Changes in content and composition of dietary fiber in yellow onions and red delicious apples during commercial storage.

Changes in pectin composition and solubility are part of the softening process in apples during ripening and postharvest storage. Lignification may also occur with long-term storage. In the United States, apples and onions are harvested once yearly and then stored and marketed for the next 12 months. The changes that occur in the dietary fiber content and composition in Red Delicious apples and yellow Spanish onions during storage were studied, and the loss of fiber in peeled apples was determined. Dietary fiber was extracted by the enzymatic-chemical method of Theander and Westerlund. Storage had no effect on total or insoluble fiber content of apples; Klason lignin concentration was greater in samples stored for 12 months than in those stored for 0, 4, or 8 months. Peeling reduced apple fiber concentration about 25% by decreasing neutral and acidic sugars and Klason lignin in the insoluble fraction. The total fiber content of onions increased with storage, primarily by increasing the insoluble fiber content of uronic acids. The results suggest that the standardized, environmentally controlled storage of apples, as used in Washington State, has little effect on dietary fiber content. In contrast, the less rigorously controlled storage conditions for yellow Spanish onions increases the insoluble fiber fraction and uronic acid content.

An unfermented gel component of psyllium seed husk promotes laxation as a lubricant in humans.

BACKGROUND: In addition to increasing stool weight, supplements of psyllium seed husk produce stools that are slick and gelatinous. OBJECTIVE: Our purpose was to test the hypothesis that a gel-forming fraction of psyllium escapes microbial fermentation and is responsible for the characteristics that enhance laxation. DESIGN: Fifteen healthy adults consumed controlled diets for two 7-d periods, one of which included 8.8 g dietary fiber provided by 15 g/d of a psyllium seed husk preparation. All stools were collected and evaluated and diet was monitored throughout. RESULTS: Psyllium significantly increased the apparent viscosity of an aqueous stool extract, stool moisture, and wet and dry stool weights. A very viscous fraction, not present in low-fiber stool and containing predominantly 2 sugars that are also found in abundance in psyllium husk, was isolated from psyllium stool. CONCLUSIONS: In contrast with other viscous fibers that are fermented completely in the colon, a component of psyllium is not fermented. This gel provided lubrication that facilitated propulsion of colon contents and produced a stool that was bulkier and more moist than were stools resulting with use of comparable amounts of other bowel-regulating fiber sources.

Plant residue and bacteria as bases for increased stool weight accompanying consumption of higher dietary fiber diets.

OBJECTIVE: Stool diluting effects of relatively inert material, such as unfermentable dietary fiber, has been proposed as an effect of fiber beneficial to the colon. Stool dilution by increasing bacterial mass may be beneficial or deleterious, depending on bacterial metabolic products. The purpose of this study was to determine the basis for stool weight when two stepwise increases of fiber from all classes of fiber-containing foods were consumed. METHODS: Stool from five men consuming three constant diets containing 15, 30 and 42 g/d of dietary fiber were fractionated into plant material and bacteria and analyzed for neutral and amino sugar content. Fecal nitrogen, fat and ash were measured. RESULTS: Daily gravimetric yield and sugar content of the plant fraction from stool increased with each fiber addition. Compared to the low fiber diet, the medium fiber diet decreased the concentration of the bacterial mass in wet stool by 11% and the high fiber diet by an additional 32%. The high fiber diet decreased stool fat concentration; the medium and high fiber diets decreased stool nitrogen concentration to the same extent. Apparent digestibility of plant-derived neutral sugars decreased with each fiber addition. CONCLUSIONS: Inherently less fermentable plant material modulates the colon environment in three beneficial ways: it is a relatively unreactive diluent of lumenal contents; it adds mass to promote distal movement of waste; it does not promote a large bacterial mass.

In vitro fermentation of swine ileal digesta containing oat bran dietary fiber by rat cecal inocula adapted to the test fiber increases propionate production but fermentation of wheat bran ileal digesta does not produce more butyrate.

This experiment evaluated three hypotheses: i) production of propionate is increased during fermentation of substrate containing oat bran (OB)(6); ii) production of butyrate is increased during fermentation of substrate containing wheat bran (WB) and iii) results of in vitro fermentations using physiological substrates and inocula agree with in vivo data. Ileal digesta collected from swine fed OB and WB were the substrates. Digesta was fermented for 0-96 h in an anaerobic in vitro system using inocula prepared from ceca of rats fed the same fiber sources. Carbohydrate and short-chain fatty acid (SCFA) contents in the fermentations were measured by gas chromatography. Fermentation of WB digesta did not produce more n-butyrate (P > 0.05) and was significantly slower (P < 0.05) than fermentation of OB digesta. OB digesta fermentation produced a significantly greater (P < 0.05) molar proportion of SCFA as propionate. Bacterial mass increased more and was maintained longer during fermentation of OB digesta than the WB digesta. Our results indicate that dilution of undigested WB fiber and not n-butyrate production is one mechanism by which WB may protect colonic mucosa; propionate production is increased during fermentation of beta-glucan in OB; and an in vitro system using physiological sources of inoculum and substrate containing WB and OB yields results that agree with in vivo findings in humans and rats.

Increasing amounts of dietary fiber provided by foods normalizes physiologic response of the large bowel without altering calcium balance or fecal steroid excretion.

Nine healthy, young men consumed constant diets to determine selected large-bowel, serum cholesterol and triacylglycerol, and calcium balance responses to 3 amounts of fiber provided by a mixture of fruit, vegetables, and grains. The diets, each consumed for 1 mo, contained 16, 30, and 42 g total fiber/d, of which 2.9, 4.8, and 7.7 g, respectively, was soluble. Mean daily wet and dry stool weights increased with each fiber addition. The first fiber addition increased defecation frequency and decreased fecal pH, bile acid concentration, and neutral steroid concentration; the second addition had no further effect. Mean weight of each defecation and stool moisture did not increase and serum cholesterol and triacylglycerol concentrations, calcium balance, and gastrointestinal transit time did not decrease as fiber intake increased. We conclude that 1) fiber provided by a mixed-food diet increases stool weight as effectively as does wheat or oat bran; 2) even high amounts of dietary fiber do not change transit time or defecation frequency if they are already approximately 1 and 2-3 d, respectively; 3) food patterns consistent with the food pyramid and incorporating legumes and whole grains are necessary to achieve recommended fiber intakes of 20-35 g/d, even if energy intake is > 12.55 MJ (3000 kcal); 4) soluble fiber provided by a mixture of whole foods has no effect on serum cholesterol concentrations or output of fecal bile acids; and 5) mixed-food fiber has little effect on calcium balance when calcium intakes are high (> or = 1.5 g/d).

Mechanisms by which wheat bran and oat bran increase stool weight in humans.

Generally, stool weight is significantly increased by adding sources of insoluble fiber to the diet. Comparable amounts of fiber provided by wheat and oat brans have the same effect on daily stool output, even though > 90% of wheat bran fiber but only 50-60% of oat bran fiber is insoluble. To determine the bases for these increases in stool weight, stool samples collected from 5 men in 2 constant diet studies that determined the effects of wheat and oat brans on large-bowel physiology were fractionated by using a physicochemical procedure into plant, bacterial, and soluble fractions, which were weighed and analyzed for sugar content and composition. Nitrogen, crude fat, and ash outputs were also determined. Wheat bran increased the fecal concentration of sugars and mass of plant material more than did oat bran, whereas oat bran increased fecal bacterial mass more. Each fiber source increased nitrogen, ash, and fat excretion, but excretion of fat was greater with oat bran. The apparent digestibility of plant-derived neutral sugars decreased significantly when wheat but not oat bran was consumed. The apparent digestibility of neutral sugars provided by wheat bran was 56%; the apparent digestibility of those provided by oat bran was 96%. We conclude that bacteria and lipids are major contributors to the increase in stool weight with oat bran consumption, whereas undigested plant fiber is responsible for much of the increase in stool weight with wheat bran consumption. Results are consistent with the hypothesis that oat bran increases stool weight by providing rapidly fermented soluble fiber in the proximal colon for bacterial growth, which is sustained until excretion by fermentation of the insoluble fiber.

Database and quick methods of assessing typical dietary fiber intakes using data for 228 commonly consumed foods.

To promote assessment of dietary fiber intakes in clinical settings, we established two objectives for this study: to provide a detailed database in grams per serving of fiber content and polymer composition for most fiber sources in the US diet, and to develop a quick method for estimating total fiber intakes. Data for 342 foods were condensed to 228 foods by combining similar foods. The comprehensive database developed includes pectin, hemicelluloses, and beta-glucan contents of the soluble and insoluble fractions of fiber and the cellulose and Klason lignin contents of the insoluble fiber. Three fourths of the 228 foods contained 2.0 g fiber per serving or less; only 10% contained more than 3.0 g per serving. The quick method consists of multiplying the number of servings in each food group by the mean total dietary fiber content of foods in that group: 1.5 g for fruits (n = 43), 1.5 g for vegetables (n = 68), 1.0 g for refined grains (n = 80), and 2.5 g for whole grains (n = 13). Actual fiber values from the database should be used in the quick method if foodstuffs concentrated from grains, legumes, and nuts and seeds are consumed. Sample menus demonstrate that quick assessment of total fiber intake yielded results similar to the sum of individual values from the database.

Fermentation of carbohydrate in rat ileal excreta is enhanced with cecal inocula compared with fecal inocula.

The differential fermentative capacities of microflora from two regions of the large bowel and how fermentation was altered by prior exposure of the microflora to the substrate to be fermented were studied using an in vitro fermentation system. Bacterial inocula were prepared from cecal contents and feces from three groups of rats fed purified diets containing 100 g/kg dietary fiber from canned peas or psyllium seed husk, or a nonpurified diet containing 170 g/kg dietary fiber. The substrate for all fermentations was ileal excreta from colectomized rats fed a purified diet containing 100 g/kg dietary fiber from canned peas. Anaerobic fermentations were conducted for 0, 3, 6, 12, 24, 48 and 96 h. Sugars of the unfermented polysaccharides were measured by gas chromatography following acid hydrolysis; disappearance was the measure of fermentation. Independent of inoculum source, > 90% of the starch and arabinose and 75% of the uronic acids, but < 30% of non-starch glucose (the measure of cellulose), were fermented by 24 h. Cecal inocula fermented arabinose and uronic acids more quickly (P < 0.05) and fermented more (P < 0.05) non-starch glucose than fecal inocula. Inocula adapted to psyllium seed husk fermented starch faster (P < 0.05) and non-starch glucose, arabinose and uronic acids more slowly (P < 0.05) than inocula adapted to peas or nonpurified diet. Bacterial efficiency of carbohydrate utilization, the increase in muramic acid/mole carbohydrate fermented, was greatest (P < 0.05) with cecal inocula adapted to peas. We conclude that using cecal microflora as the inoculum source provides a more accurate index of fermentation during transit through the large bowel and that noncellulosic and storage polysaccharides of the plant cell wall are utilized before cellulose.

Rat cecal inocula produce different patterns of short-chain fatty acids than fecal inocula in in vitro fermentations.

The effects of bacterial collection site, bacterial adaption to substrate and duration of fermentation on short-chain fatty acid (SCFA) production were studied using an in vitro fermentation system. Cecal and fecal inocula from rats fed purified diets containing 100 g/kg dietary fiber from canned peas or psyllium seed husk, or a nonpurified diet containing 170 g/kg dietary fiber fermented ileal excreta from colectomized rats fed a purified diet containing 100 g/kg dietary fiber from canned peas. The SCFA concentration, measured by gas chromatography, in anaerobic fermentations of 0, 3, 6, 12, 24, 48, 72 and 96 h, increased (P < 0.05) through 72 h. Compared with fecal inocula, cecal inocula produced SCFA at greater (P < 0.05) initial rates, more (P < 0.05) total SCFA and a greater (P < 0.05) proportion as n-butyrate. Inocula from rats fed the pea or nonpurified diets produced SCFA at greater (P < 0.05) molar proportions as acetate and less (P < 0.05) as propionate than inocula from rats fed psyllium seed husk at most time points. We conclude that collection site, adaptation of bacteria to the substrate to be fermented and duration of fermentation significantly influence the results of in vitro fermentation studies.

Mechanism of serum cholesterol reduction by oat bran.

Nine normolipidemic young men consumed a constant diet for 2 mo into which oat bran was incorporated during the second month so that we might test the hypotheses that oats lower serum cholesterol concentrations by decreasing bile acid and fat absorption and increasing bile acid synthesis. Bile acid kinetics were determined by measuring the 13C enrichment of serum cholic and chenodeoxycholic acids. Oat bran consumption decreased serum cholesterol levels (p < 0.01) and cholic acid pool size (p < 0.05). Deoxycholic acid pool size (p < 0.01) and the synthesis and fractional turnover rates of both primary bile acids (p < 0.05) increased. Total bile acid pool size did not change. Fecal excretion of total bile acids, the two secondary bile acids and fat increased significantly. The results demonstrate that oat bran lowers serum cholesterol levels in part by altering bile acid metabolism. In addition, the substantial increase in the proportion of the total bile acid pool that was deoxycholic acid is consistent with the hypothesis that oat bran also decreases cholesterol synthesis.

Mucin secretion in germfree rats fed fiber-free and psyllium diets and bacterial mass and carbohydrate fermentation after colonization.

The effect of psyllium on mucin secretion was determined by comparing water-soluble and -insoluble fractions of excreta from germfree rats fed a fiber-free (FF) diet or a diet containing psyllium seed husk (PS). Excreta from the same rats after colonization with a rat mixed cecal culture were separated into water-soluble, plant, and bacterial fractions to compare the remaining carbohydrate and the mass of bacteria. The sugar composition and water solubility of carbohydrate in excreta from germfree rats fed FF diets indicated that a primary fermentable substrate was mucin. PS increased fecal excretion of mucin-derived sugars almost threefold in germfree rats. Fecal carbohydrate was reduced from 619 to 237 mumol/g of dry feces and mostly in the bacterial fraction when rats fed an FF diet were colonized. The total sugar content and the amount of muramic acid, but not bacterial counts and mass, indicated that PS increased fecal bacteria. Fractionation of excreta from PS-fed rats was complicated by a gel which, based on sugar composition, was PS. Sugar composition of the water-soluble fraction from excreta from PS-fed rats suggested that it contained some bacterial component, possibly exopolysaccharides and some of the PS, but not mucin. PS digestibility ranged from 60 to 80%, depending on what fecal fraction was used for output. Because of the presence of PS-derived sugars in the gel and soluble fraction, it was not possible to determine which, if any, of the PS digestibilities was correct.

Determination of fermentable carbohydrate from the upper gastrointestinal tract by using colectomized rats.

The primary aim of this study was to characterize the carbohydrate that would be supplied to the colon for fermentation under physiological conditions. Colectomized rats were fed fiber-free diets or diets containing 5% (wt/wt) gum arabic. Four (fucose, galactose, glucosamine, and galactosamine) of 11 analyzed sugars accounted for 77% of the total sugar in ileal excreta from colectomized rats fed fiber-free diets. The three sugars in gum arabic, rhamnose, arabinose, and galactose, accounted for 84% of the total sugars in gum arabic ileal excreta. Comparisons of the sugar compositions of the ileal excreta, the water-soluble fractions of the excreta, and three gel filtration fractions of the water-soluble material with those of the water-soluble fraction of rat mucosa, the acetone-soluble fraction of pancreas, and pancreatin suggested that the major source of endogenous carbohydrate is mucin. Gum arabic increased the daily excretion of the four mucin-derived sugars (fucose, galactose, glucosamine, and galactosamine) by the colectomized rats from 473 mumol per day to 634 mumol per day. We conclude that mucin is the major endogenous carbohydrate excreted from the upper gut and that gum arabic increases the amount of this endogenous carbohydrate.

Content and composition of dietary fiber in 117 frequently consumed foods.

Twenty-three fruits, 33 vegetables, 41 grain products, 7 legumes, 4 nuts, and 9 miscellaneous foods were analyzed by an accurate chemical method to determine their dietary fiber content and composition. The mean (+/- standard deviation) dietary fiber content of fruits was 1.4 +/- 0.7 g/100 g (fresh weight); of vegetables, 2.0 +/- 0.8 g; of 32 refined grains (less than 5% fiber), 2.3 +/- 1.0 g; of legumes, 4.0 +/- 0.7 g; and of nuts, 6.4 +/- 2.1 g; the dietary fiber content of nine higher-fiber grains (greater than 5%) was variable. The soluble fiber fraction averaged 23% of the total fiber in refined grains, 3% in nuts, and 13% to 20% in the other food groups. Dietary fiber composition of every food group was heterogenous. Pectin, which was negligible in grains, constituted approximately 15% to 30% of the fiber in fruits, vegetables, legumes, and nuts. Hemicelluloses composed about half of the total fiber in grains, and approximately 25% to 35% of total fiber in other foods. Cellulose was one third or less of the total fiber in most foods, except for legumes, in which it was about one half. Values for total dietary fiber content generally agree with those reported previously. The soluble fiber fraction was lower than what has been reported because the distribution of total fiber between the soluble and insoluble fractions is determined by the method of analysis. The analyses used in this study demonstrated that the concentration of dietary fiber in many frequently consumed foods is 1% to 3%. The generally similar fiber concentrations of food within a group--fruits, vegetables, refined grains, and legumes--suggest that an average value for the fiber concentration in that group can be used to rank food intakes and histories into low, medium, or high dietary fiber contents.

Starch bioavailability in the upper gastrointestinal tract of colectomized rats.

Colectomized rats were used to compare the bioavailability of starch from canned peas, kidney beans, lima beans, corn, cooked rice and AIN-76A purified diet. Postoperative weight gains of colectomized and sham-operated rats were not different. Test meals of approximately 2 and approximately 3.3 g (dry weight) of each food containing chromic oxide (0.6 g/100 g) were administered to 4-6 rats by gastric intubation. Chromium recovery in ileal digesta 10 h after the meal ranged from 94 to 102%. Significant starch (11-15%) from peas, lima beans or kidney beans was recovered in the ileal digesta; 0.2-0.4% of starch from rice, corn or AIN-76A was in the ileal digesta. Pretreatment of kidney or lima beans with a heat-stable endo-alpha-amylase decreased the starch recovered in the ileal digesta to 2.7-4.3% of that fed; pretreatment with the amylase and pepsin had no further effect. Oligosaccharide extraction, the size of the test meal and the amount of starch did not affect starch bioavailability.

Simultaneous determination of neutral and amino sugars in biological materials.

A method is described for the simultaneous analysis of nine neutral and three amino sugars. Mixtures of standard sugars and biological samples were acid hydrolyzed with a two step Saeman procedure, neutralized with BaCO3, reduced with sodium borohydride, acidified, evaporated and alditol acetates prepared. Baseline resolution was achieved on a glass-capillary SP-2340 column in ca. 52 min. Reproducibility, response factors and hydrolysis losses were determined. Quantitation was linear over the range of 10-20 micrograms/ml to 2000 micrograms/ml. Conditions were defined for the reproducible quantitation of muramic acid.

Dietary fiber and long-term large bowel response in enterally nourished nonambulatory profoundly retarded youth.

The objective of this year-long research was to determine the effects of three levels of a purified dietary fiber source, soy polysaccharide, on the long-term bowel function of 11 youth, aged 7 to 17 years, who were nonambulatory, profoundly disabled, constipated, and fed by gastrostomy. From day 1 to 60 the diet was a fiber-free enteral feeding; from day 61 to 300, 12 to 20 g/day total fiber was added to the diet; and from day 301 to 360, 18 to 25 g/day. Complete stool collections were made from day 51 to 60, day 111 to 120, day 291 to 300, day 321 to 330, and day 351 to 360. The second fiber addition significantly increased mean (+/- SD) daily stool frequency from 0.6 +/- 0.2 during the fiber-free diet to 1.1 +/- 0.5. The first addition of fiber compared to the fiber-free diet significantly increased stool moisture from 70 +/- 7% to 76 +/- 8% and wet stool weight from 30 +/- 13 g/day to 53 +/- 21; mean stool weight during days 351 to 360 was 87 +/- 45 g/day. Daily dry stool weight significantly increased with the second fiber addition. Soy polysaccharide fiber improved bowel function in this nonambulatory profoundly disabled population.

Neutral sugar composition and gravimetric yield of plant and bacterial fractions of feces.

Separating dietary fiber from other polysaccharides in digesta and feces is necessary to understand its mechanisms of action. A gravimetric method that separates fecal plant and bacterial matter based on size and density was evaluated and modified to determine the plant and bacterial mass of lyophilized whole and blended rat and human feces. Three screen mech combinations (150 and 75 microns, 150 and 35 microns, 35 microns) were used with rat feces. Filtration of a homogenized rat fecal slurry sequentially through 150- and 35-microns-mesh screens versus 150- and 75-microns-mesh screens decreased the gravimetric recovery of bacteria from congruent to 35 to congruent to 25% of fecal dry weight and increased the plant fraction weight. Neutral sugar composition, determined by gas chromatography of alditol acetates, and bacterial counts of the fractions suggested that the decreased yield of bacterial fraction represented removal of plant material and not a loss of bacteria. Rat excreta contained 29.5% (dry weight) total neutral sugar, 88% of which was recovered in the plant material. Human feces containing wheat bran, fractionated with the 150- and 35-microns-mesh screens, was 21% neutral sugar, congruent to 65% of which was in the plant fraction. The plant fractions had more xylose and arabinose and less glucose than the bacterial fractions. Processing samples in a Waring blender had no adverse effect on the rat or human fecal bacterial counts. The use of this gravimetric method in combination with the sugar analysis of the fractions provided a better measure of plant and bacteria than only gravimetric yield.