All experiments were performed on mouse pancreatic islet cells. By refining our islet isolation technique, we now routinely obtain 60-140 islets per mouse. The chemical integrity of the cells was maintained by recording in the """"""""perforated patch clamp"""""""" mode. This technique uses amphotericin B to form monovalent cation-selective pores through the cell membrane so that the intracellular content of such important soluble constituents as calcium and ATP are unaffected while good electrical continuity between the cell and the recording pipette is maintained. The disadvantage is that for each cell attempted 10 to 20 minutes are required for permeation to occur, and many minutes are required even to know if the cell's resting membrane potential is suitable for further recording. Because our previous work showed that the endocannabinoid 2-arachadonoylglycerol (2-AG) blocked high voltage activated (HVA) calcium currents in the R7T1 insulinoma cell line, and because the influx of calcium is the proximal trigger for insulin release, we recorded calcium currents in beta-cells with and without the presence of 2-AG. In confirmation of the work with the R7T1 cells, 10 uM 2-AG blocked about 60% of the HVA calcium current. The fundamental question in this project was to determine if the effects of 2-AG on ion channels of beta-cells would inhibit the cell membrane depolarization that triggers insulin secretion. Our work up to this point made opposing predictions: The block of HVA calcium currents would predict that 2-AG would decrease cell depolarization, but the block of K(ATP) channels that we discovered last year would predict that by 2-AG would enhance depolarization. We showed that both of these blockades arose independently of cannabinoid receptors. To determine which ion channel effect predominates, we tested 2-AG under both current clamp and voltage clamp conditions in the mouse beta-cells. The first experiments were done in current clamp mode, whereby the recording pipette is a passive sensor of the cell's membrane potential. A proper beta-cell will depolarize when the extracellular glucose concentration is raised to 10 mM (from the low level of 2 mM used to maintain living cells) due to the increased ratio of ATP to ADP that blocks the K(ATP) channel. A beta-cell will also depolarize in the presence of a sulfonylurea antidiabetic drug, such as tolbutamide. These drugs depolarize the cell by blocking the K(ATP) channel through its SUR1 subunit. These checks were used routinely in subsequent experiments. It was immediately clear that 2-AG (10 uM) produced a small depolarization (about 5 mV) of the cells, but its effect on the glucose-induced depolarization was confounded by the observation that serial tests of high glucose (10 mM) alone gave diminished depolarizations. However, by observing that responses to high glucose in the presence of 2-AG (10 uM) were smaller than a subsequent test by glucose alone, we concluded that, 2-AG blocked the glucose-induced depolarization. To investigate which ionic currents were opposing the glucose-induced depolarization, subsequent experiments were carried out under voltage clamp mode using voltage ramps (-100 to 100 mV). Two quantities were adopted to characterize the voltage ramp experiments, namely the slope conductance between -100 and -70 mV and the null potential (the voltage at which no pipette current flows). The former measures the aggregate conductance of potassium channels, of which the K(ATP) is the predominant one at this voltage range, and the latter indicates the cell's membrane potential and consequently forecasts the activity of calcium currents and insulin secretion. In the typical experiment, a cell was depolarized with glucose (10 mM), then treated with 2-AG in the continued presence of glucose, then glucose alone, and finally washed. After glucose depolarized the cell, in 6 out of 13 cells, the addition of 2-AG caused the cell to repolarize and the slope conductance to decrease (in the other 7 cells 2-AG had no effect or (once) caused a small further depolarization). Normally, glucose raises the intracellular ATP concentration, which blocks the K(ATP) channel and depolarizes the cell. This reversal of the glucose-induced block of the K(ATP) channel was contrary to our explorations in the R7T1 cell line, which showed only that 2-AG blocked, not activated, the K(ATP) channel. To characterize this block further, we found that the depolarization induced by the K(ATP) channel blocker tolbutamide was never reversed by 2-AG, indicating that something downstream from glucose metabolism was the agent of the reversal. Furthermore we found that the endocannabinoid receptor 1 (CB1R) antagonist AM251 fully protected the cell from the reversal effect of 2-AG, indicating that, contrary to the block of the three cationic currents we characterized in the R7T1 cells, this reversal was mediated by the CB1R. Finally, because the CB1R couples to the G alpha type i protein to diminish the concentration of the signaling molecule cAMP, we overwhelmed the cell with cAMP using the adenylylcyclase-activating drug forskolin and the phosphodiesterase inhibitor IBMX. Under this condition 2-AG no longer reversed the glucose-induced depolarization, prompting the conclusion that cAMP, acting either through protein kinase A or through the effector protein Epac, ultimately reactivates the K(ATP) channel. As an offshoot to this work, we tested delta-9-tetrahydrocannabinol (THC), the active principal of the most used illegal substance, marijuana. Under both current clamp and voltage ramp conditions we saw that 1 uM THC depolarized beta-cells almost as completely as 100 uM tolbutamide. Preliminary experiments using the CB1 antagonist AM251 indicate that the THC effect is independent of the CB1R.
| Spivak, Charles E; Kim, Wook; Liu, Qing-Rong et al. (2012) Blockade of ?-cell K(ATP) channels by the endocannabinoid, 2-arachidonoylglycerol. Biochem Biophys Res Commun 423:13-8 |
