J. Phys. Chem. B 2007, 111, 7409-7414
7409
Triplet Excited Fluoroquinolones as Mediators for Thymine Cyclobutane Dimer Formation in DNA Virginie Lhiaubet-Vallet,† M. Consuelo Cuquerella,† Jose V. Castell,‡ Francisco Bosca,*,† and Miguel A. Miranda*,† Instituto de Tecnologı´a Quı´mica UPV-CSIC, UniVersidad Polite´ cnica de Valencia, AVenida de los Naranjos s/n, 46022 Valencia, Spain, and Centro de InVestigacion, Hospital UniVersitario La Fe, AVenida de Campanar 21, 46009 Valencia, Spain ReceiVed: January 9, 2007; In Final Form: April 2, 2007
A series of fluoroquinolones (FQs), including enoxacin (ENX), pefloxacin (PFX), norfloxacin (NFX), its N(4′)-acetyl derivative (ANFX), ofloxacin (OFX), and rufloxacin (RFX) have been investigated to determine their potential as DNA photosensitizers via thymine cyclobutane dimer (TT) formation in DNA. At fluoroquinolone concentrations and light doses insufficient to produce direct single strand breaks, ENX, PFX, and NFX were able to produce TT dimers in DNA, revealed by enzymatic treatment with T4 endonuclease V. By contrast, ANFX, OFX, and RFX were inefficient in this assay. The absolute values of the triplet energies of ENX, PFX, NFX, ANFX, OFX, and RFX were estimated by means of laser flash photolysis, using flurbiprofen, 4-biphenylcarboxylic acid, and naproxen as energy acceptors. They were found to be 273, 269, 269, 265, 262, and 253 kJ/mol, respectively. Other triplet excited state properties of the FQs, including quantum yields and lifetimes, were also studied. All the results indicate that the threshold ET value required for a given compound to become a potential DNA photosensitizer via TT formation is in the range defined by the triplet energies of NFX and ANFX (265-269 kJ/mol). This provides the basis for an alert rule: any chemical (drugs, cosmetics, pesticides, etc.) with higher ET has to be considered with regard to its potential photogenotoxicity.
Introduction Thymine cyclobutane dimers (TT) have been the subject of increasing interest in the past decade, as these photoproducts have been associated with mutagenic and carcinogenic processes.1 This type of damage can result from direct exposure of DNA to UV radiation or also from DNA-photosensitization by triplet-triplet energy transfer.1,2 Although TT lesions are the main bypyrimidine photoproducts, other cyclobutanedipyrimidines have recently been detected in the DNA of human skin upon UVA-photosensitization.3 As the feasibility of triplettriplet energy transfer is linked to the excited state energies of the donor and acceptor compounds, the triplet energy (ET) of thymine in DNA is a critical parameter, which appears to be different from that of free thymine or thymidine (Thd).2 It can be assumed that this value is a sequence-sensitive property, as suggested by the site-dependent efficiency of TT formation in oligonucleotides.4 In this context, on the basis of the sensitized generation of a transient T-T absorption at 380 nm in laser flash photolysis experiments, a triplet energy of 310 kJ/mol for Thy and Thd 5′-monophosphate has been estimated.2 However, more recent studies have shown that compounds with lowerlying triplet states, such as benzophenone derivatives (ET ) 290 kJ/mol), are able to photosensitize TT formation from Thd at high nucleoside concentrations, together with a more favored Paterno-Bu¨chi photocycloaddition.5,6 Thus, although the triplet energy of benzophenone is lower than that estimated * Corresponding authors. F.B.: phone, +34-963877815; fax, +34963877809; e-mail,
[email protected]. M.A.M.: phone, +34-963877807; fax, +34-963877809; e-mail,
[email protected]. † Universidad Polite ´ cnica de Valencia. ‡ Hospital Universitario La Fe.
for Thy or Thd 5′-monophosphate, the endothermic process of TT formation can be achieved using effective concentrations of both components. To address this problem in DNA, the only approach has been the use of photosensitizers of different triplet energies to produce TT lesions. In this way, the upper limit for the ET value of the thymine base in DNA has been progressively shifted from 297 kJ/mol (methoxyacetophenones2) down to 290 kJ/mol (benzophenone and phthalimidine derivatives5,6) or even lower (fluoroquinolones, FQs7). The ET of FQs has been assumed to be between 280 and 260 kJ/mol.8 However, while lomefloxacin (LFX), enoxacin (ENX), and norfloxacin (NFX) photosensitize TT lesions in DNA, ofloxacin (OFX) does not (structures are shown in Chart 1). It is reasonable to think that the ET of these molecules must be the determining factor. In addition to this type of damage, FQs may photosensitize DNA by other mechanisms, including type I/type II photooxidation and formation of highly reactive transient species.7,9-11 In fact, these drugs, which are increasingly being used as antibiotics to treat a broad range of bacterial infections, may be very efficient photosensitizers; some of them are responsible for dangerous cutaneous reactions, while others seem to be potential photomutagenic and photocarcinogenic agents.12 All FQs shown in Chart 1 share a central heterocyclic skeleton; the main difference is associated with the nature of X, which can be a carbon atom supporting different electrondonating or -withdrawing substituents, as well as a nitrogen atom. With this background, we performed a preliminary study to determine the minimum value of ET for a photosensitizer to produce TT in DNA.13 For this purpose, NFX and its acetylated derivative ANFX were selected; the former was
10.1021/jp070167f CCC: $37.00 © 2007 American Chemical Society Published on Web 05/25/2007
7410 J. Phys. Chem. B, Vol. 111, No. 25, 2007 CHART 1: Structure of Some Fluoroquinolone DNA Photosensitizers
chosen for convenience, as it is the most simple photosensitizing FQ derivative, and the latter because N(4′)-acetylation is known to result in a slightly decreased singlet energy, attributed to geometry changes in the excited state associated with N(1′) rehybridization with intramolecular charge transfer.14 Experiments carried out on supercoiled circular DNA (pBR322) showed that only NFX was able to generate TT lesions by using photosensitizer concentrations and light doses insufficient to produce direct single-strand breaks. In this paper, we report in more detail our studies on the determination of the minimum value of ET for a photosensitizer to produce TT in DNA and to show how this information can be used to predict the relative potential of a series of analogous drugs to behave as DNA photosensitizers via TT formation. For this purpose, a photophysical and photobiochemical investigation on all FQs shown in Chart 1 has been undertaken; LFX has been excluded in this study, as this compound undergoes photodehalogenation from its singlet excited state with a high efficiency and thereby exhibits reduced intersystem crossing quantum yield.8,10 Experimental Section Materials. 4-Biphenylcarboxylic acid (BPC), enoxacin (ENX), norfloxacin (NFX), ofloxacin (OFX), naproxen (NP), flurbiprofen (FBP), trifluoroacetic acid, CDCl3, and CD3OD were obtained from Sigma Chemical Company (St. Louis, MO). Pefloxacin (PFX) was extracted from AZUBEN, produced by Ipsen Pharma laboratories (Barcelona, Spain). Rufloxacin (RFX) hydrochloride was from Bracco (Milan, Italy). Supercoiled circular pBR322 DNA was obtained from Roche Diagnostics (Barcelona, Spain). The DNA repair enzyme T4 endonuclease V (endo V) and its buffer were purchased from Trevigen (Gaithersburg, MD). Other chemicals were of reagent grade and used as received. The N-acetyl derivative of NFX (ANFX) was prepared from a solution of NFX (150 mg, 0.47 mmol), in Ac2O (50 mL), which was refluxed for 7 h.15 The solution was cooled to room temperature and concentrated. Afterward, the residue was dissolved in water, neutralized to pH ∼ 7.4, extracted with CH2Cl2, concentrated under vacuum, and purified by silica-gel chromatography, eluting with a CH2Cl2/MeOH solution (85/ 15, v/v). The sodium phosphate buffer (PB, 1 mM) was prepared from reagent-grade products using deionized water; the pH of the solutions was measured through a glass electrode and adjusted with NaOH to be pH 7.4. Laser Flash Photolysis Measurements. A pulsed Nd:YAG SL404G-10 Spectron Laser Systems was used at the excitation wavelength of 355 nm. The single pulses were of ∼10 ns duration and the energy was lower than 5 mJ/pulse. The detecting light source was a pulsed Lo255 Oriel xenon lamp. The laser flash photolysis system consisted of the pulsed laser, the Xe lamp, a 77200 Oriel monochromator, and an Oriel photomultiplier tube (PMT) system made up of a 77348 sideon PMT tube, 70680 PMT housing, and a 70705 PMT power
Lhiaubet-Vallet et al. supply. The oscilloscope was a TDS-640A Tektronix. The output signal from the oscilloscope was transferred to a personal computer. Unless otherwise stated, all samples of ENX, PFX, NFX, ANFX, OFX, and RFX used for laser flash photolysis were aqueous solutions at pH 7.4 in 1 mM PB, and the absorbance was set at 0.26 at 355 nm. Each sample was deaerated by bubbling nitrogen (when specified) or N2O. Determination of the Rate Constants for Quenching of Triplet Excited State of FQs (kq) by FBP, BPC, and NP. The experiments were carried out using solutions of each FQ (ENX, PFX, NFX, ANFX, OFX, and RFX) with increasing amounts of the quenchers (FBP, BPC, and NP, 0.1-10 mM) ensuring that no changes in the pH were induced. Under these conditions more than 99% of the light was absorbed by the fluoroquinolone. Determination of the Molar Absorption Coefficient of Triplet Excited State (T). The values for ENX, PFX, NFX, and ANFX were determined by the energy transfer method using a 5 mM concentration of NP in a N2O deaerated medium (under these conditions NP is able to quench more that 95% of the 3FQs). Thus, (3FQ(λmax nm)) was calculated using eq 1
∆A(3NP(440 nm)) × (3FQ(λmax nm)) ) ∆A(3FQ(λmax nm)) × (3NP(440 nm)) (1) where ∆A(3FQ(λmax nm)) refers to the transient absorbance of at its λmax at the beginning of the reaction and ∆A(3NP(440 nm)) to that of the NP triplet state at 440 nm at the end of the reaction. The molar absorption coefficient of 3NP in aqueous solutions was taken to be (3NP(440 nm)) ) 9500 M-1 cm-1.16 Determination of the Intersystem Crossing Quantum Yields (φISC). They were estimated using the comparative method with benzophenone (BPH) in acetonitrile as standard. Thus φISC was obtained by application of eq 217
3FQ
φISC(FQ) ) φISC(BPH) × ∆A(3FQ(λmax nm)) × (3BPH(520 nm))/∆A(3BPH(520 nm)) × (3FQ(λmax nm)) (2) where ∆A (3FQ(λmax nm)) refers to the transient absorbance of each fluoroquinolone triplet (FQ) at its λmax nm under N2O atmosphere, while ∆A (3BPH (520 nm)) refers to the absorbance of 3BPH at 520 nm. The BPH triplet molar absorption coefficient and quantum yield in acetonitrile were taken to be (3BPH(520 nm)) ) 6500 M-1 cm-1 and φISC(BPH) ) 1, respectively. Photosensitized Damage to DNA. All the DNA and drug stock solutions were prepared in water. Then, air-equilibrated mixtures containing pBR322 (20 µM in base pair) and ENX, PFX, NFX, ANFX, OFX, and RFX (20-200 µM) were irradiated using a multilamp photoreactor (Philips TL8W/08 BLB, from 350 to 400 nm, 1.6 mW/cm2 at 360 nm). In the case of direct single strand breaks detection, the loading buffer (0.25% bromophenol blue, 0.25% xylene cyanol, 15% ficoll 400 in water) was added immediately after irradiation. To reveal thymine cyclobutane dimers, the irradiated samples were incubated with an excess of DNA repair enzyme (37 °C, 1 h), and loading buffer was then added to each solution. The samples were loaded on a 0.8% agarose gel containing ethidium bromide. After electrophoresis, the relative abundance of supercoiled DNA (form I) and relaxed DNA (form II) was quantified by densitometry (Kodak Digital Science 1D V.3.0). Results and Discussion The aim of this study was to provide information about the triplet energies of a series of fluoroquinolones and to compare
Fluoroquinolones as Thymine Dimer Photosensitizers
J. Phys. Chem. B, Vol. 111, No. 25, 2007 7411
Figure 1. Transient absorption spectra obtained upon 355 nm excitation of NFX, PFX, and ANFX (0.1 mM) in N2O-purged buffered aqueous solutions 50 ns after the laser pulse.
TABLE 1: Photophysical Properties of the Triplet Excited States of Fluoroquinolones in 1 mM Phosphate Buffer Solution at pH ca.7.4 parameters
ENX
PFX
NFX
ANFX
OFX
RFX
τT (µs) λmax (nm) (M-1 cm-1)a ΦTd
0.8 520 6900 0.54
2.2 620 7800 0.34
3.5 620 7900 0.52
7.1 610 7800 0.40
4.3 610 5900b 0.32c
7.3 640 5800c 0.36c
a The molar absorption coefficients (Μ-1 cm-1) were calculated at λmax of each fluoroquinolone by the energy transfer method, using NP as standard. b Value from ref 19b. c Value from ref 11. d The triplet excited state quantum yield (ΦT) was determined by the comparative method, using benzophenone as standard.
them with the triplet energy of thymine in DNA, in order to assess whether energy transfer (and therefore TT dimer formation) is feasible. The FQs were chosen as sensitizers because some drugs of this family, as ENX and NFX, are known to generate TT dimers, while others (such as OFX) do not. Hence, the triplet excited state properties of the FQs were determined before performing specific assays for TT detection. Specifically, the selected FQs were NFX, its N-acetylated (ANFX) and N-methylated (PFX) derivatives, ENX, OFX, and rufloxacin (RFX) (Chart 1). Photophysical Properties of the Triplet Excited State of Fluoroquinolones. Time-resolved nanosecond laser flash photolysis experiments were performed at 355 nm using N2O bubbled PB solutions of FQs, in order to determine their triplet excited state properties. The laser energies were attenuated to
