Effects of pharmacologic hyperglucagonemia on plasma amino acid concentrations in normal and diabetic man

Effects

of Pharmacologic Hyperglucagonemia on Plasma Amino Acid Concentrations in Normal and Diabetic Man J. E. Liljenquist, S. 3. Lewis, A. D. Cherrington, B. C. Sinclair-Smith, and W. W. Lacy

Four normal and five insulin dependent diabetic men received a 2 h pharmacologic glucagon infusion 60 ng/kg/minj resulting in plasma glucagon levels (4400 pg/mlI similar to those seen in glucagonoma patients. In normal subjects in whom plasma insulin concentrations rose significantly (239 uU/ml) and the blood level of 16 of the 18 amino acids measured fell significantly. In contrast, in the diabetic men who secreted no insulin in response to glucagon (no rise in C-peptide levels), only 10 of 18 amino acid levels fell significantly. The branched chain amino acids valine. leucine and isoleucine. as well as tyrosine and phenylalanine were among the 8 amino acids which showed no change in response to glucagon in the diabetics. Thus, glucagon appears to have no acute affect on branched chain amino acid levels in man.

I

N RECENT YEARS a condition characterized by an (Y cell tumor of the pancreas and markedly elevated plasma glucagon levels has been described.‘q2 Patients with this syndrome have been reported to have plasma glucagon concentrations ranging from 850-5000 pg/ml.‘** A second common feature of this disease state is profound hypoaminoacidemia. Mallinson et al.’ reported 9 subjects with glucagonomas and noted uniform suppression ‘of all amino acids ,levels, including the branched chain amino acids. These data, however, contrast with the data from studies in which glucagon was acutely administered and resulted in decrements in the plasma levels of only certain amino acids.3-” The effects of glucagon per se on amino acid metabolism are further clouded by the fact that glucagon administration often results in hyperinsulinemia which itself has been reported to lower plasma amino acid levels.‘2-‘4 The present study was undertaken to dissect apart the effects of hyperglucagonemia from those of concomitant hyperinsulinemia on plasma amino acid levels. This was accomplished by administering glucagon to normal men who responded with hyperinsulinemia and to insulin dependent diabetic men in whom insulin secretion was not stimulated by glucagon. MATERIALS AND METHODS

Subjects Four normal and five insulin-dependent diabetic men participated in this study. All subjects were screened prior to study and were accepted for study only when found to fulfill the following criteria. The normal subjects had no personal history of diabetes mellitus and both groups had no history of other endocrine or major disease. All normal subjects had a normal standard 3 hr oral glucose tolerance teat (40 grams of glucose per square meter of body surface area). The diabetics participating in the study were insulin-dependent, ketosis-prone, juvenile-onset patients. All subjects had normal hepatic function as assessed by measurements of SGOT, serum bilirubin. alkaline phosphatase, and BSP extraction. The complete blood count, serum urea nitrogen and urinalysis were normal in all subjects. Cardiovascular function was normal as assessed by history, physical examination, chest x-ray, and electrocardiogram. The normal and diabetic men were matched for age, height and weight. All normal subjects were on a 300 gram carbohydrate diet

~erabolism,

vol.

30, No. 12 (December). 1981

for 3 days prior to study while the diabetics adhered to their maintenance calorie intake which averaged 250 grams of carbohydrate per day. All diabetics in this study received a morning injection of an intermediate-acting insulin preparation (NPH or Lente) 24 hr prior to study. In addition two of them received 5-10 units of regular insulin at approximately 9:00 p.m. on the evening prior to the experiment in order to achieve comparable degrees of glycemic control in each subject. All subjects had fasted 12-14 hr at the time of study. The nature, purpose and possible risks of the procedure were fully explained to the subjects before obtaining their voluntary consent.

Materials and Preparation Glucagon (Eli Lilly Laboratories, Indianapolis, Ind.) and Trasylol (FBS Pharmaceuticals, Inc., New York, N.Y.) were used in this study. Phadebas Insulin Radioimmunoassay kit was purchased from Pharmacia (Uppsala, Sweden). Glucagon was mixed in saline in a sterile 100 ml volumetric flask containing 2-3 ml of the subject’s blood, which was added to eliminate binding of the glucagon to the glassware.

Surgical Procedure A Teflon catheter was inserted percutaneously in the left brachial artery. After a 30-45 min basal period, a 2-hour intravenous glucagon infusion (50 ng/kg of body wt/min) was begun. Arterial blood samples were taken frequently throughout the study.

Analytical Methods All blood samples were processed immediately after being obtained. Blood glucose was determined within 4 min using the Technicon Autoanalyzer employing the Hoffman ferricyanide reaction. After immediate centrifugation, plasma was chilled in ice for later determination of C-peptide reactivity,“* immunoreactive glucago#t and immunoreactive insulin.” Plasma samples for

*Assayed by Dr. Arthur Rubenstein, Chicago. tAssayed by Dr. Roger Unger, Dallas. From the Howord Hughes Medico1 Institute Luborotory. ond the Deportments of Medicine and Physiology. Vanderbilt University School of Medicine. Nashville. Tennessee. Received for publication December 24.1980. Address reprittr requests to Dr. Alan D. Cherrington. Deportment of Physiology. Vanderbilt University School of Medicine, Nashville. Tennessee 37232. 8 I981 by Prune & Stratton, Inc. 0026-0495/81/3012-0009$01.00~0

1195

1196

LILJENQUIST ET AL.

amino acid analysis were deproteinized by the addition of 5 volumes of 1% picric acid and the filtrate stored at -70°C until analyzed. The analyses were carried out on a Technicon amino acid analyzer as modified by Kopecky to include three 3-mm columns’s This modification allowed for the processing of 3 plasma samples per 24 hr. Each column was monitored by a separate photometer at a wave length of 570 mu and a light path of I5 mm. As proline develops very little color at this wave length, this amino acid was not quantitated. Methionine peaks were too small to quantitate and tryptophan was not recovered. The remaining 18 amino acids were well separated in this system and peak areas were measured with the Technicon Integrator. Student’s t test and where applicable the paired t test were used to determine statistical significance.”

GLUCAGON

I

d

1

-20

0

1

Fig. 1. blood

1

/

I

1

60

90

120

The

glucose

effect level

of glucagon

in normal

and

infusion

(60 “g/kg-min)

diabetic

humans

on the

off insulin

plasma levels of the gluconeogenic amino acids alanine, glycine and threonine were slightly lower (p < .05) than normal in the diabetic subjects while the combined plasma concentration of branched chain amino acids was higher (p < .05). Administration of glucagon in the normal subjects resulted in a significant fall in I5 out of 18 individual amino acids (only citrulline, half cystine, and aminobutryic acid did not fall). The mean decline for all 18 amino acids was 55%. In the diabetic subjects, however, only 10 of 18 amino acids fell significantly. In addition to citrulline, half-cystine and aminobutyric acid, the amino acids tyrosine and phenylalanine and the three branched chain amino acids did not fall in response to glucagon. When comparing the 10 amino acids that declined in both groups, the fall was always greater in the normal subjects (59% in the normal subjects versus 39% in the diabetics). DISCUSSION

in the Normal

The effect of glucagon on plasma amino acids levels has been examined previously. Bromer and Chance’ injected single doses of glucagon into rabbits and

Subjects

Response

and Plasma

to Glucagon

Glucagon

Concentrations

in the Diabetic

Subjects

in

Administration

Time, min -20

o*

5

15

35 Normal

60

90

165 ? 54

239 -f 120

120

Subjects

Plasma Insulin (pU/ml) 24 + 6

24 + 7

38 + 8

76 -+ 26

125 + 38

178 + 115

Diabetic Subjects Plasma Glucagon (pg/ml) 133 + 24

139*22

lGlucagon infusion (50

2900

+ 134

for

24 hr.

Tables II and III depict the effect of glucagon infusion on the plasma concentrations of 18 amino acids in normal and diabetic man. The combined Concentrations

L

30

Plasma Amino Acid Levels After Glucagon

Insulin

1

MINUTES

In the diabetic subjects, the 50 ng/kg/min glucagon infusion resulted in an arterial plasma glucagon concentration of 4420 pg/ml (1.3 x 10m9 M) (Table 1). This concentration was achieved within 15 min and was maintained thereafter. Glucagon was not measured in the normal subjects. The mean plasma immunoreactive insulin (IRI) level rose progressively throughout the glucagon infusion period in the normal subjects, reaching a peak of 239 pU/ml (1.6 x 1O-9 M) at 90 min (Table 1). Endogenous anti-insulin antibodies interfered with the IRI measurements in the diabetic subjects and C-peptide levels were, therefore, measured as an index of secretion of endogenous insulin.” C-peptide reactivity rose in parallel with the IRI in the normal subjects but did not rise in the diabetics, thus indicating that they did not secrete insulin in response to glucagon. The mean blood glucose concentration rose from 88 f 3 to 177 f 11 mg/dl during glucagon infusion in the normals and from 202 + 42 to 266 + 39 mg/dl in the diabetics (Fig. 1).

Plasma

1

I

’

0

RESULTS

1.

1

diabetics (n=5)

Plasma Glucagon, Insulin, C-Peptide. and Blood Glucose Values

Table

(50 ng/kg/min)

4220

f 188

ng/kg/min) begun at zero tmva in all subjects.

4325

+ 718

4420

2 273

._

4420

r 297

1197

GLUCAGON AND PLASMA AMINO ACID LEVELS

Table 2. Effect of Giucagon (56 ng/kg/min)

on Plasma Amino Acid Levels in Four Normal Postabsorptive Men Time, min

0*

46

Pt

120 123 f 9

Alenine

253 f 45

163*33

<.Ol

Glutamine

741 + 78

477 f 70

<.Ol

Lysine

141 f 9

89 + 8

<.Ol

Glycine

203 i 31

138 f 36

Threonine

106i5

64 f 5

Glutamic Acid

89 * 4

Serine

103 f; 6

Valine

199 f 17

70*

10

+ 28

<.Ol

58 f 4.5

<.Ol

<.Ol

90 + 18

<.Ol

<.Ol

41 f 1.3

c.01

34 f 5

<.Ol

NS

279.5

PS <.Ol

59 f 4

<.Ol

39 f 3

c.01

173 f 17

<.05

129 + 9

<.Ol <.Ol

Leucine

98 + 7

71 f 10

<.05

46 * 3

lsoleucine

44 f 4

33 f 6

NS

20 f 1

c.01

Arginine

70 f 6

49 f 7

<.Ol

29 * 4

<.Ol <.Ol

Phenylalanine

44 f 5

36 + 3

NS

25 t 2

Tyrosine

35 f 3

33 * 3

NS

19 * 1

<.Ol

Histidine

75 f 1

53 * 3

<.Ol

33 + 3

c.01

Omithine

59 + 3

35 + 3

<.Ol

21 f 3

<.Ol

Citrulline

26 f 4

27 ? 9

NS

24*

Half Cystine

33 * 3

31+2

NS

32 f 4

Aminobutyric Acid

23 & 10

Total .Gknxgon

2342

* 132

17

NS NS

21.8

f 10.4

NS

19.0 * 12

NS

1823

& 147

<.Ol

1082 + 81

<.Ol

infusion (50 ng/kg/min) begun at zero time and continued 120 min.

tPvalue comparing zero time amino acid levels with amino acid levels at 45 min. $P value comparing zero time amino acid levels with amino acid levels at 120 min.

noted a significant fall in the plasma level of most amino acids. Similar data have been obtained in experiments performed in newborn rats,9 in human infants,* and in normal man.“.” In all of these studies, however, hyperglycemia and hyperinsulinemia accompanied glucagon administration thus preventing clear resolution of the hormones action. In the present study glucagon was infused into diabetic man in order to circumvent modulation caused by counteregulatory Table 3. The Effect of Glucagon (60 nglkglmin)

insulin release. Lack of insulin secretion was confirmed by the demonstration that C-peptide levels did not change in response to glucagon. Under these conditions glucagon caused a significant decline in the level of the gluconeogenic amino acids but failed to change the level of the branched chain amino acids. Since the level of the branched chain amino acids fell in response to glucagon in normal man and since insulin has been shown to reduce the level of branched

on Plasma Amino Acid Levels in Five Insulin Dependent Postabsorptive Diabetic Men Time. min Pt

00

45

Alanine

206 ? 20

151 & 12

<.02

162 f 22

NS <.Ol

120

PS

Glutamine

750 + 46

535 + 31

<.Ol

440 + 28

LysilW

175 f 18

123 i 9

<.Ol

96 f 10

<.02

Glycine

164 & 12

118 * 8

<.Ol

102 f 10

<.02

Threwtine

91 f 14

67 + 10

<.Ol

58 f 0

<.Ol

Glutamic Acid

90 f 9

54 f 5

<.02

43 f 6

<.Ol

Swine

109 f 16

74 * 7

<.02

60 + 8

<.02

Valine

282 zt 45

290 * 41

NS

290 * 42

NS

Leucina

140 * 19

144 * 17

NS

144 f 20

NS

67 + 12

NS

lsdeucine

62+

10

Arginine

55 f 10

43 f 7

<.05

67+

11

35 + 8

NS <.05

Phenylelenine

44 f 5

43 f 8

NS

37 f 5

NS

Twosine

41*3

40 i 4

NS

35 f 5

NS

Histidine

70 * 7

56 + 5

1.02

43 f 6

<.02

Omithine

47 f 7

37 + 5

<.05

27 + 4

4.01

Citrulline

24 f 3

24 f 3

NS

13 f 3

NS

Half Cystine

44 f 2

43 * 3

NS

35 + 5

NS

Aminobutyric Acid Total

lGlucagon infusion (50

31 * 5 2425

it 179

27 f 4 1936 + 136

ne/kg/min) begun at zero time and continued 120 min.

tPvalue comparing zero time amino acid levels with amino acid levels et 45 min. $P valuecomparing zero time amino acid levels with amino acid levels et 120 min.

NS <.Ol

21 f 3 1708 + 155

NS <.Ol

1198

LILJENQUIST ET AL.

chain amino acids2’-24 it seems likely that the fall in BCAA triggered by glucagon in the normal individual may have been indirect occuring in response to the newly released insulin. Marliss et a1.,6 infused glucagon into 5-6 wk fasted man at a rate of 0.1 mg/24 hr for 48 hr. When infused at this rate the hormone significantly lowered the plasma levels of threonine, serine, proline, glycine, alanine, aminobutyric acid, half-cystine, tyrosine and ornithine but not of BCAA. Interestingly the plasma insulin level did not rise significantly in the above study supporting our conclusion that the glucagon induced fall in BCAA which occurs when the hormone is injected into normal man is due to mobilized insulin. In addition the data in the present study indicate that the level of 10 amino acids which did fall in response to glucagon in the diabetics (alanine, glutamine, lysine, glycine, threonine, glutamic acid, serine, histidine, arginine and ornithine) fell more in response to combined hyperinsulinemia and hyperglucagonemia than in response to the hyperglucagonemia alone. In order to conclude definatively that the larger fall was attributable to the released insulin it is necessary that the plasma glucagon levels in the two groups be equilivant. Unfortunately, direct comparison of the plasma glucagon levels in the two groups is not possible, since plasma glucagon levels were not measured in the normal subjects. There is no a priori reason to suspect, however, that the clearance of glucagon is decreased in diabetics as would have to be the case if a difference in the glucagon level were to explain the findings. In addition, with the plasma glucagon concentrations so high it is unlikely that a difference in the glucagon levels in the two groups would be of significance since the concentrations in both should still be well above the saturating level for the response. Although the alanine level is lower basally in diabetics, as reviewed elsewhere,25 it fell by only 20% in response to glucagon as opposed to 55% in the normals who secreted insulin. It is generally believed that glucagon lowers the concentration of alanine and other

amino acids by increasing their uptake by the liver as noted in studies in perfused rat 1iver,26 in the conscious dog” and in man.28 LeCam and Freychet have reported that glucagon stimulates the A transport system of the hepatocyte, the system by which alanine enters the liver ce11.29Interestingly, glutamine does not use the A transport system3’ but an effect of glucagon on glutamine transport has been reported in sheep.3’ In the present study glutamine fell dramatically in response to glucagon in both groups, suggesting that in intact man glucagon may also stimulate hepatic glutamine extraction. Glucagon infusion in the present study resulted in the induction of hyperglycemia in the normal subjects and in a further increment in the existing hyperglycemia of the diabetic subjects, thus raising the possibility of an effect of the hyperglycemia per se on plasma amino acid levels. We have previously examined the effect of hyperglycemia induced in the presence of fixed basal levels of insulin and glucagon, however, and noted that a high blood sugar level per se had no effect on the plasma level of any amino acid other than alanine. The level of alanine increased significantly in response to hyperglycemia. The latter change makes the action of glucagon to lower the alanine level in the present study even more impressive. It would thus seem that the glucagon induced changes in the amino acids levels are attributable to a direct action of the hormone on the liver. In summary, pharmacologic amounts of glucagon administered in the absence of counterregulatory insulin secretion acutely lower the plasma levels of IO amino acids, including all of the major gluconogenic precursors while leaving the level of 8 amino acids, including the branched chain amino acids, unchanged. When counteregulatory insulin secretion occurred in response to glucagon administration there was a fall in the level of an additional 5 amino acids including the BCAA and the magnitude of the fall in the level of those which fell in response to glucagon above increased.

REFEI RENCES 1. Mallinson CN, Bloom SR, Warin AP, et al: A glucagonoma syndrome. Lancet 2: l-5, 1974 2. Boden Cl, Owen OE, Rezvani I, et al: An islet cell carcinoma containing glucagon and insulin. Chronic glucagon excess and glucose homeostasis. Diabetes 26: 128-l 37, 1977 3. Brockman RP, Bergman EN: ElTect of glucagon on plasma alanine and glutamine metabolism and hepatic gluconeogenesis in sheep. Am J Physiol228:1627-1633, 1975 4. Fitzpatrick GF, Meguid MM, Gitlitz PH. et al: Glucagon infusion in normal man: Effects on 3-methylhistidine excretion and plasma amino acids. Metabolism 26:477485, 1977 5. Aoki TT, Muller WA, Brennan MF, et al: Effect of glucagon

on amino acid and nitrogen metabolism in fasting man. Metabolism 23805-8 14, 1974 6. Marliss EC, Aoki TT, Unger RH, et al: Glucagon levels and metabolic effects on fasted man. J Clin Invest 49:2256-2270, 1970 7. Bromer WW, Chance RE: Zinc glucagon depression of blood amino acids in rabbits. Diabetes 18:748-754, 1969 8. Reisner SH, Aranda JV, Colle E, et al: The effect of intravenous glucagon on plasma amino acids in the newborn. Pediat Res 7:184-191, 1973 9. Guard JR, Guillet 1, Marty T, et al: Effects of exogenous hormones on plasma levels and hepatic metabolism of amino acids in the fetus and newborn rat. Diabetologia 12:327-337, 1976

1199

GLUCAGON AND PLASMA AMINO ACID LEVELS

10. Felig P, Brown WV, Levine RA, et al: Glucose homeostasis in viral hepatitis. N Engl J Med 283:1436-1440, 1970 11. Landau RL, Lugibihl K: Effect of glucagon on concentration of several free amino acids in plasma. Metabolism 4265-276, 1969 12. Zinneman HH, Nuttall FQ, Goetz FC: Effect of endogenous insulin on human amino acid metabolism. Diabetes 15:5-8,1966 13. Harris MM, Harris RS: Effect of insulin hypoglycemia and glucose on various amino acids in blood of mental patients. Proc Sot Exp Biol Med 64:471-476, 1947 14. Crofford OB, Felts PW, Lacy WW: Effect of glucose infusion on the individual plasma free amino acids in man. Proc Sot Exp Biol Med 117:11-17, 1964 15. Melani F, Rubenstein AH, Oyer PE, et al: Identification of proinsulin and C-peptide in human serum by specific immunoassay. Prpc Nat1 Acad Sci 67:148-155,197O 16. Aguilar-Parada E, Eisentraut AM, Unger RH: Pancreatic glucagon secretion in normal and diabetic subjects.. Amer J Med Sci 257:415-419,1969 17. Wide L, Porath J: Radioimmunoassay of proteins with the use of Sephadex-coupled antibodies. Biochim Biophys Acta 130:257-260 18. Kopecky M: A modification of the Technicon autoanalyzer permitting three amino acids to be analyzed simultaneously. In Automation in Analytical Chemistry, Technicon Symposia, 1965, p 700 19. Snedecor, GW: Statistical methods, Iowa State University Press, Ames. (ed 5). 1967, p 35 20. Block MB, Mako ME, Steiner DF, et al: Circulating C-peptide immunoreactivity. Studies in normals and diabetic patients. Diabetes 21:1013-1026.1972 21. Carlsten B, Hallgren R, Jagenburg R, et al: Amino acids and free fatty acids in plasma in diabetics. I. The effect of insulin on the arterial levels. Acta Med Stand 179:361-370,1966

22. Lotapeich WD: The role of insulin in the metabolism of aminoacids. J Biol Chem 179:175-180, 1949 23. Pozefsky T, Felig P, Tobin JD, et al: Amino acid balance across tissues of the forearm in postabsorptive man. Effects of insulin at two dose levels. J Clin Invest 48:2273-2282,1969 24. Chiasson JL, Liljenquist JE, Finger FE, et al: Differential sensitivity of glycogenolysis and gluconeogenesis to insulin infusions in dogs. Diabetes 25:283-291, 1976 25. Cherrington AD: Gluconeogeneais: Its regulation by insulin and glucagon. Handbook of Diabetes Mellitus (in press) 26. Mallette LE. Exton JH, Park CR: Effect of glucagon on amino acid transport and utilization in the perfused rat liver. J Biol Chem 244:5724-5728.1969 27. Shoemaker WC, Van Itallie TB: The hepatic response to glucagon in the unanesthetized dog. Endocrinology 66:26@-268, 1966 28. Kibler RF, Taylor WT, Myers JD: The effect of glucagon on net splanchinic balances of glucose, amino acid nitrogen, urea, ketones, and oxygen in man. J Clin Invest 43:904-915.1964 29. LeCam A, Freychet P: Glucagon stimulates the A system for neutral amino acid transport in isolated hepatocytes of adult rat. Biccbmii’Biophys Res Commun 72:893-901.1976 30. Joseph SN, Bradford NM, McCiiven JD: Characteristics of the transport of alanine, serine, and glutamine across the plasma membrane of isolated rat liver cells. B&hem J 176:827-836, 1978 31. Brockman RP, Bergman EN, Joo PK, et al: Effects of glucagon and insulin on net hepatic metabolism of glucose precursors in sheep. Am J Physio1229: 1344-l 350, 1975 32. Shulman GI, Lacy WW, Liljenquist JE, et al: Effect of glucose, independent of changes in insulin and glucagon secretion, on alanine metabolism in the conscious dog. J Clin Invest 65:496505.1980 l