Chloride channels regulate HIT cell volume but cannot fully account for swelling-induced insulin secretion.

Insulin-secreting pancreatic islet beta-cells possess anion-permeable Cl- channels (I(Cl,islet)) that are swelling-activated, but the role of these channels in the cells is unclear. The Cl- channel blockers 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid (DIDS) and niflumic acid were evaluated for their ability to inhibit I(Cl,islet) in clonal beta-cells (HIT cells). Both drugs blocked the channel, but the blockade due to niflumic acid was less voltage-dependent than the blockade due to DIDS. HIT cell volume initially increased in hypotonic solution and was followed by a regulatory volume decrease (RVD). The addition of niflumic acid and, to a lesser extent, DIDS to the hypotonic solution potentiated swelling and blocked the RVD. In isotonic solution, niflumic acid produced swelling, suggesting that islet Cl- channels are activated under basal conditions. The channel blockers glyburide, gadolinium, or tetraethylammonium-Cl did not alter hypotonic-induced swelling or volume regulation. The Na/K/2Cl transport blocker furosemide produced cell shrinkage in isotonic solution and blocked cell swelling normally induced by hypotonic solution. Perifused HIT cells secreted insulin when challenged with hypotonic solutions. However, this could not be completely attributed to I(Cl,islet)-mediated depolarization, because secretion persisted even when Cl- channels were fully blocked. To test whether blocker-resistant secretion occurred via a distal pathway, distal secretion was isolated using 50 mmol/l potassium and diazoxide. Under these conditions, glucose-dependent secretion was blunted, but hypotonically induced secretion persisted, even with Cl- channel blockers present. These results suggest that beta-cell swelling stimulates insulin secretion primarily via a distal I(Cl,islet)-independent mechanism, as has been proposed for K(ATP)-independent glucose- and sulfonylurea-stimulated insulin secretion. Reverse transcriptase-polymerase chain reaction of HIT cell mRNA identified a CLC-3 transcript in HIT cells. In other systems, CLC-3 is believed to mediate swelling-induced outwardly rectifying Cl- channels. This suggests that the proximal effects of swelling to regulate cell volume may be mediated by CLC-3 or a closely related Cl- channel.

The phantom burster model for pancreatic beta-cells.

Pancreatic beta-cells exhibit bursting oscillations with a wide range of periods. Whereas periods in isolated cells are generally either a few seconds or a few minutes, in intact islets of Langerhans they are intermediate (10-60 s). We develop a mathematical model for beta-cell electrical activity capable of generating this wide range of bursting oscillations. Unlike previous models, bursting is driven by the interaction of two slow processes, one with a relatively small time constant (1-5 s) and the other with a much larger time constant (1-2 min). Bursting on the intermediate time scale is generated without need for a slow process having an intermediate time constant, hence phantom bursting. The model suggests that isolated cells exhibiting a fast pattern may nonetheless possess slower processes that can be brought out by injecting suitable exogenous currents. Guided by this, we devise an experimental protocol using the dynamic clamp technique that reliably elicits islet-like, medium period oscillations from isolated cells. Finally, we show that strong electrical coupling between a fast burster and a slow burster can produce synchronized medium bursting, suggesting that islets may be composed of cells that are intrinsically either fast or slow, with few or none that are intrinsically medium.

Modulation of the bursting properties of single mouse pancreatic beta-cells by artificial conductances.

Glucose triggers bursting activity in pancreatic islets, which mediates the Ca2+ uptake that triggers insulin secretion. Aside from the channel mechanism responsible for bursting, which remains unsettled, it is not clear whether bursting is an endogenous property of individual beta-cells or requires an electrically coupled islet. While many workers report stochastic firing or quasibursting in single cells, a few reports describe single-cell bursts much longer (minutes) than those of islets (15-60 s). We studied the behavior of single cells systematically to help resolve this issue. Perforated patch recordings were made from single mouse beta-cells or hamster insulinoma tumor cells in current clamp at 30-35 degrees C, using standard K+-rich pipette solution and external solutions containing 11.1 mM glucose. Dynamic clamp was used to apply artificial KATP and Ca2+ channel conductances to cells in current clamp to assess the role of Ca2+ and KATP channels in single cell firing. The electrical activity we observed in mouse beta-cells was heterogeneous, with three basic patterns encountered: 1) repetitive fast spiking; 2) fast spikes superimposed on brief (<5 s) plateaus; or 3) periodic plateaus of longer duration (10-20 s) with small spikes. Pattern 2 was most similar to islet bursting but was significantly faster. Burst plateaus lasting on the order of minutes were only observed when recordings were made from cell clusters. Adding gCa to cells increased the depolarizing drive of bursting and lengthened the plateaus, whereas adding gKATP hyperpolarized the cells and lengthened the silent phases. Adding gCa and gKATP together did not cancel out their individual effects but could induce robust bursts that resembled those of islets, and with increased period. These added currents had no slow components, indicating that the mechanisms of physiological bursting are likely to be endogenous to single beta-cells. It is unlikely that the fast bursting (class 2) was due to oscillations in gKATP because it persisted in 100 microM tolbutamide. The ability of small exogenous currents to modify beta-cell firing patterns supports the hypothesis that single cells contain the necessary mechanisms for bursting but often fail to exhibit this behavior because of heterogeneity of cell parameters.

Neurotransmitters and their receptors in the islets of Langerhans of the pancreas: what messages do acetylcholine, glutamate, and GABA transmit?

Although neurotransmitters are present in pancreatic islets of Langerhans and can be shown to alter hormone secretion, their precise physiological roles in islet function and their cellular mechanisms of action are unclear. Recent research has identified specific neurotransmitter receptor isoforms in islets that may be important physiologically, because selective receptor agonists activate islet ion channels, modify intracellular [Ca2+], and affect secretion. This article focuses on the putative roles of acetylcholine, glutamate, and GABA in islet function. It has been hypothesized that acetylcholine potentiates insulin secretion by either promoting Ca release from cellular stores, activating a store depletion-activated channel, or activating a novel Na channel. GABA and glutamate, in contrast, have been proposed to mediate a novel paracrine signaling pathway whereby alpha- and beta-cells communicate within the islet. The evidence supporting these hypotheses will be critically evaluated.

Temperature modulates the Ca2+ current of HIT-T15 and mouse pancreatic beta-cells.

The effects of raising temperature on the Ca2+ currents of insulin-secreting HIT and mouse pancreatic beta-cells were studied. Currents were measured in 3 mM Ca2+ containing solutions using standard whole-cell techniques. Increasing temperature from 22 degrees C to 35 degrees C increased peak Ca2+ current amplitude, percent (fast) inactivation and decreased the time-to-peak of the current. Ca2+ currents in HIT and mouse beta-cells responded in the same manner to an imposed physiological burstwave with test-pulses: (i) application of the burstwave inactivated the test-pulse Ca2+ current at both high and low temperatures; (ii) Ca2+ current inactivation leveled off during the plateau phase at 20-22 degrees C whereas there was an apparent continual decay at 33-35 degrees C; and (iii) recovery from inactivation occurred during the interburst period at both temperatures. Application of a physiological burstwave without test-pulses to mouse beta-cells before and after addition of 0.2 mM Cd2+ resulted in a Ca2+ difference current. This current activated during the hyperpolarized interburst phase, activated, inactivated and deactivated rapidly and continually during the plateau phase, and recovered from inactivation during the interburst. Although raising temperature strongly modified HIT and mouse beta-cells Ca2+ current, our work suggests that other channels, in addition to Ca2+ channels, are likely to be involved in the control of islet bursts, particularly at different temperatures. In addition, the effect of temperature on islet cell Ca2+ current may be partly responsible for the well-known temperature dependence of glucose-dependent secretion.

An ATP-sensitive Cl- channel current that is activated by cell swelling, cAMP, and glyburide in insulin-secreting cells.

Although chloride ions are known to modulate insulin release and islet electrical activity, the mechanism or mechanisms mediating these effects are unclear. However, numerous studies of islet Cl- fluxes have suggested that Cl- movements and glucose and sulfonylurea sensitive and are blocked by stilbene-derivative Cl- channel blockers. We now show for the first time that insulin-secreting cells have a Cl- channel current, which we term ICl,islet. The current is activated by hypotonic conditions, 1-10 mumol/l glyburide and 0.5 mmol/l 8-bromoadenosine 3':5'-cyclic monophosphate sodium. ICl,islet is mediated by Cl- channels, since replacing [Cl-]o with less permeant aspartate reduces current amplitude and depolarizes its reversal potential. In addition, 100 mumol/l 4,4'-diisothiocyanatostilbene-2,2'-disulfonic acid (DIDS) or glyburide, which blocks the Cl- channels of other cell types, block ICl,islet. Reducing [ATP]i reduces the amplitude of the current, suggesting that it may be under metabolic control. The current is time-independent and shows strong outward-rectification beyond approximately 0 mV. At potentials associated with the silent phase of islet electrical activity (approximately -65 mV), ICl,islet mediates a large inward current, which would be expected to depolarize islet membrane potential. Thus, activation of this novel current by increased intracellular cAMP, sulfonylureas, or ATP may contribute to the well-known depolarizing effects of these agents.

Contribution of L- and non-L-type calcium channels to voltage-gated calcium current and glucose-dependent insulin secretion in HIT-T15 cells.

The pharmacological properties of voltage-gated Ca current and glucose-dependent insulin secretion were determined using the HIT insulinoma line to understand the role of Ca channels in stimulus-secretion coupling. The L-type Ca channel antagonist nimodipine inhibited a maximum of 50-55% of the peak Ca current, suggesting that L- and non-L-type channels contribute to Ca current. The L-agonist BAY K 8644 increased Ca current by 155%, whereas the N-channel blocker omega-conotoxin MVIIA reversibly blocked 35% of the peak Ca current. Total block with nimodipine and MVIIA was additive. Conotoxin MVIIC did not affect HIT Ca current. Prolonged depolarizations elicited rapidly and slowly inactivating Ca currents. Nimodipine partially inhibited transient current, but fully inhibited slowly inactivating current, suggesting that the former is mediated by L- and N-channels, and the latter is mediated by L-channels. Like slowly inactivating Ca current, glucose-dependent insulin secretion was fully inhibited by nimodipine and insensitive to MVIIA. BAY K potentiated secretion and antagonized nimodipine block. These results suggest that persistent Ca current is mediated by L-channels and is strongly coupled to insulin secretion, whereas transient Ca current is mediated by L- and N-channels and is weakly coupled. Sustained Ca influx may be preferentially coupled because glucose persistently depolarizes HIT cells and inactivates more transient Ca channel pathways.

Effect of Napip on K0 activation of the Na-K pump in adult rat cardiac myocytes.

To determine if environmental factors influence the external K (K0) dependence of Na-K pump current (Ip), we systematically varied internal (pipette) Na (Napip) and Na-K pump activity while measuring the K0 dependence in adult rat cardiac myocytes. For each Napip, reactivation of Ip by K0 was dose dependent. The maximal Ip (Ipmax) and apparent affinity for K0 binding to the Na-K pump (K0.5) increased as Napip increased. The results of making an equimolar substitution of tetramethylammonium for K and Cs, and partial Ip inhibition with ouabain, also showed that Ipmax and K0.5 increased as Napip increased. We simulated pump activity as a function of intracellular Na (Nai) and K0 using a cyclic model of the Na-K pump and found that the model predicts K0.5 for K0 binding increases as Na increases, even when the conditions are adjusted by removing pipette K and partial pump inhibition with ouabain.

Effect of Nai on activity and voltage dependence of the Na/K pump in adult rat cardiac myocytes.

We have measured the voltage dependence of the Na/K pump in isolated adult rat cardiac myocytes using the whole-cell patch-clamp technique. In the presence of 1-2 mM Ba and 0.1 mM Cd and nominally Ca-free, Na/K pump current (Ip) was measured as the change in current due to 1 mM ouabain. Voltage dependence of Ip was measured between -140 and +40 or +60 mV using square voltage-pulse and voltage-ramp protocols, respectively. With 150 mM extracellular Na (Nao) and 5.4 mM extracellular K (Ko), we found that the Na/K pump shows a strong positive voltage dependence between -140 and 0 mV and is voltage independent at positive potentials. Removing Nao reduced the voltage dependence at negative potentials with no effect at positive potentials. When Ko was reduced, a negative slope appeared in the current-voltage (I-V) curve at positive potentials. We have investigated whether Nai (intracellular Na) might also affect the voltage dependence of Ip by varying Na in the patch pipette (Napip) between 20 and 85 mM. We found, as expected, that Ip increased markedly as Napip was raised, saturating at about 70 mM Napip under these conditions. In contrast, while Ip saturated near +20 mV and declined to about 40% of maximum at -120 mV, there was no effect of Napip under these conditions. In contrast, while Ip saturated near +20 mV and declined to about 40% of maximum at -120 mV, there was no effect of Napip on the voltage dependence of Ip. This suggests that neither Nai binding to the Na/K pump nor the conformational changes dependent on Nai binding are voltage dependent. These results are consistent with extracellular ion binding within the field of the membrane but do not rule out the possibility that other steps, such as Na translocation, are also voltage dependent.