Roberts, Stephen K.

Cholangiocytes line the intrahepatic bile ducts and regulate salt and water secretion during bile formation, but the mechanism(s) regulating ductal water movement remains obscure. A water-selective channel, the aquaporin CHIP, was recently described in several epithelia, so we tested the hypothesis that osmotic water movement by cholangiocytes is mediated by CHIP. Isolated rodent cholangiocytes showed a rapid increase in volume in the presence of hypotonic extracellular buffers; the ratio of osmotic to diffusional permeability coefficients was > 10. The osmotically induced increase in cholangiocyte volume was inversely proportional to buffer osmolality, independent of temperature, and reversibly blocked by HgCl2. Also, the luminal area of isolated, enclosed bile duct units increased after exposure to hypotonic buffer and was reversibly inhibited by HgCl2. RNase protection assays, anti-CHIP immunoblots, and immunocytochemistry confirmed that CHIP transcript and protein were present in isolated cholangiocytes but not in hepatocytes. These results demonstrate that (i) isolated cholangiocytes and intact, polarized bile duct units manifest rapid, mercury-sensitive increases in cell size and luminal area, respectively, in response to osmotic gradients and (ii) isolated cholangiocytes express aquaporin CHIP at both the mRNA and the protein level. The data implicate aquaporin water channels in the transcellular movement of water across cholangiocytes lining intrahepatic bile ducts and provide a plausible molecular explanation for ductal water secretion.

While intrahepatic bile duct epithelial cells secrete bile through transport of ions and water, the physiological mechanisms regulating ductular bile secretion are obscure, in part because of the lack of suitable experimental models. We report here the successful micropuncture of the lumen of isolated intrahepatic bile ducts and direct measurements of ductular ion secretion. Intact, polarized bile duct units (BDUs) were isolated from livers of normal rats by enzymatic digestion and microdissection. BDUs were cultured and mounted on a microscope in bicarbonate-containing buffer, and the lumens were microinjected with 2',7'-bis(2-carboxyethyl)-5-(and -6)carboxyfluorescein (BCECF)-dextran. Lumenal pH was measured by ratio imaging of BCECF fluorescence using digitized video fluorescent microscopy. After 36 hr in culture, the ends of BDUs sealed, forming closed compartments. After lumenal microinjection of BCECF-dextran, fluorescence was stable at the pH-insensitive wavelength, indicating no dye leakage. Serial changes in pH of extralumenal buffers containing pH-gradient collapsing ionophores allowed us to establish reliable standard curves relating fluorescence ratio to lumenal pH (r = 0.99; P < 0.001). By this approach, the basal pH inside the lumen of BDUs was 7.87 +/- 0.08 units (n = 9), 0.47 unit higher (P < 0.001) than the bathing buffer pH. Addition of 100 microM forskolin increased (P = 0.02) the lumenal pH from 7.78 +/- 0.06 to 7.97 +/- 0.06 units (n = 5); the forskolin effect was completely abolished by incubation of BDUs in HCO3-/CO2-free buffer. Moreover, forskolin caused a 50-fold increase in cAMP levels in BDUs. The observations are consistent with cAMP-dependent, active lumenal HCO3- secretion by BDUs. Furthermore, they demonstrate the suitability of the BDU model for studying regulatory and mechanistic aspects of ductular bile secretion.