Trafficking of a Single Photosensitizing Molecule to Different Intracellular Organelles Demonstrates Effective Hydroxyl Radical-Mediated Photodynamic Therapy in the Endoplasmic Reticulum

Abstract

Photodynamic therapy (PDT) is often used in preclinical and clinical treatment regimens. Reactive oxygen species (ROS) generated by photosensitizers (PSs) upon exposure to light induce cell death via diverse mechanisms. PSs can exert therapeutic effects in different cellular organelles, although the efficacy of organelle-specific PDT has yet to be determined as most previous studies use different PSs in different organelles. Here, we explored how a single PS, chlorin e6 (Ce6), targeted to different organelles altered the effectiveness of PDT. Ce6 intrinsically localizes to the ER after 4 h of incubation. Modification of Ce6 via conjugation with an octapeptide (LS765), a monosubstituted triphenylphosphonium (TPP) derivative (LS897), or a disubstituted TPP derivative (LS909) altered the intrinsic localization. We determined that LS765 and LS9897 predominantly accumulated in the lysosomes, but LS909 trafficked equally to both the mitochondria and the lysosomes. Moreover, the conjugation altered the type of ROS produced by Ce6, increasing the ratio of hydrogen peroxide to hydroxyl radicals. Irradiation of identical concentrations of the PSs in solution with 650 nm, 0.84 mW/cm² light for 10 min showed that the TPP conjugates nearly doubled the hydrogen peroxide production from ∼0.2 μM for Ce6 and LS765 to ∼0.37 μM for LS897 and LS909. In contrast, Ce6 produced ∼1.5-fold higher hydroxyl radicals than its conjugates. To compare the effect of each PS on cell death, we normalized the intracellular concentration of each PS. Hydrogen peroxide-producing PSs are effective PDT agents in the lysosomes while the hydroxyl-generating PSs are very effective in the ER. Compared to the PSs that accumulated in the lysosomes, only the ER-targeted Ce6 exerted >50% cell death at either low light power or low intracellular concentration. By delineating the contributions of cellular organelles and types of ROS produced, our work suggests that targeting hydroxyl radical-producing PSs to the ER is an exciting strategy to improve the therapeutic outcome of PDT.

Figure 1.

Structures of each of four Ce6 based PSs.

Figure 2.

(a) Hydrogen peroxide or (b) hydroxyl radical production by PS irradiation with 650 nm, 0.84 mW/cm² light. Bar plot shows comparison at 10 min, the time used for in vivo PDT. n ≥ 3.

Figure 3.

(a) Subcellular localization experiments where each PS (red) was coincubated with an organelle stains (green). Composite images show colocalization (yellow) between the two dyes. (b) Normalized intensity of each dye along a line of interest (shown in white). (c) Pearson correlation coefficient between the red and green channel for each PS and organelle staining. n ≥ 3 images.

Figure 4.

4T1/luc cells were incubated with 1 or 10 μM of each PS for 4 h, wash, and irradiated with 650 nm: (a) 0.84 mW/cm² or (b) 3.25 mW/cm² light for 10 min. n ≥ 3.

Figure 5.

(a) Intracellular concentration of PS after 10 μM of PS was incubated with cells for 4 h. (b) Incubation of cells with different starting concentrations of respective PS, to normalize intracellular concentration of PS after incubating for 4 h. (c) Cell viability after PDT, when all PS had the same intracellular sconcentrations. Each experiments were completed in triplicate.