点击显示 收起
1 From the Department of Clinical Nutrition, Göteborg University, Sahlgrenska University Hospital, Göteborg, Sweden.
2 Supported by grants from the Swedish Council for Forestry and Agricultural Research (project number 50.0444/98), the Göteborg Medical Society (grant number 00/91), and the Ingabritt and Arne Lundberg Foundation (project number 224/97). 3 Reprints not available. Address correspondence to L Ellegård, Department of Clinical Nutrition, Göteborg University, Sahlgrenska University Hospital, S-413 45 Göteborg, Sweden. E-mail: lasse.ellegard{at}sahlgrenska.se.
ABSTRACT
Background: Oat bran contains soluble fibers, such as ß-glucan, that increase bile acid excretion and thus decrease serum cholesterol. Bile acid synthesis correlates with serum concentrations of the metabolite 7-hydroxy-4-cholesten-3-one (-HC).
Objective: The objective was to investigate whether consumption of ß-glucan from oat bran increases bile acid synthesis, as measured by the serum -HC concentration, within hours after consumption in response to the loss of bile acids from the liver.
Design: In a randomized, single-blind, wheat bran–controlled study with crossover design, 8 subjects were served a controlled diet during 2 periods of 3 d each, with an 11-d washout between the periods. Breakfast included either 75 g extruded oat bran, of which 11 g was ß-glucan, or 75 g wheat bran, of which 1 g was ß-glucan. -HC was measured by HPLC on each day at 0, 12, and 24 h after breakfast and also at 8 h after breakfast on the first day.
Results: After 8 and 12 h of the oat bran diet period, the serum -HC concentration was 84% (P = 0.012) and 92% (P = 0.017) higher, respectively, than that before breakfast. Serum concentrations returned to the baseline value after 24 h. Wheat bran did not influence serum -HC concentrations.
Conclusions: Consumption of ß-glucan from oat bran nearly doubled the serum -HC concentration within 8 h, indicating increased bile acid synthesis. -HC in serum could be used as a marker of increased bile acid excretion induced by the diet.
Key Words: Oat bran • wheat bran • ß-glucan • 7-hydroxy-4-cholesten-3-one • bile acids • cholesterol
INTRODUCTION
Cholesterol metabolism is determined by diet, genetics, cholesterol absorption, and sterol synthesis and excretion. The synthesis and excretion of bile acids is the major pathway of elimination of cholesterol. Primary bile acids are synthesized from cholesterol in the liver, and cholesterol 7--hydroxylase (EC 1.14.13.17) is considered to catalyze the rate-limiting step in the biosynthesis process (1). Hepatic bile acid synthesis is controlled in the liver through negative and positive feedback mechanisms. Bile acids recirculating to the liver regulate their own synthesis by influencing the activity of cholesterol 7--hydroxylase (2–4). Thus, increased bile acid excretion promotes bile acid synthesis from cholesterol (2, 5, 6) through decreased feedback inhibition of cholesterol 7--hydroxylase by bile acids. The consequences of increased bile acid excretion are stimulated cholesterol uptake from the circulation (3, 5) followed by a decreased serum cholesterol concentration (7).
Oat bran has been shown to decrease serum cholesterol (8, 9). Consumption of a diet rich in soluble fiber including oat bran (10) or of a diet supplemented with oat bran (6, 11, 12) has been shown to increase fecal excretion of bile acids.
Short-term dietary intervention studies on sterol excretion in humans are possible by studying subjects who have an ileostomy without resection of the small intestine (13, 14). In such subjects, the transit time of food in the small intestine is short and fairly stable (15). With frequent changes of ileostomy bags and immediate freezing, the ileostomy effluents of subjects with an ileostomy are much less degraded by intestinal bacteria than are the feces of healthy subjects, as shown previously with dietary fiber (13). With this approach, the soluble ß-glucan fraction in oat bran increased bile acid excretion from the small bowel within 24 h after consumption (16). Pretreatment with ß-glucanase (EC 3.2.1.6) counteracted this effect (16). With daily intake of oat fiber, increased bile acid excretion was also found after 17 d (17). Thus, ß-glucan may act in a manner similar to that of cholestyramine (5, 7) and thereby increase bile acid synthesis.
7-Hydroxy-4-cholesten-3-one (-HC) is a metabolite in the synthesis of bile acids that is oxidized from 7-hydroxycholesterol (1). Serum -HC concentrations strongly correlate with the activity of the key enzyme cholesterol 7--hydroxylase (18). -HC is also a reliable marker of bile acid synthesis in humans (18–21).
The purpose of this study was to examine the short-term effect of oat bran on the synthesis of bile acids as measured by -HC. It was hypothesized that consumption of oat bran, which contains ß-glucan, but not consumption of wheat bran with a minimal content of ß-glucan, would increase bile acid synthesis shortly after consumption. If so, measuring -HC as a marker of bile acid synthesis could be an alternative method to sterol balance studies with the ileostomy model. -HC could be used to study how different food components influence cholesterol metabolism through increased bile acid excretion and, thus, increased bile acid synthesis.
SUBJECTS AND METHODS
Subjects
Nine healthy subjects volunteered to participate in the study. One subject suffered from gastrointestinal disturbances during the oat bran period and left the study. Eight subjects (5 women and 3 men) were included in the analysis. Subject data are shown in Table 1. The subjects were healthy, with no signs of anemia, inflammation, or hepatic, renal, or thyroid disease as judged by current standard laboratory tests performed at the Central Laboratory for Clinical Chemistry, Sahlgrenska University Hospital. No prescription drugs were used by any of the subjects during the study. Informed consent was obtained from each participant, and the study protocol was approved by the Ethics Committee of Sahlgrenska University Hospital.
View this table:
TABLE 1 . Clinical data for the 5 women and 3 men who participated in the present study1
Study design
The subjects were served a controlled diet in which all of their food was provided to them. During 2 test periods of 3 consecutive days each, the subjects received either oat bran or wheat bran for breakfast; the 2 test periods were separated by an 11-d washout period. The order of the test periods was randomized. All meals were eaten at fixed times, and the subjects were encouraged to eat all the food served. Breakfast was served in the metabolic ward kitchen at 0715. Lunch and dinner were eaten at work or at home at 1200 and 1800, respectively. Fruit was consumed between the main meals. Fasting blood samples were drawn on each day during the 2 test periods and on the morning after the last day of each test period at 0700. On the first day of each test period, blood samples were collected 8 and 12 h after breakfast, and on the following days, blood samples were collected 12 h after breakfast. The samples were subsequently stored at -20 °C before analysis.
Study diets
The compositions of the controlled basal diet and of 75 g oat bran or wheat bran are shown in Table 2. The diet was composed of ordinary food items and was in accordance with Nordic Nutrition Recommendation (23). In short, this includes a moderate low-fat diet with 30% of energy from fat, 10% of energy from saturated fat, and 25 g total dietary fiber/d. The controlled diet had the same nutrient content on each of the 3 d but was composed of different dishes on different days. Both test periods had the same menu except for the type of breakfast cereal served. The oat bran breakfast consisted of 75 g extruded oat bran with 11.0 g ß-glucan (Avena Group, Kokemäki, Finland). A commercial wheat bran–based breakfast cereal (All Bran Fiber Plus; Kellogg’s AB, Glostrup, Denmark) containing 1.1 g ß-glucan was served as the wheat bran breakfast. The controlled diet was served at 3 energy levels (8.4, 10.0, and 11.7 MJ/d) according to each subject’s individual energy requirements, as predicted from age, sex, body weight, and physical activity. The subjects were not allowed to eat anything except the served food, and no alcohol consumption was allowed during the study. The food was prepared in the metabolic ward kitchen and stored at -20 °C until the day it was consumed. Duplicate portions of the controlled diet from each of the 3 d were freeze-dried and analyzed for energy, nitrogen, total dietary fiber, ß-glucan, cholesterol, and plant sterols.
View this table:
TABLE 2 . Composition of the controlled diet, 75 g oat bran, and 75 g wheat bran1
Analysis
-HC was analyzed by HPLC with a modified version of the method originally described by Axelson et al (19). A 1-mL aliquot of serum was deproteinized by the addition of 1 mL acetonitrile and was centrifuged for 10 min at 2000 x g at 20 °C. The supernatant fluid was mixed with 1 mL phosphate buffer (pH 10.5) and centrifuged again. Solid-phase columns were rinsed with 2 mL each of the solvents hexane, isopropanol, and methanol and were then washed with water. The sample was applied at a flow rate of 1–2 mL/min, followed by 5 mL water and 5 mL methanol:water (70:30 by vol). The columns were dried by centrifugation for 10 min at 2000 x g at 20 °C, and the samples were then eluted with 5 mL dichloromethane:hexane (10:90 by vol). Before the measurement of -HC, the samples were dried under nitrogen and redissolved in 100 µL mobile phase [4% propanol (by vol) in hexane]. A 20-µL aliquot of the sample was injected at a flow rate of 0.5 mL/min on a Kromasil silica column (5 µm, 150 x 3 mm; Sorbent AB, Göteborg, Sweden) with a Kromasil silica FL precolumn (5 µm, 10 x 3 mm; Sorbent AB). -HC was measured with the use of a Gynkotek HPLC system (Kovalent AB, Göteborg, Sweden) featuring pump model M480 G, a Gina 50 injector, Chromeleon software version 3.14, and a UVD 340 S ultraviolet detector; detection was at 240 nm. Recovery was determined by counting the amount of radioactively labeled internal standard, [3H]25-hydroxyvitamin D3 (Amersham International, Buckinghamshire, United Kingdom), in 20 µL sample with OptiPhase Hisafe 2 scintillation solution (Wallace AB, Stockholm) on a Tricarb TR 1900 Liquid Scintillation Analyzer (Packard, Stockholm). Standard curves were prepared by dilution of stock -HC, which was a kind gift of I Björkhem, Department of Clinical Chemistry, Karolinska Hospital, Huddinge, Sweden. All the chemicals were analytic grade, and most were purchased from Merck Eurolab, Stockholm. All analyses were carried out in duplicate, with a CV of 6.5%.
Serum lathosterol was determined as trimethylsilyl derivatives by gas-liquid chromatography on a 0.32 mm x 50 m capillary column, as proposed by Miettinen (24). Lathosterol was expressed as µmol/100 mmol cholesterol, which was simultaneously derived during analysis. The CV of duplicate samples was 4.8%.
Energy content was determined by combustion in a Gallenkamp adiabatic bomb calorimeter (Loughborough, Leicestershire, United Kingdom). Nitrogen content was determined with a modified micro-Kjeldahl method after acid digestion in a Tecator T40 Digestor (Tecator AB, Höganäs, Sweden) and spectrophotometric determination of ammonia in a Technicon Auto Analyzer (method N-3b; Technicon AB, Stockholm) (13). The CV of duplicate samples was 1.6%. Total dietary fiber was measured by gravimetry after enzymatic hydrolysis (25), and the CV of duplicate samples was 4.3%. Cholesterol and plant sterols were determined by gas-liquid chromatography after both acid and alkaline hydrolysis and silylation (26). The CV of duplicate samples was 4.7%.
Statistical analyses
Data are presented as medians and ranges. Comparisons between the wheat bran diet and the oat bran diet, the effects of time and subject, and the interaction between diet and time were assessed by univariate analysis of variance on the ranked data because of the small number of subjects and the uneven distribution of the data. Comparisons between values before and after each diet, as well as post hoc analyses of differences between the diets, were performed by Wilcoxon’s signed rank test. Trends in lathosterol or cholesterol concentrations were evaluated by regression analysis. Calculations were performed by using SYSTAT for WINDOWS version 7.0.1 (SPSS Inc, Chicago) and SPSS version 10.0.5 (SPSS Inc). A P value < 0.05 was considered significant. This study was designed to be able to detect a 50% increase in -HC, with a power of 90% and a significance level of 0.05 (27). This design was based partly on earlier results with the ileostomy model, in which the same dose of ß-glucan increased bile acid excretion by 50% (16), and partly on an individual day-to-day variation in -HC concentration of 28–32% in 2 pilot investigations conducted in our laboratory.
RESULTS
All analyses were based on 8 subjects, except for those at 60 and 72 h during the wheat bran diet period, which were based on only 7 subjects. This is because one subject left the study after 48 h of the wheat bran diet period because of influenza. The subjects maintained their body weight during the study. Serum -HC concentrations at baseline and during the wheat bran and oat bran diet periods are shown in Figure 1. Serum cholesterol concentrations and ratios of serum lathosterol and cholesterol concentrations at baseline and during the wheat bran and oat bran diet periods are shown in Table 3. The median -HC concentration at baseline was 39.2 nmol/L (range: 13.8–79.2 nmol/L). The -HC concentration during the oat bran diet period was significantly higher at 8 h (P = 0.012) and 12 h (P = 0.017) after breakfast than at baseline. The increase in -HC concentration 12 h after breakfast on days 2 and 3 was also significant (P = 0.036 and 0.017, respectively). No significant increase in -HC concentration occurred during the wheat bran diet period.
FIGURE 1. . Median serum concentrations and ranges of 7-hydroxy-4-cholesten-3-one (-HC) in 8 subjects during the oat bran () and wheat bran () diet periods. The times at which the bran breakfast was given are indicated by arrows. As assessed by ANOVA on the ranked data, there was a significant difference between the diets (P < 0.001) and a significant time x diet interaction (P = 0.013).
View this table:
TABLE 3 . Serum cholesterol concentrations and ratios of lathosterol to cholesterol at baseline and after ingestion of oat bran or wheat bran breakfasts1
Univariate analysis of variance showed significant differences in -HC concentrations between the 2 diets (P < 0.001), with a significant interaction between diet and time (P = 0.013). In comparison with the change in serum -HC concentration during the wheat bran diet period, the serum -HC concentration increased 84% (P = 0.012), 92% (P = 0.017), 154% (P = 0.018), and 127% (P = 0.018) after 8, 12, 36, and 60 h, respectively, of the oat bran diet period. There were no significant differences in fasting -HC concentrations either between different days within each diet period (P = 0.94 for time) or between the oat bran and wheat bran diet periods (P = 0.86 for time x diet interaction). Prolonged ingestion of the same amount of extruded oat bran for another 7 d by one subject did not affect fasting -HC concentrations. The within-subject day-to-day variations in fasting -HC concentrations were 31% and 37% for the wheat bran and oat bran diet periods, respectively. The average between-subject variation (CV) in fasting -HC concentrations was 51%. The between-subject variation in postprandial -HC concentrations was similar at 8 and 12 h after ingestion of the oat bran breakfast (ie, 49% and 48%, respectively). Serum cholesterol concentrations, as measured by gas-liquid chromatography in the lathosterol analysis, decreased nominally during both the wheat bran and oat bran diet periods, but no significant differences were detected over time (P = 0.069) or between the 2 diet periods (P = 0.329). No significant differences in the ratio of lathosterol to cholesterol were detected over time (P = 0.414), and the time x diet interaction was also not significant (P = 0.465). However, there was a significant diet effect (P = 0.002): the ratios were significantly lower during the wheat bran diet period than during the oat bran diet period. Flatulence increased in 4 of the 8 subjects during the wheat bran diet period, whereas all of the 8 subjects had increased flatulence during the oat bran diet period.
DISCUSSION
The present study showed that consumption of 75 g extruded oat bran along with a controlled diet significantly increased serum -HC concentrations within 8 h in healthy subjects, whereas consumption of 75 g wheat bran along with the same diet did not. Within 24 h the -HC concentration returned to its initial value. Thus, if only the fasting -HC concentration had been measured, this rapid effect of oat bran on sterol metabolism would not have been found. Served daily at breakfast for 3 d, extruded oat bran had the same effect on -HC concentrations on each of the 3 d, without an additive effect. Extruded oat bran contains 15% ß-glucan. Thus, 75 g extruded oat bran, containing 11 g ß-glucan, corresponds to the concentration that has been shown to increase bile acid excretion 50% in ileostomy studies (16, 17). The ß-glucan content of Swedish rolled oats and oat bran is 4% and 8–9%, respectively (28). A daily intake of 3 g soluble oat fiber was shown in a recent meta-analysis to significantly decrease serum cholesterol (9). The dose of 11 g ß-glucan consumed daily in a single meal in the present study corresponds to 6 normal servings of oatmeal porridge. The relatively high total fiber content of the diet in the present study may explain the increased flatulence and other gastrointestinal disturbances that were observed.
It should be emphasized that there are 2 bile acid synthetic pathways. Bile acid synthesis occurs in the liver primarily via cholesterol 7--hydroxylase. An alternative pathway in humans begins with 27-hydroxylation of cholesterol via sterol 27-hydroxylase. Some reports indicate that the initial steps in this pathway take place not only in the liver but also in vascular endothelium and macrophages (29–31). Metabolites are released in serum and taken up by the liver for conversion to bile acids (32–34). Because a portion of bile acids can be formed via this pathway, thus bypassing cholesterol 7--hydroxylase as the rate-limiting step, measured concentrations of -HC do not necessarily reflect the total production of bile acids in humans under all conditions. However, this alternative pathway seems to be of minor importance, because, on average, only 9% (range: 3–18%) of total bile acid synthesis takes place via this pathway in healthy humans (35). Studies of serum concentrations of intermediates of both pathways have suggested that when the classical pathway is suppressed, the alternative pathway may become dominant (34). In addition, the alternative pathway may be less responsive to negative feedback inhibition than is the classical pathway (34, 36, 37). Intake of ß-glucan increases bile acid excretion and, subsequently, bile acid synthesis. This results in less feedback inhibition of cholesterol 7--hydroxylase. Under such conditions, it is possible that the relative contribution of the alternative pathway decreases and that the relative contribution of the classical pathway increases.
The median serum -HC concentration was in the range previously reported in healthy subjects (19), and the interindividual variation was, as expected, higher than the intraindividual variation. We found that serum -HC concentrations in the morning did not differ significantly from those in either the afternoon or the evening (ie, at 8 or 12 h) during the wheat bran diet period. However, bile acid synthesis has been reported to have a circadian rhythm in humans and has been shown to be independent of feedback inhibition (38). A small study on 3 subjects with frequent blood sampling showed a tendency toward a circadian rhythm for -HC concentration (39).
Serum lathosterol concentrations in humans have been shown to correlate with hepatic ß-hydroxy-ß-methylglutaryl coenzyme A reductase (EC 1.1.1.88) activity and thus with cholesterol synthesis (40). Furthermore, the serum concentration of lathosterol has previously been shown to be an indicator of cholesterol synthesis, with a Pearson correlation with cholesterol balance of 0.70, indicating that 50% of the variation in cholesterol balance is accounted for by changes in serum lathosterol concentration (41). Thus, the ratio of lathosterol to cholesterol in humans has been shown to correlate with cholesterol synthesis (41).
In the present study, the ratio of lathosterol to cholesterol was measured to study the hepatic response in subjects with an ileostomy to an earlier observed increase in bile acid excretion with oat fiber. No change in the ratio of lathosterol to cholesterol was observed during the oat bran diet period. This observation is in accordance with results from an ileostomy study in which no marked postprandial increase in cholesterol synthesis, measured as changes in lathosterol concentration, was observed 7 h after administration of an oat bran test meal (42). The nominal, but not significant, decrease in serum cholesterol during the first 3 d of both the oat bran and wheat bran diet periods may have been the effect of the controlled diet, which was low in saturated fat and cholesterol and high in dietary fiber and plant sterols.
In conclusion, the present study shows that ß-glucan from oat bran almost doubles the serum concentration of -HC shortly after intake, indicating increased bile acid synthesis. Measurement of this metabolite may therefore be useful even in single-meal experiments to study whether different foods influence cholesterol metabolism through increased bile acid synthesis.
ACKNOWLEDGMENTS
We thank Lena Knutsson and Mitra Ravand of the Department of Clinical Nutrition, Sahlgrenska University Hospital, Göteborg University, Göteborg, Sweden, and Homeira Nateghi for their assistance with the analysis of -HC, lathosterol, and cholesterol and Per Åman of the Department of Food Science, Swedish University of Agricultural Sciences, Uppsala, Sweden, for ß-glucan analysis.
REFERENCES