SEPARATION AND PHOTOTOXICITY IN VITRO OF SOME OF THE COMPONENTS OF HAEMATOPORPHYRIN DERIVATIVE

Abstract— The separation and fractionation of some of the components of haematoporphyrin derivative by chemical extraction, gel chromatography and membrane filtration are described. The composition of the various fractions and the purity of the individual porphyrins can be assessed by high performance liquid chromatography. Porphyrins containing vinyl side chains show an unusual ability to form relatively stable aggregates of apparently high molecular weight. The effect is most pronounced with protoporphyrin and is not observed with haematoporphyrin which has no vinyl side chains. These aggregates can be separated by gel chromatography or membrane filtration from non (or less) aggregated material. The aggregates show greater in vitro cytotoxicity than the other fractions.     

THE PHOTOTOXICITY OF PHENYLHEPTATRIYNE: OXYGEN-DEPENDENT HEMOLYSIS OF HUMAN ERYTHROCYTES AND INACTIVATION OF Escherichia coli

Abstract— Phenylheptatriyne (PHT) plus near-ultraviolet light(320–400 nm; NUV) hemolyzed human erythrocytes in an oxygen dependent manner. When the phototoxicity of PHT plus NUV was tested with a series of Escherichia coli strains carrying all four possible combinations of genes controlling excision proficiency ( uvrA6 vs uvrA ⁺) and catalase activity (HPII, katF vs katF *), the membrane was found to be an important lethal target. Consistent with this observation. PHT plus NUV did not induce histidine independent ( his-4 ⁺) mutations in the four tester strains (RT7h-RT10h). Using tester strain RT10h, it was shown that there was no inactivation by PHT plus NUV in nitrogen. Results of experiments with an E. coli fatty acid auxotroph (K1060) treated with PHT plus NUV are also consistent with membrane proteins being the chief targets for attack. Radicals were formed during the photolysis of PHT plus NUV in aqueous solutions, both in the presence of air and under nitrogen. Since PHT plus NUV did not hemolyze erythrocytes or inactivate E. coli cells under nitrogen, these radicals are not cytotoxic.     

GLUCOSE ADMINISTRATION AUGMENTS IN VIVO UPTAKE AND PHOTOTOXICITY OF THE TUMOR-LOCALIZING FRACTION OF HEMATOPORPHYRIN DERIVATIVE

Rhabdomyosarcoma tumors in rats made hyperglycemic by multiple injections of glucose exhibited a transient decrease in pH and an increased ability to accumulate derivatives of hematoporphyrin (HPD). Photoradiation of tumors in glucose/HPD-treated animals produced a greater cell kill than in galactose/HPD-treated controls. A therapeutic benefit of glucose administration in conjunction with HPD-phototherapy is suggested.     

SINGLET OXYGEN YIELDS, OPTICAL PROPERTIES, AND PHOTOTOXICITY OF REDUCED DERIVATIVES OF THE PHOTOSENSITIZER CERCOSPORIN

Abstract— The perylenequinone cercosporin (CR) is a singlet oxygen generating photosensitizer produced by Cercospora spp which plays a critical role in parasitism of plants by these fungi. Several lines of evidence suggest that the defense mechanism of Cercospora spp towards this toxin is the generation of a cell surface reducing environment that leads to transient reduction of CR. In order to demonstrate that reduced CR is less toxic, several derivatives of CR were synthesized. Hexaacetyl-dihydrocercos-porin (HAC) was prepared by reductive acetylation of CR. Noranhydrocercosporin (NAC) resulted from dehydration of CR and tetramethyl-noranhydrodihydrocercosporin (TMNAC) was a product of reductive methylation of NAC. The perylenequinones, CR and NAC, absorb more light than their respective reduced derivatives, but are much less fluorescent; the relative fluorescence intensities of HAC, TMNAC, and dithionite-reduced CR were 80–160 times greater than that of CR and NAC. Also, CR and NAC were more efficient at generating singlet oxygen. As measured by time-resolved IR luminescence, the singlet oxygen quantum yields relative to CR (adjusted to 1.00) were 0.16, 0.19 and 0.73, respectively, for HAC, TMNAC, and NAC. Toxicity was measured by assaying for inhibition of growth of CR-sensitive fungi in constant light. The reduced derivatives were less toxic than their respective oxidized forms. None of the compounds showed significant growth inhibition in the dark with any of the fungi, or when assayed in the light with the CR-resistant fungus Cercospora kikuchii. A lipid peroxidation assay with methyl linolenate also showed that HAC was less active than CR. Thus, reduction of CR leads to greater fluorescence intensities, lower production of singlet oxygen and lower phototoxicity. These data support the hypothesis of transient cercosporin reduction as a mechanism of defense against cercosporin toxicity.     

INVOLVEMENT OF DRUG-DERIVED PEROXIDES IN THE PHOTOTOXICITY OF NAPROXEN and TIAPROFENIC ACID

Photodegradation of naproxen and tiaprofenic acid in aqueous buffered solutions leads to decarboxylated products with ethyl, 1-hydroxyethyl and/or acetyl side chains. The photomixtures obtained in the presence of oxygen were clearly more toxic to cultured hepatocytes than those obtained under anaerobic conditions. This effect was more noticeable in the case of naproxen. Based on the composition of the oxygenated photomixtures and the relative toxicity of the different photoproducts, it is possible to account for most of the observed toxicity in the case of tiaprofenic acid but not in the case of naproxen. This is explained as a result of the presence of drug-derived peroxidic species in the photomixtures and their contribution to the observed toxicity. Peroxides were determined by the peroxidase-catalyzed oxidation of dichlorodihydrofluorescein to its fluorescent analog. The amount of peroxides present in naproxen photomixtures was much higher than in the case of tiaprofenic acid. A dose-dependent depletion of intracellular glutathione was observed when hepatocytes were incubated with peroxide-containing naproxen photomixtures. This effect was prevented by the addition of catalase or jV-acetylcysteine to the culture medium.     

OXYGEN DEPENDENCE OF HYPERICIN-INDUCED PHOTOTOXICITY TO EMT6 MOUSE MAMMARY CARCINOMA CELLS

Abstract
The photodynamic effect of hypericin on EMT6 mouse mammary carcinoma cells was investigated in vitro under aerobic and hypoxic conditions. Under aerobic conditions, hypericin-induced photocytotoxicity was dose dependent within a 1–50 μ M range. Under hypoxic conditions, cells were resistant to hypericin-induced phototoxicity. In the dark, no cytotoxicity was observed at any hypericin concentration tested either aerobically or hypoxically. Cellular accumulation of hypericin, examined by chemical extraction and spectroscopy, occurred under both hypoxic and aerobic conditions. Fluorescence photomicrographs of cells exposed to hypericin corroborate drug uptake in the plasma membrane and subcellular regions. Our results demonstrate that hypericin cytotoxicity to EMT6 mouse mammary carcinoma cells in vitro is both light and oxygen-dependent. These results suggest that EMT6 cell kill caused by photoactivated hypericin is mediated by an oxygen-dependent mechanism, rather than by a type I oxygen-independent mechanism.     

OXYGEN DEPENDENCE OF HYPERICIN-INDUCED PHOTOTOXICITY TO EMT6 MOUSE MAMMARY CARCINOMA CELLS

The photodynamic effect of hypericin on EMT6 mouse mammary carcinoma cells was investigated in vitro under aerobic and hypoxic conditions. Under aerobic conditions, hypericin-induced photocytotoxicity was dose dependent within a 1-50 microM range. Under hypoxic conditions, cells were resistant to hypericin-induced phototoxicity. In the dark, no cytotoxicity was observed at any hypericin concentration tested either aerobically or hypoxically. Cellular accumulation of hypericin, examined by chemical extraction and spectroscopy, occurred under both hypoxic and aerobic conditions. Fluorescence photomicrographs of cells exposed to hypericin corroborate drug uptake in the plasma membrane and subcellular regions. Our results demonstrate that hypericin cytotoxicity to EMT6 mouse mammary carcinoma cells in vitro is both light and oxygen-dependent. These results suggest that EMT6 cell kill caused by photoactivated hypericin is mediated by an oxygen-dependent mechanism, rather than by a type I oxygen-independent mechanism.     

COMPARISON OF THE PHOTOTOXICITY OF α-TERTHIENYL WITH THAT OF A SELENIUM AND OF AN OXYGEN ANALOGUE

Abstract— Two analogues of α-terthienyl, namely 2, 5di(2'-thienyl) selenophene and 2, 5di(2'-thienyl) furan have been prepared and their phototoxicities toward several microorganisms have been compared. Dose response studies with Escherichia coli indicate that α-terthienyl is more active than these analogues. α-Terthienyl was shown to be the most effective of the three compounds in the photoinactivation of yeast alcohol dehydrogenase. Diagnostic tests showed the participation of singlet oxygen in the photosensitization to different extents by these three thiophenes.     

MECHANISMS OF CITRAL PHOTOTOXICITY

Citral, a monoterpene aldehyde synthesized by several plant genera, has been reported to exhibit antimicrobial activity. For the first time, we report that critral exhibits UV-A (315-400 nm) light enhanced oxygen-dependent toxicity against a series of Escherichia coli strains differing in DNA repair and catalase proficiency. Those E. coli strains carrying a gene leading to catalase deficiency (katF) are particularly sensitized to inactivation by citral and UV-A treatment when compared to catalase proficient strains (katF+). Consistent with these in vivo observations, citral when treated with UV-A in vitro produces H2O2. When tested against Fusarium oxysporum and F. solani, fungal root pathogens of Citrus, enhanced toxicity by citral in the presence of UV-A was demonstrated, while dark toxicity was negligible. When the plasmid pBR322 was treated with citral in the presence of UV-A, a change in conformation from the covalently closed circular to the open circular and, ultimately, the linear form was observed. The change in plasmid conformation corresponded to a reduction in transforming activity. Holding plasmid DNA which had been treated with UV-A light in the presence of citral at 4 degrees C for 22 h in the dark resulted in continued degradation of the DNA and loss of transforming activity. Holding plasmid DNA treated with UV-A or citral alone under identical conditions had no detectable effect on either plasmid conformation or transforming activity.     

THE LIPID SOLUBILITY OF PORPHYRINS MODULATES THEIR PHOTOTOXICITY IN MEMBRANE MODELS

The photosensitizing activities of hematoporphyrin, Cu-hematoporphyrin, protoporphyrin. Zn-protoporphyrin, deuteroporphyrin and uroporphyrin are studied. The relative yields of ¹O₂ are measured in buffer (pH 8.0) and compared to the yields of Type II photodamages induced on cholesterol embedded in egg lecithin liposomes. Results show that the solubilization of the sensitizer in the lipid bilayer is a prerequisite for its photosensitizing activity at the membrane level. In this context, the partition coefficient represents an important parameter.     

PHOTOTOXICITY INVOLVING THE OCULAR LENS: in vivo AND in vitro STUDIES

Our laboratory has demonstrated the cataractogenic potential of UV radiation and several photosensitizing drugs in laboratory animals and in humans. We have utilized lens fluorescence measurements (which we have demonstrated to be a reliable marker for pre-cataractous and early cataractous changes), NMR pulse relaxation techniques, and our recently developed magnetic resonance imaging method to measure lens T2 values in the normal and UV exposed Degus lens (in vivo and in vitro) to detect pre-cataractous changes in the lens. These approaches will permit us to employ two parameters (increased non-tryptophan fluorescence and a decrease in T2 values) to monitor for such changes months before the lens opacities become manifest by conventional slit lamp examinations.     

INVOLVEMENT OF SINGLET OXYGEN IN THE PHOTOTOXICITY MECHANISM FOR A METABOLITE OF PIROXICAM

It has been previously shown that a metabolite of piroxicam but not piroxicam itself causes phototoxicity to cells in vitro after exposure to UVA (320–400 nm) radiation. The phototoxicity mechanism for this metabolite, 2-methyl-4-oxo-2H-l,2-benzothiazine-l,l-dioxide (Compound I), was investigated. In vitro phototoxicity to human mononuclear cells was assayed using 0.5 m M Compound I and UVA radiation. The UVA fluence required for phototoxicity of Compound I was lower by a factor of 2-3 in D₂O buffer compared to H₂O buffer. Superoxide dismutase and mannitol, which remove O₂^- and OH", respectively, do not decrease the phototoxicity. The photodecomposition of Compound I was inhibited by sodium azide, enhanced by human serum albumin and unaffected by mannitol. Stable photoproducts of Compound I were not toxic to the cells. The quantum yield of singlet oxygen based on its emission at 1270 nm was 0.19 and 0.35 for Compound I and s2 ± 10^-3 and 10^-2 for piroxicam in D₂O and C₆H₆, respectively. While the extremely low quantum yield for singlet oxygen from piroxicam appears to account for its lack of phototoxicity, the phototoxicity mechanism for its metabolite, Compound I, most likely does involve singlet oxygen.     

PHOTOPHARMACOLOGY OF THE TRANQUILIZER CHLORDIAZEPOXIDE IN RELATION TO ITS PHOTOTOXICITY

Abstract— This study demonstrates that photobiological effects by exogenous molecules are not per se restricted to the skin. As an example the photopharmacology of chlordiazepoxide (e.g. Librium^R) was studied. Rats being exposed to UV-A after administration of ¹⁴C₂ labeled chlordiazepoxide (CDZ) showed a marked change in the pattern of urinary metabolites; the quantity of metabolites without N₄-oxide function increased, whereas that of compounds with N₄-oxide function decreased. Covalent binding to organ tissue of especially liver and skin of the irradiated rats was found. This would mean that in the irradiated rats oxaziridines which may also react with tissue of the inner organs are formed from CDZ and its N₄-oxymetabolites.
Further, the combination of light and CDZ leads to a 30% decrease in weight of the liver, indicating liver damage. This is also supported by the percentage of conjugated urinary metabolites being 1.5–2.5 times lower than in the non-irradiated CDZ-treated rats, possibly caused by a decrease of enzymatic activity.
The administration of CDZ-oxaziridine to non-irradiated rats provoked the same effects, such as a decrease of the weight of the liver comparable to that of irradiated CDZ-treated rats. On the other hand the combination of light and treatment with N₄-desoxy CDZ did not have any effect, as expected.     

RETENTION AND PHOTOTOXICITY OF TETRA(4-SULFONATOPHENYL)PORPHINE IN CULTIVATED HUMAN CELLS. THE EFFECT OF FRACTIONATION OF LIGHT

Human cervix carcinoma cells of the line NHIK 3025 were incubated for 18 h with tetra(4-sulfonatophenyl)porphine (TPPS4) and further incubated for 1-29 h in sensitizer free medium before exposure to light. After 1 h in sensitizer free medium only a 20% further loss of TPPS4 was observed within the next 28 h. During the time in sensitizer free medium, each TPPS4 molecule became more efficient in sensitizing single cells to photoinactivation. This enhanced photosensitizing efficiency of TPPS4 correlated well with the enhanced fluorescence yield of TPPS4. In some experiments the cells were exposed to a light dose inactivating 10% of the cells after incubation for 1 h in sensitizer free medium and a second graded light dose given 4-28 h later. Exposure of the cells to the first light dose led to loss of 60% of TPPS4 from the cells. Despite the significant loss of sensitizer from the cells the fluorescence yield of TPPS4 from each cell was found to increase (e.g. by 100% 4 h after light exposure). The enhanced fluorescence yield of cell bound TPPS4 was followed by a 1.6-2.5-fold increase in sensitivity of each cell to second light dose. Thus, a small light dose increased the photosensitivity of TPPS4-loaded NHIK 3025 cells for several hours after the first light exposure. The advantageous effect of light fractionation was reduced by a significantly enhanced loss of sensitizer induced by the first light exposure. The optimal time between the two fractions of light seems to be 30-90 min.     

EFFECTS OF DEMETHYLCHLORTETRACYCLINE PHOTOTOXICITY ON THE STRUCTURE and FUNCTIONS OF RAT LIVER MITOCHONDRIA

The phototoxic effects of demethylchlortetracycline (DMCT) with UVA radiation on isolated rat liver mitochondria were studied. DMCT at concentration of 21.5 microM with 5.53 x 10(-2) J cm-3 of UVA was found to be a potent uncoupler of oxidative phosphorylation. DMCT alone also uncoupled mitochondria but at higher concentrations (105 microM). ATPase activity was remarkably induced in mitochondria exposed to DMCT (21.5 microM) plus UVA (120% of DNP-stimulated ATPase activity). Content of ATP in such mitochondria when measured after addition of ADP was much smaller than that in control mitochondria. Ultrastructurally, mitochondria treated either with DMCT (21.5 microM) or with UVA alone stayed in the condensed configuration. On the other hand, uncoupled mitochondria treated with DMCT plus UVA became swollen and were changed into the orthodox configuration.     

PHOTODEGRADATION AND in vitro PHOTOTOXICITY OF FENOFIBRATE, A PHOTOSENSITIZING ANTI-HYPEIUIPOPROTEINEMIC DRUG

The phototoxic anti-hyperlipoproteinemic drug fenofibrate was found to be photolabile under aerobic and anaerobic conditions. Irradiation under argon of a methanol solution of this drug produced the photoproducts isopropyl 4-(1-[4-chlorophenyl]-1,2-dihydroxy)ethylphenoxyisobutyrate, 1,2- bis (4-chlorophenyl)-1,2- bis (4-[isopro-poxycarbonylisopropoxy]phenyl)ethane-1,2-diol and 4-(4-chlorobenzoyl)phenol, while under oxygen the photoproducts were 4-chloroperbenzoic acid, methyl 4-chlorobenzoate, 4-chlorobenzoic acid and singlet oxygen, as evidenced by trapping with 2,5-dimethylfuran. These results can be rationalized through hydrogen abstraction by excited fenofibrate, to afford a free radical as key intermediate. Biologically active antioxidants such as glutathione and cysteine efficiently reduced 4-chloroperbenzoic acid to 4-chlorobenzoic acid. The involvement of an electron transfer mechanism is suggested by detection (UV-vis spectrophotometry) of the radical cation TMP⁺ during the oxidation of tetramethylphenylenediamine (TMP) with 4-chloroperbenzoic acid. Fenofibrate was phototoxic in vitro when examined by the photohemolysis test, both under oxygen and argon atmosphere, although the photohemolysis rate was markedly lower under anaerobic conditions. The photoproducts 4-(1-[4-chlorophenyl]-1,2-dihy-droxy)ethylphenoxyisobutyrate and 4-chloroperbenzoic acid induced hemolysis in the dark however, this effect was quantitatively less important than photohemolysis by fenofibrate. On the other hand, fenofibrate photosensitized peroxidation of linoleic acid, monitored by the UV detection of dienic hydroperoxides. Based on the inhibition of this process upon addition of butylated hydroxyanisole, a radical chain (type I) mechanism appears to operate. In summary, fenofibrate is phototoxic in vitro . This behavior can be explained through the involvement of free radicals, singlet oxygen and stable photoproducts.     

THE INTERACTION OF 8-AZIDOETHIDIUM AND tRNA

Abstract Optical spectroscopy and proton nuclear magnetic resonance spectroscopy have been used to investigate the non-covalent and covalent binding of 8-azidoethidium (8-azido-3-amino-5-ethyl-6-phe-nyl-phenanthridium) with transfer RNA (tRNA) molecules. The nuclear magnetic resonance measurements demonstrate that ethidium and ethidium monoazide bind to the same unique binding site on tRNA molecules. However, the optical studies demonstrate that the photochemical reaction between ethidium monoazide and tRNA generates at least three different photoproducts. One of the covalent photoproducts has fluorescence properties which mimic those of intercalated, non-covalently bound ethidium, but it is produced only in about 10% yield. The major photoproduct(s) is(are) virtually non-fluorescent. The ethidium monoazide photochemistry with tRNA molecules is obviously complex and this may complicate its use as a photolabel of polynucleotides.     

β-CARBOLINE ALKALOIDS: MECHANISMS OF PHOTOTOXICITY TO BACTERIA AND INSECTS

Abstract— Phototoxicity of five naturally occurring (3-carboline alkaloids was assayed against a series of Escherichia coli strains and against the insect Trichoplusia ni (Lepidoptera: Noctuidae). Rank order efficacy of the compounds was comparable in both organisms. Although the bacterial assay demonstrated oxygen dependence, the degree of phototoxicity did not correlate with the relative efficiency of in vitro singlet oxygen or hydrogen peroxide photoproduction by the alkaloids. A better correlation was observed between chromatographic migration distances (lipophilicity) and phototoxicity. Therefore, hydrophobic mechanisms in which the alkaloids diffuse with varying efficiencies into cells or into the vicinity of target molecules may be important modes of phototoxicity for these compounds.     

SULFONATED PHTHALIMIDOMETHYL ALUMINUM PHTHALOCYANINE: THE EFFECT OF HYDROPHOBIC SUBSTITUENTS ON THE in vitro PHOTOTOXICITY OF PHTHALOCYANINES

The photocytotoxicity of sulfonated phthalimidomethyl aluminum phthalocyanine, a more hydrophobic photosensitizer as compared to phthalocyanine substituted with sulfonate groups only, was investigated. Inclusion of 1-2 phthalimidomethyl groups into disulfonated aluminum phthalocyanine, resulted in increased partition coefficients between n-octanol and water, and a six-fold increase in both cellular uptake and photocytotoxicity towards Chinese hamster lung fibroblast cells (line V-79). Reducing the number of phthalimidomethyl groups, or increasing the degree of sulfonation, lead to a decrease in the partition coefficient, cellular uptake, and phototoxicity. The quantum yield of singlet oxygen was comparable for all dyes tested in this series, indicating that no significant change in this photophysical parameter resulted from phthalimidomethylation. These results suggest that the addition of 1-2 phthalimidomethyl groups to disulfonated aluminum phthalocyanine improves cellular uptake, but, as the relative efficiency of cell killing was not effected, the intracellular distribution on photosensitive molecules may not be modified.     

INVESTIGATIONS ON THE MECHANISM OF CHLORPROMAZINE PHOTOTOXICITY: EFFECTS ON LYSOSOMES OF CULTURED HUMAN FIBROBLASTS

Abstract The effect of chlorpromazine (CPZ) and UVA on lysosomes of cultured normal human fibroblasts has been investigated. Acid phosphatase (ACPase) activity in 12 000 g pellet of cells treated with CPZ (10 μg/m l ) and UVA (6 × 10⁴ J/m²) was found to be decreased as compared with non-treated, CPZ or UVA treated control cells. This decrease, however, was not accompanied by a concomitant increase in ACPase activity in the 12 000 g supernatant. The addition of Triton X-100 to cells pretreated with CPZ + UVA resulted in only a moderate increase in ACPase activity of the 12 000 g supernatant. ACPase activity of the cells incubated in media containing preirradiated CPZ was also found to he decreased. These results indicate that CPZ + UVA directly inactivate lysosomal enzymes, possibly without affecting the membrane.