WO1993025616A1

WO1993025616A1 - Encapsulated starch for ruminant feed

Info

Publication number: WO1993025616A1
Application number: PCT/GB1993/001245
Authority: WO
Inventors: Graeme Gordon; Colin Stanley Fitchett; Graham Stuart Plastow; Sotiris Anthimou Papasolomontos; Stuart Marsden
Original assignee: Dalgety Plc
Priority date: 1992-06-11
Filing date: 1993-06-11
Publication date: 1993-12-23
Also published as: AU4345493A; GB9212418D0

Abstract

Starch encapsulated in a matrix derived from the Maillard reaction of a soluble heterologous protein and a reducing sugar may be protected from rumen degradation and used in improved ruminant feeds. The starch may be encapsulated by combining starch with a soluble heterologous protein and a reducing sugar and heating to initiate the Maillard reaction. Feeds including the encapsulated starch may be used to improve the quality of ruminant milk and to increase the efficiency of microbial fermentation in the rumen.

Description

ENCAPSULATED STARCH FOR RUMINANT FEED

The present invention relates to encapsulated starch and its use in ruminant feeds.

In the field of dairy livestock nutrition, it is common practice to supplement grass silage with other feeds (mainly cereals) to achieve diets with a sufficiently high feeding value to sustain economically efficient levels of milk production from dairy cows. Cows are ruminants, in nature feeding largely by grazing. The rumen allows the animal to utilise low quality fibrous foods such as dietary cellulose (derived from grass), which is a major constituent of the ruminant diet. As a result, ruminant feeds are initially subjected to microbial degradation in the rumen. However, these micro-organisms also attack other dietary components which would be more beneficial to the ruminant animal if they were digested in the stomach and small intestine.

It has been estimated that starch digested postruminally would be of a 10 - 13% higher nutritive value than if it were digested in the rumen (Leng (1981), In Nutritional limits to animal production from pastures, pp.427-453, edited by J.B. Hacker, Commonwealth Agricultural Bureau, Slough, UK). Digestion of readily available starch from feed supplements such as cereals in the rumen also has a detrimental effect on fibre breakdown. The main end products of rumen starch digestion are the volatile fatty acids, acetic, propionic and butyric acids. As a

consequence, starch digestion in the rumen causes a fall in the pH of the rumen environment which in turn reduces desirable cellulolytic activity (which is pH-sensitive). This leads to inefficient utilization of dietary

cellulose. It will therefore be appreciated that protection of starch in feed such as cereals from rumen fermentation thereby rendering it available to the small intestine could be of significant commercial benefit in animal husbandry directly through increased post-ruminal

digestion and/or indirectly through a decreased rate of digestion within the rumen.

Chemical approaches to the protection of nutrients from rumen degradation has primarily been focused on proteins, amino acids and fats. Proteins can be protected by heat treatment, chemical modification or by the introduction of inhibitors of proteolytic activity with the feed.

Heat treatment is effective by making the protein less soluble, and inhibition of proteolytic enzymes can be achieved with vegetable tannins. Chemical modification has primarily relied upon formaldehyde to protect protein in the rumen based on the observation that 0.6-2.0 % of bound formaldehyde markedly reduces the solubility of protein at pH 6.0, thereby rendering it highly resistant to microbial attack. US4,95,7748 discloses the protection of proteins by reacting them with carbohydrates under conditions that promote the Maillard reaction. In this case, the products of the Maillard reaction per se are used directly as the basis for animal feedstuffs. Amino acids can be protected by encapsulation within lipid films, synthetic polymers, or proteins. The major disadvantage of this approach is the relatively high cost of such protection procedures. Fats can be protected by encapsulation in protein matrices via crosslinking with formaldehyde on air, flash, or spray drying. They can also be protected by employing calcium salts of fatty acids. WO 91/05482 discloses the excapsulation of emulsified fats using the Maillard reaction. The use of the encapsulated fats as livestock feed supplements to modify the milk and/or meat fat content of the livestock is also described.

Attempts have also been made to protect starch from rumen attack using chemical means, such as modification with crosslinking agents. There are two common types of crosslinkages, esters and ethers. Esters are stable in acid conditions and labile in alkaline. The pH's

exhibited by the gastrointestinal tract of ruminants makes this approach unattractive for starch protection. Ethers however, formed for example by condensation with formaldehyde, have the appropriate pH stability but unlike esters are permanent. Kassem et al. (1987) have reported an 8 to 10% improvement in milk yield when acid/formaldehyde treated cereals, which were estimated to exhibit a 20% protection of starch in the rumen, were fed to dairy cows. However, the use of formaldehyde is impractical due to its toxicity and at the limits

approved for food use there is no significant effect on the digestibility in the rumen. At present, there are no reported successful methods of direct starch protection from rumen degradation, even in the light of the possible benefits to feed efficiency or therapeutic approaches to certain metabolic diseases, eg. ketosis. On the contrary, most of the work on starch has focused on treatments which make it more available throughout the digestive tract (including the rumen). Thus, methods of processing cereal grains for ruminants has mainly concentrated on increasing the efficiency of utilising the grain. Steam flaking, micronisation, and rolling are examples of processes where the grain and especially its starch are made readily available to the rumen. It is therefore apparent that a prejudice exists in the art against treating starch to render it less digestible, the technical trend being towards increasing its availability to the animal.

The present invention is directed to the problem of starch digestion in the rumen, and in the face of the aforementioned technical prejudice provides starch which has been at least partially protected from rumen

degradation by encapsulation in an insoluble

protein/sugar matrix. Laboratory studies have indicated that the starch is subsequently available to digestion in the small intestine. The invention is based at least in part on the recognition that the Maillard reaction of suitable quantities of a soluble protein and a reducing sugar can produce a matrix that can act as a starch encapsulant suitable for reducing the susceptibility of the encapsulated starch to degradation in the rumen. The Maillard reaction is a well-known chemical reaction (also known as non-enzymic browning). The first step of the reaction involves condensation of a reducing sugar and an amino acid contained in the soluble protein to form a Schiff base. This is then followed by

rearrangement to the more stable Amadori product (Figure 1). Further reactions result in indigestible melanoidins (brown, high molecular weight, complex furan

ring-containing and nitrogen containing polymers). The rate of formation of Maillard products is dependent upon temperature, pH and the water content of the material, with reaction reaching a maximum at water levels of typically 25% in foodstuffs. The rate also depends on the amino acid composition of the protein, proteins with a high lysine content being particularly active in the reaction.

The starch encapsulated according to the present

invention can be derived from a source (such as wheat starch) that also contains protein (in this case, gluten). Quantities of such homologous protein may therefore remain associated with the starch. While this homologous protein may participate to some degree in the Maillard reaction, the extent of such reaction will of course depend inter alia on the chemical composition of the protein (as explained in the foregoing paragraph) and the amount present in the starch preparation. Such homologous protein may not promote effective and reproducible encapsulation - indeed, in the case of gluten present in wheat flour starch, no encapsulation whatsoever appears to occur, presumably due to the low concentration of lysine groups present in this protein.

For this reason, the present invention employs protein from a different source to the starch (ie. heterologous protein), so that suitable amounts of a heterologous protein of known activity in the Maillard reaction can be used in the encapsulation process to allow controlled encapsulation. The heterologous protein may be from a different biological source to the starch source (as is the case when wheat starch is reacted with soya protein), or may simply be from a different physical preparation (as is the case when a pea starch preparation is reacted with pea protein from a physically separate and

independently prepared pea protein source).

Thus, according to the present invention, there is provided starch encapsulated in a matrix derived from the Maillard reaction of a soluble heterologous protein and a reducing sugar.

The invention also provides a method of encapsulating starch comprising the steps of mixing air dry starch with a soluble, heterologous protein and a reducing sugar to form an air dry mix, adding water to the air dry mix, and heating to initiate the Maillard reaction.

Also included within the scope of the present invention is a method of protecting starch from rumen degradation comprising the step of encapsulating the starch in a matrix derived from the Maillard reaction of a soluble heterologous protein and a reducing sugar. The invention also covers ruminant feeds comprising starch encapsulated in a matrix derived from the Maillard reaction of a soluble heterologous protein and a reducing sugar, along with methods for the preparation of such feeds comprising the step of encapsulating starch in a matrix derived from the Maillard reaction of a soluble heterologous protein and a reducing sugar. Such feeds may be in the form of pellets, since the encapsulated starch of the invention may be incorporated in pelletized feeds without disrupting the integrity of the pellets.

The invention also comprehends methods of feeding

ruminants comprising the steps of selecting a feed, mixing the feed with the encapsulated starch according to the invention, and feeding the resulting feed to the ruminants. The invention will now be explained in more detail in the following non-limitative description.

Suitable sources of starch for use in the invention may be derived from any of wheat, barley, oats, flour, triticale, maize, sorghum, rice, rye, potato, tapioca, sweet potato, pea, bean, lupin, salseed and mango, or by-products thereof. Suitable proteins for use in the preparation of the encapsulant include casein, rapeseed, sunflower, soya, linseed, sesame, lentil, cotton, groundnut, maize.

Brewer's yeast, pea, bean, lupin, tomato pip, dried red blood, milk, fish meal, meat meal, meat and bone meal, feather meal, poultry offal, synthetic protein, synthetic amino acids or gelatin, wheat, barley, oats, rye, rice, sorghum, triticale or by products thereof. Some of these products are presently used in the feed industry in large amounts and are therefore readily available.

Particularly suitable for use as a protein in the

invention is gelatin. Even though gelatin has fewer lysine groups than some of the other proteins listed above, it exhibits film-forming properties that are a distinct advantage. However, gelatin is relatively expensive, and for this reason may advantageously be replaced with rapeseed, soya or maize.

Suitable sugars for use in the present invention include xylose, arabinose, glucose, galactose, mannose, ascorbic acid and the disaccharides maltose and lactose. Although xylose and arabinose are known to be most effective in the Maillard reaction, maltose and lactose are

particularly preferred for use in the present invention, being readily available in malt extracts and glucose syrups. Sucrose is a non-reducing sugar and as such gives no reaction, however the disaccharide hydrolyses readily under mild acid condition giving glucose and fructose. Sucrose is a major constituent of molasses which is typically used in feed formulations. Of the two sources of molasses, beet molasses contains approximately 48% sucrose and 1% reducing sugars, in contrast cane molasses has typically 30% sucrose and 22% reducing sugars. Glucose is a readily available hexose and fructose a ketose.

Xylose, a pentose sugar, was found to be very reactive. Glucose, an example of a hexose, was less reactive but gave suitable protection at elevated temperatures. A range of reducing sugars including maltose and lactose syrups from liquid by-products can also be used. Although the sugars used in the examples described below are in the purified form, this is not an essential requirement. Inexpensive impure sources of reducing sugar, such as precursors in glucose production, high fructose corn syrups or malt extracts and waste effluents from wood, milling, brewing and dairy industries could all be used.

As an alternative to using two separate sources of protein and reducing sugar, it is possible to use liquid by-products of presently operated processes as a cost-effective source of both reducing sugars and soluble proteins with the appropriate amino acid balance for Maillard reaction (eg. whey syrup and steep liquor maize). This may advantageously limit the raw material ingredients to one source.

The reducing sugar may also be derived from endogenous starches and/or polysaccharides present in the starch and/or protein sources. In this case the protein/starch components may be pre-treated with enzymes prior to the Maillard reaction step. For example, partial digestion of the protein/starch sources with enzymes such as

cellulases, lactases, amyloglucosidases, xylanases, arabinofuranosidases, beta-glucanases, invertases, galactosidases, pectinases and amylases may release sufficient quantities of reducing sugar to promote

Maillard reactions, without the need to introduce separately prepared or purified reducing sugar. The method of encapsulating starch according to the invention employs the Maillard reaction. Accordingly, the encapsulation must be conducted under conditions suitable for promoting the Maillard reaction. Important parameters in this respect include the moisture content and the pH. It has been found that the reaction proceeds at very low moisture contents (eg. 5%), though the optimum is about 38-40% by weight water.

As to the pH, it has been found that the higher the pH, the greater the extent of the Maillard reaction, though for commercial encapsulation a near neutral pH is

preferred.

The amounts of starch, protein and reducing sugar used according to the invention are not critical, and the optimum quantities vary according to the precise nature of the reactants, their physical state (eg. particle size) and the conditions under which the Maillard

reaction is carried out. For example, when wheat starch is used, the amount of sugar used can vary from about 1% to about 10% by weight (with respect to the starch), that of the heterologous protein from about 0.4% (or even lower) to about 50% by weight. Of course, even higher quantities of sugar and protein could be used with respect to the starch, though such quantities would be expected to be of little benefit to starch encapsulation.

The protein and reducing sugars employed in the present invention as reactants for generating the protective encapsulant do not in themselves significantly contribute to the nutritional value of ruminant feedstuffs based on the encapsulated starch product. The protein/reducing sugar used to generate the encapsulant are therefore not protein/carbohydrate feeds per se. Ruminant feedstuffs (such as feed pellets) based on the encapsulated starches of the invention would therefore normally include

quantities of feed protein/carbohydrate in addition to and quite distinct from the Maillard-modified

protein/carbohydrate present in the starch encapsulant. Microscopic analysis of starch grains about 1 mm in radius which had been encapsulated according to the present invention indicated that the protein and reducing sugar can react to form a continuous layer about 0.3 mm thick around the grains.

The encapsulated starch of the present invention is more slowly digested in the rumen, and so effectively by-passes the rumen to become available to the small

intestine. This produces four major benefits; (1) the nutritive value of the starch is increased because the starch is protected from unproductive microbial

degradation in the rumen, (2) cellulolytic activity in the rumen is increased thereby improving the efficiency of utilization of dietary cellulose, (3) the intake of silage by the ruminants increases, and (4) the quality of the ruminant milk is improved, in particular via an increase in milk fat (specifically, unsaturated fatty acids).

Without wishing to be bound by any theory, the

improvement in milk quality is thought to arise as follows:

Cereal starch is very fast fermented in the rumen, encouraging a low pH and starch digesting bacteria yielding high levels of propionate as opposed to acetate. Protection of starch slows down its rate of rumen

digestion, creating an environment of higher and less fluctuating pH and yielding a higher level of acetate to propionate. This also indirectly encourages conditions which favour cellulolytic bacteria, yielding still higher acetate production from better fibre digestion. Acetate is absorbed from the rumen wall and it is metabolised in part to fatty acids, which in turn are increasingly converted to milk fat. These are mostly of a saturated form. Thus the slower rumen fermentation is thought to lead to de novo synthesis of milk fat in the mammary gland from the increased levels of rumen-produced

acetate.

The invention will now be described in greater detail by way of examples. These examples are for illustrative purposes only and are in no way intended to limit the scope of the invention. In these examples, the starch encapsulated is derived from wheat, since this is one of the most convenient sources of starch. It is of course possible to use starch from other sources in varying states of purity.

Example 1: Encapsulation of wheat starch Encapsulation of wheat starch was accomplished by the following method. A milled wholemeal flour (153.37g) was mixed with a source of soluble heterologous protein

(15.34g) and a reducing sugar (12.27g) in a Kenwood mixer, setting no. 2. The flour was used either course-ground (in which the particle size ranged from 63 um to 2 mm) or fine-ground (in which the particle sizes were from 2-45 um). The air dry ingredients were thoroughly mixed. Tap water (72.02g) was added dropwise with mixing over a two to three minute period. After addition of water the mixture was mixed for a further five minutes to give a crumb. This crumb was spread evenly over baking trays and the Maillard reaction was carried out at the desired temperature in a preheated draught assisted oven.

Various combinations of sugars and proteins were

employed, and the resulting samples of encapsulated wheat starch are shown in Table 1.

Example 2: Measurement of starch protection Two methods were chosen to determine the effectiveness of protection. The first technique employs an in vitro assay under conditions that mimic the rumen, and gives an indication as to whether the encapsulated starch is protected from the degradative activities of the rumen. Although it is impossible to mimic accurately the enzymic activity of the rumen in vitro (it contains one of the most varied microbial populations known in nature, and involves both solid and liquid phases), an indication of the degree of protection conferred by a given

encapsulation technique can be obtained by incubating the protected starch in a buffer that mimics many of the. most important chemical parameters of the rumen. In this method, 5g of coarsely milled sample is treated overnight (16 hours) with bacterial amylase and protease in a 0.1M citric acid-trisodium citrate buffer, pH6.5, with 0.0043M CaCl2 at 37ºC.

The second technique employs an in vivo study of rumen digestion using Dacron (RTM) bags containing the

encapsulated starch inserted into the rumen of fistulated cows. In this technique, the starch is directly

subjected to the rumen environment and rates of digestion can be monitored.

Analyses of starch for assessment of protection were based on either direct measurement of total starch or measurements of dry matter. For direct starch

determination, the sample is first ground and then treated with a solution of pepsin. This allows hydrolysis of the starch, which is carried out in two steps; first, the addition of a fast acting thermostable alpha-amylase, Termamyl (Registered Trade Mark), to liquify the starch to soluble dextrins, and second, incubation with porcine alpha-amylase and amyloglucosidase to hydrolyse these dextrins to glucose. Finally, glucose generated from the enzymatic breakdown of the starch is measured

colourimetrically by the GoD/Perid (Registered Trade Mark) assay and the value for starch calculated.

As an alternative to direct starch measurements, the loss of starch was followed as the loss of total dry matter. This technique allowed measurement of starch protection indirectly via measurements of dry matter disappearance from Dacron bags. For the measurement of dry matter, the Dacron bags are removed after the required incubation period and then washed, dried and weighed.

Example 3: Protection of starch from rumen degradation by encapsulation

The samples of encapsulated starch prepared in Example 1 (and listed in Table 1) were subjected to rumen

conditions and measurements of starch protection were carried out as described in Example 2.

The results of these experiments are shown in Table 2. It can be seen that whereas unprotected starch was almost completely degraded after 18 hours in the rumen (see the results for the unencapsulated controls), the

encapsulated starch was protected to the extent that typically about 20% of the amount originally present remained undegraded at this time (as determined by in vivo rumen incubation experiments). This increased stability is also reflected in the longer nominal half-life of the encapsulated starch (up to 5-fold higher than that of unprotected starch). This is also shown in Table 2.

Example 4: Starch protection using a variety of proteins

The relative efficacy of a variety of different proteins was investigated using xylose as the reducing sugar in a Maillard reaction at 100 degrees centigrade for 2 hours. The in vitro rumen technique (see Example 2) was used to compare the degree of protection conferred - the results are shown in Table 3. The experiments described in Examples 1 to 4 above show that a wide range of sugars and proteins can be used to produce the protected encapsulated starch (and feeds based thereon) of the invention. Example 5: Effect of initial moisture content on

encapsulation

The effect of the initial moisture content on

encapsulation efficiency was determined by measuring the colour development of the product - the efficiency of the encapsulation is determined by that of the Maillard . reaction, which can be monitored by the extent of

browning achieved. Samples of wheat starch mixed with protein and various amounts of water were heated to 170 and 150 degrees centigrade for 20 to 60 minutes, and the extent of the Maillard reaction was measured by

determining the absorbance at 430 nm of a filtered blend of the products with deionized water. It was found that the optimum water content was about 38%, the efficiency of the Maillard reaction being reduced at higher (greater than about 40% by weight) or lower (less than 30% by weight).

Example 6: Effect of pH on encapsulation A similar experiment to that described in Example 5

(above) was conducted, this time determining the effect of changes in pH on the development of browning (as measured by absorbance at 430 nm). It was found that after 400 min the extent of the Maillard reaction was about twice as great at pH 9.2 as it was at pH 7.0, and about 20 times as great as it was at pH 4.0.

Example 7: Increased milk fat content in cows fed with encapsulated starch Two groups of cows were fed on feeds with and without encapsulated starch. The key physical performance data for the two groups of cows, after six weeks on trial, are shown below;

Control Encapsulated Sig

Milk Yield (kg) 28.03 27.44 NS

Milk Fat (%) 3.75 4.16 **

Milk Protein 3.07 3.08 NS

Milk Lactose 5.07 5.07 NS

(NS = not significant, ** = significant)

These data indicate that the incorporation of

encapsulated starch into ruminant feed improves the efficiency of rumen fermentation and so leads to

significant increases in milk fat content.

Example 8: Manufacture of Encapsulated Starch The following ingredients were mixed for 15 minutes in the following proportions to provide a dough:

Ingredient % inclusion

Ground wheat 61.7

Soya protein 6.0

Glucose 3.6

Sodium metabisulphite 0.02 Water 28 . 7

This dough was rolled into a sheet 16mm thick. The dough sheet was then cut into irregular small sized pieces using a rotary cutter. These pieces were passed through a travelling oven with a 10.5 minute bake time and a throughput of 3 tonnes per hour.

The oven parameters employed were:

Start of oven End of oven

290 290 270 270 270 270 250 250 250°C

On leaving the oven the product was cooled to ambient temperature prior to grinding (15mm sieve) and bagging.

TABLE 1 : ENCAPSULATED STARCH SAMPLES

Sample Coateat

Code Protela Sugar Process Conditions

GRL 1 Casein Xylose Oven dried at 60ºC/4hrs

GRL 3 Casein Xylose Freeze dried then oven dried at 100ºC/2hr

GRL 6 Rapeseed Xylose Oven dried at 100ºC/2hrs

GRL 7 Soya Xylose Oven dried at 100ºC /2hrs

GRL 8 Maize Xylose Oven dried at 100ºC /2hrs

GRL 9 Casein Xylose Oven dried at 100ºC/2hrs

GRL 10 Soya Glucose Oven dried at 100ºC/2hrs

GRL 11 Soya Glucose Oven dried at 150ºC/1hrs

Table 3 ; Protected Starch to in Vitro Rumen Digestion

Protein % rumen protected starch

Brewers Yeast 31.3

Gelatin 57.4

Pea Protein 44.6

Dried Red Blood 14.7

Hipro Soya 46.7

High Protein Maize 48.2

Fish Meal 47.5

Casein 49.8

Rapeseed 45.8

Claims

1. Starch encapsulated in a matrix derived from the

Maillard reaction of a soluble heterologous protein and a reducing sugar.

2. Encapsulated starch according to claim 1 wherein the starch is derived from any of wheat, barley, oats, flour, triticale, maize, sorghum, rice, rye, potato, tapioca, sweet potato, pea, bean, lupin, salseed and mango, or by-products thereof.

3. Encapsulated starch according to claim 1 or 2 wherein the protein is derived from casein, rapeseed, sunflower, soya, linseed, sesame, lentil, cotton, groundnut, maize, Brewer's yeast, pea, bean, lupin, tomato pip, dried red blood, milk, fish meal, meat meal, meat and bone meal, feather meal, poultry offal, synthetic protein, synthetic amino acids and gelatin, wheat, barley, oats, rye, rice, sorghum, triticale or by products thereof.

4. Encapsulated starch according to any one of the preceding claims wherein the reducing sugar is xylose, glucose or arabinose.

5. Encapsulated starch according to any one .of claims 1 to 3 wherein the reducing sugar is in an impure form, eg, precursors in glucose production, high fructose corn syrups, malt extracts and waste effluents from the wood, milling, brewing and dairy industries.

6. Encapsulated starch according to claim 1 or 2 wherein the soluble heterologous protein and reducing sugar are derived from liquid by-products, for example spent maize wash, steep maize liquor, whey syrup, molasses cane, molasses beet or spent wheat wash.

7. Encapsulated starch according to claim 1 or 2 wherein the reducing sugar is derived from the enzymic

degradation of endogenous starch/polysaccharide present in the starch and/or protein source.

8. Encapsulated starch according to claim 7 wherein the enzymic degradation involves any of the enzymes selected from cellulases, lactases, amyloglucosidases, xylanases, arabinofuranosidases, beta-glucanases, invertases, galactosidases, pectinases and amylases.

9. A method of encapsulating starch comprising the steps of combining starch with a soluble heterologous protein and a reducing sugar, and heating to initiate

the Maillard reaction.

10. A method of encapsulating starch according to claim 9 wherein air dry starch is combined with the heterologous protein to form an air dry mix, and water is then added before the heating step.

11. A method of encapsulating starch according to claims 9 or 10 wherein the starch is derived from any of wheat, barley, oats, flour, triticale, maize, sorghum, rice, rye, potato, tapioca, sweet potato, pea, bean, lupin, salseed and mango, or by-products thereof.

12. A method of encapsulating starch according to any one of claims 9 to 11 wherein the protein is derived from any of casein, rapeseed, sunflower, soya, linseed, sesame, lentil, cotton, groundnut, maize, Brewer's yeast, pea, bean, lupin, tomato pip, dried red blood, milk, fish meal, meat meal, meat and bone meal, feather meal, poultry offal, synthetic protein, synthetic amino acids, wheat, barley, oats, rye, rice, sorghum, triticale and gelatin, or by products thereof.

13. A method of encapsulating starch according to any one of claims 9 to 12 wherein the reducing sugar is xylose, glucose or arabinose.

14. A method of encapsulating starch according to any one of claims 9 to 13 wherein the reducing sugar is in an impure form. eg. precursors in glucose production, high fructose corn syrups, malt extracts and waste effluents from the wood, milling, brewing and dairy industries.

15. A method of encapsulating starch according to any one of claims 9 to 12 wherein the soluble heterologous protein and reducing sugar are derived from liquid by-products, for example spent maize wash, steep maize liquor, whey syrup, molasses cane, molasses beet or spent wheat wash.

16. A method of encapsulating starch according to claim 9 or 10 wherein the reducing sugar is derived from the enzymic degradation of endogenous starch/polysaccharide present in the starch and/or protein source.

17. A method of encapsulating starch according to claim 16 wherein the enzymic degradation involves any of the enzymes selected from cellulases, amyloglucosidases, xylanases, arabinofuranosidases, beta glucanases, invertases, galactosidases, pectinases, lactases and amylases.

18. A method of encapsulating starch according to any one of claims 9 to 17 wherein the starch is, mixed with 1% to 10% by weight reducing sugar (with respect to the starch), 0.4% to 50% by weight heterologous protein (with respect to the starch), and the water is added to a final amount of at least 5% with respect to the total weight of the final mixture.

19. A method of encapsulating starch according to any one of claims 9 to 18 wherein with respect to the total weight of the final mixture about 60% by weight milled wheat is mixed with about 5% by weight soluble

heterologous protein and about 5% by weight reducing sugar, the water being added to about 30% by weight.

20. A method of encapsulating starch according to any one of claims 9 to 19 wherein heating is carried out at between about 50 degrees and 300 degrees centigrade.

21. A method of encapsulating starch according to any one of claims 9 to 20 wherein heating is carried out at about 100 degrees centigrade for about 2 hours.

22. A method of encapsulating starch according to any one of claims 9 to 21 wherein heating is carried out at about 150 to 175 degrees centigrade for about 50 to 60 minutes, or at about 250 to 290 degrees centigrade for about 10 minutes.

23. A method of protecting starch from rumen degradation comprising the step of encapsulating the starch in a matrix derived from the Maillard reaction of a soluble heterologous protein and a reducing sugar by a method according to any one of claims 9 to 22.

24. A ruminant feed (for example pelletized feed) comprising encapsulated starch according to any one of claims 1 to 8.

25. A method of preparing a ruminant feed comprising the step of encapsulating starch in a matrix derived from the Maillard reaction of a soluble heterologous protein and a reducing sugar by a method according to any one of claims 9 to 21.

26. A method of feeding ruminants comprising the steps of selecting a feed, mixing the feed with the encapsulated starch according to any one of claims 1 to 8, and feeding the resulting feed to the ruminants.

27. A method of improving the quality of ruminant milk, comprising feeding ruminants according to the method of claim 26.

28. A method of according to claim 27, wherein the milk fat content is increased.

29. A method of slowing the rate of starch digestion in the rumen such that the biological efficiency of microbial fermentation therein is improved, comprising feeding ruminants according to the method of claim 26.

30. Use of a soluble protein and a reducing sugar for generating a matrix by the Maillard reaction for encapsulating starch.