Photodynamic treatment of culture medium containing serum induces long-lasting toxicity in vitro.

Photodynamic therapy (PDT) produces singlet oxygen and reactive oxygen species (ROS) that damage tumor cells and the vasculature. The resulting effect is a balance between photo-oxidations through primary or secondary ROS and scavenging activity. Sensitizers are distributed in the extracellular space before and during cell sensitization, suggesting that PDT could act directly on cell structures and on extracellular compartments, including sera. In this study we endeavored to determine whether the application of PDT to culture medium could affect cell survival. Culture medium [RPMI 1640 supplemented with fetal calf serum (FCS)] was incubated with Rose Bengal and irradiated before being added to cells for various contact times as a replacement for untreated medium. Cells were then kept in darkness until the survival assay. Treated medium reduced cell survival by up to 40% after 30 min of contact for 10 microg/ml of Rose Bengal and 20 J/cm(2). Rose Bengal or m-THPC alone or irradiated in water had no effect. This effect was dependent on the doses of Rose Bengal and light and decreased when FCS was replaced by human serum mixed with FCS. The reduction in survival observed with treated medium was more pronounced when the cell doubling time was shorter. Analysis of ROS or peroxide production in treated medium by DCFH added at the end of irradiation of Rose Bengal in serum-containing medium revealed a long-lasting oxidizing activity. Our findings support the hypothesis of an ROS- or peroxide-mediated, PDT-induced, long-lasting cell toxicity.

Secondary oxidants in human serum exposed to singlet oxygen: the influence of hemolysis.

Singlet oxygen (1O2) is produced by leucocytes during inflammatory reactions, various biochemical reactions and during photoreactions. It deactivates by reacting with a number of targets to produce reactive oxygen species (ROS) and peroxides (that in turn produce ROS). To verify whether serum had the same capability to deactivate secondary oxidants after exposure to 1O2, we provoked a photoreaction using rose bengal added to sera of 53 healthy donors and, after light delivery, reduced 2',7'-dichlorofluorescein (DCFH) was added at the end of irradiation and fluorescence of the oxidized derivative (DCF) was recorded. To avoid optical artifacts, we analyzed the influence of hemolysis. Deactivation capability of secondary oxidants after exposure to (1)O(2) was stable over a long period of time, slightly different between men and women, but standard biochemistry parameters had little influence. Hemolysis, age and platelet number reduced deactivation of 1O2-induced secondary oxidants. Addition of lysed cancer cells had no influence. Blood sampling in clot act tubes gave a better signal than in heparinized tubes. Red blood cells (RBCs) loaded with antioxidants strongly decreased deactivation of secondary oxidants. Assays are in progress to evaluate the clinical implications of these findings.

Secondary reactive oxygen species production after PDT during pulmonary tumor growth in sera of nude mice.

Photodynamic therapy (PDT), mediated by a sensitizer exposed to light to produce singlet oxygen ((1)O2), induces tumor responses varying from one person to another. Cancer growth induces oxidative stress at any step of its development from induction to treatment, which could also modify response to PDT. After the initial amount of (1)O2 delivered, secondary oxidative species (SOS) are also generated inducing additional damages. Using an in vitro assay we saw variations among mice strains concerning their serum capability to generate SOS after (1)O2 production. Nude mice had a higher capability to generate SOS as compared to the non mutated strain. Capability to generate SOS evolved during growth of orthotopically-grafted pulmonary cancers (A549), with either values corrected for hemolysis or not. Immediately after graft SOS production decreased, then increased again, reaching a plateau phase after 10 days which lasted for 20 days and finally increased steeply during the last phase of tumor growth, preceding cachexia and death. This profile differed profoundly from the one observed after heterotopic tumor grafts for which hemolysis induced artifacts masking important variations in SOS production. Our results demonstrate experimentally a relationship between the general health status of an individual, cancer progression and serum capability to generate SOS during PDT. These findings could explain some PDT failures as well as some unexpected successes on large tumors and should be taken into account when determining treatment parameters. They may also explain why different effects are observed on different experimental models with similar sensitizers.

Dynamics of oxidative stress and urinary excretion of melatonin and its metabolites during acute ischemic stroke.

Oxidative stress is a leading cause of neuronal damage in ischemic stroke. Melatonin may play a role in the antioxidant response. Melatonin and its metabolites may be involved in the modulation of oxidative stress in human acute stroke. No data are available in humans to establish this relationship. In this context, on the first and the fifth days post-stroke, we assessed serum total antioxidant capacity (TAC) and urine levels of melatonin, 6-sulfatoxymelatonin (aMT6S), and N1-acetyl-N2-formyl-5-methoxykynuramine (AFMK), the last compound being produced in the brain after reaction of melatonin with reactive oxygen species. Compared to controls' values, TAC and levels of melatonin and aMT6S were reduced, without difference between the first and the fifth days post-stroke, whereas AFMK levels remained in the normal range at both time points. Melatonin catabolism might be speeded up in acute ischemic stroke in order to increase the antioxidant response.

Influence of animal species on secondary oxidant activation in serum exposed to singlet oxygen.

Singlet oxygen ((1)O(2)) produced during inflammatory reactions and during photodynamic therapy deactivates by producing in tissues secondary reactive oxygen species and peroxides (SOS) as well as other degradation products. We investigated the influence of animal species on SOS production secondary to standardized (1)O(2) production by performing in vitro experiments with rose bengal as the (1)O(2) producer, human serum (HS) as a control, sera derived from various animal species, and dichloro-dihydro-fluorescein as a nonspecific marker that becomes fluorescent when oxidized. The overall SOS production in HS from a presumed healthy cohort of 53 donors (31 males and 22 females) gave a mean "normal" value of 0.91 compared with a previous pool of 75 male sera samples. SOS production after a photo-reaction was two or four times lower in HS than in fetal calf serum or mice sera, respectively. In mice, the "nude" characteristic increased even more than in the SOS production. In the Aves order, this production appeared to be distributed randomly according to the number of branches after the appearance of Amniotas. For primates, SOS production appeared to decrease linearly with the number of branches (R(2) = 0.98). Adding hemolysates from complete bloods to the corresponding sera induced an increase in SOS production in all species, proportional to the production in sera. These findings should be kept in mind when interpreting results from studies of secondary reactive oxygen species-induced pathways following (1)O(2) production, regardless of its origin.

Serum resistance to singlet oxygen in patients with diabetes mellitus in comparison to healthy donors.

Diabetes mellitus causes endothelial injury through oxidative stress involving reactive oxygen species and peroxides as well as inflammation, both of which consume antioxidant defenses. Singlet oxygen ((1)O(2)) is produced by leukocytes during inflammatory and biochemical reactions and deactivated by producing reactive oxygen species and peroxides. To determine whether serum was capable of deactivating (1)O(2), we triggered a photo reaction in sera from 53 healthy donors and 52 diabetic patients. Immediately after light delivery, dichlorofluorescein was added and then its fluorescence was recorded. The mean capacity of (1)O(2) or secondary oxidant deactivation was reduced in patients with diabetes mellitus. Hemolysis reduced deactivation of (1)O(2)-induced secondary oxidants in both healthy and diabetic patients. Body mass index, age, platelet counts, and blood cell numbers exerted a nonlinear influence. High levels of glycated hemoglobin were associated with an increased deactivation of oxidative species, whereas high-density lipoprotein cholesterol, total cholesterol, and the total cholesterol to high-density lipoprotein cholesterol ratio decreased the serum deactivation capacity. Oral antidiabetics bore no influence on deactivation, which was restored by insulin in women. Deactivation capacity was lower in women, who had half the complications found in men, suggesting that, with more severe diabetes mellitus, protection was maintained against complications. Resistance to (1)O(2) should be considered during the monitoring of diabetes mellitus.

Biophysical parameters influencing secondary oxidants activation in human serum exposed to singlet oxygen.

Singlet oxygen (¹O2), produced during photodynamic therapy, deactivates during its interaction with tissues by producing reactive oxygen species (ROS) and peroxides as well as other degradation products. Here we investigated the role of parameters of light delivery, O(2), and temperature on the ROS and peroxides production, secondary to ¹O2. A series of simple in vitro experiments has been performed with Rose Bengal (RB) as a ¹O2 producer, human serum (HS) as a target and dichlorofluorescein (DCFH) as a nonspecific marker, becoming fluorescent when oxidized. The overall secondary production of ROS and peroxides in HS had also been compared to fetal calf serum (FCS) or mice sera. Increasing power but with a same delivered energy decreased secondary ROS and peroxides when increasing power with a same duration for light delivery increased them. Increasing delivered energy increased linearly secondary ROS or peroxides. Delivering O2 by bubbling before light delivery increased secondary ROS or peroxides, when Ar decreased them. Delivering gases after light delivery had no influence on secondary ROS or peroxides production. Increasing temperature from 20 to 40 °C increased secondary ROS or peroxides production but freezing after light delivery either before or after measurement had only a mild influence. Secondary ROS or peroxides production was 2 or 4 times lower in HS than in FCS or nude mice sera respectively. PDT seems to consist of two subsequent phases, both linked but developing independently. The intensity of photo-reactions varied with the model, human sera producing less secondary ROS than fetal calf or mouse sera. Search for new sensitizers should consider secondary ROS-induced pathways in addition to ¹O2 production.

In vitro evaluation of Radachlorin sensitizer for photodynamic therapy.

This paper reports the evaluation of a new photosensitizer, Radachlorin in comparison with one of its well known components but used solely, Chlorin e(6). The photodynamic properties and cell uptake and localisation of the two drugs were compared. In vitro studies were conducted on human adenocarcinoma cells (HT29) and lung carcinoma cell line (A549). Both dyes showed an absorption maximum between 640 and 650 nm, but those absorption peaks are enhanced by interactions with serum, with a shifted maximum at 661 and 664 nm, and much higher absorbance. As Radachlorin is constituted of different products and as photoreactivity is dependent on absorbed light energy, we chose to adapt concentrations so that both drugs had the same absorption at the irradiation wavelength (664 nm) for photoreactivity tests, and express concentrations in optical density at 664 nm. The capacity of the two drugs to generate Reactive Oxygen Species was identical, but on HT29 cells, Radachlorin reaches its optimal LD50 sooner than Chlorin e(6). Radachlorin LD50 on HT29 cells was 0.0251 OD(664 nm) after 2 h and 0.0672 OD(664 nm) for Chlorin e(6) for a 20 J cm(-2) irradiation. Radachlorin gave very similar results on A549 cells, LD50 being 0.05 for 5 J irradiation, and 0.026 for 10 and 20 J cm(-2). Pharmacokinetics using fluorescence showed that, even if Radachlorin quickly crossed HT29 (a human colonic cancer line) cell membrane, cellular distribution evolved from a diffuse cytoplasmic repartition 1 hour after Radachlorin addition to a delimited localisation into organelles all around the nucleus. Radachlorin intracellular fluorescence decreased after 4 h, whereas we did not observe a decrease of Chlorin e(6) intracellular fluorescence for times up to 24 h. In both case, a quick decline was observed as soon as the culture medium was replaced with a drug-free one. Radachlorin appears to be an excellent photosensitizer, with similar phototoxicity to Chlorin e(6) on cell cultures, but with quicker kinetics, which could be an improvement if confirmed on further in vivo studies.

In vitro and in vivo evaluation of Radachlorin(R) sensitizer for photodynamic therapy.

This paper reports the evaluation of a new photosensitizer, Radachlorin in comparison with one of its well known components but used solely, Chlorin e6. The photodynamic properties, cell uptake and localisation of the 2 drugs were compared. In vitro studies were conducted on human adenocarcinoma cells (HT-29) and lung carcinoma cell line (A549). Both dyes showed an absorption maximum between 640 and 650 nm, that were enhanced by serum, with a shifted maximum at 661 nm. In vitro, phototoxicities of Radachlorin and Chlorin e6 were nearly identical for HT29 and A549 cells. However, Radachlorin reached its optimal LD50 sooner (0.59 microg ml(-1) for 3 h incubation followed by 20 J cm(-2) of 664 nm light (0.02 W cm(-2))) than Chlorin e6 (0.60 microg ml(-1) for 4 h incubation). For in vivo studies, Swiss athymic mice were grafted with human lung carcinoma of the line A549 15 days before intravenous photosensitizer injection. Fluorescence was recorded through an optical fibre spectrofluorimeter using the 666 nm peak for detection. Maximum Radachlorin fluorescence in tumor was observed 2 h after injection (1412 +/- 313 AU). Selectivity was expressed by the calculated tumor-to-skin and tumor-to-muscle ratios. Maximum ratios (1.45 +/- 0.14 for tumor-to-skin and 1.95 +/- 0.29 tumor-to-muscle) were observed 7 h after injection with Radachlorin. Maximal Chlorin e6 fluorescence was observed 1 h (shortest time interval measured) after injection in all organs and highest tumor-to-muscle ratio (2.56 +/- 0.97) 8 h after injection. Chlorin e6 fluorescence in skin was always at least equivalent to tumor fluorescence. Complete response of grafted tumor was achieved (no recurrence observed during 15 days) after 20 mg kg(-1) IV injection and 200 J cm(-2) irradiation (0.3 W cm(-2)) with both drugs. Optimal delays between injection and light delivery were between 1 and 7 h with Radachlorin and 3 h for Chlorin e6 but severe adverse effects were noted for both drugs when drug-light intervals were shorter than 3 h. This suggests that clinical use would be easier with Radachlorin than Chlorin e6.

Photoreactivity of indirubin derivatives.

Twenty-nine analogs of indirubin, an isomer of indigo, have been synthesized to optimize its promising kinase inhibitory scaffold. These compounds being also pigmented, have been tested for their photoreactivity. Absorption maxima were between 485 nm and 560 nm. Addition of fetal calf serum induced fluorescence and time dependent absorption modifications. Appropriate illumination induced Reactive Oxygen Species (ROS) production for nineteen compounds out of twenty-nine. The relationship between fluorescence and ROS production is discussed. Six compounds showed an important toxicity on F98 cells, a murine glioma cell line. Three of these were found to be also phototoxic, as four other non-toxic compounds. All but one phototoxic compounds were detected as ROS producers by in vitro tests. Photoreactivity assessment is important to anticipate adverse reactions for compounds that might be clinically developed. The experimental assay was found to be the only way to evaluate the photoreactivity of this family of compounds since no predictive criteria on structures could be found. Combining the vascular tumor growth inhibition induced by kinase inhibitors with the massive local blood flow arrest following photodynamic treatment may be an efficient anti-cancer strategy. These data could orientate further syntheses of either non-photoreactive compounds or compounds displaying both kinase inhibitory activity and strong phototoxicity.