Effects of intravenous infusion of I7 amino acids on the secretion of GH, glucagon, and insulin in sheep T. KUHARA, S. IKEDA, A. OHNEDA, AND Y. SASAKI Laboratory of Animal Physiology, Faculty of Agriculture, and Third Medical Department, School of Medicine, Tohoku University, Sendai 981, Japan
KUHARA, T., S. IKEDA, A. OHNEDA, AND Y. SASAKI. Effects intravenous infusion of 17 amino acids on the secretion of GH, glucagon, and insulin in sheep. Am. J. Physiol. 260 (Endocrinol. Metab. 23): E21-E26, 1991.-The effects of intravenous infusion of 17 amino acids, each at a dose of 3 mmol/kg over 30 min, on the secretion of insulin, glucagon, and growth hormone (GH) were studied in 6 castrated male sheep. Insulin-like growth factor I (IGF-I) secretion was also studied using eight of the amino acids. Plasma a-amino nitrogen reached a peak at 30 min followed by a gradual decrease thereafter. The greatest increase was obtained using aspartic acid and the smallest with methionine, responses to the remaining amino acids lying between these two. Leucine was the most effective amino acid in stimulating insulin secretion but did not produce any increase in glucagon and GH secretion. Alanine, glycine, and serine induced a greater enhancement of both glucagon and insulin secretion than other amino acids. No amino acid was able to specifically stimulate glucagon secretion without also increasing insulin or GH secretion. With regard to insulin and glucagon secretion, amino acids could be divided into groups according to their R groups. Neutral straight-chain amino acids stimulated both insulin and glucagon secretion, with a greater secretory response to shorter C-chain amino acids. Branchedchain amino acids tended to enhance insulin and suppress glucagon secretion. Acidic amino acids caused an increase in GH secretion. Aspartic acid caused the strongest stimulation of GH secretion, exceeding that induced by arginine. No changes in plasma IGF-I were brought about by any of the amino acids tested. of
metabolic
hormones
AMINO ACIDS are known to affect the secretion of metabolic hormones such as insulin, glucagon, and growth hormone (GH). As these hormones are central to the regulation of amino acid metabolism, it is worthwhile to quantify the extent to which each amino acid can stimulate the secretion of these hormones. It has been reported that each amino acid has different stimulatory effects on insulin and/or glucagon secretion in humans (9), in dog (29), in canine pancreas perfused in situ (25), and in rat pancreas perfused in vitro (1). GH secretion in response to amino acids has also been reported in humans (18, 19). Although differences in experimental design make it difficult to reach definite conclusions from these reports on the magnitude of the stimulatory effects of each amino acid on hormone secretion, there seem to be species differences in the ability of individual amino acids to enhance or abolish hormone secretion.
SEVERAL
0193-1849/91
$1.50 Copyright
Hertelendy et al. (13) have reported that significant differences existed between species in the ability of arginine to stimulate insulin and GH secretion. For ruminant species, the effects of only a few amino acids on the secretion of insulin and GH have been quantified (6). In ruminant animals, dietary carbohydrate is fermented to volatile fatty acids in the reticulorumen, so that only small amounts of glucose are absorbed from the gut under most dietary conditions. The contribution of amino acids to gluconeogenesis could be more important in fed ruminants than in other species. The control of the endocrine system by amino acids would also be of importance in regulating the metabolism of amino acids. Therefore, in the present work, using sheep, we studied the secretory responses of insulin, glucagon, and GH to intravenous infusion of 17 amino acids. The secretion of insulin-like growth factor I (IGF-I) in response to several amino acids was also investigated. MATERIALS
AND METHODS
Animals. Six cross-bred castrated adult male sheep weighing 30-45 kg were used. They were housed in metabolic cages and were offered alfalfa pellets once daily at 2% of body weight/day at 1600 h. Water was available continuously. At least 3 mo before experiments began, animals had their left common carotid artery surgically placed in a loop of skin under general anesthesia with pentobarbitone sodium (25 mg/kg). During the postoperative recovery period, the animals were trained to become accustomed to the experimental procedure and surroundings. At least 1 wk before the experiments, catheters were inserted into the jugular vein through a hypodermic needle. Catheters were kept patent by flushing and filling with 3.8% (wt/vol) sterile solution of trisodium citrate. Experimental procedure. On the morning of an experiment, feed was withdrawn, and at least 1 h before experimentation, an indwelling sterile needle (Venula V2, TOP, Tokyo, Japan) was placed in the exteriorized carotid artery after puncturing and was connected to a catheter. The arterial catheter was used for blood sampling and was filled with sterile trisodium citrate solution. Sheep were intravenously infused through the intravenous catheter with a varying volume of 250 mM amino acid solution at a rate of 3 mmol/kg over a period of 30 min, using a constant infusion pump. The amino acids used for infusion were L-alanine (Ala), L-arginine
0 1991 the American
Physiological
Society
E21
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (128.143.007.175) on August 11, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
E22
(A%) L-asparagine
METABOLIC
HORMONE
SECRETION
(Asn ) monohydrochloride, L-a .spartic L-glutamic acid (Glu), L-glutamine (Gin), glycine (Gly), L- histidine (His) monohydrochloride monohydrate, L-isoleucine (Ile), L-leucine (Leu), L-lysine (Lys) monohydrochloride, L-methionine (Met), L-phenylalanine (Phe), L-proline (Pro), L-serine (Ser), L-threonine (Thr), and L-valine (Val). Each amino acid was dissolved in warm distilled sterile water, and the solution was adjusted to pH 7.4 with sodium hydroxide or hydrochloric acid before use. The solution of Leu was adjusted to pH 2.5 to allow the concentration to be the same as for the other amino acids. Each sheep received all 17 amino acid infusions in a randomized manner at &day intervals. The timing of blood samples was at -20, -10, 0, 10, 20, 30, 45, 60, 75, and 90 min. Analyses. Arterial blood collected into heparinized syringes was immediately transferred into polyethylene test tubes cooled in ice water and centrifuged at 4OC. Each tube contained 0.05 mmol of benzamidine/ml of blood. A portion of plasma was deproteinized with 5% (wt/vol) trichloroacetic acid, and the supernatant was stored at -20°C until glucose was determined by the glucose oxidase method (14) and a-amino nitrogen by the method of Lee and Takahashi (21). A further aliquot of plasma was stored at -20°C for glucagon, insulin, GH, and IGF-I assays. Plasma glucagon was assayed (24) using dextran-coated charcoal and antiserum G-42E, which was highly specific for the COOH-terminal portion of glucagon (26). Crystalline bovine glucagon was used as the standard (Calbiochem, California), and porcine lz51-labeled glucagon was purchased (Du Pont-New England Nuclear, Boston, MA). The assay revealed a minimal detectable concentration of 65 pg/ml. Plasma insulin was assayed by the method of Morgan an d Lazarow (22) with slight modification (30). A highly purified bovine GH used for iodination and as the standard was prepared from bovine pituitaries by the method of Spitsberg (32). Bovine GH was repurified by high-performance liquid chromatography (Triroter SR, Jasco, Tokyo) on a reverse-phase (TSK-gel ODS-120T, 5 pm) column (0.4 x 24 cm) at a column temperature of 40°C and a flow rate of 1 ml/min. Iodination of bovine GH was done by the method of Johke (15). Plasma GH was assayed by the double antibody method using antiserum NIADDK-antioGH-2 (AFP-C0123080). Plasma IGF-I was assayed using the IGF-I reagent pack for radioimmunoassay (Amersham, UK) and dextran-coated charcoal. A recombinant analogue of human IGF-I was used as a standard (Amersham, UK, batch no. 12), because the amino acid sequence of ovine IGF-I is identical to that of human IGF-I (11). Plasma samples were extracted with an acidethanol solution to dissociate IGF-I from binding protein before assay (5). The acid-ethanol-extracted sheep plasma exhibited a displacement of tracer that was nearly parallel to the standard. Calculations and statistics. The data for the concentration of plasma a-amino nitrogen, glucose, insulin, glucagon, GH, and IGF-I were assigned to three periods representing pre-, during, and postinfusion values. The significance of differences between periods within each amino acid infusion were determined by analysis of variance using the General Linear Model (GLM) of the SAS
acid i Asp),
BY
AMINO
ACIDS
IN
SHEEP
program package (SAS, Cary, NC) and by using Duncan’s multiple range test. The incremental hormone areas enclosed by the hormone concentration curves above basal levels were calculated between 0 and 30 min, between 30 and 90 min, and between 0 and 90 min (the values between 0 and 90 min are only shown in Tables l-3). Areas are expressed as PU min-lo ml-’ for insulin, as pg min-l . ml-’ for glucagon and as ng . min-’ ml-l for GH. The significance of differences in areas between amino acids was determined by Duncan’s multiple range test using the GLM procedure of SAS. l
l
l
RESULTS
Plasma wamino nitrogen. The intravenous infusion of each amino acid brought about a significant (P < 0.05) increase in plasma a-amino nitrogen concentration .. The concentration gradually increased during infusion until it reached a maximum value at 30 min, followed by a gradual decrease thereafter, declining to a level of -8 mg N/l00 ml on average for each amino acid, which was slightly above the basal level of 5 mg N/l00 ml at the end of the 90-min experimental period. The increase in a-amino nitrogen differed slightly between amino acids. As shown in Fig. 1, the greatest increase in a-amino nitrogen concentration was observed with the infusion of Asp, and the smallest for Met, with the increase for the remaining amino acids lying between these two (values for the other 15 amino acids are not shown in Fig. 1) . Insulin secretion. Large differences were observed in the responses of plasma insulin concentration to the individual amino acids. Figure 2 represents typical responses of plasma insulin to the intravenous infusion of Leu, Gly, Arg, and Asp. The incremental insulin areas during a period of 90 min after the onset of the infusion of all 17 amino acids are given in Table 1. The effect of Leu, a branched-chain amino acid, on .insulin secretion during the period between 0 and 30 min was the greatest among the amino acids tested, with the plasma insulin concentration increasing from a mean basal value of 18.4 t 5.0 to a peak of 195.8 t 56.3 pU/ml 30 min after the I
-20
1
PLASMA CI-AMI NO
I
I
I
0
30
60
N 1 TROGEN
I
so (min)
FIG. 1. Changes i .n plasma a-amino nitrogen after intravenous infusions of L-aspartic acid (Asp) and L-methionine (Met). Amino acids were infused at a rate of 0.1 mmol kg-‘. min-’ over a period of 30 min. Bars indicate means t SE for 6 animals. Data were assigned to 3 periods representing pre-, during, and postinfusion values. Significant differences between periods within each amino acid infusion were determined by analysis of variance using General Linear Model of the SAS program package and Duncan’s multiple range test. A, O: significant differences between preinfusion and remaining two periods (P c 0.05).
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (128.143.007.175) on August 11, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
METABOLIC I
I
HORMONE PLASMA
I
SECRETION
INSULIN
200
-
100
I -ii =, *
0 i
GLU I
I
I
I
I
-20
0
30
60
90 (min)
2. Effects of intravenous infusion of L-leucine (Leu), L-glycine (Gly), L-arginine (Arg), and L-glutamic acid (Glu) on plasma insulin concentration. See legend to Fig. 1 for experimental and statistical details. V, 0, q I, A: significant differences between preinfusion and remaining 2 periods (P < 0.05). FIG.
1. Incremental insulin areas in response to infusions of 17 amino acids
TABLE
Amino GlY
Leu Ser Ala Phe Met Thr Asn A% Pro LYS
Gln Ile Val His ASP
Glu
Acid
Insulin
Area,
pU - min-l
- ml-l
101.6&19.2* 99.8t38.5* 88.9*18.9* 64.8k7.4”t 37.6+8.3t$ 36.1&11.9t$ 26.8&5.2-f$5 25.5kl3.8-@§ 25.5+4.5-f@ 21.4+2.9$§ 17.2+1.9$§ 15.0+3.4$§ 6.1&0.7$§ 3.7&0.5$§ -1.9+1.2gj -3.6&1.2$§ -7.8k3.75
Values represent means k SE of areas for 6 sheep. Incremental areas enclosed by hormone concentration curves above basal levels were calculated for a period of 90 min from starting infusions of amino acids. Values were separated into groups by Duncan’s multiple range test using the General Linear Model of the SAS program package to try to explain the relationship between amino acid structure and magnitude of hormone response. Means with no common superscripts are significantly different (P c 0.05).
onset of infusion, followed by a gradual decrease thereafter (Fig. 2). The stimulatory effect of the other branched-chain amino acids (Ile and Val) on insulin secretion was very low (Table 1). Insulin secretion induced by Gly, Ser, and Ala was significantly lower (P < 0.05) between 0 and 30 min than with Leu. However, a marked increase in insulin secretion was caused by Gly, Ser, and Ala during the period between 30 and 90 min, levels being significantly higher (P < 0.05) than those for the remaining amino acids (values are not shown) except for Leu, indicating a delayed and prolonged insulin response to these amino acids. Because the patterns of change in plasma insulin concentrations were similar for the infusion of each of these three amino acids, a typical response for Gly infusion is illustrated in Fig. 2, showing that plasma insulin reached its maximal value at 15 min after the cessation of infusion. The moderate enhancement of insulin secretion induced by Arg infusion was around the average response for the amino acids tested (Table 1) and tended to be greater, but not signif-
BY
AMINO
ACIDS
IN
E23
SHEEP
icantly so, than that for other basic amino acids (Lys and His). The concentration of plasma insulin reached a peak value of 57.8 t 11.3 pU/ml 10 min after starting Arg infusion, followed by a sustained significantly (P < 0.05) higher level than that during the preinfusion period (Fig. 2). As shown for Glu infusion in Fig. 2, administration of acidic amino acids brought about slight but significant (P < 0.05) decreases in the plasma insulin concentration. Glucagon secretion. The concentration of plasma glucagon increased immediately after the onset of Ala infusion, followed by a further gradual increase to a level of 411.0 t 51.5 pg/ml at 30 min, compared with the basal value of 137.9 t 20.0 pg/ml (Fig. 3). A further increase in plasma glucagon was observed even after stopping Ala infusion, rising from 539.0 t 96.2 pg/ml at 45 min to 668.4 t 197.1 pg/ml at 90 min. Such a remarkable delayed increase in plasma glucagon after the cessation of infusion was also seen for the infusion of Gly and Ser. The infusion of other neutral amino acids (Asn, Thr, Met, Phe, and Gln) also brought about significant (P < 0.05) increases in plasma glucagon concentration during the period between 0 and 30 min (concentration curves are not shown), followed by maintenance of elevated levels throughout the experimental period. The patterns of change in plasma glucagon concentration induced by these amino acids were similar to those induced by Ala, Gly, and Ser, although the increases in plasma glucagon during the experimental period were significantly (P < 0.05) less for Asn, Thr, Met, Phe, and Gln than for Ala, Gly, and Ser (Table 2). Among the three basic amino acids tested, Arg was a much more effective stimulant of glucagon secretion than was Lys or His, showing a significantly (P c 0.05) increased plasma glucagon concentration throughout the experimental period (Fig. 3). The magnitude of the glucagon secretory response to Arg was somewhat less than responses to Ala, Gly, and Ser. The infusion of the acidic amino acids, Asp and Glu, had no effect on glucagon secretion. Only the infusion of Leu among the 17 amino acids tested caused a statistically significant (P < 0.05) decrease in plasma glucagon concentration during the infusion (Fig. 3). The other branched-chain amino acids (Val and Ile) brought about either no change in glucagon secretion or a trend toward a decrease in the glucagon response area. PLASMA
800
GLUCAGON
r
600 -
400 7 z
a" 200
OL
L
I
I
I
-20
0
30
60
I
90 (min)
FIG. 3. Effects of intravenous infusion of L-alanine (Ala), Arg, and Leu on plasma glucagon concentration. See legend to Fig. 1 for experimental and statistical details. 0, q , V: significant differences between preinfusion and remaining 2 periods (P < 0.05).
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (128.143.007.175) on August 11, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
E24
METABOLIC
HORMONE
SECRETION
BY
AMINO
2. Incre mental glucago na reas in response to infus bions of 1 7 amino acids
TABLE
Amino
Acid
Glucagon
Ala
Area,
pg - min-’
Ser A% Asn Gln Thr Pro Phe Met LYS Val Glu His ASP Leu Ile
* F
ng - min-l
ASP
9.8k2.0"
Phe Glu Gln Asn Ser Leu Ala
3.3+0.8-t 3.2+0.8t$ 2.5+1.2”f$§ 1.9+0.5t$§ 1.2+0.3t$§ 1.2+0.5t$§ 1.1+0.3”f$§ 1.1+0.6”f$§ 1.0+0.4t$§ 0.8+0.3$$ 0.7&0.2§ 0.6kO.55 0.4kO.55 0.4+0.2§ 0.2+0.2gj 0.2+0.4§
GlY
Arg Met Thr Pro Val His LYS
Ile
PLASMA
GH
10
0
FIG. 4. Effects of intravenous infusion of Asp, Arg, plasma growth hormone (GH) concentration. See legend experimental and statistical details. A, q , 0: significant between preinfusion and remaining 2 periods (P < 0.05).
3. Incremental GH areas in response to infusions of 17 amino acids GH Area,
SHEEP
20
TABLE
Acid
IN
30
-7 2
Values are means t SE. See legend to Table 1 for calculations and statistical details. Means with no common letters are significantly different (P < 0.05).
Amino
40
- ml-’
307.5k62.7” 295.5t57.1” 242.5t44.6” 139.2+20.0b 135.8+20.8b 99.6+19.4bc 90.7+11.1bC 83.8k9.5bcd 77.3+15.1bCd” 62.7&8. lbcdef 45.8+14.5cdef 23.8+3.Zcdef 4.3+11.0def 3.7+5.8def -3.5k14.4ef -8.8t5.0f -10.8+5.6f-
GlY
ACIDS
- ml-l
Values are means k SE. GH, growth hormone. See legend to Table 1 for calculations and statistical details. Means with no common superscripts are significantly different (P < 0.05).
GH secretion. The GH assay using our bovine GH as a standard and for iodination was validated on the basis of dose-response parallelism between sheep plasma dilutions compared with tracer displacement by bovine GH. Sheep plasma exhibited a tracer displacement that was parallel to the extracted bovine GH standard. The assay revealed a minimal detectable concentration of 0.1 rig/ml. As shown in Table 3, the intravenous infusion of the acidic amino acid Asp caused the greatest enhancement of GH secretion among the amino acids tested. The plasma GH concentration increased from the basal value of 2.3 t 0.3 rig/ml until it reached a peak value of 30.1 t 6.2 rig/ml 30 min after the onset of Asp infusion, followed by a gradual decrease thereafter (Fig. 4). A moderate increase in GH secretion was caused by the infusion of Glu (acidic amino acid) or Phe (neutral and aromatic amino acid), reaching peak values of 11.2 t 1.9 and 9.1 t 2.0 rig/ml at 30 min, respectively. Both re-
and Gly on to Fig. 1 for differences
sponses were signi ficantly (P < 0.05) less than that to Asp (Table 3). GH responses to most other amino acids were very low. The infusion of Arg, which has been shown to be a potent stimulant for GH secretion in other species, caused a statistically significant (P < 0.05) increase in plasma GH concentration, but its stimulatory effect on GH secretion was very meager. The concentration of plasma GH was unchanged during the infusion of Gly but was slightly increased after stopping infusion (Fig. 4). IGF-I secretion. Plasma IGF-I concentration was only measured for the infusions of Gly, Ala, Met, Phe, Gln, Leu, Arg, and Asp. The basal concentration of plasma IGF-I was -140 rig/ml on average for six animals and remained unaffected by the infusion of these amino acids. PZasma glucose. The responses of plasma glucose concentration to amino acid infusions could be divided roughly into three groups. The first group of amino acids consisting of Gly, Asn, Ser, Ala, Thr, Gln, Arg, Met, Pro, and Phe induced a significant (P < 0.05) increase in plasma glucose concentrations, which reached a peak value between 45 and 60 min, followed by a gradual decrease toward the end of the period of observation. The degree of hyperglycemia induced by these amino acids decreased in the order in which they are listed above. A representative pattern of hyperglycemia for the infusion of Gly is shown in Fig. 5. Ile, Val, Asp, Glu, and Lys, the second group of amino acids, had no effect on plasma glucose (results for Asp only given in Fig. 5). The concentration of plasma glucose was significantly (P < 0.05) decreased by the infusion of Leu and His, the third group of amino acids. As shown in Fig. 5, Leu infusion caused a gradual decrease in plasma glucose during the period of infusion, with hypoglycemia being sustained throughout the experimental period. DISCUSSION
The results of the present experiment show that each amino acid infused intravenously had a different time course of response and potency for the secretion of insulin, glucagon, and GH in sheep. Leu was the most effective amino acid for stimulating insulin secretion. Ala, Gly, and Ser were most effective for glucagon and
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (128.143.007.175) on August 11, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
METABOLIC
HORMONE PLASMA
SECRETION
BY
AMINO
ACIDS
IN
SHEEP
E25
GLUCOSE
acids could be due to depolarization of the plasma membrane (2). The response pattern of glucagon and insulin to the 80 amino acids tested in the present investigation could be classified roughly into groups associated with their R group. Among the neutral straight chain group, short C60 -ichain amino acids (Gly, Ala, and Ser) were greater stimz 00 ulants for glucagon and insulin secretion than long C: 40chain amino acids. In the branched chain group, conE versely, Val (shortest C chain) had scarcely any effects on the secretion of either glucagon or insulin, whereas I 1 I I I -20 0 30 60 90 Leu (longest) stimulated insulin secretion and slightly (min) suppressed glucagon secretion. Arg and Lys, the basic FIG. 5. Effects of intravenous infusion of Gly, Asp, and Leu on amino acids, gave around average responses among the plasma glucose concentration. See legend to Fig. 1 for experimental amino acids tested for both glucagon and insulin secreand statistical details. 0, q : significant differences between preinfusion tion. The acidic amino acids slightly decreased the and remaining 2 periods (P c 0.05). plasma insulin concentration, whereas they did not affect insulin, and Asp was most effective for GH. glucagon secretion. We are unable to explain the reason In the present study, the period of observation used why most amino acids having glucagon-stimulating acwas shorter than the ideal for the dose of amino acids tivity caused further increases in plasma glucagon even used so that the changes in plasma glucagon and insulin after the cessation of infusion, whereas the increased induced by some amino acids were not back to basal by plasma insulin levels gradually decreased once the infuthe end of the experimental periods. We have however sion had ceased. The continued secretion of glucagon concluded that the maintenance of augmented levels of after stopping an amino acid infusion has also been plasma glucagon and insulin produced by some amino reported in conscious dogs (23) and sheep (31). acids were in themselves a reflection of the high potency It is conceivable that there would be some relationship of these amino acids in stimulating hormone secretion. between control of the endocrine system by amino acids Therefore, the magnitude of the effect of each amino and the regulation of amino acid metabolism and glucoacid in stimulating hormone secretion could be compared neogenesis. When amino acids known to be potent glufor each amino acid using the data for hormone areas cogenic precursors such as Ala and Gly are administered, calculated over the period of 90 min. the greater increase in glucagon secretion relative to that It has been well documented that, among amino acids of insulin might favor enhanced glucose production that have been tested, Arg had the greatest ability to through gluconeogenesis from these amino acids, resultaugment insulin secretion in most animal species, as ing in a clear elevation in plasma glucose levels, as reported in humans (9), in the perfused rat pancreas (I), observed in the present investigation. In contrast, the and in the dog pancreas (16, 17), so that Arg has been hyperinsulinemic and hypoglucagonemic responses to commonly used as a standard secretagogue for charac- Leu, a nonglucogenic amino acid, would support the terizing the insulin secretory response to other amino usefulness of branched-chain amino acids in promoting acids or metabolites. We suggest however that it may be protein anabolism by sparing amino acid catabolism in preferable to use Leu rather than Arg as a secretagogue, protein-wasting conditions (27), as well as in enhancing to induce a large stimulation of insulin secretion, at least glucose utilization in peripheral tissues. in castrated male sheep. From the results of the present The present results clearly show that, in sheep, the investigation, it also seems probable that secondary re- greatest enhancement of GH secretion is induced by Asp. sponses to other hormones, such as glucagon, causing As has been generally recognized, Arg was reported to be stimulation of insulin secretion, could be eliminated by the most effective amino acid to stimulate GH secretion using Leu, since glucagon secretion was not augmented in humans (18, 19), although Asp was not employed in by intravenous administration of Leu. that work. It has also been reported in sheep that Arg It has been proposed that amino acids stimulate insulin stimulated a significant increase in GH secretion (6). and glucagon secretions by binding to the transport Our present result that the stimulatory effect of Asp on receptor sites or transport molecules in the plasma mem- GH secretion markedly exceeds that of Arg coincides branes of pancreatic A and B cells, and/or by their with a report for rats (10) in which orally administered transport across the cell membrane (3,8,12,28). Another Asp caused a much greater increase in plasma GH than possibility that has also been proposed is that metabolic that caused by Arg. It is likely that the remarkable products of amino acids may affect the secretory process response of GH secretion to Asp in sheep may not be in the cell. The present result showing an inhibition of due to any species-related differences in the mechanism glucagon secretion in response to Leu might be partly of GH secretion in response to Asp. due to its catabolic metabolite in the A cell, as LeclercqIn our present study, we have failed to observe any Meyer et al. (20) have reported, that a-ketoisocaproate, changes in plasma IGF-I after the infusion of the amino the first catabolic product of Leu, induced an inhibition acids tested. As has been reported by other workers (4, of glucagon release and stimulated insulin release from 7), dietary protein or amino acids could be a factor the perfused rat pancreas. On the other hand, it has been modulating IGF-I secretion, with stimulatory effects only reported that the release of insulin evoked by basic amino being obvious after a longer period of ingestion. Further I
I
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (128.143.007.175) on August 11, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
E26
METABOLIC
HORMONE
SECRETION
studies will be needed to see the effect on IGF-I secretion of amino acid infusion for a longer period of time than that employed in this study. The authors are grateful to H. Kawauchi and assistants, University of Kitazato, for preparation of the bovine growth hormone. The authors are also most grateful to the National Hormone and Pituitary Program (NIADDK) for providing GH antiserum. We would like to thank Y. Ohtomo for care of the animals and skillful technical assistance. We give thanks to Dr. T. E. C. Weekes, University of Newcastle, for comments on the manuscript. This study was supported by a Grant-in-Aid for Scientific Research (B) from the Ministry of Education, Science, and Culture of Japan (no. 62480077). Present address of S. Ikeda: Miyagi Agricultural College, Hatatate, Sendai 982-02, Japan. Address for reprint requests: T. Kuhara, Faculty of Agriculture, Tohoku University l-l, Tsutsumidori Amamiyamati, Aobaku Sendai 981, Japan. Received
7 August
1989; accepted
in final
form
13 August
BY
13.
14. 15.
16.
17.
18.
1990.
REFERENCES G. BALLERIO, J. BOILLOT, AND J. R. 1. ASSAN, R., J. R. ATTALI, GIRARD. Glucagon secretion induced by natural and artificial amino acids in the perfused rat pancreas. Diabetes 26: 300-307, 1977. F., A. MOURTADA, A. SENER, AND W. J. MALAISSE. 2. BLACHIER, Stimulus-secretion coupling of arginine-induced insulin release. Uptake of metabolized and nonmetabolized cationic amino acids by pancreatic islets. Endocrinology 124: 134-141, 1989. H. N., B. HELLMAN, A. LERNMARK, J. SEHLIN, H. 3. CHRISTENSEN, S. TAGER, AND I. TALJEDAL. In vitro stimulation of insulin release by nonmetabolizable, transport-specific amino acids. Biochim. Biophys. Acta 241: 341-348, 1971. D. R., M. M. SEEK, AND L. E. UNDERWOOD. Supple4. CLEMMONS, mental essential amino acids augment the somatomedin-C/insulinlike growth factor I response to refeeding after fasting. Metab. Clin. Exp. 34: 391-395, 1985. 5. DAUGHADAY, W. H., I. K. MARIZ, AND S. L. BLETHEN. Inhibition of access of bound somatomedin to membrane receptor and immunobinding sites: a comparison of radioreceptor and radioimmunoassay of somatomedin in native and acid-ethanol-extracted serum. J. Clin. Endocrinol. Metab. 51: 781-788, 1980. 6. DAVIS, S. L. Plasma levels of prolactin, growth hormone, and insulin in sheep following the infusion of arginine, leucine and phenylalanine. Endocrinology 91: 549-555, 1972. 7. ELSASSER, T. H., T. S. RUMSEY, AND A. C. HAMMOND. Influence of diet on basal and growth hormone-stimulated plasma concentrations of IGF-I in beef cattle. J. Anim. Sci. 67: 128-141, 1989. 8. FAJANS, S. S., R. QUIBRERA, S. PEK, J. C. FLOYD, JR., H. N. CHRISTENSEN, AND J. W. CONN. Stimulation of insulin release in dog by a nonmetabolizable amino acid. Comparison with leucine and arginine. J. CZin. Endocrinol. Metab. 33: 35-41, 1971. 9. FLOYD, J. C., JR., S. S. FAJANS, J. W. CONN, R. F. KNOPF, AND J. RULL. Stimulation of insulin secretion by amino acids. J. CZin. Inuest. 45: 1487-1502, 1966. 10. FRANCHIMONT, P., G. CAMPISTRON, M. H. CREUZET, J. GROS, AND A. LUYCKX. Effects of arginine aspartate and its components on growth hormone, insulin and glucagon secretion in the rat. Arch. Int. Pharmacodyn. Ther. 267: 161-168, 1984. 11. FRANCIS, G. L., K. A. MCNEII,, J. C. WALLACE, F. J. BALLARD, AND P. C. OWENS. Sheep insulin-like growth factor I and II: sequences, activities and assays. Endocrinology 124: 1173-1183, 1989. 12. HELLMAN, B., J. SEHLIN, AND I. TALJEDAL. Transnort of L-leucine
19.
20.
21.
22. 23. 24.
25.
26.
27.
28.
29.
30. 31.
32.
AMINO
ACIDS
IN
SHEEP
and D-leucine into pancreatic ,&cells with reference to the mechanisms of amino acid-induced insulin release. Biochim. Biophys. Acta 266: 436-443,1972. HERTELENDY, F., K. TAKAHA~SHI, L. J. MACHLIN, AND D. M. KIPNIS. Growth hormone and insulin secretory responses to arginine in the sheep, pig, and cow. Gen. Comp. EndocrinoZ. 14: 72-77, 1970. HUGGETT, A. G., AND D. A. NIXON. Enzymic determination of blood glucose (Abstract). Biochem. J. 66: 12, 1957. JOHKE, T. Effects of TRH on circulating growth hormone, prolactin and triiodothyronine levels in the bovine. Endocrinol. Jpn. 25: 19-26, 1978. KANETO, A., AND K. KOSAKA. Stimulation of glucagon secretion by arginine and histidine infused intrapancreatically. EndocrinoZogy 88: 1239-1245,197l. KANETO, A., AND K. KOSAKA. Effects of leucine and isoleucine infused intrapancreatically on glucagon and insulin secretion. Endocrinology 91: 691-695, 1972. KNOPF, R. F., J. W. CONN, S. S. FAJANS, J. C. FLOYD, JR., E. M. GUNTSCHE, AND J. A. RULL. Plasma growth hormone response to intravenous administration of amino acids. J. CZin. EndocrinoZ. Metab. 25: 1140-1144, 1965. KNOPF, R. F., J. W. CONN, J. C. FLOYD, JR., S. S. FAJANS, J. A. RULL, E. A. GUNTSCHE, AND C. A. THIFFAULT. The normal endocrine response to ingestion of protein and infusions of amino acids. Sequential secretion of insulin and growth hormone. Trans. Assoc. Am. Physicians 79: 312-321, 1966. LECLERCQ-MEYER, V., J. MARCHAND, R. LECLERCQ, AND W. J. MALAISSE. Interactions of cu-ketoisocaproate, glucose and arginine in the secretion of glucagon and insulin from the perfused rat pancreas. DiabetoZogia 17: 121-126, 1979. LEE, Y. P., AND T. TAKAHASHI. An improved calorimetric determination of amino acids with the use of ninhydrin. AnaZ. Biochem. 14; 71-77, 1966. MORGAN, C. R., AND A. LAZAROW. Immunoassay of insulin: two antibody systems. Diabetes 12: 115-126, 1963. MULLER, W. A., G. R. FALOONA, AND R. H. UNGER. The effect of alanine on glucagon secretion. J. CZin. Invest. 50: 2215-2218, 1971. OHNEDA, A., S. ISHII, K. HORIGOME, AND S. YAMAGATA. Glucagon response to arginine after treatment of diabetes mellitus. Diabetes 24: 811-819, 1975. OHNEDA, A., S. ISHII, H. ITABASHI, AND K. HORIGOME. Effect of intrapancreatic administration of amino acids upon glucagon secretion in dogs. Tohoku. J. Exp. Med. 130: 227-237, 1980. OHNEDA, A., K. WATANABE, M. WAKIMATSU, AND M. FUJINO. Production of a specific antiserum by synthetic C-terminal fragment of glucagon. Horm. Metab. Res. 11: 463-468,1979. OKADA, A., S. MORI, M. TOTSUKA, K. OKAMOTO, S. USUI, H. FUJITA, T. ITAKURA, AND H. MIZOTE. Branched-chain amino acids metabolic support in surgical patients: a randomized, controlled trial in patients with subtotal or total gastrectomy in 16 Japanese institutions. J. Parenter. Enteral. Nutr. 12: 332-337, 1988. PEK, S., J. C. SANTIAGO, AND T. Y. TAI. L-leucine-induced secretion of glucagon and insulin, and the “off-response” to L-leucine in vitro. I. Characterization of the dynamics of secretion. Endocrinology 103: 1208-1218, 1978. ROCHA, D. M., G. R. FALOONA, AND R. H. UNGER. Glucagonstimulating activity of 20 amino acids in dogs. J. CZin. Inuest. 51: 2346-2351,1972. SASAKI, Y., AND H. TAKAHASHI. Insulin secretion in sheep exposed to cold. J. Physiol. Lond. 306: 323-335, 1980. SASAKI, Y., H. TAKAHASHI, H. Aso, A. OHNEDA, AND T. E. C. WEEKES. Effects of cold exposure on insulin and glucagon secretion in sheep. EndocrinoZogy 111: 2070-2076, 1982. SPITSBERG, V. L. A selective extraction of growth hormone from bovine pituitary gland and its further purification and crystallization. AnaZ. Biochem. 160: 489-495, 1987.
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (128.143.007.175) on August 11, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
KUHARA, T., S. IKEDA, A. OHNEDA, AND Y. SASAKI. Effects intravenous infusion of 17 amino acids on the secretion of GH, glucagon, and insulin in sheep. Am. J. Physiol. 260 (Endocrinol. Metab. 23): E21-E26, 1991.-The effects of intravenous infusion of 17 amino acids, each at a dose of 3 mmol/kg over 30 min, on the secretion of insulin, glucagon, and growth hormone (GH) were studied in 6 castrated male sheep. Insulin-like growth factor I (IGF-I) secretion was also studied using eight of the amino acids. Plasma a-amino nitrogen reached a peak at 30 min followed by a gradual decrease thereafter. The greatest increase was obtained using aspartic acid and the smallest with methionine, responses to the remaining amino acids lying between these two. Leucine was the most effective amino acid in stimulating insulin secretion but did not produce any increase in glucagon and GH secretion. Alanine, glycine, and serine induced a greater enhancement of both glucagon and insulin secretion than other amino acids. No amino acid was able to specifically stimulate glucagon secretion without also increasing insulin or GH secretion. With regard to insulin and glucagon secretion, amino acids could be divided into groups according to their R groups. Neutral straight-chain amino acids stimulated both insulin and glucagon secretion, with a greater secretory response to shorter C-chain amino acids. Branchedchain amino acids tended to enhance insulin and suppress glucagon secretion. Acidic amino acids caused an increase in GH secretion. Aspartic acid caused the strongest stimulation of GH secretion, exceeding that induced by arginine. No changes in plasma IGF-I were brought about by any of the amino acids tested. of
metabolic
hormones
AMINO ACIDS are known to affect the secretion of metabolic hormones such as insulin, glucagon, and growth hormone (GH). As these hormones are central to the regulation of amino acid metabolism, it is worthwhile to quantify the extent to which each amino acid can stimulate the secretion of these hormones. It has been reported that each amino acid has different stimulatory effects on insulin and/or glucagon secretion in humans (9), in dog (29), in canine pancreas perfused in situ (25), and in rat pancreas perfused in vitro (1). GH secretion in response to amino acids has also been reported in humans (18, 19). Although differences in experimental design make it difficult to reach definite conclusions from these reports on the magnitude of the stimulatory effects of each amino acid on hormone secretion, there seem to be species differences in the ability of individual amino acids to enhance or abolish hormone secretion.
SEVERAL
0193-1849/91
$1.50 Copyright
Hertelendy et al. (13) have reported that significant differences existed between species in the ability of arginine to stimulate insulin and GH secretion. For ruminant species, the effects of only a few amino acids on the secretion of insulin and GH have been quantified (6). In ruminant animals, dietary carbohydrate is fermented to volatile fatty acids in the reticulorumen, so that only small amounts of glucose are absorbed from the gut under most dietary conditions. The contribution of amino acids to gluconeogenesis could be more important in fed ruminants than in other species. The control of the endocrine system by amino acids would also be of importance in regulating the metabolism of amino acids. Therefore, in the present work, using sheep, we studied the secretory responses of insulin, glucagon, and GH to intravenous infusion of 17 amino acids. The secretion of insulin-like growth factor I (IGF-I) in response to several amino acids was also investigated. MATERIALS
AND METHODS
Animals. Six cross-bred castrated adult male sheep weighing 30-45 kg were used. They were housed in metabolic cages and were offered alfalfa pellets once daily at 2% of body weight/day at 1600 h. Water was available continuously. At least 3 mo before experiments began, animals had their left common carotid artery surgically placed in a loop of skin under general anesthesia with pentobarbitone sodium (25 mg/kg). During the postoperative recovery period, the animals were trained to become accustomed to the experimental procedure and surroundings. At least 1 wk before the experiments, catheters were inserted into the jugular vein through a hypodermic needle. Catheters were kept patent by flushing and filling with 3.8% (wt/vol) sterile solution of trisodium citrate. Experimental procedure. On the morning of an experiment, feed was withdrawn, and at least 1 h before experimentation, an indwelling sterile needle (Venula V2, TOP, Tokyo, Japan) was placed in the exteriorized carotid artery after puncturing and was connected to a catheter. The arterial catheter was used for blood sampling and was filled with sterile trisodium citrate solution. Sheep were intravenously infused through the intravenous catheter with a varying volume of 250 mM amino acid solution at a rate of 3 mmol/kg over a period of 30 min, using a constant infusion pump. The amino acids used for infusion were L-alanine (Ala), L-arginine
0 1991 the American
Physiological
Society
E21
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (128.143.007.175) on August 11, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
E22
(A%) L-asparagine
METABOLIC
HORMONE
SECRETION
(Asn ) monohydrochloride, L-a .spartic L-glutamic acid (Glu), L-glutamine (Gin), glycine (Gly), L- histidine (His) monohydrochloride monohydrate, L-isoleucine (Ile), L-leucine (Leu), L-lysine (Lys) monohydrochloride, L-methionine (Met), L-phenylalanine (Phe), L-proline (Pro), L-serine (Ser), L-threonine (Thr), and L-valine (Val). Each amino acid was dissolved in warm distilled sterile water, and the solution was adjusted to pH 7.4 with sodium hydroxide or hydrochloric acid before use. The solution of Leu was adjusted to pH 2.5 to allow the concentration to be the same as for the other amino acids. Each sheep received all 17 amino acid infusions in a randomized manner at &day intervals. The timing of blood samples was at -20, -10, 0, 10, 20, 30, 45, 60, 75, and 90 min. Analyses. Arterial blood collected into heparinized syringes was immediately transferred into polyethylene test tubes cooled in ice water and centrifuged at 4OC. Each tube contained 0.05 mmol of benzamidine/ml of blood. A portion of plasma was deproteinized with 5% (wt/vol) trichloroacetic acid, and the supernatant was stored at -20°C until glucose was determined by the glucose oxidase method (14) and a-amino nitrogen by the method of Lee and Takahashi (21). A further aliquot of plasma was stored at -20°C for glucagon, insulin, GH, and IGF-I assays. Plasma glucagon was assayed (24) using dextran-coated charcoal and antiserum G-42E, which was highly specific for the COOH-terminal portion of glucagon (26). Crystalline bovine glucagon was used as the standard (Calbiochem, California), and porcine lz51-labeled glucagon was purchased (Du Pont-New England Nuclear, Boston, MA). The assay revealed a minimal detectable concentration of 65 pg/ml. Plasma insulin was assayed by the method of Morgan an d Lazarow (22) with slight modification (30). A highly purified bovine GH used for iodination and as the standard was prepared from bovine pituitaries by the method of Spitsberg (32). Bovine GH was repurified by high-performance liquid chromatography (Triroter SR, Jasco, Tokyo) on a reverse-phase (TSK-gel ODS-120T, 5 pm) column (0.4 x 24 cm) at a column temperature of 40°C and a flow rate of 1 ml/min. Iodination of bovine GH was done by the method of Johke (15). Plasma GH was assayed by the double antibody method using antiserum NIADDK-antioGH-2 (AFP-C0123080). Plasma IGF-I was assayed using the IGF-I reagent pack for radioimmunoassay (Amersham, UK) and dextran-coated charcoal. A recombinant analogue of human IGF-I was used as a standard (Amersham, UK, batch no. 12), because the amino acid sequence of ovine IGF-I is identical to that of human IGF-I (11). Plasma samples were extracted with an acidethanol solution to dissociate IGF-I from binding protein before assay (5). The acid-ethanol-extracted sheep plasma exhibited a displacement of tracer that was nearly parallel to the standard. Calculations and statistics. The data for the concentration of plasma a-amino nitrogen, glucose, insulin, glucagon, GH, and IGF-I were assigned to three periods representing pre-, during, and postinfusion values. The significance of differences between periods within each amino acid infusion were determined by analysis of variance using the General Linear Model (GLM) of the SAS
acid i Asp),
BY
AMINO
ACIDS
IN
SHEEP
program package (SAS, Cary, NC) and by using Duncan’s multiple range test. The incremental hormone areas enclosed by the hormone concentration curves above basal levels were calculated between 0 and 30 min, between 30 and 90 min, and between 0 and 90 min (the values between 0 and 90 min are only shown in Tables l-3). Areas are expressed as PU min-lo ml-’ for insulin, as pg min-l . ml-’ for glucagon and as ng . min-’ ml-l for GH. The significance of differences in areas between amino acids was determined by Duncan’s multiple range test using the GLM procedure of SAS. l
l
l
RESULTS
Plasma wamino nitrogen. The intravenous infusion of each amino acid brought about a significant (P < 0.05) increase in plasma a-amino nitrogen concentration .. The concentration gradually increased during infusion until it reached a maximum value at 30 min, followed by a gradual decrease thereafter, declining to a level of -8 mg N/l00 ml on average for each amino acid, which was slightly above the basal level of 5 mg N/l00 ml at the end of the 90-min experimental period. The increase in a-amino nitrogen differed slightly between amino acids. As shown in Fig. 1, the greatest increase in a-amino nitrogen concentration was observed with the infusion of Asp, and the smallest for Met, with the increase for the remaining amino acids lying between these two (values for the other 15 amino acids are not shown in Fig. 1) . Insulin secretion. Large differences were observed in the responses of plasma insulin concentration to the individual amino acids. Figure 2 represents typical responses of plasma insulin to the intravenous infusion of Leu, Gly, Arg, and Asp. The incremental insulin areas during a period of 90 min after the onset of the infusion of all 17 amino acids are given in Table 1. The effect of Leu, a branched-chain amino acid, on .insulin secretion during the period between 0 and 30 min was the greatest among the amino acids tested, with the plasma insulin concentration increasing from a mean basal value of 18.4 t 5.0 to a peak of 195.8 t 56.3 pU/ml 30 min after the I
-20
1
PLASMA CI-AMI NO
I
I
I
0
30
60
N 1 TROGEN
I
so (min)
FIG. 1. Changes i .n plasma a-amino nitrogen after intravenous infusions of L-aspartic acid (Asp) and L-methionine (Met). Amino acids were infused at a rate of 0.1 mmol kg-‘. min-’ over a period of 30 min. Bars indicate means t SE for 6 animals. Data were assigned to 3 periods representing pre-, during, and postinfusion values. Significant differences between periods within each amino acid infusion were determined by analysis of variance using General Linear Model of the SAS program package and Duncan’s multiple range test. A, O: significant differences between preinfusion and remaining two periods (P c 0.05).
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (128.143.007.175) on August 11, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
METABOLIC I
I
HORMONE PLASMA
I
SECRETION
INSULIN
200
-
100
I -ii =, *
0 i
GLU I
I
I
I
I
-20
0
30
60
90 (min)
2. Effects of intravenous infusion of L-leucine (Leu), L-glycine (Gly), L-arginine (Arg), and L-glutamic acid (Glu) on plasma insulin concentration. See legend to Fig. 1 for experimental and statistical details. V, 0, q I, A: significant differences between preinfusion and remaining 2 periods (P < 0.05). FIG.
1. Incremental insulin areas in response to infusions of 17 amino acids
TABLE
Amino GlY
Leu Ser Ala Phe Met Thr Asn A% Pro LYS
Gln Ile Val His ASP
Glu
Acid
Insulin
Area,
pU - min-l
- ml-l
101.6&19.2* 99.8t38.5* 88.9*18.9* 64.8k7.4”t 37.6+8.3t$ 36.1&11.9t$ 26.8&5.2-f$5 25.5kl3.8-@§ 25.5+4.5-f@ 21.4+2.9$§ 17.2+1.9$§ 15.0+3.4$§ 6.1&0.7$§ 3.7&0.5$§ -1.9+1.2gj -3.6&1.2$§ -7.8k3.75
Values represent means k SE of areas for 6 sheep. Incremental areas enclosed by hormone concentration curves above basal levels were calculated for a period of 90 min from starting infusions of amino acids. Values were separated into groups by Duncan’s multiple range test using the General Linear Model of the SAS program package to try to explain the relationship between amino acid structure and magnitude of hormone response. Means with no common superscripts are significantly different (P c 0.05).
onset of infusion, followed by a gradual decrease thereafter (Fig. 2). The stimulatory effect of the other branched-chain amino acids (Ile and Val) on insulin secretion was very low (Table 1). Insulin secretion induced by Gly, Ser, and Ala was significantly lower (P < 0.05) between 0 and 30 min than with Leu. However, a marked increase in insulin secretion was caused by Gly, Ser, and Ala during the period between 30 and 90 min, levels being significantly higher (P < 0.05) than those for the remaining amino acids (values are not shown) except for Leu, indicating a delayed and prolonged insulin response to these amino acids. Because the patterns of change in plasma insulin concentrations were similar for the infusion of each of these three amino acids, a typical response for Gly infusion is illustrated in Fig. 2, showing that plasma insulin reached its maximal value at 15 min after the cessation of infusion. The moderate enhancement of insulin secretion induced by Arg infusion was around the average response for the amino acids tested (Table 1) and tended to be greater, but not signif-
BY
AMINO
ACIDS
IN
E23
SHEEP
icantly so, than that for other basic amino acids (Lys and His). The concentration of plasma insulin reached a peak value of 57.8 t 11.3 pU/ml 10 min after starting Arg infusion, followed by a sustained significantly (P < 0.05) higher level than that during the preinfusion period (Fig. 2). As shown for Glu infusion in Fig. 2, administration of acidic amino acids brought about slight but significant (P < 0.05) decreases in the plasma insulin concentration. Glucagon secretion. The concentration of plasma glucagon increased immediately after the onset of Ala infusion, followed by a further gradual increase to a level of 411.0 t 51.5 pg/ml at 30 min, compared with the basal value of 137.9 t 20.0 pg/ml (Fig. 3). A further increase in plasma glucagon was observed even after stopping Ala infusion, rising from 539.0 t 96.2 pg/ml at 45 min to 668.4 t 197.1 pg/ml at 90 min. Such a remarkable delayed increase in plasma glucagon after the cessation of infusion was also seen for the infusion of Gly and Ser. The infusion of other neutral amino acids (Asn, Thr, Met, Phe, and Gln) also brought about significant (P < 0.05) increases in plasma glucagon concentration during the period between 0 and 30 min (concentration curves are not shown), followed by maintenance of elevated levels throughout the experimental period. The patterns of change in plasma glucagon concentration induced by these amino acids were similar to those induced by Ala, Gly, and Ser, although the increases in plasma glucagon during the experimental period were significantly (P < 0.05) less for Asn, Thr, Met, Phe, and Gln than for Ala, Gly, and Ser (Table 2). Among the three basic amino acids tested, Arg was a much more effective stimulant of glucagon secretion than was Lys or His, showing a significantly (P c 0.05) increased plasma glucagon concentration throughout the experimental period (Fig. 3). The magnitude of the glucagon secretory response to Arg was somewhat less than responses to Ala, Gly, and Ser. The infusion of the acidic amino acids, Asp and Glu, had no effect on glucagon secretion. Only the infusion of Leu among the 17 amino acids tested caused a statistically significant (P < 0.05) decrease in plasma glucagon concentration during the infusion (Fig. 3). The other branched-chain amino acids (Val and Ile) brought about either no change in glucagon secretion or a trend toward a decrease in the glucagon response area. PLASMA
800
GLUCAGON
r
600 -
400 7 z
a" 200
OL
L
I
I
I
-20
0
30
60
I
90 (min)
FIG. 3. Effects of intravenous infusion of L-alanine (Ala), Arg, and Leu on plasma glucagon concentration. See legend to Fig. 1 for experimental and statistical details. 0, q , V: significant differences between preinfusion and remaining 2 periods (P < 0.05).
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (128.143.007.175) on August 11, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
E24
METABOLIC
HORMONE
SECRETION
BY
AMINO
2. Incre mental glucago na reas in response to infus bions of 1 7 amino acids
TABLE
Amino
Acid
Glucagon
Ala
Area,
pg - min-’
Ser A% Asn Gln Thr Pro Phe Met LYS Val Glu His ASP Leu Ile
* F
ng - min-l
ASP
9.8k2.0"
Phe Glu Gln Asn Ser Leu Ala
3.3+0.8-t 3.2+0.8t$ 2.5+1.2”f$§ 1.9+0.5t$§ 1.2+0.3t$§ 1.2+0.5t$§ 1.1+0.3”f$§ 1.1+0.6”f$§ 1.0+0.4t$§ 0.8+0.3$$ 0.7&0.2§ 0.6kO.55 0.4kO.55 0.4+0.2§ 0.2+0.2gj 0.2+0.4§
GlY
Arg Met Thr Pro Val His LYS
Ile
PLASMA
GH
10
0
FIG. 4. Effects of intravenous infusion of Asp, Arg, plasma growth hormone (GH) concentration. See legend experimental and statistical details. A, q , 0: significant between preinfusion and remaining 2 periods (P < 0.05).
3. Incremental GH areas in response to infusions of 17 amino acids GH Area,
SHEEP
20
TABLE
Acid
IN
30
-7 2
Values are means t SE. See legend to Table 1 for calculations and statistical details. Means with no common letters are significantly different (P < 0.05).
Amino
40
- ml-’
307.5k62.7” 295.5t57.1” 242.5t44.6” 139.2+20.0b 135.8+20.8b 99.6+19.4bc 90.7+11.1bC 83.8k9.5bcd 77.3+15.1bCd” 62.7&8. lbcdef 45.8+14.5cdef 23.8+3.Zcdef 4.3+11.0def 3.7+5.8def -3.5k14.4ef -8.8t5.0f -10.8+5.6f-
GlY
ACIDS
- ml-l
Values are means k SE. GH, growth hormone. See legend to Table 1 for calculations and statistical details. Means with no common superscripts are significantly different (P < 0.05).
GH secretion. The GH assay using our bovine GH as a standard and for iodination was validated on the basis of dose-response parallelism between sheep plasma dilutions compared with tracer displacement by bovine GH. Sheep plasma exhibited a tracer displacement that was parallel to the extracted bovine GH standard. The assay revealed a minimal detectable concentration of 0.1 rig/ml. As shown in Table 3, the intravenous infusion of the acidic amino acid Asp caused the greatest enhancement of GH secretion among the amino acids tested. The plasma GH concentration increased from the basal value of 2.3 t 0.3 rig/ml until it reached a peak value of 30.1 t 6.2 rig/ml 30 min after the onset of Asp infusion, followed by a gradual decrease thereafter (Fig. 4). A moderate increase in GH secretion was caused by the infusion of Glu (acidic amino acid) or Phe (neutral and aromatic amino acid), reaching peak values of 11.2 t 1.9 and 9.1 t 2.0 rig/ml at 30 min, respectively. Both re-
and Gly on to Fig. 1 for differences
sponses were signi ficantly (P < 0.05) less than that to Asp (Table 3). GH responses to most other amino acids were very low. The infusion of Arg, which has been shown to be a potent stimulant for GH secretion in other species, caused a statistically significant (P < 0.05) increase in plasma GH concentration, but its stimulatory effect on GH secretion was very meager. The concentration of plasma GH was unchanged during the infusion of Gly but was slightly increased after stopping infusion (Fig. 4). IGF-I secretion. Plasma IGF-I concentration was only measured for the infusions of Gly, Ala, Met, Phe, Gln, Leu, Arg, and Asp. The basal concentration of plasma IGF-I was -140 rig/ml on average for six animals and remained unaffected by the infusion of these amino acids. PZasma glucose. The responses of plasma glucose concentration to amino acid infusions could be divided roughly into three groups. The first group of amino acids consisting of Gly, Asn, Ser, Ala, Thr, Gln, Arg, Met, Pro, and Phe induced a significant (P < 0.05) increase in plasma glucose concentrations, which reached a peak value between 45 and 60 min, followed by a gradual decrease toward the end of the period of observation. The degree of hyperglycemia induced by these amino acids decreased in the order in which they are listed above. A representative pattern of hyperglycemia for the infusion of Gly is shown in Fig. 5. Ile, Val, Asp, Glu, and Lys, the second group of amino acids, had no effect on plasma glucose (results for Asp only given in Fig. 5). The concentration of plasma glucose was significantly (P < 0.05) decreased by the infusion of Leu and His, the third group of amino acids. As shown in Fig. 5, Leu infusion caused a gradual decrease in plasma glucose during the period of infusion, with hypoglycemia being sustained throughout the experimental period. DISCUSSION
The results of the present experiment show that each amino acid infused intravenously had a different time course of response and potency for the secretion of insulin, glucagon, and GH in sheep. Leu was the most effective amino acid for stimulating insulin secretion. Ala, Gly, and Ser were most effective for glucagon and
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (128.143.007.175) on August 11, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
METABOLIC
HORMONE PLASMA
SECRETION
BY
AMINO
ACIDS
IN
SHEEP
E25
GLUCOSE
acids could be due to depolarization of the plasma membrane (2). The response pattern of glucagon and insulin to the 80 amino acids tested in the present investigation could be classified roughly into groups associated with their R group. Among the neutral straight chain group, short C60 -ichain amino acids (Gly, Ala, and Ser) were greater stimz 00 ulants for glucagon and insulin secretion than long C: 40chain amino acids. In the branched chain group, conE versely, Val (shortest C chain) had scarcely any effects on the secretion of either glucagon or insulin, whereas I 1 I I I -20 0 30 60 90 Leu (longest) stimulated insulin secretion and slightly (min) suppressed glucagon secretion. Arg and Lys, the basic FIG. 5. Effects of intravenous infusion of Gly, Asp, and Leu on amino acids, gave around average responses among the plasma glucose concentration. See legend to Fig. 1 for experimental amino acids tested for both glucagon and insulin secreand statistical details. 0, q : significant differences between preinfusion tion. The acidic amino acids slightly decreased the and remaining 2 periods (P c 0.05). plasma insulin concentration, whereas they did not affect insulin, and Asp was most effective for GH. glucagon secretion. We are unable to explain the reason In the present study, the period of observation used why most amino acids having glucagon-stimulating acwas shorter than the ideal for the dose of amino acids tivity caused further increases in plasma glucagon even used so that the changes in plasma glucagon and insulin after the cessation of infusion, whereas the increased induced by some amino acids were not back to basal by plasma insulin levels gradually decreased once the infuthe end of the experimental periods. We have however sion had ceased. The continued secretion of glucagon concluded that the maintenance of augmented levels of after stopping an amino acid infusion has also been plasma glucagon and insulin produced by some amino reported in conscious dogs (23) and sheep (31). acids were in themselves a reflection of the high potency It is conceivable that there would be some relationship of these amino acids in stimulating hormone secretion. between control of the endocrine system by amino acids Therefore, the magnitude of the effect of each amino and the regulation of amino acid metabolism and glucoacid in stimulating hormone secretion could be compared neogenesis. When amino acids known to be potent glufor each amino acid using the data for hormone areas cogenic precursors such as Ala and Gly are administered, calculated over the period of 90 min. the greater increase in glucagon secretion relative to that It has been well documented that, among amino acids of insulin might favor enhanced glucose production that have been tested, Arg had the greatest ability to through gluconeogenesis from these amino acids, resultaugment insulin secretion in most animal species, as ing in a clear elevation in plasma glucose levels, as reported in humans (9), in the perfused rat pancreas (I), observed in the present investigation. In contrast, the and in the dog pancreas (16, 17), so that Arg has been hyperinsulinemic and hypoglucagonemic responses to commonly used as a standard secretagogue for charac- Leu, a nonglucogenic amino acid, would support the terizing the insulin secretory response to other amino usefulness of branched-chain amino acids in promoting acids or metabolites. We suggest however that it may be protein anabolism by sparing amino acid catabolism in preferable to use Leu rather than Arg as a secretagogue, protein-wasting conditions (27), as well as in enhancing to induce a large stimulation of insulin secretion, at least glucose utilization in peripheral tissues. in castrated male sheep. From the results of the present The present results clearly show that, in sheep, the investigation, it also seems probable that secondary re- greatest enhancement of GH secretion is induced by Asp. sponses to other hormones, such as glucagon, causing As has been generally recognized, Arg was reported to be stimulation of insulin secretion, could be eliminated by the most effective amino acid to stimulate GH secretion using Leu, since glucagon secretion was not augmented in humans (18, 19), although Asp was not employed in by intravenous administration of Leu. that work. It has also been reported in sheep that Arg It has been proposed that amino acids stimulate insulin stimulated a significant increase in GH secretion (6). and glucagon secretions by binding to the transport Our present result that the stimulatory effect of Asp on receptor sites or transport molecules in the plasma mem- GH secretion markedly exceeds that of Arg coincides branes of pancreatic A and B cells, and/or by their with a report for rats (10) in which orally administered transport across the cell membrane (3,8,12,28). Another Asp caused a much greater increase in plasma GH than possibility that has also been proposed is that metabolic that caused by Arg. It is likely that the remarkable products of amino acids may affect the secretory process response of GH secretion to Asp in sheep may not be in the cell. The present result showing an inhibition of due to any species-related differences in the mechanism glucagon secretion in response to Leu might be partly of GH secretion in response to Asp. due to its catabolic metabolite in the A cell, as LeclercqIn our present study, we have failed to observe any Meyer et al. (20) have reported, that a-ketoisocaproate, changes in plasma IGF-I after the infusion of the amino the first catabolic product of Leu, induced an inhibition acids tested. As has been reported by other workers (4, of glucagon release and stimulated insulin release from 7), dietary protein or amino acids could be a factor the perfused rat pancreas. On the other hand, it has been modulating IGF-I secretion, with stimulatory effects only reported that the release of insulin evoked by basic amino being obvious after a longer period of ingestion. Further I
I
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (128.143.007.175) on August 11, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
E26
METABOLIC
HORMONE
SECRETION
studies will be needed to see the effect on IGF-I secretion of amino acid infusion for a longer period of time than that employed in this study. The authors are grateful to H. Kawauchi and assistants, University of Kitazato, for preparation of the bovine growth hormone. The authors are also most grateful to the National Hormone and Pituitary Program (NIADDK) for providing GH antiserum. We would like to thank Y. Ohtomo for care of the animals and skillful technical assistance. We give thanks to Dr. T. E. C. Weekes, University of Newcastle, for comments on the manuscript. This study was supported by a Grant-in-Aid for Scientific Research (B) from the Ministry of Education, Science, and Culture of Japan (no. 62480077). Present address of S. Ikeda: Miyagi Agricultural College, Hatatate, Sendai 982-02, Japan. Address for reprint requests: T. Kuhara, Faculty of Agriculture, Tohoku University l-l, Tsutsumidori Amamiyamati, Aobaku Sendai 981, Japan. Received
7 August
1989; accepted
in final
form
13 August
BY
13.
14. 15.
16.
17.
18.
1990.
REFERENCES G. BALLERIO, J. BOILLOT, AND J. R. 1. ASSAN, R., J. R. ATTALI, GIRARD. Glucagon secretion induced by natural and artificial amino acids in the perfused rat pancreas. Diabetes 26: 300-307, 1977. F., A. MOURTADA, A. SENER, AND W. J. MALAISSE. 2. BLACHIER, Stimulus-secretion coupling of arginine-induced insulin release. Uptake of metabolized and nonmetabolized cationic amino acids by pancreatic islets. Endocrinology 124: 134-141, 1989. H. N., B. HELLMAN, A. LERNMARK, J. SEHLIN, H. 3. CHRISTENSEN, S. TAGER, AND I. TALJEDAL. In vitro stimulation of insulin release by nonmetabolizable, transport-specific amino acids. Biochim. Biophys. Acta 241: 341-348, 1971. D. R., M. M. SEEK, AND L. E. UNDERWOOD. Supple4. CLEMMONS, mental essential amino acids augment the somatomedin-C/insulinlike growth factor I response to refeeding after fasting. Metab. Clin. Exp. 34: 391-395, 1985. 5. DAUGHADAY, W. H., I. K. MARIZ, AND S. L. BLETHEN. Inhibition of access of bound somatomedin to membrane receptor and immunobinding sites: a comparison of radioreceptor and radioimmunoassay of somatomedin in native and acid-ethanol-extracted serum. J. Clin. Endocrinol. Metab. 51: 781-788, 1980. 6. DAVIS, S. L. Plasma levels of prolactin, growth hormone, and insulin in sheep following the infusion of arginine, leucine and phenylalanine. Endocrinology 91: 549-555, 1972. 7. ELSASSER, T. H., T. S. RUMSEY, AND A. C. HAMMOND. Influence of diet on basal and growth hormone-stimulated plasma concentrations of IGF-I in beef cattle. J. Anim. Sci. 67: 128-141, 1989. 8. FAJANS, S. S., R. QUIBRERA, S. PEK, J. C. FLOYD, JR., H. N. CHRISTENSEN, AND J. W. CONN. Stimulation of insulin release in dog by a nonmetabolizable amino acid. Comparison with leucine and arginine. J. CZin. Endocrinol. Metab. 33: 35-41, 1971. 9. FLOYD, J. C., JR., S. S. FAJANS, J. W. CONN, R. F. KNOPF, AND J. RULL. Stimulation of insulin secretion by amino acids. J. CZin. Inuest. 45: 1487-1502, 1966. 10. FRANCHIMONT, P., G. CAMPISTRON, M. H. CREUZET, J. GROS, AND A. LUYCKX. Effects of arginine aspartate and its components on growth hormone, insulin and glucagon secretion in the rat. Arch. Int. Pharmacodyn. Ther. 267: 161-168, 1984. 11. FRANCIS, G. L., K. A. MCNEII,, J. C. WALLACE, F. J. BALLARD, AND P. C. OWENS. Sheep insulin-like growth factor I and II: sequences, activities and assays. Endocrinology 124: 1173-1183, 1989. 12. HELLMAN, B., J. SEHLIN, AND I. TALJEDAL. Transnort of L-leucine
19.
20.
21.
22. 23. 24.
25.
26.
27.
28.
29.
30. 31.
32.
AMINO
ACIDS
IN
SHEEP
and D-leucine into pancreatic ,&cells with reference to the mechanisms of amino acid-induced insulin release. Biochim. Biophys. Acta 266: 436-443,1972. HERTELENDY, F., K. TAKAHA~SHI, L. J. MACHLIN, AND D. M. KIPNIS. Growth hormone and insulin secretory responses to arginine in the sheep, pig, and cow. Gen. Comp. EndocrinoZ. 14: 72-77, 1970. HUGGETT, A. G., AND D. A. NIXON. Enzymic determination of blood glucose (Abstract). Biochem. J. 66: 12, 1957. JOHKE, T. Effects of TRH on circulating growth hormone, prolactin and triiodothyronine levels in the bovine. Endocrinol. Jpn. 25: 19-26, 1978. KANETO, A., AND K. KOSAKA. Stimulation of glucagon secretion by arginine and histidine infused intrapancreatically. EndocrinoZogy 88: 1239-1245,197l. KANETO, A., AND K. KOSAKA. Effects of leucine and isoleucine infused intrapancreatically on glucagon and insulin secretion. Endocrinology 91: 691-695, 1972. KNOPF, R. F., J. W. CONN, S. S. FAJANS, J. C. FLOYD, JR., E. M. GUNTSCHE, AND J. A. RULL. Plasma growth hormone response to intravenous administration of amino acids. J. CZin. EndocrinoZ. Metab. 25: 1140-1144, 1965. KNOPF, R. F., J. W. CONN, J. C. FLOYD, JR., S. S. FAJANS, J. A. RULL, E. A. GUNTSCHE, AND C. A. THIFFAULT. The normal endocrine response to ingestion of protein and infusions of amino acids. Sequential secretion of insulin and growth hormone. Trans. Assoc. Am. Physicians 79: 312-321, 1966. LECLERCQ-MEYER, V., J. MARCHAND, R. LECLERCQ, AND W. J. MALAISSE. Interactions of cu-ketoisocaproate, glucose and arginine in the secretion of glucagon and insulin from the perfused rat pancreas. DiabetoZogia 17: 121-126, 1979. LEE, Y. P., AND T. TAKAHASHI. An improved calorimetric determination of amino acids with the use of ninhydrin. AnaZ. Biochem. 14; 71-77, 1966. MORGAN, C. R., AND A. LAZAROW. Immunoassay of insulin: two antibody systems. Diabetes 12: 115-126, 1963. MULLER, W. A., G. R. FALOONA, AND R. H. UNGER. The effect of alanine on glucagon secretion. J. CZin. Invest. 50: 2215-2218, 1971. OHNEDA, A., S. ISHII, K. HORIGOME, AND S. YAMAGATA. Glucagon response to arginine after treatment of diabetes mellitus. Diabetes 24: 811-819, 1975. OHNEDA, A., S. ISHII, H. ITABASHI, AND K. HORIGOME. Effect of intrapancreatic administration of amino acids upon glucagon secretion in dogs. Tohoku. J. Exp. Med. 130: 227-237, 1980. OHNEDA, A., K. WATANABE, M. WAKIMATSU, AND M. FUJINO. Production of a specific antiserum by synthetic C-terminal fragment of glucagon. Horm. Metab. Res. 11: 463-468,1979. OKADA, A., S. MORI, M. TOTSUKA, K. OKAMOTO, S. USUI, H. FUJITA, T. ITAKURA, AND H. MIZOTE. Branched-chain amino acids metabolic support in surgical patients: a randomized, controlled trial in patients with subtotal or total gastrectomy in 16 Japanese institutions. J. Parenter. Enteral. Nutr. 12: 332-337, 1988. PEK, S., J. C. SANTIAGO, AND T. Y. TAI. L-leucine-induced secretion of glucagon and insulin, and the “off-response” to L-leucine in vitro. I. Characterization of the dynamics of secretion. Endocrinology 103: 1208-1218, 1978. ROCHA, D. M., G. R. FALOONA, AND R. H. UNGER. Glucagonstimulating activity of 20 amino acids in dogs. J. CZin. Inuest. 51: 2346-2351,1972. SASAKI, Y., AND H. TAKAHASHI. Insulin secretion in sheep exposed to cold. J. Physiol. Lond. 306: 323-335, 1980. SASAKI, Y., H. TAKAHASHI, H. Aso, A. OHNEDA, AND T. E. C. WEEKES. Effects of cold exposure on insulin and glucagon secretion in sheep. EndocrinoZogy 111: 2070-2076, 1982. SPITSBERG, V. L. A selective extraction of growth hormone from bovine pituitary gland and its further purification and crystallization. AnaZ. Biochem. 160: 489-495, 1987.
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (128.143.007.175) on August 11, 2018. Copyright © 1991 American Physiological Society. All rights reserved.
