NZ312221A

NZ312221A - Method for determining the digestible reactive lysine content in foodstuffs

Info

Publication number: NZ312221A
Application number: NZ312221A
Authority: NZ
Inventors: Paul James Moughan; Shane Mcartney Rutherfurd
Original assignee: Pig Res & Dev Corp; Univ Massey
Priority date: 1995-06-30
Filing date: 1996-07-01
Publication date: 2002-05-31

Abstract

A method for determining the reactive lysine digestibility co-efficient of a foodstuff. The method comprises the steps of: (a) introducing a marker into the foodstuff to be analysed; (b) feeding the foodstuff to a non-human subject for a predetermined period of time; (c) obtaining a sample of the foodstuff digesta from the subject; (d) determining the digestible reactive lysine content of the foodstuff by: (i) introducing a lysine derivatising agent into the foodstuff; and (ii) determining the digestible reactive lysine content of the foodstuff by measuring the equivalent derivatised lysine content in the foodstuff; (e) determining the digestible reactive lysine content in the foodstuff digesta by: (i) introducing a lysine derivatising agent for the epsilon-amino group of lysine, into the foodstuff digesta; and (ii) determining the digestible reactive lysine content of the foodstuff digesta by measuring the equivalent derivatised lysine content in the foodstuff digesta; (f) measuring the marker concentration in both the foodstuff and foodstuff digesta; (g) expressing the reactive lysine content of both the foodstuff and foodstuff digesta per gram of the marker; and (h) calculating the reactive lysine digestibility co-efficient.

Description

<div class="application article clearfix" id="description"> , WO 97/02489 PCT/NZ96/00066 - 1 - METHOD FOR DETERMINING AMINO ACID CONTENT IN FOODSTUFFS TECHNICAL FIELD 5 This invention relates to methods for determining the content of an amino acid in foodstuffs and more paiticularly both the reactive lysine digestibihty co-efficient and the digestible reactive lysine content in foodstuffs BACKGROUND 10 Recently there has been renewed interest in the effects of food processing on the availability of ammo acids, in particular lysine (Knipfel 1981, Batterham 1992) in foodstuffs. Lysine is of particular interest because it is an essential amino acid and is often the first 15 limiting ammo acid in diets, particularly pig and poultry diets In foodstuffs that have undergone processing oi prolonged storage, the e-amino group of lysine can react with other compounds present in feedstuffs (e.g the Maillaid leaction) to become structurally altered and nutritionally unavailable (Hunell and Carpenter 1981) Consequently, numerous assays have been developed to allow determination of the chemically reactive 20 lysine content of foods (Hendriks el al, 1994) One such assay is the l-fluoro-l,4-dinitrobenzene (FDNB) assay (Carpenter 1960) In this assay reactive lysine in a foodstuff to be tested is determined by reacting the foodstuff with FDNB lesulting in the production of Dinitiophenvl lysine (DNP-lysine) which is 25 then detected by absorbance at 435 nm and the corresponding reactive lysine content m the foodstuff calculated An ileal amino acid digestibility assay to predict the availability of amino acids, including lysme, in a foodstuff is also known and in cunent use This tiaditional ileal amino acid 30 digestibility assay is described comprehensively in Moughan et al, 1990 and Butts ei al, 1991 incorporated heiem by reference Briefly, the tiaditional assay comprises firstly calculating the amino acid content of a foodstuff Secondly, calculating the ammo acid digestibility co-efficient from both the amino acid content of a diet formulated from the foodstuff, and the ammo acid content of the digesta of a subject fed that diet Finally, the 35 content of digestible amino acids m the foodstuff is calculated by multiplying the amino acid content m the foodstuff by the digestibility co-efficient 312221 -2- However, it has recently been noted that not all of the chemically reactive lysine in heat-treated proteins is absorbed from the small intestine (Schmitz 1988, Desrosiers et al., 1989, Moughan 1991). Consequently, methods for determining chemically reactive lysine such as the FDNB method are inappropriate for assessment of available lysine as they 5 incorrectly assume that all of the reactive lysine present in a feedstuff is dige; ted and absorbed. Furthermore, it has been shown that the traditional true ileal amino acid digestibility assay does not always accurately predict the availability of lysme in heat processed feedstuffs (Batterham 1992). The explanation for this is that during the acid hydrolysis step of ammo acid analysis, an integral part of the digestibility assay, a portion 10 of the structurally-altered nutritionally unavailable lysine derivatives in processed feedstuffs can break down, reverting back to lysine and leading to an overestimate of unaltered lysine in diets and digesta samples and therefore inaccuracy in digestibility coefficients. 15 Accordingly, there is a recognised need for an accurate method for determining both the reactive lysine digestibility co-efficient and the digestible reactive lysine content in a foodstuff, especially heat processed foodstuffs. It is therefore an object of this invention to provide a more accurate method for 20 determining the content of an amino acid in a foodstuff, as well as methods for determining both the reactive lysine digestibility co-efficient and the digestible reactive lysine content in a foodstuff, or at least to provide the public with a useful choice. SUMMARY OF THE INVENTION 25 Accordingly, in a first aspect the present invention can broadly be said to consist m a method for determining the reactive lysine digestibility co-efficient of a foodstuff which method comprises the steps of: (a) introducing a marker into the foodstuff to be analysed; (b) feeding the foodstuff to a non-human subject for a predetermined period of time; 30 (c) obtaining a sample of the foodstuff digesta from the subject; (d) determining the digestible reactive lysine content of the foodstuff by: (i) introducing a lysine derivatising agent into the foodstuff; and (ii) determining the digestible reactive lysine content of the foodstuff by measuring the equivalent derivatised lysine content in the foodstuff; 35 (e) determining the digestible reactive lysine content in the foodstuff digesta by. (i) introducing a lysine derivatising agent specific for the e-amino group of lysine, into the foodstuff digesta; and (ii) determining the digestible reactive lysine content of the foodstuff digesta by measuring the equivalent derivatised lysine content in the foodstuff digesta; INTELLECTUAL PROPERTY. OFFICE OF N.Z. 3 0 JAN 2002 RECEIVED -3 - (f) measuring the marker concentration in both the foodstuff and foodstuff digesta; (g) expressing the reactive lysine content of both the foodstuff and foodstuff digesta per gram of the marker; and (h) calculating the reactive lysine digestibility co-efficient. 5 In a further aspect, the present invention provides a method for determining the digestible reactive lysine content of a foodstuff which method comprises the steps of: (a) calculating the reactive lysine digestibility co-efficient using the above defined method; and 10 (b) determining the digestible reactive lysine content of the foodstuff by multiplying the value for the reactive lysine content of the foodstuff by the reactive lysine digestibility co-efficient. In a preferred embodiment of both methods, the foodstuff is formulated into a diet before 15 adding the marker and feeding to a subject. Conveniently, the foodstuff digesta sample is taken from the terminal ileum. It is also preferred that the foodstuff digesta is dried, before treatment with a derivatising agent. 20 Also described but not claimed is an assay for determining the content of an amino acid in a foodstuff which method comprises determining the amino acid digestibility co-efficient and the digestible amino acid content in a foodstuff according to the methods detailed above, wherein the amino acid of interest is substituted for lysine. 25 The term foodstuff as used herein refers to a foodstuff comprised at least partially of protein. The term foodstuff digesta as used herein means digesta derived from an animal, or test animal fed that foodstuff. 30 Although the present invention is broadly as defined above, it will be appreciated by those persons skilled in the art that the invention is not limited thereto and that it also includes embodiments of which the following description gives examples. In particular preferred aspects of the invention will now be described in relation to the accompanying drawings 35 in which: intellectual property office of n.z - 6 NOV 2001 RECEIVED WO 97/02489 -4- PCT/NZ96/00066 Figure 1 is a graph illustrating the homoarginine (reactive lysine) content of heatea lactose/casein incubated with 0.6 M O-methylisourea pH 10.6 in a shaking waterbath at 21±2°C for 1 to 14 days, with the reagent to lysine ratio greater than 1000. 5 Figures 2/1 and 2/2 provide graphic illustrations of the amount of homoarginine (reactive lysine) present in the digesta of rats fed unheated casein and hea.ed lactose/casein determined using the guanidination reaction. For Figure 2/1 the reaction time was varied. The guanidination conditions were 10 incubation in 0.6M O-methylisourea pH 10.6 at 21±2°C for 1 to 21 days with the reagent to lysine ratio being greater than 1000. For Fig. 2/1, n=13 for unheated casein, 1 day incubation: n=8 for heated lactose/casein, 1 day incubation: n=7 for unheated casein 3 and 7 day incubations and heated 15 lactose/casein, 7 day incubation: n=5 for heated lactose/casein, 7 day incubation: n=3 for both proteins at 14 and 21 day incubations. For Figure 2/2 the pH of the reaction mixture was varied. Guanidination conditions for the digesta of rats fed unheated casein were incubation for 1 day at 21±2°C in 0.6M O-20 methylisourea at pH 9.8 to 11.4, with the reagent to lysine ratio being greater than 1000. The guanidination conditions for the digesta of rats fed heated lactose/casein were incubation for 7 days at 21±2°C in 0.6M O-methylisourea at pH 9.8 to 11.4, with the reagent to lysine ratio being greater than 1000. 25 For Fig. 2/2, n=3 for all analyses. Values were means ± SE. Figure 3 illustrates by means of bar graphs the recovery of amino acids from various protein sources after guanidination with O-methylisourea. Recoveries were calculated as 30 follows: Recovery (%) = Moles of amino acid determined m guanidinated protem X 100 Moles of amino acid determined in unreacted protein 1 Figure 4 illustrates by means of graphs the mean (±SE) food intakes for the first five 35 meals for rats on the last day of trial. Figure 5 is a flow chart illustrating a method of determining endogenous amino acid flow. SUBSTITUTE SHEET WO 97/02489 PCT/NZ96/00066 -5- Figure 6 is a flow chart illustrating the present methods for determining both the digestible reactive lysine co-efficient and digestible reactive lysine content of a foodstuff Figure 7 is a flow chart illustrating the traditional ileal amino acid digestibility assay. 5 Figure 8 is a flow chart illustrating a current process for determining the reactive lysine content of a foodstuff. Figure 9 is a bar graph comparison of the true ileal digestability of amino acids (other than 10 lysine) as determined using conventional amino acid analysis (□) or following the guanidination reaction (■). DETAILED DESCRIPTION OF THE INVENTION 15 As the reader will appreciate, while the present invention is discussed for convemence with respect to lysme, the methods of the invention are equally applicable to other amino acids of interest, and therefore represents an alternative to the traditional amino acid assays. 20 In a first aspect, the present invention provides methods for determining both the reactive lysine digestibility co-efficient and digestible reactive lysine content of a foodstuff. Essentially, the method involves coupling a derivatising reaction (which converts chemically reactive lysine to an equivalent stable derivative) to the traditional true ileal amino acid digestibility assay An example of a derivatising reaction is to use 25 guanidination to convert lysme to the equivalent acid stable homoarginine The guanidination reaction is described m Rutherfurd and Moughan 1990 incorporated herein by reference. In the method of the present invention a derivatising reaction is used in a method foi 30 determining the reactive lysme content of both the foodstuff being tested and the foodstuff digesta of subjects fed that foodstuff When coupled with the ileal digestibility assay the true ileal reactive lysine digestibility co-efficient of the foodstuff can be determined The co-efficient is calculated according to the following equation 35 True ileal reactive lysine digestibility = Reactive lysine in diet - (Reactive lysine m digesta - Endogenous lysine) Reactive lysine in diet WO 97/02489 PCT/NZ96/00066 -6- Once the reactive lysine digestibility co-efficient has been calculated the true digestible reactive lysine content of the foodstuff can be calculated by multiplying the reactive lysme content of the foodstuff by the reactive lysine digestibility co-efficient. 5 The foodstuffs analysed according to the methods of the present invention will be comprised at least partially of protein, and may be wholly comprised of protein The foodstuffs may be naturally occurring or processed foodstuffs. Naturally occurring foodstuffs mclude crude protem such as meats and milk Processed foodstuffs range from relatively unprocessed proteins such as lysozyme, soy protein isolates, skim milk powder, 10 lactic casein, whey protein concentrate, soy protem concentrate, wheatmeal, and soyabean meal amongst others, through to severely processed foodstuffs such as meat and bone meal, cottonseed meal, bloodmeal, commercial animal feeds, and foods produced for human consumption as examples only. 15 In a preferred embodiment the foodstuff is formulated into a diet before feeding to a subject. Formulation of a diet may mclude mixing a selected protem source with additives such as water, carbohydrates, vitamins and minerals but not limited thereto. Formulation of a diet may also comprise physical treatment such as freeze drying, heating, and pulverising but again is not limited thereto 20 A marker or markers are introduced into the diet as a means for measuring the degree of concentration the diet undergoes as it is converted into digesta Selected markers will be indigestible and of low toxicity Suitable markers include spectrophotomenc markers such as chromic oxide or radiopaque markers amongst others. Chromic oxide is the 25 presently preferred marker The diet formulated together with the chromic oxide marker is then fed to a selected subject for a predetermined period of time. The term subjects as used herein generally refers to non-human animals Suitable "subjects" include by way of example only rats, 30 mice, pigs, chickens and dogs '7/7 vitro subjects" are possible These subjects constitute in vitro environments which simulate an animal gut, and produce "/« vitro digesta" Subjects may be selected according to the diet to be tested. However, rats are a convenient test subject 35 The period over which the diet is fed to the subject may be varied according to the diet to be tested, and the subject selected For example, a diet will be fed to a rat for between 5 to 20 days, and most preferably 14 days The rat may be fed the diet m one or more meals a day for a set time period It is preferable on the 14th day to adopt a feeding PCT/NZ96/00066 7 - regime whereby the rats are fed hourly from 8.30 h and 16 30 h for a total of nme meals is appropriate. Generally the feed will be available for 5 to 15 minutes and preferably 10 minutes only. 5 To obtain a sample of the foodstuff digesta from the subject it is generally necessary to sacrifice the subject, although live sampling techniques are possible. The digesta sample may be selected from any part of the ileum However, in a preferred embodiment the digesta sample is taken from the ileum immediately anterior to the ileal-caecal junction For a rat the ileum portion dissected out would be in the order of 10 to 30 cm preferably 10 20 cm. The digesta sample is obtained by dissecting out a portion of the ileum and then flushing out the digesta Convemently, the digesta may be flushed out of the ileum using a synnge filled with distilled, deionised water The digesta obtained can then be analysed directly 15 but is preferably freeze-dried Samples of in vitro digesta are readily obtainable from the in vitro environment. The reactive lysine content of the diet and digesta is then required to be determined As 20 a first step it is necessary to convert reactive lysine into a stable derivative. This is conveniently accomplished by reaction with a derivatising agent. Any derivatising agent used to determine the reactive lysine in diets and digesta must be specific for the e-amino group of lysme, to allow the determination of both free and bound reactive lysine that may be present in digesta. Secondly, the derivatised lysine compound must be acid stable 25 Thirdly, derivatisation must be quantitative FDNB, the most commonly used reagent for determining reactive lysme, fails as a suitable reagent as it can also react with the a-amino group of lysine and will therefore not detect free reactive lysine Furthermore, DNP-lysme is not particularly acid stable necessitating 30 the use of correction factors The derivatising agent, O-methylisourea, is specific for the e-amino group of lysine In a preliminary study, 95% of lysine was recovered as homoargimne or residual unguanidinated lysine from a mixture of lysme and 0 6M OMIU at pH 10.6 incubated for 35 24 hr at 20°±2°C in a shaking waterbath Furthermore, O-methylisourea appears to be acid stable In a preliminary study, 96% of homoarginine was recovered when homoarginine was incubated in 6M HCL in a sealed evacuated glass tube for 24 hr at 110°C±2°C Therefore, O-methylisourea shows promise as a suitable reagent for the ileal WO 97/02489 WO 97/02489 PCT/NZ96/00066 - 8 - reactive lysine digestibility assay, and is currently the derivatising agent of choice. O-methylisourea acts as a guanidinating agent converting reactive lysine to acid stable homoarginine. The homoarginine content as measured is therefore equivalent to the reactive lysine content Previous guanidination techniques are described in Rutherfurd 5 and Moughan 1990 It is also preferred that the guanidination reaction be carried out under optimised conditions to ensure maximum guanidination 10 Conversion of lysme to homoarginine in an unprocessed protem source may be calculated as shown below Conversion of lysine to homoarginine = Moles of homoarginine 15 x 100 Moles of unreacted lysine + Moles of homoarginine 1 The homoarginine level is measured by any suitable means known in the art. A preferred method is amino acid analysis 20 As a parallel step in the methods of the invention it is also necessary to determine the marker content of both the diet and digesta. In the case of a chromic oxide marker the chromium content of the diet and digesta are determined spectrophotometncally following the method of Costigan and Ellis (Costigan and Ellis, 1987) incorporated herein by 25 reference The reacnve lysme content of both the diet and digesta must then be expressed m terms of reactive lysme per gram of the marker according to the following equations: Reactive lysine content in diet = Reactive lysine in diet Chromium m diet 30 Reactive lysine content in digesta = Reactive lysine in digesta X Chromium in diet Chromium in digesta It will also be appreciated by the skilled worker that fractions analysed may include 35 endogenous peptides and proteins It is therefore necessary to calculate the endogenous amino acid flow, and to correct the apparent figure for ileal reactive lysme digestibility flow by subtracting from it the endogenous amino acid flow. One scheme suitable for determining the endogenous amino acid flow is illustrated in Figure 5 Generally, any peptides or proteins with molecular weight greater than 10 000 Daltons will be regarded WO 97/02489 PCT/NZ96/00066 -9- as of endogenous origin. The following equations are used to calculate the endogenous amino acid flows and true digestibility co-efficient Endogenous amino acid flows at the terminal ileum were calculated using the following 5 equation (units are jig g'1 dry matter intake (DMI)): Ileal amino acid flow = Ammo acid concentration in ileal digesta X Chromium in diet Chromium m digesta 10 True ileal ammo acid digestibihty was calculated usmg the following equations (umts are pg g1 DMI): True digestibility = Dietary ammo acid intake - (Ileal ammo acid flow - endogenous ammo acid flow) x 100 15 Dietary ammo acid intake 1 True ileal reactive lysine digestibility was calculated using the following equation (units are pg g'1 DMI): 20 True reactive lysine digestibility = Dietary reactive lysine intake - (Ileal reactive lysme flow - endogenous lysine flow) x 100 Dietary reactive lysine intake 1 The second method of the mvennon relates to the determination of the digestible reactive 25 lysine content of a foodstuff In this method the reactive lysine digestibility co-efficient is determined in accordance with the first method of the invention As a second step the reactive lysine content of the foodstuff is determined using the guanidination reaction set out above. In the case of a foodstuff which has not been formulated into a diet in any specific way this calculation will be the same for the reactive lysine content of the 30 foodstuff calculated in the first method From these two methods the digestible reactive lysine content of the foodstuff can be calculated by multiplying the digestible reactive lysme content of the foodstuff by the true reactive lysine digestibility co-efficient 35 As the reader will appreciate, the methods of the invention apart from having general applicability to any amino acid of interest, may also be performed in vitro. For in vitro applications of the methods of the invention the subject compnses an in vitro environment WO 97/02489 PCT/NZ96/00066 - 10- which simulates the gut of an animal of interest The foodstuff to be analysed is fed mto the in vitro environment, digested and analysed in the same manner as for an animal subject 5 The present invention will now be further described with reference to fie following specific non-limiting examples. EXPERIMENTAL Materials 10 l-fluoro-l,4-dinitrobenzene (FDNB), Dmitrophenyl-lysine (DNP-lysine) and O-methylisourea were obtained from Sigma chemicals, St Louis, MO. Barium hydroxide octahydrate and lactose were obtained from BDH Laboratory Supplies, Poole, England. Lactic casein, skim milk powder and whey protein concentrate were obtained from the New Zealand Dairy Board, Wellington, New Zealand Soy protem isolate and concentrate 15 were obtained from Columbit (New Zealand) Ltd, Auckland, New Zealand Wheat meal, blood meal, meat and bone meal and soybean meal were obtained from the Feed Processing Unit Massey University, New Zealand and cottonseed meal from Cargill Oilseed Ltd, Brisbane, Australia The enzymatically hydrolysed casein was obtamed from New Zealand Pharmaceuticals LTD, Palmerston North, New Zealand and contained 20 peptides no larger than 2000 Daltons. Centriprep 10 disposable ultrafiltration devices were obtained from Amicon, Inc, Beverly, MA. Laboratory rats were sourced from the Small Animal Production Unit, Massey University, Palmerston North, New Zealand EXAMPLE 1 25 FDNB Method l-fluoro-l,4-dinittobenzene (FDNB)-reactive lysine was determined according to the method of Carpenter (1960) usmg the modifications described by Booth (1971). Samples containing approximately 10 mg of reactive lysme (estimated previously using amino acid analysis), were reacted with FDNB in ethanol/NaHCO, at room temperature for 2 hr The 30 resulting Dinitrophenyl (DNP)-lysine was liberated from the protein by hydrolysis in 5.8M HCI for 16 hr under reflux conditions The unreacted FDNB was removed by diethylether extraction and the remaining DNP-lysine detected by absorbance at 435nm EXAMPLE 2 35 Preparation of 0 6M O methvhsourea solution A 0.6M O-methylisourea solution was prepared by a modified procedure based on the methods of Chervenka and Wilcox (1956), Shields et al, (1959), Mauron and Bujard (1964) and Kassell and Chow (1966) Four grams of barium hydroxide octahydrate were WO 97/02489 PCT/NZ96/00066 - 11 - added to approximately 16 ml of preboiled boiling distilled deiomsed water which had been preboiled for 10 mm to remove carbon dioxide The solution was heated to near boiling then added to 2 g of O-methylisourea (sulphate salt) in a 40 ml centrifuge tube The solution was left to cool for 30 mm before centrifugmg at 6400g for 10 nun. The 5 supernatant was retained and the precipitate was washed with approximately 2 ml of distilled deiomsed water before recentrifuging The washings were added to the supernatant and the pH checked If the pH of the solution was lower than 12 then it was assumed that conversion of the sulphate salt to the free base was incomplete and the solunon was remade. However, if the pH was above 12 then the pH was adjusted to the 10 appropriate pH for guanidination (pH 10.6-11.2), and made up to 20 ml with distilled deiomsed water EXAMPLE 3 A Preparation of a heated lactose/casein mixture 15 A heat treated lactose/casem mixture, which contained 250 g lactose and 750 g lactic casein, was prepared by mixing the two components in 4 litres of distilled deiomsed water then freeze drying the suspension and autoclaving the dried mixture for 3 5min at 121°C The autoclaved sample was ground through a 1mm mesh The resultant mixture simulated a protein having undergone early to late Maillaid damage (Gall 1989) 20 B. Preparation of protein sources To ensure that at least one of the protein sources was sufficiently heat damaged to allow sizeable differences between the assays, approximately 1 kg of skim milk powder was autoclaved for 3 mm at 121 °C before use The autoclaved skim milk powder along with 25 a selection of readily available feedstuffs including wheat meal, blood meal, soybean meal, meat and bone meal, dried maize, cottonseed meal and a pelleted lucerne based mix containing 55% lucerne, 10% meat and bone meal and 5% each of blood, wheat, barley, maize, sorghum, soybean, broil meals, were each ground through a 0 5 mm mesh The blood meal, soybean meal and wheat meal represented processed feedstuffs which were 30 expected to be of high quality whereas the other materials, being subjected to a higher degree of processing during manufacture, were expected to have a lower overall protein quality EXAMPLE 4 35 Optimisation of reaction time for guanidination of unheated partially purified proteins 5-10 mg samples were incubated for 1,2 and 3 days m 0 6M O-methylisourea, pH 10 6, at 21°C±2°C in a shaking waterbath with the reagent to lysine ratio being greater than WO 97/02489 - 12 - PCT/NZ96/00066 1000. The samples were then reduced to diyness and the homoarginine and lysine contents determined. In preliminary studies investigating the optimal time for guanidination of two unheated 5 protem sources (lysozyme and unheated casein), carried out over 1, 2 and 3 day incubation periods, near complete conversion of lysine to homoarginine (greater than 98%) was achieved in all cases EXAMPLE 5 10 Optimisation of the reaction time for guanidination of heated lactose/casem 5-10 mg samples of heated lactose/casein were incubated for 1, 3, 7 and 14 days in 0 6M O-methylisourea, pH 10 6, at 21°±2°C in a shaking waterbath, with the reagent to lysme ratio greater than 1000 The samples were subsequently reduced to dryness and the homoarginine content was determined 15 The conversion of lysine to homoargmine in heated lactose/casem was investigated using incubation times ranging from 1 to 14 days The yield of homoarginine over the 14 day period is shown in Fig 1 Maximal guanidination was achieved after 3 to 7 days mcubanon m the O-methylisourea solution, although there was no significant difference 20 between homoarginine yields observed after 1, 3, 7 or 14 days incubation EXAMPLE 6 Optimisation of the reaction time and pH for guanidination of digesta The optimal incubation time was determined after incubating 5-10 mg samples of rat ileal 25 digesta in 0 6M O-methylisourea, pH 10 6, at 21±2°C in a shaking waterbath, for 1, 3, 7, 14 and 21 days, with the reagent to lysine ratio greater than 1000 The samples were reduced to diyness and the homoarginine content was determined The ileal digesta had been obtained from rats given either an unheated casein based diet or a heated lactose/casem based diet 30 The optimal reaction mixture pH was determined after incubating 5-10 mg samples of ileal digesta from rats fed unheated casein and from rats fed heated lactose/casein in 0 6M O-methylisourea at pH 9 8, 10 2, 10 6, 11 0, 11 4 at 21±2°C in a shaking waterbath, with the reagent to lysine ratio being greater than 1000 (or for 1 to 7 days in 0 6M 35 O-methylisourea at pH 10 6 in a shaking waterbath) The samples were then ieduced to dryness and the homoargmine content was determined WO 97/02489 PCT/NZ96/00066 - 13 - The optimum incubation times for maximal guanidination of digesta from rats fed unheated casern and heated lactose/casem were determined (Fig.2a). Maximal conversion of lysine to homoarginine in the digesta of rats fed unheated casern was achieved with a 1 day incubation, after which the levels appeared to decline slightly, although this trend 5 was not statistically significant In contrast, a 7 day incubation time was required to achieve maximal guanidination of lysine m the digesta of rats fed the heated lactose/casem, although there was no significant difference between homoargmine yields for the reaction mixtures after 3, 7 or 14 days incubation There was also no significant difference in the amount of the homoarginine determined in the heated lactose/casein 10 digesta after incubation in 0.6M O-methylisourea foi 1 and 3 days. The optimum reaction mixture pH for maximal guanidination of digesta of rats fed unheated casern and heated lactose/casem was also determined (Fig 2b) The pH optimum for the guanidination of lysine in digesta of rats fed the unheated casein was 15 approximately 10 6, although the amounts of homoarginine obtained from guanidination mixtures at pH's ranging from 10.2 to 11 4 were not significantly different From the pH range examined in this experiment, the pH required for optimal guanidination of digesta from rats fed the heated lactose/casem was between 11 0 and 11 4 20 EXAMPLE 7 Determination of endogenous amino acid loss Enzymatically hydrolysed casern (EHC), containing peptides no larger than 5000 Daltons, are fed to a group of animal in the same mannei as described above for animals fed test foodstuffs The EHC fed animals are slaughtered and digesta sampled from the terminal 25 ileum The digesta is centrifuged at 6500 x g for 10 rmn The supernatant is then ultrafiltered using a Centnprep 10 ultrafiltration device The resulting filtrate (containing peptides smaller than 10000 Daltons) is discarded, while the retentate is pooled with the precipitate from the previous centrifugation step This fraction containing large endogenous peptides and protein, is then dried down and analysed for ammo acids and 30 chromium This measure of endogenous loss is then used to correct apparent amino acid digestibility in a test foodstuff to true ammo acid digestibility as shown in the equations below and Moughan 1991, and Butts el a 1, 1991 Endogenous ileal ammo acid flow = 35 Amino acid concentration in ileal digesta X Chromium in diet Chromium in digesta WO 97/02489 PCT/NZ96/00066 - 14- EXAMPLE 8 Digestibility study Sprague-Dawley male rats, of approximately 150g bodyweight, were housed individually in stainless steel wire-bottomed cages in a room maintained at 22±2°C, with a 12 h 5 light/dark cycle A. In a first study, two semi-synthetic test diets were formulated to each contain lOOg/kg crude protem Two enzymatically hydrolysed casein (EHC) based diets were also formulated to allow determination of endogenous ileal amino acid flows (Moughan et al, 10 1990, Butts et al, 1991) Chromic oxide was included in each diet as an indigestible marker The ingredient compositions of the diets are given in Table 1 below B In a second study, eight semi-synthetic test diets were formulated to each contain lOOg/kg crude protein An EHC based diet was also formulated to allow determination 15 of endogenous ileal lysine flow as above Chromic oxide was included (0.5%) in each diet as an indigestible marker The ingredient compositions of the diets are given m Table IA below WO 97/02489 PCT/NZ96/00066 - 15 - TABLE 1. Ingredient compositions (g kg'1 an dry weight) of the expenmental diets given to the laboratory rat. 5 EHC1 Unheated casein EHC2J Heated lactose/casein3 Wheat starch 625.7 639.7 583 584 9 Soybean oil 50 50 50 50 Purified cellulose 50 50 50 50 10 Sucrose 100 100 100 100 Vitamin/mineral mix4 39 3 39.3 39 3 39 3 Lactose - - 42 7 - Casein - 116 - - Heated lactose/casem - - - 170 8 15 EHC 130 - 130 - Chromic oxide 5 5 5 5 All diets were formulated to contain equal crude protein contents 'Enzymatically hydrolysed casern containing diet for determining endogenous amino acid 20 loss foi the unheated casein diet 2Enzymaftcally hydrolysed casern containmg diet for determining endogenous amino acid loss for the heated lactose/casein diet "'Heated lactose/casein was prepared as described in Expenmental example 3 JVitamin/mineral mix was formulated to meet the requirements for vitamins and minerals 25 as described by the National Reseaich Council (National Academy of Sciences, 1972) TABLE 1A. Ingredient compositions! (g kg-1 air dry weight) of the experimental diets. EHC2 Blood meal Wheat meal Meat and bone meal Soybean meal Heated skim milk powder Dned maize Lucerne based mix Cottonsei meal Wheat starch 625 7 646 7 - 572 7 542 7 495 7 - 355 7 504.7 Soybean oil 50 50 50 50 50 50 - 50 50 Purified cellulose 50 50 - 50 50 50 - 50 50 Sucrose 100 100 20 7 100 100 100 - 100 100 Vitamin/mineral mix4 39 3 39 3 39 3 39 3 39 3 39 3 39 3 39.3 39.3 EHC 130 - - - - - - - - Blood meal - 109 - - - - - - - Wheat meal - - 885 - - - - - - Meat/bone meal - - - 183 - - - - - Soybean meal - - - - 213 - - - - Heated skim milk powder - - - - - 260 - - - Dried maize - - - - - - 955 7 - Lucerne based mix3 - - - - - - - 400 Cottonseed meal - - - - - - - - 251 Chromic oxide 5 5 5 5 5 5 5 5 5 'All diets were formulated to contain equal crude protein contents 2Enzymatically hydrolysed casein diet used for determining endogenous amino acid losses at the terminal ileum, the EHC contained free amino acids and small peptides (<2000 Da) 3The lucerne based mix consisted of 55% lucerne, 10% meat and bone meal and 5% each of blood, wheat, barley, maize, sorghum, soybean and broil meals and was initially in a pelleted form.. 4Vitamin/mineral mix was formulated to meet the requirements for vitamins and minerals in the final diets as described by the National Research Council (National Academy of Sciences, 1972) WO 97/02489 PCT/NZ96/00066 - 17- For both studies A and B the diets were randomly allocated to the rats such that in study A there were six rats on each diet, and m study B there were a mimmum of five rats on each diet The rats were given the diets for a 14 day period. On each day each rat received its respective diet as nine meals given hourly (0830h to 1630h). At each meal 5 time the diet was freely available for a ten minute period The feed containers were weighed after each meal Water was available at all times On the fourteenth day of the study, from 5.5 to 7 hours after the start of feeding, the rats were asphyxiated in carbon dioxide gas and then decapitated. The 20cm of ileum immediately anterior to the lleo-caecal junction was dissected out The dissected ileum was washed with distilled 10 deiomsed water to remove any blood and hair and carefully dried on an absorbent paper towel The digesta were then gently flushed from the ileum section with distilled deiomsed water from a syringe The digesta fiom the rats fed the test diets were then freeze-dned ready for chemical analysis The pH of the digesta of rats fed the EHC diet was adjusted to approximately pH 3 with 6M HC1, to minimise protease activity The 15 EHC digesta were then centnfuged at 6400g for 30 mm at 3±1°C and the precipitate was washed and recentrifuged The washings were pooled with the supernatant the supernatant underwent ultrafiltration in a Centriprep 10 disposable ultrafiltration device after which the filtrate was discarded and the retentate washed and underwent ultrafiltration for a second time The resulting retentate was added to the precipitate from 20 the centnfugation step and freeze-dned ready for chemical analysis CHEMICAL ANALYSIS Study A Amino acids contents were determined in triplicate 5 mg digesta samples and 25 quadruplicate 5 mg diet samples using a Waters ion-exchange HPLC system, utilising post-column o-phthalaldehyde derivatisation and fluorescence detection, following hydrolysis in 6M glass-distilled HCI containing 0 1% phenol for 24 hr at 110±2°C in evacuated sealed tubes Cysteine, methionine, proline and tryptophan were not determined Where appropnate, the weight of each ammo acid was calculated using free 0 ammo acid moleculai weights For the determination of reactive lysine, the samples were incubated for 7 days in 0 6M O-methylisourea pH 10 6(pH 110 for the digesta samples), at 21 °C in a shaking waterbath, with the reagent to lysine ratio being greatei than 1000, before being dried 5 down and analysed for amino acid content as described above WO 97/02489 PCT/NZ96/00066 - 18 - Study B Amino acid contents were determined in duplicate 5 mg diet and digesta samples and quadruplicate 5 mg semi-synthetic diet samples using a Waters ion-exchange HPLC system, utilising post-column mnhydrin denvatisation and detection using absorbance at 5 570nm and 440nm, following hydrolysis in 6M glass-distilled HC1 containing 0 1% phenol for 24 hr at 110±2°C in evacuated sealed tubes. Cysteine, methionine and tryptophan were not determined as they are destroyed during acid hydrolysis The weight of each amino acid was calculated using free amino acid molecular weights 10 Reactive lysme contents were determined m duplicate 5 mg feedstuff and digesta samples and quadruplicate 5 mg diet samples by incubation for 1, 7 and 7 days respectively in 0.6M O-methylisourea, pH 10 6 (pH 11 0 for the digesta samples), at 21°C m a shaking waterbath, with the reagent to lysine ratio being greater than 1000 After incubation, the samples were dried down using a Speedvac concentrator (Savant Instruments, Inc, 15 Farmmgdale, NY, USA) and analysed for amino acid content as described above. The chromium contents of diet and ileal digesta samples for both study A and study B were determined m duplicate on an Instrumentation Laboratory atomic absorption spectrophotometer following the method of Costigan and Ellis (1987). 20 REACTIVE LYSINE CONTENT STUDY A The reactive lysme contents of the unheated casein and the heated lactose/casein were compared using the guanidination method (where homoargmine levels were equated to reactive lysine levels), the FDNB-reactive lysine method and conventional amino acid 25 analysis The guanidination conditions used were incubation for 24 hr in 0 6M O-methylisourea, pH 10 6 at 21°C±2°C in a shaking waterbath, with the reagent to lysine ratio bemg greater than 1000 The reactive lysine level of the heated lactose/casem was then extrapolated using Fig 1 to determine the ieactive lysine content using the optimal 7 day incubation period The results are shown m Table 2 below 35 WO 97/02489 PCT/NZ96/00066 - 19- TABLE 2. Reactive lysine content in unheated casein and heated lactose/casein determined using the FDNB and guanidination methods, and "total lysine" levels determined by conventional amino acid analysis. Lysine content (mmol g" 1 casein) FDNB' Guanidination2 Total3 Unheated casein 051 0 55 0.50 Heated lactose/casein 0 31 0.33 0.38 'FDNB analyses used a correction factor of 1 05 for both samples. 2Guanidination analyses consisted of 24hr incubation in 0.6M O-methylisourea, pH 10 6 15 ma shaking waterbath at 21 ±2° C with the reagent to lysine ratio bemg greater than 1000, followed by conventional acid analysis Reactive lysine values were corrected to an optimal 7 day incubation time using Fig 1 3Total analyses consisted of conventional acid hydrolysis and amino acid quantitation Values are means of triplicate analyses 20 REACTIVE LYSINE CONTENT STUDY B The reactive lysine content for five of the protein sources was determmed using the guarudination method and the FDNB-reactive lysine method and compared to total lysine content determined usmg conventional ammo acid analysis The guanidmation conditions 25 are set out above The results are shown in Table 2A below TABLE 2A. Reactive lysme contents (mg g-1 sample) of several protein sources determined using the FDNB or Guanidination methods in comparison with total lysine contents (mg g-1 sample) determined usmg conventional amino acid analysis 30 Reactive lysine Total lysine FDNB Guanidination Blood meal 84 4 88 0 89 1 Wheat meal 3 1 3 1 3 5 Meat and bone meal 30 4 34 6 36 5 Soybean meal 27 1 JZ J Cottonseed meal 14 7 14 4 20 6 The correction factors used for the FDNB method were 1 06 for blood meal, 1 03 for wheat meal, 1 08 for meat and bone, 1 04 for soybean meal and 1 05 for cottonseed meal and were determined as described in the materials and metliods WO 97/02489 PCT/NZ96/00066 -20- Results of Study A The amount of reactive lysine in the unheated casein ranged from 0.51mmol g'1 casein determined using conventional amino acid analysis to 0.55mmol g"1 casern determmed using the guanidination method. Generally the amounts of lysine m the unheated casein, 5 where it can be assumed that all the lysine is available, compared quite well between methods, with a less than 10% difference between the three methods The reactive lysine content of the heated lactose/casein determined using the FDNB and guanidination methods also agreed well (0 31mmol g'1 casern and 0 33 mmol g'1 casein respectively), and differed by less than 7% In contrast, the total lysine level determined using 10 conventional amino acid analysis was considerably higher (almost 20%), than for the other two methods Results of Study B Reactive lysme determined using guanidination was generally similar or higher than 15 FDNB-reactive lysine content for all five protein sources Generally good agreement was found between the reactive lysine contents of the feedstuffs determined using the FDNB and guanidination methods, especially for blood meal, wheat meal and cottonseed meal For soybean meal and meat and bone meal the 20 reactive lysme contents determined using the FDNB method were lower than those determined using guanidination Since theoretically the guamdination method cannot overestimate reactive lysine, it would appear that these differences are most likely an artefact of the FDNB method in which correction factors must be used 25 In an unprocessed protein source the reactive lysine content should be equivalent to the "total" lysme content, where total lysine is the lysine determined by conventional ammo acid analysis In contrast, in a protein source which has sustained early heat damage the total lysine content may be higher than the reactive lysine content due to reversion of lysine during acid hydrolysis In some processed protein sources where more severe 0 processing damage has occurred, structurally altered lysine derivatives may be acid-stable In this case reactive and total lysme values would be expected to be similar. For the blood meal, meat and bone meal and soybean meal in Study B the reactive lysine content determined using the guanidination method was similar to the total lysine content, suggesting that these protein sources either did not contain structurally-altered lysine 5 derivatives, or if they were present, they were in a form that is stable to acid conditions For wheat meal the reactive lysine content was lower than the total lysine content suggesting that some reversible modification of lysine may have occurred For dried maize, the lucerne based mix, cottonseed meal and heated skim milk powder, the reactive WO 97/02489 PCT/NZ96/00066 -21 - lysine content was considerably lower than total lysine, reflecting protein sources in which lysine had undergone early Maillard type reactions during processing. Cottonseed meal undergoes considerable heat processing, in order to reduce the toxicity of the anti-nutritional factors known to be present (Berardi and Goldblatt, 1980), while the skim 5 milk powder in the present study was subjected to controlled heating in our laboratory Determination of true ileal amino acid digestibility in unheated casein and heated lactose/casein of Study A. 10 The rats appeared healthy throughout the 14 day study There was no sign of faecal particles in the stomach contents of the rats at post-mortem, indicating that coprophagy had not occurred, at least on the last day of the study Meal intakes were relatively constant over the first five meals on the last day of study and therefore a relatively constant flow of digesta through the gut should have been achieved (Fig 4). 15 The endogenous amino acid flows at the terminal ileum, determined using the "lactose-free" EHC 1 diet, were used to correct apparent amino acid digestibilities to true ones for the unheated casem diet The lactose-containing EHC 2 diet was used to correct apparent digestibilities to true digestibilities for the heated lactose/casem diet. Although the 20 absolute endogenous ammo acid flows appeared to be higher with the lactose-containmg EHC diet compared to the lactose-free EHC diet, there was no statistical difference (p<0 05) between the two diets for all amino acids determined, with the exception of histidine (Table 3) The variation in endogenous flows observed in the rats fed the lactose-containmg EHC diet was considerably greater than that observed for the rats fed 25 the lactose-free EHC diet 30 35 WO 97/02489 PCT/NZ96/00066 -22 - TABLE 3. Endogenous amino acid flows of rats fed an enzymatically hydrolysed casein diet with and without lactose. EHC with lactose EHC without lactose 5 (n=6) (n=5) Aspartic acid 501 723 Threonine 391 448 Serine 393 555 10 Glutamic acid 587 1133 Glycme 307 305 Alanine 225 336 Valine 296 418 Isoleucine 226 337 15 Leucine 307 405 Tyrosine Phenylalanine 172 210 235 209 Histidine 156 352 Lysine 229 285 20 Argimne 175 207 The true ileal ammo acid digestibility (conventional assay) of the unheated casein was very high with a mean digestibility (excludmg lysine) of about 95% (Table 4 below) The 25 digestibihty of the heated lactose/casein was significantly lower for all amino acids except glycme, alanine, phenylalanine and argimne The mean decrease in digestibility between the unheated casein and the heated lactose/casein (excluding lysine) was 3% units but was as high as 9% units for aspartic acid (Table 4) 30 35 40 WO 97/02489 PCT/NZ96/00066 -23 - TABLE 4. Mean true ileal amino acid digestibility (%) of unheated casein and heated lactose/casein in the growing rat determined using an ileal amino acid digestibility assay based on conventional amino acid analysis. 5 Unheated casein Heated Overall SE Significance lactose/casein1 Aspartic acid 96.0 87.3 1.37 ** Threonine 93 6 88 0 0.96 ** 10 Serine 89 7 85 0 1.03 ** Glutamic acid 93.1 89 4 0.74 ** Glycme 86.0 81.2 2.70 NS Alanine 97.2 93 0 0.67 NS Valine 96 7 92 4 0 73 ** 15 Isoleucine 94 8 90 1 0 82 ** Leucine 99.1 97 1 0.27 ** Tyrosine Phenylalanine 100 4 100 4 97 1 98 0 0.29 0.30 *** NS Histidme 95 8 86 0 0.88 * 20 Argimne 98 1 96 1 1.68 NS 'Heated lactose/casein was prepared as described m Experimental example 3. NS P>0 05, * P<0 05, ** P<0 01, *** P<0 001 25 The true ileal digestibility of lysine determined using the traditional ileal digestibility assay which utilises conventional amino acid analysis was very high in the unheated casem (99%) (Table 5 below). The same coefficient in the heated lactose/casem was considerably lower than 71% In contrast, the true ileal digestibility of reactive lysme determmed using the traditional ileal amino acid digestibility assay coupled with the 30 guanidination reaction (new method) yielded a digestibility of 86%, significantly higher than the digestibility found using conventional methods The amino acid digestibility data were subjected to a one-way analysis of variance for each ammo acid singly (GLM Procedure, SAS Institute Inc USA) 35 40 WO 97/02489 PCT/NZ96/00066 -24- TABLE 5. Mean true ileal lysine digestibility (n=6) for an unheated casein determined using the rat ileal digestibility assay and based on conventional amino acid analysis and for a heated lactose/casein (n=6) using the rat ileal digestibility assay coupled with either conventional amino acid analysis or the guanidination method (reactive lysme 5 digestibility coefficient). Unheated Heated lactose/casein1 Overall Significance casein SE Conventional Guanidination 10 amino acid method analysis Digestibility of lysine 98.8 70 5 85 92 2 01 *** 15 Endogenous ammo acid flows at the terminal ileum were corrected for, using the EHC/ultrafiltration method (Butts et al, 1991) and the appropriate EHC diet (See Table 1). 'Heated lactose/casem was prepared as described in Experimental example 3 20 2The reactive lysine digestibility was calculated as follows- Reactive lysine in the diet - (Reactive lysine in the disiesta - Endogenous lvsme) X 100 Reactive lysine in the diet 1 25 Where units are pg g"1 DMI. Comparison of tine ileal lysine digestibility (conventional assay) with true ileal reactive lysine digestibility for Study B 30 The rats appeared healthy throughout the 14 day digestibility study Meal intakes were relatively constant over the first six meals on the last day of study and therefore a relatively constant flow of digesta through the gut should have been achieved Mean meal intakes (g) ±SE for the first six meals on the last day were 1 7±0 08g for the wheat meal diet, 1.8±0 05g for the cottonseed diet, 1 9±0 06g for the meat and bone 35 diet, 2.0±0.06g for the soybean diet, 0.8±0.04g for the blood meal diet, 1 7±0.12g for the heated skim milk powder diet, 0.9±0 15g for the dried maize diet, 1.9±0.28g for the lucerne based mix diet and 1 7±0 07g for the EHC based diet True ileal digestibility values based on "total" lysine as determined using conventional 40 ammo acid analysis were compared with true ileal digestibility values for reactive lysme, determined following the guanidination reaction, for eight different protein sources (Table WO 97/02489 PCT/NZ96/00066 -25- 6 below). TABLE 6. Comparison of the meanl true ileal lysine digestibility (%) determined using conventional amino acid analysis (total) and true ileal lysine digestibility 5 (%) based on determined reactive lysine (reactive). Lysine digestibility Total' Reactive3 Overall SE Blood meal 96.3 96 7 0.41 NS Wheat meal 92 6 92 1 0 45 NS Meat and bone meal 88 9 91 5 0.76 NS Soybean meal 94 5 96.5 041 * Dried Maize 80 5 84.3 1 54 * Heated skim milk powder 69.1 94 0 1 11 *** Cottonseed meal 62 1 71.9 1.75 ** Lucerne based mix 74 2 86 3 0 63 *** 'For blood meal, wheat meal, soybean meal, meat and bone meal, heated skim milk powder and cottonseed meal n=8, for the dried maize, and lucerne based mix n=5 2Lysine digestibility was determmed using a true ileal amino acid digestibility assay (rat) and conventional amino acid analysis was used to quantitate total lysme m the diets and 25 digesta. ^Lysine digestibility was determined using a true ileal amino acid digestibility assay (rat) and the guanidination reaction was used to quantitate reactive lysine in the diets and digesta 30 For blood meal, wheat meal and meat and bone meal the digestibilities of total lysine and reactive lysine were high (generally greater than 90%) and there was no significant difference between total lysme digestibility and reactive lysine digestibility. For soybean meal, the total lysine digestibility, which was also high, was statistically significantly lower than the reactive lysine digestibility, although the actual difference was less than 35 three percentage units These results again reflect protein sources containing minimal amounts of acid-labile "damaged" lysme derivatives, and as such indicates that the new true ileal reactive lysme digestibility assay, may be a suitable alternative method for determining lysine 40 digestibility. For the dried maize, lucerne based mixed diet, cottonseed meal and heated skim milk powder, there were significant differences (p<0.5) between the total lysme digestibilities WO 97/02489 PCT/NZ96/00066 -26- and the reactive lysine digestibilities. Differences between the digestibility values for the two approaches were four, twelve, ten and 25 percentage units respectively. Further, total lysine digestibility underestimated the actual digestibility of reactive lysine 5 This result indicates that the conventional true ileal amino acid digestibility assay appears to be unsuitable for assessing lysine availability in heat processed feedstuffs as it underestimates the digestibility of structurally unaltered lysine. This result also shows that the new true ileal reactive lysine digestibility assay may provide a more accurate assessment of digestibility of structurally unaltered (available) lysine in processed protem 10 sources Digestible lysine (based on total lysine determined by conventional analysis) and digestible reactive lysine contents are shown in Table 7 below 15 TABLE 7. Mean1 digestible total lysine and mean digestible reactive lysine contents (g kg-1 sample) in several protein sources Digestible lysine Total' ReactiveJ Overall SE Blood meal 89.9 85 1 0.34 NS Wheat meal 3.2 2 9 0 02 *** Meat and bone meal 32.5 31 6 0 24 NS Soybean meal 30 6 31 2 0.12 * Dned maize 2.6 1 9 0 04 *** Heated skim milk powder 19 8 16 6 0.30 *** Cottonseed meal 12 9 10 3 0 29 *** Lucerne based mix 14 4 10 8 0 10 * * * 'For blood meal, wheat meal, soybean meal, meat and bone meal, heated skim milk powder and cottonseed meal n=8, for the dried maize, and lucerne based mix n=5 2Digestible total lysine was calculated from true ileal lysine digestibility (rat), with lysine determmed by conventional ammo acid analysis and from the total lysme 35 content m the protein source, also determined using conventional amino acid analysis 3Digestible reactive lysme was calculated from true ileal reactive lysine digestibility (rat, guanidination analysis), and the reactive lysine content of the protein source, also determined using guanidination. 40 For blood meal and meat and bone meal, there were no significant differences between the two values In contrast, there were statistically significant differences between the two values for the six remaining protein sources However, for soybean meal there was less WO 97/02489 PCT/NZ96/00066 -27- than a 2% difference between digestible total lysine and digestible reactive lysine. For the lucerne based mix, the dried maize, cottonseed meal and heated skim milk powder, all of which had undergone more severe heat processing, the differences (34%, 37%, 25% and 19%, respectively) between digestible total lysine and digestible reactive lysine were 5 quantitatively significant. This result indicates that digestible reactive lysine more accurately reflects available lysine in these protein sources than the conventional true ileal lysine digestibility assay which is known to overestimate lysme availability in processed feeds (Batterham 1990) 10 EXAMPLE 9 Recovery of acid stable amino acids during guanidination The amounts of other acid stable amino acids were compared when using the guanidination method and conventional ammo acid analysis for a range of protem sources, 15 with the aim of determining if the guanidination method interfered with the quantitation of these other amino acids The recovery of acid stable ammo acids is shown in Fig 3 Again conversion of lysme to homoarginine m the relatively unprocessed proteins (lysozyme, soy protem isolate, skim milk powder, lactic casein, whey protem concentrate, soy protem concentrate, rotary dried bloodmeal and soyabean meal), was high, ranging 20 from 97 to 100% In contrast, in wheatmeal and the more severely processed protein sources such as meat and bone meal and cottonseed meal the conversion of lysine to homoarginine was considerably lower The recoveries of almost all of the ammo acids in all protein sources examined were close to 100% The mam exception was histidme, where, m skim milk powder, wheat and soybean meals, recoveries well above 100% were 25 observed This may have been associated with the chromatographic procedure and may have been due to column ageing WO 97/02489 PCT/NZ96/00066 -28- EXAMPLE 10 Comparison of the true ileal digestibility of acid-stable amino acids in protein sources determined using conventional amino acid analysis or following guanidination of the diet and digesta prior to amino acid analysis 5 The true ileal digestibilities of amino acids, other than lysine, were determined using the true ileal digestibility assay applied to unguanidinated diet and digesta samples or to samples which had undergone guanidination and the results are given in Figure 9 For most of the protein sources tested, including wheat meal, soybean meal, blood meal, dned 10 maize, meat and bone meal, skim milk powder and a lucerne based mix, there was either no statistically significant or practical difference (less than 3 percentage units) for most (89%) of the amino acids between digestibility determined using conventional amino acid analysis with or without prior guanidination. In contrast for cottonseed meal there were significant and practical differences for nine of the ammo acids examined. There was no 15 one ammo acid for which digestibility differed between the two approaches for all protein sources. Depending on the level of accuracy required it may be possible to obtain digestibility coefficients for amino acids other than lysine following guanidination of the diet and digesta samples 20 EXAMPLE 11 Comparison of the true ileal amino acid (mcludmg reactive lysme) digestibility of a heated lactose/casein determined in the rat and pig Four kilograms of lactic casein and 1 25kg of lactose were mixed in suspension then 25 freeze dried The dned mixture was then autoclaved at 121 °C for 15min and ground through a 0.5mm mesh Semi-synthetic test diets were formulated to each contain 100 g kg-1 crude protein and to meet the nutnent requirements (except for the ammo acids) of the growing rat (National 30 Research Council, 1972) and growing pig (Agricultural Research Council, 1981)(Table 8) WO 97/02489 PCT/NZ96/00066 -29- TABLE 8 Ingredient compositions (g kg'1 air dry weight) of the experimental diets for the mterspecies comparison (rat and pig) of the true ileal reactive lysine digestibility of a heated lactose/casein mixture. EHC Rat Heated lactose/casein EHC Pig Heated lactose/casein Cornstarch 561 9 598 7 555 7 592.5 Soyabean oil 50.0 50 0 50.0 50.0 Cellulose 50 0 50 0 50.0 50 0 Sucrose 100 0 100 0 100 0 100.0 Vit/Min mix 39 3 39 3 45 5 45.5 Lactose 70 0 - 70 0 EHC 123 8 - 123 8 - Unheated casein - - - - Heated lactose/casein - 157 0 - 157 0 Chromic oxide 5 0 5 0 5 0 5 0 'Vitamin/mineral mix was formulated to meet the rats requirements for vitamins and minerals m the final diets as described by the National Research Council (1972) and for 20 the pigs requirements as described by the Agricultural Research Council (1981) The heated lactose/casem was used as the sole source of protem An EHC based diet was also prepared for each species of animal to allow determination of endogenous ileal amino acid flows 25 Forty 150g bodyweight Sprague Dawley male rats were housed as described previously Forty entire-male Landrace X (Landrace X Large White) pigs of 30 kg bodyweight were housed individually in steel metabolism crates at an ambient room temperature of approximately 23°C The rats and pigs were randomly allocated to their respective diets 30 such that there were eight animals on each of the diets The animals underwent a 14 day expenmental period On each day each rat or pig was given its respective diet at nine hourly meals (0830 to I630h) For the rats, the diet was available for ten minutes at each meal time Watei was freely available The pigs weie given a set and equal amount of food at each meal time The daily intake of the pig was pioportional to their ad lib energy 35 intake, based on tlie ratio of energy intake foi tlie lats on the 13th day of study to the rat's ad lib energy intake Prior to each meal the pig diets were mixed with water (1 1, w/v) Water was freely available between meals On the fourteenth day of study the rats and pigs were killed 6 hours after the first meal 40 and digesta were immediately collected from the terminal 20cm of ileum The rats were killed and digesta collected and processed as described previously The pigs were killed by an intra-cardial injection of sodium pentobarbitone administered with the animal under WO 97/02489 PCT/NZ96/00066 -30- halothane anaesthesia The ileal digesta samples for the pigs given the test diets were freeze dried and stored frozen (-20°C) while awaiting chemical analysis The EHC digesta samples were processed as described previously Ammo acids were determined on duplicate digesta samples and quadruplicate diet samples while chromium content was 5 determined in duplicate. The apparent and true ileal amino acid digestibilities for the heated lactose/casem given to the rat and pig are presented m Tables 9 and 10 There was no significant difference in apparent digestibility values determined between the two species Further, for only 10 alanine was there a significant difference in true digestibility estimates determined in either the rat or the pig. TABLE 9. Mean apparent ileal ammo acid digestibility (%) in a heated lactose/casein mixture, for the rat (n=8) and pig (n=8) 15 Pig Rat Overall SE Significance Aspartic acid 90 4 88 3 0 89 NS Threonine 87 4 86 8 1 10 NS 20 Senne 82.9 87.5 1 99 NS Glutamic acid 92 3 93 0 0 85 NS Proline 94 6 93 5 0 55 NS Glycine 67.3 61 2 5 32 NS Valine 87 8 90 8 1 07 NS 25 Alanine 92 2 92 4 0 80 NS Isoleucine 90 1 90 9 1 08 NS Leucine 95 5 96 2 0 38 NS Tyrosine 96 6 95 7 0 36 NS Phenylalanine 96.3 96 0 0 32 NS 30 Histidine 89 7 88 5 0 70 NS Argimne 90 0 90 6 2 10 NS Reactive lysine 94.6 94 1 0 47 NS NS, non significant, P> 0 05 40 WO 97/02489 -31 - PCT/NZ96/00066 TABLE 10. Mean true ileal amino acid digestibility (%) in a heated lactose/casein mixture, for the rat (n=8) and pig (n=8). Pig Rat Overall SE Significance Aspartic acid 96.6 95.0 0.89 NS Threonine 95.6 94.0 1.10 NS Serine 89.9 92.5 1.99 NS Glutamic acid 95.6 95.7 0.85 NS Glycine 78.5 74.4 5.32 NS Valine 96.4 96.9 0.80 NS Alanine 95.1 99.2 1.07 * Isoleucine 94.7 95.5 1.08 NS Leucine 98.8 99.6 0.38 NS Tyrosine Phenylalanine 99.3 98.9 99.3 99 4 0.36 0.32 NS NS Histidine 93.8 94.6 0.70 NS Argimne 95 0 94.3 2 10 NS Reactive lysme 98.2 98 0 0 47 NS 20 NS, non significant, P> 0 05, * P<0.05. These results demonstrate the applicability of the guanidation assay for feedstuffs for a range of animals From these results it is also apparent that the laboratory rat is a suitable 25 model animal for the growing pig EXAMPLE 12 Evaluation of the accuracy of the true ileal digestible reactive lvsme assay usmg a pig growth study 30 In this experiment two comparisons were made. (a) comparison of the lysine and protem depositions and liveweight gain m pigs fed heated skim milk powder with that m pigs fed an EHC/free amino acid diet (EHC diet A) formulated to contam a lysme level equal to the digestible lysme 35 level based on the traditional true ileal digestibility assay (reactive lysme content x total lysine digestibility), and (b) comparison of the lysme deposition m pigs fed the heated skim milk powder with that in pigs fed an EHC/free amino acid diet (EHC Diet B) formulated to contam a lysine level equal to the digestible lysme level based on the new true 40 ileal reactive lysme digestibility assay (reactive lysme content x reactive lysme digestibility) Approximately 130kg of skim milk powder was autoclaved m 5kg batches at 121oC for approximately lOmin then ground through a 1mm mesh. Samples were taken and ground SUBSTITUTE SHEET WO 97/02489 PCT/NZ96/00066 -32- to 0.5mm before analysis for total and reactive lysine contents. Three semi-synthetic diets containing either unheated skim milk powder (basal diet), heated skim milk powder or EHC as the sole source of nitrogen were formulated in order to determine the digestibility of the heated skim milk powder (Table 11) The EHC diet was used to determme the 5 endogenous lysine flow Chromic oxide (0 5%) was added as an indigestible marker. Sixteen pigs were housed in metabolism crates as described previously and allocated to either a basal skim milk powder diet or an EHC based diet. After 7 days the pigs on the basal skim milk powder diet were changed to the heated skim milk powder diet and fed for a further 7 days. After a 14 day trial period the pigs were slaughtered and ileal digesta 10 collected, processed and true ileal total lysine and reactive lysine digestibilities determined as described previously TABLE 11. Ingredient compositions (g kg'1 air dry weight) of the expenmental diets for the determination of true ileal reactive and total lysine digestibility of heated skim milk 15 powder for the evaluation of the new assay using a pig growth study Basal skim milk powder Heated skim milk powder EHC Cornflour 488 2 492 4 551 2 20 Maize Oil 50 0 50 0 50 0 Cellulose 50 0 50 0 50 0 Sucrose 100 0 100 0 100 0 Vit/Minmixl 50 0 50 0 50 0 Lactose - - 70 0 25 Skim milk powder 256 8 Heated skim milk powder - 252 6 EHC - - 123 8 Chromic Oxide 5 0 5 0 5 0 30 ' Vitamin/mineral mix was formulated to meet the pigs requirements for vitamins and minerals in the final diets as described by the Agricultural Research Council (1981) The true ileal reactive lysme digestibility for the heated skim milk powder (88 5%) was significantly higher than the true ileal total lysine digestibility (67 1%) Consequently the 35 heated skim milk powder used as a test feedstuff to evaluate the accuracy of the true ileal reactive lysine digestibility assay Based on tlie true ileal total lysme and reactive lysine digestibility coefficients for heated skim milk powder determined above, three test diets were formulated (Table 12) to 40 contam similar net energy levels and fibre levels and similar amino acids balances The heated skim milk powder was the sole source of nitrogen WO 97/02489 PCT/NZ96/00066 - 33 - TABLE 12. Ingredient compositions (g kg"1 air dry weight) of the experimental diets for the pig growth study. 5 Basal skim milk powder Cornflour 365.0 10 Maize Oil 50 0 Cellulose 50.0 Sucrose 100.0 Vit/Min mix1 50 0 Citric acid 15 Sodium citrate DiCalcium phosphate Magnesium oxiae Potassium chloride Lactose - 20 Skim milk powder 385 0 Heated skim milk powder EHC Asparagine 25 Aspartic acid Threonine Senne Glutamic acid Glycine 30 Alanine Valine Cysteine Methiomne Isoleucine 35 Leucine Tyrosine Phenylalanine Tryptophan Histidine 40 Argimne Proline Heated skim milk EHC Diet A EHC DietB powder 345 0 401.4 372.1 50 0 50 0 50 0 50 0 50.0 50.0 100 0 100 0 100 0 50 0 50.0 50 0 - 64 64 - 0.8 0 8 - 2 0 2 0 - 0.007 0 007 - 1 2 1 2 - 191 7 191 7 385 0 - 90 5 108 6 - 1 7 20 - 1 7 20 - 2 0 24 - 1.8 2 1 - 8 1 9 7 11 0 9 7 11 6 9 0 8 3 10 0 - 1 9 2.3 - 06 0 7 - 0 6 0 7 - 1 3 1 5 - 4 5 5 4 - 3 2 3 9 - 3 2 3 9 - 2 2 2 6 - 1 4 1 7 - 1 0 1 2 - 2 9 3 5 1 Vitamin/mineral mix was formulated to meet the pigs requirements for vitamins and 45 minerals in the final diets as described by the Agricultural Research Council (1981) Heated skim milk powder Heated skim milk powder based diet 50 EHC Diet A EHC + free amino acids formulated to contain a lysine level equal to the digestible lysine content of the heated skim milk powder determined usmg the conventional ileal digestibility assay (reactive lysine m heated skim milk powder x true WO 97/02489 PCT/NZ96/00066 -34- digestibility of total lysine (determined using conventional methods) for the heated skim milk powder) EHC Diet B: EHC + free amino acids formulated to contain a lysme level equal to the 5 digestible lysine content of the heated skim milk powder determined using the new ileal reactive lysine digestibility assay (reactive lysine in heated skim milk powder x true digestibility of reactive lysine (determined using the new method) for the heated skim milk powder) 10 The heated skim milk powder based diet was also tested independently to ensure that lysme was limiting Six 25kg hveweight entire male pigs were housed in metabolism crates designed for complete urine collection After a six day acclimatisation period, during which the pigs were fed a basal skim milk powder based diet, the diet for three of the pigs was changed to the heated skim milk powder diet which was fed for a further six 15 days. The remaining three pigs were fed the heated skim milk powder diet supplemented with lysine The total daily urine volume for each pig was determmed by collecting the daily urine output for each of the first three days of the six day test period Total urine was collected for each day of the last three days of this six day period into a bottle containing 25ml of 1 8M H2S04 per litre of urine The walls of the metabolism crates 20 were also washed down with distilled water After this 6 day period the pig's diets were swapped over and again total urine was collected for each day of the last 3 days of the 6 day period. Each individual pig's daily urine sample was analysed for creatinine content, after which, the samples were pooled and analysed for total nitrogen and urea content The daily unnaiy total nitrogen excretion of tlie pigs fed tlie heated skim milk powder diet 25 were then compared to tlie total nitrogen excietion for the pigs ieceiving the heated skim milk powder diet supplemented with lysine Growth trial Entire littermate Landrace x (Landrace x Large white) male pigs were housed at the Pig 30 Research Umt m a temperature controlled room mam tamed at 22±2oC The pigs were fed at 10.2% of their metabolic bodyweight and were given their respective daily allowance as three equal meals Any feed refusals were collected and weighed. At the end of the acclimatisation period the pigs were weighed and the feed level recalculated, again to equal 10 2% of their metabolic weight. At this time 8 pigs were slaughtered and their 35 body lysine content determined to provide a baseline lysine level When the pigs were slaughtered, the entire gut, gall bladder and bladder contents were removed and care was taken to recover any blood that was lost from the body The pig bodies were stored at -20 °C until they could be ground. The remainder of the pigs were fed there respective WO 97/02489 PCT/NZ96/00066 -35 - test diets in a sunilar manner as during the acclimatisation penod and every 7 days the pigs were reweighed and the feed level was adjusted accordingly. At the end of the 19 day test period the pigs were slaughtered and the bodies were processed as described above After grinding the pigs, whole body samples were taken and freeze dried Fat was 5 then extracted using the Soxhlet fat extraction technique before amino acid analyses were carried out. The deposition of lysine was calculated over the trial period after subtraction of the lysine contents calculated to be present in the animals at the start of the trial penod from the lysme content measured at the end of the trial period The lysine content of the animals at the start of the trial was estimated using a regression of the lysine content of 10 the baseline" pigs on liveweight The urinary nitrogen excretion in the pigs when fed the heated skim milk powder diet was significantly higher (P=0 021) than when fed the lysine supplemented diet (9 lg/day and 6 6g/day respectively), demonstrating that lysine was limiting in the heated skim milk 15 powder diet Furthermore, since the EHC diets were formulated with the same ratio of amino acids to lysme as present m the heated skim milk powder then these diets too were deemed to be limiting in lysine The pigs appealed to be healthy dunng the acclimatisation period and during the first few 20 days of the 19 day trial However, the majority of the pigs fed the EHC control diets then began to suffer from dianhoea The diarrhoea lasted on average for about two to three days after which the pigs appeared normal Apart from dunng the acclimatisation penod, the pigs generally consumed their diets readily and there were negligible food refusals dunng the tnal period The mean initial liveweight of the pigs fed the heated skim milk 25 powder diet was 29 2kg which was significantly higher (P<0 025) than that for the pigs fed the EHC diet A (25 7kg), but was not significantly different from that for the pigs fed the EHC diet B (28 0kg) Further, there was no significant difference m initial hveweights between the pigs fed either of the EHC diets Consequently, lysine and protein depositions as well as weight gain were compared statistically using initial 30 liveweight as a covanate The lysine and protein contents of the pigs at the beginning of the trial period were determined by regression of the lysine and protein depositions against liveweight for the baseline pigs slaughtered at the onset of the tnal Over the 19 day tnal penod the pigs (for all diets) on average increased their body lysine 35 content by 38.5% The lysine depositions of the pigs fed the heated skim milk powder were significantly higher than for the pigs fed the EHC control diet A containing a lysine level equivalent to the digestible lysine in the heated skim milk powder determmed using the traditional digestibihty assay (reactive lysme content x total lysine digestibility)(Table WO 97/02489 PCT/NZ96/00066 -36- 13). In contrast, the lysine depositions of the pigs fed the heated skim milk powder were similar and not significantly different to those for the pigs fed the EHC control diet B containing a lysme level equivalent to the digestible lysme in the heated skim milk powder determined using the new assay (reactive lysine content x reactive lysine digestibility). 5 The protein depositions and liveweight gain of the pigs fed the heated skim milk powder diet were significantly higher than that found m both groups of pigs fed the EHC diets. Though the pigs fed EHC diet B deposited more protein and grew faster than those pigs given EHC diet A 10 TABLE 13. Mean lysine deposition (g day-1), protein deposition (g day-1) and bodyweight gam (g day-1) determined over 19 days m pigs fed a heated skim milk powder diet, an EHC diet A and EHC diet B 15 Heated skim milk powder EHC Diet A1 EHC Diet B2 Overall SE Significance Lysine Deposition Protein Deposition Weight gain 20 9 6" 140 8a 764 5a 5.1b 80 lb 539 0b 8.r 107 T 627 5C 1 46 13 41 46 48 *** *** *** 'EHC Diet A • EHC + free amino acids formulated to contain a lysine level equal to the digestible lysine content of the heated skim milk powder determined using the conventional ileal digestibility assay (reactive lysine in heated skim milk powder x true digestibility of total lysme (determined using conventional methods) for the healed skim 25 milk powder) 2EHC Diet B EHC + free ammo acids formulated to contain a lysine level equal to the digestible lysine content of the heated skim milk powder determined using the new ileal reactive lysine digestibility assay (reactive lysine m heated skim milk powder x true digestibility of reactive lysine 30 (determined usmg the new method) for the healed skim milk powder) Since most of the pigs fed the EHC based diets suffered from diarrhoea in the early part of the tnal, the lysme and protein depositions and weight gains of the pigs over the last 12 days of the tnal were also examined The mean initial liveweight at the 35 commencement of this 12 day period for the pigs fed the heated skim milk powder diet was 33kg which was significantly higher (P<0 023) than that for the pigs fed the EHC diet A (28.8kg), but was not significantly different from that for the pigs fed the EHC diet B (31.3kg). Further, there was no significant difference in initial liveweights between the pigs fed either of the EHC diets. Consequently again, lysine and protein depositions as 40 well as weight gain were compared statistically usmg initial liveweight fitted as a WO 97/02489 PCT/N Z96/00066 - 37 - covariate. The lysine and protein contents of the pigs at the beginning of the 12 day period were determined by regression of the lysine and protein depositions against liveweight for the baseline pigs slaughtered at the onset of the trial. 5 Over the 19 day trial period the pigs (for all diets) on average increased their body lysine content by 26 9% Again, lysine depositions of the pigs fed the heated skim milk powder were significantly higher than for the pigs fed the EHC control diet A (Table 14), the lysine depositions of the pigs fed the heated skim milk powder were similar and not significantly different to those for the pigs fed the EHC control diet B The protein 10 deposition and liveweight gain of the pigs fed the heated skim milk powder diet was significantly higher than that found in both groups of pigs fed the EHC diets TABLE 14. Mean lysine deposition (g day"1), protein deposition (g dky ) and bodyweight gam (g day'1) determmed over 12 days m pigs fed a heated skim milk powder 15 diet, an EHC diet A and EHC diet B Heated skim milk powder EHC Diet A1 EHC Diet B2 Overall SE Significance level Lysine Deposition Protein Deposition Weight gain 10.2" 156 4a 790 0a 5 0b 86 8h 575 8b 8 T 109 2b 647 9h 2.23 22.45 98.08 ** *** *** 'EHC Diet A EHC + free amino acids formula led to contain a lysine level equal to the 25 digestible lysine content of the heated skim milk powder determined using the conventional ilea! digestibility assay (reactive lysine in heated skim milk powder x true digestibility of total lysine (determined using conventional methods) for the heated skim milk powder) 2EHC Diet B EHC ^ free amino acids formulated to contain a lysine level equal to the 30 digestible lysine content of the heated skim milk powder determined using the new ileal reactive lysine digestibility assay (reactive lysine in healed skim milk powder x true digestibility of reactive lysine (determined using the new method) for the heated skim milk powder) 35 Lysme and protein depositions and liveweight gams were also determined over the final 12 days on an initial liveweight basis instead of using initial liveweight as a covariate (Table 15) The results calculated in this manner showed similar trends as those that use initial liveweight as a covariate WO 97/02489 PCT/NZ96/00066 -38- TABLE 15. Mean lysine deposition (g day-1 kg-1 initial bodyweight), protein deposition (g day-1 kg-1 initial bodyweight) and bodyweight gain (g day-1 kg-1 initial bodyweight) determined over 12 days m pigs fed a heated skim milk powder diet, an EHC diet A and EHC diet B. 5 Heated skim milk powder EHC Diet A1 EHC Diet B2 Overall SE Significance level Lysine Deposition Protein Deposition Weight gam 031" 49° 25 2a 0.18" 2.9b 19 0b 0 28a 3.5b 20 8b 0.076 0.75 3.26 ** *** ** 'EHC Diet A EHC + free amino acids formulated to contain a lysine level equal to the digestible lysine content of the heated skim milk powder determined using the 15 conventional ileal digestibility assay (reactive lysine in heated skim milk powder x true digestibility of total lysine (determined using conventional methods) for the heated skim milk powder) 2EHC Diet B EHC + free amino acids formulated to contain a lysine level equal to the digestible lysine content of the healed skim milk powder determined using the new ileal 20 reactive lysine digestibility assay (reactive lysine in heated skim milk powder x true digestibility of reactive lysine (determined using the new method) for the heated skim milk powder) The mean lysme deposition for the pigs fed the heated skim milk powder measured over 25 tlie total 19 day expenmental penod was significantly different from that for the pigs fed EHC Diet A Furthei. the lysine deposition of the pigs fed the heated skim milk powder was similar and not significantly different from the pigs fed EHC Diet B Over the final 12 days of the tnal (excluding an initial period of two to three days where many of the EHC fed pigs suffered from dianhoea) the results similar to that observed over the total 30 tnal penod Tlie protein deposition and liveweight gams of pigs fed the heated skim milk powder was significantly higher than pigs fed eitiier of the EHC control diets although the protem deposition and liveweight gams of pigs fed EHC Diet B were closer to that of the pigs fed the skim milk powder based diet than were those of the pigs fed EHC Diet A This result demonstrates firstly, the inadequacy of the traditional assay in predicting 35 available lysme in the heated skim milk powder and secondly, the accuracy of the new true ileal reactive lysine digestibility assay in determining lysine availability in a heated skim milk powder and gives confidence in the suitability of the new assay for accurately predicting lysine availability in processed feedstuffs to be used in least cost feed formulation 40 WO 97/02489 PCT/NZ96/00066 -39- CONCLUSION In accordance with the present invention there are provided methods for determining the digestibility co-efficient of an ammo acid in a foodstuff More especially, the present invention provides methods for determining both the reactive lysine digestibility co-5 efficient and the digestible reactive lysine content in a foodstuff The methods of the present invention represent an alternative ammo acid assay for many amino acids and for lysine, a significant advance on conventional assays which do not adequately take account of the effects of processing on lysine bioavailability The approach taken m the present invention was to regard altered lysine residues as "lost" to a subject for protem synthesis 10 and to attempt to directly determine the absorption of reactive lysine residues remaining in a foodstuff The present methods detected that as much as 40% of the original lysine in a heated foodstuff may be destroyed or modified by heat treatment 15 By determining the reactive lysine content of the diet and digesta rather than the total lysine content (as in the traditional digestibility assays) the problem of overestimated lysme values due to interference from lysme destroyed or modified by heat treatment, and heat induced lysme derivatives reverting back to lysine during acid hydrolysis is avoided. 20 Lysme was found to be highly digestible in the unheated or unprocessed foodstuffs in the order of 90 to 100% digestible The determined lysine digestibility co-efficient in the heated or processed foodstuffs, when lysine content was qualified using conventional amino acid analysis, was some 25 to 35 percentage units lower than for the unheated 25 foodstuffs This decrease in digestibility was considerably greater than the 2-10 percentage unit decrease observed for the other amino acids In contrast, the true reactive lysine digestibility co-efficient in heat damaged or processed foodstuffs obtained using guanidination was significantly higher at 80 to 90%, approximately 13 percentage units lower than m the unheated or unprocessed foodstuffs This shows better agreement with 30 the decrease m digestibility observed for the other amino acids It is clear that the use of conventional ammo acid analysis significantly underestimates lysine digestibility The reactive lysine digestibility estimate obtained using the guanidination method therefore clearly provides more reliable estimates than those 35 obtained using conventional amino acid analysis There are many methods for determining reactive lysine in feeds, all with their inherent advantages and disadvantages, but until now there has been no ideal method that allows WO 97/02489 PCT/NZ96/00066 -40- for the routine determination of digestible reactive lysine in processed feeds, especially heat processed feeds. The new true ileal reactive lysine digestibility methods described here are means by which this can be achieved Furthermore, the use of the rat as an experimental animal may allow for a relatively routine assessment of available lysine m 5 processed feedstuffs. This is an economically important assay where lysine is often the first limiting ammo acid particularly in pig and poultry diets Accurate analysis of available lysme m feedstuffs will allow for the formulation of feedstuffs designed to met the nutritional requirements 10 of the subjects to which they are to be fed. The methods of the invention can also be used to determine optimal lengths of time over which feeds can be stored Both factors will allow for reduction m feed costs, where the feeds utilised will have a higher nutritional value through optimised available lysine content 15 It will be further appreciated by those persons skilled m the art that the present description is provided by way of example only and that the scope of the invention is not limited thereto WO 97/02489 -41 - PCT/NZ96/00066 REFERENCES Batterham E S; Anderson, L.M.; Baigent, R.D; Darnell, R.E; Teverner, M.R.A Companson of the availability and ileal digestibility of lysine in cottonseed and soyabean meals for grower/finisher pigs. Brit J. Nut 1990, 64, 663-677. 5 Batterham, E.S Availability and utilisation of ammo acids for growing pigs. Nutr Res Rev. 1992, 5, 1-18 Berardi, L C, Goldblatt, L A 1980. Gassypol \n Toxic Constituents of Plant Feedstuffs 10 Liener, J.E. ed Academic Press, New York, USA 1980, 183-273. Booth, V H Problems in the determination of FDNB-available lysine J Sci Food Agric 1971, 22, 658-666 15 Butts, C A , Moughan, P.J , Smith, W C Endogenous amino acid flow at the terminal ileum of the rat determined under conditions of peptide alimentation J Sci. Food Agric 1991, 44, 175-187 Carpenter, K J The estimation of the available lysine in animal-protem foods Biochem, 20 J 1960, 77, 604-610 Chervenka, C H , and Wilcox, P.F Chemical derivatives of chymotrypsinogen II Reaction with O-methylisourea J. Biol Chem 1956,222,635-647 25 Costigan, P , and Ellis, K J .Analysis of fecal chiomium derived from controlled release marker devices NZ J Tech 1987,3,89-92 Desrosiers, T , Savoire, L , Bergeron, G , and Parent, G Estimation of Lysine Damage m Heated Whey Proteins by Furosine Determinations m Conjunction with the Digestion 30 Cell Technique J Agric Food Chem 1989,37. 1385-1391 Gall, M P.J Nutritional and physiological effects of short-term feeding of an early Maillard browned casein to growing pigs Ph D Dissertation, Massey University, Palmerston North, New Zealand, 1989 35 Hendriks, W H , Moughan P J , Boer, H., and van der Poel, A F.B Effects of extrusion on the dye-bmdmg, fluorodimtrobenzene-reactive and total lysme content of soybean meal and peas Amm Feed Sci Tech 1994, 48, 99-109 WO 97/02489 PCT/NZ96/00066 -42- Hurrell, R.F., and Carpenter, K.J. The estimation of available lysme in foodstuffs after Maillard reactions Prog. Fd. Nutr. Sci 1981, 5, 159-176. Kassell, B., and Chow, R.B. Modification of the basic trypsin inhibitor of bovine 5 pancreas The e-amino groups of lysme and the amino terminal sequence. Biochemistry 1966,5 3449-3453 Knipfel, J.E Nitrogen and energy availabilities m foods and feeds subjected to heating In Progress in Food and Nutritional Science Maillard Reactions in Food; Eriksson, C., 10 Ed., Pergamon Press, Oxford, 1981, Vol 5 Numbers 1-6, Chapter 2.1. Mauron, J., and Bujard, E Guanidination, an alternative approach to the determination of available lysine in foods Proc 6th hit Nutr Congr 1964, 489-490 15 Moughan, P.J Towards an improved utilisation of dietary amino acids by the growing pig In Recent Advances in Animal Nutrition, Haresign, W , and Cole, D.J.A., Eds , Butterworth-Heinemann Ltd, Oxford, 1991, pp 45-64 Moughan, P J , Dairagh, A J., Smith, W C , and Butts, C A Perchloric and trichloroacetic 20 acids as precipitants of protein in endogenous ileal digesta from the rat. J. Sci Food Agric 1990, 52, 13-21 National Academy of Sciences Nutrient requirements of the laboratory rat Ln Nutrient requirements of laboratoiy animals, National Academy of Sciences Washington, DC 25 1972, pp 56-93 Schmitz, M Moglichkeiten und Grenzen der Homoarginin-Markierungsmethode zur Messung dei Proteinverdaulichkeit beim Schwein Ph.D Dissertation, Christam-Albrechts-Umversitat, Kiel, 1988 30 Shields, G S . Hill, R L , and Smith, E L Preparation and properties of guanidmated mercuripapain J Biol Chem. 1959,234 1747-1753 35 Rutherfurd, B M and Moughan, P J Guanidination of Lysine in Selected Dietary Proteins. J Agric Food Chem 1990, 38, 209-211 </div>

Claims

<div class="application article clearfix printTableText" id="claims"> INTELLECTUAL PROPERTY OFFICE OF N.Z. 3 0 JAN 2002 RECEIVED

1. A method for determining the reactive lysine digestibility co-efficient of a foodstuff which method comprises the steps of: 5 (a) introducing a marker into the foodstuff to be analysed; (b) feeding the foodstuff to a non-human subject for a predetermined period of time; (c) obtaining a sample of the foodstuff digesta from the subject; (d) determining the digestible reactive lysine content of the foodstuff by: 10 (i) introducing a lysine derivatising agent into the foodstuff; and (ii) determining the digestible reactive lysine content of the foodstuff by measuring the equivalent derivatised lysine content in the foodstuff; (e) determining the digestible reactive lysine content in the foodstuff digesta by. (i) introducing a lysine derivatising agent specific for the e-amino 15 group of lysine, into the foodstuff digesta; and (ii) determining the digestible reactive lysine content of the foodstuff digesta by measuring the equivalent derivatised lysine content in the foodstuff digesta; (f) measuring the marker concentration in both the foodstuff and foodstuff 20 digesta; (g) expressing the reactive lysine content of both the foodstuff and foodstuff digesta per gram of the marker; and (h) calculating the reactive lysine digestibility co-efficient. 25

2. A method for determining the digestible reactive lysine content of a foodstuff which assay comprises the steps of: (a) calculating the reactive lysine digestibility co-efficient according to the method of claim 1; and .30 (b) determining the digestible reactive lysine content of the foodstuff by multiplying the value for reactive lysine content of the foodstuff by the reactive lysine digestibility co-efficient

3. A method according to claim 1 or claim 2 wherein the foodstuff is formulated into 35 a diet before adding the marker and feeding the foodstuff to the subject

4. A method according to any one of claim 1 to 3 wherein the foodstuff is a heat processed foodstuff. - 44 - ~ "

5. A method according to any one of claims 1 to 4 wherein the marker is a spectrophotometric marker.

6. A method according to claim 5 wherein the spectrophotometric marker is chromic 5 oxide. •

7. A method according to any one of claims 1 to 6 wherein foodstuff digesta is extracted from the terminal ileum 10

8. A method accordmg to any one of claims 1 to 7 wherein the foodstuff digesta is dried before treatment with the derivatising agent.

9 A method according to any one of claims 1 to 8 wherein the derivatising agent is 0-methylisourea. 15

10. A method as defined in claim 1 for determining the reactive lysine digestibility co-efficient of a foodstuff substantially as herein described with reference to any example thereof and with or without reference to the accompanying drawings. 20

11. A method as defined in claim 2 for determining the digestible reactive lysine content of a foodstuff substantially as herein described with reference to any example thereof and with or without reference to the accompanying drawings. INTELLECTUAL PROPERTY OFFICE OF NZ. 3 0 JAN 2002 RECEIVED </div>