Effect of Intravenous Glucose and Insulin on Plasma Tryptophan and Tyrosine Concentrations in Normal Subjects and Patients With Carcinoid Tumors Jerome
M.
Feldman
We evaluated the effect of intravenous (i.v.) glucose on the plasma tryptophan (TRP) and tyrosine (TYR) concentration of 12 normal subjects, six patients with carcinoid tumors and the carcinoid syndrome (carcinoid syndrome), and five patients with carcinoid tumors without the carcinoid syndrome (tumor). Following i.v. glucose administration, the plasma Trp concentmtion of the normal subiects and the tumor patients increased, while the plasma Trp concentration of the carcinoid syndrome patients decreased. Following i.v. glucose adminirtmtion, the plasma Tyr concentmtion of the normal subiects and the tumor patients decreased, while the plasma Tyr concentmtion of the carcinoid syndrome patients did not
and
James
W.
Plonk
change. The response to i.v. insulin differed in some respects from the response to i.v. glucose: the plosma Trp of normal subjects did not change while the plasma Trp of carcinoid syndrome patients decreased; the plasma Tyr of the normal subjects increased while the plasma Tyr concentmtion of the carcinoid syndrome patients did not change. The carcinoid syndrome patients had high serum serotonin concentmtions and impaired glucose tolerance and insulin secretion as compared to both normal subjects and tumor patients. We conclude that under appropriate experimental conditions, glucose administmtion can increase the plasma Trp concentration of normal human subjects.
I
N RECENT YEARS Fernstrom and Wurtman have demonstrated the unique relationship between the plasma concentration of tryptophan (Trp) and tissue utilization of glucose.’ When rats ingest a carbohydrate meal or receive a subconvulsive dose of insulin2 their plasma Trp increases while the concentration of the other amino acids in their plasma decreases.3 The rise in plasma Trp results in an increased concentration of Trp and 5hydroxytryptamine (serotonin) in brain tissue. ‘s4 Fernstrom and Wurtman have suggested that this mechanism of elevating brain serotonin may be part of a closed circle in which dietary Trp modifies appetite and food consumption.5 In contrast to the rat, when man ingests oral glucose6 or receives an injection of insulin’ plasma Trp levels show a modest decrease. During the ingestion of glucose, a glucagon like material is released from the gastrointestinal tract,* and glucagon injections decrease the plasma concentration of Trp.3 We therefore thought it would be of interest to determine the effect of intravenous (i.v.) glucose on the plasma Trp concentrations of normal volunteers. We also evaluFrom the Durham Veterans Administration Hospital and the Division of Endocrinology, Department of Medicine. Duke University Medical Center. Durham, N.C. Received for publication May 9. 1975. Supported by the Veterans Administration (2605-l). Grant AM-05620 from the National Institute of Arthritis Metabolic and Digestive Diseases and Grant MOI-RR-30 from the Clinical Research Center Branch Division of Research Facilities and Resources, U.S. Public Health Service. Reprint requests should be addressed to Jerome M. Feldman, M. D., Box 2963, Duke University Medical Center, Durham, N.C. 27710. 0 1976 by Grune & Stratton, Inc.
Metabolism,
Vol. 25, No. 1 (January),
1976
97
FELDMAN
98
AND
PLONK
ated the effect of i.v. glucose on the plasma Trp of patients with metastatic carcinoid tumors, some of whom had the carcinoid syndrome. The increased synthesis of serotonin from Trp is reported to decrease the plasma levels of this amino acid.g We compared the effect of i.v. glucose on Trp with its effect on another aromatic amino acid, tyrosine (Tyr). MATERIALS
AND
METHODS
Experimentalsubjects. All of the experimental subjects were within 15% of their ideal body weight” and prior to our studies none of the subjects were known to have diabetes mellitus. The 12 normal volunteers (ten men and two women) who underwent intravenous glucose tolerance tests (IVGTT) had a mean age of 54 yr (range 32-67 yr). The six normal volunteers (three men and three women) who underwent insulin tolerance tests (ITT) had a mean age of 35 yr (range 21-60 yr). The group of patients with metastatic carcinoid tumors and the carcinoid syndrome consisted of six men, with a mean age of 62 yr (range 54-69 yr). Their serum serotonin concentration of 1.85 f 0.44 pg/ml (M + SE) was above the normal range of 0.05-0.30 ag/ml. Their 24hr urinary 5-hydroxyindoleacetic acid excretion (5-HIAA) of 111 + 65.3 mg/day was above the normal range of 2-9 mg/day. The five patients (three men and two women) with metastatic carcinoid tumors without the carcinoid syndrome had a mean age of 49 yr (range 33-60 yr). Their serum serotonin concentration of 0.11 + 0.04 pg/ml and their 24-hr 5-HIAA excretion of 3 & 0.4 mg/day was within the normal range. The diagnosis of carcinoid tumor had been previously established by histologic examination of the tumor. The two groups of patients with carcinoid tumors had comparable metastatic involvement as evaluated with liver function tests and -Tc liver scans. Test procedure. After obtaining their informed consent for the studies, the subjects were admitted to the Clinical Research Unit at Duke University Medical Center. Hospital diets were matched as closely as possible to home diets and a normal level of physical activity was encouraged. After an overnight fast, the subjects remained at bed rest on the morning of the test. The IVGTT and ITT tests were done with previously described procedures.“*‘* The ITT was performed with 0.1 unit of pork insulin per Kg of body weight. Analytical methods. Plasma growth hormone (GH) and insulin were determined by a modification of a double antibody radioimmunoassay method usink tz51 labeled pork insulin and human GH as tracer and human insulin and GH as standards. 3 Plasma cortisol was measured by a competitive protein binding method.14 Plasma glucose was measured by a glucose oxidase method.” Plasma Trp16 plasma Tyr17 and serum serotonin’* were measured by fluorometric methods. 5-HIAA excretion was measured on 24-hr urine samples by a colorometric method.” The glucose disposal rate constant after i.v. glucose (KG) or i.v. insulin (Kt) administration and the plasma insulin half-life were calculated by previously described methods.20 These rate constants represent the fall of the plasma glucose in percent per minute. Insulin secretion in response to i.v. glucose was expressed as the area under the plasma insulin curve above the fasting level from 0 to 60 min (AI). The terms plasma Trp and plasma Tyr refer to the mean plasma Trp concentration and the mean plasma Tyr concentration of the three experimental groups. Plasma Trp and Tyr concentrations after i.v. glucose were expressed as the percent change from 0 time concentration. The changes in Trp and Tyr after i.v. glucose were also expressed as the net change in area above or below the fasting Trp (ATrp) or Tyr (ATyr) level from 0 to 60 min. The data was analyzed with the Student’s t test, and the paired t test. Correlation analysis were done with the Pearson product correlation coefficient (r). Differences were considered to be significant for p < 0.05.*’ All data is expressed as the mean +SEM.
RESULTS
IVGTT Figure 1 depicts the effect of i.v. glucose on the plasma Trp and Tyr of normal volunteers (normal), patients with metastatic carcinoid tumors with the
INTRAVENOUS
GLUCOSE
AND
99
INSULIN
%;'No'o . l
NORMAL ' TUNOR
-
-
k---------
Y
0
TIME
0 (Minutes)
60
Fig. 1. Effect of i.v. glucose on the plasma concentration of Trp and Tyr in normal subiects, patients with the carcinoid syndrome, and patients with tumor. Plasma Trp and Tyr concentmtions after i.v. glucose are expressed as the per cent change of the amino acid concentration of each subject from his O-time concentration. The absolute O-time plasma amino acid concentrations in pg/ml are (M f SE): Trp-normal, 8.3 f 0.59; carcinoid syndrome, 9.6 f 1.83; tumor, 6.4 f 0.68; Tyr-normal, 14.6 f 0.99; carcinoid syndrome, 11.4 f 0.99; tumor, 15.2 f 1.38. The points represent the mean and the brackets the SEM of 12 (normal), six (catxinoid syndrome), and five (tumor) subjects. (“) indicates a significant per cent change from the O-time plasma amino acid concentration of the group.
carcinoid syndrome (carcinoid syndrome) and patients with metastatic carcinoid tumors without the carcinoid syndrome (tumor). The plasma amino acid concentration of each patient is expressed as a percent change from the patients plasma amino acid concentration prior to glucose injection. In the normal subjects the plasma Trp is significantly elevated above 0 time values 15 min after glucose injection (+25%) and the mean plasma Trp of this group remains elevated for the 90 min of observation. In contrast the plasma Trp of the carcinoid syndrome subjects is significantly decreased below the 0 time value 30 and 60 min after glucose injection (- 20%). The plasma Trp of the tumor subjects responds more like the normal subjects than the carcinoid syndrome subjects. Plasma Trp of the tumor subjects is significantly elevated above their 0 time values at 60 and 90 min (+ 58% and 87%). In the normal subjects the plasma Tyr promptly declines after glucose injection. It is significantly below 0 time values at 60 and 90 min. In contrast the plasma Tyr of the carcinoid syndrome subjects does not significantly change from their 0 time value. The plasma Tyr of the tumor subjects decreases significantly below the 0 time value by 90 min. The plasma Tyr of the tumor subjects behaves more like the plasma Tyr levels of the normal subjects than it does the plasma Tyr of the carcinoid syndrome subjects. The Ko of the carcinoid syndrome subjects (0.85 + 0.07) is significantly less
100
FELDMAN
1 TRYPTOPHAN
I
TIME
I
AND
PLONK
TYROSINE
(Minutes)
Fig. 2. Effect of i.v. insulin on the plasma concentration of Trp and Tyr of normal subjects and patients with the carcinoid syndrome. Plasma Trp and Tyr concentrations after i.v. insulin are expressed as the per cent change of amino acid concentmtion of each subject from his O-time concentration. The abolute O-time plasma amino acid concentrations in Fg/ml am Trp-normal, 11 .O f 0.59; carcinoid syndrome, 9.6 f 1.38; Tyr-normal, 10.8 f 1.24; carcinoid syndrome, 12.2 f 0.89. The points represent the mean and the bmckets the SEM of six subjects in each group. (“) indicates a sign&ant per cent change from the O-time plasma amino acid concentration of the group.
than the K, of the normal subjects (1.83 + 0.20) or the Ko of the tumor subjects (1.42 f 0.13). The insulin secretion (AI) of the normal subjects (2540 f 378 PU-min/ml) is greater than the AI of the carcinoid subjects (1364 f 247 PU-min/ml) and the AI of the tumor subjects (1152 f 471 PI-l-min/ml). However, within the group of 12 normal subjects there is no correlation between the changes in plasma Trp or plasma Tyr and the K, or the AI of individual subjects. Within the group of normal subjects there is no relationship between the subjects age and their Ko, AI, ATrp, or ATyr. ITT Figure 2 depicts the effect of i.v. insulin on the plasma Trp and Tyr of normal subjects and carcinoid syndrome subjects. Although, insulin did not alter the plasma Trp of normal subjects it significantly reduced the plasma Trp of carcinoid syndrome subjects. In contrast insulin significantly increased the plasma Tyr of normal subjects, although it did not alter the plasma Tyr of carcinoid subjects. In response to insulin, the maximal glucose fall of the normal subjects (67 f 3.7%) was greater than the maximal fall of the carcinoid syndrome subjects (49 f 2.9%). The K, of the normal subjects (6.0 + 0.07). The normal subjects and carcinoid syndrome subjects did not differ with respect to insulin halflife, peak plasma insulin concentration, Acortisol response to hypoglycemia, or AGH in response to hypoglycemia.
INTRAVENOUS
GLUCOSE
AND
101
INSULIN
To determine if increased plasma cortisol plays a role in elevating the plasma Tyr in normal subjects, three fasting normal volunteers received an i.v. injection of 200 mg of hydrocortisone sodium succinate at 0800 hr. There was no change in their plasma Trp or plasma Tyr at 0900 hr. Four hours after the injection (1200 hr) their plasma Trp and Tyr had declined by 25%. This was prbably due to the diurnal decrease in the plasma concentration of amino acids.22 DISCUSSION
The present report indicates that i.v. insulin does not mimic the effect of i.v. glucose on the concentration of plasma Trp and Tyr. Although, glucose utilization is increased after iv. glucose or i.v. insulin administration, there are both similarities and differences in the metabolic effects produced by these substances. The administration of both i.v. insulin and i.v. glucose causes a prompt decline in plasma free fatty acids (FFA), with the FFA nadir occurring at approximately 30 min. 23*24The hypoglycemia induced by i.v. insulin results in a brisk increase in plasma levels of growth hormone, cortisol, and epinephrine with the plasma levels of these hormones peaking 20-40 min after i.v. insulin.2s-27 In contrast i.v. glucose produces a trivial increase in GH at approximately 60 min. Intravenous insulin results in a brisk rise in plasma glucagon (peak concentration 30 min after insulin injection) while i.v. glucose results in a decrease in plasma glucagon. **v*~Glucagon and cortisol decrease the plasma concentration of Tyr and Trp. 3*30*3’Although we did not evaluate the effect of glucagon on Trp and Tyr in the present study, acute administration of a large dose of cortisol did not alter the plasma concentration of these amino acids. Plasma Trp is maximally elevated 15 min after i.v. glucose administration while plasma concentration of GH, cortisol, glucagon, and epinephrine are not maximally elevated until 30 min after i.v. insulin injection.25-28 This suggests that additional factors play a role in the differential response of amino acids to i.v. glucose and i.v. insulin administration. The failure of exogenous insulin to mimic the effects of exogenous glucose has been decribed in other experimental models. Thus glucose, but not insulin administration, can restore the decreased hepatic glucokinase concentration that developes when rats receive a carbohydrate free diet.)* Lipsett et al. found that 120 min after ingesting oral glucose, normal subjects had a modest reduction in their plasma Trp.6 They also demonstrated that there is a relationship between the decline in plasma Trp and the decline in FFA.6 The maximum fall in FFA after oral glucose occurs at 30 min.6*23 The maximal increase in plasma insulin after oral glucose occurs at 30-60 min while the maximal increase after i.v. glucose occurs at 5 min. These metabolic differences, in conjuction with the release of glucagon like material after the ingestion of glucose,7 may account for the differential effect of oral and i.v. glucose on plasma Trp. Fernstrom et al. found that 2 hr after receiving i.v. insulin, normal subjects had a 39% reduction in plasma Trp.’ In our studies we measured the plasma glucose for only 1 hr after i.v. insulin administration. During this hour we noted no significant changes in plasma Trp. We also evaluated the effect of i.v. glucose on patients with metastatic carcinoid tumors. Patients with the carcinoid syndrome are reported to have
102
FELDMAN
AND
PLONK
low plasma Trp concentrations because of the increased conversion of Trp to serotonin.9 The plasma amino acid concentrations of these patients had a different response to i.v. glucose than the plasma amino acid concentrations of the normal subjects. In contrast, the plasma amino acids of patients with metastatic carcinoid tumors, who did not have the carcinoid syndrome, had a similar response to i.v. glucose as the normal subjects. This suggests that the metastatic tumor itself, is not the major factor in the altered response of patients with carcinoid syndrome to i.v. glucose. Although, the altered response may be due to elevated serum serotonin concentration, it may also be related to the fact that the carcinoid syndrome patients (but not the tumor patients) have chemical diabetes mellitus with impaired glucose tolerance and insulin secretion.33 We have not had the opportunity to evaluate the plasma Trp and Tyr response to i.v. glucose in patients with diabetes mellitus, who do not have the carcinoid syndrome. ACKNOWLEDGMENT We thank Miss Connie
Mrs. Joan Roche and Miss Barbara Chapman for Heiner for assistance in preparation of the manuscript.
expert
technical
assistance
and
REFERENCES 1. Fernstrom JD, Wurtman, RJ: Control of brain 5-HT content by dietary carbohydrates in Barchas J, Usidin E (ed): Serotonin and Behavior, London, Academic Press, 1973, p 121. 2. Fernstrom JD, Wurtman RJ: Brain serotonin content: increase following ingestion of carbohydrate diet. Science 174: 1023, 197 I 3. Fernstrom JD, Wurtman RJ: Elevation of plasma tryptophan by insulin in rat. Metabolism 21:337, 1972 4. Madras BK, Cohen EL, Messing R, Munro HN, Wurtman RJ: Relevance of free tryptophan in serum to tissue tryptophan concentration. Metabolism 23: 1107, 1974 5. Fernstrom JD, Wurtman RJ: Nutrition and the brain. Sci Am 230:84, 1974 6. Lipsett D, Madras BK, Wurtman RJ, Munro HN: Serum tryptophan level after carbohydrate ingestion: selective decline in nonalbumin-bound tryptophan coincident with reduction in serum free fatty acids, Life Sci I2 (Part II): 57, 1973 7. Fernstrom JD, Larin F, Schonfeld G, Wurtman RJ: Insulin-induced changes in plasma tryptophan in rats and man. Fed Proc 30:250, 197 I 8. Lawrence AM: Glucagon. Annual Review of Medicine 20:207, 1969 9. Levine RJ, Sjoerdsma AS: Pressor amines and the carcinoid flush. Ann Intern Med 58:818, 1963 10. Metropolitan Life Insurance Company
Statistical Bulletin. New weight standards for men and women. 40:3, 1959 1 I. Quickel KE, Jr, Feldman JM, Lebovitz HE: Enhancement of insulin secretion in adult onset diabetes mellitus by methysergide maleate: Evidence for an endogenous biogenic monoamine mechanism as a factor in the impaired insulin secretion in diabetes mellitus. J Clin Endocrinol Metab 33:877, 1971 12. Bivens CH, Levobitz HE, Feldman JM: Inhibition of hypoglycemia induced growth hormone secretion by the serotonin antagonists cyproheptadine and methysergide. N Engl J Med 289:236, 1973 13. Frohman L. Horton ES and Lebovitz HE: Growth hormone releasing action of a pseudomonas endotoxin (Piromen). Metabolism 16:57, 1967 14. Murphy BEP: Some studies on the protein-binding of steroids and their application to the routine micro and ultra micro measurement of various steroids in body fluids by competitive protein-binding radioassay. J Clin Endocrinol Metab 27:973, 1967 15. Saifer A, Gerstenfeld B: Photometric microdetermination of blood glucose with glucose oxidase. J Lab Clin Med 5:448, 1958 16. Williams CH. King DJ, and Cairns J: Microestimation of tryptophan in plasma by a fluorometric procedure. Biochem Med 6:504, 1972 17. Williams T, Besser GM: Effects of
INTRAVENOUS
GLUCOSE
AND
INSULIN
treatment on tyrosine tolerance in thyroid disease. Clin Chem 17:148, 1971 18. Davis RB: The concentration of serotonin in normal human serum as determined by an improved method. J Lab Clin Med 54334, 1959 19. Udenfriend S, Titus E and Weissbach H: The identification of 5-hydroxy-3-indoleacetic acid in normal urine and a method for its assay. J Biol Chem 216449, 1955 20. Bivens CH, Feldman JM: Metabolic and endocrine studies in identical diabetic twins. Arch Intern Med 134:639, 1974 21. Snedecor GW, Cochran WG: Statistical Methods (6th ed) Ames Iowa, Iowa State University Press, 1967 22. Wurtman RJ, Rose CM, Chou C and Larin FF: Daily rhythms in the concentration in various amino acids in human plasma. N Engl J Med 279:17l, 1968 23. Abramson EA. Arky RA: Role of beta-adrenergic receptors in cournter-regulation to insulin hypoglycemia, Diabetes 17:141, 1968 24. Schalch DS, Kipnis DM: Abnormalities in carbohydrate tolerance associated with elevated plasma nonesterified fatty acids. J Clin Invest 44:2010, 1965 25. Bivens CM, Lebovitz HE, Feldman JM: Inhibition of hypoglycemia-induced growth hormone secretion by the serotonin antagonists cyproheptadine and methysergide. N Engl J Med 289:236, 1973 26. Plonk JW, Bivens CH, Feldman JM: Inhibition of hypoglycemia-induced cortisol
103
secretion heptadine. 1974
by the serotonin J Clin Endocrinol
antagonist Metab
cypro38:836,
27. Goldfien A, Zileli S, Goodman D, Thorn GW: The estimation of epinephrine and norepinephrine in human plasma. J Clin Endocrinol Metab 21:281, 1960 28. Walter RM, Dud] RJ, Palmer JP, Ensinck JW: The effect of adrenergic blockade on the glucagon responses to starvation and hypoglycemia in man. J Clin Invest 54:1214, 1974 29. Gerich JE, Langlois M, Schnieder V, Karam JH, Noacco C: Effects of alterations of plasma free fatty acid levels on pancreatic glucagon secretion in man. J Clin Invest 53: 1284, 1974 30. Landau RL, Lugibihl K: Effect of glucagon on concentration of several free amino acids in plasma. Metabolism 28:265, 1969 31. Rivlin RS, Melmon KL: Cortisoneprovoked depression of plasma tyrosine concentration: relation to enzyme induction in man. J Clin Invest 44: 1690, 1965 32. Niemeryer H, Perez N, Rabajille E: Interrelation of actions of glucose insulin and glucagon on induction of adenosine triphosphate D-hexose phosphotransferase in rat liver. J Biol Chem 241:4055, 1966 33. Feldman JM, Marecek RL, Quickel KE, Lebovitz HE: Glucose metabolism in the carcinoid syndrome. J Clin Endocrinol Metab 35:307, 1972