Photochemistry and Photobiology, 2005, 81 : 819-822

Symposium-in-Print: Vlll ELAFOT, La Plata, Argentina, 2004

Induced and Photoinduced DNA Damage by Quinolones: Ciprofloxacin, Ofloxacin and Nalidixic Acid Determined by Comet AssayS

Georgina Sanchez*’, Maria Eliana Hidalgo2, Jose Miguel Vivanco’ and Jorge Escobar3 ’ Facultad de Farmacia, Universidad de Valparaiso, Valparaiso, Chile *Facultad de Ciencias, Universidad de Valparaiso, Valparaiso, Chile 31nstituto de Quimica Pontificia, Universidad Catolica de Valparaiso, Valparaiso, Chile

Received 30 November 2004; accepted 24 January 2005

ABSTRACT

Quinolones are degraded by light with loss of their antimi- crobial activity, generating active species or radicals. Evidence exists that some fluoroquinolones (lomefloxacin, fleroxacin and enoxacin) induce damage to the cellular membrane and DNA cleavage by photosensitization. In this study, the genotoxic potential of the quinolones ofloxacin, nalidixic acid and ciprofloxacin (three antimicrobials frequently used in therapy) was evaluated upon irradiation with UV light by using the comet assay on cells of the Jurkat line. The results demonstrate that there are significant differences between the control groups (positive control with 50 pM H202, negative controls without drug and with and without irradiation) and the groups of irradiated quinolones (ofloxacin 2.76 X M , nalidixic acid 2.15 X M and ciprofloxacin 2.01 x M), indicating that, at the dose of irradiation employed (necessary to produce 50% photodegradation), the photodecomposition of the quinolones enhanced DNA damage. The unirradiated drugs also exhibited genotoxicity significantly different from that of the negative control.

INTRODUCTION Quinolones are antimicrobials classified, according to their action spectra, into first and second generation. The first-generation quin- olones exercise their effect on gram-negative microorganisms: Escherichia coli, Pseudomonas aeruginosa, Klebsiella pneumo- niae, etc., whereas the second-generation quinolones also act on gram-positive microorganisms: Staphylococcus aureus, Staphylo- coccus pyogenes, Streptococcus pneumoniae, etc.

The first-generation quinolones lack a fluorine atom at position 6 and representatives include nalidixic acid, oxolinic acid, pipemidic acid and rosoxacin. The quinolones of the second generation are fluoroquinolones with a fluorine at position 6 and

YPosted on the website on 2 February 2005 *To whom correspondence should be addressed: Facultad de Farmacia,

Universidad de Valparafso, Gran Bretaiia 11 11, Valparaiso, Chile. Fax: 56-32-5081 1 1; e-mail: georgina.sanchez@uv.cl

0 2005 American Society for Photobiology 0031-8655/05

include ciprofloxacin, enoxacin, lomefloxacin, norfloxacin and ofloxacin.

A third generation, consisting of difluoro and trifluoro derivatives, also exists and a fourth generation is under de- velopment. Quinolones of the first generation, which lack the piperazine ring, exhibit important phototoxicity reactions (1). Conversely, the presence of this ring leads to high photolability, potentially resulting in a decrease in the therapeutic effect of the drug or the formation of photoproducts that might eventually pro- duce phototoxic reactions (1,2).

The photodegradation of quinolones probably occurs via a free radical mechanism, with loss of the carboxylic group at the ortho position. Decarboxylation upon irradiation has been demonstrated for aryl acetic acid derivatives (3), a-ketocarboxylic acids and nalidixic acid (3-5).

The comet assay is currently extensively used in in vivo and in vitro studies for the evaluation of the genotoxic potential of a variety of toxic agents such as chemical compounds, ionizing radiation and UV radiation (&8), as well as the phototoxic potential of chemical compounds such as fluoroquinolonic antibiotics (9) following irradiation with UV light. Because the comet assay re- quires the use of a fluorescence microscope, we employed a confocal microscope, which is more sensitive than a conventional micro- scope, has higher resolution and can measure directly the tail moment of a cellular population.

In the present communication, we report data bearing on the genotoxic potential of three quinolones (ofloxacin, nalidixic acid and ciprofloxacin), both unirradiated and irradiated to 50% pho- todegradation with a light intensity (5.640 mW/cm2)* equivalent to that of a clear summer day in Valparaiso, Chile.

MATERIALS AND METHODS Quinolone genotoxic capacity determination. Solutions of quinolones were prepared in phosphate-buffered saline (PBS) at pH 7.4 at concentrations comparable to typical plasmatic levels. Solutions were irradiated to 50% conversion of the initial uinolone to photoproducts in quartz cells with an irradiance of 2 mW/cm . The total radiation doses required for 50% 9

*Data provided by the Station Out Door from Universidad Tknica Federico Santa Chile (Physics Department and Solar Energy Laboratory, January-May 2002).

81 9

820 Georgina Sanchez eta/.

Nalidixic acid Ciprofloxacine

Of loxacine Figure 1. Structure of the quinolones studied.

conversion were 720 mJ/cm2 for ofloxacin (2.76 X M initially) (Hoechst Laboratories, Barcelona, Spain), 960 mJ/cmZ for nalidixic acid (2.15 X lo4 M) (Chile Laboratories, Santiago, Chile) and 360 mJ/cm2 for ciprofloxacin (2.01 X lo-’ M) (Chile Laboratories).

Comet assay. Cells of the Jurkat line (human T lymphocytes) were embedded in thin coverings of agarose on a glass microscope slide. Cells were lysed by immersing the slides in lysis buffer (2.5 M NaC1; 0.1 mM ethylenediaminetetraacetic acid; 10 mM Tris, pH lo), followed by electrophoresis of the uncoiled DNA under alkaline conditions. The DNA was visualized by staining with ethidium bromide. This procedure was applied to irradiated cells, cells incubated with quinolone unirradiated, cells incubated with quinolones and irradiated, and cells incubated with hydrogen peroxide (50 pM). The samples were analyzed with the aid of a fluorescence microscope (SLM S10 confocal microscope; Zeiss, Santiago, Chile). The damage, length and intensity of the staining of the comet were quantified by means of a program (CASP 1 .O. 1, Free Software Foundation Inc, Boston, MA) that calculates the moment of the line (tail moment).

Statistical analysis. The means and standard deviations (SD) were calculated for the tail moment. The Kruskal-Wallis test (P < 0.05) was used to compare the means of the different groups.

RESULTS The chemical structures of the quinolones studied are shown in Fig. 1. The drugs per se cause genotoxic damage in all cases; significant statistical differences were observed only between cip- rofloxacin and the other two drugs, nalidixic acid and ofloxacin (Fig. 2). All three drugs were also genotoxic (Fig. 3) under our

Figure 3. Quinolone effect on Jurkat cells with irradiation. (Tail mo- ment 2 SD.)

irradiation conditions (50% photodegradation of the initial drug), with no significant statistical differences between the three drugs. For nalidixic acid (Fig. 4) and ofloxacin (Fig. 5) , the differences between irradiated and unirradiated samples were not statisti- cally significant, i.e. the genotoxicity was not enhanced upon ir- radiation of the drug. Irradiation of ciprofloxacin increases the damage; the difference is significant relative to the unirradiated samples (Fig. 6).

DISCUSSION The sensitivity of Jurkat cells to UV light is known and they have been employed in numerous studies, such as the induction of apoptosis by UV-A1 (340-400 nm) involving singlet oxygen (10) or the determination of the cytotoxic and phototoxic effects of chemical compounds irradiated with UV-A and UV-B light (1 1). The effect of UV irradiation on the Jurkat cells is analyzed by comparing the damage to DNA (tail moment) in the irradiated control with that in the unirradiated control (Fig. 2). Comparing the damage to DNA for the three quinolones (Figs. 3-6) without irradiation, all three are genotoxic; ofloxacin and ciprofloxacin are comparable and ciprofloxacin is slightly (but the difference was statistically significant) less toxic (Fig. 3). Although irradiation of nalidixic acid clearly failed to increase the damage to the DNA (Fig. 3), a statistical study indicates that there art no significant differences between the three quinolones upon irradiation. In the case of ciprofloxacin and ofloxacin, the damage potentially might be due to the formation of a highly reactive carbene species via the loss of fluorine at the 6 position, as has been reported for

Figure 2. Quinolone effect on Jurkat cells without irradiation. (Tail moment 2 SD.)

Figure 4. Nalidixic acid effect on Jurkat cells with and without irradiation. (Tail moment f. SD.)

Photochemistry and Photobiology, 2005, 81 821

Figure 5. Ofloxacine effect on Jurkat cells with and without irradia- tion. (Tail moment 2 SD.)

enoxacin (3.3 X lo-’ M) irradiated for 20 min (310-390 nm) with an irradiance of 0.80 mW/cm2 (equivalent to a radiation dose of 960 mJ/cm2) (12,13). However, we cannot discard the possibility that the damage to the DNA is mediated by a Type-I1 mechanism with the intermediacy of singlet oxygen (14,15). The presence of a piperazinic ring would produce genotoxicity and increases the photolability (16).

If we compare irradiated and unirradiated nalidixic acid (Fig. 4), there is an increase in the damage upon irradiation, although the increase is not sufficient to be statistically significant. Irradiation of nalidixic acid generates free radicals and singlet oxygen, both of which can affect cellular structures. Irradiation of the drug generates a decarboxylated photoproduct with a higher photo- hemolytic potential than nalidixic acid itself, suggesting that the carboxylic acid group at position 3 is not responsible for the photohemolytic effect and pointing to processes mediated by singlet oxygen and/or the hydroxyl radical. On the other hand, it has been demonstrated that nalidixic acid photoproducts pro- duce photohemolysis by an oxygen-dependent mechanism (1 3). Irradiation of the drug for 24 h in methanol with a mercury lamp at low atmospheric oxygen concentrations generates photoprod- ucts that can produce phototoxicity through a mechanism involving free radicals and singlet oxygen (3). Comparatively, nalidixic acid (5-500 I*M) irradiated with a dose of 2 X lo5 d / c m 2 in cells derived from human laringo carcinoma (Hep-2) and in pri- mary cultures of chick embryo fibroblasts (CEF) is also found to be phototoxic (17). With a UV-A dose of 3750 mJ/cm2 in V79 hamster fibroblasts, nalidixic acid produces smaller comets than fleroxacin, lomefloxacin or ciprofloxacin, demonstrating that the

Figure 7. Comet Assay photograph.

(18). The damage to DNA upon UV-A radiation of the plas- mid pBR322 (the vector normally used in genetic engineering for cloning genes) is minor relative to that produced by fleroxacin and lomefloxacin, which possess two fluorine atoms in their struc- tures, at positions 6 and 8, and generate carbenes at position 8, considerably increasing their genotoxic capacity relative to norfloxacin and enoxacin (19).

Irradiation of ciprofloxacin increases the damage to DNA significantly (Fig. 6), indicating that the photoproducts are more genotoxic than the drug itself, which, as mentioned previously, might be due to the generation of a carbene, as in the case of enoxacin (12,13). Studies with the same drug in mouse lymphoma cells, irradiated with a solar simulator (with a reduced UV-B component) demonstrated similar comets, photographed directly from the confocal microscope (Fig. 7) for ciprofloxacin, fleroxacin and lomefloxacin, which could be attributed to the participation of very short-lived genotoxic species (20).

In the case of ofloxacin, irradiation decreases the damage slightly (Fig. 5) , although the difference is not statistically significant. Presumably, the resulting photoproducts are of low toxicity and low photogenotoxicity. Although ofloxacin may also generate a carbene or singlet oxygen (12,14), the efficiency of generation of these may be much lower. Indeed, work in progress has shown that the presence of an alkoxy substituent (methoxy group) at position 8 reduces both the photolability and the phototoxicity (P. Gonzalez, unpublished).

flUOrOqUinOlOneS are more photogenotoxic in this Cellular model Acknowledgements-We thank Dr. Manuel Ellahuefie for providing us with the computational program to calculate the tail moment, Dr. Miguel Rios for allowing us to use the confocal microscope, and DIPUV Project 15/2001 for financial support.

REFERENCES 1. Fernindez, E., A. M. Cirdenas and G. Martinez (1987) Phototoxicity

from nalidixic acid: oxygen dependent photohemolysis. Fdrmaco 42, 68 1-690.

2. Baes, H. (1968) Photosensitivity caused by nalidixic acid. Dermato- logica 136, 61-64.

3. Vargas, F., C. Rivas and R. Machado (1991) Decarboxylation and singlet oxygen production in the photolysis of nalidixic acid. J . Photochem. Photobiol. B: Biol. 11, 81-85.

4. Moore, D. E. (1977) Photosensitization by drugs. J . Pharm. Sci. 66(9), 1282-1 284.

5. Femindez, E. and A. M. Cirdenas (1990) The mechanism of photohaemolysis by nalidixic acid. J . Photochem. Photobiol. Bc Biol. 4, 329-333.

Figure 6. Ciprofloxacin effect on Jurkat cells with and without irradiation. (Tail moment ? SD.)

822 Georgina Sanchez eta/.

6. Collins, A., M. DuSinskL and A. Horska (2001) Detection of alkylation damage in human lymphocyte DNA with the comet assay. Acta Biochim. Polonica 48(3), 61 1-614.

7. Garaj-Vrhovac, V. and D. Zeljezic (2001) Cytogenetic monitoring of Croatian population occupationally exposed to a complex mixture of pesticides. Toxicology 165, 153-162.

8. Garaj-Vrhovac, V. and D. Zeljezic (2000) Evaluation of DNA damage in workers occupationally exposed to pesticides using single-cell gel electrophoresis (SCGE) assay: pesticide genotoxicity revealed by comet assay. Mutation Res. 469, 219-285.

9. Meunier, J., A. Sarasin and L. Marrot (2002) Photogenotoxicity of mammalian cells: a review of the different assays for in vitro testing. Photochem. Photobiol. 75(5), 4 3 7 4 7 .

10. Godar, D. E. (1999) UVAl radiation triggers two different final apoptotic pathways. J . Invest. Dermatol. 112(1), 3-12.

11. Rancan, F., S. Rosan, F. Boehm, A. Cantrell, M. Brellreich, H. Schoenberger, A. Hirsch and F. Moussa (2002) Cytotoxicity and photocytotoxicity of a dentritic C (60) monoadduct and a malonic acid C (60) tris-adduct on Jurkat cells. J . Photochem. Photobiol. B: Biol. 67(3), 157-162.

12. Sortino, S., G. Condorelli, G. De Guidi and S. Giuffrida (1998) Molecular mechanism of photosensitization XI. Membrane damage and DNA cleavage photoinduced by enoxacin. Photochem. Photobiol. 68(5), 652-659.

13. Fasani, E., A. F'rofumo and A. Albini (1998) Structure and medium- dependent photodecomposition of fluoroquinolone antibiotics. Photo- chem. Photobiol. 68(5), 666614.

14. Sauvaigo S., T. Douki, F. Odin, S. Caillat, J. L. Ravanat and J. Cadet (2001) Analysis of fluoroquinolone-mediated photosensitization of 2'- deoxyguanosine, calf thymus and cellular DNA: determination of type- I, type-I1 and triplet-triplet energy transfer mechanism contribution. Photochem. Photobiol. 73(3), 230-231.

15. Spratt, T. E., S. S. Schultz, D. E. Levy, D. Chen, G. Schluter and G. M. Williams (1999) Different mechanisms for the pho- toinduced production of oxidative DNA damage by fluoro- quinolones differing in photostability. Chem. Res. Toxicol. 12(9),

16. Domagala, J. M. (1994) Structure-activity and structure-side-effect relationships for the quinolone antibacterials. J . Antimicrobial Chrmo- ther. 33, 685-706.

17. CortBs, M. P., A. M. Cirdenas, E. Navarrete and E. Femindez (1991) Phototoxicity induced by nalidixic and oxolinic acids: decrease in cell survival of chick embryo fibroblasts and Hep-2-cells. J . Photochem. Photobiol. B: Biol. 9, 229-234.

18. Reavy, H. J., N. J. Traynor and N. K. Gibbs (1997) Photogenotoxicity of skin phototumorigenic fluoroquinolone antibiotics detected using the comet assay. Photochem. Photobiol. 66(3), 712-71 8.

19. Martinez, L. and C. F. Chignell (1998) Photocleavage of DNA by the fluoroquinolone antibacterials. J . Photochem. Photobiol. B: B i d . 45( l), 51-59.

20. Chetelat, A. A,, S. Albertini and E. Gocke (1996) The photo- mutagenicity of fluoroquinolones in tests for gene mutation, chromo- somal aberration, gene conversion and DNA breakage (Comet assay). Mutagen 11(5), 497-504.

809-8 15.