The Cyclic Voltammetric Study of Iron-Tallysomycin in the Absence and Presence of DNA

Jie Sun and Daniel K. Y. Solaiman

Department o f Chemistry, Duquesne University, Pittsburgh, Pennsylvania

ABSI 'RACT

Cyclic voltammetric studies on iron-tallysomycin complexes have been conducted with and without the presence of calf thymus DNA. Fe(ID-TLM samples exhibit a cyclic voltammogram with only a reduction peak at - 230 + 5 mV vs Ag/AgCI. The addition of DNA substrate causes the shift of this reduction peak to - 140 ± 10 mV vs Ag/AgCI. This large shift in the positive direction implies that the regeneration of Fe(II)-TLM through the reduction of Fe(HI)-TLM is facilitated with the aid of DNA. It also implies that the metal-binding/oxygen-activation domain may be directly involved in the formation of iron-tallysomycin-DNA ternary complex. Air oxidation of Fe(H)-TLM produces an activated intermediate with the following CV characteristics, lpc/Ipa = 0.90; AE ffi 65 mY; E.xh~eo~ pe~ = - 100 mV vs Ag/AgCI. Addition of DNA abolishes the redox peaks of this voltammogram, signifying inactivation of the activA_ted_ species through reaction with substrate. Air oxidation of preformed Fe(ID-TLM-DNA complex did not give a discernable cyclic voltammogram, nor did preformed Fe(III)-TLM and Fe(m)-TLM-DNA samples.

ABBREVIATIONS

DNA, deoxyribonucleic acid; Hepes, N-(2-hydroxyethyl)piperazine-N'-2-ethane- sulfonic acid; BLM, bleomycin; TLM, tallysomycin; PMS, phenazine methosulfate; EDTA, ethylenedimninetetmacetic acid; SHE, standard hydrogen electrode; SCE, standard calomel electrode; CV, cyclic voltammetry; Fe-TLM or Fe-BLM, iron complexes of tallysomycin or bleomycin, respectively, with unspecified oxidation state.

I N T R O D U C T I O N

The bleomycins [1] are a family of antineoplaslic agents that are clinically used to treat human neoplasms such as squamous ceil carcinoma, sarcoma, and malignant

Address reprint reque~ and correspondence to: Dr. Daniel K. Y. Solaiman, U.S. Department of Agriculture, Agricultural Research Service, ERRC, 600 E. Mermaid Lane, Philadelphia, Pennsylvania 19118.

Journal of Inorganic Biochemistry, 40, 271-277 (1990) 2"/1 © 1990 Elsevier Science Publishing Co., Inc., 655 Avenue of the Americas, NY, NY 10010 0162-0134/90/$3.50

212 J. Sun and D. K. Y. Soiaiman

lymphoma [2-41. A more potent derivative, tallysomycin, was later developed [5]. Their antitumor cytotoxic effect is believed to be due to their DNA strand scission activity [6-81 in the presexw of metal ions and electron acceptors such as oxygen [9- 161. The strong evidence that metal-centered redox processes are involved in the mechanism of action of bleomycins [9 - 11, 17, 181 prompted extensive research into the electrochemical properties of metallobleomycins [19-221. For iron-bleomycins, Dabrowiak and Santillo [20] mported that the iron-centered oxidation could not be observed on a hanging mercury drop electrode. In addition, no iron-centered reduction could be observed either in the presence of phosphate buffer or in unbuffered solutions. Mehtyk et al. [19] claimed that a mediator, PMS, helped the electron transfer on the platinum electrode. The ferric bleomycin prepared by air oxidation of ferrous bleomycin demo&rated a reduction potential of 129 f 12 mV vs SHE in the presence of PMS in 50 mM Hepes buffer (pH 7.0). The oxidation potential had a range of 80-120 mV vs SHE. Using cyclic voltammetric technique, Fultz [21] demonstrated that the redox peaks of ferrous bleomycin and ferric bleomycin (prepared via air oxidation of ferrous bleomycin) could be observed in the absence of mediator if a glassy carbon electrode was used. The measured Fe(III)- BLM/Fe(II)-BLM reduction potential ranged from - 90 mV to - 246 mV vs SCE in 50 mM Hepes buffer @H 6.5-8.1). In another cyclic volmmmetric experiment, Miysohi et al. [22] recorded the mdox peaks for ferric bleomycins obtained via air oxidation of ferrous bleomycin. This “active” ferric bleomycin has a reduction peak around -255 mV vs SCE on the hanging mercury electrode in borate but%%, pH 8.5.

It has been well documented that metallobleomycins combine with DNA to form a ternary complex [23-271. A recent paper [28] further suggests that, in addition to the intercalation of bithiaxole moiety into DNA, the metal-binding/oxygen-activation domain of metallobleomycins may also participate qualitatively in the interaction with DNA. The metal center of the ternary complex may thus assume different redox characteristics. In this communication, we report the results of our studies on the influence of DNA on the redox property of iron-tallysomycin, as well as the different redox properties of a variety of iron-tallysomycin and iron-tallysomycin-DNA complexes. It should be noted that unlike bleomycin, tallysomycin had a secondary metal binding site involving the mrminal amine moiety attached to the bithiaxole ring system [29]. However, the stability of this alternative metallo-species was so low in comparison to the primary “bleomycin-like” metal complex, that its significance in the inorganic biochemistry of tallysomycin was doubtful 1303.

EXPERIMENTAL

Materials

Tallysomycin S,,b, obtained as a kind gift from Dr. William Bradner of Bristol Myers Company (Syracuse, New York), was used throughout this study without further treatment. Hepes (99%) was purchased from Aldrich Chemical Co. The stock 50 mM Hepes buffer @H 7.0) was prepared with double deionized water; adventitious metals in the buffer were removed by chromatography through a Chelex-100 column. FeS04 (99 + 96) was purchased from Aldrich Chemical Co. The stock FeSO, solution (40 mM) was prepared anaerobically with Ar-flushed double deionized water and a few drops of H 2 SO,. Calf thymus DNA was purchased

DNA AND IRON-TALLYSOMYCIN REDOX PROPERTIES 273

from Sigma Chemical Co. The concentration of the prepared DNA stock solution in Hepes buffer, pH 7.0, was estimated according to its absorbance at 260 nm using e2- = 6.6 mM_’ cm-‘. Ultrapure grade Ar gas was purchased from Linde. Chelex 100 resin was obtained from Bio-Rad. KClO, and Fe,(SO& were from Fisher Scientific Co.

Metboda

The cyclic voltammetric measurements were performed with a BAS CV-27 Cyclic Voltammograph (Bioanalytical System Inc., West Lafayette, Indiana). Before the experiments, the voltammograph was tested with K,Fe(CN), as suggested by the manufacturer. Ag/AgCl electrode was used as the reference electrode with Pt as the auxihary electrode. A glassy carbon electrode was chosen as the working electrode [21] because earlier studies on metallobleomycins showed that a hanging mercury drop electrode did not yield the reduction peak [20], nor did platinum and gold electrodes [21] in the absence of chemical mediators [ 19, 2 11. A recent study [22], however, showed that the reduction peak of iron-bleomycin could be obtained using a hanging mercury drop electrode. Anaerobic condition in the reaction vessel was achieved by continuous purging of Ar gas. The reaction solutions were also bubbled with Ar gas for a few minutes immediately before the cyclic voltammetric measure- ments. All the voltammograms were taken at a scan rate of 50 mV/s and a starting potential of 0.4 V unless stated otherwise.

All cyclic voltammetric measurements were performed in 50 mM Hepes buffer with 0.1 M KClO, as the electrolyte. At this Hepes concentration, we estimated using the Henderson-Hasselbalch equation that the pH of the reaction mixtures did not change by more than 0.1 pH unit when the acidic Fe(R) stock solution was added and additional protons were released [31, 321 in the formation of the Fe-TLM complexes. Hepes was selected because it did not interact with metallobleomycins [ 191, in contrast to other buffer systems such as phosphate [33]. Without a buffer, the experimental results may be very difficult to interpret because the reduction of Fe(TII)-BLM to Fe@)-BLM is strongly pH dependent [21]. In the iron-tallysomycin complex solutions, the iron concentration was slightly less than the stoichiometric amount to ensure that no free iron exists in the systems; the typical iron/TLM ratio used was 0.9. In systems containing DNA, the concentration of nucleic acid used was equal to that of iron-tallysomycin complex. Athough excess nucleic acids were usually added in studies by others [34,35], the results of our present study showing the effects of DNA addition to the redox properties of Fe-TLM nevertheless indicated the occurrence of binding reactions between the reactants at the stoichiome- try used here.

The Fe(H)-TLM solution was prepared by anaerobically transferring appropriate amounts of the 40 mM Fe(H) stock to an Ar purged TLM solution. When needed, the At-purged calf thymus DNA was added into the Fe(H)-TLM solution anaerobi- cally. Fe(Ill)-TLM solutions were prepared either by air oxidation of Fe(R)-TLM described above, or by direct mixing of an appropriate amount of Fe,(SO,), stock with a TLM solution. Fe(IH)-TLM-DNA solutions were prepared by one of the following procedures: 1) air oxidation of Fe(H)-TLM-DNA, or 2) addition of an appropriate amount of DNA into the Fe(HI)-TLM solutions.

After every cyclic voltamme&ic measurement, the reaction vessel and electrodes were washed with 0.25 M EDTA solution to minimize residual iron contamination. This was followed by the cyclic voltammetric scan of Hepes-KClO,. Whenever

214 J. Sun and D. K. Y. Solaiman

FIGURE 1. (A) Cyclic vohmmogmm of0.4mMFe@)-TLMin0.05MHepesbufkr@H 7.0) with 0.1 M KClO, as elcctrolytc. @) Cyclic volmmmogmm of 0.3 mM &o-TLM-DNA ia0.05MHepes~~@H7.0)widrO.lMKCIO,aselectrdyte.~scanswereinitisted at -0.5 V. Scan rate: 50 mV/s.

peaks were deteaed in the range of scanning, the glassy carbon electrode was repolished and all other ele+%odes and naction vessels were again washe4l with 0.25 MEDTA.Ifnecessary,~proceduresweretepeateduntilnoironredoxpeeks were detected in the Hepes-KClO, cyclic voltamme@ic scan.

RESULTS AND DEXXJSSION

The cyclic voltammograms of anaerobically prepa& Fe(H)-TLM exhibited an irreversible reduction peak at a potential of - 230 f 5 mV vs Ag/AgCl (Fig. 1A). With the anaerobic addition of Ar-purged calf thymus DNA, this irreversible peak shiftedtoapotentialof - 140 f 10 mV vs Ag/AgCl (Fig. 1B). The cyclic voltammograms of both Fe@)-TLM and Fe(Il)-TLM-DNA displayed a broad, hdihguishable oxidation pet&. All these peak4 are assigned to the Fe(llI)- TLM/Fe(II)-TLM or Fe(III)-TLM-DNA/Fe@)-TLM-DNA, because no redox peaks were found in the cyclic ~01~ of control samples such as Hepes-KCIO,, BLM-Hepes-KClO,, and DNA-Hepes-KClO, at comparable concentrations (data not shown). Fe@)-Hepes alone showed a reducGon peak potent4 of -330 mV vs Ag/AgCl electrode in our experiment (data not shown); this peak did not occur in the voltammograms of Fe-TLM complexes. Thus, Hepes doesn’t seem to interfere with the formation of Fe-TLM complexes, as had been reported earlier hr BLM [ 191. The reduction peak potential of ferrous tallysomycin in the absence and presence of calf thymus DNA are all within the redox potential range of common physiological reductants, implying that the iron-bleomycin and its DNA-add& may react with these reagents as suggested [36,37J. In fact, ferric bleomycin can be reduced with ascorbate and cysteine [19]; and glutathione e&ances the DNA strand scission caused by iron-bleomycin [38].

In the presence of DNA, the Pe(III)-TLM/Fe(II)-TLM reduction potential shifts toward a more positive direction by about 90 mV. Therefore, it appears that in the presence of DNA, Fe@)-TLM will the-y be more favored in its equilibrium with Fe@)-TIM. In this sense, the substrate DNA seems to enhance its own degradation by Fe@)-TLM or Fe@)-BLM, in an auto-regenemkd catalysis process.

DNA AND IRON-TALLYSOMYCR’I REDOX PROPHtTXES 275

1.4 0.2 E&sAgY&l)

-0.4

FIGURE 2. Cyclic voltamnqrant ofM-hrairoxidi?md0.4mMFc(II)-TLMill0.OsM Hqm bulb (pH 7.0) with 0.1 M KClO, as elecmlytc. Starting potcdl: 0.4 V. 8caa rate: 50 mV/e.

Wudumlly, bleomycht and its derivatives are viewed as constituting three do mains, htchuiing DNA-binding domain and metalbmding/oxygen activation do- main. The humcakion of bithiaxole moiety of DNA-bii domain into the DNA constitutes a commonly accepkd model of metallobleomycin-DNA ternary complex [24-27). A meent report suggested that direct ionic interaction between the metal center and m or more negatively charged groups in DNA [28] may also be involved inthetenrarycomplexfonnotioa.nLelargeshiftinthe~~~potentialoftbe Fe(H)-TLM-DNA system seems to support the occurrt~~oe of such an hueraction. A binding mode without any interaction between the metal center (domain) and the DNA-binding domain of TLM is not likely to result in such a reduction potemhd perhubation. The patticipation of the metal-bii domain in the metallobleomycin- DNA htteraction may play an @tortant role in the site specific recognition of bleomycins, as demoWmted by the fact that different metals may give different DNA degradation produ&r [39]..

Figure 2 shows the cyclic voltammogram of the Fe@l)-TLM pmpared by air oxidation. Exposum of Fe(E)-TLM to air shifted the reduction peak potential to a less negative value. After 24 hours of air exposure, it became an almost reversible, one-electron redox process (Ipc/IP = 0.90 and AE = 65 mV) with an even less negative reduction potential (- 100 mV vs AgjAgCl). Again, these reduction potentials ate within the reach of biological redwing agents. Fulk [21] mported a similar shifting and broadening of the teduction peak of aircxposed Fe(B)-BLM sample on glassy carbon electrode. Although the author suggeskd the existence of new and multiple species during air-oxidation, no specitic assignment to the known Fe-BLM intermedky species [33, 40-441 was made. Miyoshi et al. [22] recorded the cyclic voltammogram of Fe(B)-BLM in pH 8.5 borate buffer after bubbling the samplewith1ppnO,for3omin;aadobservedthe occurrenceofareductionpeak at -255 mV vs SCE. Their tesults are similar to the observation with the 24-hr air oxidixed Fe(B)-TLM sample of this study. This species had been proposed by Miyoshi et al. based on the results of a concurtent EER study [22], to be the active/activated Fe(IH)-BLM-OOH- complex, which in turn is the pmcurso rofthe “activated” BLM-Fe(V) = 0 species [35,42-441. In seeming agreement with this assivnt, the addition of DNA sub&rate to the air-oxidixed Fe(E)-DNA results in thedisappearanceofthereductionpotentialat - 100 mV vs Ag/AgCl (data not shown). The activated Fe-TLM species had appatently undergone reaction with DNA substrate to become Fe(lH)-TLM [40], whose cyclic voltammogram lacks apparent redox peaks. Indeed, Fe(BI)-TLM pmpared directly from Fe&O,), in

216 J. 'Sun and D. K. Y. Solaiman

Hepcs-KC104 exhibited no recognizable reduction or oxidation peak either in the absence or presence of calf thymus DNA (data not shown). As expected, when near stoichiometric mixtures of Fe@), TLM, and calf thymus DNA were exposed to air, no redox peaks were discernable in CV scans (data not shown). Under this condition, DNA degradation reaction is known to occur f40], resulting in the oxidation of Fe@)-TLM to the Fe(III)-TLM with’ featureless cyclic voltammogram. Therefore, it is obvious that Fe@)-TLM and air-oxidized Fe@)-TLM are different species as proposed [40]. Furthermore, Dabrowiak and Santillo [20] demonstrated that the coordination of pyrimidine moiety to iron was different in ferric bleomycin prepared directly and via air oxidation.

This work was supported by the College of Arts and Sciences Faculty Development Award (0s) from Duquesne University.

REFERENCES

1. H. Umazawa, K. Maeda, T. Takeuchi, and Y. Okami, J. Antibiot. Ser. A19, 200 Wm.

2. H. Umezawa, in Bleomycin: Chemical, Biochemical and Biological Aspects, S. M. Hecht, Ed., Springer-Verlag, New York, 1979, pp. 24-36.

3. S. T. Crooke, in Bleomycin: Current Status and New Development, S. K. Carter, S. T. Crooke, and H. Umezawa, Eds., Academic Press, New York, 1978, pp. l-8.

4. S. K. Carter, in Bleomycin: Current Status and New Development, S. K. Carter, S. T. Crooke, and H. Umezawa, Eds., Academic Press, New York, 1978, pp. 9- 14.

5. J. E. Schurig, W. C. Rose, R. S. Hirth, A. Schleht, J. B. Huftalen, A. P. Florczyk, and W. T. Bradner, Cancer Chemother. Pharmacol. 13, 164 (1984).

6. S. M. Hecht, in Bleomycin: Chemical, Biochemical and Biological Aspects, S. M. Hecht, Ed., Springer-Verlag, New York, 1979.

7. H. Umezawa, in Anticancer Agents Based on Natural Products, J. M. Cassady and J. D. Douros, Eds., Academic Press, New York, 1980, p. 147.

8. R. M. Burger, J. Peisach, and S. B. Horwitz, L:ife Sci. 28, 715 (1981). 9. E. A. Sausville, J. Peisach, and S. B. Horwitz, B&hem. Biophys. Res. Commun. 73,

814 (1976). 10. E. A. Sausville, J. Peisach, and S. B. Horwitz, Biochemistry 17, 2740 (1978). 11. E. A. Sausville, R. W. Stein, J. Peiscach, and S. B. Horwik, Biochemistry 17, 2746

(1978). 12. T. Takita, Y. Muraoka, T. Nakatani, A. Fujii, Y. Itaka, and H. Umazawa, J. Antibiot.

31, 1073 (1978). 13. E. A. Rao, L. A. Saryan, W. E. Antholine, and D. H. Petering, J. Med. Chem. 23,

1310 (1980). 14. H. Suzaki, K. Nagai, E. Akutsu, N. Yamaki, N. Tanaka, and H. Umazawa, J. Antibiot.

23, 473 (1970). 15. Y. Itaka, N. Nakamura, T. Nakatani, Y. Mutaoka, A. Fujii, T. Takita, and H.

Umazawa, J. Antibiot. 31, 1073 (1978). 16. J. C. Dabrowiak, F. T. Greenaway, W. E. Longo, M. Van Husen and S. T. Crooke,

Biochim. Biophys. Acta 517, 517 (1978). 17. J. E. Strong and S. T. Crooke, Cancer Res. 38, 3322 (1978). 18. W. J. Lown and S. Sims, Biochem. Biophys. Rex Commun. 77, 1150 (1977). 19. D. L. Melnyk, S. B. Horwik, and J. Peisach, Biochemistty 20, 5327 (1981). 20. J. C. Dabrowski and F. S. Santillo, J. Electrochem. Sot. 126, 2091 (1979).

DNA AND IRON-TALLYSOMYCIN REDOX PROPERTIES 277

21. M. L. Fultz, PhD D-on, University of Maryland, 1982. 22. K. Miyoshi, T. Kikuchi, T. Takita, S. Murato, and K. Ishizu, Inorg. Chim. Actu 151,

45 (1988). 23. N. J_ Opp&ebner, C. Chang, L. 0. Rodrigwz, and S. M. Hecht, J. Biol. Chem. 256,

1514 (1981). 24. H. Kasai, H. Naganawa, T. Takita, and H. Umazawa, J. Antibiof. 31, 1316 (1978). 25. S. N. Roy, G. A. Grr, C. F. Brewer, and S. B. Horwitz, Cancer Res. 41,4471(1981). 26. J. D. Glickson, R. P. Pillai, and T. T. Sakai, Proc. Notl. Acad. Sci. U.S.A. 78, 2%7

(1981). 27. C. H. Huang, L. Galvan, and S. T. Crooke, Biochemistry 19, 1761 (1980). 28. M. J. Levy and S. M. He&t, B&hem&y 27,2647 (1988). 29. F. T. Greenaway, J. C. Dabrowiak, M. Van Husen, R. Grulich, and S. T. Crooke,

B&hem. Biophys. Res. Commun. 85, 1407 (1978). 30. J. C. Dabrowiak, in Metul Ions in Biologicui Systems, H. Sigel, Ed., Marcel Dekker,

New York, 1980, pp, 305-333. 31. D. Solaiman, in Metai-Based Anti-Tumour Drugs, M. F. Gielen, Ed., Fread,

London, 1988, pp. 235-256. 32. D. Solaiman, E. A. Rao, W. Antholine, and D. H. Petering, J. Znorg. B&hem. 12 201

(1980). 33. R. M. Burger, S. B. Honvitz, J. Peisach, and J. B. Wittenberg, J. Eiol. Chem. 254,

12299 (1979). 34. L. E. Rabow. J. Stubbe, and J. W. Kozarich, J. Am. Chem. Sot. 112, 31% (1990). 35. J. R. Barr, R. B. Van Atta, A. Natrajan, and S. M. Hecht, J. Am. Chem. Sot. 112,

4058 (1990). 36. D. Solaiman, E. A. Rao, D. H. Petering, R. C. Sealy, and W. E. Antholine, Znt. J.

Radiation Oncoiogv Biol. Phys. 5, 1519 (1979). 37. W. E. Antholine and D. H. Petering, B&hem. Biophys. kes. Commun. 90, 384

(1979). 38. P. L. Moseley and R. Chalkley, J. Lab. Clin. Med. 110, 618 (1987). 39. G. M. Ehrenfeld, J. B. Shipley, D. C. Heimbrook, H. Sugiyama, E. C. Long, J. H. van

Boom, G. A. van der Marel, N. J. Oppenheimer, and S. M. Hecht, Biochemistry 26, 931 (1987).

40. R. M. Burger, J. Peisach, and S. B. Horwitz, J. Biol. Chem. 256, 11636 (1981). 41. H. Kuramo~hi, K. Takahashi, T. Takita, and H. Umazawa, J. Antibiot. 34,576 (1981). 42. R. M. Burger, J. S. Blanchard, S. B. Honvitz, and J. Peisach, .I. Biol. Chem. 260,

15406 (1985). 43. L. E. Rabow, G. H. McGail, J. Stubbe, and J. W. Kozarich, J. Am. Chem. Sot. 112,

3203 (1990). 44. A. Natrajan, S. M. Hecht, G. A. van der Marel, and J. H. van Boom, .I. Am. Chem.

Sot. 112,4532 (1990).

Received January 25, 1990; accepted June 25, 1990