Relative rates of gastric emptying in rats fed nonliquid meals DAVID
L. TROUT,
JOHN
D. PUTNEY,
AND EMILY
of glucose
vs. fat
S. CONWAY
Nutrition Institute, Agricultural Research Service, United States Department of Agriculture, Beltsville, Maryland 20705
TROUT, DAVID L., JOHN D. PUTNEY, AND EMILY S. CONWAY. Relative rates of gastric emptying of glucose vs. fat in rats fed nonliquid meals. Am. J. Physiol. 234(6): E660-E666, 1978 or Endocrinol. Metab. Gastrointest. Physiol. Am. J. Physiol.: 3(6k E660-E666, 1978. - Sprague-Dawley rats were briefly starved, fed various test meals, and killed at measured intervals, and the average fractional emptying (disappearance from stomach) rates for glucose (K,,) and for fat (K,,,) were determined. The Kgl /&, ratio was calculated as a measure of the degree to which the stomach emptied glucose preferentially to fat. The size of the meal affected this ratio, which was 7.6 for a small (0.5 g) meal and 2.4 for a large (2.0 g) meal of a nutritionally complete diet. When test meals contained one of two levels of fat (0.4 and 0.1 g) and of glucose (1.2 and 0.3 g). the high level of fat depressed &, and &,I&, , whereas the high level of glucose depressed K,, and particularly Kfa, and, therefore, raised &, &, . Ku,/Kf,t was also affected by strain of rat and was reduced almost to 1 .O by mixing the meal into a viscous gel of xanthan gum. In the absence of this gel, the percentage of water existing in stomach contents shortly after the test meals varied between 53 and 79% and was suspected of influencing KRI &, .
25). When the stomach receives discrete sugar-containing meals and empties the carbohydrate preferentially to fat, it must introduce into the duodenum and, ultimately, into the metabolic system, a relatively highcarbohydrate nutrient mixture after the meal and a relatively high-fat mixture before the next meal. It has been suggested that metabolic responses to eating mono- and disaccharides versus complex carbohydrates are importantly influenced by the degree to which the relative rates of intestinal absorption of carbohydrate versus fat fluctuate between successive meals (28). In the present report, we have studied the disappearance of glucose and fat from the stomach of rats that were starved and then fed dry or relatively dry nutrient mixtures in order to identify factors that influence the degree to which the stomach empties simple sugar preferentially to fat. Factors evaluated include the strain of rat, the addition of a gel to the test meal, size of the meal, and the relative and absolute amounts of glucose and fat ingested.
test
METHODS
meals;
rates; Wistar
food deprivation; average rats; Sprague-Dawley rats
fractional
emptying
WHEN A SIMPLE SUGAR is fed with fat, the stomach empties the sugar at a greater fractional rate than the fat (21,28). Reasohs for a preferential emptying of sugar versus fat relate to the anatomy of the stomach and to a partial separation of the sugarand fat in that organ. During a meal, water may be ingested, and salivary and gastric secretions are added to the food. The water present dissolves part or all of the simple sugar but none of the fat. The fat may separate and float or may associate itself with protein and tend to sink. However, to leave the stomach, the fat must reach the level of the gastroduodenal junction, which, for a standing or sitting person or for a horizontally oriented rat, is higher than the low point of the stomach, yet lower than the high point. Thus, when either floating or sinking, fatcontaining particles tend to be retained in the stomach, allowing the simple sugar to be preferentially emptied. Ingested fat and even carbohydrate are normally delayed much longer in the stomach than in the small intestine prior to their digestion and absorption (5). The rates of absorption of carbohydrate and fat are controlled mainly at the stage of gastric emptying (19, 20,
Animals were principally male Sprague-Dawley rats weighing 200-275 g but included 19 male Wistar rats of similar body size from Hilltop Laboratory Animals, Inc., Scottdale, Pa. 1 All were maintained in our laboratory for at least 5 days and then starved for 18-22 h or for 2 days. Food deprivation for 18 h was found to reduce stomach contents of glucose and fat. usually to less than 1 and 5 mg, respectively; residual contents were ignored in calculating nutrient disappearance from the stomach after test meals. After being starved, rats were offered test meals of known size and composition, observed until they began to eat, and were allowed up to 15 min to finish. The amount of food actually consumed was determined by weighing the food cups before and after the meal and correcting for diet spillage. Five minutes before animals were scheduled to be killed, they were injected with 4050 mg amobarbital. When anesthesia developed, stomachs were promptly exposed, tied off at the intestinal and esophageal ends, and removed intact. The time of l Mention of a trademark or proprietary product does not constitute a guarantee or warranty of the product by the U.S. Department of Agriculture, and does not imply its approval to the exclusion of other products that may also be suitable.
E660
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 16, 2019.
GASTRIC
EMPTYING
OF
GLUCOSE
VS.
FAT
IN RATS
sampling was taken to be the interval between the midpoint of the eating period and the tying off of the stomachs. Animals whose stomachs were sampled after 100 or 120 min were provided water ad libitum, and their stomachs were analyzed for glucose and fat by standard methods (10, 12). Details of diet analysis and animal treatment and of procedures for recovering and sampling gastric contents were as previously described (28)
The fraction of the ingested glucose or fat recovered in the stomachs was calculated as Fgl and F,,,; time of sampling after the meal was measured in hours as t. The rate constants for the disappearance of glucose and fat from the stomach were also calculated, respectively, as (-In Fgl)/t and (-In F&/t and were called the average fractional emptying rates for glucose and fat. The KS&&, ratio, which was also called the gastric emptying ratio (28), was also calculated and was considered to be a measure of the degree to which the stomach emptied glucose preferentially to fat. Kgl, K&, and the Kgl/Kfat ratio were calculated for individual rats, and statistics were run on these normalized data. The study consisted of five experiments in which KB1, Kfat , and the Kgl/Kfat ratio were determined after rats were fed test meals. Table 1 indicates the size and composition of the test meals, the time a&r the meal when gastric contents were measured, and also the nature of the previous feeding and the period of starvation during which stomachs were emptied prior to the test meal. At least one of the test meals used in four experiments was composed of a basal diet containing, by weight, 3 parts glucose, 1 part partially hydrogenTABLE
1. Protocol
ated vegetable oil (Crisco) and 1 part of “essential nutrients,” a mixture of protein, vitamins, and minerals (28) Tie effects of varying meal size, of incorporating (or not incorporating) the test meal in a viscous, nonnutritive gel, and of feeding 1the same test meal to rats of two strains (Wistar vs. Sprague-Dawley) were determined, respectively, in experiments 1,2, and 3. Expriments 4 and 5 entailed feeding all combinations of the highest and lowest levels of glucose (1.2 and 0.3 g) and of fat (0.4 and 0.1 g partially hydrogenated vegetable oil) used in experiment 1. Early data (unpublished observations) indicated a need to promote the consumption of the test meals of experiments 4 and 5. Therefore, we familiarized these animals with the nutrient mixture of their final test meal during a 4-5 day pretest period, while feeding a second diet that insured a relatively balanced total intake. As Table 1 indicates, animals of experiment 4 ate similar daily quantities of all essential nutrients during their pretest period, but the total diets for groups B and C were high, respectively, in fat and glucose. In experiment 5, the two diets given to each rat during this time were so adjusted that animals of all groups ate similar quantities of all nutrients. Twenty animals in exprimertts 3 and 28 in experiment 4 were used to investigate a factor suspected of influencing Kg1lKfat importantly, i.e., the early addition of water, through salivary and gastric secretions, to stomach contents. Having been prefed and starved like others in these experiments, animals were denied water during and after presentation of their respective test meals, and stomachs were removed under amobarbital
of experiments Test Meal: Weight of Components,
Erpt No. and Group
Glucose
Fat”
1A 1B IC
0.3 0.6 1.2
0.1 0.2 0.4
0.1
2A 2B
0.6 0.6
0.2 0.2
0.2 0.2
3A’ 3B
L2 1.2
0.4 0.4
4A 4B 4c 40
0.3 0.3 1.2 1.2
5A 5B 5c 5D
0.3 0.3 1.2 1.2
g Total offered
Total eaten
Time of Sampling Stomachs,d min
0.5 1.0 2.0
0.50 0.98 1.96
100 100
18-22 18-22 18-22
Diet of subsequent test meal (basal) ad libiturn
1.0 0
2.0 1.0
1.99 0.99
100 100
18-22 18-22
Biscuit diet ad libitum”
0.4 0.4
0 0
2.0 2.0
1.85 1.86
120 120
ca 48 ca 48
Biscuit diet ad libitum
0.1
0.1
0.5 0.8 1.4 1.7
loo
0.1 0.1 0.1
0 0 0 0
0.49
0.4
0.76 1.25 1.57
100 100 100
18-22 18-22 18-22 18-22
Each rat except for group A was fed 2 diets in separate dishes”
0.4 0.7 1.3 1.6
0.40 0.68 1.15 1.39
100
18-22 18-22 18-22 18-22
Each rat was fed 2 diets in separate dishesh
0.1 0.4 0.1 0.4 0.1 0.4
Othersb
0.2 0.4
100
100 100 loo
F+io&~Fni
tth*
Previous
Diet for 4-5 Days
a Partially hydrogenated vegetable oil (Crisco). b Others is a mixture of essential nutrients containing, per 100 g: 75 g lactalbumin, 5 g vitamin mixture, and 20 g minerals. See ref. 28 for details. c Xanthan gum (4 g/NO g) of a viscous aqueous suspension. * Time of sampling stomach is interval between midpoint of the eating period and the tying off of the stomach under anesthesia, prior to its removal. (’ D and G Bat and Mouse Diet, Price-Wilhoite Co., Frederick, Md. f Variable is strain of rat, Sprague-Dawley (A ) vs. Wistar (B). R All rats received their respective test meal diet ad libitum during their pretest period. Those of groups B, C, and D also received restricted amounts of their test meal diet supplemented with extra essential nutrients. Each rat received a total of 2.4 g la&albumin and 3.2 g total essential nutrients daily if he consumed all the restricted diet and sufficient of the ad libitum mixture to obtain a total of 76 kcal metabolizable energy per day. h All rats were ad libitum fed the glucose-fat mixture of their subsequent test meal and also fed, in restricted amounts, a second complementary diet. Animals consumed 9.6 g glucose, 3.2 g fat, and 3.2 g essential nutrients if they ate all of the diet offered and an appropriate amount of the glucose-fat mixture. Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 16, 2019.
TROUT,
E662 anesthesia 20 min a&r the midpoint of the meal. Water was determined in stomach contents, which were weighed before and after drying overnight in a 100°C air oven. The Student t test and analysis of variance (27) and Duncan’s multiple range test (13) were used to evaluate data (P > 0.05 considered not significant).
PUTNEY,
GLUCOSE I.(
AND
CONWAY
I
-
.E RESULTS
In general, about 90% of the Sprague-Dawley rats began eating their test meals within 5 min, whereas about 20% of the Wistar rats used in a recent study (28) delayed for at least 30 min. This behavioral difference afkcted greatly the amount of time and effort required to observe when each of a group of rats begins eating. Table 1 indicates how much of the meals offered was actually eaten by our Sprague-Dawley rats. In all experiments, at least 80% of the Sprague-Dawley rats consumed 90+% of their test meals within the 15 min allowed. Data plotted in Figs. l-5 are gastric emptying rates. However, one can calculate Fgl and Ffat values (i.e., the fraction of the ingested glucose or fat recovered in the stomach t hours after the midpoint of the ‘test meal) from the following relationships: I$ = antilog, ( -K, x t) and Ffa, = antilog, ( -KPat x t). Figure 1 indicates the effect of meal size on gastric
l-l 1GLUCOSE
1 I I
IL ).5
SIZE
1 2.0 OF
MEAL
c : ::
(grams)
Effect of meal size on fractional emptying (disappearance from stomach) rates for glucose and fat in rats, i.e., Kg, and Krat. These rates equal (-In F)/t, where F is fraction of ingested glucose or fkt recovered in stomach, and t is time (in hours) of sampling stomach contents after midpoint of test meal. Test meals were of basal diet, and time of sampling stomach contents was 1 */a h. Bars and vertical lines represent mean 2 1 SE for 7 rats. Mean t SE of Kcl &a* ratio were 7.63 t 1.85, 2.72 2 0.33, and 2.41 t 0.34, respectivelv, after 0.5, 1.0, and 2.0 g meals. FIG.
1.
Y
.4
: .2
DIET
t GEL
DIET
TEST
ALONE
MEAL
FIG. 2. Effect of adding gel to basal diet on fractional emptying rates for glucose and fat. Test meals contained 1 g basal diet, which was fed alone or as mixed with 1 g gel containing 40 mg xanthan gum, and stomachs were sampled 100 min after midpoint of meals. Means + SE ofK,,/K,, were 1.09 + 0.02 and 3.23 + 0.10, respectively, for animals that received gel or no gel.
I
s c r’I
.E
emptying rates in experiment 1. For animals that received the 0.5-g meal, the I& of 1.73 observed 1.67 h after the midpoint of the test meal indicates that Flg1 equaled antilog, (-1.73 x 1.67), i.e., 0.056 (or 5.6%). The K& of 0.377 for this group indicated a Frat of 53.3%. For the group that received the 2.0-g meal, the geometric means of Fgl and Ffat were 23.1 and 53.3%, respectively. Glucose emptied from the stomachs preferentially to fat asker these and, to varying degrees, all other meals. The higher Kg1/Kfat ratio of animals fed the 0.5-g versus the 2.0-g meal indicates that the preferential emptying of glucose versus fat was particularly prominent after the small meal. Figure 2 summarizes results of a second experiment that was designed to determine how effectively the addition of a viscous gel to the basal diet tended to equalize Kg1 and Kfat . The gel, a 4% suspension (wt/wt) of xanthan gum in water, was within the concentration range that was used to bind cornstarch and relatively hydrophobic proteins into doughs suitable for breadmaking (6). The addition of gel decreased KB1, tended to raise Kfat (0.05 c P c O.lO), and decreased Kfat to 1.09. Because xanthan gum is chemically inert and nontoxic (3, 30) and should not greatly inhibit absorption, the low Kgl/Kfat ratio observed with the gel indicates that the high ratios observed in its absence were not due to glucose absorption in the stomach.
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 16, 2019.
GASTRIC
EMPTYING
OF GLUCOSE
VS. FAT IN RATS
3 was prompted by our observation that Sprague-Dawley rats were giving slightly higher gastric emptying ratios than had been observed with 2-day starved Wistar rats (28). Gastric emptying was observed in four lots of Sprague-Dawley rats and in two lots of Wistar rats after they had been starved for 2 days and fed 2 g basal diet. As Fig. 3 indicates, under these conditions, stomachs of SpragueDawley rats emptied glucose faster than those of the Wistar rats. The KgI/Krat ratios were appreciably higher for the Sprague-Dawley than for the Wistar rats and also higher than had been observed in Sprague-Dawley rats that were accustomed’ to the nutrient mixture of the test meal, then starved for only 18-22 h (see Fig. 3). In experiment 4, two levels of glucose and of fat were provided in four test meals, which contained equal amounts of essential nutrients. As Fig. 4 indicates, the high level of glucose lowered Kg1 and particularly Kfal and, therefore, raised the K,,lK,,, ratio. The high level of fat elevated Kfat and depressed Kg1 and the Kgl/Kfat ratio. A notably high Kgl/Kfat ratio of 17.1 was observed in animals receiving the high-glucose, low-fat meal. In experiment 5, test meals contained only glucose and fat, in the amounts fed in experiment 4, and results are summarized in Fig. 5. As in experiment 4, the high level of glucose lowered Kg1 and particularly Kfat and thus elevated the Kgl/Kfat ratio. The high level of fat lowered KBI and the Kg1/Kfat ratio. The Kg1/Kfat ratio was
!
Experiment
18- to 22-h starved
3.0 I 2.50
GLUCOSE
I
2.0 2.25 ! 1.75'y 150
I 1.0 .75 1.25L 50' r
z: ncc T
.25
SMALL
SPRAGUE
-DAWLEY STRAIN
HIGHMEALS
WITH
D
HIGH-GLUCOSE
ESSENTIAL
f
I
C
FAT
‘Im :
T I :
s
LARGE
NUTRIENTS
FIG. 4. Effect of amounts of glucose and fat fed on their fractional emptying rates from rat stomach. Amounts of glucose and fat fed were, respectively, 0.3 and 0.1 g for A; 0.3 and 0.4 g for B; 1.2 and 0.1 g for C; 1.2 and 0.4 g for D; and a 0.1 g mixture of protein, vitamins, and minerals was included. Stomachs were removed 100 min after midpoint of meal. Mean + SE for Kg1/I& ratios were 6.9 2 1.27 for A; 2.4 + .29 for B; 17.1 * 2.13 for C; and 2.6 st .37 for D.
W ISTAR
OF RAT
3. Fractional emptying rates for glucose and fat in 2 strains of rats. Two-day-starved animals were provided 2-g basal diet and stomachs were sampled 2 h after midpoint of test meal. Mean t SE for & /I& were 4.3 t 0.50 for Sprague-Dawley and 1.9 2 0.11 for Wistar rats. FIG.
: I3
A
TEST
1 J GLUCOSE
: : z z
nearly 30 for animals fed the high-glucose, low-fat meal. In experiments 3 and 4, the water in stomach contents was determined 20 min after the middle of the eating period in those animals deprived of drinking watey when given their test meal. Data are presented as percentage of water in stomach contents in Table 2. The percentage of water in gastric contents was greater in Sprague-Dawley than Wistar rats when animals of both strains were fed the same test meal and was greater in Wistar rats a&r meals containing 0.1 g than after meals containing 0.4 g fat. Treatment means for percentage of gastric water varied between 53 and 79%, implying ratios of 1.1-3.8 g water/g dry matter. It appears that the various test meals elicited markedly different amounts of gastric and salivary secretions per gram consumed. Table 2 also records the Kgl/Kfat ratios for animals of experiments 3 and 4, as indicated in Figs. 3 and 4. No simple relationship between KB1/Kfat and percentage of water in stomach contents was-observed; but, with one exception among six groups, increasing Kgl/Kfat values were associated with increasing percentages of gastric water. Data are consistent with the notion that the relative proportion of water in stomach contents shortly
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 16, 2019.
TROUT,
q
GLUCOSE
cl
FAT
1
PUTNEY,
AND CONWAY
was extracted by water and mixed. The stomachs were cut near the pyloric sphincter; several drops of fluid were expelled, filtered through cotton, and analyzed for glucose. Water was also determined in the remaining stomach contents. In two animals fed the high-glucose meal, no crystals of glucose were observed in the stomachs, and the concentration of glucose in the initially expelled fluid approximated the expected value if the glucose known to be present in the stomachs was all dissolved and well mixed. In two animals fed the highfat meal, however, glucose crystals were observed under low-power magnification; and the glucose concentration observed near the pyloric sphincter, if extrapolated to all the water present, would account for only 40% of the glucose in that organ. It appears that the increased proportion of fat impeded the extraction of glucose from the diet and perhaps also the mixing of the glucose solution within the stomach.
1
DISCUSSION
s : z II 3 SMALL
f
HIGH-FAT
G LUCOSE
-FAT
HIGH-GLUCOSE
MIXTURE
LARGE
FED
FIG. 5. Effect of amounts of glucose and fat fed on their fractional emptying rates from rat stomach. Glucose and fat were fed together without additional ingredients in following amounts, respectively: 0.3andO.lgforA;0.3and0.4gforB; l.ZandO.lgforC;1.2and0.4 g for D. Stomachs were removed 100 min after midpoint of meal. Means + SE for K#J..~~t ratios were 5.1 + 1.21 for A; 3.10 2 0.40 for B; 28.7 t 3.51 for C; and 4.7 + 1.43 for D.
TABLE 2. Association between percentage stomachs shortly afier meals and gastric emptying ratio Type of Meal
GastriL:zpty
of water in
ing
Water in Stomach, %
From Expt 3 Sprague-Dawley Basal 4.3 _+ 0.50 (25),,* 64 t 1(12)b* Wistar Basal 1.9 +_ 0.11 (ll), 53 + 1 (8), From Expt 4 Group A Small 6.9 -e 1.27 (7)b 79 + 2 (6), B High-fat 2.4 t 0.29 (7), 60 2 1 (6), c High-glucose 17.1 2 2.13 (7), 72 + 2 (6)b D Large 2.6 2 0.37 (7), 63 A 2 (6), Values are means + SE. Numbers in parentheses are numbers of animals studied. Gastric emptying ratios, i.e., I&,&~ , were calculated from data reported in Figs. 3 and 4. Percentage of water in stomach contents was determined for animals in the same study which had been deprived of water when fed; stomachs were taken under anesthesia 20 min afbr the midpoint of the test meal. * Means not sharing a common subscript letter are different (P < 0.05) by Duncan’s Multiple Range Test.
after a test meal is one of several factors affecting the degree to which soluble sugar is emptied from the stomach preferentially to fat. Four stomachs from experiment 4 were collected 20 min aRer the middle of the eating period and examined to determine how well the glucose within the contents
One should be careful not to overgeneralize from the data presented. First, KB1 and Kfat values are based on an assumption that the disappearance of glucose and fat from the stomach, when these are fed in dry nutrient mixtures, follows first-order kinetics. In a recent study in which basal diet was fed to Wistar rats (281, we have shown an initial delay before the disappearance of glucose and fat came to approximate a semilogarithmic relationship to time. This imperfection in our model implies inaccuracy if one extrapolates the observed values of Kgl and Krat beyond the period when stomach contents were sampled. Our primary concern is with the &1&t ratio as a measure of the degree to which glucose is emptied from the stomach preferentially to fat. We have previously observed that, with one strain of rat and one test meal, K,,lK,, ratios were relatively constant when determined over an interval of 1, 2, 4, or 6 h a&r the beginning of the meal. We have no guarantee that this relationship holds for all the conditions of feeding reported here. A third caution derives from the apparent difference in Kgl/K& ratios when determined in Sprague-Dawley rats after 18-22 h vs, 2 days of prior starvation. One should not assume that our KBt /I;krar values should apply strictly to the ad libitum-fed condition. ARer people ingest liquid meals, most of the simple sugars leave the stomach before appreciable amounts of fat are emptied (21), Our Ksl/Kfat ratios of 17 and 28 for rats fed the high-glucose, low-fat meals indicate that a similarly high degree of preferential emptying of simple sugar versus fat can occur in rats fed nonliquid meals with access to drinking water (see Figs. 4 and 5). When the glucose was reduced and the fat increased in the test meals, KBt /Kfal ratios were reduced to 2.4 and 3.1 in the same experiments. K&Krat varied widely in rats with the composition of glucose-containing meals. We have also shown that the addition of a viscous gel to a glucose-containing meal inhibited the preferential emptying of glucose versus fat and reduced the K,,lK,, ratio to 1.1. Other stabilizing agents exist normally in the stomach. Gastric mucus is a gel, although not a
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 16, 2019.
GASTRIC
EMPTYING
OF
GLUCOSE
VS.
FAT
E665
IN RATS
viscous one. Some complex carbohydrates and proteins from the diet may act together to stabilize gastric contents (26). It was further shown that the strain of rat given the test meal influenced the degree to which glucose is emptied from the stomach preferentially to fat. The mechanism of this strain effect is not clear. However, the presence of extra water in the stomachs of the Sprague-Dawley rats could be expected to speed the extraction of glucose from the test meal, to promote the physical separation of glucose from fat and protein in the stomach, and to enhance the preferential emptying of glucose. However, other factors, such as the state of gastric mucoproteins and the mixing action of gastric contractions, may also be involved. The preferential gastric emptying of carbohydrate versus fat is not unique to glucose. Unpublished data of ours indicate that glucose, sucrose, and maltose are preferentially emptied from the stomach to similar degrees. Starch is also preferentially emptied but ‘to a lesser extent than is glucose (28). Our interest in the preferential gastric emptying of carbohydrate versus fat relates particularly to sucrose, a substance whose high intake is widely suspected of . contributing importantly to atherogenesis (2, 8, 22, 24) and the appearance of maturity-onset diabetes (7-9, 24, 29) in susceptible people. The mechanisms underlying sucrose effects are not well understood. It is particularly
confusing that the size of effects of feeding sucrose in place of starch varies widely from situation to situation (2, 7, 18, 24). We suspect that the degree to which the ratio of sucrose versus fat entering the metabolic system fluctuates between successive meals mav be imnortant. This hypothesis seems reasonable in viewA of the several ways in which the absorption of fat influences the metabolic responses to simultaneously absorbed carbohydrate. First, the fat reduces the con version of carbohydrate to fat and inhibits related pathways (1, 17). Second, although fat has little effect on serum insulin when given alone (4, 11, 16), it increases greatly the insulin elevation elicited by carbohydrate (11, 14, 16, 23). Third, dietary fat tends to elevate serum levels of two other metabolic regulators, i.e., glucagon and plasma free fatty acids, even in the presence of high serum glucose (4, 11, 15, 16). Present data indicate that xanthan gum may prove useful in exploring the PhYsiological significance of gastric emptying ratios and their possible role in mediating special nutritional effects of various carbohydrates. This gum, which is nontoxic (3, 30) and provides little digestible nutrient (3), allows control of the gastric emptying ratio without altering appreciably the nutrient mixture of the diet. The authors thank tions. Received for publication
Mr.
P. P. Padovano 27 May
for preparing
the illustra-
1977.
REFERENCES 1.. ABRAHAM, S. Effect of diet on hepatic fatty acid synthesis. Am. J. CZin. Nutr. 23: 1120-1128, 1970. 2. BENDER, A. E., AND K. B. DAMJI. Some effects of dietary sucrose. World Rev. Nutr. Dietetics 15: 104-155, 1972. 3. BOOTH, A. N., A. P. HENDRICKSON, AND F. DE EDS. Physiologic effects of three microbial polysaccharides on rats. ToxicoZ. Appl. Pharmacol. 5: 478-484, 1963. 4. B~TTGER, I., R. DOBBS, G. R. FALOONA, AND R. H. UNGER. The effects of triglyceride absorption upon glucagon, insulin and gut glucagon-like immunoreactivity. J. Clin. Invest. 52: 2532-2541, 1973. 5. CANNON, W. B. The passage of different food-stuffs from the stomach and through the small intestine. Am. J. Physiol. 12: 387-418, 1904. 6. CHRISTIANSON, D. D., H. W. GARDNER, K. WARNER, B. K. BOUNDY, AND G. E. INGLETT. Xanthan gum. Protein-fortified starch bread. Food TechnoZ. 28: 23-29, 1974. 7. COHEN, A. M. Effect of dietary carbohydrate on the glucose tolerance curve in the normal and in the carbohydrate-induced hyperlipemic subject. Am. J. CZin. Nutr. 20: 126-130, 1967. 8. COHEN, A. M., S. BAVLY, AND R. POZANSKI. Change of diet of Yemenite Jews in relation to diabetes and ischemic heart disease. Lancet 2: 1399-1401, 1961. 9. COHEN, A. M., A. TEITELBAUM, AND R. SALITERNIK. Genetics and diet as factors in development of diabetes mellitus. MetaboZism 21: 235-240, 1972. 10. CONWAY, E. S., AND M. ADAMS. Determination of fat in body tissues and in food mixtures. J. Assoc. Offic. AnaZ. Chem. 58: 23-27, 1975. 11. DOBBS, R., G. R. FALOONA, AND R. H. UNGER. Effect of intravenously administered glucose on glucagon and insulin secretion during fat absorption. MetaboZism 24: 69-75, 1975. 12. DUBOWSKI, K. M. An o-toluidine method for body-fluid glucose determination. CZin. Chem. 8: 215-235, 1962. 13. DUNCAN, D. B. Multiple range and multiple F tests. B’iometrics 11: l-42, 1955. 14. FELBER, J.-P., AND A. VANNOTTI. Effects of fat infusion on
15. 16.
17.
18. 19.
20.
21.
22.
23.
24.
25.
glucose tolerance and insulin plasma levels. Med. Exptl., BaseZ 10: 153-156, 1964. FREDERICKSON, D. S., AND R. S. GORDON, JR. Transport of fatty acids. PhysioZ. Rev. 38: 585-630, 1958. G~MEZ, F., E. J~QUIER, V. CHABOT, V. B~BER, AND J.-P. FELBER. Carbohydrate and lipid oxidation in normal human subjects: its influence on glucose tolerance and insulin response to glucose. MetaboZism 21: 381-391, 1972. HILL, R., J. M. LINAZASORO, F. CHAVELLIER, AND I. L. CHAIKOFF. Regulation of hepatic lipogenesis: the influence of dietary fats. J. BioZ. Chem. 233: 305-310, 1958. HODGES, R. E., AND W. A. KREHL. The role of carbohydrates in lipid metabolism. Am. J. CZin. Nutr. 17: 334-343, 1965. HUNT, J. N., AND M. T. KNOX. Regulation of gastric emptying. In: Handbook of PhysioZogy. Alimentary CanaZ. Motility. Washington, D. C.: Am. Physiol. Sot., 1968, sect. 6, vol. IV, chapt. 94, p. 1917-1935. HUNT, J. N., AND D. F. STUBBS. The volume and energy content of meals as determinants of gastric emptying. J. Physiol., London 245: 209-225, 1975. JOHANSSON, C. Studies of gastrointestinal interactions. IV. Gastric emptying of a composite meal in man. The influence of glucose. Stand. J. Gastroenterol. 8: 533-539, 1973. QURESHI, R. U., P. A. AKINYANJU, AND J. YUDKIN. The effect of an “atherogenic” diet containing starch or sucrose upon carcass composition and plasma lipids in the rat. Nutr. Metab. 12: 347357, 1970. RANDLE, P. J., P. F. GARLAND, C. N. HALE, AND E. A: NEWSHOLME. The glucose-fatty acid cycle. Its role in insulin sensitivity and the metabolic disturbance of diabetes mellitus. Lancet 1: 785-789, 1963. REISER, S. Metabolic effects of dietary carbohydrates-a review. In: Physiological Effects of Food Carbohydrates, edited by A. Jeanes and J. Hodge. Washington, D. C.: Am. Chem. Sot., 1975, p. 20-45. REYNELL, P. C., AND G. H. SPRAY. The absorption of glucose by the intact rat. J. Physiol., London 134: 531-537, 1954.
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 16, 2019.
TROUT, S., AND E. S. NASSET. Gastric emptying and intestinal absorption of carbohydrate and protein as influenced by the nature of the test meal. J. Nutr. 66: 91-103, 1958. 27. SNEDECOR, G. W., AND W. G. C~CHRAN. Statistical Methods (6th ed.). Ames: Iowa Univ. Press, 1967. 28. TROUT, D. L., E. S. CONWAY, AND J. D. PIJTNEY. Dietary influences on gastric emptying of carbohydrate vs. fat in the rat. J.Nutr. 107:104-111,1977.
26. ROSENTHAL,
PUTNEY,
AND CONWAY
S., A. C. B. WICKS, E. KANENGONI, AND J. J. JONES. Can diet be responsible for the initial lesion in diabetes? Lancet
29. WAPNICK,
2:300-302,1972. 30. W~~DARD, AND M. T.
G., M. W. WOODARD, W. H. MCNEELY, P. KOVACS, I. CRONIN. Xanthan gum: safety evaluation by twoyear feeding studies in rats and dogs and a three-generation reproduction study in rats. Toxicol. Appl. PharmacoZ. 24: 30-36, 1973.
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 16, 2019.
L. TROUT,
JOHN
D. PUTNEY,
AND EMILY
of glucose
vs. fat
S. CONWAY
Nutrition Institute, Agricultural Research Service, United States Department of Agriculture, Beltsville, Maryland 20705
TROUT, DAVID L., JOHN D. PUTNEY, AND EMILY S. CONWAY. Relative rates of gastric emptying of glucose vs. fat in rats fed nonliquid meals. Am. J. Physiol. 234(6): E660-E666, 1978 or Endocrinol. Metab. Gastrointest. Physiol. Am. J. Physiol.: 3(6k E660-E666, 1978. - Sprague-Dawley rats were briefly starved, fed various test meals, and killed at measured intervals, and the average fractional emptying (disappearance from stomach) rates for glucose (K,,) and for fat (K,,,) were determined. The Kgl /&, ratio was calculated as a measure of the degree to which the stomach emptied glucose preferentially to fat. The size of the meal affected this ratio, which was 7.6 for a small (0.5 g) meal and 2.4 for a large (2.0 g) meal of a nutritionally complete diet. When test meals contained one of two levels of fat (0.4 and 0.1 g) and of glucose (1.2 and 0.3 g). the high level of fat depressed &, and &,I&, , whereas the high level of glucose depressed K,, and particularly Kfa, and, therefore, raised &, &, . Ku,/Kf,t was also affected by strain of rat and was reduced almost to 1 .O by mixing the meal into a viscous gel of xanthan gum. In the absence of this gel, the percentage of water existing in stomach contents shortly after the test meals varied between 53 and 79% and was suspected of influencing KRI &, .
25). When the stomach receives discrete sugar-containing meals and empties the carbohydrate preferentially to fat, it must introduce into the duodenum and, ultimately, into the metabolic system, a relatively highcarbohydrate nutrient mixture after the meal and a relatively high-fat mixture before the next meal. It has been suggested that metabolic responses to eating mono- and disaccharides versus complex carbohydrates are importantly influenced by the degree to which the relative rates of intestinal absorption of carbohydrate versus fat fluctuate between successive meals (28). In the present report, we have studied the disappearance of glucose and fat from the stomach of rats that were starved and then fed dry or relatively dry nutrient mixtures in order to identify factors that influence the degree to which the stomach empties simple sugar preferentially to fat. Factors evaluated include the strain of rat, the addition of a gel to the test meal, size of the meal, and the relative and absolute amounts of glucose and fat ingested.
test
METHODS
meals;
rates; Wistar
food deprivation; average rats; Sprague-Dawley rats
fractional
emptying
WHEN A SIMPLE SUGAR is fed with fat, the stomach empties the sugar at a greater fractional rate than the fat (21,28). Reasohs for a preferential emptying of sugar versus fat relate to the anatomy of the stomach and to a partial separation of the sugarand fat in that organ. During a meal, water may be ingested, and salivary and gastric secretions are added to the food. The water present dissolves part or all of the simple sugar but none of the fat. The fat may separate and float or may associate itself with protein and tend to sink. However, to leave the stomach, the fat must reach the level of the gastroduodenal junction, which, for a standing or sitting person or for a horizontally oriented rat, is higher than the low point of the stomach, yet lower than the high point. Thus, when either floating or sinking, fatcontaining particles tend to be retained in the stomach, allowing the simple sugar to be preferentially emptied. Ingested fat and even carbohydrate are normally delayed much longer in the stomach than in the small intestine prior to their digestion and absorption (5). The rates of absorption of carbohydrate and fat are controlled mainly at the stage of gastric emptying (19, 20,
Animals were principally male Sprague-Dawley rats weighing 200-275 g but included 19 male Wistar rats of similar body size from Hilltop Laboratory Animals, Inc., Scottdale, Pa. 1 All were maintained in our laboratory for at least 5 days and then starved for 18-22 h or for 2 days. Food deprivation for 18 h was found to reduce stomach contents of glucose and fat. usually to less than 1 and 5 mg, respectively; residual contents were ignored in calculating nutrient disappearance from the stomach after test meals. After being starved, rats were offered test meals of known size and composition, observed until they began to eat, and were allowed up to 15 min to finish. The amount of food actually consumed was determined by weighing the food cups before and after the meal and correcting for diet spillage. Five minutes before animals were scheduled to be killed, they were injected with 4050 mg amobarbital. When anesthesia developed, stomachs were promptly exposed, tied off at the intestinal and esophageal ends, and removed intact. The time of l Mention of a trademark or proprietary product does not constitute a guarantee or warranty of the product by the U.S. Department of Agriculture, and does not imply its approval to the exclusion of other products that may also be suitable.
E660
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 16, 2019.
GASTRIC
EMPTYING
OF
GLUCOSE
VS.
FAT
IN RATS
sampling was taken to be the interval between the midpoint of the eating period and the tying off of the stomachs. Animals whose stomachs were sampled after 100 or 120 min were provided water ad libitum, and their stomachs were analyzed for glucose and fat by standard methods (10, 12). Details of diet analysis and animal treatment and of procedures for recovering and sampling gastric contents were as previously described (28)
The fraction of the ingested glucose or fat recovered in the stomachs was calculated as Fgl and F,,,; time of sampling after the meal was measured in hours as t. The rate constants for the disappearance of glucose and fat from the stomach were also calculated, respectively, as (-In Fgl)/t and (-In F&/t and were called the average fractional emptying rates for glucose and fat. The KS&&, ratio, which was also called the gastric emptying ratio (28), was also calculated and was considered to be a measure of the degree to which the stomach emptied glucose preferentially to fat. Kgl, K&, and the Kgl/Kfat ratio were calculated for individual rats, and statistics were run on these normalized data. The study consisted of five experiments in which KB1, Kfat , and the Kgl/Kfat ratio were determined after rats were fed test meals. Table 1 indicates the size and composition of the test meals, the time a&r the meal when gastric contents were measured, and also the nature of the previous feeding and the period of starvation during which stomachs were emptied prior to the test meal. At least one of the test meals used in four experiments was composed of a basal diet containing, by weight, 3 parts glucose, 1 part partially hydrogenTABLE
1. Protocol
ated vegetable oil (Crisco) and 1 part of “essential nutrients,” a mixture of protein, vitamins, and minerals (28) Tie effects of varying meal size, of incorporating (or not incorporating) the test meal in a viscous, nonnutritive gel, and of feeding 1the same test meal to rats of two strains (Wistar vs. Sprague-Dawley) were determined, respectively, in experiments 1,2, and 3. Expriments 4 and 5 entailed feeding all combinations of the highest and lowest levels of glucose (1.2 and 0.3 g) and of fat (0.4 and 0.1 g partially hydrogenated vegetable oil) used in experiment 1. Early data (unpublished observations) indicated a need to promote the consumption of the test meals of experiments 4 and 5. Therefore, we familiarized these animals with the nutrient mixture of their final test meal during a 4-5 day pretest period, while feeding a second diet that insured a relatively balanced total intake. As Table 1 indicates, animals of experiment 4 ate similar daily quantities of all essential nutrients during their pretest period, but the total diets for groups B and C were high, respectively, in fat and glucose. In experiment 5, the two diets given to each rat during this time were so adjusted that animals of all groups ate similar quantities of all nutrients. Twenty animals in exprimertts 3 and 28 in experiment 4 were used to investigate a factor suspected of influencing Kg1lKfat importantly, i.e., the early addition of water, through salivary and gastric secretions, to stomach contents. Having been prefed and starved like others in these experiments, animals were denied water during and after presentation of their respective test meals, and stomachs were removed under amobarbital
of experiments Test Meal: Weight of Components,
Erpt No. and Group
Glucose
Fat”
1A 1B IC
0.3 0.6 1.2
0.1 0.2 0.4
0.1
2A 2B
0.6 0.6
0.2 0.2
0.2 0.2
3A’ 3B
L2 1.2
0.4 0.4
4A 4B 4c 40
0.3 0.3 1.2 1.2
5A 5B 5c 5D
0.3 0.3 1.2 1.2
g Total offered
Total eaten
Time of Sampling Stomachs,d min
0.5 1.0 2.0
0.50 0.98 1.96
100 100
18-22 18-22 18-22
Diet of subsequent test meal (basal) ad libiturn
1.0 0
2.0 1.0
1.99 0.99
100 100
18-22 18-22
Biscuit diet ad libitum”
0.4 0.4
0 0
2.0 2.0
1.85 1.86
120 120
ca 48 ca 48
Biscuit diet ad libitum
0.1
0.1
0.5 0.8 1.4 1.7
loo
0.1 0.1 0.1
0 0 0 0
0.49
0.4
0.76 1.25 1.57
100 100 100
18-22 18-22 18-22 18-22
Each rat except for group A was fed 2 diets in separate dishes”
0.4 0.7 1.3 1.6
0.40 0.68 1.15 1.39
100
18-22 18-22 18-22 18-22
Each rat was fed 2 diets in separate dishesh
0.1 0.4 0.1 0.4 0.1 0.4
Othersb
0.2 0.4
100
100 100 loo
F+io&~Fni
tth*
Previous
Diet for 4-5 Days
a Partially hydrogenated vegetable oil (Crisco). b Others is a mixture of essential nutrients containing, per 100 g: 75 g lactalbumin, 5 g vitamin mixture, and 20 g minerals. See ref. 28 for details. c Xanthan gum (4 g/NO g) of a viscous aqueous suspension. * Time of sampling stomach is interval between midpoint of the eating period and the tying off of the stomach under anesthesia, prior to its removal. (’ D and G Bat and Mouse Diet, Price-Wilhoite Co., Frederick, Md. f Variable is strain of rat, Sprague-Dawley (A ) vs. Wistar (B). R All rats received their respective test meal diet ad libitum during their pretest period. Those of groups B, C, and D also received restricted amounts of their test meal diet supplemented with extra essential nutrients. Each rat received a total of 2.4 g la&albumin and 3.2 g total essential nutrients daily if he consumed all the restricted diet and sufficient of the ad libitum mixture to obtain a total of 76 kcal metabolizable energy per day. h All rats were ad libitum fed the glucose-fat mixture of their subsequent test meal and also fed, in restricted amounts, a second complementary diet. Animals consumed 9.6 g glucose, 3.2 g fat, and 3.2 g essential nutrients if they ate all of the diet offered and an appropriate amount of the glucose-fat mixture. Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 16, 2019.
TROUT,
E662 anesthesia 20 min a&r the midpoint of the meal. Water was determined in stomach contents, which were weighed before and after drying overnight in a 100°C air oven. The Student t test and analysis of variance (27) and Duncan’s multiple range test (13) were used to evaluate data (P > 0.05 considered not significant).
PUTNEY,
GLUCOSE I.(
AND
CONWAY
I
-
.E RESULTS
In general, about 90% of the Sprague-Dawley rats began eating their test meals within 5 min, whereas about 20% of the Wistar rats used in a recent study (28) delayed for at least 30 min. This behavioral difference afkcted greatly the amount of time and effort required to observe when each of a group of rats begins eating. Table 1 indicates how much of the meals offered was actually eaten by our Sprague-Dawley rats. In all experiments, at least 80% of the Sprague-Dawley rats consumed 90+% of their test meals within the 15 min allowed. Data plotted in Figs. l-5 are gastric emptying rates. However, one can calculate Fgl and Ffat values (i.e., the fraction of the ingested glucose or fat recovered in the stomach t hours after the midpoint of the ‘test meal) from the following relationships: I$ = antilog, ( -K, x t) and Ffa, = antilog, ( -KPat x t). Figure 1 indicates the effect of meal size on gastric
l-l 1GLUCOSE
1 I I
IL ).5
SIZE
1 2.0 OF
MEAL
c : ::
(grams)
Effect of meal size on fractional emptying (disappearance from stomach) rates for glucose and fat in rats, i.e., Kg, and Krat. These rates equal (-In F)/t, where F is fraction of ingested glucose or fkt recovered in stomach, and t is time (in hours) of sampling stomach contents after midpoint of test meal. Test meals were of basal diet, and time of sampling stomach contents was 1 */a h. Bars and vertical lines represent mean 2 1 SE for 7 rats. Mean t SE of Kcl &a* ratio were 7.63 t 1.85, 2.72 2 0.33, and 2.41 t 0.34, respectivelv, after 0.5, 1.0, and 2.0 g meals. FIG.
1.
Y
.4
: .2
DIET
t GEL
DIET
TEST
ALONE
MEAL
FIG. 2. Effect of adding gel to basal diet on fractional emptying rates for glucose and fat. Test meals contained 1 g basal diet, which was fed alone or as mixed with 1 g gel containing 40 mg xanthan gum, and stomachs were sampled 100 min after midpoint of meals. Means + SE ofK,,/K,, were 1.09 + 0.02 and 3.23 + 0.10, respectively, for animals that received gel or no gel.
I
s c r’I
.E
emptying rates in experiment 1. For animals that received the 0.5-g meal, the I& of 1.73 observed 1.67 h after the midpoint of the test meal indicates that Flg1 equaled antilog, (-1.73 x 1.67), i.e., 0.056 (or 5.6%). The K& of 0.377 for this group indicated a Frat of 53.3%. For the group that received the 2.0-g meal, the geometric means of Fgl and Ffat were 23.1 and 53.3%, respectively. Glucose emptied from the stomachs preferentially to fat asker these and, to varying degrees, all other meals. The higher Kg1/Kfat ratio of animals fed the 0.5-g versus the 2.0-g meal indicates that the preferential emptying of glucose versus fat was particularly prominent after the small meal. Figure 2 summarizes results of a second experiment that was designed to determine how effectively the addition of a viscous gel to the basal diet tended to equalize Kg1 and Kfat . The gel, a 4% suspension (wt/wt) of xanthan gum in water, was within the concentration range that was used to bind cornstarch and relatively hydrophobic proteins into doughs suitable for breadmaking (6). The addition of gel decreased KB1, tended to raise Kfat (0.05 c P c O.lO), and decreased Kfat to 1.09. Because xanthan gum is chemically inert and nontoxic (3, 30) and should not greatly inhibit absorption, the low Kgl/Kfat ratio observed with the gel indicates that the high ratios observed in its absence were not due to glucose absorption in the stomach.
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 16, 2019.
GASTRIC
EMPTYING
OF GLUCOSE
VS. FAT IN RATS
3 was prompted by our observation that Sprague-Dawley rats were giving slightly higher gastric emptying ratios than had been observed with 2-day starved Wistar rats (28). Gastric emptying was observed in four lots of Sprague-Dawley rats and in two lots of Wistar rats after they had been starved for 2 days and fed 2 g basal diet. As Fig. 3 indicates, under these conditions, stomachs of SpragueDawley rats emptied glucose faster than those of the Wistar rats. The KgI/Krat ratios were appreciably higher for the Sprague-Dawley than for the Wistar rats and also higher than had been observed in Sprague-Dawley rats that were accustomed’ to the nutrient mixture of the test meal, then starved for only 18-22 h (see Fig. 3). In experiment 4, two levels of glucose and of fat were provided in four test meals, which contained equal amounts of essential nutrients. As Fig. 4 indicates, the high level of glucose lowered Kg1 and particularly Kfal and, therefore, raised the K,,lK,,, ratio. The high level of fat elevated Kfat and depressed Kg1 and the Kgl/Kfat ratio. A notably high Kgl/Kfat ratio of 17.1 was observed in animals receiving the high-glucose, low-fat meal. In experiment 5, test meals contained only glucose and fat, in the amounts fed in experiment 4, and results are summarized in Fig. 5. As in experiment 4, the high level of glucose lowered Kg1 and particularly Kfat and thus elevated the Kgl/Kfat ratio. The high level of fat lowered KBI and the Kg1/Kfat ratio. The Kg1/Kfat ratio was
!
Experiment
18- to 22-h starved
3.0 I 2.50
GLUCOSE
I
2.0 2.25 ! 1.75'y 150
I 1.0 .75 1.25L 50' r
z: ncc T
.25
SMALL
SPRAGUE
-DAWLEY STRAIN
HIGHMEALS
WITH
D
HIGH-GLUCOSE
ESSENTIAL
f
I
C
FAT
‘Im :
T I :
s
LARGE
NUTRIENTS
FIG. 4. Effect of amounts of glucose and fat fed on their fractional emptying rates from rat stomach. Amounts of glucose and fat fed were, respectively, 0.3 and 0.1 g for A; 0.3 and 0.4 g for B; 1.2 and 0.1 g for C; 1.2 and 0.4 g for D; and a 0.1 g mixture of protein, vitamins, and minerals was included. Stomachs were removed 100 min after midpoint of meal. Mean + SE for Kg1/I& ratios were 6.9 2 1.27 for A; 2.4 + .29 for B; 17.1 * 2.13 for C; and 2.6 st .37 for D.
W ISTAR
OF RAT
3. Fractional emptying rates for glucose and fat in 2 strains of rats. Two-day-starved animals were provided 2-g basal diet and stomachs were sampled 2 h after midpoint of test meal. Mean t SE for & /I& were 4.3 t 0.50 for Sprague-Dawley and 1.9 2 0.11 for Wistar rats. FIG.
: I3
A
TEST
1 J GLUCOSE
: : z z
nearly 30 for animals fed the high-glucose, low-fat meal. In experiments 3 and 4, the water in stomach contents was determined 20 min after the middle of the eating period in those animals deprived of drinking watey when given their test meal. Data are presented as percentage of water in stomach contents in Table 2. The percentage of water in gastric contents was greater in Sprague-Dawley than Wistar rats when animals of both strains were fed the same test meal and was greater in Wistar rats a&r meals containing 0.1 g than after meals containing 0.4 g fat. Treatment means for percentage of gastric water varied between 53 and 79%, implying ratios of 1.1-3.8 g water/g dry matter. It appears that the various test meals elicited markedly different amounts of gastric and salivary secretions per gram consumed. Table 2 also records the Kgl/Kfat ratios for animals of experiments 3 and 4, as indicated in Figs. 3 and 4. No simple relationship between KB1/Kfat and percentage of water in stomach contents was-observed; but, with one exception among six groups, increasing Kgl/Kfat values were associated with increasing percentages of gastric water. Data are consistent with the notion that the relative proportion of water in stomach contents shortly
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 16, 2019.
TROUT,
q
GLUCOSE
cl
FAT
1
PUTNEY,
AND CONWAY
was extracted by water and mixed. The stomachs were cut near the pyloric sphincter; several drops of fluid were expelled, filtered through cotton, and analyzed for glucose. Water was also determined in the remaining stomach contents. In two animals fed the high-glucose meal, no crystals of glucose were observed in the stomachs, and the concentration of glucose in the initially expelled fluid approximated the expected value if the glucose known to be present in the stomachs was all dissolved and well mixed. In two animals fed the highfat meal, however, glucose crystals were observed under low-power magnification; and the glucose concentration observed near the pyloric sphincter, if extrapolated to all the water present, would account for only 40% of the glucose in that organ. It appears that the increased proportion of fat impeded the extraction of glucose from the diet and perhaps also the mixing of the glucose solution within the stomach.
1
DISCUSSION
s : z II 3 SMALL
f
HIGH-FAT
G LUCOSE
-FAT
HIGH-GLUCOSE
MIXTURE
LARGE
FED
FIG. 5. Effect of amounts of glucose and fat fed on their fractional emptying rates from rat stomach. Glucose and fat were fed together without additional ingredients in following amounts, respectively: 0.3andO.lgforA;0.3and0.4gforB; l.ZandO.lgforC;1.2and0.4 g for D. Stomachs were removed 100 min after midpoint of meal. Means + SE for K#J..~~t ratios were 5.1 + 1.21 for A; 3.10 2 0.40 for B; 28.7 t 3.51 for C; and 4.7 + 1.43 for D.
TABLE 2. Association between percentage stomachs shortly afier meals and gastric emptying ratio Type of Meal
GastriL:zpty
of water in
ing
Water in Stomach, %
From Expt 3 Sprague-Dawley Basal 4.3 _+ 0.50 (25),,* 64 t 1(12)b* Wistar Basal 1.9 +_ 0.11 (ll), 53 + 1 (8), From Expt 4 Group A Small 6.9 -e 1.27 (7)b 79 + 2 (6), B High-fat 2.4 t 0.29 (7), 60 2 1 (6), c High-glucose 17.1 2 2.13 (7), 72 + 2 (6)b D Large 2.6 2 0.37 (7), 63 A 2 (6), Values are means + SE. Numbers in parentheses are numbers of animals studied. Gastric emptying ratios, i.e., I&,&~ , were calculated from data reported in Figs. 3 and 4. Percentage of water in stomach contents was determined for animals in the same study which had been deprived of water when fed; stomachs were taken under anesthesia 20 min afbr the midpoint of the test meal. * Means not sharing a common subscript letter are different (P < 0.05) by Duncan’s Multiple Range Test.
after a test meal is one of several factors affecting the degree to which soluble sugar is emptied from the stomach preferentially to fat. Four stomachs from experiment 4 were collected 20 min aRer the middle of the eating period and examined to determine how well the glucose within the contents
One should be careful not to overgeneralize from the data presented. First, KB1 and Kfat values are based on an assumption that the disappearance of glucose and fat from the stomach, when these are fed in dry nutrient mixtures, follows first-order kinetics. In a recent study in which basal diet was fed to Wistar rats (281, we have shown an initial delay before the disappearance of glucose and fat came to approximate a semilogarithmic relationship to time. This imperfection in our model implies inaccuracy if one extrapolates the observed values of Kgl and Krat beyond the period when stomach contents were sampled. Our primary concern is with the &1&t ratio as a measure of the degree to which glucose is emptied from the stomach preferentially to fat. We have previously observed that, with one strain of rat and one test meal, K,,lK,, ratios were relatively constant when determined over an interval of 1, 2, 4, or 6 h a&r the beginning of the meal. We have no guarantee that this relationship holds for all the conditions of feeding reported here. A third caution derives from the apparent difference in Kgl/K& ratios when determined in Sprague-Dawley rats after 18-22 h vs, 2 days of prior starvation. One should not assume that our KBt /I;krar values should apply strictly to the ad libitum-fed condition. ARer people ingest liquid meals, most of the simple sugars leave the stomach before appreciable amounts of fat are emptied (21), Our Ksl/Kfat ratios of 17 and 28 for rats fed the high-glucose, low-fat meals indicate that a similarly high degree of preferential emptying of simple sugar versus fat can occur in rats fed nonliquid meals with access to drinking water (see Figs. 4 and 5). When the glucose was reduced and the fat increased in the test meals, KBt /Kfal ratios were reduced to 2.4 and 3.1 in the same experiments. K&Krat varied widely in rats with the composition of glucose-containing meals. We have also shown that the addition of a viscous gel to a glucose-containing meal inhibited the preferential emptying of glucose versus fat and reduced the K,,lK,, ratio to 1.1. Other stabilizing agents exist normally in the stomach. Gastric mucus is a gel, although not a
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 16, 2019.
GASTRIC
EMPTYING
OF
GLUCOSE
VS.
FAT
E665
IN RATS
viscous one. Some complex carbohydrates and proteins from the diet may act together to stabilize gastric contents (26). It was further shown that the strain of rat given the test meal influenced the degree to which glucose is emptied from the stomach preferentially to fat. The mechanism of this strain effect is not clear. However, the presence of extra water in the stomachs of the Sprague-Dawley rats could be expected to speed the extraction of glucose from the test meal, to promote the physical separation of glucose from fat and protein in the stomach, and to enhance the preferential emptying of glucose. However, other factors, such as the state of gastric mucoproteins and the mixing action of gastric contractions, may also be involved. The preferential gastric emptying of carbohydrate versus fat is not unique to glucose. Unpublished data of ours indicate that glucose, sucrose, and maltose are preferentially emptied from the stomach to similar degrees. Starch is also preferentially emptied but ‘to a lesser extent than is glucose (28). Our interest in the preferential gastric emptying of carbohydrate versus fat relates particularly to sucrose, a substance whose high intake is widely suspected of . contributing importantly to atherogenesis (2, 8, 22, 24) and the appearance of maturity-onset diabetes (7-9, 24, 29) in susceptible people. The mechanisms underlying sucrose effects are not well understood. It is particularly
confusing that the size of effects of feeding sucrose in place of starch varies widely from situation to situation (2, 7, 18, 24). We suspect that the degree to which the ratio of sucrose versus fat entering the metabolic system fluctuates between successive meals mav be imnortant. This hypothesis seems reasonable in viewA of the several ways in which the absorption of fat influences the metabolic responses to simultaneously absorbed carbohydrate. First, the fat reduces the con version of carbohydrate to fat and inhibits related pathways (1, 17). Second, although fat has little effect on serum insulin when given alone (4, 11, 16), it increases greatly the insulin elevation elicited by carbohydrate (11, 14, 16, 23). Third, dietary fat tends to elevate serum levels of two other metabolic regulators, i.e., glucagon and plasma free fatty acids, even in the presence of high serum glucose (4, 11, 15, 16). Present data indicate that xanthan gum may prove useful in exploring the PhYsiological significance of gastric emptying ratios and their possible role in mediating special nutritional effects of various carbohydrates. This gum, which is nontoxic (3, 30) and provides little digestible nutrient (3), allows control of the gastric emptying ratio without altering appreciably the nutrient mixture of the diet. The authors thank tions. Received for publication
Mr.
P. P. Padovano 27 May
for preparing
the illustra-
1977.
REFERENCES 1.. ABRAHAM, S. Effect of diet on hepatic fatty acid synthesis. Am. J. CZin. Nutr. 23: 1120-1128, 1970. 2. BENDER, A. E., AND K. B. DAMJI. Some effects of dietary sucrose. World Rev. Nutr. Dietetics 15: 104-155, 1972. 3. BOOTH, A. N., A. P. HENDRICKSON, AND F. DE EDS. Physiologic effects of three microbial polysaccharides on rats. ToxicoZ. Appl. Pharmacol. 5: 478-484, 1963. 4. B~TTGER, I., R. DOBBS, G. R. FALOONA, AND R. H. UNGER. The effects of triglyceride absorption upon glucagon, insulin and gut glucagon-like immunoreactivity. J. Clin. Invest. 52: 2532-2541, 1973. 5. CANNON, W. B. The passage of different food-stuffs from the stomach and through the small intestine. Am. J. Physiol. 12: 387-418, 1904. 6. CHRISTIANSON, D. D., H. W. GARDNER, K. WARNER, B. K. BOUNDY, AND G. E. INGLETT. Xanthan gum. Protein-fortified starch bread. Food TechnoZ. 28: 23-29, 1974. 7. COHEN, A. M. Effect of dietary carbohydrate on the glucose tolerance curve in the normal and in the carbohydrate-induced hyperlipemic subject. Am. J. CZin. Nutr. 20: 126-130, 1967. 8. COHEN, A. M., S. BAVLY, AND R. POZANSKI. Change of diet of Yemenite Jews in relation to diabetes and ischemic heart disease. Lancet 2: 1399-1401, 1961. 9. COHEN, A. M., A. TEITELBAUM, AND R. SALITERNIK. Genetics and diet as factors in development of diabetes mellitus. MetaboZism 21: 235-240, 1972. 10. CONWAY, E. S., AND M. ADAMS. Determination of fat in body tissues and in food mixtures. J. Assoc. Offic. AnaZ. Chem. 58: 23-27, 1975. 11. DOBBS, R., G. R. FALOONA, AND R. H. UNGER. Effect of intravenously administered glucose on glucagon and insulin secretion during fat absorption. MetaboZism 24: 69-75, 1975. 12. DUBOWSKI, K. M. An o-toluidine method for body-fluid glucose determination. CZin. Chem. 8: 215-235, 1962. 13. DUNCAN, D. B. Multiple range and multiple F tests. B’iometrics 11: l-42, 1955. 14. FELBER, J.-P., AND A. VANNOTTI. Effects of fat infusion on
15. 16.
17.
18. 19.
20.
21.
22.
23.
24.
25.
glucose tolerance and insulin plasma levels. Med. Exptl., BaseZ 10: 153-156, 1964. FREDERICKSON, D. S., AND R. S. GORDON, JR. Transport of fatty acids. PhysioZ. Rev. 38: 585-630, 1958. G~MEZ, F., E. J~QUIER, V. CHABOT, V. B~BER, AND J.-P. FELBER. Carbohydrate and lipid oxidation in normal human subjects: its influence on glucose tolerance and insulin response to glucose. MetaboZism 21: 381-391, 1972. HILL, R., J. M. LINAZASORO, F. CHAVELLIER, AND I. L. CHAIKOFF. Regulation of hepatic lipogenesis: the influence of dietary fats. J. BioZ. Chem. 233: 305-310, 1958. HODGES, R. E., AND W. A. KREHL. The role of carbohydrates in lipid metabolism. Am. J. CZin. Nutr. 17: 334-343, 1965. HUNT, J. N., AND M. T. KNOX. Regulation of gastric emptying. In: Handbook of PhysioZogy. Alimentary CanaZ. Motility. Washington, D. C.: Am. Physiol. Sot., 1968, sect. 6, vol. IV, chapt. 94, p. 1917-1935. HUNT, J. N., AND D. F. STUBBS. The volume and energy content of meals as determinants of gastric emptying. J. Physiol., London 245: 209-225, 1975. JOHANSSON, C. Studies of gastrointestinal interactions. IV. Gastric emptying of a composite meal in man. The influence of glucose. Stand. J. Gastroenterol. 8: 533-539, 1973. QURESHI, R. U., P. A. AKINYANJU, AND J. YUDKIN. The effect of an “atherogenic” diet containing starch or sucrose upon carcass composition and plasma lipids in the rat. Nutr. Metab. 12: 347357, 1970. RANDLE, P. J., P. F. GARLAND, C. N. HALE, AND E. A: NEWSHOLME. The glucose-fatty acid cycle. Its role in insulin sensitivity and the metabolic disturbance of diabetes mellitus. Lancet 1: 785-789, 1963. REISER, S. Metabolic effects of dietary carbohydrates-a review. In: Physiological Effects of Food Carbohydrates, edited by A. Jeanes and J. Hodge. Washington, D. C.: Am. Chem. Sot., 1975, p. 20-45. REYNELL, P. C., AND G. H. SPRAY. The absorption of glucose by the intact rat. J. Physiol., London 134: 531-537, 1954.
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 16, 2019.
TROUT, S., AND E. S. NASSET. Gastric emptying and intestinal absorption of carbohydrate and protein as influenced by the nature of the test meal. J. Nutr. 66: 91-103, 1958. 27. SNEDECOR, G. W., AND W. G. C~CHRAN. Statistical Methods (6th ed.). Ames: Iowa Univ. Press, 1967. 28. TROUT, D. L., E. S. CONWAY, AND J. D. PIJTNEY. Dietary influences on gastric emptying of carbohydrate vs. fat in the rat. J.Nutr. 107:104-111,1977.
26. ROSENTHAL,
PUTNEY,
AND CONWAY
S., A. C. B. WICKS, E. KANENGONI, AND J. J. JONES. Can diet be responsible for the initial lesion in diabetes? Lancet
29. WAPNICK,
2:300-302,1972. 30. W~~DARD, AND M. T.
G., M. W. WOODARD, W. H. MCNEELY, P. KOVACS, I. CRONIN. Xanthan gum: safety evaluation by twoyear feeding studies in rats and dogs and a three-generation reproduction study in rats. Toxicol. Appl. PharmacoZ. 24: 30-36, 1973.
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (129.081.226.078) on January 16, 2019.
