(Systemic) phototoxicity of drugs and other xenobiotics.

Xenobiotics extensively used in drugs, cosmetics, food and agricultural chemicals can produce adverse biological effects. These toxic effects are separated into classes, e.g. hepatotoxicity, genotoxicity and neurotoxicity. Skin allergy, part of immunotoxicity, is also a subdivision of toxicology. When light is an essential condition for toxicity, the xenobiotic is called phototoxic. Thus it fits into the logic of toxicology that photoallergic compounds are a subdivision of phototoxic compounds. Phototoxicons as a group do not differ from the group of phototherapeutics with regard to their eventual biological effects. The primary photoreactions, secondary molecular processes, biomolecules involved and cellular and tissue damage are similar. The difference between the two groups is in the appreciation of the photobiological effects: adverse vs. desired. The aim of research is to determine the part of the molecular structure which makes a given compound phototoxic. With that knowledge the structure of the phototoxicon can be changed. This can result in a derivative which still has the desired properties of the parent compound, but is no longer phototoxic. This aim can be reached by combining data from both in vitro and in vivo research. The variety and number of phototoxic compounds is large. This, together with the limited research effort devoted to this subject so far, means that for most phototoxic xenobiotics a relationship between structure and in vivo photoreactivity is not available. In this review, emphasis is placed on xenobiotics whose in vitro and in vivo photochemistry have been studied. Furthermore, possible phototoxic effects which do not concern the skin but involve inner organs (systemic effects) are considered. References in this review mostly concern investigations over the last 10 years. For older literature or for additional information, references to other reviews are given. Important groups of phototoxic xenobiotics not dealt with in this article were already sufficiently covered in the reviews referred to.     

Photobinding of 8-methoxypsoralen, 4,6,4'-trimethylangelicin and chlorpromazine to Wistar rat epidermal biomacromolecules in vivo.

Photoinduced binding of drugs to endogenous biomacromolecules may cause both toxic and therapeutic effects. For example, photobinding of certain phenothiazines to biomolecules possibly underlies their phototoxic and photoallergic potential, whereas photobinding of furocoumarins to epidermal DNA is held responsible for their advantageous effects in the photochemotherapy of psoriasis. Usually, the in vitro photobinding of drugs is investigated. However, under in vivo conditions, the metabolism and distribution of the drug and the light absorption by endogenous compounds will significantly affect the photobinding of drugs to biomolecules. Therefore, in the present study, the photobinding of 8-methoxypsoralen (8-MOP), 4,6,4'-trimethylangelicin (TMA) (two therapeutically used furocoumarins) and chlorpromazine (CPZ) (a member of the phenothiazines) was investigated in vivo. The compounds were applied topically on the shaven skin of Wistar rats; one group was exposed to UVA and the other was kept in a dimly lit environment. Immediately, and at certain time intervals after UVA exposure, members of the two groups were sacrificed. By separating epidermal lipids, DNA/RNA and proteins by a selective extraction method, irreversible binding of 8-MOP, TMA or CPZ to each of these biomacromolecules was determined. In contrast with in vitro experiments, photobinding of CPZ to epidermal DNA/RNA was not found in vivo; apparently the bioavailability in the nucleus is very low. Compared with TMA, 8-MOP was observed to bind more extensively to epidermal DNA/RNA (again in contrast with findings from in vitro experiments) and proteins, but less extensively to lipids. The rates of removal of photobound 8-MOP and TMA were comparable. Photobound CPZ was more slowly removed from epidermal proteins and lipids than the furocoumarins. The observed in vivo photobinding is discussed with respect to the UVA-induced (side) effects of these drugs.     

Identification of Phototoxic Polycyclic Aromatic Hydrocarbons in Sediments Through Sample Fractionation and QSAR Analysis

Abstract

The toxicity of certain polycyclic aromatic hydrocarbons (PAHs) can be greatly increased by simultaneous exposure of test organisms to ultraviolet (UV) wavelengths present in sunlight. This phenomenon, commonly termed photoinduced toxicity, had been evaluated extensively in laboratory settings where only one chemical of concern was present. However, more recent studies have demonstrated that complex mixtures of PAHs present, for example in sediments, also can cause phototoxicity to a variety of aquatic species when the samples are tested in simulated sunlight. Unfortunately, because these types of samples can contain thousands of substituted and unsubstituted PAHs it is difficult, if not impossible, to use conventional analytical techniques to identify those responsible for photoinduced toxicity. The objective of the present study was to link two powerful ecotoxicology tools, toxicity-based fractionation techniques and QSAR models, to identify phototoxic chemicals in a sediment contaminated with PAHs emanating from an oil refinery. Extensive chromatographic fractionation of pore water from the sediment, in conjunction with toxicity testing, yielded a simplified set of sample fractions containing 12 PAHs that were identified via mass spectroscopy. Evaluation of these compounds using a recently developed QSAR model revealed that, based upon their HOMO-LUMO gap energies, about half were capable of producing photoinduced toxicity. We further evaluated the phototoxic potential of the reduced set of PAHs by determining their propensity to bioaccumulate in test organisms, through calculation of octanol-water partition coefficients for the chemicals. These studies represent a novel linkage of sample fractionation methods with QSAR models for conducting an ecological risk assessment.     

Photobinding of ketoprofen in vitro and ex vivo

Ketoprofen (KP), a non-steroidal anti-inflammatory drug of the 2-aryl propionic class, has been shown to produce photoallergic side effects as well as cutaneous photosensitizing properties that induce other phototoxic effects. In the present study we investigated photobinding of ketoprofen to both human serum albumin (HSA), a model protein, and to ex vivo pig skin and its photodegradation. Results demonstrate that photoadduct formation and photodegradation progressively increased with irradiation time where they reach a maximum. Maximum photobinding to the viable layer of the epidermis was about 7-8% of the initial radiolabelled KP added, in the region of 15-30 min UV irradiation. These results were comparable to in vitro results that were seen with photobinding of KP to HSA; in this case, the quantity of covalently bound material was approximately 10% of the initial, after a maximum of 18 min irradiation. It was found by HPLC analysis that the KP decrease is accompanied by an increase of the corresponding photoproduct, decarboxylated ketoprofen (DKP). The yield of DKP reaches a maximum at around 15 min. DKP appears to play an important role in vitro and ex vivo, being the major photoproduct and responsible for the photobinding process. Using micro-autoradiographical techniques we investigated the penetration and distribution of ketoprofen in ex vivo pig skin in greater detail. It was apparent that percutaneous absorption was taking place and that most of the ketoprofen was predominately localised in fibroblasts in the papillary dermis. No other specific localisation within the skin architecture was identified. Although there were differences in the quantities of bound ketoprofen within the different layers of the skin, these levels did not appear to correlate with irradiation time.     

Photoreactivity of tiaprofenic acid and suprofen using pig skin as an ex vivo model

The skin is repeatedly exposed to solar ultraviolet radiation. Photoreaction of drugs in the body may result in phototoxic or photoallergic side effects. Non-steroidal anti-inflammatory drugs, such as tiaprofenic acid (TPA) and the closely related isomer suprofen (SUP) are frequently associated with photosensitive disorders; they may mediate photosensitised damage to lipids, proteins and nucleic acids. Using ex vivo pig skin as a model, we investigated the photodegradation of TPA and SUP, and photobinding of these drugs to protein by means of HPLC analysis and drug-directed antibodies. Both with keratinocytes, which were first isolated from the pig skin and thereafter exposed to UVA and with keratinocytes which were isolated from pig skin after the skin was UVA exposed, time-dependent photodegradation of TPA and SUP was found, beside photoadduct formation to protein. The results of this work show that: (a) TPA and SUP were photodecomposed with similar efficiency; major photoproducts detected were decarboxytiaprofenic acid (DTPA) and decarboxysuprofen (DSUP), respectively. (b) Both drugs form photoadducts, as concluded from recognition by drug-specific antibodies. Pig skin appears to be a good model for studying the skin photosensitising potential of drugs.     

Analyses of cutaneous fluoroquinolones photoreactivity using the integrated model for the differentiation of skin reactions

Currently available test models for the differentiation of photoallergic and photoirritant reactions are extremely time consuming and the protocols are very heterogeneous. In vitro tests are of proven value in predicting irritant or toxic effects, but these tests fail to predict chemical-induced allergic side effects. We developed test systems for this endpoint which is not easily detected by existing assays. In a previous publication we were able to discriminate between a contact sensitizer and a skin irritant with a combination of primary ear swelling analysis and cell counting of the ear-draining lymph nodes [Toxicol. Appl. Pharm. 153 (1998) 83; Arch. Toxicol. 73 (2000) 501]. This combination of tests was called the Integrated Model for the Differentiation of chemical-induced allergic and irritant Skin reactions (IMDS). In addition, it had been shown before that inclusion of UV irradiation in the local lymph node assay enables discrimination of photoallergic from photoirritant reactions after dermal application [Photodermatol. Photoimmunol. Photomed. 10 (1994) 57]. Because of the fact that fluoroquinolones are known to induce photoreactions after oral but not dermal treatment, the aim of the present study was to apply the IMDS for the fast and reliable differentiation of photoreactions due to fluoroquinolones after oral treatment. Enoxacin, lomefloxacin, ofloxacin, sparfloxacin and BAY y 3118 were tested in this system. We found a good correlation between the results of UV light-irradiated IMDS and a guinea pig model with the quinolones as far as photoirritancy was concerned. This holds true also for the photoallergic standard olaquindox and the photoirritant standard 8-methoxypsoralen. However, in contrast to the guinea pig assays the IMDS is fast and extremely predictive for the risk of both photosensitization and photoirritancy depending on the route of exposure. Thus, the UV light-irradiated IMDS turned out to be a good tool for the preclinical risk assessment procedure in terms of discriminating photoreactions. In addition, flow cytometric analyses were used to underline the fact that antigen-independent activation occurred after the induction of photoirritant reactions.     

PHOTOALLERGIC RESPONSES TO CHEMICALS

Abstract— –Photochemical and immunologic knowledge about photoallergy to chemicals is briefly summarized. Studies in in vitro systems have demonstrated that photoallergic compounds can covalently bond to proteins through a photochemical reaction. The immunologic nature of the photoallergic response is based mainly on clinical observations, induction of photoallergy in man and in guinea-pigs and on results of in vitro immunologic tests.
Studies of the photoreactions of the photoallergic compound, 3,3',4',5-tetrachlorosalicylanilide (TCSA), with proteins are discussed. TCSA noncovalently bonds to human serum albumin prior to irradiation. Prior interaction is essential for formation of a photoaddition product indicating that a short-lived reactive species derived from TCSA is involved in the photoaddition and limiting the number of skin proteins which can participate in antigen formation. By fragmentation of the TCSA-albumin photoadduct with CNBr, it was determined that TCSA can bond to at least three sites on the albumin molecule. TCSA also can sensitize the photooxidation of histidine in albumin.     

A CE Method for Measuring Phototoxicity in Vitro

This paper presents an experimental setup which employs capillary electrophoresis with electrochemical and UV detection to test phototoxicity of plant extracts and components in terms of oxygen consumption and generation of reactive oxygen species upon irradiation with visible light. The experimental setup was used to test the phototoxicity of different buckwheat extracts and individual plant derived substances. The buckwheat extracts showed differences in their phototoxic behavior which might be due to different phytochemical composition. Screening of individual components revealed that rutin and quercetin alone were not phototoxic, but quercetin in combination with hypericin and chlorophyll caused considerable oxygen consumption. It has been demonstrated that the apparatus is a valuable tool to screen in vitro potential phototoxic reactions of plant extracts and individual constituents.     

THE STEPWISE BIPHOTONIC PHOTOIONIZATION OF CHLORPROMAZINE AS SEEN BY LASER FLASH PHOTOLYSIS

It is generally accepted that both promazine (PZ) and chlorpromazine (CPZ) photionize monophotonically to their respective cation radicals and the corresponding hydrated electrons. It is also supposed that this photoinization has a role in the phototoxic effects of these drugs. However, using laser flash photolysis, we have observed that photoionization of CPZ during S1 excitation (lambda greater than 300 nm) is a stepwise biphotonic process. In the case of PZ our flash photolysis results are less clearcut, but are consistent with stepwise biphotonic photoionization for S1 excitation. We demonstrate, using computer simulation of the intramolecular kinetics, that the estimated triplet state lifetime of CPZ is sufficiently long (23 ns at room temperature) to account for the apparent monophotonic photoionization that has been observed by others at high light intensities and short pulse times. Our laser flash photolysis results also suggest that the photo-ionization mechanism of PZ and CPZ is wavelength-dependent. Both drugs exhibit apparent monophotonic photoionization when they are excited at 266 nm under conditions of laser pulse width and intensity similar to those at 355 nm. We suggest that photoionization is not an important mechanism in the observed phototoxic and photoallergic effects of PZ and CPZ in sunlight.     

AFLOQUALONE PHOTOSENSITIVITY. IMMUNOGENICITY OF AFLOQUALONE-PHOTOMODIFIED EPIDERMAL CELLS

Abstract Afloqualone (AQ) is an oral muscle relaxant and evokes ultraviolet A (UVA)-induced photosensitivity dermatitis as a side effect. Histologic observations of the skin eruption suggest that AQ photosensitivity is mediated not only by phototoxic but also by photoallergic mechanisms. To address the immunological mechanisms of AQ photosensitivity, we examined the immunogenicity of AQ-photomodified epidermal cells in mice. Afloqualone was covalently coupled with bovine serum albumin by irradiation with UVA but not UVB. Because of this ability of AQ to photobind to protein, murine epidermal cells were easily modified with AQ by exposure to UVA. Subcutaneous inoculation of AQ-photomodified epidermal cells successfully induced an antigen-specific delayed-type hypersensi-tivity in mice. These findings suggest that AQ-photoderivatized epidermal cells are highly immunogenic and that photomodification of epidermal cells with AQ is an initial event to evoke AQ photosensitivity dermatitis.     

Protecting the protectors: reducing the biological toxicity of UV sunscreens by zeolite encapsulation

The phototoxic and photoallergic effects of the once popular UV sunscreen p-aminobanzoic acid are related, in part, to its ability to sensitize the formation of singlet oxygen as well as other reactive oxygen species. In this work we demonstrate that the sunscreen-photoinduced inactivation of a model protein, horseradish peroxidase, is reduced by approximately a factor of three when the sunscreen is encaspsulated in zeolite sodium Y. These results provide evidence that using the technology of zeolite encapsulation to prepare a supramolecular sunscreen that minimizes the skin contact of active ingredients may reduce the adverse effects of "naked" sunscreens on biological systems. These radiation-induced effects, unfortunately, frequently accompany the desirable UV-screening role of these products. These results provide an important benchmark for the use of zeolite encapsulation as a means of improving the safety of UV sunscreens for topical application.     

A blocked diketo form of avobenzone: photostability, photosensitizing properties and triplet quenching by a triazine-derived UVB-filter

Novel sunscreens are required providing active protection in the UVA and UVB regions. On the other hand, there is an increasing concern about the photosafety of UV filters, as some of them are not sufficiently photostable. Avobenzone is one of the most frequently employed sunscreen ingredients, but it has been reported to partially decompose after irradiation. In the present work, photophysical and photochemical studies on a methylated avobenzone-derivative have shown that the diketo form is responsible for photodegradation. A transient absorption was observed at 380 nm after laser flash photolysis excitation at 308 nm. It was assigned to the triplet excited state of the diketo form, as inferred from quenching by oxygen and β-carotene. This transient also interacted with key building blocks of biomolecules by triplet–triplet energy transfer (in the case of thymidine) or electron transfer processes (for 2'-deoxyguanosine, tryptophan and tyrosine). Irradiation of the avobenzone derivative in the presence of a triazine UV-B filter (E-35852) diminished the undesirable effects of the compound by an efficient quenching of the triplet excited state. Thus, sunscreen formulations including triplet quenchers could provide effective protection from the potential phototoxic and photoallergic effects derived from poor photostability of avobenzone.     

In Vitro Evaluation of the Photoreactivity and Phototoxicity of Natural Polyphenol Antioxidants

Polyphenols are a large family of natural compounds widely used in cosmetic products due to their antioxidant and anti-inflammatory beneficial properties and their ability to prevent UV radiation-induced oxidative stress. Since these compounds present chromophores and are applied directly to the skin, they can react with sunlight and exert phototoxic effects. The available scientific information on the phototoxic potential of these natural compounds is scarce, and thus the aim of this study was to evaluate the photoreactivity and phototoxicity of five phenolic antioxidants with documented use in cosmetic products. A standard ROS assay was validated and applied to screen the photoreactivity of the natural phenolic antioxidants caffeic acid, ferulic acid, p-coumaric acid, 3,4-dihydroxyphenylacetic acid (DOPAC), and rutin. The phototoxicity potential was determined by using a human keratinocyte cell line (HaCaT), based on the 3T3 Neutral Red Uptake phototoxicity test. Although all studied phenolic antioxidants absorbed UV/Vis radiation in the range of 290 to 700 nm, only DOPAC was able to generate singlet oxygen. The generation of reactive oxygen species is an early-stage chemical reaction as part of the phototoxicity mechanism. Yet, none of the studied compounds decreased the viability of keratinocytes after irradiation, leading to the conclusion that they do not have phototoxic potential. The data obtained with this work suggests that these compounds are safe when incorporated in cosmetic products.     

Comparison of the DNA Damage Photoinduced by Fenofibrate and Ketoprofen, Two Phototoxic Drugs of Parent Structure

Fenofibrate and ketoprofen (KP) are two drugs of similar structure derived from that of benzophenone. Both are photoallergic and promote cross reactions in patients. However, the cutaneous photosensitizing properties of KP also include phototoxic effects and are more frequently mentioned. To account for this difference in their in vivo properties, their in vitro photosensitizing properties on DNA were compared. First, it was shown that under irradiation at 313 nm, fenofibric acid (FB), the main metabolite of fenofibrate, photosensitized DNA cleavage by a radical mechanism similar to that proposed for KP but with a 50 times lower efficiency. Furthermore, FB did not photosensitize the formation of pyrimidine dimers into DNA in contrast to KP, which did promote this type of DNA damage. Their difference in efficiency as DNA breakers was compared to their relative photochemical reactivity and the quantum yield of FB photolysis was found to be eightfold lower than that of KP. The reactivity of these drugs cannot explain alone the difference in their photosensitizing properties. Other factors such as the magnitude of the ionic character of the pho-todecarboxylation pathway of these benzophenone-like drugs are considered in the discussion.     

Thioridazine induces immediate and delayed erythema in photopatch test

Thioridazine is a phenothiazine derivative that has been used as an antipsychotic; it rarely causes photosensitization. However, we noticed that this drug induced an erythematous reaction in a photopatch test. Six volunteers were patch tested with various concentrations of thioridazine and irradiated with a range of UVA doses, and the time courses of the color of and blood flow to the test sites were monitored. The free-radical metabolites of thioridazine generated under UVA irradiation and its effects on ascorbate radical formation were examined with an electron paramagnetic resonance (EPR) spectrometer in vitro. As a result, immediate erythema developed during UVA irradiation in most subjects when 1% thioridazine was applied for 48 h and irradiation doses were higher than 4 J cm(-2). Another peak of erythematous reaction was observed 8-12 h after irradiation. The in vitro examination detected an apparent EPR signal, which appeared when 2 mM thioridazine in air-saturated phosphate buffer was irradiated with UVA, whereas this reaction was attenuated under anaerobic conditions. The EPR signal of the ascorbate radical was augmented under both aerobic and anaerobic conditions. Thioridazine-derived oxidants and/or thioridazine radicals generated during UVA irradiation seem to play an important role in this unique phototoxic reaction.     

In-silico screening of high production volume chemicals for mutagenicity using the MCASE QSAR expert system

Computational screening is suggested as a way to set priorities for further testing of high production volume (HPV) chemicals for mutagenicity and other toxic endpoints. Results are presented for batch screening of 2484 HPV chemicals to predict their mutagenicity in Salmonella typhimurium (Ames test). The chemicals were tested against 15 databases for Salmonella strains TA100, TA1535, TA1537, TA97 and TA98, both with metabolic activation (using rat liver and hamster liver S9 mix test) and without metabolic activation. Of the 2484 chemicals, 1868 are predicted to be completely nonmutagenic in all of the 15 data modules and 39 chemicals were found to contain structural fragments outside the knowledge of the expert system and therefore suggested for further evaluation. The remaining 616 chemicals were found to contain different biophores (structural alerts) believed to be linked to mutagenicity. The chemicals were ranked indescending order according to their predicted mutagenic potential and the first 100 chemicals with highest mutagenicity scores are presented. The screening result offers hope that rapid and inexpensive computational methods can aid in prioritizing the testing of HPV chemicals, save time and animals and help to avoid needless expense.     

Searching for an enhanced predictive tool for mutagenicity

The Multiple Computer Automated Structure Evaluation (MCASE) program was used to evaluate the mutagenic potential of organic compounds. The experimental Ames test mutagenic activities for 2513 chemicals were collected from various literature sources. All chemicals have experimental results in one or more Salmonella tester strains. A general mutagenicity data set and fifteen individual Salmonella test strain data sets were compiled. Analysis of the learning sets by the MCASE program resulted in the derivation of good correlations between chemical structure and mutagenic activity. Significant improvement was obtained as more data was added to the learning databases when compared with the results of our previous reports. Several biophores were identified as being responsible for the mutagenic activity of the majority of active chemicals in each individual mutagenicity module. It was shown that the multiple-database mutagenicity model showed a clear advantage over normally used single-database models. The expertise produced by this analysis can be used to predict the mutagenic potential of new compounds.     

Phototoxicity induced by nalidixic and oxolinic acids: decrease in cell survival of chick embryo fibroblasts and Hep-2 cells

The phototoxic effects of nalidixic and oxolinic acids were evaluated in two types of cultured cells: chick embryo fibroblast and Hep-2 (human laryngo carcinoma cell line). In order to evaluate the phototoxicity induced by nalidixic and oxolinic acids, both cell types were irradiated for 5 min in the presence of each drug. The results showed an inverse relationship between cell survival and the concentration of the drug added to the culture medium. The concentrations of nalidixic and oxolinic acids necessary to induce a phototoxic effect were in the range of therapeutic blood levels. Both chick embryo fibroblasts and Hep-2 cells were more sensitive to the phototoxic effect induced by nalidixic acid than oxolinic acid.     

SPECTROSCOPIC STUDIES OF CUTANEOUS PHOTOSENSITIZING AGENTS–VIII. A SPIN-TRAPPING STUDY OF LIGHT INDUCED FREE RADICALS FROM CHLORPROMAZINE and PROMAZINE

Abstract— The clinically important phenothiazine drugs, particularly chlorpromazine, often elicit phototoxic and photoallergic reactions. We have used the spin traps 2-methyl-2-nitrosopropane (MNP) and 5,5-dimethyl-pyrroline-N-oxide (DMPO) to define the radical photolysis pathways of chlorpromazine and promazine. In the absence of oxygen the dechlorination product of chlorpromazine is trapped by MNP. The reactivity of the dechlorination product is similar to that of the phenyl radical as shown by its ability to extract hydrogen atoms from donors. Our results suggest that the dechlorination product is sufficiently reactive to account for the observation that chlorpromazine is more phototoxic than its parent promazine. In the presence of oxygen both chlorpromazine and promazine form a superoxide-dismutase-insensitive oxygen-centered intermediate which, when trapped by DMPO, rapidly decays to DMPO-OOH and subsequently to DMPO-OH. In addition, chlorpromazine readily undergoes photoelectron ejection only when it is excited into the second excited singlet state (Δ < 280 nra). This previously unknown wavelength dependence of photoionization should be considered in establishing the mechanism of chlorpromazine photosensitization.     

In-Silico Screening of High Production Volume Chemicals for Mutagenicity using the mcase QSAR Expert System

Computational screening is suggested as a way to set priorities for further testing of high production volume (HPV) chemicals for mutagenicity and other toxic endpoints. Results are presented for batch screening of 2484 HPV chemicals to predict their mutagenicity in Salmonella typhimurium (Ames test). The chemicals were tested against 15 databases for Salmonella strains TA100, TA1535, TA1537, TA97 and TA98, both with metabolic activation (using rat liver and hamster liver S9 mix test) and without metabolic activation. Of the 2484 chemicals, 1868 are predicted to be completely nonmutagenic in all of the 15 data modules and 39 chemicals were found to contain structural fragments outside the knowledge of the expert system and therefore suggested for further evaluation. The remaining 616 chemicals were found to contain different biophores (structural alerts) believed to be linked to mutagenicity. The chemicals were ranked in descending order according to their predicted mutagenic potential and the first 100 chemicals with highest mutagenicity scores are presented. The screening result offers hope that rapid and inexpensive computational methods can aid in prioritizing the testing of HPV chemicals, save time and animals and help to avoid needless expense.