Method of Making Soluble Dietary Fiber Compositions From Cereals

1. Field of the Invention

This invention relates to the conversion of milled cereal products to soluble dietary fiber compositions which are useful in a variety of foods as sources of soluble dietary fiber.

Dietary fiber is considered to be the soluble and insoluble components of food that are not digested by enzymes in the human gastrointestinal tract. The primary sources of dietary fiber include such cell wall materials as cellulose, hemicelluloses, lignin, and pectins, along with gums and mucilages. Dietary fiber has been considered an important food component since early times. Recently, Burkitt et al. [Lancet 2: 1408-1411 (1972)] concluded that dietary fiber has a role in the prevention of certain large-intestine diseases, including cancer of the colon and diverticulitis. Diets containing large amounts of dietary fiber lead to stools that are softer and larger, and bowel movements are generally more frequent. Burkitt also mentioned that the serum cholesterol rises when dietary fiber is removed from the diet, and that eating a fiber-rich diet lowers serum cholesterol. Trowell [Am. J. Clin. Nutr. 25: 464-465 (1972) reached a similar conclusion regarding the relationship between fiber and health benefits.

It is now known that not all dietary fiber is the same and that different fibers provide different health benefits. For example, wheat bran is very rich in insoluble dietary fiber (mainly cellulose and hemicelluloses) and is excellent for decreasing the transit time of food through the digestive tract [Anderson et al., Am. J. Clin. Nutr. 32: 346-363 (1979)]. Some fiber are reported to reduce total plasma cholesterol [Munoz et al., Am. J. Clin. Nutr. 32: 580-592 (1979)].

2. Description of the Prior Art

In recent years, considerable attention has been given to oats as a source of dietary fiber. It is the soluble fiber component that is effective in lowering serum cholesterol levels. Oatmeal, or rolled oats, and especially oat bran are the best sources of this soluble fiber. The first indication of serum cholesterol lowering by rolled oats was observed in rats by Degroot et al. [Lancet 2: 303-304 (1963)]. Fisher et al. [Pro. Soc. Exp. Biol. Med. 126: 108-111 (1967)] report that the fiber fraction of oats is responsible for its unique effects on cholesterol. Over the years, numerous experiments with animals have shown that oat fiber has a strong hypocholesterolemic effect. Anderson et al. [Am. J. Clin. Nutr. 34: 824-829 (1981); 40: 1146-1155 (1984)] have confirmed hypocholesterolemic effects of oats in humans. Oat fiber also reduces the amount of low-density lipoprotein (LDL) without lowering the beneficial high-density lipoprotein (HDL). In fact, Anderson et al. [66th Annual Meeting, Am. Assoc. Cer. Chem., Abstr. No. 112 (1981)] teach that oat bran fed to humans can reduce LDL 58% while increasing HDL 82%. Other water-soluble fibers, such as pectin and guar gum, can lower serum cholesterol, but they are frequently accompanied by undesirable side effects such as nausea and vomiting. The results of another study by Anderson et al. (supra 1984) indicate that oat bran diets decrease total serum cholesterol 19% and LDL 23% and that oat bran increases bile acid excretion 65%. These studies clearly document the hypocholesterolemic effects in humans of oat products which are rich in soluble fiber.

Maltodextrins are successful commercial products prepared by hydrolysis of pure starch. The wet milling of corn is the primary source of starches from which maltodextrins are obtained. Other sources of starch for commercial products are tapioca, potato, and rice. Cereal flours are generally not used for maltodextrin production because of the flavors and colors that develop during processing [Katz, Cer. Foods World, 31: 866-867 (1986)]. Also, when cereal flours are treated with amylases, a colloidal hydrolysate is obtained that is nonfilterable (U.S. Pat. No. 4,894,242). However, some breakfast and related food products have been produced in which the characteristic cereal flavor is desired. Whole cereal flours have been subjected to starch-hydrolyzing conditions and have yielded, for example, a whole-grain hydrolyzed product [Conrad, U.S. Pat. No. 4,377,602] and a ready-to-eat, enzyme-saccharified cereal [Fulger et al., U.S. Pat. No. 4,710,286]. Ronai [U.S. Pat. No. 3,640,729] arrives at a similar product by adding prehydrolyzed starch to oat flour to yield an instant oat cereal product.

I have now surprisingly discovered a process for converting milled cereal products to soluble dietary fiber compositions by enzymatic hydrolysis with .alpha.-amylase and recovery of the soluble fraction. These compositions are colorless and devoid of inherent undesirable cereal flavors.

In accordance with this discovery, it is an object of the invention to provide a novel source of water-soluble dietary fiber compositions for incorporation into ingestible formulations.

It is also an object of the invention to provide a novel protein-rich byproduct useful as an ingredient in food or animal feed compositions.

Other objects and advantages of this invention will become readily apparent from the ensuring description.

Suitable substrates contemplated for use in the invention include cereal flours, milled cereal brans, cereal starches, tuber starches, and blends thereof. Of particular interest are the whole flours of barley, oats, wheat, corn, rice, rye, triticale, and milo, as well as the flours prepared from bran or other fractions of the milled grain.

The substrate is slurried in a sufficient amount of water to give a concentration in the range of about 10-40% by weight. The water should contain a suitable calcium salt in an amount sufficient to stabilize the subsequently added .alpha.-amylase [preferably about 25-50 part per million (ppm) of calcium]. The slurried substrate may be gelatinized prior to enzymatic treatment, using any method known in the art. The pH of the ungelatinized slurry or the gelatinized dispersion is adjusted to about 5.5-7.5, preferably about 6.0, with sodium hydroxide or other alkali, and the .alpha.-amylase is added.

It is advantageous to use thermostable .alpha.-amylases referred to as 1,4-.alpha.-D-glucan glucanohydrolases and having the essential enzymatic characteristics of those produced by the Bacillus stearothermophilus strains ATCC Nos. 31,195; 31,196; 31,197; 31,198; 31,199; and 31,783. These strains are described in U.S. Pat. No. 4,284,722, which is herein incorporated by reference. Other sources of this enzyme include organisms as B. subtilis which have been genetically modified to express the thermostable .alpha.-amylase of B. stearothermophilus as described in U.S. Pat. No. 4,493,893, herein incorporated by reference. These enzymes are available commercially under the name "G-zyme G995" (formerly called "Enzeco Thermolase"; Enzyme Development Div., Biddle Sawyer Corp., New York, N.Y.).

Other suitable .alpha.-amylases are those having the essential enzymatic characteristics of those produced by B. licheniformis var. as described in U.S. Pat. Nos. 4,717,662 and 4,724,208, herein incorporated by reference. These enzymes are available commercially under the name "Taka-Therm L-340" (formerly called "Takalite" Solvay Enzyme Products, Inc., Elkart, Ind.). Of course, any .alpha.-amylase which is useful in the thinning of the starch is contemplated for use herein.

The conditions of enzyme treatment, including the enzyme concentration and the time and temperature of reaction, are selected to achieve liquefaction of the starch in the substrate. When using a thermostable .alpha.-amylase, a preferred treatment temperature is in the range of 70.degree.-100.degree. C., preferably about 95.degree. C. At these temperatures, gelatinization of the starch in the substrate occurs concurrently with the hydrolysis. The duration of the treatment at the desired conversion temperature depends on the desired product properties and will generally range from about 2-60 min.

After completion of the enzymatic hydrolysis, the enzyme is inactivated, such as by passing the mixture through a steam injection pressure cooker at a temperature of about 140.degree. C. Alternatively, the enzyme may be inactivated by acidification (pH 3.5-4.0) at 95.degree. C. for about 10 min. Optional neutralization with alkali increases the salt concentration of the product. After the enzyme has been inactivated, the soluble fraction comprising the soluble dietary fiber and the maltodextrins (maltooligosaccharides) is separated from the insoluble residue by centrifugation of the hydrolysates. In a preferred embodiment of the invention, temperatures during centrifugation are maintained less than 70.degree. C., and most preferably within the range of 5.degree.-50.degree. C. Under these conditions of separation, the levels of lipids and proteins in the dietary fiber products are significantly reduced. Water is then removed from the soluble fraction by any of a variety of conventional techniques, whereby the products of this invention comprising the dietary fiber and maltodextrins are recovered. The maltodextrins produced by the process of the invention have a D.E. of 20 or less. These maltodextrins are substantially water soluble at elevated temperatures (e.g., 70.degree.-100.degree. C.).

The soluble dietary fiber recovered from the centrifugate is principally in the form of .beta.-glucans and pentosans. Of course the relative amount of each fiber type varies with the species of substrate. Oat and barley substrates yield mostly the .beta.-glucans; whereas wheat, rice, and corn yield the pentosans.

It was surprising to find that the maltodextrin products of this invention are readily obtainable in the colorless, white, and smooth-textured form, devoid of inherent undesirable color, flavor, and grittiness associated with the starting materials. They are remarkably adapted for use as functional and nutritional components of many foods comprising a digestible food component including dairy products, dairy product substitutes, high-soluble fiber bakery products, salad dressings, meats, frozen foods, yoghurt, snacks, confectioneries, coatings, dietary-fiber beverages, breakfast foods, and various low-calorie/low-fat products. Consumer preference dictates that additives intended for these applications be characterized by the virtual absence of the flavor, color, and grittiness inherent to the starting substrate and to most dietary fibers currently marketed.

Characterization of the insoluble residue from the conversion mixture (supra) reveals a high proportion of protein. The protein is believed to complex with available lipid and to become insolubilized by heat denaturation. The residue also contains the insoluble fiber and the majority of the flavor and color components. This byproduct, therefore, has potential as an ingredient in foods of animal feeds and is considered an ancillary asset to the inventive process.

The following examples are presented only to further illustrate the invention and are not intended to limit the scope of the invention which is defined by the claims.

All percentages herein disclosed are by weight unless otherwise specified.

For each preparation, 100 g (dry basis) of oat flour (The Quaker Oats Company, Cedar Rapids, Iowa) was slurried in 400 ml of water containing 25 ppm of calcium (0.09 g/l CaCl.sub.2.2H.sub.2 O) and gelatinized by passage through a steam injection cooker at 138.degree.-143.degree. C. (30-40 psi of steam pressure). The gelatinized mixture was collected in a container, and the pH was adjusted with 1.0N sodium hydroxide as desired (Table I). "Enzeco Thermolase" (supra) was added to the mixture at 95.degree. C. in an amount sufficient to provide 24 units per gram of oat flour, where 1 unit of amylase activity is the amount of enzyme required to hydrolyze 10 mg of starch per minute under specified conditions [Enzyme Development Div., Biddle Sawyer Corp., New York, N.Y., Technical Bulletin No. 20 (Revised 7/86)]. After 20 min. of stirring at 95.degree. C., the starch was liquefied, and the enzyme was inactivated by passing the mixture through a steam injection cooker (supra). The mixture was then allowed to cool to about 70.degree. C., and it was centrifuged 30 min. at 5000 rpm. The water-soluble fiber product in the supernatant solution was recovered by decanting the solution and freeze-drying. The insoluble residue obtained from centrifuging was removed and air-dried. The results in Table I show that the higher pH's (9 and 11) are accompanied by undesirably high levels of protein in the soluble products and decreased protein amounts in the insoluble residues.

Compositions were prepared as described in Examples 1-4 except that the starting oat four was obtained from National Oats Company (Cedar Rapids, Iowa), the process pH was 6.0 for all preparations, and enzyme was added in varying amounts as reported in Table II. The DP distribution of starch oligomers in the products was determined by high-pressure liquid chromatography (Inglett, supra). The data in Table II show that lower enzyme amounts result in higher DP starch oligomers.

Two hundred grams (dry basis) of oat flour [National Oats Company (supra)] were slurried in 800 ml of water containing 50 ppm of calcium (0.185 gl CaCl.sub.2.2H.sub.2 O). The pH of the slurry was adjusted to 6.0 with 1.0N sodium hydroxide, and "Enzeco Thermolase" (supra) was added to the stirred slurry in an amount sufficient to provide 48 units (supra) per gram of oat flour. The temperature was increased to 80.degree. C., and stirring was continued until the starch was liquefied. After 20 min., the pH was adjusted to 4.0 with 1N sulfuric acid, and the mixture was heated at 95.degree. C. for 10 min. to inactivate the enzyme.

The pH was then adjusted to 6.2 with 1N sodium hydroxide, and the soluble product was separated from the insoluble residue following the procedure described in Examples 1-4. The total dietary fiber [Prosky et al., J. Assoc. Off. Anal. Chem. 67: 1044 (1984); 68: 399 (1985)] amounted to 4.8% of the product, compared to 7-8% normally obtained in the products of this invention. Because of the acid inactivation of the enzyme and subsequent neutralization with sodium hydroxide, the sodium content of this soluble product was 5215 .mu.g/g, in contrast to the range of 200-900 .mu.g/g for products having the enzyme inactivated by steam injection cooking.

Four kilograms of oat bran [National Oats Company (supra)] was slurried in 8 l of water containing 25 ppm of calcium (supra). The pH of the slurry was adjusted to 6.0 with 1.0N sodium hydroxide, and "Taa-Therm L-340" (supra) was added to the slurry in a large Sigma mixer in an amount sufficient to provide 1632 Modified Wohlgemuth Units (MWU) per gram of substrate, where 1 MWU is that activity which will dextrinize 1 mg of soluble starch to a defined size dextrin in 30 min. under specified conditions (Solray Enzyme Products, Inc., Elkart, Ind.). During a period of about 10-15 min., the temperature was increased to about 95.degree. C. with steam heat. After 20 min., the enzyme was inactivated by passing the mixture through a steam injection cooker. The warm slurry was centrifuged at 15,000 rpm by a large "Sharples" centrifuge to separate the soluble components from the insoluble components. The soluble components of the oligomer composition (expressed on a percentage basis) were: DP> 9, 32.3; DP 9, 0; DP 8, 0; DP 7, 1.0; DP 6, 13.0; DP 5, 13.9; DP 4, 6.4; DP 3, 13.0; DP 2, 12.1; DP 1, 5.6.

Six kilograms of oat flour [National Oats Company (supra)] was slurried in 18 l of water containing 25 ppm of calcium (supra). The pH of the slurry was 5.75. After gelatinization by passage of the mixture through a steam injection cooker, the slurry was collected in a 30-gal. steam-heated kettle. "Enzeco Thermolase" (supra) was added to the slurry in an amount sufficient to provide 1 unit per gram of oat flour. After 5 min. of stirring at 80.degree.-90.degree. C., the enzyme was inactivated by passing the slurry through a steam injection cooker (supra). The warm slurry was centrifuged at 15,000 rmp by a large "Sharples" centrifuge to separate the soluble and insoluble components. The products were dried separately on hot rolls. The oligomer composition indicated 98% DP 9 and larger.

The soluble oat fiber product of Example 7 and the starting oat flour from which it was obtained (supra) were evaluated as ingredients in ice milks. For each preparation, 484 g of 0.5%-fat milk, 100 g of sugar, 1.9 g of vanilla extract, and 0.4 g of uniodized salt were mixed for 1 min. in a blender with the oat products and in the amounts reported in Table III. Each mixture was treated in a 2-qt. ice cream mixer (Oster) for 30 min.

An 18-member trained panel experienced in testing cereal products evaluated the ice milk samples for flavor, texture, and overall quality by a modification of the method of Warner et al. [Cer. Chem. 60: 102-106 (1983)], herein incorporated by reference. At each panel session, the testers received a control sample of laboratory-prepared ice cream and two randomized, unidentified ice milk samples. The panelists were instructed to rate the three samples for intensity of the descriptors listed in Table III, using the scales in the Table footnotes. Scores were statistically analyzed for significant differences at the 95% confidence level (P<0.05).

The results in Table III show that the sweetness of the oat-containing samples did not vary significantly from that of the plain ice milk, and two of the oat samples (11D and 11G) did not vary significantly from the ice cream control. The vanilla flavor intensity was highest in the ice cream control, and two of the oat samples (11D and 11F) were not significantly different; all the oat-containing samples ranged between 4.9 and 5.5 except 11G and 11H, which were significantly lower than the others. The creamy flavor intensity also was highest in the ice cream control; next in creamy flavor were the three ice milk samples containing the invention product. Cereal flavor intensity was significantly lower in the three samples containing the invention product than in the three samples containing the starting oat flour, and two of the invention samples were not significantly different from the ice cream control and the ice milk that contained no oat product.

The creamiest texture among the ice milks were exhibited by 11E; all the other ice milks did not vary significantly from sample to sample. Graininess was significantly lower in the three invention samples than in the other three oat-containing samples. Cohesiveness of 11E did not vary significantly from that of the ice cream control; all the other ice milks ranged from 2.9-4.0. Density values ranged from a low of 2.2 for the plain ice milk to a high of 7.6 for the ice cream control. Both the invention samples and the other oat samples yielded increasing density with increasing concentration of additive.

The ice milks were significantly lower in overall quality than the ice cream control, and the highest levels of oat flour starting material (11G and 11H) resulted in significantly lower overall quality than that of the other ice milks. The ice milks containing the invention product had the highest scores among the ice milks.

In summary, the soluble oat fiber product of the invention increased the creamy flavor, creamy texture, cohesiveness, and density of ice milks without imparting the inherent cereal flavor and grainy texture of the starting oat flour.

The oat materials listed in Table IV were blended into milk and evaluated for flavor characteristics as in Example 11. For each preparation, 300 g of 0.5%-fat milk and 30 g of oat product were mixed for 1 min. in a blender, poured into a flask, and swirled before serving to the test panel. Each panelist received a control of 0.5%-fat milk identified as having a milky flavor intensity score of 8, plus two randomized, unidentified test samples. The panelists were instructed to score the three samples for intensity of the descriptors listed in Table IV. Scoring was based on a 0-10 scale (10=strong; 0=none). Scores were statistically analyzed as in Example 11. Values that were not significantly different are followed by the same letter in Table IV. The results reported in the Table show that the soluble oat fiber products of the invention (12H and 12F) had significantly lower cereal flavor intensity scores than the other oat materials.

A yoghurt was prepared from retail 2% milk, instant nonfat dry milk, sugar (sucrose), water, potassium sorbate, and the soluble oat fiber obtained in Example 10. A control yoghurt without the fiber was also prepared. Compositions of the starting mixtures are set forth in Table V below. The ingredients were blended together with a wire whip. The resulting mixtures were heated to 185.degree. F. in stainless steel beakers, held in temperature for 30 min., cooled to 100.degree. F., and then inoculated with 2% bulk starter (Hansen's CH.sub.3 propagated once in 11% nonfat milk). The inoculated mixtures were cultured to a pH of 4.35-4.40 (270 min. for the control; 165 min. for the test sample). The yoghurt was then stirred, poured into containers, and cooled overnight, achieving a final pH of 4.0-4.2. The samples were evaluated in a Brookfield viscometer Model RVT equipped with a No. 2 spindle at 0.5 rpm and at 50.degree. F. They were also evaluated for sensory properties.

The control was characterized by a viscosity of 48,000 cp and a good yoghurt flavor with smooth texture. The test sample with the oat fiber of Example 10 had a viscosity of 22,4000 cp, a slight cereal flavor, and a smooth texture.

It is understood that the foregoing detailed description is given merely by way of illustration and that modification and variations may be made therein without departing from the spirit and scope of the invention.