 |
INTRODUCTION |
Clinical studies have shown that consumption of sources of viscous polysaccharides lowers fasting plasma cholesterol (Glore et al. 1994
, Ripsin et al. 1992). Sources of dietary fiber that contain viscous polysaccharides can affect digestion, absorption and subsequent metabolism of dietary fat and cholesterol. These actions can occur via slowed gastric emptying, slowed small intestinal transit time, interference with digestive enzyme activity, binding of micellar components or slowed diffusion within the small intestinal contents. Various soluble fibers delay the absorption of dietary fat in the rat (Ebihara and Schneeman, 1989, Vahouny et al. 1988). Consequently, a greater proportion of lipid from the diet is absorbed from the more distal portion of the small intestine. This effect of fiber may prolong intestinal contribution to postprandial lipids or alter the physical properties and subsequent handling of the lipoprotein particles. The ability of sources of viscous polysaccharides such as psyllium or oat bran to alter lipid digestion and absorption may contribute to their plasma cholesterol-lowering effects.
Preadaptation of rats to cellulose, oat bran or psyllium did not alter postprandial lipids when a fiber-free, high fat test meal was fed (Redard et al. 1992). In addition, the presence of nutrients in the lower part of the small intestine of rats was insufficient to lower plasma lipids (Middleton and Schneeman 1995). Therefore the acute effects of fiber on lipid digestion and absorption may be a more important factor leading to changes in postprandial lipemia and subsequent changes in fasting lipids when soluble fibers are fed chronically. Our objective was to compare the pattern of postprandial lipemia in animals fed acutely a source of viscous polysaccharide (oat bran) or a source of insoluble, non-viscous polysaccharide (cellulose). The pattern was compared in rats consuming a meal that contained either of these fiber sources without additional treatment (Experiment 1) with those injected with Triton WR-1339 to block clearance of triglyceride-rich lipoproteins (TRL)4 (Experiment 2).
| |
MATERIALS AND METHODS |
Male Wistar rats weighing 275-300 g were obtained from Hilltop Laboratories (Scottsdale, PA) and adapted to meal-feeding using a semipurified diet containing 25% energy from fat (Table 1). Rats were individually housed in a temperature- and humidity-controlled room with a 12-h reverse light:dark cycle. The protocols for animal use were approved by the University of California, Davis Animal Use and Care Committee.
Analytical methods.
Lipoproteins were separated by sequential ultracentrifugation (Ney et al. 1986) with a Sorvall fixed angle rotor (TFT 45.6) in a Sorvall OTD-65B ultracentrifuge (Dupont Biomedical Products, Sorvall Instruments, Wilmington, DE). Lipoproteins were isolated at the following density ranges: triglyceride-rich lipoproteins (TRL, containing both chylomicrons and VLDL), d < 1.006 kg/L; LDL, d = 1.006-1.050 kg/L; and HDL, d = 1.050-1.1963 kg/L.
Plasma and lipoprotein cholesterol were measured by the cholesterol oxidase method (Allain et al. 1974, Cooper et al. 1982). Plasma and lipoprotein triglycerides (TG) were determined by a colorimetric enzymatic method that measures TG by glycerol release (Buculo and David 1973, Fletcher 1968, kit #336, Sigma Chemical, St. Louis, MO). The recovery rate of TRL-TG was lower in the 8-h groups compared with the other time points; thus, the TRL-TG values for the 8-h groups were corrected to the same recovery rate as the remaining time points.
Plasma apolipoproteins (apo) A-I, A-IV, B and E were determined by rocket immunoelectrophoresis. Antisera were a gift from P. S. Roheim (Louisiana State University Medical Center, New Orleans, LA); methods used were based on techniques developed in his laboratory (Dory and Roheim 1981). A dilution series of standardized pooled plasma samples with known apolipoprotein concentrations was analyzed simultaneously to generate a standard curve.
The concentration of retinyl palmitate (RP) was measured in plasma and the TRL fraction by reverse-phase HPLC. Plasma or TRL, 100-200 µL adjusted to 500 µL with distilled water, was mixed with 500 µL of ethanol (ETOH). Internal standard (173 pmol retinyl heptadecanoate in 1 mL hexane, synthesized from retinol and fatty acid anhydride) and 2 mL hexane were mixed together and the sample centrifuged. The top layer was removed and dried under nitrogen. The sample was redissolved in 50 µL ethanol/chloroform 3:1 and sonicated. Retinyl esters were separated by a reverse-phase HPLC column with methanol/isopropanol 2:1 as the liquid phase at a flow rate of 2 mL/min. RP was detected at 325 nm and quantitated by the area ratio method.
Experiment 1.
After 18 h without food, one group of rats was killed (0 h); the remaining rats were given 5 g of a nutritionally complete test meal. Meals contained 5 g/100 g fiber as either oat bran (OB) or cellulose (CL), and 40% energy from fat (Table 1). Meals were consumed within 15 min. Groups of rats were killed at 1.5, 4.5 and 8.5 h after consumption of the meal. Rats were anesthetized with ketamine/xylazine/acepromazine maleate (50:5:0.75 mg/kg), and blood was drawn from the inferior vena cava into a 12-mL syringe containing 200 µL 5% EDTA. Plasma was separated by centrifugation at 1200 g for 20 min at 4°C.
An overall difference between the two fiber types was determined by two-way ANOVA with time and fiber as factors, excluding the 0-h group. To determine an effect of time for each group, however, the 0-h group was included and a one-way ANOVA performed. If the effects of fiber or the interaction term were significant in the two-way ANOVA, differences between individual groups were estimated using Fischer's least significant difference (LSD) test (Bartlett 1937, Statview 512+, Brain Power, Calabasas, CA). Differences were considered significant at P 
0.05.
Experiment 2.
Rats were adapted to a specially designed jacket for 1 wk, during which time they were also adapted to meal-feeding with a purified diet (Table 2) and given free access to water. The oat bran content was higher in the diets for Experiment 2, based on a lower fiber content of this batch of oat bran. After adaptation, food was withheld overnight and rats were fitted with jugular catheters as previously described (Ney et al. 1986). After surgery, an analgesic (Buprenex, 0.1 mg/kg) was administered.
View this table:
[in this window]
[in a new window]
|
Table 3.
Plasma and triglyceride-rich lipoprotein (TRL) triglyceride
concentrations in rats fed meals containing
cellulose or oat bran1,2
|
|
Catheters were protected by a spring, which was attached to the jacket via a metal button. The spring and catheter were connected to a freely rotating infusion swivel (Instech Laboratories, Plymouth Meeting, PA), allowing relatively unrestricted movement.
Rats recovered for 2-3 d, during which time meal-feeding was continued and sterile saline infused at the rate of 0.1 mL/h to maintain catheter patency. Food was then withheld overnight. Each of the two fed groups was given a 5-g test meal containing 6% fiber from either cellulose (CL) or oat bran (OB) (Table 2). Meals were consumed within 15 min. Retinyl palmitate (Sigma Chemical) was included in the test meals to estimate the presence of intestinally derived lipoproteins (Goodman et al. 1965). Thirty minutes after administration of the meal, Triton WR-1339 (Sigma Chemical) was infused via the venous catheter at a dose of 400 mg/kg in 0.5 mL of saline. The two unfed control groups included a saline control group that remained unfed (SUF) and was infused via the catheter with 0.5 mL of saline and a Triton group, also unfed (TUF) that received the same dose of Triton via the catheter.
At 2.5 h after infusion of either Triton or saline (3 h after administration of the meal for the fed groups), rats were anesthetized via the catheters with sodium pentobarbital at a dose of 36 mg/kg. Blood was drawn via cardiac puncture into a 12-mL syringe containing 200 µL of 5% EDTA. Plasma was prepared and TRL isolated for determination of total and free cholesterol and plasma and TRL-TG.
An overall difference between the groups was determined by one-way ANOVA. If the one-way ANOVA was significant, differences between individual groups were estimated using Fisher's LSD test (Statview 512+). For plasma and TRL-TG, the variance was not homogeneous, and the data were transformed via natural logs and analyzed with one-way ANOVA (Bartlett 1937). Differences were considered significant at P 0.05.
| |
RESULTS |
Experiment 1.
Contents within the stomach increased at 1.5 h and were emptied by 8.5 h, indicating that the meal was eaten and emptied into the intestine. There were no differences in the emptying of stomach contents due to the type of fiber in the diet (data not shown).
Plasma TG was higher in the fed than unfed state only in the OB group, as seen by a higher concentration at 4.5 h than at 0 h (Table 3). Plasma TG was significantly higher in the OB than in the CL group at 4.5 h. Similarly, TRL-TG was significantly higher for the OB than the CL group at 4.5 h; only in the OB group was TRL-TG significantly lower at 0 h than at 1.5 and 4.5 h (Table 3). There was no effect of fiber or time on either LDL or HDL TG (data not shown).
Fiber type did not affect plasma or lipoprotein cholesterols, but cholesterol concentration changed with time (Table 4). For rats fed both fiber types, plasma cholesterol concentration was significantly lower than for the 0-h group at all time points. For both OB and CL, TRL-C was lower at 1.5 h than at 0 h. TRL-C was higher at 8.5 h than at 0 h, but this was significant only for the OB group. In both groups, LDL cholesterol changed significantly with time, and followed a pattern similar to that of TRL-C. LDL cholesterol was lower at 1.5 h than at 0 h but higher at 8.5 h than at 0 h. However, only the OB group was significantly different at 8.5 h than at 0 h. HDL cholesterol was significantly lower at 8.5 h than at 0 h for the OB group only.
Changes in the apolipoproteins are shown in Table 5. Apo A-I concentration ranged from 167 ± 25 to 238 ± 24 mg/L and did not differ due to time or fiber treatment. Apo A-IV ranged from 422 ± 35 to 529 ± 28 mg/L, but also was not different due to fiber or time. The fiber types were not different from each other in their apo B response. Both OB and CL changed significantly over time, with the 1.5- and 4.5-h groups fed both fiber types lower than the 0-h group. Apo E did not significantly differ between the two fiber types. Both fiber groups changed over time, and apo E was lower in the OB and CL groups at 4.5 h than at 0 h. Apo E was also lower in the OB group at 8.5 h than in the 0 h group.
Experiment 2.
Plasma TG concentrations were significantly higher in the OB, CL and TUF groups than in the SUF group (Fig. 1A). This difference in TG between the three groups infused with Triton and the group infused with saline was similar in the TRL fraction. The difference between the saline-treated and Triton-treated rats illustrates the TG accumulation due to triton administration. The slightly higher OB and CL bars suggest additional accumulation due to feeding but was not significant.
View larger version (53K):
[in this window]
[in a new window]
| Fig 1.
Plasma and triglyceride-rich lipoprotein (TRL) cholesterol and esterified/total cholesterol in rats fed meals containing cellulose or oat bran A) The concentration of triglycerides in plasma and the triglyceride-rich (TRL) fraction of plasma. B) The concentration of cholesterol in plasma and the TRL fraction of plasma. The saline and triton control groups were not fed and infused with saline and Triton, respectively. The remaining two groups were fed 5-g test meals containing either oat bran or cellulose and infused with Triton 30 min later. All rats were killed 2.5 h post-infusion. abcValues with different superscripts differ significantly (P < 0.05). Data were transformed with natural logs for statistical analysis.
|
|
Plasma retinyl palmitate was not different between the OB and CL groups, with a level in the OB group of 577 ± 225 mmol/L, and in the CL group of 735 ± 307 mmol/L.
Plasma total cholesterol was significantly higher in the two fed groups, OB and CL, than in the SUF group (Fig. 1B). Triton infusion caused a significant accumulation of cholesterol in the TRL fraction. In addition, TRL cholesterol (TRL-C) was significantly greater in the TUF group than in the two fed groups, OB and CL. TRL-C as a proportion of plasma total cholesterol was greater in the groups infused with Triton than in the SUF group. The plasma TRL/cholesterol ratio was significantly higher in the TUF group than in the two fed groups (Table 6).
View this table:
[in this window]
[in a new window]
|
Table 6.
Plasma and triglyceride-rich lipoprotein (TRL) cholesterol and esterified/total cholesterol in rats fed meals
containing cellulose or oat bran1
|
|
The fraction of esterified cholesterol in plasma appeared slightly lower after feeding, but the difference was not significant. The fraction of esterified cholesterol in the TRL fraction was significantly lower in all rats infused with Triton (Table 6).
| |
DISCUSSION |
Viscous polysaccharides may delay lipid digestion and absorption; thus one would predict that appearance of lipid in the plasma after a fiber-containing meal might be delayed and prolonged relative to a meal without fiber. Several studies have been conducted in humans to determine if fiber will attenuate the appearance of triglycerides in the plasma after a meal. No difference in lipemia was reported in a study in which bran or psyllium was added to a bolus of cream; however, the number of subjects was small and the bolus of cream not representative of a typical fiber-containing meal (Miettinen 1987). Other studies using guar gum (Gatti et al. 1984, Irie et al. 1982) or oat bran, wheat fiber or wheat germ (Cara et al. 1992) found that fiber blunted the rise in plasma triglyceride after a test meal, relative to meals without the fibers. However, other investigators have reported that certain dietary fibers enhanced postprandial lipemia. Jenkins (1978) gave subjects Lundh test meals with and without 12 g dietary fiber from wheat bran, pectin or guar gum. With both pectin and guar gum, the rises in serum TG were significantly above the control during the first 3 h. The author explained the elevation by a decreased clearance of chylomicrons secondary to a flatter glucose and insulin response. Redard et al. (1990) reported that guar gum and oat bran added to a high fat test meal resulted in an enhanced lipemia relative to a low fiber meal. Recently, Dubois et al. (1996) reported that feeding young adult men a high oat bran diet chronically resulted in an enhanced lipemia when oat bran was fed with a meal, relative to when a low fiber diet was chronically fed.
In this study, rats had an enhanced lipemia following a high fat test meal containing oat bran, compared with rats fed the test meal containing cellulose. The higher plasma TG values are reflected almost entirely in the TRL fraction. TRL is composed of both chylomicrons (CM) and VLDL, and both fractions contribute to postprandial lipemia (Cohn et al. 1989, Schneeman et al. 1993). Higher TG concentration in the OB group at 4.5 h could be due to enhanced secretion or delayed clearance of either CM or VLDL. The accumulation of TRL after a meal is most likely due to delayed clearance (Schneeman et al. 1993). Consumption of soluble fiber may contribute to such a delay, based on its previously reported ability to blunt the glucose and insulin response to a meal (Vachon et al. 1988, Wolever and Jenkins 1986). Insulin indirectly affects lipoprotein clearance through its effect on lipoprotein lipase and VLDL secretion.
The decrease over time in plasma cholesterol and HDL cholesterol postprandially that occurred with both fiber sources has been demonstrated in other studies (Cohn et al. 1988 and 1989; Dubois et al. 1993, Patsch and Braunsteiner 1991, Redard et al. 1990). Dubois et al. (1993) found that HDL cholesterol decreased slightly in human subjects fed a low fiber test meal, and this decrease was amplified by the addition of OB to the meal. Recently, the same investigators found that chronic oat bran intake followed by a high fat, oat bran-supplemented test meal resulted in increased plasma and HDL free cholesterol, and decreased plasma and HDL-cholesterol esters (Dubois et al. 1996). These results and this study indicate that lipoproteins are remodeled postprandially, with an exchange of the triglyceride, cholesterol and the apolipoproteins among the lipoprotein particles. These effects were seen most consistently in the OB groups. In the oat bran-fed rats, some movement of cholesterol from HDL to TRL and LDL appeared to have occurred by 8.5 h. This movement of cholesterol among the lipoproteins in the rats fed oat bran may be stimulated by the enhanced triglyceridemia observed at 4.5 h. Plasma from unfed rats does not have measurable cholesterol ester transfer protein (CETP) activity (Oschry and Eisenberg 1982); however, tissues contain the gene and mRNA for CETP (Jiang et al. 1991). CETP-like activity or non-enzyme-mediated cholesterol transfer may occur in rats adapted to meal-feeding and fed a high fat test meal, as in this study. Thus, the enhanced lipemia observed when oat bran is fed may stimulate reverse cholesterol transport due to a prolonged presence of TRL, which provides an acceptor for cholesterol transfer from HDL.
Similar to plasma and lipoprotein cholesterol concentrations, the apolipoproteins tended to be lower in the 1.5- or 4.5-h groups compared with the 0-h group, returning to the 0-h levels by 8.5 h. Cohn et al. (1988) demonstrated small but significant decreases in apo A-I and apo B in human subjects fed a fat-rich meal. Clearance of lipoprotein particles containing the various apolipoproteins could account for the reductions in concentration. At 8.5 h, apo E in the OB group was still lower than in the 0-h group, whereas the CL group was above the 0-h group levels, suggesting an enhanced clearance of the lower density lipoproteins with oat bran feeding.
The data in the first study indicate that a soluble fiber, oat bran, enhances postprandial lipemia in rats fed a high fat test meal. These changes in triglyceride and cholesterol indicate remodelling of the lipoprotein particles. The presence of TG-rich lipoproteins stimulates reverse cholesterol transport. Such alterations in postprandial lipemia contribute, at least in part, to the lowering of fasting plasma lipids observed when soluble fibers are consumed long term.
Investigators have used the capacity of the nonhemolytic detergent Triton WR-1339 to block lipid clearance to investigate lipid metabolism (Otway and Robinson, 1967, Scanu, 1965). Triton has been hypothesized to coat lipoprotein particles and inhibit the activity of the enzymes lecithin cholesterol acyl transferase (LCAT) and lipoprotein lipase (Scanu 1965, Schotz et al. 1957, Steiner et al. 1984), resulting in accumulation of lipoprotein particles in the bloodstream for 24-36 h. Ishikawa and Fidge have suggested that Triton dissociates apo A-I and C-II from HDL, thereby preventing their transfer to TRL. This lack of cofactors for the LCAT and lipoprotein enzymes may prevent the catabolism of TRL and result in massive accumulation of lipid in the plasma (Ishikawa and Fidge 1979). No study previous to this one has used Triton to investigate TRL metabolism in meal-fed rats.
In the second study, the infusion of Triton resulted in an increase in plasma triglyceride of more than 10-fold over the SUF group, and as expected, most of this TG accumulation was in the TRL fraction. Although plasma and TRL-TG were higher in the two fed groups than in the TUF group, this difference was not significant. The saline and Triton control plasma TG concentrations were compared with the oat bran and cellulose concentrations to estimate plasma TG attributable to the meal-induced TG secretion. In the first study in which clearance was not blocked, plasma TG concentration was higher after an OB-containing meal than after a cellulose-containing meal. The second study indicates that accumulation of TG did not differ in OB- and CL-fed rats. In addition, accumulation of retinyl palmitate was not enhanced. During the first 6-8 h of alimentary lipemia, RP serves as a marker of intestinally derived TRL. TG accumulation tended to be slightly lower in the OB than in the CL group, which would be consistent with the hypothesis that sources of viscous polysaccharides slow digestion and absorption of fat. Taken together these observations suggest that the lack of enhanced TG or retinyl palmitate accumulation with oat bran feeding in the second study, in which TG catabolism was blocked, indicates that the enhanced lipemia demonstrated in the first study was not due to a higher intestinal TRL secretion rate. Rather, slower TRL clearance rates are more likely responsible.
Although the plasma esterified cholesterol/total cholesterol ratio was not different among the groups, the ratio in the TRL fraction was significantly lower in the groups infused with Triton. Unlike humans, the rat carries cholesterol primarily in the free form, and the reduction in the ratio suggests an accumulation of free cholesterol in the TRL fraction. This may be due to a blockage or inhibition of cholesterol esterification reaction by Triton.
These data indicate that an enhanced TRL secretion rate is not responsible for the enhanced lipemia shown in the first study in rats fed oat bran relative to those fed cellulose. In fact, the slightly lower plasma TG and retinyl palmitate accumulation in the rats fed Triton-treated oat bran relative to those fed cellulose suggests that oat bran may delay intestinal lipid digestion and absorption. Thus, enhanced lipemia with oat bran feeding demonstrated in the previous study is likely due to differences in TRL clearance rate, with the TRL being cleared more slowly in rats fed oat bran. This difference in clearance rate may be secondary to hormonal differences resulting from an accompanying delay in carbohydrate digestion and absorption. Dubois et al. (1996) investigated the interaction between chronic oat bran intake and the postprandial response to an oat bran-supplemented test meal. Relative to subjects fed a low fiber diet for 14 d, subjects chronically fed oat bran had an enhanced triglyceride response to the test meal. In addition, plasma and HDL free cholesterol were greater, whereas the esterified fractions were lower. Chronic fiber feeding has been reported to increase bile acid excretion and therefore de novo synthesis of bile acids from cholesterol in the liver (Amigo et al. 1992, Everson et al. 1992). HDL particles serve as carriers to deliver cholesterol to the liver for this synthesis, and the greatest demand may occur in the postprandial period when TRL are present. Taken together with previous studies, the data from these two animal studies suggest that acute changes in the handling of dietary lipid when oat bran is fed contribute to the ability of oat bran to lower fasting cholesterol levels when it is fed chronically.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Abstract
Introduction
Abstract
