Singlet oxygen, O(2)(a(1)Δ(g)), was produced upon pulsed-laser irradiation of an intracellular photosensitizer and detected by its 1275 nm O(2)(a(1)Δ(g)) → O(2)(X(3)Σ(g)(-)) phosphorescence in time-resolved experiments using (1) individual mammalian cells on the stage of a microscope and (2) suspensions of mammalian cells in a 1 cm cuvette. Data were recorded using hydrophilic and, independently, hydrophobic sensitizers. The microscope-based single cell results are consistent with a model in which the behavior of singlet oxygen reflects the environment in which it is produced; nevertheless, the data also indicate that a significant fraction of a given singlet oxygen population readily crosses barriers between phase-separated intracellular domains. The singlet oxygen phosphorescence signals reflect the effects of singlet-oxygen-mediated damage on cell components which, at the limit, mean that data were collected from dead cells and, in some cases, reflect contributions from both intracellular and extracellular populations of singlet oxygen. Despite the irradiation-induced changes in the environment to which singlet oxygen is exposed, the "inherent" intracellular lifetime of singlet oxygen does not appear to change appreciably as the cell progresses toward death. The results obtained from cell suspensions reflect key features that differentiate cell ensemble from single cell experiments (e.g., the ensemble experiment is more susceptible to the effects of sensitizer that has leaked out of the cell). Overall, the data clearly indicate that measuring the intracellular lifetime of singlet oxygen in a O(2)(a(1)Δ(g)) → O(2)(X(3)Σ(g)(-)) phosphorescence experiment is a challenging endeavor that involves working with a dynamic system that is perturbed during the measurement. The most important aspect of this study is that it establishes a useful framework through which future singlet oxygen data from cells can be interpreted.

Selected singlet oxygen photosensitizers have been examined from the perspective of obtaining a molecule that is sufficiently stable under conditions currently employed to study singlet oxygen behavior in single mammalian cells. Reasonable predictions about intracellular sensitizer stability can be made based on solution phase experiments that approximate the intracellular environment (e.g., solutions containing proteins). Nevertheless, attempts to construct a stable sensitizer based solely on the expected reactivity of a given functional group with singlet oxygen are generally not sufficient for experiments in cells; it is difficult to construct a suitable chromophore that is impervious to all of the secondary and/or competing degradative processes that are present in the intracellular environment. On the other hand, prospects are reasonably positive when one considers the use of a sensitizer encapsulated in a specific protein; the local environment of the chromophore is controlled, degradation as a consequence of bimolecular reactions can be mitigated, and genetic engineering can be used to localize the encapsulated sensitizer in a given cellular domain. Also, the option of directly exciting oxygen in sensitizer-free experiments provides a useful complementary tool. These latter systems bode well with respect to obtaining more accurate control of the "dose" of singlet oxygen used to perturb a cell; a parameter that currently limits mechanistic studies of singlet-oxygen-mediated cell signaling.

In this article, the laser fluence used to irradiate sensitizers in single cell experiments was erroneously given as 7 kJ/cm 2 . The correct laser fluence used in these experiments was 0.29 J/cm 2 .