Detection and prevention of ocular phototoxicity of ciprofloxacin and other fluoroquinolone antibiotics

Abstract

Fluoroquinolone (FLQ) drugs are a potent family of antibiotics used to treat infections including ocular infections. To determine if these antibiotics may be phototoxic to the eye, we exposed human lens epithelial cells to 0.125-1 mm FLQs (ciprofloxacin [Cipro], lomefloxacin [Lome], norfloxacin [Nor] and ofloxacin [Ofl]), the precursor quinolone nalidixic acid (Nalid) and UVA radiation (2.5 J cm(-2)). Based on fluorescence confocal microscopy, FLQs are diffused throughout the cytoplasm and preferentially located in the lysosomes of lens epithelial cells. Neither FLQ exposure alone nor UVA exposure alone reduced cell viability. However, with exposure to UVA radiation the FLQs studied (Cipro, Nor, Lome and Ofl) induced a phototoxic reaction that included necrosis, apoptosis, loss of cell viability as measured by MTS, and membrane damage as determined by the lactate dehydrogenase assay. Both Nalid and all FLQs studied (Cipro, Nor, Lome and Ofl) photopolymerized the lens protein alpha-crystallin. Phototoxic damage to lens epithelial cells and/or alpha-crystallin will lead to a loss of transparency of the human lens. However, if precautions are taken to filter all UV radiation from the eye while taking these antibiotics, eye damage may be prevented.

Figure 1

Chemical structures of the quinolone Nalid and the four fluoroquinolones studied.

Figure 2

UV-visible absorption spectra (AU, Absorbance Units) of FLQs/quinolone and emission spectra of UVA lamp (W m^-2 nm^-1) from four parallel fluorescent UVA lamps.

Figure 3

UV-visible absorption spectra (AU, Absorbance Units) of ciprofloxacin (10 μM) in aqueous solutions (pH 7.4) containing 0.0 (1), 0.1 (2), 0.2 (3), 0.3 (4), 0.4 (5) and 0.5 mg/mL (6) α-crystallin lens protein.

Figure 4

Photolysis of α-crystallin lens protein by (A) Nalid; (B) Nor; (C) Ofl ; (D) Cipro; (E) Lome, and (F) control experiments after UVA radiation. From left to right of each figure, UVA duration 0, 10, 20 min. (G) Plot of quantification (ratio of high-to-low molecular weight) for each gel. *p < 0.05 compared with control experiments.

Figure 5

Effect of FLQs/quinolone exposure on the metabolic activity of HLE B-3 cells in the dark (A) or irradiated with UVA (B) as a function of FLQs/quinolone concentration, as measured by the MTS assay. Results were presented as the mean ± SEM from three independent experiments in duplicate. *p < 0.05 compared with cells without FLQ treatment.

Figure 6

Effect of FLQs/quinolone exposure on the lactate dehydrogenase release of HLE B-3 cells in the dark (A) or irradiated with UVA (B) as a function of FLQs/quinolone concentration, as measured by the LDH assay. Results were presented as the mean ± SEM from three independent experiments in duplicate. * p < 0.05 compared with cells without FLQ treatment.

Figure 7

FLQ/quinolone-induced apoptotic and necrotic death in HLE B-3 cells with UVA radiation (A: Cipro, B: Nor, C: Lome, D: Ofl, E: Nalid, and F: Control). Cells were seeded in plastic petri dishes (60 cm²) and pretreated with different FLQs at 500 μM concentration in HBSS for 1 hour and then exposed to UVA (10 min). After radiation, the cells were incubated overnight in cell culture medium and then stained with annexin V-FITC and propidium iodide. Apoptotic and necrotic cell death were determined with flow cytometry. The lower left quadrant shows normal viable cells, the lower right and upper right quadrants show apoptotic cells, while the upper left quadrant shows necrotic cells. (G) Graphs illustrating apoptosis and necrosis induced by UVA with different FLQs/quinolone.

Figure 8

Visualization of intracellular fluorescence of HLE B-3 cells using filter sets specific for FLQs/quinolone (in green; excitation: 364 nm, emission: 385-545 nm), the MitoTracker or LysoTracker (in red; excitation: 543 nm, emission: 560-615 nm), and the corresponding superimposed images.