- Email: [email protected]
Overnutrition induced decrease in insulin action for glucose storage: In vivo and in vitro in man
Overnutrition
Induced
Decrease in Insulin Action for Glucose In Vivo and in Vitro in Man D.M. Mott,
Storage:
S. Lillioja, and C. Bogardus
The effect of short-term overnutrition on insulin action for glucose disposal was assessed in 16 Southwest American Indians (mean wt = 74 + 6 kg). After two weeks of weight maintenance and again after two weeks of 62% greater caloric intake (constant ratio of fat:carbohydrate:protein), insulin action for glucose disposal was measured using the euglycemic clamp technique with plasma insulin concentrations of about 110 and 1800 uU/mL. Simultaneous indirect calorimetry was used to estimate carbohydrate oxidation and storage rates. Following overnutrition, mean weight gain was 3.0 f 0.2 kg, P < 0.01. Overnutrition induced a decrease in glucose storage at the low and high insulin concentrations: 1.2 of:0.3 to 0.2 f 0.3, P < 0.01, and 6.4 f 0.3 to 4.3 + 0.5, mgl kg FFM min, P < 0.001. Carbohydrate oxidation was significantly increased at both insulin concentrations. The mean total insulin mediated glucose disposal rate decreased from 11.6 * 0.5 to 10.3 f 0.7, P < 0.01, at the high insulin concentration. This decrease was due entirely to the reduction in carbohydrate storage and was correlated with increased fasting insulin concentration (r = 0.7, P < 0.01). Overnutrition also induced a significant decrease in the percent muscle glycogen synthase active measured fasting and at the end of the high-dose insulin infusion. The results indicate that short-term overnutrition results in reduced insulin action for glucose storage and disposal which is correlated with increased fasting insulin concentrations. Reduced glycogen synthase activity may contribute to the effect of overnutrition on in vivo insulin-mediated glucose storage. m 1988 by Grune & Stratton, Inc.
0
BESE subjects have reduced rates of insulin-mediated glucose disposal’,* compared to lean subjects. It is not clear, however, whether this is the result of increased body mass, increased caloric intake, or some other factor of the obese condition. Olefsky et al’ reported that short-term overnutrition, with no change in diet composition, was associated with hyperinsulinemic responses to oral glucose and a meal of mixed composition. This increased insulin response, however, was not accompanied by a significant decrease in insulin action as measured by the steady state plasma glucose technique. The apparently inconsistent results from different measures of insulin action for glucose disposal, suggested that further studies were needed to define the possible importance that overnutrition per se, independent of large changes in weight, could have in reducing insulin action on glucose disposal. In this study, we have determined the effect of short-term overnutrition on in vivo insulin action in 15 Pima Indian subjects. Insulin action for glucose disposal was measured using the euglycemic clamp technique with simultaneous indirect calorimetry to estimate carbohydrate oxidation and storage rates. Insulin action was assessed in vitro by measuring muscle glycogen content and glycogen synthase activity in biopsies collected before and after overnutrition. In this study, short-term overnutrition induced a reduction in insulin-mediated glucose storage rates, which was associated with increased fasting plasma insulin concentrations and a reduction in muscle glycogen synthase activities.
From the Phoenix Clinical Research Section, National Institute of Arthritis. Diabetes, Digestive and Kidney Diseases, National Institutes of Health, 4212 North 16th St, Phoenix, AZ 85016. Address reprint requests to Dr Mott, Phoenix Clinical Research Section, NIADDK, 4212 North 16th St, Rm 541, Phoenix, AZ 85016. o 1986 by Grune & Stratton. Inc. 0026-0495~86/3S02-0013$03.00/0
160
MATERIALS AND METHODS Subjects
Fifteen Southwest American Indian males were admitted to the Phoenix clinical research ward. Their characteristics are shown in Table 1. After written informed consent was obtained, all subjects were physically examined and a 12-lead electrocardiogram was recorded. After an overnight fast, blood was drawn for complete blood count, liver function tests, BUN, creatinine, electrolytes, calcium, total protein, and albumin. After at least two days of a weight-maintaining diet containing at least 200 g of carbohydrate, an oral glucose tolerance test was performed.’ Results are reported only on subjects with a normal examination and 2-hour postglucose load plasma glucose concentrations of less than 160 mg/lOO mL. The percent body fat of each volunteer was estimated by underwater weighing with correction for the simultaneously measured residual lung volume.s Protocol
Subjects were placed on a weight-maintaining diet containing 45% carbohydrate, 40% fat, and 15% protein for 14 days. On days 9 through 14, a 24-hour urine C-peptide, oral glucose tolerance test, meal test, and euglycemic clamp with indirect calorimetry and muscle biopsies were performed in that order. Over the next 3 days the diet was stepwise increased to 62% above weight-maintenance requirements using a liquid supplement (Ensure Plus, Ross Laboratories, Columbus, Ohio) while keeping the same relative proportion of protein, carbohydrate, and fat. This diet was maintained for the next 11 days including the period of repeat testing. We have previously observed that a weight-maintenance diet during 10 days in the clinical research unit produced no effect on the parameters reported here from the euglycemic clamp, indirect calorimetry, or muscle biopsy (unpublished observations). Tolerance
Tests
The oral glucose tolerance test was performed after ingestion of 75 g of glucose.’ The meal consisted of orange juice, beef patty, eggs, toast, butter, and jelly, and contained 47% fat, 29% carbohydrate, and 24% protein (690 calories). It was consumed over a IO-minute
Metabolism, Vol35, No 2 (February), 1986: pp 160- 165
OVERNUTRITION
161
AND INSULIN ACTION
Table 1. Clinical Data Before and After Overnutrition Fasting
2-Hour Plasma
Plasma Glucose BMI Subject
Mean
25
(%)
WM
WM
2 3
GlUCose
hg/dl)
(mg/a)
Bodv Weight Gain Cki
WM
ON
WM
ON
3.0
88
85
22
1.5
98
89
110
135
30
3.2
89
75
159
112
28
25
3.3
87
93
121
116
18
26
18
3.8
83
86
96
145
6
23
21
16
2.5
80
74
83
77
7
23
22
15
3.3
81
86
107
126
8
30
27
27
4.1
80
86
146
129
9
24
23
22
3.0
77
77
117
138
10
20
23
24
2.0
88
92
129
131
11
21
25
24
2.0
79
85
88
103
12
27
23
20
3.9
87
95
104
99
13
18
21
27
3.4
82
90
90
106
14
28
28
29
2.4
95
86
148
112
15
22
21
21
2.8
83
87
118
105
23 k 1
3.0 k 0.2
85
27
29
21
29
22
30
4
26
5
i
23 + 1
SE
Abbreviations: + , percent extreme
lkalm’) G
1
Fat +
WM, fat
case where
was
weight not
25 k 1 maintenance;
determined
increased
fat mass
after was
ON,
f
2
86
98
*
2
114
128
*
6
117
+ 5
overnutrition.
overnutrition the only source
but
using
of weight
the
subject’s
gain,
period of time. During both tolerance tests blood samples were drawn for determination of plasma glucose and insulin concentrations at 0, 30.60, 120, and 180 minutes.
Euglycemic Clamp At 5 AM hours after a IO-hour overnight fast, an intravenous catheter was placed in an antecubital vein for infusion of insulin, glucose, and 3-H’ glucose. A primed (30 uCi)-continuous (0.30 uCi/min) infusion of 3-H’ glucose was then begun and continued until the end of the study. Another catheter was placed retrograde in a dorsal vein of the contralateral hand for blood withdrawal. The hand was kept in a warming box at 70 “C. After 2.5 hours four blood samples were drawn over 30 minutes for 3-H’ glucose-specific activity determination. A primed continuous insulin infusion of 40 mu/m* was then started and continued for 100 minutes (low dose). Another primed insulin infusion of 400 mu/m* was then started and continued for an additional 100 minutes (high dose). After the start of the initial insulin infusion, a variable infusion of 20% glucose was given, as necessary, to maintain the plasma glucose concentration at approximately 90 mg/ 100 mL. Blood for plasma glucose concentration was drawn every 5 minutes throughout the test. Blood for plasma insulin and 3-H3 glucose-specific activity was drawn every 20 minutes from 60 to 100 (low dose) and from 160 to 200 minutes (high dose). Steady-state glucose and insulin levels were determined during these two time intervals (Table 2). The high plasma insulin concentration achieved with 400 mu/m’ infusion was considered to be a maximally insulin stimulating concentration.2.6.7
mean
weight
% fat would
maintenance
have increased
23%
of 75
(WM)
+ 2kg
to 26%
and
assuming
the
(ON).
zirconium cell analyzer (Applied Electrochemistry, Sunnyvale, Calif) and the carbon dioxide analyzer was an infrared analyzer (Applied Electrochemistry, Sunnyvale, Calif). The analyzers and flowmeter outputs were connected to a desktop computer (HewlettPackard, Palo Alto, Calif). This recorded continuous, integrated calorimetric measurements every 5 minutes for the hour before and for the duration of the euglycemic clamp. The protein oxidation during the test was estimated from the urinary urea production rate. The nonprotein respiratory quotient was then calculated and the substrate oxidation rates determined from the equations of Lusk.’
Biopsy and Determination of Glycogen Synthase and Glycogen Content
Muscle
Three hours after the start of the 3-H’ glucose infusion and just prior to starting the insulin infusion, a percutaneous biopsy of the vastus lateralis muscle was performed. The specimen was immediately frozen in liquid nitrogen, stored at - 70 “C and later assayed for glycogen content and glycogen synthase activity. The muscle biopsy was repeated at the end of insulin infusion. The method of Nuttal et aI9 was modified for homogenization of muscle biopsies. Aliquots of the homogenate were stored at -20 “C for the determination of glycogen content. The remaining homogenate was centrifuged at 8000 x G for ten minutes and aliquots of the supernatant
Table 2. Mean Steady-State
Glucose and insulin Concentration
During Euglycemic Clamp Weight
Indirect Calorimetry For one hour before and for the duration of the euglycemic clamp procedure, a clear, plastic, ventilated hood was placed over the subject’s head. Room air was drawn through the hood and the flow rate measured by a pneumotachograph (Gould, Netherlands). A constant fraction of expired air was withdrawn and analyzed for oxygen and carbon dioxide content. The oxygen analyzer was a
weight
from
Maintenance
Low Dow, Glucose
(%)
Insulin C.V.
Low Dose
High Dose
(mg/
dL) C.V.
Overnutrition
High Dose
(ulJ/mLI
(%I
P < 0.05
89.6
+ 0.4
90.6
f
0.3
0.5
89.4
2.5
-c 0.2
3.6
r
0.3
2.3
+ 0.2
3.0
108
I
1810
i
100
117
+ 5’
5.5
+ 0.7
5.9
* 0.9
compared
4
to weight
5.9
+ 0.7
maintenance
90.3
f
by paired
1820 6.3 f-test.
+ 0.4 + 0.2 *
110
+ 0.6
162
MOlT ET AL
were stored at -20 OC for protein assay or used immediately for determination of glycogen synthase activity. The active form of glycogen synthase was assayed at low glucose-6-phosphate (G-6-P) concentration (0.17 mmol/L) in the presence of 0.14 mmol/L UDP glucose.” Total activity was measured at high G-6-P (7.2 mmol/L), also in the presence of 0.14 mmol/L UDP glucose. Activity was measured using UDP “C-glucose and is expressed as nanomoles of glucose incorporated into glycogen per minute per milligram of protein. The percent glycogen synthase active is calculated from activity at low G-6-P concentration divided by total activity. Frozen aliquots for glycogen determination were thawed and assayed according to the method of Hultman et al.”
Oral GlucoseTobrmce Test
(?sqGlreerd
s-
Meal
1690 C&ml- 475cFat. [email protected]. Rolsnl RdOn%Am .-m-w.
‘601
Calculations The appearance rate (Ra) of glucose in the plasma was calculated from the blood 3-H’ glucose-specific activities using the equations of SteeIe,12 in their steady-state form for the basal period and the nonsteady form for the Ra during the euglycemic clamp. During the basal period the Ra equals the endogenous glucose production rate. During the euglycemic clamp the endogenous glucose production rate equals the difference between the rate of exogenously infused glucose and the Ra determined from Steele’s equations. When the Ra equals the exogenous glucose infusion rate, the endogenous glucose production rate is assumed to be completely suppressed so that the total glucose disposal rate equals the exogenous glucose infusion rate. In these experiments these data were calculated for each 20 minutes of the last 40 minutes during the low- and high-dose euglycemic clamp. Data from each 40-minute period was averaged to calculate the total glucose disposal or “M” value. The basal and stimulated carbohydrate oxidation rates were calculated from the indirect calorimetry data by averaging the data for 40 minutes prior to the beginning of the insulin infusion and for the last 40 minutes during each of the two rates of insulin infusion. The glucose storage rate was estimated by subtracting the carbohydrate oxidation rate from the total glucose disposal rate. The glucose disposal values are expressed as milligram of carbyhydrate per kilogram of fat-free mass per minute. The fat-free mass is determined from underwater weighing.
Analytical Methods and Statistics Plasma insulin concentration were determined using the Herbert modification” of the radioimmunoassay of Yalow.” Urinary C-peptide concentrations were determined by radioimmunoassay as synthesized human C-peptide previously described. I5 Bacterially was used as standardI and was also used to prepare the tracer, I’Z5-t-Boc-tyr-C-peptide” and goat anti-human-C-peptide antiserum. Materials for the C-peptide assay were generously provided by Lilly laboratories (Indianapolis, Ind). Trititated glucose-specific activity in blood samples was determined after precipitating protein with perchloric acid as described by others.‘* Samples for fatty acid determinations were collected in tubes containing 1.l mg/mL diethyl p-nitrophenyl phosphate. (Sigma, St. Louis, Miss) and kept on ice. Concentrations of free fatty acids in plasma were measured using the microfluorometric method of Miles et a1.19 All data are expressed as the mean 5 the standard error of the mean. All statistical analyses were calculated using Statistical Analyses System, SAS Institute, Inc, Cary, N.C. RESULTS Overnutrition 3.0 k 0.2 kg,
for 13 days
produced
P < 0.01. There
a mean
weight
was no significant
gain
change
of in
glucose concentration, but the mean fasting plasma insulin concentration, measured on the mornmean
fasting
plasma
l-laws Fig 1. Effect of overnutrition on the response of glucose and insulin to 75 g oral glucose (left panels) and a 690 calorie standard meal (right panels). Results are the mean values for 15 subjects measured after weight maintenance (0) and overnutrition (0).
ing of the euglycemic clamp, increased significantly from 20 -+ 1 to 27 + 3, uU/mL, P < 0.005. The 24-hour urinary C-peptide also increased from 37 2 7 to 69 + 13 nmol, P < 0.01. Plasma-free fatty acids decreased from 7.6 + 0.9 to 5.4 f 0.7 mg/dL (NS). Following overnutrition, the increases in the insulin and glucose responses to oral glucose were not significant. However, there was a significant increase in the insulin response 240 + 30 v 302 f 33 uU/mL . 3 h f SE, P < 0.005 to the test meal with no significant change in the mean area under the glucose curve (Fig 1). Overnutrition was associated with increased basal and steady-state plasma insulin concentrations during the lowdose insulin infusion (Fig 2). In spite of this, glucose disposal was not significantly changed in the basal state or during the low-dose insulin infusion. Overnutrition significantly reduced_glucose storage and increased carbohydrate oxidation at all three insulin concentration. Overnutrition induced changes in these two parameters were significantly correlated in the basal state (r = -0.9, P < 0.001) at the end of the low-dose clamp (r = 0.7, P < O.Ol), but not at the end of the high-dose clamp (r = 0.5, P -G0.06). At the high insulin concentration, the decrease in glucose storage completely accounted for the significant reduction in the glucose disposal rate (see Fig 2). The muscle glycogen content and glycogen synthase activity was measured in muscle biopsies obtained from 9 subjects before (fasting) and after the insulin infusion for the euglycemic clamp. Similar to the results in Fig 2, this subgroup of
OVERNUTRITION
AND INSULIN
Fasting ~sposd
stomga~
oxidotimn
163
ACTION
LOWDOW
is.
.
Hiih Dose
.
IiS.
.
.
co.0 I
aooo5
am
aoool
Relationship between the overnutrition-induced Fig 3. decrease in glucose storage rates at the high dose of insulin and the overnutrition-induced increase in fasting insulin, r = 0.7; P -c 0.005.
106’4 WM
after (P < 0.05) the euglycemic clamp. In regard to the effect of overnutrition on glycogen concentration and glycogen synthase activity measured at the end of the clamp on these 9 subjects, the largest increase in muscle glycogen content (overnutrition-weight maintenance) was associated with the smallest decrease in the active form of glycogen synthase (r = 0.9, P < 0.005). The results in Fig 3 show that decreases in glucose storage during the high-dose clamp are correlated (r = 0.7, P .C 0.005) with the overnutrition-induced increases in fasting insulin concentration. The decreases in glucose disposal rates are also correlated with increased fasting insulin concentrations (r = 0.7, P < 0.005). Following overnutrition, those subjects with the largest overfeeding-associated increase in fasting insulin concentrations, had the largest decrease in the active form of the enzyme (r = 0.53, P = 0.14) and the smallest increase in glycogen (r = 0.7, P -c0.05), measured after the insulin infusion. No significant correlation was observed between the decrease in glucose storage rates and the change in muscle glycogen content or glycogen synthase activity.
117*5* ON
Fig 2. Effect of overnutrition on glucose disposal (height of bar) as measured by hepatic glucose production and the euglycemic clamp at basal, low. and high insulin levels at the end of the weight maintenance (WM) and overfeeding (ON) periods. Mean basal and steady state insulin concentrations are shown below the bars. Indirect calorimetry was used to measure glucose oxidation (clear area) and glucose storage (shaded area). All results are mean f SE. The SE values for glucose oxidation and storage were similar for each glucose disposal rate. and error bars are shown only for glucose disposal and storage. l = P < 0.05 and l * = P i 0.005 compared to weight-maintenance insulin concentration.
nine subjects had significant overnutrition induced reductions in carbohydrate storage before and during the low-dose and high-dose clamp (data not shown). On the weightmaintenance diet, muscle glycogen and the active form of glycogen synthase were not correlated before the euglycemic clamp but were positively correlated at the end of the clamp (r = 0.9. P < 0.001). The results in Table 3 show that overnutrition was associated with an increase in muscle glycogen content at the end of the euglycemic clamp (P < 0.05). Overnutrition also induced a significant decrease in the percent glycogen synthase active before (P < 0.05) and
DISCUSSION
This study was designed to determine if short-term overnutrition is associated with reduced insulin action for glucose disposal. The results indicate that overnutrition results in
Table 3. Effect of Overnutrition on Muscle Glycogen Content and Glycogen Synthase Activity Before and After Clamp Procedure* Wmght
Maintenance
Overnutrition
BeforeClamp
After Clamp
(g/ 100 g tissue) % Glycogen
1.5 + 0.2
1.6 + 0.1
2.5 + 0.7
Synthase Active Total Glycogen
28 + 4
59 t 4t
22 zk 3$
9.3 * 0.9
7.5 ? 0.9
BeforeClamp
After Clamp
Glycogen 2.0
r 0.2$
48 t 4tt
Synthase Activity (nmol/min. mg protein) *Comparisons made by paired t-test. tf
< 0.00 1 compared to before clamp.
$P i 0.05 compared to weight maintenance.
a.9 i
1.0
7.9
k 0.9
164
reduced in vivo insulin action for glucose disposal as estimated by the insulin and glucose response to a test meal and by the euglycemic clamp. Indirect calormetry during the euglycemic clamp indicated that overnutrition decreased glucose storage. In addition, overnutrition is associated with decreased skeletal muscle glycogen synthase activity. The mean results of the oral glucose tolerance test on 15 subjects were suggestive of an overnutrition-induced decrease in glucose tolerance but increases in both the area under the glucose and insulin curves failed to reach statistical significance. The significant increase in the insulin response to a meal of mixed composition could be interpreted as an overnutrition induced loss of insulin action. The results in Fig 2 indicate that short-term overnutrition reduces insulin action on glucose storage at fasting, low- and high-dose concentrations of insulin. The decreases in storage and disposal at the high insulin dose was correlated with the overnutrition-associated increase in fasting insulin concentrations. It is not clear, however, whether the increase in fasting insulin is the cause of or secondary to the overnutrition-associated decrease in glucose storage. In a previous report” using an identical high-dose clamp technique and diet composition, the mean glucose disposal differed from 13.1 2 0.5 mg/kg FFM ; min for 10 lean subjects (mean percent fat = 13 * 1) to 10.5 + 0.6 for 20 obese subjects (mean percent fat = 38 + 1). This represents a 20% decrease and 70% of this difference was due to reduced glucose storage. Our group of 15 subjects on weight maintenance diet had a glucose disposal rate of 11.6 + 0.5 mg/kg FFM - min, and a mean % fat of 23 ? 1, intermediate between the latter groups in both glucose disposal rate and % fat. As a result of overnutrition (mean Cal/d = 4,400) these subjects increased mean % fat by no more than 3% but decreased their glucose disposal rate to 10.3 + 0.7 mg/kg FFM - min (11% decrease), which is similar to the glucose disposal rate reported for the obese group (mean Cal/d = 3,100). All of this decrease was due to reduced glucose storage. These results suggest that overnutrition might be important for the mechanism of reduced rates of insulinmediated glucose disposal in obese subjects. However, decreased insulin action in obese subjects is characterized by both decreased glucose oxidation as well as storage.*’ In contrast, the overnutrition-induced decrease in glucose storage was significantly correlated with an increase in glucose oxidation. Although increased carbohydrate oxidation appears to compensate for part of the loss in glucose storage, the mechanisms coordinating these pathways are not known. The effect of decreased plasma free fatty acids to increase carbohydrate oxidation has been extensively reviewed.*’ Because plasma-free fatty acids in this report were not significantly changed by overnutrition, other factors may have stimulated carbohydrate oxidation. Results of an earlier study by Olefsky et al3 on the effects of overnutrition on insulin action for glucose disposal are similar to some of the results reported here. Both studies demonstrate no overnutrition-induced change in insulinmediated glucose disposal when insulin infusion raised plasma insulin concentrations about 100 uU/mL. The previous study, however, only observed evidence of reduced
MO-IT ET AL
insulin-mediated glucose disposal during a mixed meal test. Here insulin action for glucose disposal appears to be decreased by overnutrition when measured during the mixed meal test and when measured during high insulin dose infusion (+ 1800 uU/mL). The observed overnutritioninduced decrease in glucose disposal at the high but not the low insulin dose is probably caused by the following: (1) The decrease in glucose disposal is the result of an apparent decrease in the maximal velocity for glucose storage measured at the high insulin dose. This decrease in glucose storage is less prominent at the low insulin dose. (2) The overnutrition-induced increase in carbohydrate oxidation completely compensates for the reduced glucose storage at the low insulin dose, but not at the high insulin dose. It is, therefore, possible that the apparent inconsistent results of the low-dose clamp and the mixed meal test could be explained on the basis of a different response of each test to changes in maximal velocity of storage and/or to changes in the relative contribution of carbohydrate oxidation and storage. Further information on the regulation of glucose disposal during these two tests is needed to explain the apparent inconsistencies. The pathways of carbohydrate metabolism that account for the carbohydrate storage component measured by indirect calorimetry have not been fully identified. In a previous report, basal glycogen synthase activity in human muscle biopsies correlated with carbohydrate storage rates in glycogen-depleted subjects.22 We have reported that the stimulation of muscle glycogen synthase activity measured in biopsies obtained after insulin infusion, correlated with the range of glucose storage rates observed in glucose tolerant and intolerant subjects.23 These observations suggested that regulation of glycogen synthase activity may be a determinant of in vivo insulin-mediated glucose storage and disposal. In this study, overnutrition reduced the mean values for the percent active glycogen synthase measured before and after the clamp. This observation suggests that the reduction in the active form of the enzyme could contribute to the decreased rate of insulin-mediated glucose storage. It is not clear why overnutrition-associated increases in fasting insulin appear related to reduced glucose storage. The rise in insulin concentration may be secondary to the effects of overnutrition which include loss of glycogen synthase activity. Decreased glycogen synthase activity could be a result of increased muscle glycogen content which has been shown to inhibit glycogen synthase activity.24 This is an unlikely explanation, however, considering that subjects with the largest overnutrition-associated increase in muscle glycogen content had the smallest decrease in active glycogen synthase. A complete explanation for the overnutrition-induced decrease in carbohydrate storage and its relationship to carbohydrate oxidation will require further study to determine if most or only a small part of insulin-mediated glucose storage measured by indirect calorimetry is accounted for by glycogen synthesis. Another possible explanation for the effects of overnutrition on glucose storage would be a decrease in glucose transport leading to reduced G6P and an associated decrease in glycogen synthase activity.*’ This explanation seems
OVERNUTRITION
AND INSULIN ACTION
165
unlikely because reduced transport would probably also decrease carbohydrate oxidation. It is possible, however, that the principle sources of carbohydrate storage and oxidation occur in different tissues and would not necessarily be expected to change in the same direction. It is also possible that hyperinsulinemia induces a reduction in glucose disposal that is secondary to decreased insulin binding.26*27Based on studies in fat cells, none of these mechanisms seem likely. We have recently reported that overnutrition produced a significant increase in human basal and maximal insulin stimulated fat cell glucose transport with no significant change in insulin binding.** Results in the fat cell, however, may not reflect insulin binding and glucose transport in the tissues responsible for the overnutrition-induced decrease in glucose storage. In summary, reduced glucose storage is the primary component of reduced rates of glucose disposal during the high-dose clamp of both overfed normal glucose tolerant subjects (compared to weight maintenance) and in a previous
report of obese subjects (compared to lean). We have also previously reported that subjects on weight-maintenance diet with low glucose storage and disposal rates have reduced muscle glycogen synthase activity. Here we observe that normal glucose tolerant subjects with overnutrition-reduced glucose storage also have reduced glycogen synthase activity. These results suggest that overnutrition, independent of increased body mass, may contribute to the reduced insulinmediated glucose storage rates observed in obese subjects and that a reduction in glycogen synthase activity could be important to this mechanism.
ACKNOWLEDGMENT We thank Marjorie Robinson and the nursing staff of the Clinical Research Section for nursing support; Verna hoyioma for secretarial help; Karen Stone, Vera Rodriguez, Mongillo, and Tom Anderson for technical assistance. Most thank the Indian volunteers.
Phoenix KuwanVictoria of all we
REFERENCES 1. Rabinovitz D, Zierler KL: Forearm metabolism in obesity and its response to intra-arterial insulin: Characterization of insulin resistance and evidence for adaptive hyperinsulinism. J Clin Invest 41:2173-2181.1962 2. Kolterman OG, Insel J, Saekow M, et al: Mechanisms of insulin resistance in human obesity. Evidence for receptor and post receptor defects. J Clin Invest 651272-1284, 1980 3. Olefsky JM, Crapo PA, Ginsberg H, et al: Metabolic effects of increased caloric intake in man. Metabolism 24:495-503, 1975 4. National Diabetes Data Group: Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. Diabetes 28:1039-1057, 1979 5. Goldman RF, Buskirk ER: A method of underwater weighing and the determination of body density, in Brozek J, Hershel A (eds): Techniques for Measuring Body Composition. Washington DC, National Academy of Sciences, National Research Council, 196 1, pp 78-79 6. DeFronzo RA: Insulin secretion, Int J Obesity 6:73-82, 1982 (suppl)
insulin resistance
and obesity.
7. Rizza RH, Mandarin0 LJ, Gerich JE, et al: Dose-response characteristics for effects of insulin on production and utilization of glucose in man. Am J Physiol240:E63&E639, 1981 8. Lusk G: Animal calorimetry. Analysis of oxidation of carbohydrate and fat. J Biol Chem 59:41-42, 1924
of mixtures
9. Nuttal FQ, Barbosa J, Gannon MC, et al: The glycogen synthase system in skeletal muscle of normal humans and patients with myotonic dystrophy: Effect of glucose and insulin administration. Metabolism 23:561-568, 1974 10. Guinovart JJ, Salavert J, Massague J, et al: Glycogen synthase: A new activity ratio assay expressing a high sensitivity to the phosphorylation state. FEBS Letters 106:284-288, 1979
11. Hultman E: Muscle glycogen in man determined in needle biopsy specimens. Method and normal values. Stand J Clin Lab Invest 19:163, 1967 (suppl) 12. Steele R: Influences of glucose loading and of injected insulin on hepatic glucose output. Ann NY Acad Sci 82:420430, 1959 13. Herbert V, Lau K, Gotlieb CW, et al: Coated charcoal immunoassay of insulin, J CIin Endocrinol Metab 25:1375-1384, 1964 14. Yalow RS, Berson SA: Immunoassay of endogenous insulin in man. J Clin Invest 39:1157-l 167, 1960
plasma
15. Heding LG: Radioimmunological determination of human C-peptide in serum. Diabetologia 11:541-548, 1975 16. Frank BH, Pettee JM, Zimmerman RE, et al: The production of human proinsulin and its transformation to human insulin and C-peptide, in Rich EH, Gross E (eds): Peptides: SynthesisStructure-Function. Pierce Chemical Co, 1981 17. Frank BH, Beckage MJ, Wiley KA: High performance liquid chromatographic preparation of single-site carrier-free pancreatic polypeptide hormone radiotracers. J Chromatog 266:239-248, 1983 18. Best JD, Judzewitsch MA, Pfeifer JC, et al: The effect of chronic sulfonylurea therapy on hepatic glucose production in noninsulin dependent diabetes. Diabetes 31:333-338, 1982 19. Miles J, Glasscock R, Aikena J, et al: A microfluorometric method for the determination of free fatty acids in plasma. J Lipid Res 24196-99, 1983 20. Bogardus C, Lillioja S, Mott D, et al: Relationship between obesity and maximal insulin stimulated glucose uptake in vivo and in vitro in Pima Indians. J Clin Invest 73:80&805, 1984 21. Ruderman NB, Toews CJ, Shafrir E: Role of Free Fatty Acids in glucose homeostasis. Arch Intern Med 123:299-3 13, 1969 22. Bogardus CP, Thuillez P, Ravussin E, et al: Effect of muscle glycogen depletion and glycogen synthase activation on in vivo insulin action in man. J Clin Invest 72:1605-1610, 1983 23. Bogardus C, Lillioja S, Stone K, et al: Correlation between muscle glycogen synthase activity and in vivo insulin action in man. J Clin Invest 73:1185-l 190, 1984 24. Hultman E, Bergstrom J, Roth-Norlund E: Glycogen storage in human skeletal muscle, in Pernow, Saltin B (eds): Muscle Metabolism During Exercise. New York, Plenum Press, 1971, pp 273-288 25. Lawrence JC Jr, Larner J: Activation of glycogen synthase in rat adipocytes by insulin and glucose involves increased glucose transport and phosporylation. J Biol Chem 253:210&2113, 1978 26. Archer JA, Gordon P, Gavin JR, et al: Insulin receptors in human circulating lymphocytes: Application to the study of insulin resistance in man. J CIin Endocrinol Metab 36:627-633, 1973 27. Olefsky JM: Decreased insulin binding to adipocytes and circulating monocytes from obese subjects. J Clin Invest 57: 11651172.1976 28. Kashiwagi A, Mott D, Bogardus C, et al: The effects of short-term overfeeding on adipocyte metabolism in Pima Indians. Metabolism 34:364370, 1984
Induced
Decrease in Insulin Action for Glucose In Vivo and in Vitro in Man D.M. Mott,
Storage:
S. Lillioja, and C. Bogardus
The effect of short-term overnutrition on insulin action for glucose disposal was assessed in 16 Southwest American Indians (mean wt = 74 + 6 kg). After two weeks of weight maintenance and again after two weeks of 62% greater caloric intake (constant ratio of fat:carbohydrate:protein), insulin action for glucose disposal was measured using the euglycemic clamp technique with plasma insulin concentrations of about 110 and 1800 uU/mL. Simultaneous indirect calorimetry was used to estimate carbohydrate oxidation and storage rates. Following overnutrition, mean weight gain was 3.0 f 0.2 kg, P < 0.01. Overnutrition induced a decrease in glucose storage at the low and high insulin concentrations: 1.2 of:0.3 to 0.2 f 0.3, P < 0.01, and 6.4 f 0.3 to 4.3 + 0.5, mgl kg FFM min, P < 0.001. Carbohydrate oxidation was significantly increased at both insulin concentrations. The mean total insulin mediated glucose disposal rate decreased from 11.6 * 0.5 to 10.3 f 0.7, P < 0.01, at the high insulin concentration. This decrease was due entirely to the reduction in carbohydrate storage and was correlated with increased fasting insulin concentration (r = 0.7, P < 0.01). Overnutrition also induced a significant decrease in the percent muscle glycogen synthase active measured fasting and at the end of the high-dose insulin infusion. The results indicate that short-term overnutrition results in reduced insulin action for glucose storage and disposal which is correlated with increased fasting insulin concentrations. Reduced glycogen synthase activity may contribute to the effect of overnutrition on in vivo insulin-mediated glucose storage. m 1988 by Grune & Stratton, Inc.
0
BESE subjects have reduced rates of insulin-mediated glucose disposal’,* compared to lean subjects. It is not clear, however, whether this is the result of increased body mass, increased caloric intake, or some other factor of the obese condition. Olefsky et al’ reported that short-term overnutrition, with no change in diet composition, was associated with hyperinsulinemic responses to oral glucose and a meal of mixed composition. This increased insulin response, however, was not accompanied by a significant decrease in insulin action as measured by the steady state plasma glucose technique. The apparently inconsistent results from different measures of insulin action for glucose disposal, suggested that further studies were needed to define the possible importance that overnutrition per se, independent of large changes in weight, could have in reducing insulin action on glucose disposal. In this study, we have determined the effect of short-term overnutrition on in vivo insulin action in 15 Pima Indian subjects. Insulin action for glucose disposal was measured using the euglycemic clamp technique with simultaneous indirect calorimetry to estimate carbohydrate oxidation and storage rates. Insulin action was assessed in vitro by measuring muscle glycogen content and glycogen synthase activity in biopsies collected before and after overnutrition. In this study, short-term overnutrition induced a reduction in insulin-mediated glucose storage rates, which was associated with increased fasting plasma insulin concentrations and a reduction in muscle glycogen synthase activities.
From the Phoenix Clinical Research Section, National Institute of Arthritis. Diabetes, Digestive and Kidney Diseases, National Institutes of Health, 4212 North 16th St, Phoenix, AZ 85016. Address reprint requests to Dr Mott, Phoenix Clinical Research Section, NIADDK, 4212 North 16th St, Rm 541, Phoenix, AZ 85016. o 1986 by Grune & Stratton. Inc. 0026-0495~86/3S02-0013$03.00/0
160
MATERIALS AND METHODS Subjects
Fifteen Southwest American Indian males were admitted to the Phoenix clinical research ward. Their characteristics are shown in Table 1. After written informed consent was obtained, all subjects were physically examined and a 12-lead electrocardiogram was recorded. After an overnight fast, blood was drawn for complete blood count, liver function tests, BUN, creatinine, electrolytes, calcium, total protein, and albumin. After at least two days of a weight-maintaining diet containing at least 200 g of carbohydrate, an oral glucose tolerance test was performed.’ Results are reported only on subjects with a normal examination and 2-hour postglucose load plasma glucose concentrations of less than 160 mg/lOO mL. The percent body fat of each volunteer was estimated by underwater weighing with correction for the simultaneously measured residual lung volume.s Protocol
Subjects were placed on a weight-maintaining diet containing 45% carbohydrate, 40% fat, and 15% protein for 14 days. On days 9 through 14, a 24-hour urine C-peptide, oral glucose tolerance test, meal test, and euglycemic clamp with indirect calorimetry and muscle biopsies were performed in that order. Over the next 3 days the diet was stepwise increased to 62% above weight-maintenance requirements using a liquid supplement (Ensure Plus, Ross Laboratories, Columbus, Ohio) while keeping the same relative proportion of protein, carbohydrate, and fat. This diet was maintained for the next 11 days including the period of repeat testing. We have previously observed that a weight-maintenance diet during 10 days in the clinical research unit produced no effect on the parameters reported here from the euglycemic clamp, indirect calorimetry, or muscle biopsy (unpublished observations). Tolerance
Tests
The oral glucose tolerance test was performed after ingestion of 75 g of glucose.’ The meal consisted of orange juice, beef patty, eggs, toast, butter, and jelly, and contained 47% fat, 29% carbohydrate, and 24% protein (690 calories). It was consumed over a IO-minute
Metabolism, Vol35, No 2 (February), 1986: pp 160- 165
OVERNUTRITION
161
AND INSULIN ACTION
Table 1. Clinical Data Before and After Overnutrition Fasting
2-Hour Plasma
Plasma Glucose BMI Subject
Mean
25
(%)
WM
WM
2 3
GlUCose
hg/dl)
(mg/a)
Bodv Weight Gain Cki
WM
ON
WM
ON
3.0
88
85
22
1.5
98
89
110
135
30
3.2
89
75
159
112
28
25
3.3
87
93
121
116
18
26
18
3.8
83
86
96
145
6
23
21
16
2.5
80
74
83
77
7
23
22
15
3.3
81
86
107
126
8
30
27
27
4.1
80
86
146
129
9
24
23
22
3.0
77
77
117
138
10
20
23
24
2.0
88
92
129
131
11
21
25
24
2.0
79
85
88
103
12
27
23
20
3.9
87
95
104
99
13
18
21
27
3.4
82
90
90
106
14
28
28
29
2.4
95
86
148
112
15
22
21
21
2.8
83
87
118
105
23 k 1
3.0 k 0.2
85
27
29
21
29
22
30
4
26
5
i
23 + 1
SE
Abbreviations: + , percent extreme
lkalm’) G
1
Fat +
WM, fat
case where
was
weight not
25 k 1 maintenance;
determined
increased
fat mass
after was
ON,
f
2
86
98
*
2
114
128
*
6
117
+ 5
overnutrition.
overnutrition the only source
but
using
of weight
the
subject’s
gain,
period of time. During both tolerance tests blood samples were drawn for determination of plasma glucose and insulin concentrations at 0, 30.60, 120, and 180 minutes.
Euglycemic Clamp At 5 AM hours after a IO-hour overnight fast, an intravenous catheter was placed in an antecubital vein for infusion of insulin, glucose, and 3-H’ glucose. A primed (30 uCi)-continuous (0.30 uCi/min) infusion of 3-H’ glucose was then begun and continued until the end of the study. Another catheter was placed retrograde in a dorsal vein of the contralateral hand for blood withdrawal. The hand was kept in a warming box at 70 “C. After 2.5 hours four blood samples were drawn over 30 minutes for 3-H’ glucose-specific activity determination. A primed continuous insulin infusion of 40 mu/m* was then started and continued for 100 minutes (low dose). Another primed insulin infusion of 400 mu/m* was then started and continued for an additional 100 minutes (high dose). After the start of the initial insulin infusion, a variable infusion of 20% glucose was given, as necessary, to maintain the plasma glucose concentration at approximately 90 mg/ 100 mL. Blood for plasma glucose concentration was drawn every 5 minutes throughout the test. Blood for plasma insulin and 3-H3 glucose-specific activity was drawn every 20 minutes from 60 to 100 (low dose) and from 160 to 200 minutes (high dose). Steady-state glucose and insulin levels were determined during these two time intervals (Table 2). The high plasma insulin concentration achieved with 400 mu/m’ infusion was considered to be a maximally insulin stimulating concentration.2.6.7
mean
weight
% fat would
maintenance
have increased
23%
of 75
(WM)
+ 2kg
to 26%
and
assuming
the
(ON).
zirconium cell analyzer (Applied Electrochemistry, Sunnyvale, Calif) and the carbon dioxide analyzer was an infrared analyzer (Applied Electrochemistry, Sunnyvale, Calif). The analyzers and flowmeter outputs were connected to a desktop computer (HewlettPackard, Palo Alto, Calif). This recorded continuous, integrated calorimetric measurements every 5 minutes for the hour before and for the duration of the euglycemic clamp. The protein oxidation during the test was estimated from the urinary urea production rate. The nonprotein respiratory quotient was then calculated and the substrate oxidation rates determined from the equations of Lusk.’
Biopsy and Determination of Glycogen Synthase and Glycogen Content
Muscle
Three hours after the start of the 3-H’ glucose infusion and just prior to starting the insulin infusion, a percutaneous biopsy of the vastus lateralis muscle was performed. The specimen was immediately frozen in liquid nitrogen, stored at - 70 “C and later assayed for glycogen content and glycogen synthase activity. The muscle biopsy was repeated at the end of insulin infusion. The method of Nuttal et aI9 was modified for homogenization of muscle biopsies. Aliquots of the homogenate were stored at -20 “C for the determination of glycogen content. The remaining homogenate was centrifuged at 8000 x G for ten minutes and aliquots of the supernatant
Table 2. Mean Steady-State
Glucose and insulin Concentration
During Euglycemic Clamp Weight
Indirect Calorimetry For one hour before and for the duration of the euglycemic clamp procedure, a clear, plastic, ventilated hood was placed over the subject’s head. Room air was drawn through the hood and the flow rate measured by a pneumotachograph (Gould, Netherlands). A constant fraction of expired air was withdrawn and analyzed for oxygen and carbon dioxide content. The oxygen analyzer was a
weight
from
Maintenance
Low Dow, Glucose
(%)
Insulin C.V.
Low Dose
High Dose
(mg/
dL) C.V.
Overnutrition
High Dose
(ulJ/mLI
(%I
P < 0.05
89.6
+ 0.4
90.6
f
0.3
0.5
89.4
2.5
-c 0.2
3.6
r
0.3
2.3
+ 0.2
3.0
108
I
1810
i
100
117
+ 5’
5.5
+ 0.7
5.9
* 0.9
compared
4
to weight
5.9
+ 0.7
maintenance
90.3
f
by paired
1820 6.3 f-test.
+ 0.4 + 0.2 *
110
+ 0.6
162
MOlT ET AL
were stored at -20 OC for protein assay or used immediately for determination of glycogen synthase activity. The active form of glycogen synthase was assayed at low glucose-6-phosphate (G-6-P) concentration (0.17 mmol/L) in the presence of 0.14 mmol/L UDP glucose.” Total activity was measured at high G-6-P (7.2 mmol/L), also in the presence of 0.14 mmol/L UDP glucose. Activity was measured using UDP “C-glucose and is expressed as nanomoles of glucose incorporated into glycogen per minute per milligram of protein. The percent glycogen synthase active is calculated from activity at low G-6-P concentration divided by total activity. Frozen aliquots for glycogen determination were thawed and assayed according to the method of Hultman et al.”
Oral GlucoseTobrmce Test
(?sqGlreerd
s-
Meal
1690 C&ml- 475cFat. [email protected]. Rolsnl RdOn%Am .-m-w.
‘601
Calculations The appearance rate (Ra) of glucose in the plasma was calculated from the blood 3-H’ glucose-specific activities using the equations of SteeIe,12 in their steady-state form for the basal period and the nonsteady form for the Ra during the euglycemic clamp. During the basal period the Ra equals the endogenous glucose production rate. During the euglycemic clamp the endogenous glucose production rate equals the difference between the rate of exogenously infused glucose and the Ra determined from Steele’s equations. When the Ra equals the exogenous glucose infusion rate, the endogenous glucose production rate is assumed to be completely suppressed so that the total glucose disposal rate equals the exogenous glucose infusion rate. In these experiments these data were calculated for each 20 minutes of the last 40 minutes during the low- and high-dose euglycemic clamp. Data from each 40-minute period was averaged to calculate the total glucose disposal or “M” value. The basal and stimulated carbohydrate oxidation rates were calculated from the indirect calorimetry data by averaging the data for 40 minutes prior to the beginning of the insulin infusion and for the last 40 minutes during each of the two rates of insulin infusion. The glucose storage rate was estimated by subtracting the carbohydrate oxidation rate from the total glucose disposal rate. The glucose disposal values are expressed as milligram of carbyhydrate per kilogram of fat-free mass per minute. The fat-free mass is determined from underwater weighing.
Analytical Methods and Statistics Plasma insulin concentration were determined using the Herbert modification” of the radioimmunoassay of Yalow.” Urinary C-peptide concentrations were determined by radioimmunoassay as synthesized human C-peptide previously described. I5 Bacterially was used as standardI and was also used to prepare the tracer, I’Z5-t-Boc-tyr-C-peptide” and goat anti-human-C-peptide antiserum. Materials for the C-peptide assay were generously provided by Lilly laboratories (Indianapolis, Ind). Trititated glucose-specific activity in blood samples was determined after precipitating protein with perchloric acid as described by others.‘* Samples for fatty acid determinations were collected in tubes containing 1.l mg/mL diethyl p-nitrophenyl phosphate. (Sigma, St. Louis, Miss) and kept on ice. Concentrations of free fatty acids in plasma were measured using the microfluorometric method of Miles et a1.19 All data are expressed as the mean 5 the standard error of the mean. All statistical analyses were calculated using Statistical Analyses System, SAS Institute, Inc, Cary, N.C. RESULTS Overnutrition 3.0 k 0.2 kg,
for 13 days
produced
P < 0.01. There
a mean
weight
was no significant
gain
change
of in
glucose concentration, but the mean fasting plasma insulin concentration, measured on the mornmean
fasting
plasma
l-laws Fig 1. Effect of overnutrition on the response of glucose and insulin to 75 g oral glucose (left panels) and a 690 calorie standard meal (right panels). Results are the mean values for 15 subjects measured after weight maintenance (0) and overnutrition (0).
ing of the euglycemic clamp, increased significantly from 20 -+ 1 to 27 + 3, uU/mL, P < 0.005. The 24-hour urinary C-peptide also increased from 37 2 7 to 69 + 13 nmol, P < 0.01. Plasma-free fatty acids decreased from 7.6 + 0.9 to 5.4 f 0.7 mg/dL (NS). Following overnutrition, the increases in the insulin and glucose responses to oral glucose were not significant. However, there was a significant increase in the insulin response 240 + 30 v 302 f 33 uU/mL . 3 h f SE, P < 0.005 to the test meal with no significant change in the mean area under the glucose curve (Fig 1). Overnutrition was associated with increased basal and steady-state plasma insulin concentrations during the lowdose insulin infusion (Fig 2). In spite of this, glucose disposal was not significantly changed in the basal state or during the low-dose insulin infusion. Overnutrition significantly reduced_glucose storage and increased carbohydrate oxidation at all three insulin concentration. Overnutrition induced changes in these two parameters were significantly correlated in the basal state (r = -0.9, P < 0.001) at the end of the low-dose clamp (r = 0.7, P < O.Ol), but not at the end of the high-dose clamp (r = 0.5, P -G0.06). At the high insulin concentration, the decrease in glucose storage completely accounted for the significant reduction in the glucose disposal rate (see Fig 2). The muscle glycogen content and glycogen synthase activity was measured in muscle biopsies obtained from 9 subjects before (fasting) and after the insulin infusion for the euglycemic clamp. Similar to the results in Fig 2, this subgroup of
OVERNUTRITION
AND INSULIN
Fasting ~sposd
stomga~
oxidotimn
163
ACTION
LOWDOW
is.
.
Hiih Dose
.
IiS.
.
.
co.0 I
aooo5
am
aoool
Relationship between the overnutrition-induced Fig 3. decrease in glucose storage rates at the high dose of insulin and the overnutrition-induced increase in fasting insulin, r = 0.7; P -c 0.005.
106’4 WM
after (P < 0.05) the euglycemic clamp. In regard to the effect of overnutrition on glycogen concentration and glycogen synthase activity measured at the end of the clamp on these 9 subjects, the largest increase in muscle glycogen content (overnutrition-weight maintenance) was associated with the smallest decrease in the active form of glycogen synthase (r = 0.9, P < 0.005). The results in Fig 3 show that decreases in glucose storage during the high-dose clamp are correlated (r = 0.7, P .C 0.005) with the overnutrition-induced increases in fasting insulin concentration. The decreases in glucose disposal rates are also correlated with increased fasting insulin concentrations (r = 0.7, P < 0.005). Following overnutrition, those subjects with the largest overfeeding-associated increase in fasting insulin concentrations, had the largest decrease in the active form of the enzyme (r = 0.53, P = 0.14) and the smallest increase in glycogen (r = 0.7, P -c0.05), measured after the insulin infusion. No significant correlation was observed between the decrease in glucose storage rates and the change in muscle glycogen content or glycogen synthase activity.
117*5* ON
Fig 2. Effect of overnutrition on glucose disposal (height of bar) as measured by hepatic glucose production and the euglycemic clamp at basal, low. and high insulin levels at the end of the weight maintenance (WM) and overfeeding (ON) periods. Mean basal and steady state insulin concentrations are shown below the bars. Indirect calorimetry was used to measure glucose oxidation (clear area) and glucose storage (shaded area). All results are mean f SE. The SE values for glucose oxidation and storage were similar for each glucose disposal rate. and error bars are shown only for glucose disposal and storage. l = P < 0.05 and l * = P i 0.005 compared to weight-maintenance insulin concentration.
nine subjects had significant overnutrition induced reductions in carbohydrate storage before and during the low-dose and high-dose clamp (data not shown). On the weightmaintenance diet, muscle glycogen and the active form of glycogen synthase were not correlated before the euglycemic clamp but were positively correlated at the end of the clamp (r = 0.9. P < 0.001). The results in Table 3 show that overnutrition was associated with an increase in muscle glycogen content at the end of the euglycemic clamp (P < 0.05). Overnutrition also induced a significant decrease in the percent glycogen synthase active before (P < 0.05) and
DISCUSSION
This study was designed to determine if short-term overnutrition is associated with reduced insulin action for glucose disposal. The results indicate that overnutrition results in
Table 3. Effect of Overnutrition on Muscle Glycogen Content and Glycogen Synthase Activity Before and After Clamp Procedure* Wmght
Maintenance
Overnutrition
BeforeClamp
After Clamp
(g/ 100 g tissue) % Glycogen
1.5 + 0.2
1.6 + 0.1
2.5 + 0.7
Synthase Active Total Glycogen
28 + 4
59 t 4t
22 zk 3$
9.3 * 0.9
7.5 ? 0.9
BeforeClamp
After Clamp
Glycogen 2.0
r 0.2$
48 t 4tt
Synthase Activity (nmol/min. mg protein) *Comparisons made by paired t-test. tf
< 0.00 1 compared to before clamp.
$P i 0.05 compared to weight maintenance.
a.9 i
1.0
7.9
k 0.9
164
reduced in vivo insulin action for glucose disposal as estimated by the insulin and glucose response to a test meal and by the euglycemic clamp. Indirect calormetry during the euglycemic clamp indicated that overnutrition decreased glucose storage. In addition, overnutrition is associated with decreased skeletal muscle glycogen synthase activity. The mean results of the oral glucose tolerance test on 15 subjects were suggestive of an overnutrition-induced decrease in glucose tolerance but increases in both the area under the glucose and insulin curves failed to reach statistical significance. The significant increase in the insulin response to a meal of mixed composition could be interpreted as an overnutrition induced loss of insulin action. The results in Fig 2 indicate that short-term overnutrition reduces insulin action on glucose storage at fasting, low- and high-dose concentrations of insulin. The decreases in storage and disposal at the high insulin dose was correlated with the overnutrition-associated increase in fasting insulin concentrations. It is not clear, however, whether the increase in fasting insulin is the cause of or secondary to the overnutrition-associated decrease in glucose storage. In a previous report” using an identical high-dose clamp technique and diet composition, the mean glucose disposal differed from 13.1 2 0.5 mg/kg FFM ; min for 10 lean subjects (mean percent fat = 13 * 1) to 10.5 + 0.6 for 20 obese subjects (mean percent fat = 38 + 1). This represents a 20% decrease and 70% of this difference was due to reduced glucose storage. Our group of 15 subjects on weight maintenance diet had a glucose disposal rate of 11.6 + 0.5 mg/kg FFM - min, and a mean % fat of 23 ? 1, intermediate between the latter groups in both glucose disposal rate and % fat. As a result of overnutrition (mean Cal/d = 4,400) these subjects increased mean % fat by no more than 3% but decreased their glucose disposal rate to 10.3 + 0.7 mg/kg FFM - min (11% decrease), which is similar to the glucose disposal rate reported for the obese group (mean Cal/d = 3,100). All of this decrease was due to reduced glucose storage. These results suggest that overnutrition might be important for the mechanism of reduced rates of insulinmediated glucose disposal in obese subjects. However, decreased insulin action in obese subjects is characterized by both decreased glucose oxidation as well as storage.*’ In contrast, the overnutrition-induced decrease in glucose storage was significantly correlated with an increase in glucose oxidation. Although increased carbohydrate oxidation appears to compensate for part of the loss in glucose storage, the mechanisms coordinating these pathways are not known. The effect of decreased plasma free fatty acids to increase carbohydrate oxidation has been extensively reviewed.*’ Because plasma-free fatty acids in this report were not significantly changed by overnutrition, other factors may have stimulated carbohydrate oxidation. Results of an earlier study by Olefsky et al3 on the effects of overnutrition on insulin action for glucose disposal are similar to some of the results reported here. Both studies demonstrate no overnutrition-induced change in insulinmediated glucose disposal when insulin infusion raised plasma insulin concentrations about 100 uU/mL. The previous study, however, only observed evidence of reduced
MO-IT ET AL
insulin-mediated glucose disposal during a mixed meal test. Here insulin action for glucose disposal appears to be decreased by overnutrition when measured during the mixed meal test and when measured during high insulin dose infusion (+ 1800 uU/mL). The observed overnutritioninduced decrease in glucose disposal at the high but not the low insulin dose is probably caused by the following: (1) The decrease in glucose disposal is the result of an apparent decrease in the maximal velocity for glucose storage measured at the high insulin dose. This decrease in glucose storage is less prominent at the low insulin dose. (2) The overnutrition-induced increase in carbohydrate oxidation completely compensates for the reduced glucose storage at the low insulin dose, but not at the high insulin dose. It is, therefore, possible that the apparent inconsistent results of the low-dose clamp and the mixed meal test could be explained on the basis of a different response of each test to changes in maximal velocity of storage and/or to changes in the relative contribution of carbohydrate oxidation and storage. Further information on the regulation of glucose disposal during these two tests is needed to explain the apparent inconsistencies. The pathways of carbohydrate metabolism that account for the carbohydrate storage component measured by indirect calorimetry have not been fully identified. In a previous report, basal glycogen synthase activity in human muscle biopsies correlated with carbohydrate storage rates in glycogen-depleted subjects.22 We have reported that the stimulation of muscle glycogen synthase activity measured in biopsies obtained after insulin infusion, correlated with the range of glucose storage rates observed in glucose tolerant and intolerant subjects.23 These observations suggested that regulation of glycogen synthase activity may be a determinant of in vivo insulin-mediated glucose storage and disposal. In this study, overnutrition reduced the mean values for the percent active glycogen synthase measured before and after the clamp. This observation suggests that the reduction in the active form of the enzyme could contribute to the decreased rate of insulin-mediated glucose storage. It is not clear why overnutrition-associated increases in fasting insulin appear related to reduced glucose storage. The rise in insulin concentration may be secondary to the effects of overnutrition which include loss of glycogen synthase activity. Decreased glycogen synthase activity could be a result of increased muscle glycogen content which has been shown to inhibit glycogen synthase activity.24 This is an unlikely explanation, however, considering that subjects with the largest overnutrition-associated increase in muscle glycogen content had the smallest decrease in active glycogen synthase. A complete explanation for the overnutrition-induced decrease in carbohydrate storage and its relationship to carbohydrate oxidation will require further study to determine if most or only a small part of insulin-mediated glucose storage measured by indirect calorimetry is accounted for by glycogen synthesis. Another possible explanation for the effects of overnutrition on glucose storage would be a decrease in glucose transport leading to reduced G6P and an associated decrease in glycogen synthase activity.*’ This explanation seems
OVERNUTRITION
AND INSULIN ACTION
165
unlikely because reduced transport would probably also decrease carbohydrate oxidation. It is possible, however, that the principle sources of carbohydrate storage and oxidation occur in different tissues and would not necessarily be expected to change in the same direction. It is also possible that hyperinsulinemia induces a reduction in glucose disposal that is secondary to decreased insulin binding.26*27Based on studies in fat cells, none of these mechanisms seem likely. We have recently reported that overnutrition produced a significant increase in human basal and maximal insulin stimulated fat cell glucose transport with no significant change in insulin binding.** Results in the fat cell, however, may not reflect insulin binding and glucose transport in the tissues responsible for the overnutrition-induced decrease in glucose storage. In summary, reduced glucose storage is the primary component of reduced rates of glucose disposal during the high-dose clamp of both overfed normal glucose tolerant subjects (compared to weight maintenance) and in a previous
report of obese subjects (compared to lean). We have also previously reported that subjects on weight-maintenance diet with low glucose storage and disposal rates have reduced muscle glycogen synthase activity. Here we observe that normal glucose tolerant subjects with overnutrition-reduced glucose storage also have reduced glycogen synthase activity. These results suggest that overnutrition, independent of increased body mass, may contribute to the reduced insulinmediated glucose storage rates observed in obese subjects and that a reduction in glycogen synthase activity could be important to this mechanism.
ACKNOWLEDGMENT We thank Marjorie Robinson and the nursing staff of the Clinical Research Section for nursing support; Verna hoyioma for secretarial help; Karen Stone, Vera Rodriguez, Mongillo, and Tom Anderson for technical assistance. Most thank the Indian volunteers.
Phoenix KuwanVictoria of all we
REFERENCES 1. Rabinovitz D, Zierler KL: Forearm metabolism in obesity and its response to intra-arterial insulin: Characterization of insulin resistance and evidence for adaptive hyperinsulinism. J Clin Invest 41:2173-2181.1962 2. Kolterman OG, Insel J, Saekow M, et al: Mechanisms of insulin resistance in human obesity. Evidence for receptor and post receptor defects. J Clin Invest 651272-1284, 1980 3. Olefsky JM, Crapo PA, Ginsberg H, et al: Metabolic effects of increased caloric intake in man. Metabolism 24:495-503, 1975 4. National Diabetes Data Group: Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. Diabetes 28:1039-1057, 1979 5. Goldman RF, Buskirk ER: A method of underwater weighing and the determination of body density, in Brozek J, Hershel A (eds): Techniques for Measuring Body Composition. Washington DC, National Academy of Sciences, National Research Council, 196 1, pp 78-79 6. DeFronzo RA: Insulin secretion, Int J Obesity 6:73-82, 1982 (suppl)
insulin resistance
and obesity.
7. Rizza RH, Mandarin0 LJ, Gerich JE, et al: Dose-response characteristics for effects of insulin on production and utilization of glucose in man. Am J Physiol240:E63&E639, 1981 8. Lusk G: Animal calorimetry. Analysis of oxidation of carbohydrate and fat. J Biol Chem 59:41-42, 1924
of mixtures
9. Nuttal FQ, Barbosa J, Gannon MC, et al: The glycogen synthase system in skeletal muscle of normal humans and patients with myotonic dystrophy: Effect of glucose and insulin administration. Metabolism 23:561-568, 1974 10. Guinovart JJ, Salavert J, Massague J, et al: Glycogen synthase: A new activity ratio assay expressing a high sensitivity to the phosphorylation state. FEBS Letters 106:284-288, 1979
11. Hultman E: Muscle glycogen in man determined in needle biopsy specimens. Method and normal values. Stand J Clin Lab Invest 19:163, 1967 (suppl) 12. Steele R: Influences of glucose loading and of injected insulin on hepatic glucose output. Ann NY Acad Sci 82:420430, 1959 13. Herbert V, Lau K, Gotlieb CW, et al: Coated charcoal immunoassay of insulin, J CIin Endocrinol Metab 25:1375-1384, 1964 14. Yalow RS, Berson SA: Immunoassay of endogenous insulin in man. J Clin Invest 39:1157-l 167, 1960
plasma
15. Heding LG: Radioimmunological determination of human C-peptide in serum. Diabetologia 11:541-548, 1975 16. Frank BH, Pettee JM, Zimmerman RE, et al: The production of human proinsulin and its transformation to human insulin and C-peptide, in Rich EH, Gross E (eds): Peptides: SynthesisStructure-Function. Pierce Chemical Co, 1981 17. Frank BH, Beckage MJ, Wiley KA: High performance liquid chromatographic preparation of single-site carrier-free pancreatic polypeptide hormone radiotracers. J Chromatog 266:239-248, 1983 18. Best JD, Judzewitsch MA, Pfeifer JC, et al: The effect of chronic sulfonylurea therapy on hepatic glucose production in noninsulin dependent diabetes. Diabetes 31:333-338, 1982 19. Miles J, Glasscock R, Aikena J, et al: A microfluorometric method for the determination of free fatty acids in plasma. J Lipid Res 24196-99, 1983 20. Bogardus C, Lillioja S, Mott D, et al: Relationship between obesity and maximal insulin stimulated glucose uptake in vivo and in vitro in Pima Indians. J Clin Invest 73:80&805, 1984 21. Ruderman NB, Toews CJ, Shafrir E: Role of Free Fatty Acids in glucose homeostasis. Arch Intern Med 123:299-3 13, 1969 22. Bogardus CP, Thuillez P, Ravussin E, et al: Effect of muscle glycogen depletion and glycogen synthase activation on in vivo insulin action in man. J Clin Invest 72:1605-1610, 1983 23. Bogardus C, Lillioja S, Stone K, et al: Correlation between muscle glycogen synthase activity and in vivo insulin action in man. J Clin Invest 73:1185-l 190, 1984 24. Hultman E, Bergstrom J, Roth-Norlund E: Glycogen storage in human skeletal muscle, in Pernow, Saltin B (eds): Muscle Metabolism During Exercise. New York, Plenum Press, 1971, pp 273-288 25. Lawrence JC Jr, Larner J: Activation of glycogen synthase in rat adipocytes by insulin and glucose involves increased glucose transport and phosporylation. J Biol Chem 253:210&2113, 1978 26. Archer JA, Gordon P, Gavin JR, et al: Insulin receptors in human circulating lymphocytes: Application to the study of insulin resistance in man. J CIin Endocrinol Metab 36:627-633, 1973 27. Olefsky JM: Decreased insulin binding to adipocytes and circulating monocytes from obese subjects. J Clin Invest 57: 11651172.1976 28. Kashiwagi A, Mott D, Bogardus C, et al: The effects of short-term overfeeding on adipocyte metabolism in Pima Indians. Metabolism 34:364370, 1984