This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Gao, Z. Articles by Matschinsky, F. M. Search for Related Content PubMed PubMed Citation Articles by Gao, Z. Articles by Matschinsky, F. M. Pubmed/NCBI databases Compound via MeSH Substance via MeSH Hazardous Substances DB CALCIUM COMPOUNDS CALCIUM, ELEMENTAL GLUCOSE

Distinguishing Features of Leucine and -Ketoisocaproate Sensing in Pancreatic ß-Cells

Departments of Pathology and Laboratory Medicine (Z.G., R.A.Y., S.S.S., R.K.W., B.A.W.), Biochemistry and Biophysics (Z.G., G.L., H.N., C.B., F.M.M.), University of Pennsylvania School of Medicine; and Department of Pathology and Laboratory Medicine (Z.G., R.A.Y., S.S.S., R.K.W., B.A.W.), The Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania 19104

Address all correspondence and requests for reprints to: Dr. Franz M. Matschinsky, Diabetes Research Center, 501 Stemmler Hall, 36th and Hamilton Walk, Philadelphia, Pennsylvania 19104-6015. E-mail: matsch{at}mail.med.upenn.edu.

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Culturing rat islets in high glucose (HG) increased 1-¹⁴C--ketoisocaproate (KIC) oxidation compared with culturing them in low glucose. Leucine caused insulin secretion (IS) in low glucose but not in HG rat islets, whereas KIC did so in both. Pretreatment with HG for 40 min abolished leucine stimulation of IS by mouse islets and prevented the cytosolic Ca²⁺ rise without inhibiting IS and Ca²⁺ increments caused by KIC. When islets were pretreated without glucose and glutamine, aminooxyacetic acid (AOA) markedly decreased KIC effects. When islets were pretreated without glucose and with glutamine, AOA potentiated leucine effects but attenuated KIC effects. AOA stimulated glutamine oxidation in the presence but not the absence of ±2-amino-2-norbornane-carboxylic acid, a nonmetabolized leucine analog. Pretreatment with HG and glutamine partially reversed AOA inhibition of KIC effects. Glucose increased intracellular ATP and GTP, whereas it decreased ADP and GDP in ßHC9 cells. Glutamate dehydrogenase activity of ßHC9 cell extracts was increased by leucine and attenuated by GTP, but it was potentiated by ADP. In conclusion, leucine and KIC stimulated ß-cells via distinct mechanisms. Glutamate dehydrogenase is the sensor of leucine, whereas transamination plays an important role in KIC stimulation of pancreatic ß-cells.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
AMINO ACIDS PLAY an important role in the regulation of insulin secretion from pancreatic ß-cells. Although a general molecular mechanism by which this class of fuel triggers insulin secretion is lacking, the action of individual amino acids applied at high concentrations is partly understood. For example, it was realized in the 1980s that glutamate dehydrogenase (GDH) is involved in insulin secretion stimulated by leucine (1), but the central physiological role of GDH in ß-cell function was not fully appreciated until more recently. The discovery of hyperinsulinism in patients with mutations in the regulatory site of GDH (2) has triggered the reinvestigation of this complex issue (3).

Many questions pertaining to the role of energy metabolism in ß-cells stimulated physiologically by amino acids remain unresolved (4). The mechanisms by which leucine and -ketoisocaproate (KIC) stimulate insulin release are special cases in this complex of questions. It was demonstrated that pancreatic islets oxidize leucine and KIC, and it is believed that enhanced ATP production triggers insulin secretion (5). However, simplistic explanations are not satisfactory. Thus, the rates of oxidation and the influence on the rates of ATP production of a group of ketoacids do not correlate with their capacities to stimulate insulin secretion (5). Some amino acids, such as valine and isoleucine, are oxidized as efficiently as leucine and KIC, but they do not share the same potency as insulin secretagogues (5) or in ATP production (6), and sometimes they even antagonize KIC actions (5). It has been reported that ATP production from KIC oxidation of islet mitochondria plateaus at high micromolar concentrations (7), whereas the threshold for KIC-induced insulin secretion is found at low millimolar concentrations. It is noteworthy in this context that insulin secretion of patients with mutations of branched-chain ketoacid dehydrogenase (BCKDH), the apparent rate-limiting enzyme of leucine/KIC metabolism, appears to be normal (8, 9).

One possible fate of KIC is the production of leucine by transamination with the concurrent generation of -ketoglutarate. The operation of this pathway in pancreatic islets has been demonstrated previously, but its physiological importance remains a matter of debate (10, 11, 12). For instance, it was concluded that the KIC action in ß-cells is entirely indirect through leucine (7).

The aim of the current study is to continue the exploration of metabolic pathways that might be involved when pancreatic ß-cells are stimulated by leucine or KIC. The present and most previous results support the hypothesis that leucine and KIC may stimulate insulin release from pancreatic ß-cells by distinct mechanisms.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Islet preparation
Wistar rat and C57/black mouse islets were isolated by collagenase digestion and cultured in RPMI containing 0.05 mM leucine and 0.2 mM glutamine, in addition to other amino acids, as specified by the supplier (Sigma, St. Louis, MO). The concentration of all amino acids combined in RPMI is 6.51 mM. RPMI was supplemented with 10% fetal bovine serum, 100 U/ml penicillin, and 50 µg/ml streptomycin, and cells were cultured in a 5% CO₂/95% air humidified incubator. Culture and incubations were at 37 C.

Insulin secretion in islets
The first set of insulin secretion experiments was designed to mimic those of MacDonald et al. (13). Rat islets were cultured in RPMI containing 0, 5, or 20 mM glucose for 20–24 h. Batches of 10 islets were then quickly washed and incubated for 10 min in Krebs-Ringer-bicarbonate buffer [KRB; 115 mM NaCl, 24 mM NaHCO₃, 5 mM KCl, 1 mM MgCl₂, 2.5 mM CaCl₂, 25 mM HEPES, and 1% BSA (pH 7.4)] without glucose (G0). Islets were finally incubated for 20 min in the absence or presence of stimuli to study secretion. Mouse islets were also cultured and pretreated without or with 30 mM glucose (PreG30) for 40 min at 37 C in KRB. They were then perifused in KRB for 10 min G0 followed by 10 mM of either KIC or leucine. Perfusate was collected every minute, and its insulin concentration was determined by RIA.

Oxidation of 1-¹⁴C-KIC by islets
Rat islets were cultured for 20–24 h in 0, 5, 10, or 25 mM glucose to approximate conditions used by MacDonald et al. (13). Batches of 100 islets were thereafter incubated for 1 h in 0.1 ml G0 KRB containing various concentrations of nonradioactive KIC and 1 µCi/tube 1-¹⁴C-KIC (specific activity, 2 mCi/mmol; NEN Life Science Products, Boston, MA) in Eppendorf tubes, which were placed in 20-ml scintillation vials. Separate blanks were obtained for each KIC level in the absence of islets to allow correction for nonspecific decarboxylation. The experiment was stopped by adding 2 M sodium acetate (pH 3.4). A trap filter was used to collect the ¹⁴CO₂, and the amount of radioactivity was determined by liquid scintillation counting. The results are expressed as decarboxylated KIC (picomoles per islet per hour).

Oxidation of U-¹⁴C-glutamine by islets
Batches of 100 rat islets were incubated for 1 h at 37 C in 100 µl of KRB containing various concentrations of nonradioactive glutamine, as indicated in Results, and 1 µCi/tube of radioactive U-¹⁴C-glutamine (specific activity, 244 mCi/mmol; NEN Life Science Products) in Eppendorf tubes, which were placed in 20-ml scintillation vials. A trap filter was also placed in each tightly sealed scintillation vial to collect the ¹⁴CO₂ produced by the islets, and the amount of radioactivity was determined by liquid scintillation counting.

Cytosolic free Ca²⁺ measurement
Mouse islets were isolated and cultured for 3–4 d. They were then loaded with fura-2 during a 40-min pretreatment at 37 C in 2 ml KRB buffer supplemented with 1 µM fura-2 acetoxymethylester (Molecular Probes, Inc., Eugene, OR). During this pretreatment period, islets were exposed to various glucose concentrations and/or various glutamine concentrations as specified in Results. The loaded islets were then fixed by slight suction onto the tip of a micropipette in a perifusion chamber placed on the homeothermic platform of an inverted Zeiss microscope (Carl Zeiss, Thornwood, NY). The islets were perifused with KRB at 37 C at a flow rate of 1 ml/min, while various treatments were applied to the islets. The microscope was used with a 40x oil immersion objective. Fura-2 was successively excited at 334 and 380 nm by means of two narrow band-pass filters. The emitted fluorescence was filtered through a 520-nm filter, captured with an Attofluor charge coupled device video camera at a resolution of 512 x 480 pixels, digitized into 256 gray levels, and analyzed with version 6.07 of the Attofluor RatioVision software (Atto Instrument, Rockville, MD).

Intracellular nucleotide measurement
ßHC9 cells were plated in six-well dishes and cultured for 2–3 d in RPMI containing 10 mM glucose (G10). Cells (10⁶ per well) were washed with G0 KRB and incubated in G0 KRB for 6 h. They were then incubated for 40 min in the absence or presence of 5 or 15 mM glucose before nucleotides were extracted using 1 ml of ice-cold 5% trichloroacetic acid. Cellular nucleotide contents were then measured via HPLC as described previously (14).

Allosteric modification of GDH activity in extract of ßHC9 cells
ßHC9 cells were plated in 10-cm dishes and cultured in RPMI containing G10. Cells were washed with G0 KRB and incubated with G0 KRB for 6 h. They were then incubated for 40 min before the cells were scraped off the plate. Cells were then homogenized by sonication in 1 ml of homogenizing buffer [sucrose 70 mM, mannitol 230 mM, EDTA 0.1 mM, potassium phosphate 10 mM (pH 7.0)]. GDH activity in cellular homogenates was measured at a wavelength of 340 nm using a Beckman spectrophotometer (model DU640, Beckman Coulter, Inc., Fullerton, CA). Assay conditions were as follows: imidazole base, 35 mM; imidazole-HCl, 15 mM; ammonium acetate, 25 mM; BSA, 0.1%; -ketoglutarate, 2 mM; nicotinamide adenine dinucleotide phosphate (reduced), 0.1 mM (pH 7.4). Leucine, GTP, and ADP were added as described in Results.

Data analysis
Student’s t test was performed when two groups were compared. ANOVA was used, followed by the Newman-Keuls test when multiple groups were compared. Differences were considered significant for P values less than 0.05.

Materials
All chemicals were from Sigma (St. Louis, MO) unless otherwise indicated.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Insulin secretion and KIC oxidation in islets cultured with various glucose concentrations (Figs. 1 and 2)
Experiments were designed to study glucose modification of leucine and KIC effects and KIC oxidation in islets. In islets cultured 20–24 h at G0, a subsequent incubation without glucose showed insulin secretion (IS) of 1.76 ± 0.12 ng/10 islets per 20 min. When tested at 10 mM, glucose was ineffective (2.08 ± 0.20 ng/10 islets per 20 min; P > 0.05), in contrast to Leucine10 (3.11 ± 0.35 ng/10 islets per 20 min; P < 0.05) and 10 mM KIC (KIC10; 4.87 ± 0.68 ng/10 islets per 20 min; P < 0.01). In 5 mM glucose (G5)-cultured islets, IS in G10 (3.07 ± 0.54 ng/10 islets per 20 min) was not significantly different from IS in G0 (2.05 ± 0.42 ng/10 islets per 20 min; P > 0.05). Leucine10 (5.20 ± 0.39 ng/10 islets per 20 min) and KIC10 (20.18 ± 1.11 ng/10 islets per 20 min) stimulated IS significantly (P < 0.05). In 20 mM glucose (G20)-cultured islets, IS in G10 (12.48 ± 1.42 ng/10 islets per 20 min) was higher (P < 0.05) than in G0 (5.48 ± 1.05 ng/10 islets per 20 min), and Leucine10 (5.94 ± 1.55 ng/10 islets per 20 min) did not affect IS, in contrast to KIC10 (24.77 ± 2.48 ng/10 islets per 20 min).

View larger version (17K): [in this window] [in a new window]   Figure 1. Effects of brief culture with glucose on insulin secretion in rat islets. Islets were cultured in 0, 5, or 20 mM glucose for 20–24 h, before insulin secretion was tested, as indicated in Materials and Methods. Data are the mean ± SEM of four experiments. *, Significant difference compared with G0.

View larger version (22K): [in this window] [in a new window]   Figure 2. Effects of brief culture with glucose on [1-14C]-KIC oxidation in rat islets. Islets were first cultured for 3–4 d in 10 mM glucose, and culture was then continued in 0 (G0), 5 (G5), 10 (G10), or 25 (G25) mM glucose for 20–24 h. They were then used for [1-14C]-KIC oxidation studies to assess BCKDH. Data are the mean ± SEM for four to five experiments.

Glucose modulation of leucine effects on insulin secretion and cytosolic Ca²⁺ of islet cells (Fig. 3)
When cultured mouse islets were first treated with G0 for 40 min (PreG0) followed by 10 min perifusion in G0, the insulin secretion rate was low and stable at 6.4 ± 1.9 pg/islet·min at 10 min (Fig. 3A). Adding 10 mM leucine (Fig. 3A) increased secretion to a peak of 39.8 ± 6.1 pg/islet·min at 15 min. The average of leucine-induced secretion was 22.9 ± 5.8 pg/islet·min, which was significantly higher than the basal secretion (P < 0.05). When islets were first treated with 30 mM glucose for 40 min (PreG30) followed by 10 min perifusion with G0, insulin secretion gradually decreased, reaching 14.7 ± 4.9 pg/islet·min at 10 min. Adding 10 mM leucine did not increase insulin secretion (P > 0.05), which continued to decrease, eventually reaching 6.3 ± 1.7 pg/islet·min. The net change of secretion from baseline in PreG0 islets (16.3 ± 5.9 pg/isletxmin) was significantly higher than that of PreG30 islets (-8.4 ± 3.3 pg/islet·min; P < 0.01). Thus, pretreatment of high glucose inhibited leucine-induced secretion in isolated mouse islets.

View larger version (30K): [in this window] [in a new window]   Figure 3. Effects of leucine or KIC on insulin secretion and cytosolic Ca2+ of mouse islets pretreated with various concentrations of glucose. Islets were cultured for several days at 10 mM glucose before they were pretreated with 0, 5, 10, 15, or 30 mM glucose for 40 min in KRB. They were perifused with G0 KRB for 10 min and then with 10 mM leucine or KIC. A, 10 mM leucine-induced insulin secretion of islets pretreated without glucose (PreG0, open circle) or with 30 mM glucose (PreG30, filled circle). B, 10 mM KIC-induced insulin secretion of islets pretreated without glucose (PreG0, open circle) or with 30 mM glucose (PreG30, filled circle). C, The cytosolic Ca2+ time courses of islets pretreated with 30 mM glucose (PreG30, filled circle) or without glucose (PreG0, open circle). D, The average cytosolic Ca2+ change above baseline induced by leucine. All experiments in this figure were done without glutamine (Q0). Each insulin secretion curve is the mean ± SEM of at least 4 experiments, and each Ca2+ trace is the mean ± SEM of 9–24 islets from 3–6 islet batches.

When islets were glucose deprived (PreG0) before Ca²⁺ measurement, basal Ca²⁺ was 131 ± 10 nM, and leucine (10 mM) initially decreased it to 102 ± 9 nM, followed by a peak at 345 ± 43 nM (P < 0.01, compared with the absence of leucine), and a plateau of approximately 200 nM. In glucose-pretreated islets (PreG30), Ca²⁺ was higher and gradually decreased to 206 ± 24 nM at the time leucine was added, in contrast to the marked increase (1000 nM) caused by 30 mM glucose (data not shown). The pretreatment of islets with a range of glucose concentrations showed a clear dose-dependency of the inhibitory glucose effect. Thus, in the absence of glutamine, glucose pretreatment dose-dependently inhibited the leucine-induced Ca²⁺ response, with an IC₅₀ of about 4–5 mM and a maximally effective concentration of about 15 mM.

Aminooxyacetic acid (AOA) modulation of leucine-induced cytosolic Ca²⁺ and glutamine oxidation of islet cells (Fig. 4)
AOA was used to test the role of transaminases in glutaminolysis. Cultured mouse islets were loaded with fura-2 in G0 KRB with 2 mM glutamine (Q2) for 40 min before being perifused in KRB with Q2 and without or with 5 mM AOA (AOA5). Various concentrations of leucine were added to record the Ca²⁺ response (Fig. 4A). With AOA5, basal Ca²⁺ (178 ± 18 nM) was not significantly different from control (164 ± 17; P > 0.05). The leucine-induced Ca²⁺ increase occurred about 2 min earlier and reached its highest level of 503 ± 128 nM (P < 0.01, compared with the absence of AOA) and thus a larger overall stimulation. AOA5 shifted the leucine dose curve to the left and increased the maximal effect from 292 ± 45 nM to 438 ± 119 nM. Both curves are sigmoidal (Fig. 4B).

View larger version (20K): [in this window] [in a new window]   Figure 4. AOA effects on leucine-induced cytosolic Ca2+ changes and glutamine oxidation. A, Islets were cultured for several days at 10 mM glucose before they were pretreated for 40 min without glucose, but with 2 mM glutamine (Q2) ± 5 mM AOA (AOA5). Islets were then perifused without glucose and stimulated by 2.5 mM leucine. Data are the mean ± SEM of 9–24 islets from 3–6 separate isolations. B, Average cytosolic Ca2+ change above baseline induced by various concentrations of leucine. C, Rat islets cultured for 4–5 d at 10 mM glucose were preincubated for 40 min without glucose before [U-14C]glutamine oxidation was performed in the presence or absence of 10 mM BCH with or without 5 mM AOA. The mean ± SEM of four experiments is shown for each condition. *, Significant differences between control and BCH10; **, significant differences between BCH and AOA + BCH.

Glucose modification of KIC-induced cytosolic Ca²⁺ in islets (Fig. 5)
The basal Ca²⁺ of islets pretreated without glucose (PreG0) was 130 ± 32 nM (Fig. 5A, trace a). KIC10 rapidly increased Ca²⁺ to 510 ± 108 nM in approximately 2 min, and Ca²⁺ gradually plateaued at approximately 300 nM. The KIC effect was faster and larger than the leucine effect (compared with Fig. 4). Basal Ca²⁺ of islets pretreated with 30 mM glucose (PreG30) was approximately 330 nM at 0 min and declined to approximately 200 nM at 2 min (Fig. 5A, trace b). KIC at 10 mM instantaneously raised Ca²⁺ to 717 ± 189 nM, and Ca²⁺ plateaued at approximately 500 nM (170 nM higher than PreG0). However the net increase was not significantly different (Fig. 5A; P > 0.05). KIC at 2.5 mM had no effect on cytosolic Ca²⁺ (Fig. 5E; P > 0.05). The stimulation of 5 mM KIC was larger (P < 0.05) in PreG30 (103 ± 2 nM) than in PreG0 (51 ± 6 nM). Glucose pretreatment augmented or did not affect the KIC-induced islet Ca²⁺ response, whereas it prevented that of leucine as shown above. Figure 5E summarizes the net increase of Ca²⁺ by KIC at different concentrations.

View larger version (41K): [in this window] [in a new window]   Figure 5. AOA effects on KIC-induced cytosolic Ca2+ changes. Islets were pretreated with 30 mM glucose (PreG30) or without (PreG0) for 40 min, with 5 mM AOA (AOA5) or without (AOA0), and with 2 mM glutamine (Q2) or without (Q0). Cytosolic Ca2+ was then measured in G0 and with the same concentrations of AOA and Q as during pretreatment. Specifically, without AOA and without Q (A), without AOA but with 2 mM Q (B), with 5 mM AOA and without Q (C), and with 5 mM AOA and with 2 mM Q (D). E, The average net Ca2+ level above baseline during the entire period of 2.5, 5, or 10 mM KIC stimulation. F, The cytosolic Ca2+ level when perifused with 0 (G0), 10 (G10), or 20 (G20) mM glucose with or without AOA. Data are the mean ± SEM of 9–15 islets from 3–5 isolations.

AOA effects on KIC-induced cytosolic Ca²⁺ in islet cells (Fig. 5)

Glucose effects on intracellular nucleotide contents of ßHC9 cell
Intracellular ADP levels were 478 ± 36 pmol/100 µg protein in G0, which decreased to 357 ± 23 pmol/100 µg protein with G5 (P < 0.05) and 385 ± 40 pmol/µg protein with G15 (P < 0.05). GDP was 226 ± 47 pmol/100 µg protein in G0, 141 ± 7 pmol/100 µg protein in G5 (P < 0.05). ATP was 2411 ± 68 pmol/100 µg protein in G0, and 3626 ± 169 (P < 0.05) or 3608 ± 211 pmol/100 µg protein in G5 or G15, respectively (P < 0.05). GTP was 178 ± 105 pmol/100 µg protein in G0 and 409 ± 21 or 428 ± 27 pmol/100 µg protein in G5 or G15, respectively (P < 0.05).

Nucleotide regulation of GDH activity in extracts of ßHC9 cells (Fig. 6)
GDH activity was 476 ± 123 pmol/min/100 µg protein when measured without leucine and nucleotides (Fig. 6A). Leucine dose-dependently activated GDH with a maximally effective dose of 2.5 mM (1271 ± 153 pmol/min/100 µg protein). The EC₅₀ of leucine was about 1 mM, as shown by the dotted line in both panels. ADP at 1 µM without leucine activated GDH (838 ± 40 pmol/min/100µg protein) significantly (P < 0.05). The maximal leucine effect was produced at 1 mM (1452 ± 105 pmol/min/100 µg protein) with 1 µM ADP (P < 0.05). With 10 µM ADP, GDH was maximally activated without leucine (1313 ± 12 pmol/min/100 µg protein; P < 0.05). Experiments with 0.01, 0.1, and 100 µM ADP were also performed, and ADP shifted the leucine dose-response curve to the left as shown in Fig. 6A.

View larger version (23K): [in this window] [in a new window]   Figure 6. Leucine effects on GDH activity in the presence of nucleotides. ßHC9 cells were washed and incubated with G0 KRB for 6 h before being homogenized. GDH activity in cellular homogenates was measured in the absence or presence of different concentrations of ADP or GTP. The velocity of the initial 5-min record is presented as mean ± SEM of three experiments. A, Dashed line, without nucleotide; open square, with 1 µM ADP; filled square, with 10 µM ADP. B, Dashed line, without nucleotide; open circle, with 1 µM GTP; filled circle, with 10 µM GTP.

Isoleucine produced a significant stimulation of GDH only at 10 mM (from 504 ± 75 to 884 ± 207 pmol/min/100 µg protein; P < 0.05). BCH, a nonmetabolized leucine analog, increased GDH activity dose-dependently with a threshold of 1 mM (from 273 ± 8 to 377 ± 17 pmol/min/100 µg protein), EC₅₀ of approximately 2.5 mM (705 ± 73 pmol/min/100 µg protein), and maximally effective concentration of 5 mM (1247 ± 61 pmol/min/100 µg protein). Methionine was ineffective.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
We have hypothesized in a previous study that activation of GDH, and thus glutaminolysis, is the process by which leucine regulates ß-cells as judged from both insulin secretion and islet Ca²⁺ levels (3). Our present findings that high glucose completely blocks the effects of leucine, but not those of KIC, suggest that these two stimuli may act through different mechanisms. Mouse islet Ca²⁺ was used as one index of ß-cell fuel sensing because it is a fast and sensitive indicator allowing a continuous sampling and is proximal to insulin secretion, complemented by studies of insulin secretion from perifused mouse islets. Rat islets were used for oxidation experiments because they require large amounts of tissue.

Mechanisms of KIC effects
KIC has at least two metabolic fates in ß-cells: it is oxidized, or it produces leucine through transamination with glutamate with concomitant generation of -ketoglutarate (Fig. 7). Both pathways probably operate simultaneously with corresponding contributions of stimulatory signals. KIC is readily decarboxylated and also converted to leucine through transamination in pancreatic islet homogenate (5). The importance of KIC transamination is further supported by the present experiments with AOA. AOA markedly inhibited KIC stimulation of cytosolic Ca²⁺ and decreased insulin secretion in a previous study (10). The involvement of two pathways of KIC metabolism may explain that sometimes it is a more potent stimulant than leucine at equal molar concentration (10, 15).

View larger version (27K): [in this window] [in a new window]   Figure 7. The roles of GDH, GKAT, and BCKDH. GDH serves as a leucine sensor in pancreatic ß-cells, translating the intracellular leucine concentration into insulin secretion by controlling the activity of GDH and thus glutaminolysis. GDH is under glucose control through the GTP/ADP ratio. GKAT (gutamate-keto acid transaminase) serves as the KIC and auxiliary leucine sensor in pancreatic ß-cells. BCH is a nonmetabolized leucine analog. BCKDH does not appear to be limiting under the conditions listed here.

We propose the following explanation: the effects of a branched-chain ketoacid on ß-cell activities (Ca²⁺ flux/insulin secretion) depend on unique features of their metabolism. They depend on the suitability of the ketoacid to generate the corresponding amino acids by transamination to function as GDH activators. They also depend on the concomitant production of -ketoglutarate from glutamate. A ketoacid such as KIC, which produces a strong GDH activator, such as leucine, in addition to -ketoglutarate, is a potent stimulus. A ketoacid, such as -ketoisovalerate (16), which produces -ketoglutarate at rates comparable to KIC via transamination, generates a poor GDH activator (such as valine) and is a poor stimulus on its own, but it antagonizes the KIC effects by competing for the transamination reaction. The higher the capacity for transamination of a ketoacid, the greater its antagonism to KIC, as demonstrated previously (16).

The effects of a branched-chain amino acid on insulin secretion depend on its activity as GDH activator and its consumption of -ketoglutarate. Valine and isoleucine inhibit KIC effects probably because they effectively consume -ketoglutarate in transamination and thus decrease ATP production. It has been shown that these amino acids, when added alone, are poor stimuli of insulin secretion (16) and are poor activators of GDH, as shown in our data and a previous report (17). Their transamination is very effective, but complete oxidation and ATP generation from them is slow (16).

Recently, it has been reported that KIC, but not leucine, closes ATP-sensitive K⁺ channels of the plasma membrane of ob/ob mouse pancreatic ß-cells by a direct action (18), uncovering a distinguishing feature between KIC and leucine effects. However, the approach used by the authors is unable to distinguish the relative contributions of direct inhibition of ATP-sensitive K⁺ channels vs. indirect mechanisms involving metabolism. A direct KIC effect on channels might explain, at least in part, the glucose-insensitive KIC effects on Ca²⁺ reported here (Fig. 5, trace b). However, the dramatic AOA inhibition of the KIC effects (Fig. 5, comparing traces a and c) is unlikely to involve direct interactions of KIC with the channel.

Role of BCKDH
MacDonald et al. (13) and Wollheim (19) consider BCKDH the rate-limiting step of leucine and KIC actions on the ß-cells. This hypothesis is partly based on the inhibition of leucine effects on insulin secretion in rat islets cultured overnight with high glucose (13). Such an inhibition is also reproduced in our study, however glucose inhibition on insulin secretion and cytosolic Ca²⁺ was not demonstrated when KIC was used as the stimulus. In a separate study from our group, glucose also completely inhibited insulin secretion induced by leucine but not that by KIC in perifused rat islets cultured for 3–4 d in 10 mM glucose before they were pretreated with various glucose concentrations (20). Thus, differential glucose actions on leucine and KIC stimulations of ß-cells have been shown in both rat and mouse islets by measuring insulin secretion and Ca²⁺.

The evidence presented here argues against an important regulatory role of BCKDH in pancreatic ß-cells. KIC oxidation was not inhibited, but instead was slightly stimulated by brief culture with high glucose, indicating that BCKDH was not down-regulated by glucose and contradicting the results of a previous report (13). In islets cultured in high glucose, KIC was a potent stimulus of insulin secretion, whereas leucine had no effect, suggesting that oxidation is not the critical step. This interpretation is supported by findings in maple syrup urine disease caused by BCKDH mutations, which block the oxidation of leucine and KIC but have apparently no effect on serum insulin (8).

Glucose regulates GDH and transaminase
Leucine stimulates insulin secretion from rat islets cultured in low, but not in high glucose, confirming previous findings (13). This desensitization was explained by an inhibition of the leucine sensor GDH (3), a view further strengthened by the current study. It confirms a previous finding that glucose increases the GDH inhibitor GTP, whereas it decreases the GDH activator ADP (21). It is also shown that leucine and BCH, a nonmetabolized leucine analog, dose-dependently activated GDH in ß-cells. Leucine stimulation of GDH was dose-dependently regulated by GTP and ADP at nanomolar to low micromolar concentrations. Changes of glucose metabolism may thus be translated into altered leucine sensitivity through nucleotide regulation of GDH. Transamination appears to be important for both leucine and KIC effects, but in different ways. It is not needed for leucine stimulation of ß-cells, but it may consume -ketoglutarate produced by GDH and thus diminish the leucine effects. Therefore, blockage of transamination by AOA augmented the effects of leucine. On the other hand, transamination is critical for KIC effects as discussed above. The concepts of anaplerosis and cataplerosis have been applied to improve understanding of ß-cell intermediate metabolism in stimulus secretion coupling (22). We do not have sufficient information to take advantage of the conceptual approach interpreting the present results.

Conclusion
GDH is the ß-cell leucine sensor, and its leucine sensitivity is allosterically modulated by the GTP/ADP ratio. As the GTP/ADP ratio increases because of enhanced glucose metabolism, GDH is inhibited and desensitized to leucine. GDH in turn controls glutaminolysis, which contributes to the regulation of insulin release. Leucine stimulation of insulin release is primarily caused by enhanced glutaminolysis. KIC, on the other hand, stimulates the ß-cells by a combination of processes, including its catabolism, transamination to leucine, and production of -ketoglutarate from glutamate. Transamination of KIC to leucine may be more important than its catabolism in triggering cytosolic Ca²⁺ and insulin secretion. The differential effect of glucose pretreatment on leucine and KIC responsiveness of ß-cells is explained by these distinguishing features of leucine and KIC actions.

	Footnotes

 
This work was supported by National Institutes of Health (NIH) Research Grants DK-19525, DK-20212, DK-43354, and DK-49814. B.A.W. is a recipient of NIH Research Career Development Award K04DK-02217.

Abbreviations: AOA, Aminooxyacetic acid; BCH, ±2-amino-2-norbornane-carboxylic acid; BCKDH, branched-chain ketoacid dehydrogenase; G, glucose; G0, without glucose; G5, 5 mM glucose; GDH, glutamate dehydrogenase; IS, insulin secretion; KIC, -ketoisocaproate; KIC10, 10 mM KIC; KRB, Krebs-Ringer-bicarbonate buffer; PreG0, pretreated without glucose; PreG30, pretreated with 30 mM glucose; Q, glutamine.

Received November 25, 2002.

Accepted for publication February 3, 2003.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Sener A, Malaisse WJ 1980 L-Leucine and a nonmetabolized analogue activate pancreatic islet glutamate dehydrogenase. Nature 288:187–189[CrossRef][Medline]
2. Stanley CA, Lieu YK, Hsu BY, Burlina AB, Greenberg CR, Hopwood NJ, Perlman K, Rich BH, Zammarchi E, Poncz M 1998 Hyperinsulinism and hyperammonemia in infants with regulatory mutations of the glutamate dehydrogenase gene. N Engl J Med 338:1352–1357[Abstract/Free Full Text]
3. Gao ZY, Li G, Najafi H, Wolf BA, Matschinsky FM 1999 Glucose regulation of glutaminolysis and its role in insulin secretion. Diabetes 48:1535–1542[Abstract]
4. Newgard CB, Matschinsky FM 2001 Substrate control of insulin secretion. In: Handbook of physiology, section 7: The endocrine system, vol II. The endocrine pancreas and regulation of metabolism. London: Oxford University Press; 125–151
5. Malaisse WJ, Sener A, Malaisse-Legae F, Hutton JC, Christophe J 1981 The stimulus-secretion coupling of amino acid-induced insulin release. Metabolic interaction of L-glutamine and 2-ketoisocaproate in pancreatic islets. Biochim Biophys Acta 677:39–49[Medline]
6. Panten U, Zielmann S, Langer J, Zunkler BJ, Lenzen S 1984 Regulation of insulin secretion by energy metabolism in pancreatic B-cell mitochondria. Studies with a non-metabolizable leucine analogue. Biochem J 219:189–196[Medline]
7. Lembert N, Idahl LA 1998 -Ketoisocaproate is not a true substrate for ATP production by pancreatic ß-cell mitochondria. Diabetes 47:339–344[Abstract]
8. Morris AA, Leonard JV 1997 Early recognition of metabolic decompensation. Arch Dis Child 76:555–556[Free Full Text]
9. Wajner M, Vargas CR 1999 Reduction of plasma concentrations of large neutral amino acids in patients with maple syrup urine disease during crises. Arch Dis Child 80:579–579[Free Full Text]
10. Malaisse WJ, Malaisse-Lagae F, Sener A 1982 The stimulus-secretion coupling of glucose-induced insulin release: effect of aminooxyacetate upon nutrient-stimulated insulin secretion. Endocrinology 111:392–397[Abstract]
11. Malaisse WJ, Sener A, Malaisse-Lagae F, Welsh M, Matthews DE, Bier DM, Hellerstrom C 1982 The stimulus-secretion coupling of amino acid-induced insulin release: metabolic response of pancreatic islets of L-glutamine and L-leucine. J Biol Chem 257:8731–8737[Abstract/Free Full Text]
12. Sener A, Owen A, Malaisse-Lagae F, Malaisse WJ 1982 The stimulus-secretion coupling of amino acid-induced insulin release. XI. Kinetics of deamination and transamination reactions. Horm Metab Res 14:405–409[Medline]
13. MacDonald MJ, McKenzie DI, Kaysen JH, Walker TM, Moran SM, Fahien LA, Towle HC 1991 Glucose regulates leucine-induced insulin release and the expression of the branched chain ketoacid dehydrogenase E1 subunit gene in pancreatic islets. J Biol Chem 266:1335–1340[Abstract/Free Full Text]
14. Xu GG, Gao ZY, Borge Jr PD, Wolf BA 1999 Insulin receptor substrate 1-induced inhibition of endoplasmic reticulum Ca2+ uptake in ß-cells. Autocrine regulation of intracellular Ca2+ homeostasis and insulin secretion. J Biol Chem 274:18067–18074[Abstract/Free Full Text]
15. Lenzen S, Formanek H, Panten U 1982 Signal function of metabolism of neutral amino acids and 2-keto acids for initiation of insulin secretion. J Biol Chem 257:6631–6633[Abstract/Free Full Text]
16. Hutton JC, Sener A, Malaisse WJ 1980 Interaction of branched chain amino acids and keto acids upon pancreatic islet metabolism and insulin secretion. J Biol Chem 255:7340–7346[Abstract/Free Full Text]
17. Lenzen S, Schmidt W, Rustenbeck I, Panten U 1986 2-Ketoglutarate generation in pancreatic B-cell mitochondria regulates insulin secretory action of amino acids and 2-keto acids. Biosci Rep 6:163–169[CrossRef][Medline]
18. Branstrom R, Efendic S, Berggren PO, Larsson O 1998 Direct inhibition of the pancreatic ß-cell ATP-regulated potassium channel by -ketoisocaproate. J Biol Chem 273:14113–14118[Abstract/Free Full Text]
19. Wollheim CB 2000 ß-Cell mitochondria in the regulation of insulin secretion: a new culprit in type II diabetes. Diabetologia 43:265–277[CrossRef][Medline]
20. Li C, Najafi H, Daikhin Y, Nissim I, Collins HW, Yudkoff M, Matschinsky FM, Stanley CA 2003 Regulation of leucine stimulated insulin secretion and glutamine metabolism in isolated rat islets. J Biol Chem 278:2853–2858[Abstract/Free Full Text]
21. Detimary P, Van den Berghe G, Henquin JC 1996 Concentration dependence and time course of the effects of glucose on adenine and guanine nucleotides in mouse pancreatic islets. J Biol Chem 271:20559–20565[Abstract/Free Full Text]
22. Prentki M, Joly E, El-Assaad W, Roduit R 2002 Malonyl-CoA signaling, lipid partitioning, and glucolipotoxicity: role in ß-cell adaptation and failure in the etiology of diabetes. Diabetes 51:S405–S413

This article has been cited by other articles:

				P. Newsholme, L. Brennan, and K. Bender Amino Acid Metabolism, {beta}-Cell Function, and Diabetes Diabetes, December 1, 2006; 55(Supplement_2): S39 - S47. [Abstract] [Full Text] [PDF]

				P. J. Rozance, S. W. Limesand, and W. W. Hay Jr. Decreased nutrient-stimulated insulin secretion in chronically hypoglycemic late-gestation fetal sheep is due to an intrinsic islet defect Am J Physiol Endocrinol Metab, August 1, 2006; 291(2): E404 - E411. [Abstract] [Full Text] [PDF]

				M. Daunt, O. Dale, and P. A. Smith Somatostatin Inhibits Oxidative Respiration in Pancreatic {beta}-Cells Endocrinology, March 1, 2006; 147(3): 1527 - 1535. [Abstract] [Full Text] [PDF]

				H. Heissig, K. A. Urban, K. Hastedt, B. J. Zunkler, and U. Panten Mechanism of the Insulin-Releasing Action of {alpha}-Ketoisocaproate and Related {alpha}-Keto Acid Anions Mol. Pharmacol., October 1, 2005; 68(4): 1097 - 1105. [Abstract] [Full Text] [PDF]

				M. E. Rabaglia, M. P. Gray-Keller, B. L. Frey, M. R. Shortreed, L. M. Smith, and A. D. Attie {alpha}-Ketoisocaproate-induced hypersecretion of insulin by islets from diabetes-susceptible mice Am J Physiol Endocrinol Metab, August 1, 2005; 289(2): E218 - E224. [Abstract] [Full Text] [PDF]

				R. G. Radu, S. Fujimoto, E. Mukai, M. Takehiro, D. Shimono, K. Nabe, M. Shimodahira, R. Kominato, Y. Aramaki, Y. Nishi, et al. Tacrolimus suppresses glucose-induced insulin release from pancreatic islets by reducing glucokinase activity Am J Physiol Endocrinol Metab, February 1, 2005; 288(2): E365 - E371. [Abstract] [Full Text] [PDF]

				M. J. MacDonald, L. A. Fahien, L. J. Brown, N. M. Hasan, J. D. Buss, and M. A. Kendrick Perspective: emerging evidence for signaling roles of mitochondrial anaplerotic products in insulin secretion Am J Physiol Endocrinol Metab, January 1, 2005; 288(1): E1 - E15. [Abstract] [Full Text] [PDF]

				V. Delghingaro-Augusto, F. Ferreira, S. Bordin, M. E. C. do Amaral, M. H. Toyama, A. C. Boschero, and E. M. Carneiro A Low Protein Diet Alters Gene Expression in Rat Pancreatic Islets J. Nutr., February 1, 2004; 134(2): 321 - 327. [Abstract] [Full Text] [PDF]

				J. Shao, L. Qiao, and J. E. Friedman Prolactin, progesterone, and dexamethasone coordinately and adversely regulate glucokinase and cAMP/PDE cascades in MIN6 {beta}-cells Am J Physiol Endocrinol Metab, February 1, 2004; 286(2): E304 - E310. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Gao, Z. Articles by Matschinsky, F. M. Search for Related Content PubMed PubMed Citation Articles by Gao, Z. Articles by Matschinsky, F. M. Pubmed/NCBI databases Compound via MeSH Substance via MeSH Hazardous Substances DB CALCIUM COMPOUNDS CALCIUM, ELEMENTAL GLUCOSE