Differential toxic effect of dissolved triamcinolone and its crystalline deposits on cultured human retinal pigment epithelium (ARPE19) cells

Abstract

The aim of the study was to evaluate the antiproliferative and cytotoxic properties of triamcinolone acetonide (TA) on human retinal pigment epithelium cells (ARPE19) and the role of epicellular crystalline deposits. Monolayer cultures of ARPE19 cells were used. Purified or unpurified crystalline TA suspension (0.01–1.0 mg/ml) or the vehicle alone (benzyl alcohol, 0.025%–0.00025%), diluted in culture medium, were added to the cells that were either grown on cell culture dishes covered by a protecting membrane filter insert or without a filter. After 1, 3, 5 and 7 days mitochondrial activity was measured using the MTT assay and the morphology assessed microscopically. Cellular proliferative activity was monitored by BrdU-incorporation into cellular DNA. For cytotoxicity assays ARPE19 cells were grown to confluence and then cultured in a serum-deficient medium to ensure a static milieu. Annexin V-FITC and propidium iodide co-staining was performed and analyzed by flow cytometry. Exposure to TA without direct cellular contact showed a moderate antiproliferative activity resulting in a dose-dependent suppression of DNA synthesis (maximum 42.7%), but not a cytotoxic effect. In contrast, adherent deposits of crystalline TA particles on top of the cell layer caused a rapid-progressive and dose-dependent cell death preceded by an early phosphatidylserine externalization to the outer leaflet of the plasma membrane. Within a healthy, confluent cell layer the number of viable cells decreased by 14.2, 20.8 and 68.8%, respectively, after one day of direct exposure. Exposure to the vehicle alone caused only a slight growth inhibitory effect in a proliferating cell layer, but early signs of cell death were detected even at the lowest concentration tested. In conclusion, the effect of the vehicle is less pronounced than formerly assumed, but not negligible, thus indicating a beneficial effect of purification. While non-adherent TA, if purified, appears to be safe in clinically used concentrations, direct physical contact with crystalline particles might cause a local, rapid-progressive cytotoxicity that involves the induction of the apoptotic cascade. Therefore, epiretinal deposits after intravitreal TA administration might be critical in terms of long-term biocompatibility.

Introduction

Triamcinolone acetonide (TA), a synthetic, lipophilic corticosteroid with low solubility in aqueous solution, has been used as a depot drug for decades. Its advantageous pharmacokinetic profile with rapid bioavailability and sustained release characteristics let Machemer to suggest the intravitreal use of crystalline TA to maintain therapeutic drug levels within the vitreous cavity (Machemer et al., 1979, Tano et al., 1980). Since then, intravitreal administration of crystalline TA suspension has become increasingly popular for the treatment of various intraocular disorders such as chronic uveitis (Antcliff et al., 2001), diabetic macular edema (Jonas et al., 2003, Martidis et al., 2002), and age-related macular degeneration (Danis et al., 2000, Jonas et al., 2005). While the clinical side effects such as cataractogenesis and glaucoma are well known (Jaissle et al., 2004), the underlying mode of action and a potential cytotoxicity have still not been conclusively defined. Although generally considered as safe, the biocompatibility of crystalline TA in direct contact with cells is controversial.

The safety of intravitreal TA administration has been supported by prior animal studies (Kivilcim et al., 2000, McCuen et al., 1981, Schindler et al., 1982) and human trials (Danis et al., 2000, Jonas et al., 2003). In contrast, other studies have found a marked ocular toxicity of some commercial TA preparations (Schlaegel and Wilson, 1974). Hida and co-workers investigated the effect of the preservatives and osmolarity of the vehicle and postulated that adverse effects are probably caused by the vehicles rather than the corticosteroid itself (Hida et al., 1986). Hence, several authors suggested the use of various purification techniques in order to remove the vehicle prior to injection (Hernaez-Ortega and Soto-Pedre, 2004, Jaissle et al., 2005, Jonas et al., 2001, Nishimura et al., 2003).

In a recent laboratory work, however, Yeung and colleagues comprehensively investigated the growth inhibitory effect of TA and its corresponding vehicle on ocular cell lines in clinically relevant concentrations. They found the impact of the vehicle to be less pronounced than formerly assumed. The inhibitory corticosteroid effect on cellular proliferation was shown to be by far more pronounced. They concluded that TA is generally cytotoxic and assumed an apoptotic cell death to be involved (Yeung et al., 2003, Yeung et al., 2004).

These results are highly contradictory to the hitherto good clinical experience and, if confirmed will have a strong impact on the clinical use of intravitreal TA. However, several fundamental ambiguities still remain to be clarified. Firstly, the pharmacological effect of TA within the vitreous cavity must be distinguished from a potential direct toxicity of sedimented crystalline particles. The latter are the equivalent to localized epiretinal deposits that are often seen in the lower periphery, mostly in vitrectomized eyes, and might be more critical in terms of biocompatibility. Secondly, the well-known and beneficial antiproliferative drug effect of corticosteroids is barely discriminable from irreversible cellular toxicity in a proliferating cell culture, if standard assays are used, although this is decisive for a conclusive biocompatibility assessment. Finally, the significance of apoptosis induction has to be further analyzed.

The discrepancy between the good clinical results and recent laboratory evidence for a potential cytotoxicity of TA is currently unsettled. In view of the wide-spread clinical application and the off-label use of intravitreal TA this ambiguity may warrant additional investigations for the safety and efficacy of TA, if to be used with confidence clinically. In this laboratory study, crystalline TA is investigated for its antiproliferative properties, its direct cell toxicity and the significance of epicellular deposits. Finally, the effect of the vehicle and the significance of purification are further evaluated.