B.M. Aveline and R.W. Redmond
Exclusive Free Radical Mechanisms of Cellular Photosensitization

1. Introduction

Over the last decade, there has been increasing interest in photosensitization mechanisms in biological systems, in relation to both deleterious and therapeutic aspects of this phenomenon [1,2] . Photosensitization reactions are generally considered as belonging to either the type I (radical mediated) or type II (singlet oxygen mediated) definitionÊ [3] . Most photosensitizers used in photodynamic therapy (PDT) (e.g., porphyrins, chlorins or phthalocyanines) display significant production of the long-lived triplet excited state, which generally undergoes facile energy transfer to molecular oxygen, leading to the formation of singlet oxygen. For a type I reaction to take place, a radical-generating reaction must compete with energy transfer to oxygen. However, even in cases where type I reactions may occur for these compounds, the type II reaction will usually take place in tandem and it is difficult to differentiate the photobiological effects which are exclusively due to radical species.

Although experimentally one is still confronted with the problem of how to specifically generate free radicals, the study of radical-induced damage in living systems has become a topic of great interest in biology and medicine [1,2]. Attention has been increasingly focused on the role of free radicals in normal physiological conditions (such as the so-called "respiratory burst" [4], important in the human host defense) as well as in various pathologies (e.g., cancer [5,6], inflammation [7], reperfusion injury [8], ocular damage [9,10], aging [11,12] or drug photosensitivity [13]). Radical reactions are also known to mediate biochemical phenomena at the cellular level such as lipid peroxidation [14], protein damage [15] and strand-breaks in DNA [16].

We have been working towards the development of photolytic radical generating molecules, to facilitate the study of pure radical effects in biological systems, where the free radical species are produced in a selective manner without complications arising from concomitant formation of long-lived triplet states via which type II photosensitization may occur. In this context, we have recently investigated the photochemistry of thiohydroxamic esters by 355 nm laser flash photolysis [17]. The primary photoprocess undergone by these molecules, upon pulsed excitation, is a homolytic nitrogen-oxygen bond cleavage which can be summarized by reaction (1).


N-O bond homolysis in these ester derivatives leads to simultaneous formation of the 2-pyridylthiyl radical (PyS) and RCOO, an acyloxy radical. In the case of compounds 2a, 2b and 2c (see Figure 1), decarboxylation occurs rapidly (<10 ns) to yield primary, secondary and tertiary carbon-centered radicals, respectively, in addition to pys. On the other hand, irradiation of the aroyl derivative 2d produces C₆H₅COO, an oxygen-centered radical, which decarboxylates with a much lower rate [18] (tau = 310 ns in acetonitrile [17]) and can therefore participate in subsequent reactions before decarboxylation takes place.

The parent compound, N-hydroxypyridine-2(1H)-thione, was also observed to undergo efficient homolytic N-O bond cleavage upon pulsed excitation. As shown by equation (2), this process gives rise to the formation of PyS and the hydroxyl radical, OH, one of the most powerful oxidant species known.

However, an extensive investigation of the photochemistry of compound 1 by laser flash photolysis demonstrated that this molecule cannot be considered as a clean photochemical source of hydroxyl radicals for the selective study of OH reactions in biological systems since it also undergoes other primary photoprocesses, which produce potentially toxic species [19,20].

In this work, photolytic, selective (compounds 2a-2d) and non-selective (compound 1) radical generators have been applied to cellular systems (murine L1210 lymphocytic leukemia cells). Our main goal was to investigate the effects of unambiguous radical photosensitization in biological systems. In order to provide some insight into their phototoxicity reaction mechanisms, thiopyridones have also been studied for their ability to initiate lipid peroxidation and to induce apoptosis. This report provides evidence that exposure of murine leukemia cells to UVA light (355 nm) in the presence of thiopyridone derivatives, triggers biological damage, which can be exclusively attributed to free radicals and constitutes a first step toward the comparison of the relative bioreactivities of different radical types (sulfur-, carbon- and oxygen-centered radicals) and singlet oxygen in cells.