Pancreatic beta-cell glucose hypersensitivity in hyperglycemic, athymic "nude" mice.

Male athymic "nude" mice (ANM) of the USC colony manifest spontaneous fasting hyperglycemia and reduced glucose tolerance; it has been proposed that they may represent a model of nonobese non-insulin-dependent diabetes. Following the recent demonstration that insulin secretion from the isolated, perfused pancreas of the male ANM appears to be hypersensitive to glucose, the function of individual pancreatic islet beta cells was investigated by measuring the membrane potential electrical activity. Initial studies demonstrated that the cyclic pattern of electrical activity in isolated female ANM islets is indistinguishable from that in control mouse islets. In contrast to control and female ANM beta cells, in which 11.1 mM glucose evoked approximately 50% maximal electrical activity, this concentration evoked almost 80% maximal activity in male ANM beta cells (p < 0.01). Investigating electrical responses at different glucose concentrations demonstrated that this increased sensitivity to glucose extends across the concentration range 2.8 to 22 mM. Assuming that in these islets, as in normal islets, electrical activity is associated with insulin release, these data indicate that the glucose-versus-insulin secretion dose-response is shifted to lower glucose concentrations at the level of the individual beta cell. Although this study demonstrates that altered beta-cell function occurs in the isolated islet of the male ANM, further investigation is under way to determine how the observed beta-cell glucose hypersensitivity is related to the hyperglycemia and impaired glucose tolerance that develop in these animals in vivo.

Abnormal cardiac rhythm in diabetic rats.

A significant occurrence of abnormal rhythm was observed in perfused working hearts of diabetic rats. The incidence of arrhythmias was 19/51 in diabetics as compared with 2/38 in normal controls. In considering possible pathogenetic mechanisms, conduction system defects appear to merit particular attention.

Electrical activity, cAMP concentration, and insulin release in mouse islets of Langerhans.

The influence of forskolin and 3-iso-butyl-1-methylxanthine (IBMX) on mouse pancreatic beta-cell electrical activity, whole islet cAMP content, and insulin release were investigated. The two drugs potentiated to a similar extent both glucose-stimulated electrical activity and insulin release. In terms of the electrical response, both drugs potentiated the silent depolarization of the membrane in response to low (substimulatory) glucose concentrations, whereas at higher (stimulatory) glucose concentrations they caused an increase in the plateau fraction, with a response similar to the effect of increasing the glucose concentration. Both phases of insulin release were increased by each of the drugs. Ten micromolar forskolin and 100 microM IBMX caused an increase in intraislet adenosine 3',5'-cyclic monophosphate (cAMP) in the presence of 11.1 mM glucose, the former a 17-fold and the latter a 2-fold increase over the cAMP concentration in the presence of glucose alone. Because the two drugs lead to an increase in islet cAMP content, it is proposed that protein phosphorylation resulting from an activation of beta-cell cAMP-dependent protein kinases is responsible for the potentiation of the glucose-induced insulin release and beta-cell electrical activity. The observed effects on electrical activity are compatible with the hypothesis that cAMP-dependent phosphorylation induces alteration of the kinetics of the calcium-sensitive potassium permeability of the beta-cell plasma membrane. The increase in calcium entry into the beta-cell that would result from these alterations may be responsible for the cAMP-dependent potentiation of insulin release.

Pancreatic beta-cell electrical activity: the role of anions and the control of pH.

The influence of chloride on the mouse pancreatic beta-cell membrane potential and the cell membrane mechanisms controlling intracellular pH (pHi) have been investigated using glass microelectrodes to monitor the membrane potential. It has been shown that chloride is distributed passively across the beta-cell membrane such that chloride potential is equal to the membrane potential. Withdrawal of perifusate chloride or bicarbonate and the application of the drugs 4-acetamido-4'-isethiocyanostilbene-2,2'-disulfonic acid (SITS) and probenecid, both blockers of transmembrane anion movement, have been used to establish that a chloride-bicarbonate exchange system is operative in the cell membrane and that it is one of the control mechanisms of pHi. Amiloride, a specific blocker of the transmembrane sodium proton exchange, has been used to demonstrate that this mechanism is also operative in the beta-cell membrane in the control of pHi. The hypothesis that the calcium-activated potassium permeability is proton sensitive at an intracellular site, a fall in pHi causing a fall in permeability and an increase in pHi causing an increase in permeability, has been used to explain many of the effects observed in this study.

Effects of sodium on beta-cell electrical activity.

The present studies, designed to evaluate the contribution of Na+ to the mouse pancreatic beta-cell membrane potential, were performed utilizing intracellular microelectrodes. Complete removal of external sodium, in the presence of glucose, did not significantly affect spike peak potential. However, it caused a negative shift of the resting membrane potential, both in the presence and absence of glucose. After this initial hyperpolarization, the membrane gradually depolarized, the rate of depolarization being slower in the absence of glucose. This two-phase hyperpolarization-depolarization pattern remained when ouabain was added, both in the presence and absence of glucose. An increase of input resistance was associated with the slow depolarization. During this depolarization the maximum rate of rise (dV/dtmax) of the action potential ("spike") decreased. There was no direct relationship between dV/dtmax and [Na]0. Readdition of low [Na]0 (14 mM) to a glucose medium reactivated the postburst hyperpolarization (PBH), even in the presence of ouabain. These observations indicate that there is a significant resting sodium permeability (PNa). However, the action potential (spike) is not generated by activation of a voltage-dependent (gated) sodium channel. The membrane depolarization after Na+ removal reflects concomitant inhibition of the Na+-K+ pump and decrease of potassium permeability (PK). The blockage of PBH in the absence of Na+ is not related to the inhibition of an oscillatory Na+-K+ pump but to the inactivation of a PK. Aside from its effect on the Na+-K+ pump, ouabain may stimulate PNa.

Effects of divalent cations on beta-cell electrical activity.

The effects of various divalent cations, added to the external medium, upon beta-cell action potential were studied using intracellular microelectrodes. Changes of spike peak potential, as a function of external cation concentration, indicate that Sr2+ or Ba2+ may substitute for Ca2+ as a charge carrier. Complete blockage by Mn2+ of electrical activity elicited by Sr2+, Ba2+, or Ca2+ suggests that these cations penetrate the membrane though the same Ca2+ channel. The increase of maximum rate of depolarization, dV(d)/dtmax, and decrease of maximum rate of repolarization, dV(r)/dtmax, when Sr2+ is substituted for Ca2+ suggest that Sr2+ penetrates more readily the Ca2+ channel but is less effective than Ca2+ in activating K permeability. Reversal of these effects by addition of equimolar Ca2+ to Sr2+ indicates that Ca2+ has a greater affinity than Sr2+ for the receptor site. The blockage of electrical activity by Ba2+ at a depolarized membrane level suggests that Ba2+ markedly reduces all K+ permeabilities. Analysis of dV(d)/dtmax at various Ca2+ concentrations, in the presence of nonpermeant divalent cations (Co2+, Mn2+, and Mg2+), shown that these cations bind competitively at the same receptor site with differing dissociation constants, For all of these divalent cations, the order of binding would be Co2+ greater than Mn2+ greater than Ca2+ greater than Sr2+, Mg2+.

Calcium action potentials and potassium permeability activation in pancreatic beta-cells.

Ionic control mechanisms of mouse pancreatic beta-cell action potentials ("spikes"), in response to glucose, were studied by measuring membrane potentials with intracellular microelectrodes. The curve relating the peaks of the spikes to the log of the external calcium concentration above 10 mM has a slope of 25 mV/10-fold increase of Ca2+. This approaches the value predicted by the Nernst equation for a pure Ca2+ electrode. Increasing the external [Ca2+]o from 0 to 42.5 mM caused an increase in rates of spike depolarization and repolarization. Lowering [Ca2+]o or applying Ca2+ conductance blockers, including Co2+ (1.25 mM), Mn2+ (2mM), and D-600 (2 X 10(-4) M), caused a decrease in rates of spikes depolarization and repolarization, with an increase of [Ca2+]o reversing this effect. Higher concentrations of these Ca2+-conductance blockers eliminated the spike activity. Quinidine at a high concentration (10(-3) M) blocked spike repolarization. Tetraethylammonium (TEA, 25 mM) increased spike amplitude and duration. Therefore, it is concluded that Ca2+ entry during the spike affects potassium permeability, which is inhibited by TEA. Also, there is a competitive binding between Co2+, Mn2+, Mg2+, and Ca2+, the charge carrier. These cations may have an additional action of substituting for Ca2+ to "stabilize" the membrane.

Beta-cell electrical activity in response to high glucose concentration.

Glucose-induced beta-cell electrical activity, recorded with glass microelectrodes, is characterized by trains of fast action potentials ("spikes"). The membrane depolarizes before each train of spikes and then repolarizes. This pattern is termed a "burst." There is a characteristic biphasic response to a square wave of 11.1 mM glucose. Pulses at higher glucose concentrations (22.2 mM or more) evoke transient, constant spike activity. The duration of this activity is lengthened and the lag period shortened in proportion to the concentration and length of the glucose pulse. The lag period between removal of glucose and the cessation of spike activity is also proportional to the glucose concentration.

Cyclic variation of K+ conductance in pancreatic beta-cells: Ca2+ and voltage dependence.

Pulses of hyperpolarizing current were injected through the microelectrode recording the electrical activity of beta-cells in order to measure input resistance. Increase in resistance during depolarization of the slow oscillation ("burst") indicates inactivation of an outward current, probably K+. Decrease in resistance as the plateau commences suggests that the previous depolarization causes activation of an inward current, probably calcium. The postburst hyperpolarization, caused by a late activation of potassium permeability (PK), would result from the increase of intracellular free calcium. An intracellular buffering system may control this intracellular free calcium level. By restoring the silent phases, in the presence of ouabain or high potassium, injection of hyperpolarizing current shows also a voltage dependency of the PK involved in the postburst hyperpolarization. Glucose, by stimulating intracellular binding of calcium, would cause membrane depolarization at glucose levels below threshold and elongation of the plateau phase at higher concentrations.

Cytophysiological studies on isolated pancreatic islets in vitro.

Single, isolated pancreatic islets of mice and rats were incubated for varying time intervals (0.5-60) minutes with high (300 mg%) and low (50 mg%) levels of glucose. The structural integrity of islets decreased progressively with time regardless of glucose concentration. Degeneration of islets was greatest after 60 minutes of incubation. The total amount of insulin released from cytologically intact mouse islets incubated with high glucose levels was always greater than that with low glucose except following 30 seconds of incubation when no difference was observed. Peaks of insulin secretion noted after 2 and 15 minutes of incubation were correlated with light microscopic and fine structural changes indicative of active secretion in beta-cells, i.e., degranulation, granule margination. At 5 and 30 minutes of incubation many beta-cells contained enlarged Golgi zones and abundant profiles of swollen rough endoplasmic reticulum containing pale, amorphous granular material presumably indicating insulin synthesis. Emphasis is placed on the desirability of correlating physiological and biochemical studies of isolated pancreatic islets with cytologic examination.

Insulin from individual isolated islets of Langerhans 2. Effect of glucose in varying concentrations.

1. Insulin secretion of individual isolated islets of Langerhans from mice, determined at 15 second intervals, demonstrated a pulsatile pattern. Glucose stimulated increased amounts of insulin to be released from these individual islets. 2. Insulin secretion of individual isolated mouse islets of Langerhans incubated for 5-15 minute intervals showed a direct relationship to concentration. There was negligible "leakage" in the absence of glucose. 3. Serial determination of insulin release from individual islets at 15 sec. intervals also demonstrated augmented magnitude of insulin release in response to increase of glucose in the medium. More important, a pulsatile pattern of insulin release was evident, which may be related to inherent physiological periodicity of islet physiology as also reflected in the pattern of islet electrical activity.

Dynamic characteristics of electrical activity in pancreatic beta-cells. I. - Effects of calcium and magnesium removal.

1. Electrical activity in beta-cells of mouse islets of Langerhans was studied using ultra-fine micro-electrodes. 11.1 mM glucose stimulated electrical activity in bursts of 12 to 24 sec duration, including an active phase, characterized by depolarization and rapid fluctuations of the potential (spikes), and a silent phase when the membrane repolarized. Analysis of the spike frequency distribution during the active phase supports the hypothesis that the spikes result from exocytosis. 2. Calcium removal, in the presence of magnesium, did not alter membrane potential. Burst activity in 11.1 mM glucose was inhibited, and spike activity appeared in some cells. The frequency of these spikes increased as the magnesium was reduced. Re-introduction of calcium induced depolarization and increased spike frequency. 3. Omission of calcium and magnesium reversibly induced depolarization. In 11.1 mM glucose, there was further depolarization, and an altered pattern of electrical activity. Bursts were reversed in polarity, the silent phase being depolarized with respect to the active phase, and duration of reversed bursts was increased while the number of spikes during the active phase was decreased. Caffeine, when added, increased the number of spikes. Re-introduction of calcium and magnesium induced a transient hyperpolarization with suppression of spike activity. 4. Simultaneous reduction of calcium and magnesium to one-half or one-tenth of the control concentration enhanced the overall spike activity in 11.1 mM glucose by shortening or blocking the silent phase between bursts. 5. Quinidine, 0.14 mM, irreversibly inhibited the silent phases of bursts and modified the biphasic electrical response to glucose.