0013-7227/90/1272-0107$02.00/0 Endocrinology Copyright© 1990 by The Endocrine Society
Vol. 127, No. 1 Printed in U.S.A.
Stimulus-Secretion Coupling of Arginine-Induced Insulin Release: Comparison with Histidine-Induced Insulin Release* A. SENER, F. BLACHIER, J. RASSCHAERT, AND W. J. MALAISSE Laboratory of Experimental Medicine, Brussels Free University, Brussels B-1000, Belgium
ABSTRACT. L-Histidine, when tested at a 10-mM concentration, caused a rapid and sustained stimulation of insulin release from rat islets exposed to either D-glucose (7.0 or 8.3 mM) or Lleucine (10.0 mM). The stimulation of insulin release could not be ascribed to an increase in oxygen uptake, to the generation of histamine from L-histidine, or to its participation in a transglutaminase-catalyzed reaction. Like other cationic amino acids, however, L-histidine rapidly accumulated in islet cells, increased ^Rb outflow from prelabeled islets perifused in the presence or absence of extracellular Ca2+, and stimulated the entry of Ca2+
T
HE INSULINOTROPIC action of cationic amino acids, such as L-arginine or L-ornithine, is currently ascribed to the accumulation of these positively charged molecules in pancreatic islet cells, leading to depolarization of the plasma membrane and subsequent gating of voltage-sensitive Ca2+ channels (1-3). Although Lhistidine is also ranged among basic amino acids, the pK for its side chain amounts to 6.0, so that at pH 7.0 the ratio of ionized to nonionized side chains does not exceed 1:10, as distinct from 3.2 x 103 for L-ornithine and 3.2 X 105 for L-arginine. In the present study, advantage was taken of the weakly basic imidazolium function of Lhistidine to further evaluate the relevance of positive charges on the side chain of cationic amino acids to their insulinotropic action.
into islet cells. Yet, the amount of exogenous L-histidine present in the islet cells with a positively charged side chain was estimated to be below the threshold value required for stimulation of insulin release by fully ionized cationic amino acids, such as L-arginine. Hence, the present findings argue against the view that the insulinotropic action of cationic amino acids is solely attributable to the accumulation of these positively charged molecules inside the islet B cell with subsequent depolarization of the plasma membrane. {Endocrinology 127: 107-113, 1990)
islets prelabeled with either [t/-14C]palmitate (8) or L-[£/-14C] glutamine (9), the net uptake 46Ca (10), and the efflux of 45Ca (11) or ^Rb (12) from prelabeled islets were all described in the cited references. The method used to measure the uptake of either unlabeled cationic amino acids or L-[[/-14C]histidine was identical to that previously described (3). For the separation of L-histidine and histamine, however, the mobile phase used in the reversed-phase HPLC consisted of a mixture in equal volume of the buffers described as A and B by Seiler and Knodgen (13). The activity of transglutaminase in islet homogenates was measured as described elsewhere (14). All results are presented as the mean value (±SEM) together with the number of individual determinations (n). The statistical significance of differences between mean values was assessed by use of Student's t test.
Results
Materials and Methods
Secretory data
All experiments were conducted in islets isolated by the collagenase method from the pancreas of fed Wistar female rats (4). The methods used to measure the release of insulin in incubated (4) or perifused (5) islets, the oxidation of exogenous nutrients (6), the uptake of O2 (7), the output of 14CO2 from
In static experiments, over 90-min incubation, L-histidine (10 mM) failed to affect insulin release from islets deprived of any other exogenous nutrient (Table 1, Exp 1). In the presence of D-glucose (7.0 mM), however, Lhistidine stimulated insulin output ( P < 0.001), provided that the concentration of the amino acid amounted to 10 mM (Table 1, Exp 2). The insulinotropic action of Lhistidine was less marked (P < 0.001) than that of either L-arginine or L-lysine, all cationic amino acids being tested at the same concentration (10 mM). L-Histidine (10 mM) also augmented insulin release (P < 0.005) from
Received January 18,1990. Address all correspondence and requests for reprints to: Willy J. Malaisse, Laboratory of Experimental Medicine, Brussels Free University, 115 Boulevard de Waterloo, Brussels B-1000, Belgium. * This work was supported by grants from the Belgian Foundation for Scientific Medical Research and Belgian Ministry of Scientific Policy, and a predoctoral fellowship (to J.R.) from the Belgian Institute for Scientific Research in Industry and Agriculture. 107
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 19 November 2015. at 11:06 For personal use only. No other uses without permission. . All rights reserved.
108
HISTIDINE-INDUCED INSULIN RELEASE
islets incubated either at a higher concentration of Dglucose (8.3 mM) or in the presence of L-leucine (Table 1, Exp 3). Because of the possible generation of histamine from exogenous L-histidine, the effect of the amine was also investigated (Table 1, Exp 4). Up to a concentration of 2.0 mM, histamine failed to significantly affect the release of insulin evoked by D-glucose (8.3 mM). However, at a concentration of 10 mM, histamine augmented the secretory response to the hexose (P < 0.001). TABLE 1. Effect of L-histidine and histamine upon insulin release Agent(s)
Exp
Insulin output (MU/islet-90min)
(mM)
1
None L-Histidine (10.0)
15.1 ± 4.8 (9) 12.9 ± 2.0 ( 9 )
2
D-Glucose (7.0) D-Glucose (7.0) + D-Glucose (7.0) + D-Glucose (7.0) + D-Glucose (7.0) + D-Glucose (7.0) + D-Glucose (7.0) +
34.6 ± 33.0 ± 35.2 ± 33.1 ± 48.6 ± 63.1 ± 65.8 ±
L-histidine (0.5) L-histidine (2.0) L-histidine (5.0) L-histidine (10.0) L-arginine (10.0) L-lysine (10.0)
3
D-Glucose (8.3) D-Glucose (8.3) + L-histidine (10.0) L-Leucine (10.0) L-Leucine (10.0) + L-histidine (10.0)
4
D-Glucose (8.3) D-Glucose (8.3) + D-Glucose (8.3) + D-Glucose (8.3) + D-Glucose (8.3) + D-Glucose (8.3) +
histamine histamine histamine histamine histamine
63.0 ± 102.9 ± 49.9 ± 78.6 ± 61.5 ± 55.4 ± 56.8 ± 64.8 ± 69.6 ± 97.2 ±
(0.2) (0.4) (1.0) (2.0) (10.0)
2.1 3.8 1.5 2.7 1.3 1.7 1.3
(30) (10) (30) (20) (30) (30) (30)
4.6 (8) 11.0'(9) 5.0 (18) 3.2 (18) 4.4 5.8 4.3 4.1 6.3 6.3
(15) (15) (15) (14) (15) (14)
1.5 r
5 to
0.5
0L
30
40
50 TIME(min)
60
70
FIG. 1. Time course for the secretory response to L-histidine (10 mM) of islets perifused in the presence of 7.0 mM D-glucose. Mean values (±SEM) refer to 12 individual experiments.
Endo • 1990 Vol 127 • No 1
The dynamics of the secretory response to L-histidine are illustrated in Fig. 1. In the presence of D-glucose (7.0 mM), the secretory rate averaged after 45 min of perifusion 576 ± 51 nU/islet-min (n = 12). The administration of L-histidine (10 mM) provoked an immediate stimulation of insulin release. The peak response was reached within 2 min, being 818 ± 5 1 nU/islet-min higher than the paired control value (n = 12). During the late period of exposure to L-histidine, the output of insulin slowly declined, but even after 24 min of exposure to the amino acid, it remained 209 ± 35 nU/islet-min higher than the control value reached at the 45th min of perifusion (Fig. 1). Uptake and fate of L- fU-uC]histidine Preliminary experiments indicated that, over 10-min incubation, the net uptake of unlabeled L-histidine (10 mM) was not different from that of L-arginine (also 10 mM), with an overall mean value of 37.9 ± 8.6 pmol/islet (n = 6). In the islets exposed to L-arginine, the cellular ornithine/arginine ratio, after subtraction of the control readings found in islets deprived of exogenous amino acid, averaged 0.31 ± 0.03 (n = 2), in fair agreement with a prior observation (3). In the case of L-histidine, the control readings recorded in islets not exposed to any exogenous amino acid represented 50.3 ± 2.5% (n = 4) of the increment in fluorescence attributable to the uptake of exogenous L-histidine by the islet cells. Because of such a high background, further experiments were conducted with L-[f/-14C]histidine. As shown in Fig. 2, the net uptake of L-[£/-14C]histidine was concentration-related in the 0.5- to 10.0-mM range. Virtually all cellular radioactivity was identified by HPLC as true L-histidine. In the islets exposed to L[U-14C]histidine (0.5 mM), no radioactivity was found at the elution site of [3H(G)]histamine. It was calculated that the limit of detection for 14C-labeled histamine was somewhat below 0.1 pmol/islet. The possible incorporation of L-[U-UC]histidine into protein as catalyzed by the Ca2+-sensitive enzyme transglutaminase was explored in islet homogenates (Table 2). The limited incorporation of radioactivity into trichloroacetic acid (TCA) precipitable material could not be ascribed to a transglutaminase-catalyzed process, it being unaffected by either a change in Ca2+ concentration or the prior heating of the islet homogenate for 10 min at 85 C. In contrast, the incorporation of [3H(G)]histamine was inhibited in the absence of Ca2+, in the presence of glycine methylester, or after heating of the islet homogenate. A Ca2+-responsive and glycine methylestersensitive incorporation of radioactivity into TCA-precipitable material was also observed in the presence of L[£/-14C]arginine, which is rapidly converted to L-orni-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 19 November 2015. at 11:06 For personal use only. No other uses without permission. . All rights reserved.
HISTIDINE-INDUCED INSULIN RELEASE
2
5 L- [u-l4c]histidine(mM)
FIG. 2. Uptake of L-[t/-14C]histidine (10 mM) over 10 min incubation. Mean values (±SEM) are derived from 4-6 individual determinations and refer to the total radioactive content (O and —) and L-histidine content ( • and — ) of PCA extracts, after correction for extracellular contamination.
thine in islet homogenates (15). L-Ornithine is known to act as a substrate for transglutaminase (16).
109
stricted to 20 min, the oxidation of D-[6-14C]glucose was virtually identical in the absence (2.81 ± 0.14 pmol/islet) or presence (2.90 ± 0.43 pmol/islet) of L-histidine (10 mM; n = 7 in each case). The basal rate of O2 uptake, as measured at low HCO3 concentration (
Vol. 127, No. 1 Printed in U.S.A.
Stimulus-Secretion Coupling of Arginine-Induced Insulin Release: Comparison with Histidine-Induced Insulin Release* A. SENER, F. BLACHIER, J. RASSCHAERT, AND W. J. MALAISSE Laboratory of Experimental Medicine, Brussels Free University, Brussels B-1000, Belgium
ABSTRACT. L-Histidine, when tested at a 10-mM concentration, caused a rapid and sustained stimulation of insulin release from rat islets exposed to either D-glucose (7.0 or 8.3 mM) or Lleucine (10.0 mM). The stimulation of insulin release could not be ascribed to an increase in oxygen uptake, to the generation of histamine from L-histidine, or to its participation in a transglutaminase-catalyzed reaction. Like other cationic amino acids, however, L-histidine rapidly accumulated in islet cells, increased ^Rb outflow from prelabeled islets perifused in the presence or absence of extracellular Ca2+, and stimulated the entry of Ca2+
T
HE INSULINOTROPIC action of cationic amino acids, such as L-arginine or L-ornithine, is currently ascribed to the accumulation of these positively charged molecules in pancreatic islet cells, leading to depolarization of the plasma membrane and subsequent gating of voltage-sensitive Ca2+ channels (1-3). Although Lhistidine is also ranged among basic amino acids, the pK for its side chain amounts to 6.0, so that at pH 7.0 the ratio of ionized to nonionized side chains does not exceed 1:10, as distinct from 3.2 x 103 for L-ornithine and 3.2 X 105 for L-arginine. In the present study, advantage was taken of the weakly basic imidazolium function of Lhistidine to further evaluate the relevance of positive charges on the side chain of cationic amino acids to their insulinotropic action.
into islet cells. Yet, the amount of exogenous L-histidine present in the islet cells with a positively charged side chain was estimated to be below the threshold value required for stimulation of insulin release by fully ionized cationic amino acids, such as L-arginine. Hence, the present findings argue against the view that the insulinotropic action of cationic amino acids is solely attributable to the accumulation of these positively charged molecules inside the islet B cell with subsequent depolarization of the plasma membrane. {Endocrinology 127: 107-113, 1990)
islets prelabeled with either [t/-14C]palmitate (8) or L-[£/-14C] glutamine (9), the net uptake 46Ca (10), and the efflux of 45Ca (11) or ^Rb (12) from prelabeled islets were all described in the cited references. The method used to measure the uptake of either unlabeled cationic amino acids or L-[[/-14C]histidine was identical to that previously described (3). For the separation of L-histidine and histamine, however, the mobile phase used in the reversed-phase HPLC consisted of a mixture in equal volume of the buffers described as A and B by Seiler and Knodgen (13). The activity of transglutaminase in islet homogenates was measured as described elsewhere (14). All results are presented as the mean value (±SEM) together with the number of individual determinations (n). The statistical significance of differences between mean values was assessed by use of Student's t test.
Results
Materials and Methods
Secretory data
All experiments were conducted in islets isolated by the collagenase method from the pancreas of fed Wistar female rats (4). The methods used to measure the release of insulin in incubated (4) or perifused (5) islets, the oxidation of exogenous nutrients (6), the uptake of O2 (7), the output of 14CO2 from
In static experiments, over 90-min incubation, L-histidine (10 mM) failed to affect insulin release from islets deprived of any other exogenous nutrient (Table 1, Exp 1). In the presence of D-glucose (7.0 mM), however, Lhistidine stimulated insulin output ( P < 0.001), provided that the concentration of the amino acid amounted to 10 mM (Table 1, Exp 2). The insulinotropic action of Lhistidine was less marked (P < 0.001) than that of either L-arginine or L-lysine, all cationic amino acids being tested at the same concentration (10 mM). L-Histidine (10 mM) also augmented insulin release (P < 0.005) from
Received January 18,1990. Address all correspondence and requests for reprints to: Willy J. Malaisse, Laboratory of Experimental Medicine, Brussels Free University, 115 Boulevard de Waterloo, Brussels B-1000, Belgium. * This work was supported by grants from the Belgian Foundation for Scientific Medical Research and Belgian Ministry of Scientific Policy, and a predoctoral fellowship (to J.R.) from the Belgian Institute for Scientific Research in Industry and Agriculture. 107
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 19 November 2015. at 11:06 For personal use only. No other uses without permission. . All rights reserved.
108
HISTIDINE-INDUCED INSULIN RELEASE
islets incubated either at a higher concentration of Dglucose (8.3 mM) or in the presence of L-leucine (Table 1, Exp 3). Because of the possible generation of histamine from exogenous L-histidine, the effect of the amine was also investigated (Table 1, Exp 4). Up to a concentration of 2.0 mM, histamine failed to significantly affect the release of insulin evoked by D-glucose (8.3 mM). However, at a concentration of 10 mM, histamine augmented the secretory response to the hexose (P < 0.001). TABLE 1. Effect of L-histidine and histamine upon insulin release Agent(s)
Exp
Insulin output (MU/islet-90min)
(mM)
1
None L-Histidine (10.0)
15.1 ± 4.8 (9) 12.9 ± 2.0 ( 9 )
2
D-Glucose (7.0) D-Glucose (7.0) + D-Glucose (7.0) + D-Glucose (7.0) + D-Glucose (7.0) + D-Glucose (7.0) + D-Glucose (7.0) +
34.6 ± 33.0 ± 35.2 ± 33.1 ± 48.6 ± 63.1 ± 65.8 ±
L-histidine (0.5) L-histidine (2.0) L-histidine (5.0) L-histidine (10.0) L-arginine (10.0) L-lysine (10.0)
3
D-Glucose (8.3) D-Glucose (8.3) + L-histidine (10.0) L-Leucine (10.0) L-Leucine (10.0) + L-histidine (10.0)
4
D-Glucose (8.3) D-Glucose (8.3) + D-Glucose (8.3) + D-Glucose (8.3) + D-Glucose (8.3) + D-Glucose (8.3) +
histamine histamine histamine histamine histamine
63.0 ± 102.9 ± 49.9 ± 78.6 ± 61.5 ± 55.4 ± 56.8 ± 64.8 ± 69.6 ± 97.2 ±
(0.2) (0.4) (1.0) (2.0) (10.0)
2.1 3.8 1.5 2.7 1.3 1.7 1.3
(30) (10) (30) (20) (30) (30) (30)
4.6 (8) 11.0'(9) 5.0 (18) 3.2 (18) 4.4 5.8 4.3 4.1 6.3 6.3
(15) (15) (15) (14) (15) (14)
1.5 r
5 to
0.5
0L
30
40
50 TIME(min)
60
70
FIG. 1. Time course for the secretory response to L-histidine (10 mM) of islets perifused in the presence of 7.0 mM D-glucose. Mean values (±SEM) refer to 12 individual experiments.
Endo • 1990 Vol 127 • No 1
The dynamics of the secretory response to L-histidine are illustrated in Fig. 1. In the presence of D-glucose (7.0 mM), the secretory rate averaged after 45 min of perifusion 576 ± 51 nU/islet-min (n = 12). The administration of L-histidine (10 mM) provoked an immediate stimulation of insulin release. The peak response was reached within 2 min, being 818 ± 5 1 nU/islet-min higher than the paired control value (n = 12). During the late period of exposure to L-histidine, the output of insulin slowly declined, but even after 24 min of exposure to the amino acid, it remained 209 ± 35 nU/islet-min higher than the control value reached at the 45th min of perifusion (Fig. 1). Uptake and fate of L- fU-uC]histidine Preliminary experiments indicated that, over 10-min incubation, the net uptake of unlabeled L-histidine (10 mM) was not different from that of L-arginine (also 10 mM), with an overall mean value of 37.9 ± 8.6 pmol/islet (n = 6). In the islets exposed to L-arginine, the cellular ornithine/arginine ratio, after subtraction of the control readings found in islets deprived of exogenous amino acid, averaged 0.31 ± 0.03 (n = 2), in fair agreement with a prior observation (3). In the case of L-histidine, the control readings recorded in islets not exposed to any exogenous amino acid represented 50.3 ± 2.5% (n = 4) of the increment in fluorescence attributable to the uptake of exogenous L-histidine by the islet cells. Because of such a high background, further experiments were conducted with L-[f/-14C]histidine. As shown in Fig. 2, the net uptake of L-[£/-14C]histidine was concentration-related in the 0.5- to 10.0-mM range. Virtually all cellular radioactivity was identified by HPLC as true L-histidine. In the islets exposed to L[U-14C]histidine (0.5 mM), no radioactivity was found at the elution site of [3H(G)]histamine. It was calculated that the limit of detection for 14C-labeled histamine was somewhat below 0.1 pmol/islet. The possible incorporation of L-[U-UC]histidine into protein as catalyzed by the Ca2+-sensitive enzyme transglutaminase was explored in islet homogenates (Table 2). The limited incorporation of radioactivity into trichloroacetic acid (TCA) precipitable material could not be ascribed to a transglutaminase-catalyzed process, it being unaffected by either a change in Ca2+ concentration or the prior heating of the islet homogenate for 10 min at 85 C. In contrast, the incorporation of [3H(G)]histamine was inhibited in the absence of Ca2+, in the presence of glycine methylester, or after heating of the islet homogenate. A Ca2+-responsive and glycine methylestersensitive incorporation of radioactivity into TCA-precipitable material was also observed in the presence of L[£/-14C]arginine, which is rapidly converted to L-orni-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 19 November 2015. at 11:06 For personal use only. No other uses without permission. . All rights reserved.
HISTIDINE-INDUCED INSULIN RELEASE
2
5 L- [u-l4c]histidine(mM)
FIG. 2. Uptake of L-[t/-14C]histidine (10 mM) over 10 min incubation. Mean values (±SEM) are derived from 4-6 individual determinations and refer to the total radioactive content (O and —) and L-histidine content ( • and — ) of PCA extracts, after correction for extracellular contamination.
thine in islet homogenates (15). L-Ornithine is known to act as a substrate for transglutaminase (16).
109
stricted to 20 min, the oxidation of D-[6-14C]glucose was virtually identical in the absence (2.81 ± 0.14 pmol/islet) or presence (2.90 ± 0.43 pmol/islet) of L-histidine (10 mM; n = 7 in each case). The basal rate of O2 uptake, as measured at low HCO3 concentration (
