Mouse hepatocytes in primary monolayer culture (4 hr) were exposed for 10 min at 37°C to anisosmotic medium of altered NaCl concentration. Hepatocytes maintained constant relative cell volume (experimental volume/control volume) as a function of external medium relative osmolality (control mOsm/experimental mOsm), ranging from 0.8 to 1.5. In contrast, the relative cell volume fit a predicted Boyle-Van't Hoff plot when the experiment was done at 4°C. Mouse liver slices were used for electrophysiologic studies, in which hepatocyte transmembrane potential (Vm) and intracellular K+ activity (aKi) were recorded continuously by open-tip and liquid ion-exchanger ion-sensitive glass microelectrodes, respectively. Liver slices were superfused with control and then with anisosmotic medium of altered NaCl concentration. Vm increased (hyperpolarized) with hypoosmotic medium and decreased (depolarized) with hyperosmotic medium, and ln [10(experimental Vm/control Vm)] was a linear function of relative osmolality (control mOsm/experimental mOsm) in the range 0.8-1.5. The aKi did not change when medium osmolality was decreased 40-70 mOsm from control of 280 mOsm. Similar hypoosmotic stress in the presence of either 60 mm K+ or 1 mm quinine HCl or at 27°C resulted in no change in Vm compared with a 20-mV increase in Vm without the added agents or at 37°C. We conclude that mouse hepatocytes maintain their volume and aKi in response to anisosmotic medium; however, Vm behaves as an osmometer under these conditions. Also, increases in Vm by hypoosmotic stress were abolished by conditions or agents that inhibit K+ conductance.

Hepatocyte transmembrane potential (Vm) behaves as an osmometer and varies with changes in extracellular osmotic pressure created by altering the NaCl concentration in the external medium (Howard, L.D. and Wondergem, R. (1987) J. Membr. Biol. 100, 53). We now have demonstrated similar effects on Vm by increasing external osmolality with added sucrose and not altering ionic strength. We also have demonstrated that hyperosmotic stress-induced depolarization of Vm results from changes in membrane K+ conductance, gK, rather than from changes in the K+ equilibrium potential. Vm and aki of hepatocytes in liver slices were measured by conventional and ion-sensitive microelectrodes, respectively. Cell water vols. were estimated by differences in wet and dry weights of liver slices after 10-min incubations. Effect of hyperosmotic medium on membrane transference number for K+, tk, was measured by effects on Vm of step-changes in external [K+]. Hepatocyte Vm decreased 34, 52 and 54% when tissue was superfused with medium made hyperosmotic with added sucrose (50, 100 and 150 mM). Correspondingly, aKi increased 10, 18 and 29% with this hyperosmotic stress of added sucrose. Tissue water of 2.92 ± 0.10 kg H2O/kg dry weight in control solution decreased to 2.60 ± 0.05, 2.25 ± 0.06 and 2.22 ± 0.05 kg H2O/kg dry weight with additions to medium of 50, 100 and 150 mM sucrose, respectively. Adding 50 mM sucrose to medium decreased tK from 0.20 ± 0.01 to 0.05 ± 0.01. Depolarization by 50% with hyperosmotic stress (100 mM sucrose) also occurred in Cl-free medium where Cl- was substituted with gluconate. We conclude that hepatocytes shrink during hyperosmotic stress, and the aKi increases. The accompanying decrease in Vm is opposite to that expected by an increase in aKi, and at least in part results from a concomitant decrease in gK. Changes in membrane Cl- conductance most likely do not contribute to osmotic stress-induced depolarization, since equivalent decreases in Vm occurred with added sucrose in cells depleted of Cl- by superfusing tissue with Cl-free medium.