You are on page 1of 7

0021-972X/00/$03.

00/0 The Journal of Clinical Endocrinology & Metabolism Copyright 2000 by The Endocrine Society

Vol. 85, No. 10 Printed in U.S.A.

Chronic Alcohol Intake Differently Influences Glucose Metabolism According to Nutritional Status*
BONG SOO CHA, CHUL WOO AHN, YOUNG DUK SONG, SUNG KIL LIM, KYUNG RAE KIM, KAP BUM HUH, AND HYUN CHUL LEE
Division of Endocrinology, Department of Internal Medicine, College of Medicine, Yonsei University, Seoul, Korea
ABSTRACT We investigated the potential different effects of a chronic alcohol intake on glucose metabolism according to nutritional status in growing rats. Eighty weanling 4-week-old male Sprague Dawley rats were fed with low (5%, wt/wt) or control (22%) protein diet for 8 weeks. Each group was subdivided into alcohol (5 g/kg 1 day 1) or saline gavage rats during the last 4 weeks. At 12 weeks of age, we measured the weights of the body, pancreas, and epididymal fat; glycogen synthase activity of gastrocnemius muscle; and insulin content of the pancreas. We performed an ip glucose tolerance test and a euglycemic hyperinsulinemic clamp test. Weight gain was almost arrested in proteindeficient rats. The relative weight and insulin content of the pancreas and glycogen synthase activity were not different among the four groups, but the relative amount of epididymal fat decreased only in protein-deficient saline rats. Insulin response after glucose challenge and glucose disposal rate during the euglycemic clamp were both markedly decreased in protein-deficient saline rats, but not changed in protein-deficient alcohol rats. Protein-deficiency per se causes deterioration both in insulin secretory function and in sensitivity, but these defects are protected by a chronic alcohol intake. In a proteinsufficient state, alcohol intake gives no additional effects on glucose metabolism. Therefore, according to individual nutritional status, the metabolic effect of alcohol intake appears differently. (J Clin Endocrinol Metab 85: 3646 3652, 2000)

LUCOSE INTOLERANCE is a common feature of protein malnutrition, especially in children suffering from severe protein deficiency (1). Dysfunction of cells occurs in a malnourished state through a combination of islet damage, impaired repair mechanism, and an overall reduction in cell mass (1, 2). However, some confusion still exists in the literature concerning this point because high fasting plasma insulin levels or sustained insulin secretion have been found in some malnourished children (3). The effects of alcohol on glucose metabolism are complex. Several studies have indicated that alcohol infusion or ingestion per se does not alter basal insulin secretion in human (4) or in rats (5). Meanwhile, other experiments have demonstrated that alcohol inhibits (6) or stimulates (4, 7, 8) glucose-stimulated insulin secretion and increases (9, 10) or decreases (11) insulin sensitivity. With regard to the risk of noninsulin-dependent diabetes mellitus, heavy alcoholics have positive correlation (12) and light-to-moderate drinkers have negative (9, 13) correlation. A chronic, moderate amount of alcohol intake doesnt seem to deteriorate glucose metabolism even in noninsulin-dependent diabetes mellitus patients (14, 15). These diverse results suggest that alcohol influences glucose metabolism differently according to the amount and duration of alcohol intake. Presently, the combined effect of protein malnutrition and chronic alcohol intake on glucose metabolism has not been elucidated.
Received February 18, 2000. Revision received June 23, 2000. Accepted June 28, 2000. Address correspondence and requests for reprints to: Hyun Chul Lee, M.D., Department of Internal Medicine, School of Medicine, Yonsei University, 134 Shinchon-Dong, Seodaemoon-Ku, P.O. Box 120-749, Seoul, Korea. E-mail: endohclee@yumc.yonsei.ac.kr. * Supported by grants from the Korean Diabetes Association and Medical Research Foundation of Yonsei University Medical School.

In the present study, we investigated whether protein deficiency from the weaning period influences insulin secretory function and peripheral insulin sensitivity and how a chronic alcohol intake could modulate the metabolic effects caused by protein deficiency in growing rats.
Materials and Methods Materials
We purchased human biosynthetic insulin and BSA (fraction V) from Eli Lilly & Co. (Indianapolis, IN) and Boehringer Mannheim (Indianapolis, IN). Uridine diphosphate (UDP)-[14C]glucose was purchased from Amersham Pharmacia Biotech (Buckinghamshire, UK). Whole blood glucose analyzer was obtained from Johnson & Johnson (Milpitas, CA). Kits for determination of rat plasma insulin concentration were from Linco Research, Inc. (St. Charles, MO). Other biochemicals and chemicals were from Sigma (St. Louis, MO).

Animals
The principles of laboratory animal care of the NIH guidelines were followed in all these experiments. Male Sprague Dawley rats were obtained from a local colony. They were weaned at 21 days of age and received a control diet for 1 week. They weighed 70 75 g and were randomly allocated into four experimental groups, matched for initial body weight. A total of 80 rats were used in this experiment. During 8 weeks, each group received a protein-deficient or protein-sufficient diet, respectively, until the age of 12 weeks. During the last 4 weeks, half of the rats were fed with alcohol or saline. The groups were defined as protein-deficient alcohol rats (group I), protein-deficient saline rats (group II), protein-sufficient alcohol rats (group III), and protein-sufficient saline rats (group IV or control rats). Rats were housed in groups of four. Foods and water were provided ad libitum and maintained under 12L:12D lighting conditions. The amount of dietary intake was measured everyday. At the age of 12 weeks, 10 rats from each group were randomly selected to undergo an ip glucose tolerance test (ipGTT) and then were sacrificed for extraction of epididymal fat (for weight measure). The remaining rats in each group were tested for euglycemic hyperinsulinemic clamp and sacrificed to obtain the pancreas (for ex-

3646

COMBINED EFFECT OF ALCOHOL AND MALNUTRITION


traction of insulin) and gastrocnemius muscle [for measurement of glycogen synthase (GS) activity]. Body weight was recorded every week. Experiments were performed on 12- to 14-h fasted rats. This experimental protocol was approved by the Committee on Animal Investigation of the Yonsei University.

3647

before and 30 and 60 min after injection. The plasma glucose level was measured by glucose oxidase method. The plasma insulin concentration was measured by rat insulin RIA kit. Because repetitive sampling is a problem in rats that leads them to hypovolemic stress (5.5 6.5 mL/kg blood loss) due to their small blood volume (17), we obtained the smallest amount of blood required.

Diet composition and alcohol administration


For the compensation of the deficient calories in low protein diet, we increased the proportion of carbohydrate. Both protein-deficient and protein-sufficient diets were, thus, isocaloric and identical, except for their protein content. The powdered semisynthetic experimental diets containing low (5%, wt/wt) or control (22%) protein were generously provided by Prof. J. H. Lee (College of Ecology, Yonsei University, Seoul, Korea). The protein-sufficient diet contained by weight: 50 g/100 g dextrose, 8 g/100 g corn starch, 5 g/100 g cellulose-powder, 5 g/100 g corn oil, 0.2 g/100 g dl-methionine, and 22 g/100 g protein (casein); and by calories: 64.5% carbohydrate, 12.6% lipid, and 22.9% protein. The low-protein diet contained: 67 g/100 g dextrose, 8 g/100 g corn starch, 5 g/100 g cellulose-powder, 5% corn oil, 0.05% dl-methionine, and 5% protein; and by calories: 81.6% carbohydrate, 12.6% lipid, and 5.8% protein. Both semisynthetic diets contained 1.8 g/100 g CaCO3 and the same mineral mixture (3 g/100 g) and vitamin mixture (5 g/100 g). Energy content by 100-g diet was the same (353 calories) in both diets. The vitamin mixture included 0.02 mg/g folic acid, 0.004 mg/g biotin, 0.04392 mg/g vitamin A, l 0.0005 mg/g cholecalciferol, 0.01 mg/g menadione, 0.2 mg/g thiamin, 0.12 mg/g riboflavin, 0.12 mg/g pyridoxine, 0.32 mg/g calcium pantothenate, 0.5 mg/g niacin, 1.0 mg/g ascorbic acid, 2.0 mg/g -tocoperol, 30.0 mg/g choline, 0.001 mg/g vitamin B12, and sucrose to make 1 g. The mineral mixture included 339.47 mg/g CaHPO4, 216.57 mg/g KH2PO4, 45.96 mg/g K2CO3, 78.27 mg/g NaCl, 21.6 mg/g K3C6H5O7 H2O, 51.28 mg/g MgCO3 H2O, 1.5343 mg/g ZnCO3, 4.065 mg/g MnCO3, 0.3044 mg/g CuCO3, 6.3355 mg/g ferric citrate, 0.0112 mg/g KIO3, 0.0112 mg/g NaSeO3 5H2O, 0.6403 mg/g CrK(SO4)2 12H2O, and sucrose to make 1 g. For alcohol rats, ethanol was given once daily in the morning between 0800 and 0900 h by gastric tube as a 25% (vol/vol) solution in tap water. Saline rats were intubated with an equal volume of saline. To avoid acute alcohol intoxication and to induce alcohol tolerance, the alcohol rats were intubated with ethanol at a daily dose of 1.25 g/kg for 2 days, 2.5 g/kg for next 2 days, 3.75 g/kg for another 2 days, and a maintaining dose of 5 g/kg for the remainder of the period (16). Blood alcohol concentrations in control rats fed with alcohol for 4 weeks are shown in Fig. 1.

Euglycemic hyperinsulinemic clamp test


This test was performed at 0800 h in rats fasted from the previous day at 2000 h according to a previously detailed procedure (18). Rats were anesthetized with pentobarbital (50 mg/kg, ip). Body temperature was maintained at 3738 C with a heating lamp. The right femoral artery was catheterized for blood sampling. A blood sample was obtained 20 min after surgery for the determination of basal blood glucose concentration. Insulin was infused at a constant rate of 6 mU/kg 1 min 1 via the left femoral vein with an infusion pump (Harvard Apparatus 22; Harvard, Natick, MA). Human insulin was dissolved in 0.9% NaCl containing 1% BSA (fraction V). Glucose (25%, wt/vol) infusion was started 5 min after insulin infusion through a second iv catheter with its infusion rate adjusted to sustain plasma glucose at 6 mmol/L using PACBERG algorithm (19). Then, 25 l blood were sampled from the right femoral artery at 10-min intervals, and blood glucose concentrations were determined within 30 sec with a glucose analyzer. Stress from blood loss of more than 7 mL/kg may be a source of errors in the evaluation of glucose turnover and insulin sensitivity during the clamp experiments in rat (17). Because we limited maximum sampling volume as 1 mL/kg per sampling, the total blood loss was less than 7 mL/kg of body weight.

Muscle GS activity
At the end of the euglycemic clamp test, gastrocnemius muscle was excised rapidly and frozen in liquid N2. About 50 mg frozen state gastrocnemius muscle was immediately placed into 1 mL ice-cold GS buffer solution [20 mmol/L EDTA, 25 mmol/L NaF, and 50 mmol/L Tris-HCl (pH 7.8)] and homogenized for 30 sec in a glass homogenizer. The extract was centrifuged for 20 sec at 20,000 g, and the supernatant, which contained more than 95% of GS activity, was removed and retained on ice for subsequent GS assay. GS activity was measured using a modified method of Thomas et al. (20). In summary, the supernatant was diluted five times by GS buffer solution, and 30 l were taken for incubation in a 30 C water bath with 60 l substrate mixture consisting of 0.37 mmol/L UDP-glucose, 1.5% glycogen, 0 10 mmol/L glucose6-phosphate (G6P), and 0.7 Ci UDP-[14C]glucose for 25 min. After incubation, 75 l were taken and wetted in filtering paper (Whatman No 2.5 2.5; Whatman). These filter papers were immediately placed and washed in a 66% ethanol solution that had been stored at 20 C three times (5 min, 30 min 2) and washed again with acetone for 5 min, and then dried. The radioactivity of the remaining 14C- in the filter paper was measured by liquid scintillation spectrophotometry. Protein was assayed using the Lowry protein measuring method. GS activity was expressed as nanomoles of UDP-[14C]glucose incorporated into glycogen per min/mg protein or as a fractional velocity (FV), a percentage of the ratio of activity at 0 mmol/L G6P per 10 mmol/L G6P.

ipGTT
An ipGTT was performed under light ether anesthesia at 0800 h in rats that were fasting overnight (1214 h) before the test. They were injected ip with 30% (wt/vol) glucose solution at a dose of 2 g glucose/kg body weight. Blood samples were withdrawn from the tail vein immediately

Pancreatic insulin content assay


After a 2-h euglycemic clamp test, rats were killed and the whole pancreas was rapidly dissected free from surrounding tissues, weighed in air, frozen in liquid N2, and stored at 70 C. Later on the same day, the frozen whole pancreases were homogenized at 20,500 rpm for 20 sec using a tissue homogenizer (Ultra-Turrax T25; IKA-Labortechnik, Staufen, Germany) in 10 times volume (10 mL buffer solution per 1 g tissue) of acid ethanol buffer [1.5 mL HCl 12 mol/L in 100 mL 70% (vol/vol) ethanol] and incubated overnight at 4 C for further extraction. On the next day, samples were centrifuged at 3000 g for 15 min at 4 C, and supernatants were diluted 1:100 and refrozen at 70 C, prepared for assay. The insulin level was measured with an insulin RIA kit for rat insulin and was normalized with pancreas and total body weight.

Measurement of periepididymal fat


FIG. 1. Blood alcohol level in control rats fed with alcohol for 4 weeks (n 6). Data are the mean SEM. After killing the rats, both epididymal fat pads were dissected, cleared of blood vessels and reproductive organs, and weighed (21).

3648 Statistical analysis

CHA ET AL.

JCE & M 2000 Vol. 85 No. 10

Data are shown as the mean sem. Statistical analyses were done with SAS statistical software (version 6.04; SAS Institute, Inc., Cary, NC). Differences between groups were analyzed by two-way ANOVA with two levels (2 2 classification) in each type of treatment. An alcohol effect was considered to be present when ANOVA showed that rats chronically fed with alcohol were significantly different from those fed with saline. A dietary protein effect was considered to be present when rats fed with a protein-deficient diet were significantly different from those fed with a protein-sufficient diet. An interactive effect was considered to be present when the combined effects of alcohol and protein deficiency were significantly different from those of either variable analyzed separately. If ANOVA indicated significant treatment effects, the significance of the differences between individual means were deter-

mined using the Tukey test. A P value less than 0.05 (two-tailed) was considered significant.

Results Dietary intake and growth curve

During the first 4 weeks of dietary treatment, the average total calorie intake of protein-deficient rats (346 16 kcal/ kg 1 day 1) was not different from that of protein-sufficient rats [328 18 kcal/kg 1 day 1, not significant (NS)]. Fourweek-old rats fed a protein-deficient diet stopped growing for 1 week and then gained weight at a considerably lower rate than rats fed a protein-sufficient diet (Fig. 2). Body weight was apparently different after 1 week of diet between proteindeficient and protein-sufficient rats (76.8 1.6 vs. 82.0 2.0 g, P 0.05). After 4 weeks of protein-deficient diet, they weighed only half compared with the control rats (Table 1). During the last 4 weeks of diet and alcohol treatment, daily calorie intake from diet only was not changed between the alcohol and saline rats both in protein-deficient (255 14 vs. 260 15 kcal.kg 1 day 1, NS) and protein-sufficient (205 11 vs. 221 10 kcal/kg 1 day 1, NS) conditions. After 3 weeks of alcohol administration in protein-sufficient rats, alcohol rats (253.6 9.3 g) showed growth retardation compared with control rats (282.5 9.0 g, P 0.05). However, the effect of alcohol did not appear in rats fed a protein-deficient diet (Fig. 2).
Pancreas and epididymal fat weight

FIG. 2. Growth curves of rats fed four different diet protocols. Rats were fed with either a protein-deficient or protein-sufficient diet from 4 weeks of age (Diet) and administered either alcohol or saline from 8 weeks of age (Alcohol). I, Protein-deficient alcohol rats; II, proteindeficient saline rats; III, protein-sufficient alcohol rats; IV, proteina sufficient saline rats. Data are the mean SEM. Dietary protein effect, P 0.01 (one-way ANOVA), I vs. III, I vs. IV, II vs. III, II vs. IV, P less than 0.05 (Tukey test); bDietary protein effect, P 0.001, alcohol effect, P 0.01, interaction, P 0.05 (two-way ANOVA), I vs. III, I vs. IV, II vs. III, II vs. IV, III vs. IV, P less than 0.05 (Tukey test).

Pancreatic weights of protein-deficient rats were significantly lower than those of protein-sufficient rats, but relative weights after adjustment for their body weights were not significantly different among the four groups (Table 1). The relative weight of epididymal fat adjusted for body weight was decreased in protein-deficient saline rats compared with control rats, but there was no difference between protein-deficient alcohol rats and control rats (Table 1).
Glucose tolerance and insulin secretory response to glucose challenge

Fasting plasma glucose and insulin concentrations in the overnight fasting rats were not significantly different among

TABLE 1. Effects of the four different diet protocols on body, pancreas, and epididymal fat weight
Group Protein-deficient alcohol rats (I) Protein-deficient saline rats (II) Protein-sufficient alcohol rats (III) Protein-sufficient saline rats (IV) Effect of Treatment (ANOVA)

Body weight (g) 4 week 8 weeka 12 weekb Pancreas weight Total (g)c Percentage of body weight (%) Epididymal fat weight Total (g)d Percentage of body weight (%)e

73.6 82.5 91.5 0.30 0.32 0.92 0.97

2.1 2.3 6.4 0.02 0.03 0.09 0.10

74.7 82.5 95.0 0.31 0.34 0.72 0.79

2.1 2.5 5.0 0.02 0.02 0.32 0.12

73.6 160.5 272.7 0.85 0.31 2.92 1.05

1.8 4.1 9.3 0.05 0.02 0.14 0.04

73.6 161.2 320.0 0.96 0.31 3.59 1.15

1.8 4.8 9.6 0.06 0.01 0.26 0.06

Protein Protein, alcohol, interaction Protein Protein Protein, interaction

Data are the mean SEM. a Dietary protein effect, P 0.001 (one-way ANOVA); I vs. III, I vs. IV, II vs. III, and II vs. IV, P 0.05 (Tukey test). b Dietary protein effect, P 0.001; alcohol effect, P 0.01; interaction, P 0.05 (two-way ANOVA); I vs. III, I vs. IV, II vs. III, II vs. IV, and III vs. IV, P 0.05 (Tukey test). c Dietary protein effect, P 0.001 (two-way ANOVA); I vs. III, I vs. IV, II vs. III, and II vs. IV, P 0.05 (Tukey test). d Dietary protein effect, P 0.001 (two-way ANOVA); I vs. III, I vs. IV, II vs. III, and II vs. IV, P 0.05 (Tukey test). e Dietary protein effect, P 0.001; interaction, P 0.05 (two-way ANOVA); II vs. IV, P 0.01 (Tukey test).

COMBINED EFFECT OF ALCOHOL AND MALNUTRITION

3649

all four groups (Fig. 3). Plasma glucose concentrations after 30 min following ip glucose load and glucose-area under the curve (AUC) in protein-deficient alcohol rats were lower than those of the other three groups (Table 2). As for the insulin response to glucose challenge, plasma insulin concentration at 30 min and 60 min after glucose load and insulin-AUC were markedly lower in protein-deficient saline rats compared with those of control rats. The insulin to glucose ratio-AUC was also reduced in protein-deficient saline rats than control rats (Table 2). The dietary protein effect was consistently observed in all parameters of ipGTT, except fasting plasma glucose and insulin concentrations.
Glucose disposal rate (GDR) during euglycemic hyperinsulinemic clamp

Under a similar hyperinsulinemic state (mean, 512 27 pmol/L) during the euglycemic ( 6 pmol/L) clamp, the GDR was decreased only in protein-deficient saline rats compared with control rats. In alcohol rats, both protein-deficient and protein-sufficient groups, they had a tendency toward a lower GDR than control rats, but without significance (Table 3).
Muscle GS activity

within 1 week of diet treatment. In protein-sufficient rats, chronic alcohol administration diminished normal weight gain after 3 weeks of alcohol treatment. But, in proteindeficient rats, additional alcohol effect on growth was not seen, probably due to absolute growth retardation. In rats, the proportion of islet cells in normal whole pancreas is less than 1% in weight, and pancreatic weight, therefore, represents that of an exocrine pancreas. In this study, the relative pancreatic weight was unchanged by protein deficiency and alcohol intake. This suggests that an inflammatory event like pancreatitis was not occurred during the experiment and we confirmed normal architecture of pancreas by light microscopy (data not shown). In rodents, the epididymal fat pad is composed mainly of white adipose tissue and is significantly correlated with the amount of total body fat, and the optimal plasma insulin concentration is a prerequisite for fat accumulation in that area (21). In this experiment, a decreased relative amount of epididymal fat in protein-deficient saline rats suggested that the plasma insulin concentration of these rats might not be

After a 2-h euglycemic hyperinsulinmic clamp, muscle GS activities at different concentrations of G6P (0 10 mmol/L) and FV of GS were not significantly different among the four groups (Table 4).
Insulin content of whole pancreas

The total insulin content of the whole pancreas of proteindeficient rats (groups I and II) was lower than that of control rats, but the relative amounts of insulin adjusted to body and pancreatic weights were not different among the four groups (Table 4).
Discussion

This study demonstrated that chronic alcohol intake protects and reverses protein deficiency-induced insulin secretory dysfunction and insulin resistance, whereas in a protein-sufficient state alcohol does not influence glucose metabolism in growing rats. Normal growth is inhibited in rats fed with a diet that contains less than 6% protein content of total calories (2, 22), and young animals given a low protein diet voluntarily reduce their food intake (23). In this experiment, total calorie intake was not changed by protein deficiency or alcohol intake. The present experimental model, therefore, produced selective protein deficiency instead of mixed protein-calorie malnutrition. In humans, moderate alcohol intake indicates a daily intake up to 2 U ( 240 kcal) (15). In this study, the amount of alcohol administered to rats consists of 10 15% of the required daily calories for rats (24), and this corresponds to moderate alcohol intake in humans. But small animals require more calories per weight compared with human (24), the amount of alcohol administered to rats in this experiment was an alcoholic dose during several hours after ingestion. Growth retardation in protein-deficient rats was apparent

FIG. 3. Plasma glucose (A) and insulin (B) concentrations after an ip glucose load in four different groups. I, Protein-deficient alcohol rats; II, protein-deficient saline rats; III, protein-sufficient alcohol rats; IV, a protein-sufficient saline rats. Data are the mean SEM. Alcohol effect, P 0.05 (two-way ANOVA), I vs. II, I vs. III, I vs. IV, P less than 0.05 (Tukey test); bDietary protein effect, P 0.01 (two-way ANOVA), II vs. IV, P less than 0.05 (Tukey test).

3650

CHA ET AL.

JCE & M 2000 Vol. 85 No. 10

TABLE 2. Plasma glucose and insulin concentrations during ip glucose tolerance tests in four different groups
Group Protein-deficient alcohol rats (I) Protein-deficient saline rats (II) Protein-sufficient alcohol rats (III) Protein-sufficient saline rats (IV) Effect of treatment (ANOVA)

Plasma glucose 0 min (mmol/L) 30 min (mmol/L)a 60 min (mmol/L)b AUC (mmol/L h)c Plasma insulin 0 min (pmol/L)d 30 min (pmol/L)e 60 min (pmol/L)f AUC (pmol/L h)g Insulin/glucose ratio (IGR) 0 minh AUCi

5.4 12.5 8.4 9.7 66 261 89 171 12.2 17.6

0.2 0.8 0.2 0.2 6 48 11 27 1.7 3.6

5.7 15.7 10.3 11.8 45 137 68 96 7.9 8.1

0.2 1.0 0.7 0.6 12 24 11 14 2.0 1.4

5.3 14.7 10.4 11.4 101 333 137 230 19.1 20.2

0.3 1.1 0.6 0.8 20 83 18 47 3.8 5.6

5.5 15.0 9.4 11.2 108 398 177 263 19.6 23.5

0.2 0.9 0.4 0.6 24 53 33 38 3.6 2.5

Alcohol Interaction Protein, interaction Protein Protein Protein Protein Protein Protein

Data are the mean SEM. IGR, Plasma insulin (pmol/L)/plasma glucose (mmol/L). a Alcohol effect, P 0.05 (two-way ANOVA); I vs. II, I vs. III, and I vs. IV, P 0.05 (Tukey test). b Interaction, P 0.05 (two-way ANOVA); comparison of means, not significant (NS) (Tukey test). c Dietary protein effect, P 0.01; interaction, P 0.05 (two-way ANOVA); I vs. II, I vs. III, and I vs. IV, P d Dietary protein effect, P 0.01 (two-way ANOVA); comparison of means, NS (Tukey test). e Dietary protein effect, P 0.01 (two-way ANOVA); II vs. IV, P 0.05 (Tukey test). f Dietary protein effect, P 0.001 (two-way ANOVA); II vs. IV, P 0.05 (Tukey test). g Dietary protein effect, P 0.01 (two-way ANOVA); II vs. IV, P 0.05 (Tukey test). h Dietary protein effect, P 0.01 (two-way ANOVA); comparison of means, NS (Tukey test). i Dietary protein effect, P 0.01 (two-way ANOVA); II vs. IV, P 0.05 (Tukey test). TABLE 3. Peripheral glucose disposal rate during euglycemic clamp tests in four different groups
Proteindeficient alcohol rats (I) Proteindeficient saline rats (II) Proteinsufficient alcohol rats (III) Proteinsufficient saline rats (IV)

0.05 (Tukey test).

Group

Effect of treatment (ANOVA)

GDR (mg/kg 1 min)a Insulin-steady state (pmol/L)

15.1 517

2.0 116

12.2 423

1.2 82

15.3 562

1.1 109

21.8 548

2.3 121

Protein, interaction

Data are the mean SEM. GDR, Glucose disposal rate. a Dietary protein effect, P 0.05; interaction, P 0.05 (two-way ANOVA); II vs. IV, P

0.05 (Tukey test).

TABLE 4. Muscle glycogen synthase activity and pancreatic insulin content in four different groups
Proteindeficient alcohol rats (I) Proteindeficient saline rats (II) Proteinsufficient alcohol rats (III) Proteinsufficient saline rats (IV) Effect of Treatment (ANOVA)

Group

Glycogen synthase activity 0 nmol/L G6P (nmol/mg 1 min 1) 10 mmol/L G6P (nmol/mg 1 min 1)a FVb Insulin content ( mol) Totalc Per 100 mg pancreas weight Per 100 g body weightd

0.69 2.83 0.24 1098 344 1136

0.15 0.56 0.03 120 20 134

0.65 3.75 0.20 969 293 1008

0.13 0.72 0.03 18 32 104

0.62 4.68 0.16 2653 309 964

0.08 0.62 0.04 114 14 42

0.61 4.68 0.15 2933 306 894

0.09 0.76 0.02 81 8 41

Protein Protein Protein Interaction

Data are the mean SEM. GS, Glycogen synthase; G6P, glucose-6-phosphate; FV, fractional velocity GS activity at 0 mmol/L G6P/GS activity at 10 mmol/L G6P. a Dietary protein effect, P 0.05 (two-way ANOVA); comparison of means, NS (Tukey test). b Dietary protein effect, P 0.05 (two-way ANOVA); comparison of means, NS (Tukey test). c Dietary protein effect, P 0.001 (two-way ANOVA); I vs. III, I vs. IV, II vs. III, and II vs. IV, P 0.05 (Tukey test). d Interaction, P 0.01 (two-way ANOVA); comparison of means, NS (Tukey test).

maintained enough to achieve normal epididymal fat accumulation compared with control rats. In contrast to this, with similar growth retardation, alcohol-fed rats could preserve their fat deposition properly. These results indicate that the plasma insulin concentrations in protein-deficient alcohol rats might be maintained at a similar level compared with control rats. After an ip glucose load, protein-deficient saline rats

showed a marked decrease in insulin response, as well as relatively low insulin concentration to glucose level compared with control rats, and this result was similar to previous experiments (22, 25). Interestingly, protein-deficient rats fed with alcohol showed a 2-fold increase in insulin response to glucose load compared with protein-deficient saline rats and secreted optimum insulin comparable with control rats. These results suggest that protein deficiency

COMBINED EFFECT OF ALCOHOL AND MALNUTRITION

3651

during the weaning period causes deterioration of insulin secretory function, but chronic alcohol intake might play a role in protecting or reversing the insulin secretory capacity resulting from protein deficiency. How protein deficiency affects insulin secretion is uncertain. In this study, relative insulin content adjusted to weight of body and pancreas was not different among four groups, and this coincided with previous experiments (26, 27). A strong correlation is noted between islet cell volume and pancreatic insulin content (28). Therefore, decreased insulin secretion in protein malnutrition may not be due to a reduction in cell number, but due to the functional impairment of the cells in ability to respond to glucose. There is no clear evidence that alcohol exerts a direct effect on pancreatic cells. A long-term administration of excessive alcohol acts as a toxin to pancreatic cells (6), but a moderate amount of alcohol intake may stimulate cells without influencing glucose metabolism (29). We previously found that a small amount of dexamethsone, in which the dose was not enough to affect glucose metabolism, administered chronically in rats with a partially dissected pancreas induces hypertrophy and hyperplasia of pancreatic islet cells, thereby improving hyperglycemia that came with dissection of the pancreas (30). Here, a small amount of steroid was responsible for steadily maintaining insulin resistance, which, in turn, induced insulin release and cell proliferation. In this study, it can be speculated that a optimum amount of alcohol per se or its metabolites (acetaldehyde, acetate, excess NADH) might play a stimulating role on enhancing stimulus-induced insulin secretion (8), like dexamethasone in the previous experiment. Pancreatic cells are known to be proliferated by functional stimuli even in the absence of hyperglycemia (30). On the other hand, we cannot rule out impaired glucose tolerance before the experiment causing the islet growth. There is no direct evidence that alcohol is metabolized in the cells. But, the optimal level of alcohol or its metabolites might prevent atrophy of pancreatic cells and induce a sufficient plasma insulin level, which, in turn, efficiently maintains normal glucose tolerance and preserves adequate masses of epididymal fat and possibly skeletal muscles in protein-deficient rats. The effect of protein deficiency on insulin sensitivity is known to be variable increased (31), decreased (25, 32), or unchanged (22), depending on experimental models. Acute alcohol administration does not alter glucose tolerance because its diabetogenic effects on glucose disappearance and oxidation (11) were counterbalanced by its potentiating action on glucose-stimulated insulin release (33) and reduction in hepatic glucose production (11). Peripheral insulin sensitivity measured by euglycemic clamp was decreased in protein-deficient rats at a similar hyperinsulinemic state compared with control rats. But, surprisingly, protein-deficient rats fed with alcohol maintained optimal glucose use to the level of control rats. One possible mechanism to explain these results is the amount of skeletal muscle (18). Unfortunately, we did not measure the net skeletal muscle mass of experimental rats, but it can be speculated that protein-deficient saline rats had less relative muscle mass due to low protein intake with insufficient concentration of plasma insulin. Whereas protein-deficient alcohol rats could preserve their

muscle mass and maintain normal insulin sensitivity both from the protein-sparing effect of alcohol (34) and sufficient plasma insulin level. In a protein-sufficient state chronic alcohol intake did not affect insulin sensitivity. Muscle GS is known to be a rate-limiting enzyme for the synthesis of glycogen and can be modulated by environmental factors (35, 36). In this study, the activity of GS at the gastrocnemius muscle was not changed by protein-deficiency and/or alcohol intake, like other experiments (33). Factors other than GS that influence glucose disposal [i.e. nonoxidative glucose uptake), enzymes involved in oxidative pathway, like pyruvate dehydrogenase (probably due to high lactate concentration)] may be modulated by chronic alcohol administration (37). In summary, protein deficiency during the growth results in deterioration of both insulin secretory capacity and sensitivity. But, these defects can be reversed to a normal state following chronic alcohol intake in growing rats. An adequate amount of alcohol or its metabolites may compensate the metabolic derangement caused by protein deficiency. In a well-nourished state, alcohol does not add additional benefit in glucose metabolism. Therefore, the metabolic impact of chronic alcohol intake may be expressed differently according to the individual nutritional status.
Acknowledgments
We thank Dr. J. K. Sung for helpful discussion, Drs. H. I. Lee and Y. S. Yoon for generous technical support, Drs. D. K. Kim and Y. S. Kim for outstanding statistical analysis, and Drs. Robert Ross and H. Y. Maeng for critical reading of the manuscript.

References
1. Rao RH. 1988 Diabetes in the undernourished: coincidence or consequence? Endocr Rev. 9:67 87. 2. Swenne I, Crace C, Milner RD. 1987 Persistent impairment of insulin secretory response to glucose in adult rats after limited period of protein-calorie malnutrition early in life. Diabetes. 36:454 458. 3. Becker DJ. 1983 The endocrine responses to protein calorie malnutrition. Ann Rev Nutr. 3:187212. 4. Friedrenberg R, Wetz R, Mako M, Surmaczyuska B. 1971 Differential plasma insulin response to glucose stimulation following ethanol priming. Diabetes. 20:397 403. 5. Singh SP, Patel DG. 1976 Effects of alcohol on carbohydrate metabolism: I. Influence on oral glucose tolerance test. Metabolism. 25:239 243. 6. Tiengo A, Valerio A, Molinari M, Meneghel A, Lapolla A. 1981 Effect of ethanol, acetaldehyde and acetate on insulin and glucagon secretion in the perfused rat pancreas. Diabetes. 30:705709. 7. Walsh CH, OSullivan DJ, Cork I. 1974 Effect of moderate alcohol intake on control of diabetes. Diabetes. 23:440 442. 8. Holley DC, Bagby GJ, Curry DL. 1981 Ethanol-insulin interrelationships in the rat studied in vitro and in vivo: evidence for direct ethanol inhibition of biphasic glucose-induced insulin secretion. Metabolism. 30:894 899. 9. Kiechl S, Willeit J, Poewe W, et al. 1996 Insulin sensitivity and regular alcohol consumption: large, prospective, cross sectional populational study. Br Med J. 313:1040 1044. 10. Facchini F, Chen YI, Reaven GM. 1994 Light-to-moderate alcohol intake is associated with enhanced insulin sensitivity. Diabetes Care. 17:1519. 11. Shelmet JJ, Reichard GA, Skutches CL, Hoeldtke RD, Owen OE, Boden G. 1988 Ethanol causes acute inhibition of carbohydrate, fat, and protein oxidation and insulin resistance. J Clin Invest. 81:11371145. 12. Holbrook TL, Barrett-Conner E, Wingard DL. 1990 A prospective populationbased study of alcohol use and non-insulin-dependent diabetes mellitus. Am J Epidemiol. 132:902909. 13. Balkau B, Randrianjohany A, Papoz L. 1988 A prospective population based study of alcohol use and non-insulin dependent diabetes mellitus. Am J Epidemiol. 128:549 558. 14. Koivisto VA, Tulkos S, Toivonen M, Haapa E, Pelkonen R. 1993 Alcohol with a meal has no adverse effects on postprandial glucose homeostasis in diabetic patients. Diabetes Care. 16:16121614. 15. Bell DS. 1996 Alcohol and the NIDDM patient. Diabetes Care. 19:509 513.

3652

CHA ET AL.

JCE & M 2000 Vol. 85 No. 10

16. Lieber CS, DeCarli LM, Sorrell MF. 1989 Experimental methods of ethanol administration. Hepatology. 10:501510. 17. Rao RH. 1992 Changes in insulin sensitivity from stress during repetitive sampling in anesthetized rats. Am J Physiol. 262:R1033R1039. 18. Kergoat M, Portha B. 1985 In vivo hepatic and peripheral insulin sensitivity in rats with non-insulin dependent diabetes induced with streptozotocin: assessment with the insulin-glucose clamp technique. Diabetes. 34:610 617. 19. Pacini G, Bergman RN. 1983 PACBERG: an adaptive program for controlling the blood sugar. Comput Methods Programs Biomed. 16:1320. 20. Thomas J, Schlender K, Larner J. 1968 A rapid filter paper assay for UDP glucose-glycogen glucosyltransferase including an improved biosynthesis of UDP-14C-glucose. Anal Biochem. 25:489 499. 21. Barzilai N, Rossetti L. 1996 Age-related changes in body composition are associated with hepatic insulin resistance in conscious rats. Am J Physiol. 270:E930 E936. 22. Okitolonda W, Brichard SM, Henquin JC. 1987 Repercussions of chronic protein-calorie malnutrition on glucose homeostasis in the rat. Diabetologia. 30:946 951. 23. Platt BS, Stewart RJ. 1967 Experimental protein-calorie deficiency: histological change in the endocrine glands of pigs. J Endocrinol. 38:121143. 24. Thorbek G, Chwalibog A, Eggum BO, Christensen K. 1982 Studies on growth, nitrogen and energy metabolism in rats. Arch Tierenahr. 32:827 840. 25. Rao RH. 1991 Chronic undernutrition accentuates insulin deficiency in rats with mild streptozocin-induced diabetes. Diabetes. 40:1404 1409. 26. Weinkove C, Weinkove EA, Pimstone BC. 1976 Glucose tolerance and insulin release in malnourished rats. Clin Sci Mol Med. 50:153163. 27. Levine LS, Wright PG, Marcus F. 1983 Failure to secrete immunoreactive insulin by rats fed a low-protein diet. Acta Endocrinol. 102:240 245.

28. McEvoy RC. 1981 Changes in the volumes of the A-, B-, and D-cell populations in the pancreatic islets during the postnatal development of the rat. Diabetes. 30:813 817. 29. McMonagle J, Felig P. 1975 Effects of ethanol ingestion on glucose tolerance and insulin secretion in normal and diabetic subjects. Metabolism. 24:625 632. 30. Lee HC, Bonner-Weir S, Weir GC, Leahy JL. 1989 Compensatory adaptation to partial pancreatectomy in the rat. Endocrinology. 124:15711575. 31. Escriva F, Kergoat M, Bailbe D, Pascual-Leone AM, Portha B. 1991 Increased insulin action in the rat after protein malnutrition early in life. Diabetologia. 34:559 564. 32. Ozanne SE, Nave BT, Wang CL, Shepherd PR, Prins J, Smith GD. 1997 Poor fetal nutrition causes long-term changes in expression of insulin signaling components in adipocytes. Am J Physiol. 273:E46 E51. 33. Trojan N, Pavan P, Iori E, et al. 1999 Effect of different times of administration of a single ethanol dose on insulin action, insulin secretion and redox state. Diabet Med. 16:400 407. 34. Berneis K, Ninnis R, Keller U. 1997 Ethanol exerts acute protein-sparing effects during postabsorptive but not during anabolic condition in man. Metabolism. 46:750 755. 35. Henry RR, Ciaraldi TP, Mudaliar S, Abrams L, Nikoulina SE. 1996 Acquired defects of glycogen synthase activity in cultured human skeletal muscle cells: influence of high glucose and insulin levels. Diabetes. 45:400 407. 36. Xa D, Heng JK, Palmer YN. 1993 The mechanism(s) of the alcohol-induced impairment in glycogen synthesis in oxidative skeletal muscles. Biochem Mol Biol Int. 30:169 176. 37. Vary TC. 1996 Sepsis-induced alterations in pyruvate dehydrogenase complex activity in rat skeletal muscle: effects on plasma lactate. Shock. 6:89 94.