Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
[CANCER RESEARCH 44, 613-618, February 1984] Effects of Adriamycin on Supercoiled DMA and Calf Thymus Nucleosomes Studied with Fluorescent Probes1 Henry Simpkins,2 Leslie F. Pearlman, and Leslie M. Thompson Departments of Pathology and Biological Chemistry, California College of Medicine, University of California, Irvine, California 92717 ABSTRACT The interaction of the antitumor drug Adriamycin with nucleotides, polynucleotides, RNA, calf thymus nucleosomes, and DMA (including pBR322 supercoiled DMA) has been studied using fluorescent probes. The lanthanide terbium is known to interact with guanine and xanthosine to produce high fluorescence en hancement. The nature of the interaction of the lanthanide with the heterocyclic ring in guanine appears to involve both the C-2 and N-7 groups. A striking decrease in fluorescence enhance ment was observed with all of the polynucleotides, RNA, DNA, and nucleosomes after treatment with Adriamycin at molar ratios of 1:200 or less. It appears that Adriamycin interacts with the guanine ring, displacing or preventing terbium access to its second site of binding. However, with supercoiled DNA and nucleosomes, the displacement followed a destabilization of the helix at very low drug concentrations. The binding affinities of calf thymus DNA, pBR322 DNA, and calf thymus nucleosomes at 37° for Adriamycin were of the same order of magnitude. Reaction with A/-pyrene maleimide, a fluorescent probe which binds to histone H3, showed that Adriamycin interacted with the nucleosome to increase the binding of the probe (only, however, at drug ratios far greater than those required to produce effects with DNA). No compositional changes of supercoiled or nucleosomal DNA or nucleosomal histones were observed by agarose gel or sodium dodecyl sulfate:polyacrylamide gel electrophoresis, respectively. The classic intercalating agent, ethidium bromide, produced minimal displacement of the lanthanide from DNA, although an effect with RNA at high drug concentrations was observed. INTRODUCTION Adriamycin is an anthracycline antibiotic currently used as a potent chemotherapeutic agent in the treatment of certain leukemias and solid tumors (2). The primary target site of this drug has generally been accepted to be nuclear DNA. Adriamycin has been assumed to intercalate into native DNA (9) and, when added to cells in culture, it is known to inhibit both DNA and RNA synthesis (6, 20). The physicochemical effects of the drug on isolated DNA have generally been examined at very high drug:DNA ratios (22). Adriamycin has recently been shown to produce single-strand breaks when cells are treated in vivo, although it has been suggested that some of the breaks are really of a doublestranded nature. It has been postulated that a topoisomerase binds to the terminus of one of the strands (23, 26). Since some 1 Financial support was received from the Department of Pathology, California College of Medicine, University of California, Irvine, Irvine, CA. 2To whom requests for reprints should be addressed. Received March 29,1983; accepted November 8,1983. FEBRUARY 1984 derivatives of Adriamycin do not bind to DNA but are cytotoxic (16), the necessity for a priori binding has to be questioned. It has also been shown that Adriamycin is reduced by microsomal enzymes to semiquinone-free radicals which do not have to intercalate into DNA to produce strand breakage (1, 26). With the exception of 2 recent articles, very little has been reported of the interaction of this drug with nuclei or nuclear chromatin. Adriamycin has been shown to produce single-stranded DNA breaks in isolated nuclei, as determined by the sensitivity of the treated nuclei to Neurospora crassa nuclease (3). Electron mi croscopy as well as the analysis by alkaline sucrose gradients of DNA from the treated chromatin has shown morphological and compositional changes (33). These effects, however, were not observed if the nuclear proteins in the chromatin were hydrolyzed by proteases prior to drug treatment. This report prompted us to examine the effects of Adriamycin on DNA and protein structure within the nucleosome utilizing 2 highly sensi tive and specific fluorescent probes, terbium and A/-pyrene mal eimide. Terbium produces strong fluorescent enhancement when bound to guanine and xanthosine bases when they are in singlestranded structures (12, 29). Since it has been proposed that Adriamycin acts primarily on G-C sequences (7), terbium is a potentially powerful tool for examining the drug-DNA interaction within the nucleosome. A/-pyrene maleimidfc reacts with thymocyte and rat liver chromatin in situ, labeling the cysteine groups in histone H3 (probably cysteine 96), thus providing a means of monitoring the conformation of the histone octamer in the chro matin strand. We have used this probe to show conformational changes in chromatin induced by salt (21), liver regeneration (25). and the antitumor drug c/s-dichlorodiammineplatinum(ll) (28). In addition, the intrinsic fluorescence of Adriamycin has been used to monitor its binding to DNA and polynucleotides (7, 37). We have continued these observations using calf thymus mononucleosomes. MATERIALS AND METHODS Reagents. Adriamycin was obtained from Sigma Chemical Company and was used without further purification. Commercially prepared nucleic acids were obtained from Sigma Chemical Company, P-L Biochemicals, Miles Biochemicals, and Boehringer-Mannheim. They were suspended directly into Tris-HCI, pH 7.4, at a final concentration of 1.00 mw nucleotide phosphate, as determined by the respective extinction coef ficients. Identical results were obtained with DNA which had been further purified with RNase T2 digestion followed by chloroformiisoamyl alcohol extraction. Supercoiled DNA from plasmid pBR322 was prepared by amplification in the presence of chloramphenicol according to the method of Clewell (5), and the DNA was purified by the clear lysate technique of Guerryefa/. (13). Preparation of Nucleosomes. Calf thymus nucleosomes were pre pared using a modification of the procedure described by Levy-Wilson ef al. (17). Calf thymus (5 g) was homogenized in Buffer A [0.25 M 613 Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1984 American Association for Cancer Research. H. Simpkins et al. sucrose:10mw MgCI2:1 mw CaCI2; 1 mM PMSF3:0.1% (w/v) Triton X- 100:50 mM Tris-HCI, pH 7.4] then washed 3 times ¡nBuffer A without Triton X-100. Nuclei were suspended in Buffer B (10 mM NaCI:3 mw MgCI2:1 mM PMSF:10 mM Tris-HCI, pH 7.5) and then washed twice in the same buffer containing 1 mM CaCI2 and digested for 30 min with micrococcal nuclease (Worthington Biochemicals, 25 units/ml) at a DMA concentration of 10 mg/ml. The reaction was terminated by chilling the tubes and centrifuging for 5 min at 4050 x g in a Sorvall HB-4 rotor. The pellet was resuspended in Buffer C (1 mM EDTA:1 mM PMSF:10 mM Tris-HCI, pH 7.5), swelled on ice for 20 min, and centrifuged for 20 min at 16,300 x g. The last step was repeated, the 2 supematants were combined, and the nucleosomes were recovered by precipitation with 3 volumes of ethanol in 0.1 M NaCI. After centrifugation, the nucleosomes were suspended in storage buffer (0.2 mM EDTA:1 mM PMSF:10 mM Tris-HCI, pH 8.0) and kept frozen at -20° until needed for use. These nucleosomes possessed a full complement of histones, including H1, when monitored by SDS gel electrophoresis. Terbium Fluorescence Measurements. Samples containing the ap propriate concentrations of nucleic acid and terbium were made up to 1.2 ml in 20 mM Tris-HCI, pH 7.4. The samples were equilibrated for 30 min at room temperature, by which time the reaction was determined to be complete. Experiments involving RNA were equilibrated at 4°for 30 min and then brought to room temperature, and fluorescence was measured immediately, since lanthanides have been shown to hydrolyze the nucleotide bonds in RNA, although slowly, at room temperature (8). The fluorescence intensity was measured using 1.0-cm quartz cells, in a Perkin Elmer MPF 43A Fluorescence Spectrophotometer equipped with a high-pressure xenon lamp. An excitation wavelength of 290 nm and an emission wavelength of 544 nm were optimal for maximizing energy transfer and minimizing background light scattering. Entrance and exit slit widths were maintained at 10 nm for all fluorescence measurements. The concentration of terbium chloride-6H2O (Aldrich Chemicals) was made up to 100 mM in distilled water and precisely determined by titration against EDTA (12). Titrations with the terbium ion were performed on samples containing identical quantities of the nucleic acid by the addition of concentrated terbium chloride (1 or 10 mM). Titration of the Nucleosomes with At-Pyrene Maleimide. N-pyrene maleimide was synthesized from 3-aminopyrene as described by Weltman ef al. (34). Prior to use, it was dissolved as a 0.1 mM solution in ethanol. The nucleosomes [20 ^g of protein as determined by the method of Lowry ef al. (18) using total calf thymus histone as standard] were incubated with increasing concentrations of Adriamycin for 20 min at 37° in the nucleosomal storage buffer. The treated nucleosomes were then precipitated with 3 volumes of ethanol in 0.1 M NaCI, pelleted by centrifugation for 15 min at 3000 x g, resuspended in storage buffer, and reacted with 0 to 200 nM A/-pyrene maleimide for 25 min at room temperature. The fluorescence intensity was measured using an excita tion wavelength of 342 nm and an emission wavelength of 377 nm. Similar results were obtained if the treated nucleosomes were recovered by centrifugation at 400,000 x g for 4 hr in lieu of ethanol precipitation and then labeled with N-pyrene maleimide. Treated and control nucleosomes (0.5 mg) were incubated with trypsin (0.1 mg) for 60 min at 37°.DNA fragments were subsequently extracted using a 10-fold excess of chlorofornvisoamyl alcohol (24:1, v/v). After centrifugation for 10 min at 3000 x g, the aqueous layer was carefully removed, and the DNA precipitated with 3 volumes of ethanol in 0.1 M NaCI. The DNA (5 Mg) was analyzed on agarose gels according to the procedure of McMaster ef al. (19). The gels were stained with ethidium bromide (1 Mg/ml) tor 30 min, destained overnight at 4°with distilled water, and then photographed using Polaroid type 55 film, utilizing a 63B Transilluminator (Ultra-Violet Products, San Gabriel, CA). Thermal Melting Profiles. Adriamycin-treated and control DNA and nucleosomes were made up to 1.0 mM DNA phosphate in 10 mM sodium phosphate buffer, pH 7.4, with 0.2 mM EDTA and 1 mM PMSF, and diluted in the same buffer to give an absorbance at 260 nm of 0.6 to 0.8. The samples were subjected to an increase in temperature at a rate of 1°/minusing a Haake FJ thermal circulating pump. The absorbance was read at 260 nm on a single-beam Gilford 250 Spectrophotometer using buffer as the blank. RESULTS When DNA was treated with Adriamycin at a 1:20 or 1:50 (w/ w) ratio for 20 min at 37°and precipitated with ethanol in 0.1 M NaCI and when terbium binding curves were performed, a marked dimnuition in the terbium fluorescence was observed, almost to within 25% of that observed with the nontreated samples. The reaction appears complete by this time, because similar drug concentrations at 10, 30, and 60 min of incubation produced the same decreases in fluorescence. We documented this decrease with calf thymus DNA and a mixture of 16S and 23S fragments of Escherichia coli rRNA by treatment with in creasing concentrations of Adriamycin at 100 pM terbium (Chart 1). Both nucleic acids, with markedly different secondary struc tures (a double-stranded molecule versus one containing singlestranded regions), show a substantial decrease in terbium fluo rescence, the intensity reaching a plateau at approximately 1 to 1.5 pM Adriamycin. The decrease in fluorescence intensity ap pears at lower drug concentrations with the less structured molecule. When the guanine-containing homoplymers poly(rG) and poly(dG) were titrated with Adriamycin, the polyribonucleotide showed a greater decrease than did the deoxyribose derivative lOO.t Compositional Studies. The proteins of the treated and control nucleosomes were analyzed by SDS polyacrylamide gel electrophoresis using a modification of the procedure of Laemmli (15). The stacking gel contained 4.2% (w/v) acrylamide with a bisacrylamide:acrylamide ratio of 1:37.5 (w/w) in 0.1% SDS:0.125 M Tris-HCI, pH 6.8; the separating gel contained 12% (w/v) acrylamide in 0.1% SDS:0.375 M Tris-HCI, pH 8.8; and a reservoir buffer contained 0.1% SDS:0.19 M glycine:0.125 M Tris-HCI, pH 6.8. The current was kept constant at 7.5 mA until the samples had run through the stacking gel and then at 12.5 mA until the tracking dye was within 0.5 cm of the bottom. The gels were stained with 0.1% (w/v) Coomassie Blue in methanokacetic acid:water (10:7:83, v/v) and destained with the same methanol:acetic acid:water mixture. 3 The abbreviations used are: PMSF, phenylmethylsulfonylfluoride; SDS, sodium dodecyl sulfate; poly(rG). polynboguanylate; poly(dG), polydeoxyguanylate. 614 02 03 04 0.5 06 07 08 Adriamycin 09 10 Chart 1. Terbium fluorescence (F) expressed as a percentage of the control (Fc). Increasing concentrations (0 to 4 MM)of Adriamycin were incubated with 50 MMcalf thymus DNA (•)or 50 MM 16S and 23S RNA (A) at 37°for 20 min in 20 mM Tris-HCI, pH 7.4. The terbium was then added at a constant concentration of 100 MM. CANCER RESEARCH Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1984 American Association for Cancer Research. VOL. 44 Effects ofAdriamycin on DNA and Nucleosomes (Chart 2). The decrease in fluorescence intensity of poly(rG) occurred at 0.5 pM Adriamycin as compared to 1 U.M with poly(dG). These polymers are essentially nonstructured at this concentration and ionic strength, although poly(dG) forms a 4stranded helix at higher concentrations (35). Since it is known that terbium fluoresces more intensely when reacted with guanine residues in a single-stranded random coil than in a doublestranded helical form, it was important to determine whether this result was due to single-stranded breaks or to nicks in the linear DNA fragments. To this end, the titration of Adriamycin with supercoiled plasmid pBR322 DNA was performed (Chart 3). A similar decrease in overall fluorescence was found. Fluorescence intensity plateaued at 1 to 1.5 ^M Adriamycin, as with calf thymus DNA and poly(dG). A marked increase in fluorescence was observed at low Adriamycin concentrations (0.05 to 0.2 ¿IM). This effect was not apparent with linear DNA fragments from calf thymus, nor did it occur when the drug was added to a terbium:nucleic acid mixture at 25°and the titration was performed directly, where the shape of the curve is more hyperbolic. A curve similar to that generated with the pBR322 DNA was obtained when calf thymus nucleosomes were titrated with the drug. In order to determine whether the observed decreases in terbium fluorescence were caused by an intrinsic effect on the guanine residues themselves, the nucleotides GMP, XMP, and several derivatives of guanine were titrated (at far higher con centrations, since their intrinsic fluorescence with the lanthanide is so much lower). A marked decrease in terbium fluorescence was observed (Table 1). The possibility that the drug, which exhibits minimal fluores cence at 544 nm when excited at 290 nm, quenches terbium fluorescence was considered. However, the fluorescence of Adriamycin is only 5 to 10% of that observed with terbium-DNA complexes at a 1 to 2 MMdrug concentration (highest concentra tion used) and less than 1% for polynucleotide:terbium com plexes (due to the higher intrinsic fluorescence of these com plexes with terbium). In addition, when the DNA or polynucleotide was treated with Adriamycin, precipitated, and resuspended (thereby removing any unbound Adriamycin) and a terbium titra tion curve was performed, values similar to our previous data were obtained. 0 01 02 03 04 05 0.6 07 08 09 10 I.I pM Adriamycin Chart 3. Terbium fluorescence(F)expressed as a percentageof the control (Fc) of a 50 fiu solution of supercoiledp8R322 DNA to which increasingconcentrations of Adriamycin had been added at 37°for 20 min in 20 mw Tris-HCI, pH 7.4, prior to the terbium (O). G, fluorescence of a solution of 50 /JMpBR322 DNA reacted with 100 liM terbium at 25°,to which Adriamycin was added at the concentrations indicated. A, effect of treatment for 20 min at 37°with increasing concentrations of Adriamycin on the terbium fluorescence of calf thymus nucleosomes in 10 mm triethanolamineHCI, pH 7.4. Table 1 Effect ofAdriamycin on terbium fluorescenceof guanine nucleotides The nudeotide and drug were incubated at the concentrations described below for 20 min at 37°in 20 mw Tris-HCI, pH 7.4, prior to the addition of the terbium. Data are expressed as a percentage of the fluorescenceof the control, which was incubated with buffer substituted for the Adriamycin. inten sity at 100 pu ter bium10062 NucleotidePoWrG)5--GMP5'-XMP7-Methyl-5'-GMPNucleotide50 (Õ.M)0 (2)"27 + 5a 0.100.4800.311.5300.271.3600.180.90Fluorescence 5050200200200200200200200200200Adriamycin (2)10063 +3 (2)17 +7 (2)10053 +2 (2)13 +5 (2)10080 +2 (2)45 +7 + 5 (2) * Mean ±S.E. " Numbers in parentheses, number of determinations. 02 0.3 04 0.5 0.6 07 0.8 09 1.0 /U.MAdriamycin Chart 2. Terbium fluorescence (F) expressed as a percentage of the control (Fc). Solutions (25 pM)of the polynucleotidespoly(dG)(•) or poly(rG)(A) to which increasing concentrations of Adriamycin had been added at 37°for 20 min in 20 mMTris-HCI, pH 7.4, are shown. Terbium was then added at a constant concen tration of 100 MM FEBRUARY 1984 When the effect of low concentrations of the drug was ana lyzed by thermal melting of the Adriamycin:DNA complexes (1:50 or greater molar ratio), it was found that the effect was minimal, a stabilization of the order of 1-2° in the melting temperature (Tm).This was true whether isolated linear DNA fragments from E. coli, calf thymus, or calf thymus nucleosomes were used. The 15-16° stabilization in Tm (22) at high drug:DNA (1:10) ratios was not observed at these relatively low concentrations short incubation times. and 615 Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1984 American Association for Cancer Research. H. Simpkins et al. The effect of a classic intercalating agent, ethidium bromide, on terbium displacement was investigated. It was found that this drug caused no appreciable decrease in the terbium fluorescence of calf thymus DNA but did diminish, although at far higher drug concentrations, the fluorescence of rRNA from E. coli (Chart 4). The effect of Adriamycin on nucleic acid:protein complexes was investigated by reacting calf thymus nucleosomes with the drug. The treated nucleosomes were pelleted, the terbium bind ing curves were performed in 10 mM triethanolamine-HCI, pH 7.4 (Chart 5). It is observed that, in contrast to the curves obtained with polynucleotides and isolated DNA molecules, saturation by terbium was not observed even at 700 tiM and that an almost linear increase in fluorescence was observed with increasing terbium concentrations. Terbium fluorescence, however, was still markedly decreased by prior treatment of the nucleosomes 130 120 110 100 with findings of other authors (7, 37), decreases in fluorescence were observed which could be analyzed by a Scatchard analysis (See Table 2). Nucleosomes and calf thymus DNA showed similar affinities and number of binding sites for the drug. Supercoiled DNA showed a somewhat lower number of binding sites, al though the affinity constants were of the same order of magni tude. The values obtained with calf thymus DNA are of the same order of magnitude, as reported by other authors at lower temperatures and higher ionic strengths (7, 37). Minimal effects of temperature were observed (data not presented). To analyze the effect on the histone protein conformation within the nucleosome, the fluorescent probe A/-pyrene maleimide was used. This probe has been shown by us (21, 25) to label histone H3 in situ within the calf thymus or rat liver nucleosome, probably at the cysteine 96 residue. When the binding of A/pyrene maleimide was analyzed subsequent to Adriamycin treat ment of calf thymus nucleosomes, it was found that an increase in W-pyrene maleimide fluorescence was observed only at very high drug concentrations (Table 3). When the proteins of the treated nucleosomes were analyzed by polyacrylamide gel electrophoresis, no changes in protein composition were observed, even at high Adriamycin concentrations (data not presented). In addition, when calf thymus DNA, pBR322 DNA, or calf thymus 90 80 B 70 £60 -50 40 30 20 10 h 0 with increasing Adriamycin concentrations, even at a Adriamycin:DNA molar ratio as low as 1:200. Titration of Adhamycin with calf thymus DNA, pBR322 DNA, and calf thymus nucleosomes was performed at 37°.In accord O.I 0.2 0.3 0.4 05 06 07 08 0.9 1.0 fiM Ethidium Bromide Chart 4. Terbium fluorescence (F) expressed as a percentage of the controls (Fc). Calf thymus DNA (50 JIM) (A) and 16S and 23S rRNA (O) treated with increasing concentrations of ethidium bromide in 20 mu Tris-HCI, pH 7.4, at 37° for 20 min, followed by the addition of 100 /IM terbium, are shown. Table 2 Association constants and the number of binding sites for Adriamycin binding to RNA, DNA. and calf thymus nucleosomes The drug was kept at a constant concentration of 15 MM, and the DNA-P concentration was increased until fluorescence intensity leveled. The buffer used was 10 mM sodium phosphate, pH 7.4, unless otherwise noted. Experiments were performed at 37°.An excitation wavelength of 480 nm and an emission wavelength of 560 nm were used. The curves were analyzed as described by Scatchard (24). acidCalf no. of binding sites0.28 Nucleic thymus DNA (10 mM Tris-HCI:0.2 mM EDTA:1 [20 2(2)]20 ± mM PMSF, pH 7.4) ±3(2) pBR322 DNA (10 mM Tris-HCI:0.2 mM EDTA:1 [19 ±3(2)]10 mM PMSF, pH 7.4) ±2(2)Apparent Calf thymus nucleosomes (10 mM Tris-HCI:0.2 mM EDTA:1 mM PMSF, pH 8.0)Ka(x10>)M-1'i10±2"(2f " Association constant. 6 Mean ±S.E. c Numbers in parentheses, number of determinations. ±0.05 [0.25 0.05]0.21 ± ±0.04 [0.19 0.03]0.30 + ±0.05 Table 3 Effecf of Adriamycin on N-pyrene maleimide fluorescence of calf thymus nucleosomes Calf thymus nucleosomes were incubated with Adriamycin at the indicated ratios for 20 min at 37°in 10 mw Tris-HCI, pH 8.0, with 0.2 mM EDTA and 1 mM PMSF, precipitated with ethanol, and resuspended in the same buffer prior to the addition of N-pyrene maleimide. The data are expressed as a percentage of the control fluorescence, which was incubated with buffer substituted for the Adriamycin. 800 Chart 5. Calf thymus nucleosomes were reacted with increasing molar ratios of Adriamycin for 20 min at 37°. O, no drug; A, 1:200 (drug:DNA-P); D, 1:20 (drug:DNA-P). The nucleosomes were pelleted following drug treatment and resuspended in 10 mM triethanolamine-HCI, pH 7.4. Binding curves were performed by increasing the concentration of terbium at 50 ¿IM DNA-P. 616 Concentra tions of Npyrene mal eimidenM20 ratios1:10150 intensities at various drug:nucleosomal (w/w) protein ±15a + 15 ±1585 ±15 ±15 130 ±15 200 ±20 185 ±20 50 110± 15 95 ±15 240 ±20 170 ±20 100200Fluorescence 135 ±15 100 ±151:500100 150 ±151:20150 190 ±201:200120 a Mean ±S.E. CANCER RESEARCH VOL. 44 Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1984 American Association for Cancer Research. Effects of Adriamycin on DNA and Nucleosomes nucleosomes were treated with varying concentrations of Adri amycin and the DNA was analyzed by agarose gel electrophoresis, no reproducible difference in electrophoretic mobility was observed in any of the samples. This suggests that modification of the DNA or strand breakage of the supercoiled DNA had not occurred. DISCUSSION The mode of action of Adriamycin has become a matter of discussion, in view of the recent reports showing that it produces single- and double-strand breaks in DNA in vivo and that a priori binding to the DNA may not be necessary for this reaction to occur (16, 23, 36). In addition, recent reports that the drug is reduced by microsomal enzymes to free radicals opens up new avenues of research (1, 26), as do the many reports showing effects at the membrane level (10, 27, 30, 31). A recent article (32) which shows that the drug, when coupled to agarose (which renders it impermeable to cells), results in a product still markedly cytotoxic to L1210 leukemic cells, raises more important ques tions as to its mode of action. It is obvious that a simple intercalation model with DNA cannot totally explain its mecha nism of action. This idea is strengthened by our results, which show that displacement of terbium from DNA produced by this drug is not duplicated by ethidium bromide. However, it should be pointed out that ethidium produces a much greater unwinding of DNA than does Adriamycin (26°as opposed to 10-13°); thus, the mode of intercalation may be different. We have concluded that Adriamycin reacts with guanine resi dues, somehow distorting the ring such that the intrinsic fluores cence of the ion which requires 2-site binding (12) does not occur with either nucleotide, polynucleotide, RNA, singlestranded DNA, supercoiled DNA, or the intact nucleosome. It is noteworthy that the effects are observed at relatively low Adria mycin concentrations, suggesting a very high specificity and affinity of the drug for guanine residues. This may be biologically relevant. This interaction is very interesting when compared to results obtained with displacement studies using cations (14). It was found that Cu(ll), which is known to interact with both the phosphate group and the base ring (11), was the most effective counterion to terbium, having twice the affinity exhibited by terbium itself. Adriamycin produces its effect at 0.5 to 1.0 ^M with 50 ftM nucleotide, showing approximately a 10- to 20-fold increase in affinity over the cuprous ion. This may be due to both its interaction with the base ring and its specificity for guanine bases, which is not apparent with the cuprous ion. The Km for terbium for supercoiled DNA is approximately 105 M"1 (12), and we and others (7, 37) report values of 107 w~1 for the affinity of Adriamycin for DNA and calf thymus nucleosomes. However, it is possible that the reaction of the drug with A-T regions proximal to guanine residues may result in the observed change. This cannot be discounted, in view of the result of Chandra ef al. (4) showing drug-induced inhibition of DNA polymerase activity with A-T polynucleotide templates. The initial destabilization of the helix at low drug concentrations observed with supercoiled pBR 322 DNA and nucleosomes prior to the displacement of terbium from its second binding site (12) may also be biologically relevant. This results in a marked in crease in fluorescence intensity. This effect on terbium fluores cence has also been observed when supercoiled DNA is titrated FEBRUARY 1984 with low salt concentrations (11). Finally, the drug appears to have little effect on nucleosomal histones as monitored by A/-pyrene maleimide binding studies. This probe was used successfully (28) to show marked effects of the potent antitumor drug c/'s-dichlorodiammineplatinum(ll) on nucleosomal proteins at very low drug concentrations and short incubation times, conditions under which no changes in DNA conformation could be detected. Conventional polyacrylamide gel electrophoresis also showed no detectable effect on nuclear proteins following reaction with Adriamycin, although marked changes in terbium fluorescence of the nucleosomes treated with very low concentrations were observed. Thus, the drug appears to react in vitro at very low concentra tions with guanine residues, perturbing the interaction of terbium with the guanine ring so that the fluorescence enhancement of the lanthanide produced by this base is abrogated. This effect is also observed with supercoiled DNA as well as with protein:DNA complexes, i.e., nucleosomes. However, the classic intercalating agent, ethidium bromide, did not produce any effect with DNA. ACKNOWLEDGMENTS The authors wish to thank Luis de la Maza, M.D., Ph.D., and Marie Pollack for the generous gift of pBR 322 supercoiled DNA and Rachel Hwei-Ping Yu for her technical assistance. REFERENCES 1. Berlin, V., and Haseltine, W. A. Reduction of Adriamycin to a semiquinone-free radical by NADPH-cytochrome P450 reducÃ-ase produces DNA cleavage in a reaction mediated by molecular oxygen. J. Biol. Chem., 256: 4747-4756, 1981. 2. Blum, R. H., and Carter, S. K. Adriamycin. A new anticancer drug with significant clinical activity. Ann. Intern. Med., 80: 249-259,1979. 3. Center, M. S. Induction of single stranded regions in nuclear DNA by Adria mycin. Biochem. Biophys. Res. Commun., 89: 1231-1238,1979. 4. Chandra, P., Zunino, F., Gotz, A., Gericke, D., and Thorbeck, R. Specific inhibition of DNA-polymerases from RNA tumor viruses by some new daunomycin derivatives. FEBS Lett., 27: 264-268, 1972. 5. Clewell, D. B. Nature of col E plasmid replication in Escherichia coli in the presence of chloramphenicol. J. Bacteriol., 770: 667-676,1972. 6. Crooke, S. T., Duvemay, V. H., Galvan, L., and Prestayko, A. W. Structureactivity relationship of anthracyclines relative to effects on macromolecular syntheses. Mol. Pharmacol., 74:290-298,1978. 7. Duvemay, V. H., Pachter, J. A., and Crooke, T. S. Deoxyribonucleic acid binding studies on several new anthracyclic antitumor antibiotics. Sequence preference and structure activity relationship of marcellomycin and its ana logues as compared to Adriamycin. Biochemistry, 18: 4024-4030,1979. 8. Eichhorn, G. L., and Butzow, J. J. Interaction of metal ions with polynucleotides and related compounds. III. Degradation of polyribonucleotides by lanthanum ions. Biopolymers, 3: 79-94, 1965. 9. Gabbay, E. J., Grier, D., Fingerle, R. E., Reimer, R., Levy, R., Pearce, S. W., and Wilson, W. D. Interaction specificity of the anthracyclines with deoxyribonucleic acid. Biochemistry, 75: 2062-2070,1976. 10. Goormaghtigh, E., Châtelain, P., Caspers, J., Ruysschaert, J. M. Adriamycin inhibits the formation of non-bilayer lipid structures in cardiolipin-containing model membranes. Biochim. Biophys. Acta, 597:1-14,1980. 11. Gross. D. S., Rice, S. W., and Simpkins, H. Influence of inorganic cations and histone proteins on the terbium(lll)-nucleic acid interaction. Biochim. Biophys. Acta, 656: 167-176,1981. 12. Gross, D. S., and Simpkins, H. Evidence for two site binding in the terbium(lll)nucleic acid interaction. J. Bid. Chem., 256: 9593-9598, 1981. 13. Guerry, P., LeBlanc, D. J., and Falkow, S. General method for the isolation of plasmid deoxyribonucleic add. J. Bacteriol., 776:1064-1066,1973. 14. Izatt, R. M., Christensen, J. J., and Rytting, J. H. Sites and thermodynamic quantities associated with protons and metal ion interactions with ribonucleic acid and deoxyribonucleic acid and their constituent bases, nudeosides and nucleotides. Chem. Rev., 77: 439-481, 1971. 15. Laemmli, V. K. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (Lond.), 227. 680-685,1971. 16. Levin, M., Silber, R., Israel, M., Goldfeder, A., Khetarpal, V. K., and Potmesil, M. Protein-associated DNA breaks and DNA-protein cross-links caused by DNA nonbinding derivatives of Adriamycin in L1210 cells. Cancer Res., 47: 1006-1010,1981. 17. Levy-Wilson, B., Connor, W., and Dixon, G. H. A subset of trout testis nucleosomes enriched in transcribed DNA sequences containing high mobility 617 Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1984 American Association for Cancer Research. H. Simpkins et al. group proteins as major structural components. J. Biol. Chem., 254:609-620, 1979. 18. Lowry, 0. H., Rosebrough, N. J., Fair, A. L, and Randall, R. J. Protein measurement with the Polin phenol reagent. J. Biol. Chem., 193: 265-275, 1951. 19. McMaster, G. K., and Carmichael, G. C. Analysis of single and double stranded nucleic acids on polyacrylamide and agarose gels by using glyoxal and acridine orange. Proc. Nati. Acad. Sei. U. S. A„74: 4835-4838,1977. 20. Meriwether, W. D., and Bachur, N. R. Inhibition of DNA and RNA metabolism by duanorubicin and Adriamycin in U 210 mouse leukemia. Cancer Res., 32: 1137-1142.1972. 21. Mooney, D., Thompson, L. M., and Simpkins, H. The in situ labeling of historie H-3 in chromatin by a fluorescent probe. Biochim. Biophys. Acta, 625: 51-63, 1980. 22. Neidle, S. Daunomycin and related compounds. Prog. Med. Chem., ?6: 196221,1979. 23. Ross, W. E., and Bradley, M. 0. DNA double-stranded breaks in mammalian cells after exposure to interacalating agents. Biochim. Biophys. Acta, 654: 129-134, 1981. 24. Scatchard, A. The attraction of proteins for small molecules and ions. Ann. N. Y. Acad. Sci., 5T: 660-672, 1949. 25. Simpkins, H., Thompson, L. M., Waldeck, N., Gross, D. S., and Mooney, D. Conformational changes in rat liver chromatin after liver regeneration. Biochem. J., 793:671-678, 1981. 26. Sinha, B. K., and Chignell, C. F. Binding mode of chemically activated semiquinone-free radicals from quinone antica.icer agents to DNA. Chem.-Biol. Interact., 28:301-308, 1979. 27. Solle, T. N., and Yuncker, C. Adriamycin induced changes in translocation of sodium ions in transporting epithelial cells. Life Sci., 22:1907-1920,1978. 28. Thompson, L. M., Arquilla, M., and Simpkins, H. The interaction of platinum 618 29. 30. 31. 32. 33. 34. 35. 36. 37. complexes with nucleosomes investigated with fluorescent probes. Biochim. Biophys. Acta, 698: 173-182,1982. Topal, M. D., and Fresco, J. R. Fluorescence of terbium ion-nucleic acid complexes: a sensitive specific probe for unpaired residues in nucleic acids. Biochemistry, Õ9:5537-5542,1980. Tritton, T. R., Murphree, S. A., and Sartorelli, A. C. Characterization of drugmembrane interactions using the liposome system. Biochem. Pharmacol., 26: 2319-2323,1977. Tritton, T. R., Murphree, S. A., and Sartorelli, A. C. Adriamycin: a proposal on the specificity of drug action. Biochem. Biophys. Res. Commun., 84:802-808, 1978. Tritton, T. R., and Yee, A. The anticancer agent Adriamycin can be actively cytotoxic without entering cells. Science (Wash. D. C.) 277: 248-250, 1982. Waldes, H., and Center, M. S. The interaction of Adriamycin with nuclear DNA: evidence for a drug induced compaction of nuclear chromatin. Biochem. Biophys. Res. Commun., 98: 95-101,1981. Weltman, J. K., Szaro, R. P., Frackelton, A. R., Jr., Dowben, R. M., Bunting, J. R., and Cathou, R. E. N-<3-Pyrene)maleimide: a long lifetime fluorescent sulfhydryl reagent. J. Biol. Chem., 240: 3173-3177,1973. Zimmerman, S. B., Cohen, G. H., and Davies, D. R. X-ray fiber diffraction and model-building study of polyguanylic acid and polysinosinic acid. J. Mol. Biol., 92: 181-192,1975. Zwelling, L. A. Michaels, S., Erickson, L. C., Unserieider, R. S., Nichols, M., and Kohn, K. W. Protein-associated deoxyribonucleic acid strand breaks in U 210 cells treated with deoxyribonucleic acid intercalating agents 4-{9acridinylamino)methane sulfon-m-c/s-anisidide and Adriamycin. Biochemistry, 20:6553-6563,1981. Zunino, F., DiMarco. A., Zaceara, A., and Gambetta, R. A. The interaction of daunorubicin and doxorubicin with DNA and chromatin. Biochim. Biophys. Acta, 607. 206-214,1980. CANCER RESEARCH Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1984 American Association for Cancer Research. VOL. 44 Effects of Adriamycin on Supercoiled DNA and Calf Thymus Nucleosomes Studied with Fluorescent Probes Henry Simpkins, Leslie F. Pearlman and Leslie M. Thompson Cancer Res 1984;44:613-618. Updated version E-mail alerts Reprints and Subscriptions Permissions Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/44/2/613 Sign up to receive free email-alerts related to this article or journal. To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at [email protected]. To request permission to re-use all or part of this article, contact the AACR Publications Department at [email protected]. Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1984 American Association for Cancer Research.