Functionalized fullerenes mediate photodynamic killing of cancer cells: Type I versus Type II photochemical mechanism

Photodynamic therapy (PDT) employs the combination of nontoxic photosensitizers (PS) and harmless visible light to generate reactive oxygen species (ROS) and kill cells. Most clinically studied PS are based on the tetrapyrrole structure of porphyrins, chlorines, and related molecules, but new nontetrapyrrole PS are being sought. Fullerenes are soccer-ball shaped molecules composed of 60 or 70 carbon atoms and have attracted interest in connection with the search for biomedical applications of nanotechnology. Fullerenes are biologically inert unless derivatized with functional groups, whereupon they become soluble and can act as PS. We have compared the photodynamic activity of six functionalized fullerenes with 1, 2, or 3 hydrophilic or 1, 2, or 3 cationic groups. The octanol-water partition coefficients were determined and the relative contributions of Type I photochemistry (photogeneration of superoxide in the presence of NADH) and Type II photochemistry (photogeneration of singlet oxygen) were studied by measurement of oxygen consumption, 1270-nm luminescence and EPR spin trapping of the superoxide product. We studied three mouse cancer cell lines: (J774, LLC, and CT26) incubated for 24 h with fullerenes and illuminated with white light. The order of effectiveness as PS was inversely proportional to the degree of substitution of the fullerene nucleus for both the neutral and the cationic series. The monopyrrolidinium fullerene was the most active PS against all cell lines and induced apoptosis 4-6 h after illumination. It produced diffuse intracellular fluorescence when dichlorodihydrofluorescein was added as an ROS probe, suggesting a Type I mechanism for phototoxicity. We conclude that certain functionalized fullerenes have potential as novel PDT agents and phototoxicity may be mediated both by superoxide and by singlet oxygen.