The swelling independent activation of chloride channels
The swelling-independent activation of chloride channels by glycerol is proved by our further experiments. When glycerol concentration in the isoosmotic solution was reduced to the micromolar scale or even lower concentration, a Cl− current was still induced. The HOAt were not swollen by these treatments. Instead, the cells were shrunk. These data suggest that there may be two pathways for glycerol to activate chloride channels: the cell swelling-dependent and/or -independent pathways. The opening of the channels allows the outflows of ions and the accompanying water, resulting in a cell volume decrease. It was shown in this study, that the AQP blocker CuCl2 and specific AQP-3 siRNA significantly reduced the glycerol-induced chloride currents. These may be caused by the blockage or reduction of the pathways for glycerol to rapidly enter the cells. However, attenuation of the interaction between AQPs and the chloride channels by these treatments (CuCl2 and AQP-3 siRNA) may be another explanation for the reduction in the glycerol-induced currents. Our results showed that after AQP-3 expression was knocked down, the chloride current induced by the 140mM glycerol isoosmotic solution (with an increase of 22% in cell volume) was lower than that induced by the 140mM glycerol hyperosmotic solution (with an increase of 8% in cell volume) in AQP-3-intact cells. Considering that AQP-3 down-regulation has no effect on ClC-3 chloride channel expression (Zhang et al., 2014), these results suggest that the interaction between AQP-3 and chloride channels play an important role in cell volume regulation and that AQP-3 may be a modulator for the activation of chloride channels. Our present study demonstrated that glycerol (14μM) in the extracellular isoosmotic solution could activate a cell swelling-independent Cl− current, which was decreased by CuCl2 and AQP-3 siRNA. A similar current could also directly induced by dialyzing glycerol (14μM) into the cells with the recording pipette solution. It is then speculated that AQP-3 may act as the pathway for glycerol to enter the cells and then to activate the chloride channels directly from the inside of the cells. However, when AQP-3 was knocked down by the siRNA, the Cl− current induced by the intracellular dialyzing of glycerol with the recording pipette was significantly reduced. This result indicates that the cell swelling-independent Cl− current cannot be directly induced by glycerol in its own. It needs the help from AQP-3 to activate the chloride channels. AQP-3 may form a protein complex with the chloride channel protein and work as a modulator of the chloride channel (Zhang et al., 2014). These data suggest that AQP-3 is essential for the opening of chloride channels. ClC-3B, a splice variant of human ClC-3, was discovered to possess a PDZ domain-binding C terminus, by which ClC-3B could interact with other PDZ proteins (Gentzsch et al., 2003). We are exploring the mechanism of their interactions by combining use of bioinformatics and SPRi (Surface Plasmon Resonance imaging); we have identified some amino acid residue segments in AQP-3 and ClC-3 proteins where they may interact (data no shown). These residues will be further studied to clarify how they interact. Glycerol is a small and simple molecule produced in the breakdown of glucose, proteins, pyruvate, triacylglycerols and other glycerolipid, as well as release from dietary fats. Its metabolism is involved in cancer development, not only via the glycerol phosphate shuttle, which provides essential energy to cell proliferation and development, but also via the synthesis of glycerophospholipids (Brisson et al., 2001, Verkman et al., 2008). Thus, further investigation is needed to clarify the mechanism by which glycerol activates chloride channels in metabolism. In conclusion, aquaporins and chloride channels are functionally-related integral membrane channel proteins. In nasopharyngeal carcinoma cells, the swelling-dependent or independent Cl− current activated by glycerol was modulated by AQP-3.