Cancer Letters, 13 (1981) 265-268 o Elaevier/North-Holland Scientific Publishers Ltd.

265

PHOTOLABILITY OF BLEOMYCIN AND ITS COMPLEXES

NUTAN THAKRAR and KENNETH THOMAS DOUGLAS

Department of Chemistry, University of Essex, Wivenhoe Park, Colchester, Co4 3SQ (U.K.)

(Received 8 April 1981) (Accepted 13 May 1981)

SUMMARY

Bleomycin has been found to be very sensitive to photolysis and on irradiation with the full spectral range of a medium-pressure mercury lamp undergoes a number of photo-induced reactions; there is a process in which the absorbance at 310 nm decreases, and a slower process in which it increases with photolysis time. The faster process can be studied by irradia- tion at -300-350 nm, is complete in -8 min, obeys first order kinetics, but is itself biphasic. The Fe(H) and Cu(I1) complexes of bleomycin are also photolabile, as is the bleomycin-DNA complex.

The bleomycins, a family of strongly-chelating, glycopeptide antibiotics discovered by Umezawa et al., [7] are useful in the treatment of certain tumours [ 41. Bleomycin induces single-stranded and double-stranded breaks in DNA in the presence, in vitro, of Fe(H) and molecular oxygen [4]. In spite of its probable action as the Fe(I1) complex, bleomycin is believed to enter the cell as its Cu(I1) complex, the form in which it is isolated [4]. Consequently, considerable recent effort has gone into its inorganic chemistry [ 31. It has been postulated that the bisthiazole group and terminal amine groups of bleomycin bind to DNA [ 1,6], and other functional groups serve as ligands for metal ions. However, the detailed mechanisms by which bleo- mycin causes strand breaks is yet unclear, although the participation of superoxide free radical is likely [3,6]. Mizuno [5] has pointed out that other sources of free radicals should also be considered and commented on the need to control light in such experiments. For this reason we investigated the photosensivity of bleomycin sulphate (Sigma Chemical Co .).

Photoirradiation was carried out by means of an Applied Photophysics Quantum Yield Photoreactor with a high-intensity, medium-pressure mercury lamp as source mounted on an optical rail. Wavelength selection, of incident radiation, was done by means of chemical-combination filters

266

[8]. Spectral measurements were made at 25°C on a Carlo Erba Spectracomp 601 Microcomputer&d Spectrophotometer.

Figure 1 shows the spectral changes of bleomycin irradiated at pH 7.00 in 0.05 M MES buffer at 22°C as above, with no incident wavelength selection at 25 cm from the source. It is clear that with this wide spectrum of incident radiation there are 2 photo-induced processes. A relatively rapid decom- position, seen as a decrease in absorbance at -290 nm over the first -8 min and with an isosbestic point at 319 nm, is followed by a slower process, over the next hour or so, which leads to a new absorption band at -340 nm. The inset to Fig. 1 shows these 2 processes reflected in the absorption changes at 310 nm as a function of irradiation time. By irradiating with a ‘wave- length window’ between -300-350 nm, the secondary process was slowed sufficiently that the faster primary process could be isolated and studied. This process is complex for although it has a reasonably tight isosbestic point at -319 nm (see Fig. 2), there is a transitory isosbestic point at -270 nm.

In spite of this the time course of the spectral changes at 295 nm closely approximates firstorder behaviour (by non-linear regression analysis of the data with the first-order exponential equation, see inset to Fig. 2), with a half-life of a few minutes under the conditions used. We have also found that the Cu(I1) and Fe(I1) complexes of bleomycin photolyse smoothly but more slowly on irradiation with a wavelength window of 300-350 nm. However,

0 800

0600

\o--0’ I I I I

20 40 60

305 345 385 425

Wavelength ( nm 1

K a

Fig. 1. Spectra of 6 x 10“ M bleomycin (Lot. No. 89C-2040) in 0.05 M ME8 buffer (pH 7.00) as a function of photolysis time at 22°C and 26 cm from a medium-pressure mercury lamp with no filtering of incident light. ‘Ihe numbers for each of the spectra refer to irradiation times in minutes. The inset shows the time course of the absorbance at 310 nm of this bleomycin solution as a function of photolysis time.

267

l-200

o-300

0

25.28

I I I

06

270 290 310 330 350

Wavelength ( nm 1

Fig. 2. Spectra of 9.6 x lo-’ M bleomycin (Lot. No. 89C-0240) at pH 7.00 (0.05 M MES buffer) as a function of photolysis tie at 22°C and 15 cm from a medium-pressure mercury lamp with incident light in the wavelength region -300-350 nm. The numbers indicate photolysis times in minutes. The inset shows the time course of the photolysis at 295 nm: points are experimental; line is theoretical for a first-order process with k Ous = 0.209 min-‘; A,,, at t = 0 of 1.35; and A,,, at t = - of 0.616, calculated by non- linear regression analysis and the first-order exponential equation.

TABLE 1

INFLUENCE OF DNA ON THE RATE OF PHOTOLYSIS OF BLEOMYCIN AT 22°C IN pH 7.00 ME6 BUFFER (0.05 M)

[DNA1 (r&d)” kobs (min - ’ )b

0 0.251 1.55 0.249 3.10 0.202 4.65 0.426 6.20 0.795

The bleomycin concentration was 6 X 10e5 M. a Linear form DNA was prepared from trimethoprim-resistant E. coli by Dr. Wayne Davies of the Department of Biology, by a procedure with C&l ultra centrifugation as the final step. For this DNA an optical density of 1.0 corresponded to 50 pg/ml. b The rate constants were determined as described in the text for photolysis with a wave- length window between - 300-350 nm. Under these conditions no spectral changer were detectable over the times used for DNA alone.

268

the concentration effects on the firstorder rate constant and the absorbance change during reaction are complex. In addition, increasing the concentration of DNA present in a solution of bleomycin photolysing under the above conditions causes an increase in rate, but this is not linear in the amount of DNA added. The increased rate of photolysis with DNA present must indicate that the DNA bleomycin complex (detectable by difference UV- spectroscopy) is photoactive and indeed more so than free bleomycin. The results with DNA present are summarised in Table 1. It is not clear yet whether these photo-processes have an importance to in vivo studies or to DNA-nicking, but we are currently investigating this aspect.

ACKNOWLEDGEMENTS

We are grateful to Dr. Wayne Davies of the Department of Biology for the gift of DNA and to the Cancer Research Campaign for partial support.

REFERENCES

1 Chien, M., Grollman, A.P. and Horwitz, S.B. (1977) Bleomycin-DNA interactions: fluorescence and proton magnetic resonance studies. Biochemistry, 16, 3641-3647.

2 Dabriowak, J.C. (1980) The coordination chemistry of bleomycin: a review, J. Inorg. Biochem., 13,317-337.

3 Grollman, A.P. and Takeshita, M. (1980) Interactions of bleomycin with DNA. Adv. Enzyme Regul., 1867-83.

4 Hecht, S.M. (ed.) (1979) Bleomycin: Chemical, Biochemical and Biological Aspects. Springer-Verlag, New York.

5 Mizuno, D. (1979) Bleomycin: Chemical, Biochemical and Biological Aspects, p. 338. Editor: S.M. Hecht. Springer-Verlag, New York.

6 Takeshita, M. and Grollman, A.P. (1979) A molecular basis for the interaction of bleomycin with DNA. In: Bleomycin: Chemical, Biochemical and Biological Aspects, pp. 207-221. Editor: S.M. Hecht, Springer-Verlag, New York.

7 Umezawa, H., Maeda, K., Takeuchi, T. and Okami, Y. (1966) New antibiotics, bleo- mycin A and B. J. Antibiot., 19, 200-209.

8 Zimmermann, H.E. (1971) Apparatus for quantitative preparative photolysis: The Wisconsin black box. Mol. Photochem., 3, 281-292.