins HMG, andHMG2 change the DNA helical structure. Science (Wash. D. C.), 799: 1345-

1346. 1978.23. Kasai, H., Naganawa, H., Takita, T., and Umezawa, H. Chemistry of bleomycin,

XXII. Interaction of bleomycin with nucleic acids preferential binding to guaninebase and electrostatic effect of the terminal amine. J. Antibiot. (Tokyo), 37:1316-1320,1978.

24. Liu, L. F., Liu, C. C., and Alberts, B. M. Type II DNA topoisomerases: enzymesthat can unknot a topologically knotted DNA molecule via a reversible double-strand break. Cell, 79: 697-707, 1980.

25. Lloyd, R. S., Haidle, C. W., and Hewitt, R. R. Bleomycin-induced alkaline-labiledamage and direct strand breakage of PM2 DNA. Cancer Res., 38: 3191-3196, 1978.

26. Lloyd, R. S., Haidle, C. W., and Robberson, D. L. Bleomycin-specific fragmentation of double-stranded DNA. Biochemistry, 77: 1890-1896,1978.

27. Lown, J. W., and Sim, S. The mechanism of the bleomycin-induced cleavageof DNA. Biochem. Biophys. Res. Commun., 77: 1150-1157,1977.

28. Luchnik, A. N., and Glaser, V. M. Decrease in the number of DNA topologicalturns during Friend erythroleukemia differentiation. Mol. Gen. Genet., 778:459-463,1980.

29. Mattem, M. R., and Painter, R. B. Dependence of mammalian DNA replicationon DNA supercoiling. Biochim. Biophys. Acta, 563: 293-305, 1979.

30. Mirabelli, C. K., Huang, C-H., and Crooke, S. T. Comparison of DNA damageand single- and double-strand breakage activities on PM-2 DNA by talisomycinand bleomycin analogs. Cancer Res., 40: 4173-4177,1980.

31. Mirabelli, C. K., Mong, S., Huang, C. H., and Crooke, S. T. Comparison ofbleomycin A2 and talisomycin A specific fragmentation of linear duplex DNA.Biochem. Biophys. Res. Commun., 97: 871-877,1979.

32. Mong, S., Huang, C. H., Prestayko, A. W., and Crooke, S. T. Interaction ofc/s-diamminedichloroplatinum(ll) with PM2 DNA. Cancer Res., 40:3313-3317,1980.

33. Muller, W. E. G., Yamazaki, Z., Breter, H. J., and Zahn, R. K. Action of




bleomycin on DNA and RNA. Eur. J. Biochem., 31: 518-525,1972.34. Newman, S. A. The un body: a possible structure for transcriptionally inacti

vated chromatin subunits. J. Theor. Biol., 79: 55-66, 1979.

35. Olins, A. L, and Olins, D. E. Spheroid chromatin units (v bodies). Science(Wash. D. C.), 183: 330-332,1974.

36. Oostra, B. A., Goert, A. B., and Bruger, M. Involvement of DNA gyrase in thetranscription of ribosomal RNA. Nucleic Acids Res., 8: 4235-4246,1980.

37. Oppenheimer, N. J., Rodriguez, L. O., and Hecht, S. M. Metal binding tomodified bleomycins. Zinc and ferrous complexes with an acetylated bleomy-cin. Biochemistry, 19: 4096-4103, 1980.

38. Oudet, P., Gross-Bellard, M., and ChambÃ³n, P. Electron microscopic andbiochemical evidence that chromatin structure is a repeating unit. Cell, 4: 281-300.1975.

39. Paoletti, J. Relaxation of chromatin structure induced by ethidium binding.Involvement of the interacalation process. Eur. J. Biochem., 700; 531-539,1979.

40. Piper, P. W., Cells, J., Kaltoft, K., Leer, J. C., Nielsen, 0. F., and Westergaard,O. Tetrahymena ribosomal RNA gene chromatin is digested by micrococcalnuclease at sites which have the same regular spacing on the DNA ascorresponding sites in the bulk nuclear chromatin. Nucleic Acids Res., 3: 493-505.1976.

41. Povirk, L. F., Hogan, M., and Dattagupta, N. Binding of bleomycin to DNA:intercalating of the bithiazole rings. Biochemistry, 18: 96-101,1979.

42. Povirk, L. F., Hogan, M., Dattagupta, N., and Buechner, M. Coppelli)â€¢bleomycin, iron(lll)-bleomycin, and coppelli)-phleomycin: comparative study ofdeoxyribonucleic acid binding. Biochemistry, 20: 655-670,1981.

43. Prunell, A., Strauss, F., and Leblanc, B. Photographic quantitation of DNA ingel electrophoresis. Anala. Biochem., 78: 57-65,1977.

44. Renz, M., Niehls, P., and Hozier, J. Involvement of histone HI in the organization


of the chromosome fiber. Proc. Nati. Acad. Sei. U. S. A., 74: 1879-1883,1979.

45. Roques, B. P., Pelaprat, D., Gruen, l. L., Porcher, G., Gosse, C., and LePecq,J. B. DNA bifunctional intercalators: antileukemic activity of new pyridocarba-zole dimers. Biochem. Pharmacol., 28: 1811-1815,1979.

46. Sausville, E. A., Peisach, J., and Horowitz, S. B. Effect of chelating agentsand metal ions on the degradation of DNA by bleomycin. Biochemistry, 17:2740-2746, 1978.

47. Sugiura, Y., Muraoka, Y., Fugii, A., Takita, T., and Umezawa, H. Chemistry ofbleomycin. XXIV. Deamido bleomycin from view point of metal coordinationand oxygen activation. J. Antibiot. (Tokyo), 32: 756-758,1979.

48. Suzuki, H., Nagai, K., Yamaki, H., Tanaka, N., and Umezawa, H. On themechanism of action of bleomycin-scission of DNA strands in vitro and in vivo.J. Antibiot. (Tokyo), 22: 446-448, 1969.

49. Takeshita, M., Grollman, A. P., Ohtsubo, E., and Ohtsubo, H. Interaction ofbleomycin with DNA. Proc. Nati. Acad. Sei. U. S. A., 75: 5983-5987, 1978.

50. Takeshita, M., Horwitz, S. B., and Grollman, A. P. Bleomycin, an inhibitor ofvaccinia virus replication. Virology, 60: 455-465,1974.

51. WÃ¤hle,E., and Mueller, K. Involvement of DNA gyrase in rRNA synthesis Invivo. Mol. Gen. Genet., 779: 661-667,1980.

52. Waring, M. J., and Wakelin, L. P. B. Echinomycin: a bifunctional intercalatingantibiotic. Nature (Lond.), 252: 653-657, 1974.

53. Weintraub, H., and Groudine, M. Chromosomal subunits in active genes havean altered conformation. Science (Wash. D. C.), 793: 848-856, 1976.

54. Worcel, A. Molecular architecture of the chromatin fiber. Cold Spring HarborSymp. Quant. Biol., 42: 313-324, 1977.

55. Wu, H. M., Dattagupta, N., Hogan, M., and Crothers, D. M. Unfolding ofnucleosomes by ethidium binding. Biochemistry, 79: 626-643,1980.

Fig. 2. Gel electrophoretic pattern of DNA products after treatment of PM2 Form Io DNA (113.6 JIM)with a fixed amount of BLM A2 (30.7 nw). Isolated Forni Io DNA

was pretreated with increasing concentrations of EB. Gel mobility, top to bottom. Both gel and electrophoretic buffers contain EB (0.5 ng/ml) to convert the undegradedForm Io DNA into positively supercoiled form which has a higher mobility than that of Form II or Form III DNA. Slow-moving band, contains Form II DNA; intermediate

band, Form III DNA. Lane a, control; EB/DNA concentration ratios: 0 (Lane b), 0.022 (Lane c), 0.038 (Lane d), 0.050 (Lane e), 0.066 (Lane f), 0.080 (Lane g), 0.124 (Lanen), 0.164 (Lane /), 0.201 (Lane /).

Fig. 3. Gel electrophoretic pattern of DNA products of pBR322 plasmid Form I DNA preparations (25 ng) relaxed partially by treatments with increasing amounts ofcalf thymus topoisomerase (1 hr at 37Â°).Before being applied to gels, all DNA preparations went through EB/CsCI gradients to remove Forms II and III DNA produced

as a result of topoisomerase treatment. Lane a, no topoisomerase; Lane b, 1 iÂ¿topoisomerase; Lane c, 2.5 n\ topoisomerase; Lane d, 5 iÂ¿topoisomerase.

Fig. 4. Gel electrophoretic pattern of DNA products after treatment of PM2 Form I DNA with a fixed amount of PLM Di (20.7 nM) in the presence of increasing EBconcentrations. Gel mobility, top to bottom. Lane A, untreated Form I DNA; Lanes B to M, EB/DNA ratios of 0 (Lane B), 0.020 (Lane C), 0.045 (Lane D), 0.070 (Lane E),0.10 (Lane F), 0.12 (Lane G), 0.14 (Lane H), 0.16 (Lane 0, 0.17 (Lane J), 0.18 (Lane K), 0.22 (Lane L), and 0.26 (Lane M).

JUNE 1983 2855



C-H.Huangetal.

a bcde fg h ij

B

ABCDEFGH I J KLM




1983;43:2849-2856. Cancer Res Cheng-Hsiung Huang, Christopher K. Mirabelli, Seymour Mong, et al. Antibiotics, Bleomycin and PhleomycinBromide on the DNA-degrading Activity of Two Antitumor Effects of DNA Superhelical Changes Induced by Ethidium

Updated version

http://cancerres.aacrjournals.org/content/43/6/2849

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/43/6/2849To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



Documents

Effects of DMA Superhelical Changes Induced by Ethidium … · DMA-degrading Activity of Two Antitumor Antibiotics, Bleomycin and Phleomycin1 Cheng-Hsiung Huang,2-3 Christopher K