- Email: [email protected]
Mechanism of carbohydrate-induced hypertriglyceridemia: Plasma lipid metabolism in mice
Mechanism
of Carbohydrate-Induced Hypertriglyceridemia: Plasma Lipid Metabolism in Mice V. S. Sheorain, M. B. Mattock, and D. Subrahmanyam
A group of mice were fed on a basal diet for 2 wk and thereafter maintained on a fat free high sucrose or starch diet for 5-6 wk. Hypertriglyceridemia was induced in the mice on feeding the high carbohydrate diets. The triglyceride levels reached maximum in 12 and 14-16 days on high sucrose and starch diets, respectively. This was associated with increased pre-&lipoproteins in plasma. Though males had higher levels of triglycerides than females on both the diets, this attained statistical significance only in the sucrose-fed group. Cholesterol and phospholipids recorded a significant increase only towards the end of the experimental period. The plasma lipid levels correlated well with their secretion rates from the liver following an injection of Triton WR 1939. Futhermore, the hypertriglyceridemia appeared to be due to defective plasma triglyceride removal as seen by the reduced postheparin lipolytic activity (PHLA) of plasma and decreased triglyceride removal (intravenous fat tolerance test) in these animals. The triglycerides returned to almost normal levels when feeding was continued on the diets. During this adaptation, PHLA and fat tolerance of animals increased significantly. Higher triglycerides in the sucrose group seem to be due to a higher rate of secretion as well as slower removal of triglycerides in this group. The sex difference noticed in the plasma triglycerides at peak hypertriglyceridemia appeared to be mainly due to sex difference in removal rates. Blood sugar raised significantly in the animals maintained on both the diets.
Initially it was thought that this kind of hyperlipidemia occurred only in certain individuals with an abnormal response to dietary carbohydrates,’ but recent studies have shown that carbohydrate-rich diets can induce hypertriglyceridemia both in norma14*5 and obese people.’ Several studies in rat@’ and micegs9have also indicated the occurrence of this type of hypertriglyceridemia, but the mechanism(s) underlying it are not clear. Thus, the present study was designed to elucidate the mechanisms underlying carbohydrate-induced hypertriglyceridemia as well as the phenomenon of adaptation using the mouse as an experimental model. Since a sex-dependence in the response to dietary carbohydrates in the induction of hypertriglyceridemia in hurnans’%12 and mice’* has been reported, this aspect has also been studied in the present investigation. The results of this study indicated that both increased secretion from the liver as well as decreased removal rate of triglycerides were the important factors in the induction of hypertriglyceridemia.
A
Sexually mature male and female Swiss mice (VRC, Poona strain) weighing an average 22 g and 18 g, respectively, were used for this study, Bovine serum albumin (Cohn fraction V) was purchased from Sigma Chemical Co., St. Louis, MO., lntralipid (I 0% fat emulsion) was obtained from Vitrum, Stockholm, Sweden and Triton WR 1339 was obtained from Winthrop Laboratories, New Castle Upon Tyne, U.K. All other reagents were of analytical grade.
CARBOHYDRATES have LTHOUGH been the staple diet of much of the world’s population for many years, their consumption was thought to be harmless until 1960. In 1961 Ahrens et al.’ described two types of hyperlipidemia in man, one of which was carbohydrateinduced. This type was considered to be associated with various cardiovascular diseases such as atherosclerosis* and ischaemic heart disease.3 From the Department of Biochemistry, Postgraduate Institute of Medical Education and Research, Chandigarh, India. Received for publication September 17. 1979. Presented in part at the XIth International Congress of Biochemistry, Jury 8-l 3. 1979, Toronto, Canada. Address reprint requests to D. Subrahmanyam. DirectorDesignate. CIBA-GIEGY Research Centre. Post Bag 9002. Bombay 400 063. India. 0 1980 by Grune & Stratton, Inc. 0026-o495/80/291~04$01.00/0
924
MATERIALS AND METHODS
Materials
Feeding of Animals These animials were given food and water ad libitum. Before switching to the experimental diets, the animals were fed on Hindustan Lever pellet diet. These animals were divided into several groups and then adapted to basal diets (containing either sucrose or starch) for 2 wk. The groups of animals adapted on sucrose containing basal diet were fed on high sucrose diet, whereas animals adapted on basal diet containing starch were fed on high starch diet for 32 and 45 days, respectively. The percent composition of high sucrose and corresponding basal diet have been described elsewhere.” High starch and its basal diet were same except that sucrose was replaced by starch. All experiments were carrried out on animals previously fasted for 16 hr.
Metabolism, Vol. 29, No. 10 (October), 1980
CARBOHYDRATE-INDUCED HYPERTRIGLYCERIDEMIA
Analytical Methods The animals were sacrificed under light ether anaesthesia and blood was collected by cardiac puncture into heparinized vials and processed according to the parameter under study. Plasma triglycerides were estimated, at different intervals during high carbohydrate feeding, to establish the hypertriglyceridemic and adaptation stages (Figs 1 and 2). Triglycerides were estimated by the method of Van Handel and Zilversmit.‘4 Blood sugar,‘* total cholesterol,‘6 and phospholipid-phosphorus” of plasma were estimated at the initial, peak triglyceridemia and adaptation stages. Serum lipoproteins were separated and analyzed by electrophoresis on cellulose acetate’s from fresh serum of individual animals in each group. Cellulose acetate strips were scanned on a densitometer and the area under each peak was taken as an index for comparison. Postherparin lipolytic activity of plasma was assayed using serum activated intralipid (10%) as substrate (final triglyceride concentration being 1.0 mg/ml of assay medium). Postheparin plasma, the source of enzyme, was isolated 30 min after an intraperitoneal injection of heparin at an optimal dose of 2.5 x IO4 lU/kg body weight. I9 The intravenous fat tolerance test was performed by injecting 0.1 ml of intralipid into the tail vein.* Thereafter, blood was drawn every 2 min for 20 min. The amount of intralipid of the samples was determined by nephelometry. The results of fat tolerance tests were given as &-values (%, min-‘) as described elswhere. r’ For each test disappearance curve was drawn which appeared to be exponential, suggesting that removal of exogenously fat emulsion followed first order kinetics (K,-phase). Secretion rates of triglycerides, cholesterol, and phospholipids were studied 8 hr after an intraperitoneal infection of Triton WR 1339 (10 mg/mouse). The percentage increase for each fraction was calculated (lipid levels after triton injection-lipid levels after saline infection x loo/lipid levels after saline injection). All the results were statistically analyzed by Student’s t test. RESULTS
Although studies were conducted on mice of both sexes, results on males only have been reported here unless there was a sex difference in the data. The dietary intake of animals during the experimental period remained fairly constant when expressed in terms of calories consumed. There was no significant difference in the weight gain of the animals fed on different diets.
-0
4
8 HIGH
12
16
SUCROSE
20
24
FEEDING,
20
32
DAYS
Plasma triglycerides of mice fed on high sucrose Fig. 1. diet at different intervals. Male is indicated by a solid line, and female by a broken line. Values are mean? SEM of at least six animals at each stage.
The triglyceride levels of the sucrose-fed group were in general higher than those of the starchfed group with a significant difference at peak hypertriglyceridemia (p < 0.05). The plasma triglycerides returned to normal (adaptation stage) in around 32 and 45 days of feeding with high sucrose and high starch diets respectively. Although males had higher levels than females in all the groups, a significant sex difference was recorded only at the peak of hypertriglyceridemia in the high sucrose group (p < 0.01).
Plasma Triglycerides Figures 1 and 2 show the triglyceride levels of animals at various stages of experimental periods. There was an initial fall in plasma triglycerides followed by a gradual increase reaching a maximum and, thereafter, the levels returned practically to normal. The peak of hypertriglyceridemia was seen after 12 days in the high sucrose group while it took between 14 and 18 days for a similar effect in the high starch group.
c
s
IC HlGH
15 STARCH
20
LS FEEDING.
30
35
40
LS
DAYS
Plasma triglycerides of mice fed on high starch Fig. 2. diet at different intervals. Male is indicated by a solid line, and female by a broken line. Values are mean f SEM of at least six animals at each atege.
926
SHEORAIN, MAlTOCK,
AND SUBRAHMANYAM
Table 1. Blood Sugar and Plasma Cholesterol and Phoapholipida of Mice, Given Either High Sucrose or High Starch Diet, at Various Stages of Triglyceridemia (Values are Expressed as Average + SEMI sucrose
Parameter (mg/lOO
ml)
Days/O
Blood sugar
85.0
Starch
12
+ 3.3 (6)
95.0
32
+ 2.3 (6)
102.5
p < 0.05 Cholesterol
85.5
+ 3.5 (6)
90.0
105.0
+ 4.0 (6)
+ 3.7 (7)
105.5
+ 2.8 (6)
84.0
16
+ 4.3 (6)
p < 0.01 100.5
NS Phosphdipids
0
98.2
zk 2.7 (71
115.0
k 2.9 (7)
p < 0.05
+ 2.6 (6)
81.5
+ 2.9 (7)
90.5
p < 0.01
NS
45
+ 2.3 (6)
102.6
+ 2.4 (7)
108.7
t 1.8 (6)
p < 0.01
k 3.8 (8) NS
p < 0.05
101.2 105.0
t 2.2 (6)
p < 0.001
+ 2.2 (8) NS
110.0
k 2.1 (6)
p < 0.05
In parentheses are the number of animals in each group. Statistical significance of the data was estimated in comparison with the initial levels in the respective
groups.
The p values
-C 0.05
were considered significant, NS, not significant.
Blood Sugar The blood sugar increased to a similar extent in the animals on both the carbohydrate-rich diets (Table 1). There was a gradual rise in plasma cholesterol and phospholipids that reached statistical significance only at the end of the experimental period in both the diets. There was no difference in the response to the two dietary carbohydrates. Serum Lipoproteins Fractionation of serum lipoproteins indicated that pre+lipoprotein (VLDL) fraction increased to 35%-40% at induction stage (p < 0.001) from an initial level of 4%-5% of total lipoproteins. It returned to normal (6%8%) at adaptation stage. There was also an increase in LDL and HDL fraction at adaptation stage compared to their initial levels. Secretion
Rates of Lipids
Secretion of various lipid fractions was studied after a single intraperitoneal injection of Table 2.
Triton WR 1339, 10 mg/mouse (Table 2). The secretion of triglycerides during high sucrose feeding increased in the animals. The secretion rate increased almost twofold at the end of the experimental period. A similar trend was noticed in the starch-fed animals also. When the two carbohydrates were compared, the triglyceride secretion rate was found to be higher in the case of sucrose than starch. Cholesterol and phospholipids also showed a gradual increase in their secretion rates with the duration of the feeding period, but it was not as marked as with triglycerides. Unlike triglycerides, these two fractions had no differences in their secretion rates when sucrose and starch-rich diets were compared. Removal
Rate of Triglycerides
The capacity of animals to remove circulating triglyceride was assessed by assaying PHLA and by the estimation of the removal rate of exogenous triglyceride emulsion. The results with PHLA indicated a significant fall in the plasma enzyme activity at the peak of hypertriglyceri-
Secretion Rates (% Increase) of Plasma Lipids on Triton WR 1339 Treatment of Mice Fed on Either High Sucrose or
High Starch Diets, at Various Stages of Triglyceridemia. Values for Various Lipids in Saline and Triton Treated Animals Have Been Expressed as mg % (Mean k SEMI SUWO~
Starch
Lipid Fraction
Treatment
Saline TG
Triton % Increase
CHOL
68.5 + 4.2 121.9
+ 3.8
78.0’
0
32
45
16
122.5
+ 6.5
69.4
+ 3.9
70.0
* 3.5
104.8
+ 5.6
72.5
_t 5.0
260.2
f 8.0
163.6
+ 7.1
120.8
k 4.0
205.9
+ 7.9
157.5
i 6.8
112.4
135.7
72.5
96.5
117.2
Saline
81.5
? 2.8
90.9
+ 3.2
105.0
-t 5.6
78.0
t 2.7
91.4
f 2.5
104.8
+ 4.4
Triton
140.6
_t 6.0
163.9
2 7.5
201.4
+ 9.0
128.0
+_ 3.9
162.9
f 3.8
204.4
k 7.5
% Increase
PL
12
Days/O
72.5
80.5
91.8
61.5
78.2
95.0
Saline
108.0
i 4.9
105.3
f 4.6
115.2
+ 5.0
102.1
k 4.0
108.6
+ 4.9
112.5
-t 7.3
Triton
189.5
f 5.2
181.4
j, 4.4
226.3
t 8.0
177.5
+ 5.2
191.7
-r 6.8
227.7
? 8.1
% Increase
75.5
72.8
96.8
74.0
76.5
102.4
*Percent increase was calculated from the mean values of two groups of animals (see Materials and Methods). Statistical analysis of percent increase could not be done because separate group of animals were used for saline and Triton treatment.
CARBOHYDRATE-INDUCED
Table 3. Poatheparin
HYPERTRIGLYCERIDEMIA
927
Phaams Lipolytic Activity of Mice Fed on High Sucrose or High Starch Diets at Various Stages of
Triglyceridemia.
Average
Activity is Expressed
as pmolea of FFA/ml/hr
sucrose Dad0
88.81
32
12
+ 1.80 (8)
50.42 0 <
2 SEM
Starch
+ 1.43 (10)
88.90
0.001
16
0
f 2.00 (8)
84.04
k 1.84 (8)
58.08
D < 0.001
45
* 2.00”
(8)
98.50
P < 0.02
_+ 2.45’
(7)
p < 0.001
In parentheses are the number of animals in each group. Statistical analysis was done as described under Table 1. *Sucrose versus starch, p < 0.05 at the corresponding intervals.
demia and this was more pronounced in the sucrose-fed animals (Table 3). On adaptation, the animals showed a marked increase in the enzyme activity over and above the initial values, There was significantly higher enzyme activity in the starch-fed group than in the sucrose-fed group at this stage. The removal rate of intravenously injected fat emulsion (Intralipid) in the two groups of animals is shown in Table 4. The removal rate (K2, % min-‘) decreased at the peak of hypertriglyceridemia whereas a significant increase was found on continued feeding of the two diets. Unlike PHLA, removal rates were different in males and females and this was particularly evident after 12 days of high sucrose feeding (p < 0.01). As seen with PHLA, the removal rate was faster in the high starch-fed group than in the sucrose-fed group at induction stage. DISCUSSION
The experiments of Mancini et a1.4 in man demonstrated that feeding a high carbohydrate, low fat diet induced a transient hypertriglyceridemia which returned towards normal levels on prolonged feeding. The results of the present study revealed a similar response in mice on feeding with high sucrose or starch diets and has further elucidated the mechanisms involved in this phenomenon. The two major factors responsible for maintenance of plasma levels of triglycerides are: (1) the secretion rate of triglycerides Table 4. Removal Rate (K2, % min-‘),
of Exogenoualy
Injected
from the liver to the circulation and (2) the removal rate of circulating triglycerides by extrahepatic tissues. The increased plasma triglyceride levels consequent to carbohydrate feeding appeared to be mainly due to an increase in the pre-/Ilipoprotein (VLDL) fraction, a carrier for endogenous triglycerides in plasma, which is typical of carbohydrate-induced hypertriglyceridemia. This fraction returned to normal levels after adaptation on these diets. Induction was associated with an increased rate of secretion and a decreased capacity of animals to remove circulating plasma triglycerides as indicated by the fall in PHLA and a decreased removal rate of intralipid. The decreased removal rate cannot be attributed to the increased levels of triglycerides because Chait et a1.23and Nicoli et a1.24have previously shown that K2 is independent of plasma triglyceride levels. Secretion rates were slightly higher in the sucrose group than in the starch group. This agrees with earlier reports in rats fed on either high fructose or high glucose diet2’ and mice given 10% fructose or glucose in drinking water for 7 days.26 But the difference in the response to the two carbohydrate diets was in the removal rates. The sex difference observed at the induction stage with sucrose also appeared to be due mainly to a slower removal rate in males than females rather than the rate at which triglycerides were secreted from the liver. Although several reports have indicated higher Fat Emulsion jlntralipid.
Sucrose or High Starch Diets at Various Stages of Triglyceridemia. sex
Days/O
12
32
Male
0.25
k 0.02 (8)
0.08
+ 0.01 (8)
0.31
Female
0.28
k 0.01 (7)
0.18
-Co.00 1 LIZ0.01 (8)”
0.34
k 0.02 (7)
10%). of Mice Fed on Either High
Results are Expressed 0
16
0.25
f 0.02 (8)
0.17
0.27
f 0.02 (8)
0.21
<0.05 f 0.02 (7) <0.05
as Mean k SEM
+ 0.01 (8I.j CO.05 k 0.01 (7)$ -Co.05
In parentheses are the number of animals in each group. Statistical analysis was done as described under Table 1. Male vevsus female ‘p < 0.0 1. Sucrose versus starch tp .z 0.001,
$p -s 0.05.
45
0.33 0.38
f 0.02 (8) to.05 + 0.03 18)
SHEORAIN, MATTOCK, AND SUBRAHMANYAM
928
secretion rates of VLDL-triglycerides from liver in female rats than their male counterparts under normal conditions.*‘-29 But unlike these findings, we did not observe any sex difference in carbohydrate-fed mice. This may be the reason why MacDonald” observed higher levels in men than in premenopausal women. The sex difference was also observed in this laboratory in humans and mice”3’2 where females had faster removal rates than males. The fall in plasma triglycerides on continued feeding of the two dietary carbohydrates appeared to be related to increased PHLA, thus increasing the capacity of animals to remove the circulating triglycerides at a faster rate. Although secretion of triglycerides at this stage was still higher than at the induction stage, this had no effect on the plasma triglyceride levels. Therefore, it was concluded that the factor responsible for adaptation to these diets in terms of plasma triglycerides was removal rate, whereas for induction it was the resultant effect of increased secretion and reduced removal. The factors responsible for regulating removal rate could be the levels of LPL activity in extrahepatic tissue (VS. Sheorain and D. Subrahmanyam, submitted for publication). At initial stages the activity of this enzyme was low, especially in adipose tissue, but this activity increased significantly towards the end (adaptation). The increase in PHLA as well as removal rate may be due to increased LPL activity in tissues. The LPL activity in adipose
tissue has been shown to be influenced by circulating insulin.” The levels of insulin in plasma following carbohydrate feeding have been reported to rise.3’ The other possible reason for increased tissue LPL activity could be due to induction of the enzyme by the circulating substrate, i.e., VLDL-triglyceride. This has been shown to be the case in acute experiments with rat heart where infusion of chylomicrons and VLDL led to increased activity of LPL in this tissue.32 If the first possibility is true, i.e., increased insulin following carbohydrate ingestion leading to increase in enzyme activity, then one would not expect a rise in blood sugar of these animals. Since there was a continuous rise in blood sugar in both the high sucrose and high starch diets, it appeared that either the increase in insulin levels was not adequate to significantly affect the increased glucose load or adipocytes had been saturated and thus became resistant to circulating insulin.33 Thus it was concluded from this study that induction of hypertriglyceridemia by high carbohydrate diets was due to both increased secretion rate of triglycerides (present data) as well as reduced uptake of circulating triglycerides (present data and earlier findings of Mancini et al.4). The adaptation of animals appeared to be mainly due to increased uptake of triglycerides by extrahepatic tissues (data from present study as well as observations of Mancini et al.4 in humans).
REFERENCES 1. Ahrens EH Jr, Hirsch J, Oette K, et al: Carbohydrateinduced and fat-induced lipaemia. Trans Assoc Am Physicians 74:134-146, 1961 2. Fredrickson DS, Levy RI, Lees RS: Fat transport in lipoproteins. An integrated approach to mechanisms and disorders. N Engl J Med 276:34-44, 1967 3. Carlson LA, Bottiger LE: Ischaemic heart disease in relation to the fating values of plasma triglycerides and cholesterol. Lancet 1:865-868, 1972 4. Mancini M, Mattock MB, Rabaya E, et al: Studies on the mechanism of carbohydrate-induced lipaemia in normal man. Atherosclerosis 1744454, 1973 5. Wu CH, Hoshi M, Shreeve WW: Human plasma triglyceride labelling after high-sucrose feeding. I. Incorporation of sucrose-I-l-“C. Metabolism 23: 1125-l 140, 1974 6. Bar-on H, Stein Y: Effect of glucose and fructose on lipid metabolism in the rat. J Nutr 94:95-105, 1968 7. Eaton RP, Kipnis DM: Effect of glucose feeding on lipoprotein synthesis in the rat. Am J Physiol 217:11531159,1969 8. Shafrir E, Teitelbaum A, Cohen AM: Hyperlipidemia
and impaired glucose tolerance in Acomys Cahirinus maintained on synthetic carbohydrate diets. Isr J Med Sci 8:99& 992,1972 9. Shafrir E, Benchimol A, Orevi M: Hyperlipidemia and hepatic hyperlipogenesis on spiny mice (Acomys Cahirinus) on carbohydrate diets: Relationship to peripheral underresponsiveness and low insulin release. Isr J Med Sci I I :738752,1915 10. MacDonald I: Influence of fructose and glucose on serum lipid levels in men and pre- and post-menopausal women. Am J Clin Nutr 18:369-372, 1966 11. Mattock MB, Kaufman BM, Spears AB, et al: Studies with intravenous fat tolerance test. Bull Postgrad Inst 8:58-62, 1974 12. Sheorain VS, Mattock MB, Subrahmanyam D: Sexdependence in triglyceride metabolism in response to dietary carbohydrates. Experientia 35:162-163, 1979 13. Mattock MB, Sheorain VS, Subrahmanyam D: Lecithin-cholesterol acyltransferase activity in carbohydrateinduced hypertriglyceridemia in mice. Experientia 34:304305,1978
CARBOHYDRATE-INDUCED
HYPERTRIGLYCERIDEMIA
14. Somogyi M: Determination of blood sugar. J Biol Chem 160:69-73,1945 15. Van Handel E, Zilversmit DB: Micromethcd for direct determination of serum triglyceride. J Lab Clin Med 50:152-157, 1957 16. Zak B: Simple rapid microtechnic for total cholesterol. Am J Clin Path01 27:583-588, 1957 17. Bartlett GR: Phosphorus assay in column chromatography. J Biol Chem 234:466-468. 1959 18. Chin HP, Blankenhorn DH: Separation and quantitative analysis of serum lipoproteins by means of electrophoresis on cellulose acetate. Clin Chim Acta 20:305-3 14, 1968 19. Sheorain VS, Mattock MB, Subrahmanyam, D: Lipolytic activity of post-heparin plasma and various tissues of mouse. Indian J Biochem Biophys 14:79-81, 1977 20. Salmon DMW, Hems DA: Plasma lipoproteins and the synthesis and turnover of plasma triglyceride in normal and genetically obese mice. Biochem J 136:55 l-563, 1973 21. Boberg J, Carlson LA, Hallberg D: Application of a new intravenous fat tolerance test in the study of hypertriglyceridemia in man, J Atherosclerosis Res 9: 159-l 69, 1969 22. Schurr PE, Schultz JR, Parkinson TM: Tritoninduced hyperlipidaemia in rats as an animal model for screening hyperlipidaemic drugs. Lipids 7:68-74, 1972 23. Chait A, Mancini M, February AW, et al: Clinical and metabolic study of alcoholic hyperlipidemia. Lancet 2~62-64, 1972 24. Nicoli A, Sigurdsson G, March A, et al: Intravenous fat tolerance: Correlation with very low density lipoprotein apoprotein B kinetics in man. Atherosclerosis 26:17-22, 1977 25. Waddell M, Fallon HJ: The effect of high carbohy-
929
drate diets on liver triglyceride formation in the rat. J Clin Invest 52:2725-2731, 1973 26. Sheorain VS, Mattock MB, Subrahmanyam D: Mechanism of hexose-induced alterations in triglyceride metabolism of mouse. Nutr Rep Int 20:33-38, 1979 27. Watkins ML, Fizette N, Heimerg M: Sexual influences on hepatic secretion of triglyceride. Biochim Biophys Acta 280:82-85.1972 28. Wilcox HG, Woodside WF, Breen KJ, et al: The effect of sex on certain properties of the very low density lipoprotein secreted by the liver. Biochem Biophys Res Commun 58:919-926, 1974 29. Soler-Argilaga C, Danon A, Wilcox HG, et al: Effects of sex on formation and properties of plasma very low density lipoproteins in vivo. Lipids 11:517-525, 1976 30. Cryer A, Riley SE, Williams ER, et al: Effect of nutritional status on rat adipose tissue, muscle and postheparin plasma clearing factor lipase activities, their relationship to triglyceride fatty uptake by fat cells and to plasma insulin concentration. Clin Sci Mol Med 50:213-221, 1976 3 1. Vrana A, Siabochova Z, Fabry P, et al: Influence of diet with high starch or sucrose content on glucose tolerance, serum insulin and insulin sensitivity in rats. Physiol Bohemoslov 23:305-3 10, 1974 32. Shafrir E, Biale Y: Effect of experimental hypertriglyceridemia on tissue and serum lipoprotein lipase activity. Eur J Clin Invest 1:19-24, 1970 33. Vrana A, Fabry P: Dietary carbohydrates and adipose tissue metabolism. Prog Biochem Pharmacol 8: 189-2 15, 1973
of Carbohydrate-Induced Hypertriglyceridemia: Plasma Lipid Metabolism in Mice V. S. Sheorain, M. B. Mattock, and D. Subrahmanyam
A group of mice were fed on a basal diet for 2 wk and thereafter maintained on a fat free high sucrose or starch diet for 5-6 wk. Hypertriglyceridemia was induced in the mice on feeding the high carbohydrate diets. The triglyceride levels reached maximum in 12 and 14-16 days on high sucrose and starch diets, respectively. This was associated with increased pre-&lipoproteins in plasma. Though males had higher levels of triglycerides than females on both the diets, this attained statistical significance only in the sucrose-fed group. Cholesterol and phospholipids recorded a significant increase only towards the end of the experimental period. The plasma lipid levels correlated well with their secretion rates from the liver following an injection of Triton WR 1939. Futhermore, the hypertriglyceridemia appeared to be due to defective plasma triglyceride removal as seen by the reduced postheparin lipolytic activity (PHLA) of plasma and decreased triglyceride removal (intravenous fat tolerance test) in these animals. The triglycerides returned to almost normal levels when feeding was continued on the diets. During this adaptation, PHLA and fat tolerance of animals increased significantly. Higher triglycerides in the sucrose group seem to be due to a higher rate of secretion as well as slower removal of triglycerides in this group. The sex difference noticed in the plasma triglycerides at peak hypertriglyceridemia appeared to be mainly due to sex difference in removal rates. Blood sugar raised significantly in the animals maintained on both the diets.
Initially it was thought that this kind of hyperlipidemia occurred only in certain individuals with an abnormal response to dietary carbohydrates,’ but recent studies have shown that carbohydrate-rich diets can induce hypertriglyceridemia both in norma14*5 and obese people.’ Several studies in rat@’ and micegs9have also indicated the occurrence of this type of hypertriglyceridemia, but the mechanism(s) underlying it are not clear. Thus, the present study was designed to elucidate the mechanisms underlying carbohydrate-induced hypertriglyceridemia as well as the phenomenon of adaptation using the mouse as an experimental model. Since a sex-dependence in the response to dietary carbohydrates in the induction of hypertriglyceridemia in hurnans’%12 and mice’* has been reported, this aspect has also been studied in the present investigation. The results of this study indicated that both increased secretion from the liver as well as decreased removal rate of triglycerides were the important factors in the induction of hypertriglyceridemia.
A
Sexually mature male and female Swiss mice (VRC, Poona strain) weighing an average 22 g and 18 g, respectively, were used for this study, Bovine serum albumin (Cohn fraction V) was purchased from Sigma Chemical Co., St. Louis, MO., lntralipid (I 0% fat emulsion) was obtained from Vitrum, Stockholm, Sweden and Triton WR 1339 was obtained from Winthrop Laboratories, New Castle Upon Tyne, U.K. All other reagents were of analytical grade.
CARBOHYDRATES have LTHOUGH been the staple diet of much of the world’s population for many years, their consumption was thought to be harmless until 1960. In 1961 Ahrens et al.’ described two types of hyperlipidemia in man, one of which was carbohydrateinduced. This type was considered to be associated with various cardiovascular diseases such as atherosclerosis* and ischaemic heart disease.3 From the Department of Biochemistry, Postgraduate Institute of Medical Education and Research, Chandigarh, India. Received for publication September 17. 1979. Presented in part at the XIth International Congress of Biochemistry, Jury 8-l 3. 1979, Toronto, Canada. Address reprint requests to D. Subrahmanyam. DirectorDesignate. CIBA-GIEGY Research Centre. Post Bag 9002. Bombay 400 063. India. 0 1980 by Grune & Stratton, Inc. 0026-o495/80/291~04$01.00/0
924
MATERIALS AND METHODS
Materials
Feeding of Animals These animials were given food and water ad libitum. Before switching to the experimental diets, the animals were fed on Hindustan Lever pellet diet. These animals were divided into several groups and then adapted to basal diets (containing either sucrose or starch) for 2 wk. The groups of animals adapted on sucrose containing basal diet were fed on high sucrose diet, whereas animals adapted on basal diet containing starch were fed on high starch diet for 32 and 45 days, respectively. The percent composition of high sucrose and corresponding basal diet have been described elsewhere.” High starch and its basal diet were same except that sucrose was replaced by starch. All experiments were carrried out on animals previously fasted for 16 hr.
Metabolism, Vol. 29, No. 10 (October), 1980
CARBOHYDRATE-INDUCED HYPERTRIGLYCERIDEMIA
Analytical Methods The animals were sacrificed under light ether anaesthesia and blood was collected by cardiac puncture into heparinized vials and processed according to the parameter under study. Plasma triglycerides were estimated, at different intervals during high carbohydrate feeding, to establish the hypertriglyceridemic and adaptation stages (Figs 1 and 2). Triglycerides were estimated by the method of Van Handel and Zilversmit.‘4 Blood sugar,‘* total cholesterol,‘6 and phospholipid-phosphorus” of plasma were estimated at the initial, peak triglyceridemia and adaptation stages. Serum lipoproteins were separated and analyzed by electrophoresis on cellulose acetate’s from fresh serum of individual animals in each group. Cellulose acetate strips were scanned on a densitometer and the area under each peak was taken as an index for comparison. Postherparin lipolytic activity of plasma was assayed using serum activated intralipid (10%) as substrate (final triglyceride concentration being 1.0 mg/ml of assay medium). Postheparin plasma, the source of enzyme, was isolated 30 min after an intraperitoneal injection of heparin at an optimal dose of 2.5 x IO4 lU/kg body weight. I9 The intravenous fat tolerance test was performed by injecting 0.1 ml of intralipid into the tail vein.* Thereafter, blood was drawn every 2 min for 20 min. The amount of intralipid of the samples was determined by nephelometry. The results of fat tolerance tests were given as &-values (%, min-‘) as described elswhere. r’ For each test disappearance curve was drawn which appeared to be exponential, suggesting that removal of exogenously fat emulsion followed first order kinetics (K,-phase). Secretion rates of triglycerides, cholesterol, and phospholipids were studied 8 hr after an intraperitoneal infection of Triton WR 1339 (10 mg/mouse). The percentage increase for each fraction was calculated (lipid levels after triton injection-lipid levels after saline infection x loo/lipid levels after saline injection). All the results were statistically analyzed by Student’s t test. RESULTS
Although studies were conducted on mice of both sexes, results on males only have been reported here unless there was a sex difference in the data. The dietary intake of animals during the experimental period remained fairly constant when expressed in terms of calories consumed. There was no significant difference in the weight gain of the animals fed on different diets.
-0
4
8 HIGH
12
16
SUCROSE
20
24
FEEDING,
20
32
DAYS
Plasma triglycerides of mice fed on high sucrose Fig. 1. diet at different intervals. Male is indicated by a solid line, and female by a broken line. Values are mean? SEM of at least six animals at each stage.
The triglyceride levels of the sucrose-fed group were in general higher than those of the starchfed group with a significant difference at peak hypertriglyceridemia (p < 0.05). The plasma triglycerides returned to normal (adaptation stage) in around 32 and 45 days of feeding with high sucrose and high starch diets respectively. Although males had higher levels than females in all the groups, a significant sex difference was recorded only at the peak of hypertriglyceridemia in the high sucrose group (p < 0.01).
Plasma Triglycerides Figures 1 and 2 show the triglyceride levels of animals at various stages of experimental periods. There was an initial fall in plasma triglycerides followed by a gradual increase reaching a maximum and, thereafter, the levels returned practically to normal. The peak of hypertriglyceridemia was seen after 12 days in the high sucrose group while it took between 14 and 18 days for a similar effect in the high starch group.
c
s
IC HlGH
15 STARCH
20
LS FEEDING.
30
35
40
LS
DAYS
Plasma triglycerides of mice fed on high starch Fig. 2. diet at different intervals. Male is indicated by a solid line, and female by a broken line. Values are mean f SEM of at least six animals at each atege.
926
SHEORAIN, MAlTOCK,
AND SUBRAHMANYAM
Table 1. Blood Sugar and Plasma Cholesterol and Phoapholipida of Mice, Given Either High Sucrose or High Starch Diet, at Various Stages of Triglyceridemia (Values are Expressed as Average + SEMI sucrose
Parameter (mg/lOO
ml)
Days/O
Blood sugar
85.0
Starch
12
+ 3.3 (6)
95.0
32
+ 2.3 (6)
102.5
p < 0.05 Cholesterol
85.5
+ 3.5 (6)
90.0
105.0
+ 4.0 (6)
+ 3.7 (7)
105.5
+ 2.8 (6)
84.0
16
+ 4.3 (6)
p < 0.01 100.5
NS Phosphdipids
0
98.2
zk 2.7 (71
115.0
k 2.9 (7)
p < 0.05
+ 2.6 (6)
81.5
+ 2.9 (7)
90.5
p < 0.01
NS
45
+ 2.3 (6)
102.6
+ 2.4 (7)
108.7
t 1.8 (6)
p < 0.01
k 3.8 (8) NS
p < 0.05
101.2 105.0
t 2.2 (6)
p < 0.001
+ 2.2 (8) NS
110.0
k 2.1 (6)
p < 0.05
In parentheses are the number of animals in each group. Statistical significance of the data was estimated in comparison with the initial levels in the respective
groups.
The p values
-C 0.05
were considered significant, NS, not significant.
Blood Sugar The blood sugar increased to a similar extent in the animals on both the carbohydrate-rich diets (Table 1). There was a gradual rise in plasma cholesterol and phospholipids that reached statistical significance only at the end of the experimental period in both the diets. There was no difference in the response to the two dietary carbohydrates. Serum Lipoproteins Fractionation of serum lipoproteins indicated that pre+lipoprotein (VLDL) fraction increased to 35%-40% at induction stage (p < 0.001) from an initial level of 4%-5% of total lipoproteins. It returned to normal (6%8%) at adaptation stage. There was also an increase in LDL and HDL fraction at adaptation stage compared to their initial levels. Secretion
Rates of Lipids
Secretion of various lipid fractions was studied after a single intraperitoneal injection of Table 2.
Triton WR 1339, 10 mg/mouse (Table 2). The secretion of triglycerides during high sucrose feeding increased in the animals. The secretion rate increased almost twofold at the end of the experimental period. A similar trend was noticed in the starch-fed animals also. When the two carbohydrates were compared, the triglyceride secretion rate was found to be higher in the case of sucrose than starch. Cholesterol and phospholipids also showed a gradual increase in their secretion rates with the duration of the feeding period, but it was not as marked as with triglycerides. Unlike triglycerides, these two fractions had no differences in their secretion rates when sucrose and starch-rich diets were compared. Removal
Rate of Triglycerides
The capacity of animals to remove circulating triglyceride was assessed by assaying PHLA and by the estimation of the removal rate of exogenous triglyceride emulsion. The results with PHLA indicated a significant fall in the plasma enzyme activity at the peak of hypertriglyceri-
Secretion Rates (% Increase) of Plasma Lipids on Triton WR 1339 Treatment of Mice Fed on Either High Sucrose or
High Starch Diets, at Various Stages of Triglyceridemia. Values for Various Lipids in Saline and Triton Treated Animals Have Been Expressed as mg % (Mean k SEMI SUWO~
Starch
Lipid Fraction
Treatment
Saline TG
Triton % Increase
CHOL
68.5 + 4.2 121.9
+ 3.8
78.0’
0
32
45
16
122.5
+ 6.5
69.4
+ 3.9
70.0
* 3.5
104.8
+ 5.6
72.5
_t 5.0
260.2
f 8.0
163.6
+ 7.1
120.8
k 4.0
205.9
+ 7.9
157.5
i 6.8
112.4
135.7
72.5
96.5
117.2
Saline
81.5
? 2.8
90.9
+ 3.2
105.0
-t 5.6
78.0
t 2.7
91.4
f 2.5
104.8
+ 4.4
Triton
140.6
_t 6.0
163.9
2 7.5
201.4
+ 9.0
128.0
+_ 3.9
162.9
f 3.8
204.4
k 7.5
% Increase
PL
12
Days/O
72.5
80.5
91.8
61.5
78.2
95.0
Saline
108.0
i 4.9
105.3
f 4.6
115.2
+ 5.0
102.1
k 4.0
108.6
+ 4.9
112.5
-t 7.3
Triton
189.5
f 5.2
181.4
j, 4.4
226.3
t 8.0
177.5
+ 5.2
191.7
-r 6.8
227.7
? 8.1
% Increase
75.5
72.8
96.8
74.0
76.5
102.4
*Percent increase was calculated from the mean values of two groups of animals (see Materials and Methods). Statistical analysis of percent increase could not be done because separate group of animals were used for saline and Triton treatment.
CARBOHYDRATE-INDUCED
Table 3. Poatheparin
HYPERTRIGLYCERIDEMIA
927
Phaams Lipolytic Activity of Mice Fed on High Sucrose or High Starch Diets at Various Stages of
Triglyceridemia.
Average
Activity is Expressed
as pmolea of FFA/ml/hr
sucrose Dad0
88.81
32
12
+ 1.80 (8)
50.42 0 <
2 SEM
Starch
+ 1.43 (10)
88.90
0.001
16
0
f 2.00 (8)
84.04
k 1.84 (8)
58.08
D < 0.001
45
* 2.00”
(8)
98.50
P < 0.02
_+ 2.45’
(7)
p < 0.001
In parentheses are the number of animals in each group. Statistical analysis was done as described under Table 1. *Sucrose versus starch, p < 0.05 at the corresponding intervals.
demia and this was more pronounced in the sucrose-fed animals (Table 3). On adaptation, the animals showed a marked increase in the enzyme activity over and above the initial values, There was significantly higher enzyme activity in the starch-fed group than in the sucrose-fed group at this stage. The removal rate of intravenously injected fat emulsion (Intralipid) in the two groups of animals is shown in Table 4. The removal rate (K2, % min-‘) decreased at the peak of hypertriglyceridemia whereas a significant increase was found on continued feeding of the two diets. Unlike PHLA, removal rates were different in males and females and this was particularly evident after 12 days of high sucrose feeding (p < 0.01). As seen with PHLA, the removal rate was faster in the high starch-fed group than in the sucrose-fed group at induction stage. DISCUSSION
The experiments of Mancini et a1.4 in man demonstrated that feeding a high carbohydrate, low fat diet induced a transient hypertriglyceridemia which returned towards normal levels on prolonged feeding. The results of the present study revealed a similar response in mice on feeding with high sucrose or starch diets and has further elucidated the mechanisms involved in this phenomenon. The two major factors responsible for maintenance of plasma levels of triglycerides are: (1) the secretion rate of triglycerides Table 4. Removal Rate (K2, % min-‘),
of Exogenoualy
Injected
from the liver to the circulation and (2) the removal rate of circulating triglycerides by extrahepatic tissues. The increased plasma triglyceride levels consequent to carbohydrate feeding appeared to be mainly due to an increase in the pre-/Ilipoprotein (VLDL) fraction, a carrier for endogenous triglycerides in plasma, which is typical of carbohydrate-induced hypertriglyceridemia. This fraction returned to normal levels after adaptation on these diets. Induction was associated with an increased rate of secretion and a decreased capacity of animals to remove circulating plasma triglycerides as indicated by the fall in PHLA and a decreased removal rate of intralipid. The decreased removal rate cannot be attributed to the increased levels of triglycerides because Chait et a1.23and Nicoli et a1.24have previously shown that K2 is independent of plasma triglyceride levels. Secretion rates were slightly higher in the sucrose group than in the starch group. This agrees with earlier reports in rats fed on either high fructose or high glucose diet2’ and mice given 10% fructose or glucose in drinking water for 7 days.26 But the difference in the response to the two carbohydrate diets was in the removal rates. The sex difference observed at the induction stage with sucrose also appeared to be due mainly to a slower removal rate in males than females rather than the rate at which triglycerides were secreted from the liver. Although several reports have indicated higher Fat Emulsion jlntralipid.
Sucrose or High Starch Diets at Various Stages of Triglyceridemia. sex
Days/O
12
32
Male
0.25
k 0.02 (8)
0.08
+ 0.01 (8)
0.31
Female
0.28
k 0.01 (7)
0.18
-Co.00 1 LIZ0.01 (8)”
0.34
k 0.02 (7)
10%). of Mice Fed on Either High
Results are Expressed 0
16
0.25
f 0.02 (8)
0.17
0.27
f 0.02 (8)
0.21
<0.05 f 0.02 (7) <0.05
as Mean k SEM
+ 0.01 (8I.j CO.05 k 0.01 (7)$ -Co.05
In parentheses are the number of animals in each group. Statistical analysis was done as described under Table 1. Male vevsus female ‘p < 0.0 1. Sucrose versus starch tp .z 0.001,
$p -s 0.05.
45
0.33 0.38
f 0.02 (8) to.05 + 0.03 18)
SHEORAIN, MATTOCK, AND SUBRAHMANYAM
928
secretion rates of VLDL-triglycerides from liver in female rats than their male counterparts under normal conditions.*‘-29 But unlike these findings, we did not observe any sex difference in carbohydrate-fed mice. This may be the reason why MacDonald” observed higher levels in men than in premenopausal women. The sex difference was also observed in this laboratory in humans and mice”3’2 where females had faster removal rates than males. The fall in plasma triglycerides on continued feeding of the two dietary carbohydrates appeared to be related to increased PHLA, thus increasing the capacity of animals to remove the circulating triglycerides at a faster rate. Although secretion of triglycerides at this stage was still higher than at the induction stage, this had no effect on the plasma triglyceride levels. Therefore, it was concluded that the factor responsible for adaptation to these diets in terms of plasma triglycerides was removal rate, whereas for induction it was the resultant effect of increased secretion and reduced removal. The factors responsible for regulating removal rate could be the levels of LPL activity in extrahepatic tissue (VS. Sheorain and D. Subrahmanyam, submitted for publication). At initial stages the activity of this enzyme was low, especially in adipose tissue, but this activity increased significantly towards the end (adaptation). The increase in PHLA as well as removal rate may be due to increased LPL activity in tissues. The LPL activity in adipose
tissue has been shown to be influenced by circulating insulin.” The levels of insulin in plasma following carbohydrate feeding have been reported to rise.3’ The other possible reason for increased tissue LPL activity could be due to induction of the enzyme by the circulating substrate, i.e., VLDL-triglyceride. This has been shown to be the case in acute experiments with rat heart where infusion of chylomicrons and VLDL led to increased activity of LPL in this tissue.32 If the first possibility is true, i.e., increased insulin following carbohydrate ingestion leading to increase in enzyme activity, then one would not expect a rise in blood sugar of these animals. Since there was a continuous rise in blood sugar in both the high sucrose and high starch diets, it appeared that either the increase in insulin levels was not adequate to significantly affect the increased glucose load or adipocytes had been saturated and thus became resistant to circulating insulin.33 Thus it was concluded from this study that induction of hypertriglyceridemia by high carbohydrate diets was due to both increased secretion rate of triglycerides (present data) as well as reduced uptake of circulating triglycerides (present data and earlier findings of Mancini et al.4). The adaptation of animals appeared to be mainly due to increased uptake of triglycerides by extrahepatic tissues (data from present study as well as observations of Mancini et al.4 in humans).
REFERENCES 1. Ahrens EH Jr, Hirsch J, Oette K, et al: Carbohydrateinduced and fat-induced lipaemia. Trans Assoc Am Physicians 74:134-146, 1961 2. Fredrickson DS, Levy RI, Lees RS: Fat transport in lipoproteins. An integrated approach to mechanisms and disorders. N Engl J Med 276:34-44, 1967 3. Carlson LA, Bottiger LE: Ischaemic heart disease in relation to the fating values of plasma triglycerides and cholesterol. Lancet 1:865-868, 1972 4. Mancini M, Mattock MB, Rabaya E, et al: Studies on the mechanism of carbohydrate-induced lipaemia in normal man. Atherosclerosis 1744454, 1973 5. Wu CH, Hoshi M, Shreeve WW: Human plasma triglyceride labelling after high-sucrose feeding. I. Incorporation of sucrose-I-l-“C. Metabolism 23: 1125-l 140, 1974 6. Bar-on H, Stein Y: Effect of glucose and fructose on lipid metabolism in the rat. J Nutr 94:95-105, 1968 7. Eaton RP, Kipnis DM: Effect of glucose feeding on lipoprotein synthesis in the rat. Am J Physiol 217:11531159,1969 8. Shafrir E, Teitelbaum A, Cohen AM: Hyperlipidemia
and impaired glucose tolerance in Acomys Cahirinus maintained on synthetic carbohydrate diets. Isr J Med Sci 8:99& 992,1972 9. Shafrir E, Benchimol A, Orevi M: Hyperlipidemia and hepatic hyperlipogenesis on spiny mice (Acomys Cahirinus) on carbohydrate diets: Relationship to peripheral underresponsiveness and low insulin release. Isr J Med Sci I I :738752,1915 10. MacDonald I: Influence of fructose and glucose on serum lipid levels in men and pre- and post-menopausal women. Am J Clin Nutr 18:369-372, 1966 11. Mattock MB, Kaufman BM, Spears AB, et al: Studies with intravenous fat tolerance test. Bull Postgrad Inst 8:58-62, 1974 12. Sheorain VS, Mattock MB, Subrahmanyam D: Sexdependence in triglyceride metabolism in response to dietary carbohydrates. Experientia 35:162-163, 1979 13. Mattock MB, Sheorain VS, Subrahmanyam D: Lecithin-cholesterol acyltransferase activity in carbohydrateinduced hypertriglyceridemia in mice. Experientia 34:304305,1978
CARBOHYDRATE-INDUCED
HYPERTRIGLYCERIDEMIA
14. Somogyi M: Determination of blood sugar. J Biol Chem 160:69-73,1945 15. Van Handel E, Zilversmit DB: Micromethcd for direct determination of serum triglyceride. J Lab Clin Med 50:152-157, 1957 16. Zak B: Simple rapid microtechnic for total cholesterol. Am J Clin Path01 27:583-588, 1957 17. Bartlett GR: Phosphorus assay in column chromatography. J Biol Chem 234:466-468. 1959 18. Chin HP, Blankenhorn DH: Separation and quantitative analysis of serum lipoproteins by means of electrophoresis on cellulose acetate. Clin Chim Acta 20:305-3 14, 1968 19. Sheorain VS, Mattock MB, Subrahmanyam, D: Lipolytic activity of post-heparin plasma and various tissues of mouse. Indian J Biochem Biophys 14:79-81, 1977 20. Salmon DMW, Hems DA: Plasma lipoproteins and the synthesis and turnover of plasma triglyceride in normal and genetically obese mice. Biochem J 136:55 l-563, 1973 21. Boberg J, Carlson LA, Hallberg D: Application of a new intravenous fat tolerance test in the study of hypertriglyceridemia in man, J Atherosclerosis Res 9: 159-l 69, 1969 22. Schurr PE, Schultz JR, Parkinson TM: Tritoninduced hyperlipidaemia in rats as an animal model for screening hyperlipidaemic drugs. Lipids 7:68-74, 1972 23. Chait A, Mancini M, February AW, et al: Clinical and metabolic study of alcoholic hyperlipidemia. Lancet 2~62-64, 1972 24. Nicoli A, Sigurdsson G, March A, et al: Intravenous fat tolerance: Correlation with very low density lipoprotein apoprotein B kinetics in man. Atherosclerosis 26:17-22, 1977 25. Waddell M, Fallon HJ: The effect of high carbohy-
929
drate diets on liver triglyceride formation in the rat. J Clin Invest 52:2725-2731, 1973 26. Sheorain VS, Mattock MB, Subrahmanyam D: Mechanism of hexose-induced alterations in triglyceride metabolism of mouse. Nutr Rep Int 20:33-38, 1979 27. Watkins ML, Fizette N, Heimerg M: Sexual influences on hepatic secretion of triglyceride. Biochim Biophys Acta 280:82-85.1972 28. Wilcox HG, Woodside WF, Breen KJ, et al: The effect of sex on certain properties of the very low density lipoprotein secreted by the liver. Biochem Biophys Res Commun 58:919-926, 1974 29. Soler-Argilaga C, Danon A, Wilcox HG, et al: Effects of sex on formation and properties of plasma very low density lipoproteins in vivo. Lipids 11:517-525, 1976 30. Cryer A, Riley SE, Williams ER, et al: Effect of nutritional status on rat adipose tissue, muscle and postheparin plasma clearing factor lipase activities, their relationship to triglyceride fatty uptake by fat cells and to plasma insulin concentration. Clin Sci Mol Med 50:213-221, 1976 3 1. Vrana A, Siabochova Z, Fabry P, et al: Influence of diet with high starch or sucrose content on glucose tolerance, serum insulin and insulin sensitivity in rats. Physiol Bohemoslov 23:305-3 10, 1974 32. Shafrir E, Biale Y: Effect of experimental hypertriglyceridemia on tissue and serum lipoprotein lipase activity. Eur J Clin Invest 1:19-24, 1970 33. Vrana A, Fabry P: Dietary carbohydrates and adipose tissue metabolism. Prog Biochem Pharmacol 8: 189-2 15, 1973