Somatostatin Inhibits Oxidative Respiration in Pancreatic -Cells
http://www.100md.com
《内分泌学杂志》
Institute of Cell Signalling, School of Biomedical Sciences, University of Nottingham, Medical School, Nottingham NG7 2UH, United Kingdom
Abstract
Somatostatin potently inhibits insulin secretion from pancreatic -cells. It does so via activation of ATP-sensitive K+-channels (KATP) and G protein-regulated inwardly rectifying K+-channels, which act to decrease voltage-gated Ca2+-influx, a process central to exocytosis. Because KATP channels, and indeed insulin secretion, is controlled by glucose oxidation, we investigated whether somatostatin inhibits insulin secretion by direct effects on glucose metabolism. Oxidative metabolism in -cells was monitored by measuring changes in the O2 consumption (O2) of isolated mouse islets and MIN6 cells, a murine-derived -cell line. In both models, glucose-stimulated O2, an effect closely associated with inhibition of KATP channel activity and induction of electrical activity (r > 0.98). At 100 nM, somatostatin abolished glucose-stimulated O2 in mouse islets (n = 5, P < 0.05) and inhibited it by 80 ± 28% (n = 17, P < 0.01) in MIN6 cells. Removal of extracellular Ca2+, 5 mM Co2+, or 20 μM nifedipine, conditions that inhibit voltage-gated Ca2+ influx, did not mimic but either blocked or reduced the effect of the peptide on O2. The nutrient secretagogues, methylpyruvate (10 mM) and -ketoisocaproate (20 mM), also stimulated O2, but this was unaffected by somatostatin. Somatostatin also reversed glucose-induced hyperpolarization of the mitochondrial membrane potential monitored using rhodamine-123. Application of somatostatin receptor selective agonists demonstrated that the peptide worked through activation of the type 5 somatostatin receptor. In conclusion, somatostatin inhibits glucose metabolism in murine -cells by an unidentified Ca2+-dependent mechanism. This represents a new signaling pathway by which somatostatin can inhibit cellular functions regulated by glucose metabolism.
Introduction
GLUCOSE IS THE chief secretagogue for the secretion of insulin from the pancreatic -cell. By its oxidative metabolism, glucose produces an increase in the cytosolic ATP of the -cell, which in turn closes the metabolically sensitive ATP-sensitive K+-channel (KATP) that regulates the plasma-membrane potential. This leads to cell depolarization; opening of voltage-gated Ca2+ channels; and induction of Ca2+-dependent electrical activity, Ca2+ influx, and insulin secretion (1).
The peptide somatostatin is universally accepted to be a ubiquitous inhibitor of stimulus-secretion coupling in a wide variety of neurons, exocrine and endocrine cells (2, 3, 4), with the inhibition of insulin secretion the most studied to date. The major source of somatostatin that affects insulin release in vivo is from enterocytes of the proximal gut; these release the long form of somatostatin (SRIF-28) into the portal circulation in response to alimentation (5). In vivo immunoneutralization studies demonstrate that the gastric release of SRIF-28 reduces postprandial release of insulin to avoid unwanted hypoglycemia and prevent decreases in insulin sensitivity of target tissues, which may result from hyperinsulinemia (5). In addition, the short form of somatostatin (SRIF-14), secreted locally from -cells within the pancreatic islet may also have inhibitory paracrine actions on -cell function (6). Clinically the importance of somatostatin and its analogs are mainly in their therapeutic deployment to inhibit secretion from a wide range of neuroendocrine and exocrine tumors (4, 7, 8). For example, somatostatin and its analogs are used to inhibit the hypersecretion of insulin that occurs with insulinomas, thereby alleviating hyperinsulinemia and associated hypoglycemia (4, 8). More recently somatostatin has been used to treat the hyperinsulinemia associated with hypothalamic obesity and also primary insulin hypersecretion (7, 8), the latter a risk factor for obesity and insulin resistance. It is also becoming apparent that the pharmacology of somatostatin receptors in the -cell changes with age (9). Clearly pharmacological knowledge of the decretin action of somatostatin is crucial in furthering our understanding of the role and therapeutic uses of this peptide and its analogs.
To date, five somatostatin receptor types (sstr1–5) of the G protein-coupled superfamily have been identified, cloned, and characterized. Although widely distributed throughout the body, they are differentially expressed (2, 3, 4). Pancreatic -cells from mouse (10), rat (11), and human (12) islets express all five receptors subtypes. However, data from combinatorial chemical methods, immunocytochemistry, and gene knockout studies strongly support the idea that the inhibition of insulin secretion by somatostatin is mediated predominantly by sstr5 in -cells of mouse (6, 10, 13, 14), rat (11), and man (15).
The major mechanism by which somatostatin inhibits insulin secretion is to decrease Ca2+ influx via a pertussis toxin-sensitive hyperpolarization of the -cell membrane potential (10, 16, 17). This occurs via the opening of two types of K+ channel: KATP (10, 18, 19) and GIRK, a member of the G protein-regulated inward rectifier ion channel family (10, 20). Somatostatin can also inhibit insulin secretion at steps distal to Ca2+ influx: it decreases the levels of cAMP, a permissive sensitizer of insulin release (21), and it inhibits the exocytotic process via direct G protein interactions (22). However, because the inhibition of Ca2+ influx by somatostatin precedes, and precludes, these other inhibitory mechanisms and because the effect of somatostatin via ionic mechanisms has the greatest potency (10), hyperpolarization of the -cell membrane potential is probably the primary mechanism by which the peptide inhibits insulin secretion.
GIRK is activated by direct G protein receptor coupling (20), whereas KATP channel activity is predominantly regulated by metabolism, foremost by changes in the concentrations of intracellular adenine nucleotides (1, 23), although it too can be controlled by direct G protein receptor interactions (18). However, this latter mechanism is unlikely to account for the activation of KATP by somatostatin because direct G protein interactions with the channel are impaired when intracellular adenine nucleotide levels are similar to those associated with glucose-stimulated insulin secretion (18). An alternative, simpler explanation is that somatostatin inhibits glucose metabolism directly and that KATP channels are then activated secondary to the subsequent alterations in intracellular adenine nucleotide concentrations. In support of this idea, somatostatin has been reported to potently (EC50 < 10 nM) inhibit glucose oxidation in rat islets (24), although others have failed to confirm this observation (25). Because the inhibition of glucose oxidation is well established to activate KATP channels (1, 23), we investigated the idea that somatostatin inhibits electrical activity in murine -cells via direct affects on glucose metabolism. To do this, we used polarography to evaluate oxidative respiration (26, 27), fluorescence techniques to monitor the mitochondrial events (28, 29), and electrophysiology to monitor the associated plasma membrane electrical events (10).
Materials and Methods
Cells
For islets studies, male NMRI (National Memorial Research Institute) mice, 3–4 months of age, fed and watered ad libitum, were killed by a cervical dislocation using methods that comply with local guidelines and are approved by the University and U.K. home office. Pancreatic islets were then isolated from excised pancreata by collagenase digestion (26). However, several factors confound interpretation of metabolic measurements made on pancreatic islets: 1) pancreatic islets consist of a multitude of cell types, with the exact contribution of -cells to their overall O2 consumption unknown; 2) paracrine influences can be problematic with isolated islet studies in which normal islet perfusion is disrupted, especially the secretion of somatostatin from pancreatic -cells located within the islet mantle; and 3) large numbers of islets are required. To overcome these problems, we also used the endogenous, insulin-secreting, mouse pancreatic -cell line MIN6 (30), which provided large numbers of cells homogenous in phenotype and free from paracrine influences. MIN6 cells were used with the permission of Prof. Jun-Ichi Miyazaki (Osaka University Medical School, Osaka, Japan) (31). Cells were grown in RPMI 1640 tissue culture media containing 11 mM glucose, 12.5 mM HEPES, and 10% fetal calf serum, kept at 37 C in a humidified atmosphere of 5% CO2-95% air.
O2 consumption
The O2 consumption of islets or MIN6 -cell suspensions of known cell density was measured polarographically using Clark oxygen electrodes (Rank Brothers, Bottisham UK). The partial pressure of O2 (PO2) was measured at a polarographic voltage of –0.6 V with electrodes previously calibrated at 100% air saturation (vigorous gassing with air, 0.25 mM) and 0% (addition of Na2S2O4). All additions were made from stocks in H2O. The control for metabolic substrate addition was 3-O-methyl-glucose (MOG), a metabolically inactive analog of glucose (32). The control for peptide addition was vehicle alone. The rate of oxygen consumption (O2), was measured as the change in PO2 level over a 300-sec period; given that most changes were linear, this is the slope of PO2. For metabolic substrates or Ca2+ channel antagonists, the slope was measured 360 sec after reagent addition, whereas for somatostatin and its receptor agonists, to avoid tachyphylaxis, 60 sec after addition of the drug.
Mitochondrial membrane potential (mit)
To monitor the inner membrane mit we used rhodamine-123 (Rh-123) as previously described (28). Cells were loaded with 50 μg/ml–1 Rh-123 in RPMI 1640 (11 mM glucose) at 37 C for 10 min. The dye was excited at 480 nm and the emitted fluorescence monitored at 530 nm. It is well established that this cationic lipophilic dye is predominantly localized to mitochondria. Depolarization of mit, which would, for example, occur with uncoupling or blocking of mitochondrial respiration, results in the redistribution of the dye across the inner membrane with an associated increase in fluorescence, whereas an increased flux through the mitochondrial respiratory electron transport chain is associated with a decrease in fluorescence (see Fig. 4) (28, 29).
Electrophysiology
To maintain cellular metabolism and second-messenger systems intact, membrane potentials (Vm) were recorded using the perforated-patch patch-clamp technique as described previously (10). Patch pipettes were filled with (in millimoles): 76 K2SO4, 10 KCl, 10 NaCl, 55 sucrose, 10 HEPES (pH 7.2 with KOH), and 1 MgCl2. Perforation was achieved by the addition of 100 μg/ml–1 amphotericin B to the pipette solution and was considered adequate when the series conductance exceeded 40 nS (<25 M).
To mirror the experimental conditions used for the metabolic measurements as closely as possible, KATP channel activity was derived from the current clamp records without recourse to voltage-clamp measurements. If the membrane conductances that contribute to the passive Vm are assumed Ohmic, then the conductance associated with KATP channel activity (GATP) can be estimated from equation 1:
(1)
where Rin is the input resistance of the cell, Vm is the membrane potential, and Ek and El are the reversal potentials for the KATP channel and the leak conductances, respectively. Ek is estimated to be approximately –69 mV, the Vm when KATP is maximally opened by diazoxide; El is estimated to be approximately –39 mV, the interspike Vm when KATP is inhibited by 200 μM tolbutamide. Rin was calculated from the membrane potential deflections that were elicited in response to –10 pA current pulses, 200 msec in duration, applied at 0.1 Hz.
Solutions
To permit integration with previously published electrophysiological data, all experiments were carried out at 32 C in a modified Hanks’ solution that contained (in millimoles): 5.6 KCl, 138 NaCl, 4.2 NaHCO3, 1.2 NaH2PO4, 2.6 CaCl2, 1.2 MgCl2, 10 HEPES (pH 7.4 with NaOH), and 0.1% (wt/vol) BSA. For nominally Ca2+-free solutions, CaCl2 was replaced by equimolar substitution with MgCl2 (total Mg2+ = 3.8 mM, free Ca2+ 10 μM).
SRIF-14 and SRIF-28 were from Peninsula Laboratories (St. Helens, UK), whereas CH-275, BIM-23027, BIM-, L-362, 855, and NNC-26, 9100 were purchased from commercial suppliers of synthetic peptides. The structural integrity and molecular weight of the peptides were confirmed by M@LDI mass spectroscopy (J. Kyte, School of Biomedical Sciences, University of Nottingham). All other drugs and reagents were purchased from Sigma (Poole, UK) or Fluka (Buchs, Switzerland).
Glucose-response relationships
The glucose-response relationships were quantified by best fits of the data with the following equation:
where G is the concentration of glucose, EC50 is the concentration that produces the half-maximal response, Y is the response magnitude, YMAX is the maximal response, YMIN is the minimal response, and h is an index of slope.
Statistics
Unless stated otherwise for statistical comparison of two-sample populations, Wilcoxon signed rank test was used; for multiple comparisons, Kruskal-Wallis test was used with Dunn’s multiple comparison posttest. These procedures were performed using PRISM 3 (GraphPad Software Inc., San Diego, CA). Data statistics are given as means ± SEM, with n as the number of preparations given in parentheses. EC50 and h values are quoted with 95% confidence intervals. Correlation coefficients are Pearson. Statistical significance is defined as P < 0.05 and in graphics is flagged as , when P < 0.01 data are flagged with and when P < 0.001 data are flagged with .
Results
Simulation of O2 consumption by glucose and its inhibition by somatostatin
In the absence of added substrate, both islets and MIN6 possessed a linear basal respiratory rate, within 1 min of adding 20 mM glucose O2 increased by approximately 63% and approximately 80%, respectively (Fig. 1). Subsequent addition of SRIF-14 significantly inhibited O2 in both islets and MIN6 cells (Fig. 1). In some cases the inhibition of O2 was greater than that stimulated by the sugar, indicative of inhibition of the basal respiration. The inhibitory effect of the peptide was sustained for at least 10 min, the longest period tested. In control experiments, addition of H2O the vehicle for somatostatin was either without effect or slightly stimulated O2 (Figs. 1B and 2D). The inhibitory action of SRIF-14 was independent of the time of addition; the peptide had a comparable inhibitory effect when added 60 min after glucose. These data also confirm the functional integrity of the -cells over time and lack of cellular damage due to stirring. The protonophores, carbonyl cyanide 3-chlorophenylhydrazone (CCCP, 100 nM) and carbonyl-cyanide-p-trifluoro-methoxy-phenylhydrazone (FCCP; 100 nM), dramatically increased O2 (Fig. 1, C and D), effects consistent with their action as uncouplers of mitochondrial oxidative phosphorylation. Subsequent application of 3 mM azide, a blocker of cytochrome oxidase, partly blocked O2 consumption in both islets and MIN6 cells (Fig. 3A). These latter two experimental paradigms also confirm that the majority of O2 consumption measured reflects mitochondrial respiration. For reasons explained in Materials and Methods, all further experimentation solely used MIN6 cells.
Glucose stimulated O2 in a concentration-dependent manner with an EC50 of 0.27 mM (0.21–0.33 mM) and h of 1 (Fig. 5D, solid line). The control for glucose, MOG (20 mM), a glucose analog that is taken up but not metabolized by -cells (32), did not stimulate O2 consumption (Fig. 2A), an observation consistent with the inability of the analog to stimulate the insulin-secretion pathway. Instead, however, it actually blocked the basal rate of O2 consumption by 28 ± 9% (n = 14; P < 0.01). To further explore the metabolic pathways affected by somatostatin, two other metabolic substrates that stimulate insulin secretion via entry into different steps of the oxidation pathway for glucose were used. Methylpyruvate (10 mM), a membrane-permeable form of the mitochondrial fuel pyruvate (33), and -ketoisocaproate (20 mM), a fuel that feeds directly into the mitochondrial tricarboxylic acid cycle (TCA) (34), stimulated O2 by amounts approaching that produced by 20 mM glucose (Fig. 2A). However, 100 nM SRIF-14 had no effect on O2 stimulated by either of these two alternative mitochondrial substrates (Fig. 2D). Moreover, 100 nM SRIF-14 also failed to inhibit O2 stimulated by glucose concentrations less than 5 mM (Fig. 2D). Surprisingly, the peptide did, however, cause a further 30 ± 4% inhibition of basal respiration (P < 0.007, n = 16) when it was already suppressed by 20 mM MOG. SRIF-14 had on effect on O2 in cell suspensions that failed to respond to glucose.
Ca2+ dependency of oxygen consumption and its inhibition by somatostatin
Because evidence exists that suggests intracellular Ca2+ regulates glucose oxidation in -cells (35), we explored the possibility that the inhibitory effect of somatostatin we observed on respiration was actually secondary to the decrease in Ca2+ influx (16) that occurs with the inhibition of electrical activity produced by the peptide (10). Removal of extracellular Ca2+ had little effect on glucose-stimulated O2 consumption (Fig 2B). Similarly, blockade of voltage-gated Ca2+ influx with 5 mM Co2+ or 20 μM nifedipine, a dihydropyridine-sensitive L-type Ca2+-channel antagonist, also had no effect on O2. Under conditions in which Ca2+ influx was impaired by Co2+ or nifedipine, 100 nM SRIF-14 still inhibited O2 (Fig. 2, B and C) but to a lesser extent than seen under control conditions. In the absence of extracellular Ca2+, SRIF-14 failed to have a significant inhibitory effect on O2.
cAMP dependency of oxygen consumption and its inhibition by somatostatin
To investigate whether a decrease in cytoplasmic cAMP concentration is involved in the inhibition of O2 consumption by somatostatin, the effect of the peptide was explored in cells preincubated with the stable membrane-permeable cAMP analog 8-(4-chlorophenylthio)-adenosine 3',5'-cyclic monophosphate (8-CPT). Incubation in 50 μM 8-CPT, a concentration that also inhibits cAMP-specific phosphodiesterases, did not change O2 (–5.3 ± 1.4, compared with –4.6 ± 1.1 nmol 107 cells–1 min–1 in control, n = 10). 8-CPT did not affect the ability of SRIF-14 to inhibit the O2 stimulated by glucose: 100 nM SRIF-14 inhibited O2 by 50 ± 10% to –2.4 ± 0.7 nmol 107 cells–1 min–1 in the presence of 8-CPT (P < 0.05 repeated-measure ANOVA), compared with 49 ± 13% inhibition in vehicle alone (n = 9).
Identification of somatostatin receptor subtype
All five sstr subtypes have been identified in murine pancreatic -cells, including MIN6 cells (3, 10). Figure 3E shows that L-362, 855, a partial agonist at the sstr5 (36, 37), partly mimicked the action of SRIF-14 (n = 17). L-362, 855 inhibited O2 by 51%, an effect less pronounced than that produced by SRIF-14 (81% Fig. 2, D and E). CH-275, BIM-23056, BIM-23027, and NNC-26, 9100, specific agonists at the sstr1 (38), sstr2/3 (37, 39), sstr3 (37, 39), and sstr4 (40), respectively, were all without effect on O2 (n = 12–13; Fig. 2E).
Oxygen consumption by nicotinamide adenine dinucleotide phosphate [NAD(P)H] oxidase and its inhibition by somatostatin
Figure 3 shows that SRIF-14 had no effect on O2 in the combined presence of glucose and azide. Application of 0.5 mM apocynin, an inhibitor of NAD(P)H oxidase, significantly decreased O2 by 72 ± 4% (n = 7, Fig. 3D), an amount greater than that produced by 0.5% (vol/vol) (70 μM) dimethylsulfoxide (DMSO), the vehicle control (25 ± 5%, n = 8, Fig. 3C), although the difference between these effects was insignificant. The block produced by 100 nM SRIF-14 in the presence of apocynin (12 ± 7%, n = 7) was less than that produced in the presence of vehicle alone (34 ± 8%, n = 8); however, this difference was insignificant. The final overall block of O2 was similar in both cases: 84 ± 5%, with 0.5 mM apocynin plus 100 nM SRIF-14 (n = 7), and 59 ± 11%, with 70 μM DMSO plus 100 nM SRIF-14 (n = 8). Addition of 1 μM diphenyleneiodonium chloride, a nonselective inhibitor of flavoproteins, which include NAD(P)H oxidases, abolished O2 (Fig. 3B).
Effect of glucose and somatostatin on mit
Addition of 5 mM glucose to MIN-6 cells, previously incubated for 20 min in the absence of exogenous metabolic substrate, produced a sustained decrease in Rh-123 fluorescence (Fig. 4A). The effect was complete with 1 min of sugar addition and amounted to an approximately 20% drop in fluorescence (Fig. 4B), a result consistent with the redistribution and subsequent quench of the dye that occurs with its accumulation by energized mitochondria (28, 29). Addition of 100 nM SRIF-14 in the maintained presence of glucose led to a partial reversal of the change in Rh-123 fluorescence produced by the sugar. This effect was slower than that elicited by glucose, with the maximum change observed approximately 2 min after addition of the peptide, which amounted to 38 ± 8% inhibition of the glucose-sensitive change in Rh-123 fluorescence (n = 6). Subsequent addition of 1 μM CCCP resulted in an approximately 100% increase in Rh-123 fluorescence (Fig. 4B), a result consistent with the redistribution and loss of quenching of the dye that occurs with the collapse of mit produced by this protonophoric uncoupler of mitochondrial oxidative phosphorylation (28).
Electrophysiology
Elevation of glucose from 0 to 10 mM depolarized the Vm and evoked Ca2+-dependent action potential activity; the latter event is well established to promote Ca2+ influx (1, 16). In the presence of 10 mM glucose, 100 nM SRIF-14, 3 mM azide, 20 μM nifedipine, or removal of extracellular Ca2+ all abolished the action potential activity induced by the sugar and, with the exception of the dihydropyridine, hyperpolarized the membrane potential (data not shown): electrophysiological changes associated with a decrease in Ca2+-influx and insulin-secretion (1, 16).
The steady-state relationships among the concentration of glucose, Vm, Rin, and the whole-cell conductance of KATP (GATP) are explored in Fig. 5. Glucose depolarized Vm with an EC50 of 1.3 mM (0.9–1.7 mM) and a slope coefficient, h, of 1.2 (0.9–1.7, solid line, Fig. 5A). This effect was mirrored by a parallel increase (r = 1, P < 0.0001) in the Rin; EC50 of 1.4 mM glucose (0.5–3.7 mM), and h of 1.0 (0.4–1.7, Fig. 5B, solid line). Of the 21 -cells that depolarized in response to glucose, 17 depolarized sufficiently to reach the voltage threshold for action potentials (–50 ± 2 mV, n = 17) and evoke electrical activity. In these cells, electrical activity was elicited at a median glucose concentration of 2.5 mM (1–3.4, 95% confidence limits). The relationship between glucose concentration and the percentage of cells that responded with electrical activity is drawn in Fig. 5C: EC50 of 1.2 mM glucose (0.9–1.5 mM) and h of 1.8 (1.1–2.6, Fig. 5C, solid line).
As expected from Rin, GATP decreased with increased glucose concentration (Fig. 5D). The relationship of GATP with glucose concentration (closed symbols) closely mirrored that obtained for the rate of O2 consumption (open symbols): EC50 of 0.3 mM (0.2–0.4 mM) and h of 1.8 (0.6–2.9, Fig. 5D, solid line). The close correlation between GATP and the rate of O2 consumption is supported by the clear inverse linear relationship (r > 0.98, P < 0.0001) that exists between these two variables (Fig. 5E).
Discussion
We have unequivocally shown that SRIF-14 inhibits O2 stimulated by glucose concentrations that evoke electrical activity and stimulate insulin secretion in both mouse islets and murine-derived -cells. These findings support and corroborate the original observation made in rat islets (24). One possible reason for the failure of others (25) to confirm this observation in rat may be simply explained by the small sample size employed in that study (i.e. n = 3).
Substrate dependency of O2 consumption
The 2-fold stimulation in the rate of O2 consumption produced by 20 mM glucose for both -cell models are comparable with that previously demonstrated for MIN6 cells (27) and -cells within intact rodent islets (26, 41). Indeed the absolute values for O2 consumption stimulated by 20 mM glucose are comparable with those already published: 10 pmol O2 islet–1 min–1, compared with approximately 4 pmol O2 islet–1 min–1 for rat islets (41) and 5–10 nmol O2 107 cell–1 min–1, compared with approximately 10 nmol O2 107 cell–1 min–1 for MIN6 (27). Assuming an islet possesses approximately 1000 -cells, then 20 mM glucose stimulates O2 consumption by approximately 40 nmol O2 107 cell–1 min–1, an amount similar to that measured for the cell line. Methylpyruvate and -ketoisocaproate are insulinotropes metabolized predominantly within the mitochondria via the TCA cycle (33, 34). They are thought to act via the same metabolic mechanism engaged by the stimulus-secretion pathway of glucose: generation of ATP and closure of KATP with the resultant electrical activity and insulin secretion (1, 33, 34), an idea supported here in MIN6 by the ability of these substrates to stimulate oxidative respiration (41, 42), with -ketoisocaproate also known to stimulate insulin secretion in MIN6 cells (43).
The EC50 for glucose-stimulated O2 (0.27 mM) is substantially less than that measured for mouse islets: 8 mM (42). The phosphorylation of glucose to glucose-6-phosphate is well established to be the key rate-limiting step for the use of the sugar and subsequent stimulus-secretion cascade for insulin release in the pancreatic -cell (44, 45). In native -cells, phosphorylation occurs predominantly via a low-affinity glucokinase (Km 5–10 mM). Consequently many events of the stimulus-secretion pathway for insulin in native -cells possess EC50 values for glucose around 5–10 mM (42, 44, 45, 46). The low millimolar values we obtained for the EC50s of various steps within the stimulus-secretion pathway in MIN6 suggest that the activity of a high-affinity hexokinase (Km 0.3 mM) predominates over that of glucokinase (43), an idea supported by the increased sensitivities to glucose for the inhibition of GATP and excitation of electrical activity, compared with those found in native mouse -cells [0.29 vs.3.2 mM (44) and 1.2 vs. 11.5 mM (46), respectively]. Nevertheless, the marked correlation observed among the stimulation of O2 consumption, inhibition of GATP, and induction of electrical activity by glucose strongly supports the idea that oxidation of the sugar in MIN6 results in closure of the KATP channel like that occurring in native -cells (1, 44, 46). The inhibition of basal respiration by 3-O-methyl-D-glucose was surprising, given that this glucose analog is not thought to be metabolized; however, our finding is consistent with its known adverse affect on -cell metabolism at low glucose concentrations (32).
Ca2+ dependence of O2 consumption and its inhibition by SRIF-14
Whether extracellular Ca2+ affects glucose metabolism in pancreatic -cell is disputable; some studies report effects on respiration (47) and glucose oxidation (35), whereas others do not (41). We found blockade of Ca2+ influx did not affect O2 consumption and did not mimic the effect of SRIF-14. However, the subsequent inhibition of O2 by the peptide was abolished in the absence of extracellular Ca2+ and was reduced when voltage-gated Ca2+ influx was blocked with Co2+ or nifedipine. These data suggest that the mechanism by which the peptide decreases O2 is not secondary to, but requires, Ca2+ influx; with the associated increase in cytosolic Ca2+ a permissive factor, an idea supported by the fact that SRIF-14 was most effective at glucose concentrations that elicited electrical activity and voltage-dependent Ca2+ influx (cf. Figs. 2D and 5C). Because somatostatin had no effect on O2 stimulated by the mitochondrial fuels, methylpyruvate or -ketoisocaproate, SRIF-14 must inhibit metabolism at a step/s before the TCA cycle. The observation that SRIF-14 inhibited basal respiration only when it was already suppressed by the presence of 20 mM 3-O-methyl-D-glucose, suggests that the molecular action of the peptide may also involve the presence and/or transport of hexose.
The inability of somatostatin to affect O2 stimulated by the mitochondrial fuels refutes the idea that the effect of the peptide is due to changes in the respiratory control of oxidative phosphorylation: changes that arise from the inhibition of secretion previously stimulated by these substrates and the resultant attenuation in energy demand (48). Because we also show that somatostatin depolarizes mit in the presence of glucose, these data similarly refute the idea that the decrease in O2 produced by the peptide is due to the drop in the energetic requirements of the -cell that results from the associated inhibition of the exocytotic process. In fact, the increase in the cytosolic ATP to ADP ratio that is predicted on inhibition of the exocytotic process is expected to inhibit KATP channel activity and promote polarization of mit (29), neither of which are seen; instead the opposite is observed with the peptide (Ref.10 and present study, respectively). Overall the Rh-123 data further support the notion that the peptide directly inhibits oxidative phosphorylation at some unknown step before the TCA cycle.
The inability of 8-CPT, a stable membrane-permeable cAMP analog, to affect the inhibition of O2 caused by somatostatin refutes the idea that a decrease in cytosolic cAMP (21) is involved in this particular effect of the peptide. In fact, decreases in cAMP levels are classically associated with stimulation, not inhibition, of glucose catabolism. The identity of the molecular target/s for this inhibitory action of somatostatin on glucose stimulated O2 remain/s unknown.
The pharmacological profile that was observed for the inhibition of O2 by the sstr-selective analogs is consistent with the peptide prosecuting its affects solely through binding to sstr5, the same receptor isoform that mediates activation of the KATP channel (10) and inhibition of insulin secretion in mouse (6, 10, 13, 14), rat (11), and man (15). The observation that neither NNC-26, 9100 nor CH-275, agonists specific for the sstr4 and sstr1, respectively, affected O2 rules out participation of these receptor subtypes in the inhibition of O2 by the peptide. Furthermore, the inability of BIM-23027 and BIM-23056, agonists more potent at the sstr2 and sstr3 than L-362, 855 (39), to inhibit O2 argues against an involvement of these receptor subtypes. The limited ability of L-362, 855 to mimic the effect of SRIF-14 is consistent with its action as a partial agonist at sstr5 (36).
Oxygen consumption NAD(P)H oxidase and its inhibition by somatostatin
The observation that azide did not always abolish O2 consumption necessitates an additional mechanism of O2 reduction in addition to that attributed to the mitochondrial electron transport chain. Reduction and consumption of O2 can also occur by NAD(P)H oxidase activity and superoxide production. RT-PCR and Western blotting have identified a range of NAD(P)H oxidase isoforms expressed within pancreatic islet tissue. Of these, the p47PHOX isoform is predominantly expressed by the pancreatic -cell at the plasma membrane (49, 50). The observation that apocynin, an inhibitor of NAD(P)H oxidases, did not significantly decrease O2, compared with vehicle control (DMSO), does not appear to support the idea that some O2 is via activity of this particular enzyme. However, such an interpretation of these data must be treated with caution because DMSO is itself an antioxidant, which may affect NAD(P)H oxidase activity; a fact that confounds whether this enzyme contributes to O2. Also, the finding that somatostatin could still inhibit the remaining O2 after the effects of apocynin suggests that the peptide inhibits O2 by mechanism/s other that that involving NAD(P)H oxidase activity. Furthermore, because the blockade of mitochondrial reduction of O2 by azide, which increases the availability of NAD(P)H for oxidation by NAD(P)H oxidases, did not enhance the effect of somatostatin on O2 but in fact abolished it, also suggests that the peptide does not affect NAD(P)H oxidase activity. The observation that diphenyleneiodonium chloride, an inhibitor of NAD(P)H oxidases, abolished O2 is attributed to its action as a potent inhibitor of most flavoproteins, which includes complex I of the mitochondrial respiratory electron transport chain. Together, these data do not readily support an involvement of NAD(P)H oxidases in either O2 consumption of pancreatic -cells or the effect of somatostatin on O2.
Physiological implications
The inhibition of glucose metabolism by somatostatin is expected to decrease cytosolic ATP of the -cell and contribute to the pleiotropic inhibitory mechanisms of the peptide on insulin secretion. Unfortunately, due to compartmentalization of ATP within the -cell, this idea is difficult to test directly (51); however, the very fact that the activity of KATP channels reflects intracellular nucleotide levels lends strong support to the above idea (1, 23). In addition to the observed activation of KATP channels (10, 19), a decrease in cytosolic ATP is expected to inhibit both Ca2+ channel activity and exocytosis, processes in the -cell enhanced by glucose metabolism (52, 53) but also inhibited by somatostatin (54), although these may also involve direct G protein pathways. Involvement of a direct G protein pathway in the activation of the KATP channel by somatostatin is unlikely because direct activation of KATP by G-proteins is impaired when intracellular ATP is elevated to levels that are associated with glucose-stimulated insulin secretion (18). Consequently the inhibition of metabolism by somatostatin is a simple mechanism to explain the activation of KATP channels by the peptide that occurs during glucose-stimulated insulin secretion. Whether such a similar simple mechanism operates in other cell types in which KATP is activated by somatostatin, e.g. in pituitary gonadotrophs (55) and CA3 hippocampal neurons (56), remains to be tested.
In conclusion, we have described a novel Ca2+-dependent signaling mechanism, namely a decrease in glucose metabolism, which enables somatostatin via activation of sstr5 to inhibit insulin secretion from the murine pancreatic -cell at a step early in its stimulus-secretion coupling cascade.
Acknowledgments
The authors thank the Royal Society and the Institute for Cell Signaling for support.
Footnotes
The authors have no conflict of interest.
First Published Online December 15, 2005
Abbreviations: CCCP, Carbonyl cyanide 3-chlorophenylhydrazone; 8-CPT, 8-(4-chlorophenylthio)-adenosine 3',5'-cyclic monophosphate; DMSO, dimethylsulfoxide; GATP, KATP channel activity; GIRK, member of the G protein-regulated inward rectifier ion channel family; KATP, ATP-sensitive K+-channels; mit, mitochondrial membrane potential; MOG, 3-O-methyl-glucose; NAD(P)H, nicotinamide adenine dinucleotide phosphate; O2, O2 consumption; PO2, partial pressure of O2; Rh-123, rhodamine-123; Rin, input resistance of the cell; SRIF-14, short form of somatostatin; SRIF-28, long form of somatostatin; sstr, somatostatin receptor; TCA, tricarboxylic acid cycle; Vm, membrane potential.
Accepted for publication December 1, 2005.
References
Ashcroft FM, Rorsman P 1989 Electrophysiology of the pancreatic -cell. Prog Biophys Mol Biol 54:87–143
Csaba Z, Dournaud P 2001 Cellular biology of somatostatin receptors. Neuropeptides 35:1–23
Reisine T, Bell GI 1995 Molecular properties of somatostatin receptors. Neuroscience 67:777–790
Weckbecker G, Lewis I, Albert R, Schmid HA, Hoyer D, Bruns C 2003 Opportunities in somatostatin research: biological, chemical and therapeutic aspects. Nat Rev Drug Discov 2:999–1017
Ensinck JW, Vogel RE, Laschansky EC, Koerker DJ, Prigeon RL, Kahn SE, D’Alessio DA 1997 Endogenous somatostatin-28 modulates postprandial insulin secretion. Immunoneutralization studies in baboons. J Clin Invest 100:2295–2302
Tirone TA, Norman MA, Moldovan S, DeMayo FJ, Wang XP, Brunicardi FC 2003 Pancreatic somatostatin inhibits insulin secretion via SSTR-5 in the isolated perfused mouse pancreas model. Pancreas 26:67–73
Boehm BO 2003 The therapeutic potential of somatostatin receptor ligands in the treatment of obesity and diabetes. Exp Opin Investig Drugs 12:1501–1509
Hansen JB, Arkhammar PO, Bodvarsdottir TB, Wahl P 2004 Inhibition of insulin secretion as a new drug target in the treatment of metabolic disorders. Curr Med Chem 11:1595–1615
Wang XP, Yang J, Norman MA, Magnusson J, DeMayo FJ, Brunicardi FC 2005 SSTR5 ablation in islet results in alterations in glucose homeostasis in mice. FEBS Lett 579:3107–3114
Smith PA, Sellers LA, Humphrey PPA 2001 Somatostatin activates two types of inwardly rectifying K+ channels in MIN6 cells. J Physiol 532:127–142
Mitra SW, Mezey E, Hunyady B, Chamberlain L, Hayes E, Foor F, Wang YN 1999 Colocalization of somatostatin receptor sst5 and insulin in rat pancreatic -cells. Endocrinology 140:3790–3796
Kumar U, Sasi R, Suresh S, Patel A, Thangaraju M, Metrakos P, Patel SC, Patel YC 1999 Subtype-selective expression of the five somatostatin receptors (hSSTR1–5) in human pancreatic islet cells: a quantitative double-label immunohistochemical analysis. Diabetes 48:77–85
Fagan SP, Azizzadeh A, Moldovan S, Ray MK, Adrian TE, Ding XZ, Coy DH, Brunicardi FC 1998 Insulin secretion is inhibited by subtype five somatostatin receptor in the mouse. Surgery 124:254–258
Strowski MZ, Kohler M, Chen HY, Trumbauer ME, Li Z, Szalkowski D, Gopal-Truter S, Fisher JK, Schaeffer JM, Blake AD, Zhang BB, Wilkinson HA 2003 Somatostatin receptor subtype 5 regulates insulin secretion and glucose homeostasis. Mol Endocrinol 17:93–106
Zambre Y, Ling Z, Chen MC, Hou X, Woon CW, Culler M, Taylor JE, Coy DH, Van Schravendijk C, Schuit F, Pipeleers DG, Eizirik DL 1999 Inhibition of human pancreatic islet insulin release by receptor-selective somatostatin analogs directed to somatostatin receptor subtype 5. Biochem Pharmacol 57:1159–1164
Nilsson T, Arkhammar P, Rorsman P, Berggren PO 1989 Suppression of insulin release by galanin and somatostatin is mediated by a G-protein. An effect involving repolarization and reduction in cytoplasmic free Ca2+ concentration. J Biol Chem 264:973–980
Pace CS, Tarvin JT 1981 Somatostatin: mechanism of action in pancreatic islet -cells. Diabetes 30:836–842
Ribalet B, Eddlestone GT 1995 Characterization of the G-protein coupling of a somatostatin receptor to the K-ATP channel in insulin-secreting mammalian HIT and RIN cell lines. J Physiol 485:73–86
de Weille JR, Schmid-Antomarchi H, Fosset M, Lazdunski M 1989 Regulation of ATP-sensitive K+ channels in insulinoma cells: activation by somatostatin and protein kinase C and the role of cAMP. Proc Natl Acad Sci USA 86:2971–2975
Yamada M, Inanobe A, Kurachi Y 1998 G protein regulation of potassium ion channels. Pharm Rev 50:723–760
Schuit FC, Derde MP, Pipeleers DG 1989 Sensitivity of rat pancreatic and cells to somatostatin. Diabetologia 32:207–212
Ullrich S, Prentki M, Wollheim CB 1990 Somatostatin inhibition of Ca2+-induced insulin secretion in permeabilized HIT-T15 cells. Biochem J 270:273–276
Tarasov A, Dusonchet J, Ashcroft F 2004 Metabolic regulation of the pancreatic -cell ATP-sensitive K+ channel: a pas de deux. Diabetes 53:S113–S122
Hahn HJ, Gottschling HD 1976 Somatostatin-induced inhibition of insulin secretion by isolated pancreatic rat islets prepared by micro-dissection or collagenase digestion. Diabetes Metab 2:107–111
Gorus FK, Malaisse WJ, Pipeleers DG 1984 Differences in glucose handling by pancreatic A- and B-cells. J Biol Chem 259:1196–1200
Hellerstrm C 1967 Effects of carbohydrates on the O2 consumption of isolated pancreatic islets of mice. Endocrinology 81:105–112
Soejima A, Inoue K, Takai D, Kaneko M, Ishihara H, Oka Y, Hayashi JI 1996 Mitochondrial DNA is required for regulation of glucose-stimulated insulin secretion in a mouse pancreatic cell line, MIN6. J Biol Chem 271:26194–26199
Duchen MR, Smith PA, Ashcroft FM 1993 Substrate-dependent changes in mitochondrial function, intracellular free-calcium concentration and membrane channels in pancreatic -cells. Biochem J 294:35–42
Baracca A, Sgarbi G, Solaini G, Lenaz G 2003 Rhodamine 123 as a probe of mitochondrial membrane potential: evaluation of proton flux through F(0) during ATP synthesis. Biochim Biophys Acta 1606:137–146
Ishihara H, Asano T, Tsukuda K, Katagiri H, Inukai K, Anai M, Kikuchi M, Yazaki Y, Miyazaki J, Oka Y 1993 Pancreatic cell line MIN6 exhibits characteristics of glucose metabolism and glucose-stimulated insulin secretion similar to those of normal islets. Diabetologia 36:1139–1145
Miyazaki J, Araki K, Yamato E, Ikegami H, Asano T, Shibasaki Y, Oka Y, Yamamura K-I 1990 Establishment of a pancreatic cell line that retains glucose-inducible insulin secretion: special reference to expression of glucose transporter isoforms. Endocrinology 127:126–132
Sener A, Scruel O, Louchami K, Jijakli H, Malaisse WJ 1999 Inhibition of glucose-induced insulin release by 3-O-methyl-D-glucose: enzymatic, metabolic and cationic determinants. Mol Cell B 194:133–145
Jijakli H, Nadi AB, Cook L, Best L, Sener A, Malaisse WJ 1996 Insulinotropic action of methyl pyruvate: enzymic and metabolic aspects. Arch Biochem Biophys 335:245–257
Gao Z, Young RA, Li G, Najafi H, Buettger C, Sukumvanich SS, Wong RK, Wolf BA, Matschinsky FM 2003 Distinguishing features of leucine and -ketoisocaproate sensing in pancreatic -cells. Endocrinology 144:1949–1957
Hellman B, Idahl LA, Lernmark A, Sehlin J, Taljedal IB 1974 The pancreatic -cell recognition of insulin secretagogues. Effects of calcium and sodium on glucose metabolism and insulin release. Biochem J 138:33–45
Carruthers A, Warner AJ, Michel AD, Feniuk W, Humphrey PPA 1999 Activation of adenylate cyclase by human recombinant sst5 receptors expressed in CHO-K1 cells and involvement of Gs proteins. Br J Pharmacol 126:1221–1229
Raynor K, O’Carroll AM, Kong H, Yasuda K, Mahan LC, Bell GI, Reisine T 1993 Characterization of cloned somatostatin receptors sstr4 and sstr5. Mol Pharm 44:385–392
Liapakis G, Hoeger C, Rivier J, Reisine T 1996 Development of a selective agonist at the somatostatin receptor subtype sstr1. J Pharm Exp Ther 276:1089–1094
Raynor K, Murphy WA, Coy DH, Taylor JE, Moreau J-P, Yasuda K, Bell GI, Reisine T 1993 Cloned somatostatin receptors: identification of subtype-selective peptides and demonstration of high affinity binding of linear peptides. Mol Pharm 43:838–844
Ankersen M, Crider M, Liu S, Ho B, Andersen HS, Stidsen C 1998 Discovery of a novel non-peptide somatostatin agonist with sst4 selectivity. J Am Chem Soc 120:1368–1373
Hutton JC, Malaisse WJ 1980 Dynamics of O2 consumption in rat pancreatic islets. Diabetologia 18:395–405
Panten U, Klein H 1982 O2 consumption by isolated pancreatic islets, as measured in a microincubation system with a Clark-type electrode. Endocrinology 111:1595–1600
Ishihara H, Asano T, Tsukuda K, Katagiri H, Inukai K, Anai M, Kikuchi M, Yazaki Y, Miyazaki J, Oka Y 1994 Overexpression of hexokinase I but not GLUT1 glucose transporter alters concentration dependence of glucose-stimulated insulin secretion in pancreatic -cell line MIN6. J Biol Chem 269:3081–3087
Sakura H, Ashcroft SJ, Terauchi Y, Kadowaki T, Ashcroft, FM 1998 Glucose modulation of ATP-sensitive K-currents in wild-type, homozygous and heterozygous glucokinase knock-out mice. Diabetologia 41:654–659
Matschinsky FM 1996 Banting Lecture 1995. A lesson in metabolic regulation inspired by the glucokinase glucose sensor paradigm. Diabetes 45:223–241
Martin F, Andreu E, Rovira JM, Pertusa JA, Raurell M, Ripoll C, Sanchez-Andres JV, Montanya E, Soria B 1999 Mechanisms of glucose hypersensitivity in -cells from normoglycemic, partially pancreatectomized mice. Diabetes 48:1954–1961
Porterfield DM, Corkey RF, Sanger RH, Tornheim K, Smith PJ, Corkey BE 2000 Oxygen consumption oscillates in single clonal pancreatic -cells (HIT). Diabetes 49:1511–1516
Sweet IR, Cook DL, Dejulio E, Wallen AR, Khalil G, Callis J, Reems J 2004 Regulation of ATP/ADP in pancreatic islets. Diabetes 53:401–409
Oliveira HR, Verlengia R, Carvalho CR, Britto LR, Curi R, Carpinelli AR 2003 Pancreatic -cells express phagocyte-like NAD(P)H oxidase. Diabetes 52:1457–1463
Tsubouchi H, Inoguchi T, Inuo M, Kakimoto M, Sonta T, Sonoda N, Sasaki S, Kobayashi K, Sumimoto H, Nawata H 2005 Sulfonylurea as well as elevated glucose levels stimulate reactive oxygen species production in the pancreatic -cell line, MIN6—a role of NAD(P)H oxidase in -cells. Biochem Biophys Res Commun 326:60–65
Detimary P, Jonas JC, Henquin JC 1995 Possible links between glucose-induced changes in the energy state of pancreatic B cells and insulin release. J Clin Invest 96:1738–1745
Smith PA, Rorsman P, Ashcroft FM 1989 Modulation of dihydropyridine-sensitive Ca channels by glucose metabolism in mouse pancreatic -cells. Nature 342:550–553
Eliasson L, Renstrm E, Ding WG, Proks P, Rorsman P 1997 Rapid ATP-dependent priming of secretory granules precedes Ca2+-induced exocytosis in mouse pancreatic B-cells. J Physiol 503:399–412
Hsu WH, Xiang HD, Rajan AS, Kunze DL, Boyd AE 1991 Somatostatin inhibits insulin secretion by a G-protein-mediated decrease in Ca2+ entry through voltage-dependent Ca2+ channels in the cell. J Biol Chem 266:837–843
De Weille JR, Fosset M, Epelbaum J, Lazdunski M 1992 Effectors of ATP-sensitive K+ channels inhibit the regulatory effects of somatostatin and GH-releasing factor on growth hormone secretion. Biochem Biophys Res Commun 187:1007–1004
Ben-Ari Y, Krnjevic K, Crepel V 1990 Activators of ATP-sensitive K+ channels reduce anoxic depolarization in CA3 hippocampal neurons. Neuroscience 37:55–60(Mathew Daunt, Oliver Dale and Paul A. Sm)
Abstract
Somatostatin potently inhibits insulin secretion from pancreatic -cells. It does so via activation of ATP-sensitive K+-channels (KATP) and G protein-regulated inwardly rectifying K+-channels, which act to decrease voltage-gated Ca2+-influx, a process central to exocytosis. Because KATP channels, and indeed insulin secretion, is controlled by glucose oxidation, we investigated whether somatostatin inhibits insulin secretion by direct effects on glucose metabolism. Oxidative metabolism in -cells was monitored by measuring changes in the O2 consumption (O2) of isolated mouse islets and MIN6 cells, a murine-derived -cell line. In both models, glucose-stimulated O2, an effect closely associated with inhibition of KATP channel activity and induction of electrical activity (r > 0.98). At 100 nM, somatostatin abolished glucose-stimulated O2 in mouse islets (n = 5, P < 0.05) and inhibited it by 80 ± 28% (n = 17, P < 0.01) in MIN6 cells. Removal of extracellular Ca2+, 5 mM Co2+, or 20 μM nifedipine, conditions that inhibit voltage-gated Ca2+ influx, did not mimic but either blocked or reduced the effect of the peptide on O2. The nutrient secretagogues, methylpyruvate (10 mM) and -ketoisocaproate (20 mM), also stimulated O2, but this was unaffected by somatostatin. Somatostatin also reversed glucose-induced hyperpolarization of the mitochondrial membrane potential monitored using rhodamine-123. Application of somatostatin receptor selective agonists demonstrated that the peptide worked through activation of the type 5 somatostatin receptor. In conclusion, somatostatin inhibits glucose metabolism in murine -cells by an unidentified Ca2+-dependent mechanism. This represents a new signaling pathway by which somatostatin can inhibit cellular functions regulated by glucose metabolism.
Introduction
GLUCOSE IS THE chief secretagogue for the secretion of insulin from the pancreatic -cell. By its oxidative metabolism, glucose produces an increase in the cytosolic ATP of the -cell, which in turn closes the metabolically sensitive ATP-sensitive K+-channel (KATP) that regulates the plasma-membrane potential. This leads to cell depolarization; opening of voltage-gated Ca2+ channels; and induction of Ca2+-dependent electrical activity, Ca2+ influx, and insulin secretion (1).
The peptide somatostatin is universally accepted to be a ubiquitous inhibitor of stimulus-secretion coupling in a wide variety of neurons, exocrine and endocrine cells (2, 3, 4), with the inhibition of insulin secretion the most studied to date. The major source of somatostatin that affects insulin release in vivo is from enterocytes of the proximal gut; these release the long form of somatostatin (SRIF-28) into the portal circulation in response to alimentation (5). In vivo immunoneutralization studies demonstrate that the gastric release of SRIF-28 reduces postprandial release of insulin to avoid unwanted hypoglycemia and prevent decreases in insulin sensitivity of target tissues, which may result from hyperinsulinemia (5). In addition, the short form of somatostatin (SRIF-14), secreted locally from -cells within the pancreatic islet may also have inhibitory paracrine actions on -cell function (6). Clinically the importance of somatostatin and its analogs are mainly in their therapeutic deployment to inhibit secretion from a wide range of neuroendocrine and exocrine tumors (4, 7, 8). For example, somatostatin and its analogs are used to inhibit the hypersecretion of insulin that occurs with insulinomas, thereby alleviating hyperinsulinemia and associated hypoglycemia (4, 8). More recently somatostatin has been used to treat the hyperinsulinemia associated with hypothalamic obesity and also primary insulin hypersecretion (7, 8), the latter a risk factor for obesity and insulin resistance. It is also becoming apparent that the pharmacology of somatostatin receptors in the -cell changes with age (9). Clearly pharmacological knowledge of the decretin action of somatostatin is crucial in furthering our understanding of the role and therapeutic uses of this peptide and its analogs.
To date, five somatostatin receptor types (sstr1–5) of the G protein-coupled superfamily have been identified, cloned, and characterized. Although widely distributed throughout the body, they are differentially expressed (2, 3, 4). Pancreatic -cells from mouse (10), rat (11), and human (12) islets express all five receptors subtypes. However, data from combinatorial chemical methods, immunocytochemistry, and gene knockout studies strongly support the idea that the inhibition of insulin secretion by somatostatin is mediated predominantly by sstr5 in -cells of mouse (6, 10, 13, 14), rat (11), and man (15).
The major mechanism by which somatostatin inhibits insulin secretion is to decrease Ca2+ influx via a pertussis toxin-sensitive hyperpolarization of the -cell membrane potential (10, 16, 17). This occurs via the opening of two types of K+ channel: KATP (10, 18, 19) and GIRK, a member of the G protein-regulated inward rectifier ion channel family (10, 20). Somatostatin can also inhibit insulin secretion at steps distal to Ca2+ influx: it decreases the levels of cAMP, a permissive sensitizer of insulin release (21), and it inhibits the exocytotic process via direct G protein interactions (22). However, because the inhibition of Ca2+ influx by somatostatin precedes, and precludes, these other inhibitory mechanisms and because the effect of somatostatin via ionic mechanisms has the greatest potency (10), hyperpolarization of the -cell membrane potential is probably the primary mechanism by which the peptide inhibits insulin secretion.
GIRK is activated by direct G protein receptor coupling (20), whereas KATP channel activity is predominantly regulated by metabolism, foremost by changes in the concentrations of intracellular adenine nucleotides (1, 23), although it too can be controlled by direct G protein receptor interactions (18). However, this latter mechanism is unlikely to account for the activation of KATP by somatostatin because direct G protein interactions with the channel are impaired when intracellular adenine nucleotide levels are similar to those associated with glucose-stimulated insulin secretion (18). An alternative, simpler explanation is that somatostatin inhibits glucose metabolism directly and that KATP channels are then activated secondary to the subsequent alterations in intracellular adenine nucleotide concentrations. In support of this idea, somatostatin has been reported to potently (EC50 < 10 nM) inhibit glucose oxidation in rat islets (24), although others have failed to confirm this observation (25). Because the inhibition of glucose oxidation is well established to activate KATP channels (1, 23), we investigated the idea that somatostatin inhibits electrical activity in murine -cells via direct affects on glucose metabolism. To do this, we used polarography to evaluate oxidative respiration (26, 27), fluorescence techniques to monitor the mitochondrial events (28, 29), and electrophysiology to monitor the associated plasma membrane electrical events (10).
Materials and Methods
Cells
For islets studies, male NMRI (National Memorial Research Institute) mice, 3–4 months of age, fed and watered ad libitum, were killed by a cervical dislocation using methods that comply with local guidelines and are approved by the University and U.K. home office. Pancreatic islets were then isolated from excised pancreata by collagenase digestion (26). However, several factors confound interpretation of metabolic measurements made on pancreatic islets: 1) pancreatic islets consist of a multitude of cell types, with the exact contribution of -cells to their overall O2 consumption unknown; 2) paracrine influences can be problematic with isolated islet studies in which normal islet perfusion is disrupted, especially the secretion of somatostatin from pancreatic -cells located within the islet mantle; and 3) large numbers of islets are required. To overcome these problems, we also used the endogenous, insulin-secreting, mouse pancreatic -cell line MIN6 (30), which provided large numbers of cells homogenous in phenotype and free from paracrine influences. MIN6 cells were used with the permission of Prof. Jun-Ichi Miyazaki (Osaka University Medical School, Osaka, Japan) (31). Cells were grown in RPMI 1640 tissue culture media containing 11 mM glucose, 12.5 mM HEPES, and 10% fetal calf serum, kept at 37 C in a humidified atmosphere of 5% CO2-95% air.
O2 consumption
The O2 consumption of islets or MIN6 -cell suspensions of known cell density was measured polarographically using Clark oxygen electrodes (Rank Brothers, Bottisham UK). The partial pressure of O2 (PO2) was measured at a polarographic voltage of –0.6 V with electrodes previously calibrated at 100% air saturation (vigorous gassing with air, 0.25 mM) and 0% (addition of Na2S2O4). All additions were made from stocks in H2O. The control for metabolic substrate addition was 3-O-methyl-glucose (MOG), a metabolically inactive analog of glucose (32). The control for peptide addition was vehicle alone. The rate of oxygen consumption (O2), was measured as the change in PO2 level over a 300-sec period; given that most changes were linear, this is the slope of PO2. For metabolic substrates or Ca2+ channel antagonists, the slope was measured 360 sec after reagent addition, whereas for somatostatin and its receptor agonists, to avoid tachyphylaxis, 60 sec after addition of the drug.
Mitochondrial membrane potential (mit)
To monitor the inner membrane mit we used rhodamine-123 (Rh-123) as previously described (28). Cells were loaded with 50 μg/ml–1 Rh-123 in RPMI 1640 (11 mM glucose) at 37 C for 10 min. The dye was excited at 480 nm and the emitted fluorescence monitored at 530 nm. It is well established that this cationic lipophilic dye is predominantly localized to mitochondria. Depolarization of mit, which would, for example, occur with uncoupling or blocking of mitochondrial respiration, results in the redistribution of the dye across the inner membrane with an associated increase in fluorescence, whereas an increased flux through the mitochondrial respiratory electron transport chain is associated with a decrease in fluorescence (see Fig. 4) (28, 29).
Electrophysiology
To maintain cellular metabolism and second-messenger systems intact, membrane potentials (Vm) were recorded using the perforated-patch patch-clamp technique as described previously (10). Patch pipettes were filled with (in millimoles): 76 K2SO4, 10 KCl, 10 NaCl, 55 sucrose, 10 HEPES (pH 7.2 with KOH), and 1 MgCl2. Perforation was achieved by the addition of 100 μg/ml–1 amphotericin B to the pipette solution and was considered adequate when the series conductance exceeded 40 nS (<25 M).
To mirror the experimental conditions used for the metabolic measurements as closely as possible, KATP channel activity was derived from the current clamp records without recourse to voltage-clamp measurements. If the membrane conductances that contribute to the passive Vm are assumed Ohmic, then the conductance associated with KATP channel activity (GATP) can be estimated from equation 1:
(1)
where Rin is the input resistance of the cell, Vm is the membrane potential, and Ek and El are the reversal potentials for the KATP channel and the leak conductances, respectively. Ek is estimated to be approximately –69 mV, the Vm when KATP is maximally opened by diazoxide; El is estimated to be approximately –39 mV, the interspike Vm when KATP is inhibited by 200 μM tolbutamide. Rin was calculated from the membrane potential deflections that were elicited in response to –10 pA current pulses, 200 msec in duration, applied at 0.1 Hz.
Solutions
To permit integration with previously published electrophysiological data, all experiments were carried out at 32 C in a modified Hanks’ solution that contained (in millimoles): 5.6 KCl, 138 NaCl, 4.2 NaHCO3, 1.2 NaH2PO4, 2.6 CaCl2, 1.2 MgCl2, 10 HEPES (pH 7.4 with NaOH), and 0.1% (wt/vol) BSA. For nominally Ca2+-free solutions, CaCl2 was replaced by equimolar substitution with MgCl2 (total Mg2+ = 3.8 mM, free Ca2+ 10 μM).
SRIF-14 and SRIF-28 were from Peninsula Laboratories (St. Helens, UK), whereas CH-275, BIM-23027, BIM-, L-362, 855, and NNC-26, 9100 were purchased from commercial suppliers of synthetic peptides. The structural integrity and molecular weight of the peptides were confirmed by M@LDI mass spectroscopy (J. Kyte, School of Biomedical Sciences, University of Nottingham). All other drugs and reagents were purchased from Sigma (Poole, UK) or Fluka (Buchs, Switzerland).
Glucose-response relationships
The glucose-response relationships were quantified by best fits of the data with the following equation:
where G is the concentration of glucose, EC50 is the concentration that produces the half-maximal response, Y is the response magnitude, YMAX is the maximal response, YMIN is the minimal response, and h is an index of slope.
Statistics
Unless stated otherwise for statistical comparison of two-sample populations, Wilcoxon signed rank test was used; for multiple comparisons, Kruskal-Wallis test was used with Dunn’s multiple comparison posttest. These procedures were performed using PRISM 3 (GraphPad Software Inc., San Diego, CA). Data statistics are given as means ± SEM, with n as the number of preparations given in parentheses. EC50 and h values are quoted with 95% confidence intervals. Correlation coefficients are Pearson. Statistical significance is defined as P < 0.05 and in graphics is flagged as , when P < 0.01 data are flagged with and when P < 0.001 data are flagged with .
Results
Simulation of O2 consumption by glucose and its inhibition by somatostatin
In the absence of added substrate, both islets and MIN6 possessed a linear basal respiratory rate, within 1 min of adding 20 mM glucose O2 increased by approximately 63% and approximately 80%, respectively (Fig. 1). Subsequent addition of SRIF-14 significantly inhibited O2 in both islets and MIN6 cells (Fig. 1). In some cases the inhibition of O2 was greater than that stimulated by the sugar, indicative of inhibition of the basal respiration. The inhibitory effect of the peptide was sustained for at least 10 min, the longest period tested. In control experiments, addition of H2O the vehicle for somatostatin was either without effect or slightly stimulated O2 (Figs. 1B and 2D). The inhibitory action of SRIF-14 was independent of the time of addition; the peptide had a comparable inhibitory effect when added 60 min after glucose. These data also confirm the functional integrity of the -cells over time and lack of cellular damage due to stirring. The protonophores, carbonyl cyanide 3-chlorophenylhydrazone (CCCP, 100 nM) and carbonyl-cyanide-p-trifluoro-methoxy-phenylhydrazone (FCCP; 100 nM), dramatically increased O2 (Fig. 1, C and D), effects consistent with their action as uncouplers of mitochondrial oxidative phosphorylation. Subsequent application of 3 mM azide, a blocker of cytochrome oxidase, partly blocked O2 consumption in both islets and MIN6 cells (Fig. 3A). These latter two experimental paradigms also confirm that the majority of O2 consumption measured reflects mitochondrial respiration. For reasons explained in Materials and Methods, all further experimentation solely used MIN6 cells.
Glucose stimulated O2 in a concentration-dependent manner with an EC50 of 0.27 mM (0.21–0.33 mM) and h of 1 (Fig. 5D, solid line). The control for glucose, MOG (20 mM), a glucose analog that is taken up but not metabolized by -cells (32), did not stimulate O2 consumption (Fig. 2A), an observation consistent with the inability of the analog to stimulate the insulin-secretion pathway. Instead, however, it actually blocked the basal rate of O2 consumption by 28 ± 9% (n = 14; P < 0.01). To further explore the metabolic pathways affected by somatostatin, two other metabolic substrates that stimulate insulin secretion via entry into different steps of the oxidation pathway for glucose were used. Methylpyruvate (10 mM), a membrane-permeable form of the mitochondrial fuel pyruvate (33), and -ketoisocaproate (20 mM), a fuel that feeds directly into the mitochondrial tricarboxylic acid cycle (TCA) (34), stimulated O2 by amounts approaching that produced by 20 mM glucose (Fig. 2A). However, 100 nM SRIF-14 had no effect on O2 stimulated by either of these two alternative mitochondrial substrates (Fig. 2D). Moreover, 100 nM SRIF-14 also failed to inhibit O2 stimulated by glucose concentrations less than 5 mM (Fig. 2D). Surprisingly, the peptide did, however, cause a further 30 ± 4% inhibition of basal respiration (P < 0.007, n = 16) when it was already suppressed by 20 mM MOG. SRIF-14 had on effect on O2 in cell suspensions that failed to respond to glucose.
Ca2+ dependency of oxygen consumption and its inhibition by somatostatin
Because evidence exists that suggests intracellular Ca2+ regulates glucose oxidation in -cells (35), we explored the possibility that the inhibitory effect of somatostatin we observed on respiration was actually secondary to the decrease in Ca2+ influx (16) that occurs with the inhibition of electrical activity produced by the peptide (10). Removal of extracellular Ca2+ had little effect on glucose-stimulated O2 consumption (Fig 2B). Similarly, blockade of voltage-gated Ca2+ influx with 5 mM Co2+ or 20 μM nifedipine, a dihydropyridine-sensitive L-type Ca2+-channel antagonist, also had no effect on O2. Under conditions in which Ca2+ influx was impaired by Co2+ or nifedipine, 100 nM SRIF-14 still inhibited O2 (Fig. 2, B and C) but to a lesser extent than seen under control conditions. In the absence of extracellular Ca2+, SRIF-14 failed to have a significant inhibitory effect on O2.
cAMP dependency of oxygen consumption and its inhibition by somatostatin
To investigate whether a decrease in cytoplasmic cAMP concentration is involved in the inhibition of O2 consumption by somatostatin, the effect of the peptide was explored in cells preincubated with the stable membrane-permeable cAMP analog 8-(4-chlorophenylthio)-adenosine 3',5'-cyclic monophosphate (8-CPT). Incubation in 50 μM 8-CPT, a concentration that also inhibits cAMP-specific phosphodiesterases, did not change O2 (–5.3 ± 1.4, compared with –4.6 ± 1.1 nmol 107 cells–1 min–1 in control, n = 10). 8-CPT did not affect the ability of SRIF-14 to inhibit the O2 stimulated by glucose: 100 nM SRIF-14 inhibited O2 by 50 ± 10% to –2.4 ± 0.7 nmol 107 cells–1 min–1 in the presence of 8-CPT (P < 0.05 repeated-measure ANOVA), compared with 49 ± 13% inhibition in vehicle alone (n = 9).
Identification of somatostatin receptor subtype
All five sstr subtypes have been identified in murine pancreatic -cells, including MIN6 cells (3, 10). Figure 3E shows that L-362, 855, a partial agonist at the sstr5 (36, 37), partly mimicked the action of SRIF-14 (n = 17). L-362, 855 inhibited O2 by 51%, an effect less pronounced than that produced by SRIF-14 (81% Fig. 2, D and E). CH-275, BIM-23056, BIM-23027, and NNC-26, 9100, specific agonists at the sstr1 (38), sstr2/3 (37, 39), sstr3 (37, 39), and sstr4 (40), respectively, were all without effect on O2 (n = 12–13; Fig. 2E).
Oxygen consumption by nicotinamide adenine dinucleotide phosphate [NAD(P)H] oxidase and its inhibition by somatostatin
Figure 3 shows that SRIF-14 had no effect on O2 in the combined presence of glucose and azide. Application of 0.5 mM apocynin, an inhibitor of NAD(P)H oxidase, significantly decreased O2 by 72 ± 4% (n = 7, Fig. 3D), an amount greater than that produced by 0.5% (vol/vol) (70 μM) dimethylsulfoxide (DMSO), the vehicle control (25 ± 5%, n = 8, Fig. 3C), although the difference between these effects was insignificant. The block produced by 100 nM SRIF-14 in the presence of apocynin (12 ± 7%, n = 7) was less than that produced in the presence of vehicle alone (34 ± 8%, n = 8); however, this difference was insignificant. The final overall block of O2 was similar in both cases: 84 ± 5%, with 0.5 mM apocynin plus 100 nM SRIF-14 (n = 7), and 59 ± 11%, with 70 μM DMSO plus 100 nM SRIF-14 (n = 8). Addition of 1 μM diphenyleneiodonium chloride, a nonselective inhibitor of flavoproteins, which include NAD(P)H oxidases, abolished O2 (Fig. 3B).
Effect of glucose and somatostatin on mit
Addition of 5 mM glucose to MIN-6 cells, previously incubated for 20 min in the absence of exogenous metabolic substrate, produced a sustained decrease in Rh-123 fluorescence (Fig. 4A). The effect was complete with 1 min of sugar addition and amounted to an approximately 20% drop in fluorescence (Fig. 4B), a result consistent with the redistribution and subsequent quench of the dye that occurs with its accumulation by energized mitochondria (28, 29). Addition of 100 nM SRIF-14 in the maintained presence of glucose led to a partial reversal of the change in Rh-123 fluorescence produced by the sugar. This effect was slower than that elicited by glucose, with the maximum change observed approximately 2 min after addition of the peptide, which amounted to 38 ± 8% inhibition of the glucose-sensitive change in Rh-123 fluorescence (n = 6). Subsequent addition of 1 μM CCCP resulted in an approximately 100% increase in Rh-123 fluorescence (Fig. 4B), a result consistent with the redistribution and loss of quenching of the dye that occurs with the collapse of mit produced by this protonophoric uncoupler of mitochondrial oxidative phosphorylation (28).
Electrophysiology
Elevation of glucose from 0 to 10 mM depolarized the Vm and evoked Ca2+-dependent action potential activity; the latter event is well established to promote Ca2+ influx (1, 16). In the presence of 10 mM glucose, 100 nM SRIF-14, 3 mM azide, 20 μM nifedipine, or removal of extracellular Ca2+ all abolished the action potential activity induced by the sugar and, with the exception of the dihydropyridine, hyperpolarized the membrane potential (data not shown): electrophysiological changes associated with a decrease in Ca2+-influx and insulin-secretion (1, 16).
The steady-state relationships among the concentration of glucose, Vm, Rin, and the whole-cell conductance of KATP (GATP) are explored in Fig. 5. Glucose depolarized Vm with an EC50 of 1.3 mM (0.9–1.7 mM) and a slope coefficient, h, of 1.2 (0.9–1.7, solid line, Fig. 5A). This effect was mirrored by a parallel increase (r = 1, P < 0.0001) in the Rin; EC50 of 1.4 mM glucose (0.5–3.7 mM), and h of 1.0 (0.4–1.7, Fig. 5B, solid line). Of the 21 -cells that depolarized in response to glucose, 17 depolarized sufficiently to reach the voltage threshold for action potentials (–50 ± 2 mV, n = 17) and evoke electrical activity. In these cells, electrical activity was elicited at a median glucose concentration of 2.5 mM (1–3.4, 95% confidence limits). The relationship between glucose concentration and the percentage of cells that responded with electrical activity is drawn in Fig. 5C: EC50 of 1.2 mM glucose (0.9–1.5 mM) and h of 1.8 (1.1–2.6, Fig. 5C, solid line).
As expected from Rin, GATP decreased with increased glucose concentration (Fig. 5D). The relationship of GATP with glucose concentration (closed symbols) closely mirrored that obtained for the rate of O2 consumption (open symbols): EC50 of 0.3 mM (0.2–0.4 mM) and h of 1.8 (0.6–2.9, Fig. 5D, solid line). The close correlation between GATP and the rate of O2 consumption is supported by the clear inverse linear relationship (r > 0.98, P < 0.0001) that exists between these two variables (Fig. 5E).
Discussion
We have unequivocally shown that SRIF-14 inhibits O2 stimulated by glucose concentrations that evoke electrical activity and stimulate insulin secretion in both mouse islets and murine-derived -cells. These findings support and corroborate the original observation made in rat islets (24). One possible reason for the failure of others (25) to confirm this observation in rat may be simply explained by the small sample size employed in that study (i.e. n = 3).
Substrate dependency of O2 consumption
The 2-fold stimulation in the rate of O2 consumption produced by 20 mM glucose for both -cell models are comparable with that previously demonstrated for MIN6 cells (27) and -cells within intact rodent islets (26, 41). Indeed the absolute values for O2 consumption stimulated by 20 mM glucose are comparable with those already published: 10 pmol O2 islet–1 min–1, compared with approximately 4 pmol O2 islet–1 min–1 for rat islets (41) and 5–10 nmol O2 107 cell–1 min–1, compared with approximately 10 nmol O2 107 cell–1 min–1 for MIN6 (27). Assuming an islet possesses approximately 1000 -cells, then 20 mM glucose stimulates O2 consumption by approximately 40 nmol O2 107 cell–1 min–1, an amount similar to that measured for the cell line. Methylpyruvate and -ketoisocaproate are insulinotropes metabolized predominantly within the mitochondria via the TCA cycle (33, 34). They are thought to act via the same metabolic mechanism engaged by the stimulus-secretion pathway of glucose: generation of ATP and closure of KATP with the resultant electrical activity and insulin secretion (1, 33, 34), an idea supported here in MIN6 by the ability of these substrates to stimulate oxidative respiration (41, 42), with -ketoisocaproate also known to stimulate insulin secretion in MIN6 cells (43).
The EC50 for glucose-stimulated O2 (0.27 mM) is substantially less than that measured for mouse islets: 8 mM (42). The phosphorylation of glucose to glucose-6-phosphate is well established to be the key rate-limiting step for the use of the sugar and subsequent stimulus-secretion cascade for insulin release in the pancreatic -cell (44, 45). In native -cells, phosphorylation occurs predominantly via a low-affinity glucokinase (Km 5–10 mM). Consequently many events of the stimulus-secretion pathway for insulin in native -cells possess EC50 values for glucose around 5–10 mM (42, 44, 45, 46). The low millimolar values we obtained for the EC50s of various steps within the stimulus-secretion pathway in MIN6 suggest that the activity of a high-affinity hexokinase (Km 0.3 mM) predominates over that of glucokinase (43), an idea supported by the increased sensitivities to glucose for the inhibition of GATP and excitation of electrical activity, compared with those found in native mouse -cells [0.29 vs.3.2 mM (44) and 1.2 vs. 11.5 mM (46), respectively]. Nevertheless, the marked correlation observed among the stimulation of O2 consumption, inhibition of GATP, and induction of electrical activity by glucose strongly supports the idea that oxidation of the sugar in MIN6 results in closure of the KATP channel like that occurring in native -cells (1, 44, 46). The inhibition of basal respiration by 3-O-methyl-D-glucose was surprising, given that this glucose analog is not thought to be metabolized; however, our finding is consistent with its known adverse affect on -cell metabolism at low glucose concentrations (32).
Ca2+ dependence of O2 consumption and its inhibition by SRIF-14
Whether extracellular Ca2+ affects glucose metabolism in pancreatic -cell is disputable; some studies report effects on respiration (47) and glucose oxidation (35), whereas others do not (41). We found blockade of Ca2+ influx did not affect O2 consumption and did not mimic the effect of SRIF-14. However, the subsequent inhibition of O2 by the peptide was abolished in the absence of extracellular Ca2+ and was reduced when voltage-gated Ca2+ influx was blocked with Co2+ or nifedipine. These data suggest that the mechanism by which the peptide decreases O2 is not secondary to, but requires, Ca2+ influx; with the associated increase in cytosolic Ca2+ a permissive factor, an idea supported by the fact that SRIF-14 was most effective at glucose concentrations that elicited electrical activity and voltage-dependent Ca2+ influx (cf. Figs. 2D and 5C). Because somatostatin had no effect on O2 stimulated by the mitochondrial fuels, methylpyruvate or -ketoisocaproate, SRIF-14 must inhibit metabolism at a step/s before the TCA cycle. The observation that SRIF-14 inhibited basal respiration only when it was already suppressed by the presence of 20 mM 3-O-methyl-D-glucose, suggests that the molecular action of the peptide may also involve the presence and/or transport of hexose.
The inability of somatostatin to affect O2 stimulated by the mitochondrial fuels refutes the idea that the effect of the peptide is due to changes in the respiratory control of oxidative phosphorylation: changes that arise from the inhibition of secretion previously stimulated by these substrates and the resultant attenuation in energy demand (48). Because we also show that somatostatin depolarizes mit in the presence of glucose, these data similarly refute the idea that the decrease in O2 produced by the peptide is due to the drop in the energetic requirements of the -cell that results from the associated inhibition of the exocytotic process. In fact, the increase in the cytosolic ATP to ADP ratio that is predicted on inhibition of the exocytotic process is expected to inhibit KATP channel activity and promote polarization of mit (29), neither of which are seen; instead the opposite is observed with the peptide (Ref.10 and present study, respectively). Overall the Rh-123 data further support the notion that the peptide directly inhibits oxidative phosphorylation at some unknown step before the TCA cycle.
The inability of 8-CPT, a stable membrane-permeable cAMP analog, to affect the inhibition of O2 caused by somatostatin refutes the idea that a decrease in cytosolic cAMP (21) is involved in this particular effect of the peptide. In fact, decreases in cAMP levels are classically associated with stimulation, not inhibition, of glucose catabolism. The identity of the molecular target/s for this inhibitory action of somatostatin on glucose stimulated O2 remain/s unknown.
The pharmacological profile that was observed for the inhibition of O2 by the sstr-selective analogs is consistent with the peptide prosecuting its affects solely through binding to sstr5, the same receptor isoform that mediates activation of the KATP channel (10) and inhibition of insulin secretion in mouse (6, 10, 13, 14), rat (11), and man (15). The observation that neither NNC-26, 9100 nor CH-275, agonists specific for the sstr4 and sstr1, respectively, affected O2 rules out participation of these receptor subtypes in the inhibition of O2 by the peptide. Furthermore, the inability of BIM-23027 and BIM-23056, agonists more potent at the sstr2 and sstr3 than L-362, 855 (39), to inhibit O2 argues against an involvement of these receptor subtypes. The limited ability of L-362, 855 to mimic the effect of SRIF-14 is consistent with its action as a partial agonist at sstr5 (36).
Oxygen consumption NAD(P)H oxidase and its inhibition by somatostatin
The observation that azide did not always abolish O2 consumption necessitates an additional mechanism of O2 reduction in addition to that attributed to the mitochondrial electron transport chain. Reduction and consumption of O2 can also occur by NAD(P)H oxidase activity and superoxide production. RT-PCR and Western blotting have identified a range of NAD(P)H oxidase isoforms expressed within pancreatic islet tissue. Of these, the p47PHOX isoform is predominantly expressed by the pancreatic -cell at the plasma membrane (49, 50). The observation that apocynin, an inhibitor of NAD(P)H oxidases, did not significantly decrease O2, compared with vehicle control (DMSO), does not appear to support the idea that some O2 is via activity of this particular enzyme. However, such an interpretation of these data must be treated with caution because DMSO is itself an antioxidant, which may affect NAD(P)H oxidase activity; a fact that confounds whether this enzyme contributes to O2. Also, the finding that somatostatin could still inhibit the remaining O2 after the effects of apocynin suggests that the peptide inhibits O2 by mechanism/s other that that involving NAD(P)H oxidase activity. Furthermore, because the blockade of mitochondrial reduction of O2 by azide, which increases the availability of NAD(P)H for oxidation by NAD(P)H oxidases, did not enhance the effect of somatostatin on O2 but in fact abolished it, also suggests that the peptide does not affect NAD(P)H oxidase activity. The observation that diphenyleneiodonium chloride, an inhibitor of NAD(P)H oxidases, abolished O2 is attributed to its action as a potent inhibitor of most flavoproteins, which includes complex I of the mitochondrial respiratory electron transport chain. Together, these data do not readily support an involvement of NAD(P)H oxidases in either O2 consumption of pancreatic -cells or the effect of somatostatin on O2.
Physiological implications
The inhibition of glucose metabolism by somatostatin is expected to decrease cytosolic ATP of the -cell and contribute to the pleiotropic inhibitory mechanisms of the peptide on insulin secretion. Unfortunately, due to compartmentalization of ATP within the -cell, this idea is difficult to test directly (51); however, the very fact that the activity of KATP channels reflects intracellular nucleotide levels lends strong support to the above idea (1, 23). In addition to the observed activation of KATP channels (10, 19), a decrease in cytosolic ATP is expected to inhibit both Ca2+ channel activity and exocytosis, processes in the -cell enhanced by glucose metabolism (52, 53) but also inhibited by somatostatin (54), although these may also involve direct G protein pathways. Involvement of a direct G protein pathway in the activation of the KATP channel by somatostatin is unlikely because direct activation of KATP by G-proteins is impaired when intracellular ATP is elevated to levels that are associated with glucose-stimulated insulin secretion (18). Consequently the inhibition of metabolism by somatostatin is a simple mechanism to explain the activation of KATP channels by the peptide that occurs during glucose-stimulated insulin secretion. Whether such a similar simple mechanism operates in other cell types in which KATP is activated by somatostatin, e.g. in pituitary gonadotrophs (55) and CA3 hippocampal neurons (56), remains to be tested.
In conclusion, we have described a novel Ca2+-dependent signaling mechanism, namely a decrease in glucose metabolism, which enables somatostatin via activation of sstr5 to inhibit insulin secretion from the murine pancreatic -cell at a step early in its stimulus-secretion coupling cascade.
Acknowledgments
The authors thank the Royal Society and the Institute for Cell Signaling for support.
Footnotes
The authors have no conflict of interest.
First Published Online December 15, 2005
Abbreviations: CCCP, Carbonyl cyanide 3-chlorophenylhydrazone; 8-CPT, 8-(4-chlorophenylthio)-adenosine 3',5'-cyclic monophosphate; DMSO, dimethylsulfoxide; GATP, KATP channel activity; GIRK, member of the G protein-regulated inward rectifier ion channel family; KATP, ATP-sensitive K+-channels; mit, mitochondrial membrane potential; MOG, 3-O-methyl-glucose; NAD(P)H, nicotinamide adenine dinucleotide phosphate; O2, O2 consumption; PO2, partial pressure of O2; Rh-123, rhodamine-123; Rin, input resistance of the cell; SRIF-14, short form of somatostatin; SRIF-28, long form of somatostatin; sstr, somatostatin receptor; TCA, tricarboxylic acid cycle; Vm, membrane potential.
Accepted for publication December 1, 2005.
References
Ashcroft FM, Rorsman P 1989 Electrophysiology of the pancreatic -cell. Prog Biophys Mol Biol 54:87–143
Csaba Z, Dournaud P 2001 Cellular biology of somatostatin receptors. Neuropeptides 35:1–23
Reisine T, Bell GI 1995 Molecular properties of somatostatin receptors. Neuroscience 67:777–790
Weckbecker G, Lewis I, Albert R, Schmid HA, Hoyer D, Bruns C 2003 Opportunities in somatostatin research: biological, chemical and therapeutic aspects. Nat Rev Drug Discov 2:999–1017
Ensinck JW, Vogel RE, Laschansky EC, Koerker DJ, Prigeon RL, Kahn SE, D’Alessio DA 1997 Endogenous somatostatin-28 modulates postprandial insulin secretion. Immunoneutralization studies in baboons. J Clin Invest 100:2295–2302
Tirone TA, Norman MA, Moldovan S, DeMayo FJ, Wang XP, Brunicardi FC 2003 Pancreatic somatostatin inhibits insulin secretion via SSTR-5 in the isolated perfused mouse pancreas model. Pancreas 26:67–73
Boehm BO 2003 The therapeutic potential of somatostatin receptor ligands in the treatment of obesity and diabetes. Exp Opin Investig Drugs 12:1501–1509
Hansen JB, Arkhammar PO, Bodvarsdottir TB, Wahl P 2004 Inhibition of insulin secretion as a new drug target in the treatment of metabolic disorders. Curr Med Chem 11:1595–1615
Wang XP, Yang J, Norman MA, Magnusson J, DeMayo FJ, Brunicardi FC 2005 SSTR5 ablation in islet results in alterations in glucose homeostasis in mice. FEBS Lett 579:3107–3114
Smith PA, Sellers LA, Humphrey PPA 2001 Somatostatin activates two types of inwardly rectifying K+ channels in MIN6 cells. J Physiol 532:127–142
Mitra SW, Mezey E, Hunyady B, Chamberlain L, Hayes E, Foor F, Wang YN 1999 Colocalization of somatostatin receptor sst5 and insulin in rat pancreatic -cells. Endocrinology 140:3790–3796
Kumar U, Sasi R, Suresh S, Patel A, Thangaraju M, Metrakos P, Patel SC, Patel YC 1999 Subtype-selective expression of the five somatostatin receptors (hSSTR1–5) in human pancreatic islet cells: a quantitative double-label immunohistochemical analysis. Diabetes 48:77–85
Fagan SP, Azizzadeh A, Moldovan S, Ray MK, Adrian TE, Ding XZ, Coy DH, Brunicardi FC 1998 Insulin secretion is inhibited by subtype five somatostatin receptor in the mouse. Surgery 124:254–258
Strowski MZ, Kohler M, Chen HY, Trumbauer ME, Li Z, Szalkowski D, Gopal-Truter S, Fisher JK, Schaeffer JM, Blake AD, Zhang BB, Wilkinson HA 2003 Somatostatin receptor subtype 5 regulates insulin secretion and glucose homeostasis. Mol Endocrinol 17:93–106
Zambre Y, Ling Z, Chen MC, Hou X, Woon CW, Culler M, Taylor JE, Coy DH, Van Schravendijk C, Schuit F, Pipeleers DG, Eizirik DL 1999 Inhibition of human pancreatic islet insulin release by receptor-selective somatostatin analogs directed to somatostatin receptor subtype 5. Biochem Pharmacol 57:1159–1164
Nilsson T, Arkhammar P, Rorsman P, Berggren PO 1989 Suppression of insulin release by galanin and somatostatin is mediated by a G-protein. An effect involving repolarization and reduction in cytoplasmic free Ca2+ concentration. J Biol Chem 264:973–980
Pace CS, Tarvin JT 1981 Somatostatin: mechanism of action in pancreatic islet -cells. Diabetes 30:836–842
Ribalet B, Eddlestone GT 1995 Characterization of the G-protein coupling of a somatostatin receptor to the K-ATP channel in insulin-secreting mammalian HIT and RIN cell lines. J Physiol 485:73–86
de Weille JR, Schmid-Antomarchi H, Fosset M, Lazdunski M 1989 Regulation of ATP-sensitive K+ channels in insulinoma cells: activation by somatostatin and protein kinase C and the role of cAMP. Proc Natl Acad Sci USA 86:2971–2975
Yamada M, Inanobe A, Kurachi Y 1998 G protein regulation of potassium ion channels. Pharm Rev 50:723–760
Schuit FC, Derde MP, Pipeleers DG 1989 Sensitivity of rat pancreatic and cells to somatostatin. Diabetologia 32:207–212
Ullrich S, Prentki M, Wollheim CB 1990 Somatostatin inhibition of Ca2+-induced insulin secretion in permeabilized HIT-T15 cells. Biochem J 270:273–276
Tarasov A, Dusonchet J, Ashcroft F 2004 Metabolic regulation of the pancreatic -cell ATP-sensitive K+ channel: a pas de deux. Diabetes 53:S113–S122
Hahn HJ, Gottschling HD 1976 Somatostatin-induced inhibition of insulin secretion by isolated pancreatic rat islets prepared by micro-dissection or collagenase digestion. Diabetes Metab 2:107–111
Gorus FK, Malaisse WJ, Pipeleers DG 1984 Differences in glucose handling by pancreatic A- and B-cells. J Biol Chem 259:1196–1200
Hellerstrm C 1967 Effects of carbohydrates on the O2 consumption of isolated pancreatic islets of mice. Endocrinology 81:105–112
Soejima A, Inoue K, Takai D, Kaneko M, Ishihara H, Oka Y, Hayashi JI 1996 Mitochondrial DNA is required for regulation of glucose-stimulated insulin secretion in a mouse pancreatic cell line, MIN6. J Biol Chem 271:26194–26199
Duchen MR, Smith PA, Ashcroft FM 1993 Substrate-dependent changes in mitochondrial function, intracellular free-calcium concentration and membrane channels in pancreatic -cells. Biochem J 294:35–42
Baracca A, Sgarbi G, Solaini G, Lenaz G 2003 Rhodamine 123 as a probe of mitochondrial membrane potential: evaluation of proton flux through F(0) during ATP synthesis. Biochim Biophys Acta 1606:137–146
Ishihara H, Asano T, Tsukuda K, Katagiri H, Inukai K, Anai M, Kikuchi M, Yazaki Y, Miyazaki J, Oka Y 1993 Pancreatic cell line MIN6 exhibits characteristics of glucose metabolism and glucose-stimulated insulin secretion similar to those of normal islets. Diabetologia 36:1139–1145
Miyazaki J, Araki K, Yamato E, Ikegami H, Asano T, Shibasaki Y, Oka Y, Yamamura K-I 1990 Establishment of a pancreatic cell line that retains glucose-inducible insulin secretion: special reference to expression of glucose transporter isoforms. Endocrinology 127:126–132
Sener A, Scruel O, Louchami K, Jijakli H, Malaisse WJ 1999 Inhibition of glucose-induced insulin release by 3-O-methyl-D-glucose: enzymatic, metabolic and cationic determinants. Mol Cell B 194:133–145
Jijakli H, Nadi AB, Cook L, Best L, Sener A, Malaisse WJ 1996 Insulinotropic action of methyl pyruvate: enzymic and metabolic aspects. Arch Biochem Biophys 335:245–257
Gao Z, Young RA, Li G, Najafi H, Buettger C, Sukumvanich SS, Wong RK, Wolf BA, Matschinsky FM 2003 Distinguishing features of leucine and -ketoisocaproate sensing in pancreatic -cells. Endocrinology 144:1949–1957
Hellman B, Idahl LA, Lernmark A, Sehlin J, Taljedal IB 1974 The pancreatic -cell recognition of insulin secretagogues. Effects of calcium and sodium on glucose metabolism and insulin release. Biochem J 138:33–45
Carruthers A, Warner AJ, Michel AD, Feniuk W, Humphrey PPA 1999 Activation of adenylate cyclase by human recombinant sst5 receptors expressed in CHO-K1 cells and involvement of Gs proteins. Br J Pharmacol 126:1221–1229
Raynor K, O’Carroll AM, Kong H, Yasuda K, Mahan LC, Bell GI, Reisine T 1993 Characterization of cloned somatostatin receptors sstr4 and sstr5. Mol Pharm 44:385–392
Liapakis G, Hoeger C, Rivier J, Reisine T 1996 Development of a selective agonist at the somatostatin receptor subtype sstr1. J Pharm Exp Ther 276:1089–1094
Raynor K, Murphy WA, Coy DH, Taylor JE, Moreau J-P, Yasuda K, Bell GI, Reisine T 1993 Cloned somatostatin receptors: identification of subtype-selective peptides and demonstration of high affinity binding of linear peptides. Mol Pharm 43:838–844
Ankersen M, Crider M, Liu S, Ho B, Andersen HS, Stidsen C 1998 Discovery of a novel non-peptide somatostatin agonist with sst4 selectivity. J Am Chem Soc 120:1368–1373
Hutton JC, Malaisse WJ 1980 Dynamics of O2 consumption in rat pancreatic islets. Diabetologia 18:395–405
Panten U, Klein H 1982 O2 consumption by isolated pancreatic islets, as measured in a microincubation system with a Clark-type electrode. Endocrinology 111:1595–1600
Ishihara H, Asano T, Tsukuda K, Katagiri H, Inukai K, Anai M, Kikuchi M, Yazaki Y, Miyazaki J, Oka Y 1994 Overexpression of hexokinase I but not GLUT1 glucose transporter alters concentration dependence of glucose-stimulated insulin secretion in pancreatic -cell line MIN6. J Biol Chem 269:3081–3087
Sakura H, Ashcroft SJ, Terauchi Y, Kadowaki T, Ashcroft, FM 1998 Glucose modulation of ATP-sensitive K-currents in wild-type, homozygous and heterozygous glucokinase knock-out mice. Diabetologia 41:654–659
Matschinsky FM 1996 Banting Lecture 1995. A lesson in metabolic regulation inspired by the glucokinase glucose sensor paradigm. Diabetes 45:223–241
Martin F, Andreu E, Rovira JM, Pertusa JA, Raurell M, Ripoll C, Sanchez-Andres JV, Montanya E, Soria B 1999 Mechanisms of glucose hypersensitivity in -cells from normoglycemic, partially pancreatectomized mice. Diabetes 48:1954–1961
Porterfield DM, Corkey RF, Sanger RH, Tornheim K, Smith PJ, Corkey BE 2000 Oxygen consumption oscillates in single clonal pancreatic -cells (HIT). Diabetes 49:1511–1516
Sweet IR, Cook DL, Dejulio E, Wallen AR, Khalil G, Callis J, Reems J 2004 Regulation of ATP/ADP in pancreatic islets. Diabetes 53:401–409
Oliveira HR, Verlengia R, Carvalho CR, Britto LR, Curi R, Carpinelli AR 2003 Pancreatic -cells express phagocyte-like NAD(P)H oxidase. Diabetes 52:1457–1463
Tsubouchi H, Inoguchi T, Inuo M, Kakimoto M, Sonta T, Sonoda N, Sasaki S, Kobayashi K, Sumimoto H, Nawata H 2005 Sulfonylurea as well as elevated glucose levels stimulate reactive oxygen species production in the pancreatic -cell line, MIN6—a role of NAD(P)H oxidase in -cells. Biochem Biophys Res Commun 326:60–65
Detimary P, Jonas JC, Henquin JC 1995 Possible links between glucose-induced changes in the energy state of pancreatic B cells and insulin release. J Clin Invest 96:1738–1745
Smith PA, Rorsman P, Ashcroft FM 1989 Modulation of dihydropyridine-sensitive Ca channels by glucose metabolism in mouse pancreatic -cells. Nature 342:550–553
Eliasson L, Renstrm E, Ding WG, Proks P, Rorsman P 1997 Rapid ATP-dependent priming of secretory granules precedes Ca2+-induced exocytosis in mouse pancreatic B-cells. J Physiol 503:399–412
Hsu WH, Xiang HD, Rajan AS, Kunze DL, Boyd AE 1991 Somatostatin inhibits insulin secretion by a G-protein-mediated decrease in Ca2+ entry through voltage-dependent Ca2+ channels in the cell. J Biol Chem 266:837–843
De Weille JR, Fosset M, Epelbaum J, Lazdunski M 1992 Effectors of ATP-sensitive K+ channels inhibit the regulatory effects of somatostatin and GH-releasing factor on growth hormone secretion. Biochem Biophys Res Commun 187:1007–1004
Ben-Ari Y, Krnjevic K, Crepel V 1990 Activators of ATP-sensitive K+ channels reduce anoxic depolarization in CA3 hippocampal neurons. Neuroscience 37:55–60(Mathew Daunt, Oliver Dale and Paul A. Sm)