Cytotoxicity of Indocyanine Green on Retinal Pigment Epithelium: from Side Effect to Discovery of a Transporter

• Main Authors: Jau-Der Ho, 何昭德
• Other Authors: Ray Jui-Fang Tsai
• Format: Others
• Language: zh-TW
• Published: 2003
• Online Access: http://ndltd.ncl.edu.tw/handle/86625847211216093805

博士 === 長庚大學 === 臨床醫學研究所 === 91 === ABSTRACT Since indocyanine green (ICG; an organic anion) was firstly used to stain internal limiting membrane (ILM) during macular hole surgery in 2000, there have been controversies about its toxicity to the intraocular cells (esp. retinal pigment epithelium; RPE). The aims of this study include (1) to evaluate the cytotoxic effects of ICG on cultured human RPE, (2) to develop approaches to reduce the cytotoxicity, (3) to explore the mechanisms of ICG uptake by RPE. Firstly, we set up an in vitro model of ICG-induced RPE cytotoxicity. We demonstrated the toxic effects of prolonged ICG exposure (3 hours) on RPE, even at a very low ICG concentration (0.01 mg/mL). Besides, the cell nuclei revealed characteristics of necrosis. The next, we established an in vitro model of ICG-induced photodynamic toxicity. After short term exposure (2 min) of 2.5 mg/mL ICG followed by 40 minutes of light illumination (4 × 104 lux; compared to the intensity of an endoillumination probe), RPE cells were apoptotic (i.e. nuclear condensation and fragmentation, p53 nuclear translocation). However, RPE in both models revealed phosphatidylserine exposure on the cell surface (a marker of apoptosis). Further study found that increasing ATP in the ICG-treated RPE cells with glucose switched the cell death pattern from necrosis to apoptosis, implying a determinant role of ATP between apoptosis and necrosis. We also found that replacement of Na+ in the solvent by other cations significantly reduced ICG-induced RPE cytotoxicity (including those caused by prolonged exposure and photodynamic toxicity). With spectrophotometry, we demonstrated it was caused by reduced ICG uptake in the absence of Na+. This result was constructive clinically because it would enhance the safety margin of intravitreal ICG usage. Through pharmacokinetic study, we found RPE-mediated ICG uptake was composed of two parts: Na+-dependent and Na+-independent. Both parts conformed to the Michaelis-Menten kinetics. That is, the ICG uptake was saturable with increasing substrate concentration, indicative of a transporter-mediated process. The Michaelis-Menten parameters were Km = 0.1774 mM, Vmax = 1.032 nmol/105 cells‧min (Na+-dependent part) and Km = 0.454 mM, Vmax = 1.545 nmol/105 cells‧min (Na+-independent part). According to our results, we deduced that there was Na+-dependent organic anion transporter(s) on RPE cells which was responsible for the Na+-dependent portion of ICG uptake. The organic anion transporters found to date on RPE cells (OATP2 and MRP1) would not account for our observation, since OATP2 was Na+-independent and MRP1 mediated the extrusion of organic anions from cells, rather than uptake into cells. To address this issue, we demonstrated the existence of a Na+-dependent organic anion transporter OAT3, on human and rat RPE by RT-PCR and immunohistochemistry. In summary, we demonstrated the ICG-induced RPE cytotoxicity, either after prolonged exposure or after short-term exposure followed by light illumination. In addition, intracellular ATP content was a determining factor for the mechanism of cell death. Removal of Na+ in the solvent reduced the cytotoxicity and photodynamic toxicity of ICG by reducing the ICG uptake by RPE. We also demonstrated the presence of an organic anion transporter (OAT3) on RPE cells. Such results may provide a new view in reducing clinical side effects of intraocular ICG use and the study of the transport function of RPE.