Fluoxetine Inhibits K + Transport Pathways (K + Efflux, Na + -K + -2Cl - Cotransport, and Na + Pump) Underlying Volume Regulation in Corneal Endothelial Cells E. Hara 1 , P.S. Reinach 1,2,3 , Q. Wen 1 , P. Iserovich 1 , J. Fischbarg 1,2 1 Departments of Ophthalmology and 2 Physiology and Cellular Biophysics, College of Physicians & Surgeons, Columbia University, New York, NY 10032, USA 3 Department of Biological Sciences, SUNY, College of Optometry, New York, NY 10010, USA Abstract. We have studied regulatory volume responses of cultured bovine corneal endothelial cells (CBCEC) using light scattering. We assessed the contributions of fluoxetine (Prozac) and bumetanide-sensitive membrane ion transport pathways to such responses by determining K + efflux and influx. Cells swollen by a 20% hypo- osmotic solution underwent a regulatory volume de- crease (RVD) response, which after 6 min restored rela- tive cell volume by 98%. Fluoxetine inhibited RVD re- covery; 20 M by 26%, and 50 M totally. Fluoxetine had a triphasic effect on K + efflux; from 20 to 100 M it inhibited efflux 2-fold, whereas at higher concentrations the efflux first increased to 1.5-fold above the control value, and then decreased again. Cells shrunk by a 20% hyperosmotic solution underwent a regulatory volume increase (RVI) which also after 6 min restored the cell volume by 99%. Fluoxetine inhibited RVI; 20 M by 25%, and 50 M completely. Bumetanide (1 M) inhib- ited RVI by 43%. In a Cl - -free medium, fluoxetine (50– 500 M) progressively inhibited bumetanide-insensitive K + influx. The inhibitions of RVI and K + influx induced by fluoxetine 20 to 50 M were similar to those induced by 1 M bumetanide and by Cl - -free medium. A com- puter simulation suggests that fluoxetine can interact with the selectivity filter of K + channels. The data sug- gest that CBCEC can mediate RVD and RVI in part through increases in K + efflux and Na-K-2Cl cotransport (NKCC) activity. Interestingly, the data also suggest that fluoxetine at 20 to 50 M inhibits NKCC, and at 100–1000 M inhibits the Na + pump. One possible ex- planation for these findings is that fluoxetine could in- teract with K + -selective sites in K + channels, the NKC cotransporter and the Na + pump. Key words: Na + -K + -2Cl - cotransporter — Na + /K + ATPase — Prozac (fluoxetine) — Regulatory volume decrease (RVD) — Regulatory volume increase (RVI) — Corneal endothelium Introduction The cornea is composed of the epithelium, the underly- ing stroma, and the endothelium. The stromal ground substance can imbibe fluid, which would result in cor- neal swelling and loss of transparency. However, meta- bolically dependent mechanisms in the corneal limiting layers transport fluid out of the stroma and maintain normal vision. Fluid transport across the endothelium is secondary to ionic transport across the layer; such trans- port depends on endothelial Na + :K + ATPase activity (Trenberth & Mishima, 1968) and includes a bicarbonate secretory mechanism (Hodson & Miller, 1976). It has been hypothesized that fluid transport across epithelial layers may result from cyclic regulatory vol- ume responses (Fischbarg, 1997). During a cycle, polar- ized epithelial cells would first gain fluid across one side during RVI, and would then expel it across the opposite side during RVD. It is currently difficult to assess the validity of this hypothesis for the corneal endothelium. Volume regulation and its underlying ionic transport mechanisms in this layer have been identified but not yet extensively described (Fischbarg et al., 1993; Srinivas & Bonanno, 1997). In many other systems, membrane ion transport mechanisms underlying RVD include an increase in the loss of K + and/or Cl - which can occur through parallel K + and anion channels or via KCl symport activity (Hal- lows & Knauf, 1994). Likewise, mechanisms underlying RVI include the bumetanide-sensitive NKC cotrans- Correspondence to: J. Fischbarg J. Membrane Biol. 171, 75–85 (1999) The Journal of Membrane Biology © Springer-Verlag New York Inc. 1999