|
|
||||||||
RESEARCH COMMUNICATION |
a Department of Science and Technology and Institute of Bioengineering, Campus de San Juan, Miguel Hernandez University, Alicante, Spain
b Department of Biochemistry, Veterinary Faculty, Complutense University, Madrid, Spain
c Department of Physiology and Institute of Bioengineering, Campus de San Juan, Miguel Hernandez University, Alicante, Spain
| ABSTRACT |
|---|
|
|
|---|
Key Words: KATP channel pancreatic ß-cell stimulus-secretion coupling diabetes
| INTRODUCTION |
|---|
|
|
|---|
The current hypothesis indicates that, in pancreatic ß-cells, KATP channels act as metabolic sensors and close as a consequence of glucose metabolism, which raises the intracellular ATP-to-ADP ratio (8). KATP channel inhibition depolarizes the cell membrane, leading to the activation of voltage-dependent Ca2+ channels and to an ensuing rise in the free intracellular concentration of Ca2+ ([Ca2+]i) (9, 10), which then initiates the release of insulin (11). In a previous study, we demonstrated that glucose challenge of pancreatic ß-cells increased intracellular concentrations of ApnA by 30- to 50-fold (7). Remarkably, glucose raises ApnA values from ineffective cytosolic concentrations to values sufficient to close KATP channel activity, with an efficacy similar to that observed for intracellular ATP ([ATP]i).2 Thus, [ATP]i is found at millimolar concentrations, far higher than needed to inhibit channel activity, whereas ApnA concentrations are in the micromolar range, much closer to the EC50 for both ATP and ApnA, and in the suitable concentration interval that effectively regulates the channel. Despite the increasing physiological relevance of ApnA at the intracellular level, knowledge of the mechanisms through which ApnA cellular concentrations are regulated in mammalian cells is restricted to a study showing that cytosolic concentrations of dinucleotide polyphosphates must be under strict control and a description of the enzymes able to degrade ApnA (12). However, nothing is known about the metabolic pathways by which ApnA are synthesized in pancreatic ß-cells or the link between nutrient metabolism and ApnA increase. Moreover, a precise role for ApnA-induced KATP channel inhibition remains to be established.
The aim of this study was to identify the metabolic pathways that contribute to nutrient-induced ApnA synthesis in vivo and the role this ApnA output plays in the regulation of ß-cell functional activity.
| MATERIALS AND METHODS |
|---|
|
|
|---|
-ketoisocaproate (KIC), iodoacetate (iodoacetic acid sodium salt), fluoroacetate (phenylmethylsulfonyl fluoride), tolbutamide, and sodium fluoride (NaF) were obtained from Sigma (St. Louis, Mo.). Alkaline phosphatase (EC 3.1.3.1.) and phosphodiesterase (EC 3.1.15.1) were purchased from Pharmacia (Uppsala, Sweden). Glucose and KCl were from Panreac (Barcelona, Spain).
Cell isolation and culture
Islets of Langerhans were isolated as previously described (13). Briefly, after pancreas digestion with collagenase in a stationary bath at 37°C, islets were separated by centrifugation and hand picked under a stereomicroscope. Once isolated, islets were dissociated into single cells in a low-calcium medium as previously described (7). Cells were then centrifuged, resuspended in culture medium RPMI 1640 supplemented with 10% fetal calf serum, penicillin (100 IU/ml), streptomycin (0.1 mg/ml), and 5.6 mM glucose, and plated on glass coverslips. Cells were kept at 37°C in a humidified atmosphere of 95% O2 and 5% CO2, and used within 13 days of culture.
ApnA and ATP studies
After collagenase isolation, islets were incubated for 1 h at 37°C in modified Krebs Ringer buffer (KRB) supplemented with 5.6 mM glucose and 3% bovine serum albumin (BSA). The buffer was continuously bubbled with a mixture of O2 (95%) and CO2 (5%) for a final pH of 7.4. Batches of 250 islets were then incubated in 5 ml of fresh modified KRB buffer with 1% BSA plus the test agents, for different time intervals, at 37°C. After removing the incubation medium, islets were sonicated twice in 1 ml of water/ethanol 50% for 30 s using a Sonic dismembrator (Artek System Corp., Farmingdale, N.Y.) at 65% maximal power. Cytosolic fractions were obtained by centrifugation of the samples for 30 min in an Optima TLX Ultracentrifuge (Beckman Instruments Inc., Palo Alto, Calif.) at 65,000 r.p.m. and 4°C. Afterward, proteins in the supernatant were precipitated in acetone at -20°C, for 20 min. The final suspension was lyophilized (Virtis Co. Inc., Gardiner, N.Y.). ApnA and ATP were then analyzed, in the same sample, by high-performance liquid chromatography (HPLC) as previously described (7).
HPLC procedures
The HPLC equipment (Waters, Milford, Mass.) consisted of a 600 E delivery system, an injector autosampler 717 plus, a Scanning Fluorescence Detector 474, a
max 481 Spectrophotometer, and a Millenium 2010 Chromatography Manager System. The mobile phase conditions were as follows: 100 mM KH2PO4, 3% methanol, pH 6.0. The column was a Spherisorb ODS-2 25 cm in length and with an inner diameter of 0.4 (from Scharlau Science, Barcelona, Spain). The system was equilibrated overnight at a flow rate of 0.1 ml/min. Chromatographic analysis was performed at a flow rate of 1.5 ml/min and detection was monitored at 260 nm wavelength. Areas of the mono- and dinucleotides were transformed into concentrations by comparisons with commercial standards (14).
The highly sensitive luciferin-luciferase method was used to determine Ap4A and confirm the measures performed by HPLC, especially when this dinucleotide was present at very low concentrations (15). Briefly, samples were treated with 1 U/ml of alkaline phosphatase (EC 3.1.3.1.) at 37°C for 3 h to guarantee the complete hydrolysis of ATP and other adenine mononucleotides. The enzyme was inactivated by boiling the sample for 5 min at 100°C. Dinucleotides are highly resistant to temperature, as previously described (16). The presence of Ap4A was studied by applying to 1 ml of 25 mM HEPES, 10 mM MgCl2, 2 mM EDTA, and 20 µl of luciferin-luciferase reagent (pH 7.5) an appropriate volume of each corresponding biological sample. After application of the sample, phosphodiesterase from Crotalus durissus (EC 3.1.15.1) in a concentration of 1 U/ml was added. The increase of light was measured in a BioHit 1250 Luminometer. Luminescence values were transformed into Ap4A concentrations by comparing them with a standard curve of luminescence prepared with graded concentrations of Ap4A.
Electrophysiology
Patch pipettes were pulled from Clark Electromedical glass capillaries (Reading, U.K.) using a two-stage puller (Mecanex BB-CH, Geneva, Switzerland) with resistances in the range of 312 M
when filled with a standard solution (in mM): 5 KCl, 135 NaCl, 10 HEPES, 2.5 CaCl2, 1.1 MgCl2, pH 7.4. Bath solution contained (in mM): 140 KCl, 1 CaCl2, 1 MgCl2, 10 HEPES, and 1 EGTA, pH 7.2. Solutions containing ApnA and ATP were applied through an RSC-100 rapid solution changer (Biologic, Claix, France). KATP channel unitary currents were registered from excised membrane patches in the inside-out configuration (17). Currents were measured using an Axopatch 200 amplifier (Axon Instruments Inc. Foster City, Calif.) and stored in a tape recorder (DAT, DTR-1202, Biologic, Claix, France) for subsequent analysis with a homemade program. Experiments were filtered through an eight-pole Bessel filter (Frequency Devices, Haverhill, Mass.) at 1 kHz and sampled at 10 kHz. Pipette potential was held at 0 mV throughout the record. The experiments were carried out at room temperature (2024°C). Concentration-response curves were obtained by integrating the area corresponding to channel activity in the presence or absence of different ApnA concentrations and adjusting these data to the Hill equation.
Data analysis
Patch clamp results were expressed as mean and standard deviation of the mean (±SD). The rest of the results were expressed as mean and standard error of the mean (±SE). For comparisons between two groups, the unpaired Student's t test (two-tailed) was used. P < 0.01 was considered significant.
| RESULTS |
|---|
|
|
|---|
|
In
Fig. 1C, D, the [ApnA]i and [ATP]i values were measured at the end of 300 s incubation in the presence of increasing glucose concentrations. As shown in
Fig. 1C, Ap4A and Ap3A cytosolic concentrations increased 30- to 50-fold between 2.75 and 16.7 mM glucose (EC50=7.81 and 7.42 mM, respectively), remaining constant at higher glucose concentrations. The first significant increase in [Ap4A]i and [Ap3A]i was observed at 5 mM glucose. Therefore, [ApnA]i enhancement was significant (P<0.01) between 5 and 16.7 mM glucose, the glucose concentration interval in which insulin secretion is increased. Glucose-induced [ATP]i rise was parallel to changes observed in cytosolic ApnA concentrations (EC50=7.5 mM) (
Fig. 1D). Cytosolic ATP values showed only a two- to threefold increase between 2.75 and 22 mM glucose. Ap4A measurements gave very similar results with the luciferin-luciferase assay.
Amino acid- and glucose-induced ApnA cytosolic increase is blocked by metabolism inhibition
Other metabolizable fuel secretagogues also increased [Ap4A]i and [Ap3A]i to levels similar to those reached with stimulatory glucose concentrations. Islets incubated for 300 s in the presence of 10 mM leucine or KIC showed an [Ap4A]i 66- and 80-fold increase, respectively. Analogous results were observed for [Ap3A]i (44- and 70-fold increase, respectively) (
Fig. 2A).
Simultaneous measurements of [ATP]i reached values of 3.83±0.84 mM for 10 mM leucine and 6.10±0.96 mM for 10 mM KIC (
Fig. 2B). These values correspond to a two- to threefold [ATP]i increase relative to basal values. On the other hand, islets incubated with the nonmetabolizable amino acid arginine (20 mM) showed no modifications of basal cytosolic Ap4A (0.62±0.03 µM) and Ap3A (0.25±0.02 µM) values (
Fig. 2A). Likewise, [ATP]i remained at basal values (1.03±0.26 mM) in the presence of 20 mM arginine (
Fig. 2B).
|
The influence of glycolysis and mitochondrial metabolism was determined by incubating islets with iodoacetate, a glyceraldehyde 3-phosphate dehydrogenase inhibitor, or with fluoroacetate, a citric acid cycle inhibitor. In
Fig. 2C, D, islets were preincubated with 1 mM iodoacetate for 20 min and then with 1 mM iodoacetate plus 22 mM glucose for another 300 s. Blocking glycolysis at the oxidation of glyceraldehyde 3-phosphate step completely inhibited glucose-induced [Ap4A]i (0.22±0.01 µM) and [Ap3A]i (0.27±0.03 µM) increase (
Fig. 2C). Halting the mitochondrial metabolism by incubating the islets with 2 mM fluoroacetate plus 22 mM glucose for 300 s showed a 59% inhibition for Ap4A and a 68% inhibition for Ap3A cytosolic concentrations (
Fig. 2C). Since both fluoroacetate and iodoacetate are metabolic inhibitors that specifically block one metabolic pathway, they do not counteract all the energy-providing systems. Thus, [ATP]i of islets incubated with 1 mM iodoacetate plus 22 mM glucose or 2 mM fluoroacetate plus 22 mM glucose were not completely depleted (89 and 64% respectively) (
Fig. 2D).
Nonnutrient secretagogues do not increase cytosolic ApnA concentrations
To study whether increases in [Ca2+]i contribute by themselves to ApnA synthesis, we investigated the effects of tolbutamide (to block KATP channels) and high K+ concentrations on [ApnA]i. Islets were incubated in the presence of 100 µM tolbutamide or 10 mM KCl for 300 s in the absence of glucose (
Fig. 3).
Neither tolbutamide nor KCl led to increases in Ap4A (0.40±0.13 and 0.29±0.02 µM, respectively) and Ap3A cytosolic concentrations (0.33±0.11 and 0.15±0.04 µM, respectively).
|
Inorganic pyrophosphatase inhibition prevents glucose-induced ApnA cytosolic increase
The contribution of pyrophosphate and aminoacyl-adenylate to glucose-induced ApnA synthesis was examined by incubating islets with the inorganic pyrophosphatase inhibitor NaF. Islets were preincubated with 50 µM NaF for 30 min and then with 50 µM NaF plus 22 mM glucose for another 300 s. As observed in
Fig. 4A,
22 mM glucose-induced Ap4A and Ap3A synthesis was almost completely blocked by 50 µM NaF (67% and 68% inhibition, respectively). Conversely, cytosolic ATP values were not affected by 50 µM NaF inhibition (5.47±0.41 mM without NaF and 4.77±0.29 mM with 50 µM NaF, in the presence of 22 mM glucose) (
Fig. 4B).
|
Ap4A interaction with KATP current differs from that of ATP
When KATP channel activity is recorded in the inside-out configuration, it declines spontaneously after a short while, a phenomenon termed `rundown'. ATP, the primer ligand for KATP channels, acts not only as inhibitor, but also restores channel activity after rundown has occurred. The maintenance of channel activity can be observed after ATP removal. The next experiment was aimed at testing whether Ap4A, in addition to imitating ATP by decreasing channel activity, was able to reactivate KATP channel as well.
Figure 5A
shows a sample record of current in a membrane patch containing about eight channels shortly after isolation of the patch into zero ATP. ATP (2 mM) applied at the beginning of the record completely blocked channel activity. When ATP was removed, the patch current initially rose to a maximum and then started to decline. On return to 2 mM ATP, KATP channel activity was completely inhibited. ATP was applied twice. It can be observed that after ATP removal, the current returns to its initial value. The record shows that this is not the case for Ap4A. The presence of 200 µM Ap4A almost completely blocked channel activity. But upon Ap4A removal, patch current did not rise to maximal activity; it only increased to the level of the patch current after suffering rundown. Finally, a new exposure to ATP showed that channel recovery in the patch was still present. This experiment demonstrates that, in contrast to ATP, rundown of KATP channel activity could not be reversed by the addition of Ap4A to the cytoplasmic side of the membrane.
|
We also examined the interactions between ATP and Ap4A in modulating channel activity.
Figure 5B shows the dependence of channel activity on ATP in the absence and presence of 10 µM Ap4A. When normalized to maximum channel activity in zero ATP, the ATP concentration-response curve was not consistently altered by Ap4A. Likewise, the Ap4A dose-response curve normalized to zero Ap4A was not consistently altered in the presence of 10 µM ATP (not shown). At a given control level of channel activity, the relative blocking efficacy of Ap4A was the same for channels partially inhibited by ATP. These results suggest that Ap4A inhibition of channel activity is independent of channel inhibition by ATP and that they do not compete with each other for the same binding site.
| DISCUSSION |
|---|
|
|
|---|
aa~AMP + PPi); 2) the aminoacyl-adenylate reacts with ATP or ADP to form respectively Ap4A or Ap3A and the free amino acid (aa~AMP + ATP or ADP
Ap4A or Ap3A + aa) (18). It could therefore be assumed that cytosolic ATP and ADP concentrations are involved in glucose-induced [ApnA]i increase. In addition, the concentration-response relationship observed in the glucose- and time-dependent cytosolic increase between ApnA and ATP supports the view that adenine nucleotides are direct regulators of ApnA synthesis. However, several pieces of evidence do not favor this hypothesis. First, recent studies have shown that glucose causes a very modest elevation in ß-cell ATP values (19), whereas [ApnA]i increases by 30- to 50-fold (7). Second, [ATP]i is always in the millimolar range, more than enough to induce ApnA synthesis, which nonetheless never reaches values higher than tens of micromolar. Third, free [ADP]i, which is the source for Ap3A, appears to fall when glucose is elevated (20) whereas [Ap3A]i rises. That other nutrient stimuli such as leucine and KIC, which are directly metabolized in the mitochondria (21), also increase [ApnA]i to levels similar to those reached with stimulatory glucose concentrations, together with the absence of effect observed for nonnutrient and nonmetabolizable secretagogues (KCl, tolbutamide and arginine), suggest that high-energy currency molecules common to different metabolic pathways mediate the synthesis of ApnA. Consistent with this view, both glycolytic [iodoacetate, a glyceraldehyde 3-phosphate dehydrogenase inhibitor (22)] and mitochondrial inhibitors [fluoroacetate, a citric acid cycle inhibitor (23)] should block [ApnA]i increase, as confirmed by our findings. Inorganic pyrophosphate greatly inhibits ApnA production by aminoacyl-tRNA synthetases, probably through competitive inhibition of the ATP reaction with aminoacyl-adenylate. The Km values for pyrophosphate in the reaction with aminoacyl-adenylate to produce amino acid and ATP can be as small as 25 µM, whereas the corresponding Km value for ATP in the reaction of Ap4A formation is usually between 2 and 50 mM (24). This explains why [ApnA]i never reaches values higher than 2030 µM. Consequently, inorganic pyrophosphatase should contribute to ApnA synthesis. Sodium fluoride inhibits many enzymes, but only a few (including inorganic pyrophosphatase) are sensitive to micromolar concentrations (25). The evidence that sodium fluoride does not affect [ATP]i levels, while decreasing [ApnA]i, indicates that cellular metabolism is undamaged and that inorganic pyrophosphatase is the enzyme that controls nutrient-induced ApnA synthesis.
Looking at alternative nutrient-activated pathways for ApnA synthesis, we see that acetyl-CoA appears to be a good candidate for two reasons: 1) Glucose (26), leucine (21), and KIC (27) are metabolized to acetyl-CoA causing insulin release. Thus, a rise in an external signal (nutrient) is transduced into accelerated mitochondrial acetyl-CoA production. 2) Acetyl-CoA causes marked alterations in pyrophosphate profiles (28) through mitochondrial pyrophosphatase. Therefore, we propose that nutrient metabolism generates high-energy compounds (besides ATP) that increase ApnA synthesis through pyrophosphatase activation.
In this study we have also compared the mechanisms by which ATP and Ap4A exert their respective inhibitory actions on the KATP channel in mouse pancreatic ß-cells. We show that both ATP and Ap4A induce a distinct effect on KATP channel activity. In contrast to ATP, Ap4A is not able to prevent or remove channel rundown. Thus, Ap4A appears to be a pure channel inhibitor without the ability to maintain channel opening through hydrolysis or phosphorylation. In this regard, the action of Ap4A on KATP channels more closely resembles that of nonhydrolyzable analogs such as ATP-
-S (29).
It seems unlikely that ATP and Ap4A bind to the same site, since an effect of ATP on the Ap4A dose-response curve (and vice versa) would be expected. The present data do not indicate any competition between ATP and Ap4A for inhibition of KATP channels, and are more consistent with both intracellular compounds acting independently and additively to inhibit channel activity. Functional interactions between them likely occur not by competition at a particular binding site, but rather at another level of control within the channel protein. The constitutive KATP channel protein complex possesses several nucleotide binding and phosphorylation sites. The sites of action of Ap4A and ATP are probably different nucleotide binding domains (30).
We propose that KATP channel activity relies on at least two inhibitory signaling pathwaysthe conventional ATP-dependent and the ApnA-dependentwhich are not mutually exclusive. We want to emphasize that our data do not disprove a role for ATP, but at the very least the evidence indicates that the notion that ATP is the most important metabolic coupling factor of ß-cells must not be taken for granted. Endogenous ApnA have a considerably longer intracellular half-life compared to ATP (6), which makes them better candidates as messenger molecules. Consequently, these results agree with the notion of the ß-cell stimulated in response to nutrients through KATP channels, broadening the role of KATP channel as effector system: since the channel is able to recognize several metabolites from different pathways, channel activity reflects more widely the metabolic state of the cell (
Fig. 6).
Furthermore, the outcome of the interaction between Ap4A and ATP on KATP channel is not constant, and the contributions of each would depend on metabolic conditions. This provides a physiological meaning for the existence of two different metabolic pathways that converge in KATP channel regulation. For instance, Ap4A may counteract the consequence of locally decreased intracellular ATP concentration under cellular challenges (31).
|
In conclusion, this study provides new insights for the stimulus-secretion coupling, finding the link between nutrient secretagogues and ApnA synthesis, which may act as a new second messenger mediating nutrient-induced blockade of KATP channel in pancreatic ß-cell, and independence of cytosolic ATP concentrations.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
2 Abbreviations: Ap4A, P1,P4-di(adenosine 5')-tetraphosphate; [Ap4A]i, intracellular Ap4A concentration; Ap3A, P1,P3-di(adenosine 5')-triphosphate; [Ap3A]i, intracellular Ap3A concentration; ApnA, diadenosine polyphosphates in general; [ApnA]i, intracellular ApnA concentrations; [ADP]i, intracellular ADP concentration; [ATP]i, intracellular ATP concentration; BSA, bovine serum albumin; [Ca2+]i, cytosolic Ca2+ concentration; HPLC, high-performance liquid chromatography; KIC,
-ketoisocaproate; KRB, Krebs Ringer buffer; KATP channel, ATP-sensitive K+ channel; NaF, sodium fluoride. ![]()
Received for publication April 17, 1998.
Revision received June 4, 1998.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R.-T. Guo, Y. E. Chong, M. Guo, and X.-L. Yang Crystal Structures and Biochemical Analyses Suggest a Unique Mechanism and Role for Human Glycyl-tRNA Synthetase in Ap4A Homeostasis J. Biol. Chem., October 16, 2009; 284(42): 28968 - 28976. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Dabrowski, A. Tarasov, and F. M. Ashcroft Mapping the architecture of the ATP-binding site of the KATP channel subunit Kir6.2 J. Physiol., June 1, 2004; 557(2): 347 - 354. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. Soria, I. Quesada, A. B. Ropero, J. A. Pertusa, F. Martin, and A. Nadal Novel Players in Pancreatic Islet Signaling: From Membrane Receptors to Nuclear Channels Diabetes, February 1, 2004; 53(90001): S86 - 91. [Abstract] [Full Text] |
||||
![]() |
I. Quesada, F. Martin, E. Roche, and B. Soria Nutrients Induce Different Ca2+ Signals in Cytosol and Nucleus in Pancreatic {beta}-Cells Diabetes, February 1, 2004; 53(90001): S92 - 95. [Abstract] [Full Text] |
||||
![]() |
D. Schulze, M. Rapedius, T. Krauter, and T. Baukrowitz Long-chain acyl-CoA esters and phosphatidylinositol phosphates modulate ATP inhibition of KATP channels by the same mechanism J. Physiol., October 15, 2003; 552(2): 357 - 367. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. Quesada, J. M. Rovira, F. Martin, E. Roche, A. Nadal, and B. Soria Nuclear KATP channels trigger nuclear Ca2+ transients that modulate nuclear function PNAS, July 9, 2002; 99(14): 9544 - 9549. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. B. Wollheim and P. Maechler {beta}-Cell Mitochondria and Insulin Secretion: Messenger Role of Nucleotides and Metabolites Diabetes, February 1, 2002; 51(90001): S37 - 42. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. Quesada, F. Martin, and B. Soria Nutrient modulation of polarized and sustained submembrane Ca2+ microgradients in mouse pancreatic islet cells J. Physiol., May 15, 2000; 525(1): 159 - 167. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |