Αlpha-amylase is ubiquitous in nature. It has a central

Transcription

1 1096 MCCLEARY ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 85, NO. 5, 2002 FOOD COMPOSITION AND ADDITIVES Measurement of -Amylase Activity in White Wheat Flour, Milled Malt, and Microbial Enzyme Preparations, Using the Ceralpha Assay: Collaborative Study BARRY V. MCCLEARY, MARIAN MCNALLY, and DYMPNA MONAGHAN Megazyme International Ireland Ltd., Bray Business Park, Bray, County Wicklow, Ireland DAVID C. MUGFORD BRI Australia Ltd., North Ryde, New South Wales 2113, Australia Collaborators: C. Black; R. Broadbent; M. Chin; M. Cormack; R. Fox; C. Gaines; P. Gothard; S. Home; E. Howes; C. Johnson; R. Keeping; M. Koliatsou; M. Lindhauer; R. Marins de Sa; R. Martin; D. Monaghan; U. Nees; R. Nishwitz; G. Palmer; J. Panozzo; J. Recabarren; S. Roumeliotis; S. Seddig; V. Solah; M. Sonnet; H. Themeier This study was conducted to evaluate the method performance of a rapid procedure for the measurement of -amylase activity in flours and microbial enzyme preparations. Samples were milled (if necessary) to pass a 0.5 mm sieve and then extracted with a buffer/salt solution, and the extracts were clarified and diluted. Aliquots of diluted extract (containing -amylase) were incubated with substrate mixture under defined conditions of ph, temperature, and time. The substrate used was nonreducing end-blocked p-nitrophenyl maltoheptaoside (BPNPG7) in the presence of excess quantities of thermostable -glucosidase. The blocking group in BPNPG7 prevents hydrolysis of this substrate by exo-acting enzymes such as amyloglucosidase, -glucosidase, and -amylase. When the substrate is cleaved by endo-acting -amylase, the nitrophenyl oligosaccharide is immediately and completely hydrolyzed to p-nitrophenol and free glucose by the excess quantities of -glucosidase present in the substrate mixture. The reaction is terminated, and the phenolate color developed by the addition of an alkaline solution is measured at 400 nm. Amylase activity is expressed in terms of Ceralpha units; 1 unit is defined as the amount of enzyme required to release 1 mol p-nitrophenyl (in the presence of excess quantities of -glucosidase) in 1 min at 40 C. In the present study, 15 laboratories analyzed 16 samples as blind duplicates. The analyzed samples were white wheat flour, white wheat flour to which fungal -amylase had been added, milled Submitted for publication May The recommendation was approved by the Methods Committee on Commodity Foods and Commodity Products as First Action. See Official Methods Program Actions, (2002) Inside Laboratory Management, July/August issue. Corresponding author s barry@megazyme.com. malt, and fungal and bacterial enzyme preparations. Repeatability relative standard deviations ranged from 1.4 to 14.4%, and reproducibility relative standard deviations ranged from 5.0 to 16.7%. Αlpha-amylase is ubiquitous in nature. It has a central role in the mobilization of starch reserves in the germination of cereals and many legumes (1), and it is a key quality parameter in the commercial utilization of most cereals. In malting, the aim is to induce maximum synthesis of α-amylase with minimum respiration of starch reserves. This amylase then plays a central role (along with β-amylase and limit dextrinase) in the mashing process in converting starch reserves to fermentable sugars (2). In baking, it is essential that the α-amylase level is adequate to hydrolyze starch from damaged starch granules to fermentable sugars, but not so high as to result in excessive starch dextrinization during the baking process (3); this would lead to sticky crumb and problems in the slicing of bread. Microbial amylases find widespread application in the starch processing industries (4). Numerous food-related problems can be attributed to the presence of trace levels of α-amylase (usually bacterial α-amylase), e.g., instability of starch-based products such as long-life custards and fillings. Many methods for the assay of α-amylase are described in the literature. The simplest and most direct are those based on the measurement of reducing-sugar increase on hydrolysis of starch (5, 6). Such methods directly relate to international units of activity (1 unit is the amount of enzyme that releases 1 µmol glucose reducing-sugar equivalents per minute under the defined assay conditions). The major limitation of these assays is that they cannot be performed on materials that contain high levels of endogenous reducing sugar (7), e.g., wheat flours and malts. To overcome this problem, a range of other methods has been developed over the years. Some of these use dye-labeled starch substrates, either soluble (8, 9) or insoluble (10), whereas other methods are based on the decrease in

2 Table 1. Interlaboratory study results a for determination of -amylase in white wheat flour, milled malt, and microbial preparations White wheat flour, Ceralpha units/g Milled malt, Ceralpha units/g Fungal enzyme preparation, Ceralpha units/g Bacterial enzyme preparation, Ceralpha units/g Lab A D B C E G F H I J K L M P N O b 302 b c 103 c b 170 b b 224 b c 513 c b 257 b b 352 b No. of labs d No. of outliers e Mean f g s r h s R RSD r, % i RSD R, % j r k R l a Each value is the average of duplicate assays. For each laboratory each test sample (coded A O) was extracted once, and then the assay was performed in duplicate. Duplicate determinations for a single extraction are a method requirement. b Cochran (repeatability) outlier. c Single Grubbs (reproducibility) outlier. d Number of laboratories included in calculations. e Number of outliers omitted from calculations. f Mean of replicate data. g Repeatability (within laboratory) standard deviation. h Reproducibility (between laboratories) standard deviation. i Repeatability relative standard deviation. j Reproducibility relative standard deviation. k Repeatability value (2.8 s r ). l Reproducibility value (2.8 s R ). MCCLEARY ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 85, NO. 5,

3 1098 MCCLEARY ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 85, NO. 5, 2002 the starch iodine color on hydrolysis of β-limit dextrin by α-amylase (11, 12). Both of these types of methods are very dependent on the nature of the starch, or starch fraction, used as substrate, and cannot be regarded as primary assays because they need to be related to another assay that uses a defined substrate. Even with reducing-sugar assays, activity values obtained are dependent on the type of starch used as substrate. In 1987, the Ceralpha method was introduced (13, 14) for the measurement of cereal and microbial α-amylases. This method is based on the use of a defined oligosaccharide substrate, end-blocked p-nitrophenyl maltoheptaoside, in the presence of 2 exo-acting enzymes, namely, amyloglucosidase and α-glucosidase (maltase). The blocking group on the nonreducing end of the oligosaccharide prevents hydrolysis of the oligosaccharide by the exo-acting enzymes. When the oligosaccharide is hydrolyzed by endo-acting α-amylase, the nonblocked nitrophenyl maltosaccharide produced is instantaneously hydrolyzed to glucose and free p-nitrophenol by the combined action of amyloglucosidase and α-glucosidase (i.e., the amyloglucosidase hydrolyzes nitrophenyl maltooligosaccharides to glucose and p-nitrophenyl α-d-glucoside, and the α-glucosidase hydrolyzes the p-nitrophenyl α-d-glucoside to glucose and p-nitrophenol). The reaction is terminated by the addition of an alkaline solution, which also develops the phenolate color. This is the only assay for cereal and microbial α-amylases that uses a defined substrate. This procedure was adopted as a standard procedure by the International Association for Cereal Science and Technology (ICC; Method No. 303). More recently, the Ceralpha reagent mixture was modified by replacing the amyloglucosidase plus α-glucosidase (maltase) by a single enzyme, thermostable α-glucosidase (15). The modified Ceralpha method, after extensive interlaboratory evaluation (16), recently replaced the Farrand procedure (11) as the standard method of the United Kingdom Milling and Baking Industry (16). The aim of the current collaborative study was to determine the performance characteristics of the method for selected flours and microbial enzyme preparations. Collaborative Study Eight well-mixed (in a plastic jar) test materials (containing α-amylase) of white wheat flour (2 materials, one containing fungal α-amylase), milled malt (2 materials from different malt houses), 2 powder fungal (Aspergillus niger) preparations and 2 powder bacterial (Bacillus licheniformis) preparations were provided as randomly coded blind duplicates to 15 collaborators. The fungal and bacterial preparations were prepared from original commercial materials from Novozymes (Bagsvaerd, Denmark) by freeze-drying in the presence of lactose. Two in-house reference samples (white wheat flour and milled malt) were also provided (along with the blind duplicates for analysis) to familiarize collaborators with the method. To facilitate the selection of the buffers and the extraction/dilutions to use, laboratories were advised which samples were white wheat flour, milled malt, fungal preparations, and bacterial preparations. Collaborators were requested to perform single extractions of each material by the enclosed method, but duplicate assays with aliquots of test material extract were also requested. Duplicate determinations are a necessary part of the described methodology, and the values shown in Table 1 are averages of duplicate assays. Results were evaluated according to AOAC guidelines for calculating interlaboratory performance statistics and identifying outliers (17). The significance level for outlier identification was set at α = HORRAT values were not determined because the HORRAT equation does not readily apply to units of enzyme activity. AOAC Official Method Measurement of -Amylase Activity in White Wheat Flour, Milled Malt, and Microbial Enzyme Preparations Ceralpha Assay First Action 2002 (Applicable to the measurement of α-amylase activity in white wheat flour, milled malt, fungal enzyme preparations, and bacterial enzyme preparations. α-amylase activity is described empirically in terms of Ceralpha units. Absorbance range is linear between 0.10 and 1.2.) See Table A for the results of the interlaboratory study supporting acceptance of the method. A. Principle Test portions are extracted with salt solution or buffer at room temperature or 40 C. Extracts are clarified by centrifugation or filtration and diluted. Aliquots of diluted extract are incubated with the substrate mixture under defined conditions of ph, temperature, and time. The substrate is nonreducing end-blocked p-nitrophenyl maltoheptaoside (BPNPG7) in the presence of excess quantities of thermostable α-glucosidase. The blocking group in BPNPG7 prevents hydrolysis of this substrate by exo-acting enzymes such as amyloglucosidase, α-glucosidase, or β-amylase. When the substrate is cleaved by endo-acting α-amylase, the nitrophenyl oligosaccharide is immediately and completely hydrolyzed to p-nitrophenol and free glucose. The reaction is terminated, and the phenolate color developed by the addition of an alkaline solution is measured at 400 nm. Results are expressed in Ceralpha units; 1 unit is defined as the amount of enzyme required to release (in the presence of excess α-glucosidase) 1 µmol p-nitrophenol per minute at 40 C. B. Apparatus (a) Grinding mill. Centrifugal, air-cooled, with 12-tooth rotor and 0.5 mm sieve, or similar device. Alternatively, cyclone mill can be used for small samples. (b) Bench centrifuge. Holding test tubes, (l), with rating of ca 1000 g. (c) Water bath. Maintaining 40 ± 0.1 C. (d) Vortex mixer (e) ph Meter (f) Stop-clock timer (digital)

4 MCCLEARY ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 85, NO. 5, Table A. Interlaboratory study results for determination of -amylase in white wheat flour, milled malt, and microbial preparations, using the Ceralpha assay Material Mean, α-amylase units a No. of labs b s a r s a R RSD r, % RSD R, % r c R d White wheat flour A (0) White wheat flour B (0) (containing fungal α-amylase) Milled malt A (3) Milled malt B (0) Fungal preparation A (2) Fungal preparation B (0) Bacterial preparation A (2) Bacterial preparation B 68 15(0) a b Ceralpha units. Each value is the number of laboratories retained after elimination of outliers; each value in parentheses is the number of laboratories removed as outliers. c r = 2.8 s r. d R = 2.8 s R. (g) Top-loading balance. Correct to 0.01 g. (h) Spectrophotometer. With 10 mm pathlength cell, or 10 mm pathlength flow-through cell; operating at 400 nm. (i) Pipets. 200 µl with disposable tips. Alternatively, motorized hand-held dispenser can be used. (j) Positive displacement pipetters. To accurately deliver 200 µl and 3 ml. (k) Dispenser. For dispensing stopping reagent, C(e); set to deliver 3 ml. (l) Glass test tubes. Approximately mm, 17 ml capacity, suitable for centrifugation at ca 1000 g in centrifuge, (b). (m) Test tube racks. Holding mm tubes, (l). (n) Thermometer. Reading 40 ± 0.1 C. (o) Glass fiber filter. Whatman GF/A, 9 cm; or equivalent. (p) Volumetric flasks. 25, 50, and 100 ml. C. Reagents (a) Concentrated sodium malate buffer. 1M, ph 5.4. Dissolve g malic acid, 70 g NaOH, and 58.4 g NaCl in 900 ml deionized water. Add 6.0 g CaCl 2 2H 2 O and dissolve. Adjust ph to 5.4 by dropwise addition of NaOH (4M) or HCl (4M). Then add 1.0 g sodium azide as preservative. (Addition of sodium azide is optional. It is added as an antimicrobial agent. It is a poisonous chemical and should be used according to supplier s safety instructions.) In the presence of sodium azide, this reagent is stable at room temperature for >1 year. If sodium azide is not added, the reagent is stable at 4 C for 2 weeks. (b) Malate buffer (for cereal and fungal -amylases). 50mM, ph 5.4. Dilute 50 ml concentrated sodium malate buffer, (a), to 1 L with deionized water. Store at room temperature for 6 months if sodium azide is added. The reagent is stable at 4 C for 1 week if sodium azide is not added. (c) Sodium maleate buffer (for bacterial -amylase). 0.1M, ph 6.5. Dissolve 23.2 g maleic acid and 11.6 g NaCl in 1600 ml deionized water, and adjust ph to 6.5 with 4M (160 g/l) NaOH. Add 0.6 g CaCl 2 2H 2 O and 0.2 g sodium azide, and adjust volume to 2 L. Store at room temperature for 6 months. (d) Sodium chloride (1%, w/v) plus calcium chloride (0.03%, w/v) solution. Dissolve 10 g NaCl and 0.3 g CaCl 2 2H 2 O in 1 L deionized water. To improve stability of bacterial α-amylase extracts, add bovine serum albumin (BSA; 0.5 g/l) plus sodium azide (0.2 g/l). The reagent is stable at room temperature for 6 months in the presence of sodium azide. In the absence of sodium azide, the reagent is stable at room temperature for 6 h. (e) Stopping reagent (trisodium phosphate solution). 2% (w/v), ph Dissolve 20 g Na 3 PO 4 in 900 ml deionized water. Adjust ph to 11.0 by careful addition of 4M HCl, and then adjust volume to 1 L. The reagent is stable at room temperature for 6 months. (f) Substrate solution. Amylase HR Reagent (Megazyme International Ireland Ltd., Bray Business Park, Bray, County Wicklow, Ireland). Dissolve contents of 1 vial [containing 54.5 mg blocked (4,6-O-benzilidine)-p-nitrophenyl maltoheptaoside and 125 units (U) thermostable α-glucosidase] in 10 ml boiled and cooled water. Divide substrate into ca 3 ml portions and store at 20 C in polypropylene containers between use. On thawing, store cold. The dissolved substrate is stable for 7 days at 4 C, and >12 months at 20 C. (g) p-nitrophenol solution. 10mM. Dissolve mg p-nitrophenol (spectroscopic grade) in 90 ml water. Adjust to 100 ml. The reagent is stable for 6 months when stored in the dark at 4 C. Items (a), (e), and (f) are supplied in the Ceralpha α-amylase Assay Kit available from Megazyme International Ireland

5 1100 MCCLEARY ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 85, NO. 5, 2002 Ltd., but preparations of reagents and buffers that meet the criteria may also be used. D. Preparation of Test Samples and Standards (a) Test samples. Grind ca 50 g grain or malt in grinding mill to pass 0.5 mm sieve, B(a). Transfer all material to wide-mouth plastic jar, and mix well by shaking and inversion. Industrial enzyme preparations supplied as liquids or fine powders do not require grinding. (b) p-nitrophenol standard for spectrophotometic calibration. 0.05mM. Dilute 1 ml p-nitrophenol solution, C(g), to 200 ml with stopping reagent, C(e). The assay relies on measurement of the absolute quantity of p-nitrophenol released. It is therefore essential that the spectrophotometer reading is correct for a standard quantity of p-nitrophenol. The absorbance of the diluted standard (0.05mM in stopping reagent) should be ± at ph 11.0 (400 nm). Perform this check each day that a batch of test samples is analyzed. Check the linearity of the spectrophotometer readings by using a series of dilutions of the concentrated p-nitrophenol standard solution, C(g), prepared with stopping reagent, C(e). Measure absorbance of selected dilutions at 400 nm, using stopping reagent, C(e), as the blank. Table B shows such a series of dilutions and the corresponding expected absorbance values. Plot absorbance vs p-nitrophenol concentration, and determine linear region of spectrophotometric response. E. Extraction of -Amylase (a) White wheat flour. Accurately weigh 3.0 g flour into 50 ml flask. Add 20.0 ml ph 5.4 sodium malate buffer, C(b), and vigorously swirl flask. Let enzyme extract for min at 40 C in temperature-controlled water bath, B(c), with occasional swirling. Filter solution through Whatman GF/A glass fiber filter, B(o), or centrifuge aliquot at 1000 g for 10 min. Initiate assay for enzyme activity within 1 h. (b) Milled malt and fungal preparations. Accurately weigh 0.5 g milled malt or fungal preparation into 100 ml volumetric flask. Dilute to volume with solution of NaCl (1%, w/v) plus CaCl 2 2H 2 O (0.03%, w/v), C(d). Let enzyme extract for min at room temperature with occasional swirling. Filter, if necessary, aliquot of extract through Whatman GF/A glass fiber filter, B(o), or centrifuge, B(b), at 1000 g for 10 min. Perform serial dilutions [i.e., dilute 1.0 ml filtrate to 10.0 ml with sodium malate buffer (ph 5.4), C(b), etc.] of filtrate to obtain suitable concentration of enzyme resulting in absorbance value in the range of ). Initiate assay for enzyme activity within 1 h. (c) Bacterial preparations. Accurately weigh 0.5 g powdered bacterial preparation into 100 ml volumetric flask (or dispense 0.5 ml liquid preparation with positive displacement dispenser). Dilute to volume with solution of NaCl (1%, w/v) plus CaCl 2 2H 2 O (0.03%, w/v), C(d). Let enzyme extract for min at room temperature with occasional swirling. If necessary for clarification, centrifuge aliquot of extract at 1000 g for 10 min. (Do not filter these extracts because some bacterial α-amylases are unstable when filtered.) Table B. Dilution of concentrated p-nitrophenol standard and the corresponding expected absorbance values Dilution of 10mM p-nitrophenol a Perform serial dilutions of supernatant to obtain suitable concentration (i.e., concentration that gives reaction absorbance value of ) of enzyme for assay [i.e., dilute 1.0 ml filtrate to 10.0 ml with sodium maleate buffer (ph 6.5), C(c), etc.]. Assay enzyme activity immediately. F. Assay of -Amylase Final concn of p-nitrophenol, mm Expected absorbance at 400 nm 200 µl to 25 ml µl to 25 ml µl to 25 ml µl to 25 ml µl to 50 ml µl to 100 ml a Microliter volumes should be measured with a micropipet, B(i), and volumes should be adjusted to 25, 50, or 100 ml in volumetric flasks, B(p). (a) Extracts from white wheat flour. Dispense 200 µl substrate solution, C(f), into test tubes, B(l), in duplicate, and preincubate tubes (uncapped) at 40 C for 5 min. Concurrently, preincubate extracts of white wheat flour, E(a), (ca 5 ml in uncapped tubes) at 40 C for 5 min. To each tube containing 200 µl substrate solution, add 200 µl preequilibrated wheat flour extract directly to bottom of tube. Mix tube contents vigorously on vortex mixer, B(d), and immediately incubate at 40 C for exactly 20 min. Terminate reaction by adding 3.0 ml stopping reagent, C(e), and mix tube contents vigorously on vortex mixer. Read absorbance of solutions and reaction blank, (c), at 400 nm vs water set at zero. Average absorbance readings for duplicate tests. If absorbance reading exceeds 1.2, dilute extract further with malate extraction buffer, C(b), and repeat assay of extract and reaction blank, (c). (b) Extracts from malt, fungal, and bacterial preparations. Dispense 200 µl substrate solution, C(f), into test tubes, B(l), and preincubate tubes (uncapped) at 40 C for 5 min. Concurrently, preincubate appropriately buffered and diluted extracts of milled malt, E(b), fungal preparation, E(b), or bacterial preparation, E(c), in uncapped tubes at 40 C for 5 min. To each tube containing 200 µl substrate solution, add 200 µl preequilibrated extract of milled malt, E(b), fungal preparation, E(b), or bacterial preparation, E(c), directly to bottom of tube. Mix tube contents vigorously on vortex mixer, B(d), and immediately incubate at 40 C for exactly 10 min from time of addition. At end of incubation period, add 3.0 ml stopping reagent, C(e), and mix tube contents vigorously. Read absorbance of solutions and reaction blank, (c), at 400 nm vs water set at zero. Average absorbance readings for duplicate tests. If absorbance reading exceeds 1.20, dilute ex-

6 MCCLEARY ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 85, NO. 5, tract further, and repeat assay of extracts and reaction blank. If results of duplicate tests exceed 3% of average absorbance value, repeat assay. (c) Reaction blank. Add 200 µl flour extract or diluted extract of milled malt, E(b), fungal preparation, E(b), or bacterial preparation, E(c), (containing α-amylase) to 3.0 ml stopping reagent, C(e), and mix well with vortex mixer, B(d). Add 200 µl substrate solution, C(f), to each tube and mix well. Read absorbance of solution at 400 nm vs water. Blanks compensate for any possible yellow color in the extract, which is often the case for white wheat flour. G. Calculations One Ceralpha unit (a name assigned by Megazyme) is defined as the amount of enzyme which, in the presence of excess quantities of thermostable α-glucosidase, will release 1 µmol p-nitrophenol from end-blocked p-nitrophenyl maltoheptaoside in 1 min at 40 C, under the defined assay conditions. Calculate α-amylase content (units/g or ml) in test sample as follows: α-amylase (Ceralpha units/g or ml) = A A I V 1 V Z ε W F a b r e where A a = average absorbance of the 2 assay solutions at 400 nm; A b = absorbance of reaction blank at 400 nm;i=incubation time (20 min for white wheat flour extract or 10 min for milled malt, fungal preparation, and bacterial preparation extracts); V r = volume in the reaction tube (3.4 ml); Z = volume of diluted extract used in the assay (0.2 ml); ε = the absorptivity of p-nitrophenol at 400 nm in 2% trisodium phosphate solution, ph 11.0 (1.81 mmol 1 mm 1 ; or 18.1 for a 10 mm cell path length); absorbance values are measured in a 10 mm spectrophotometer cuvette; V e = extraction volume (20 ml for white wheat flour or 100 ml for milled malt, fungal preparation, and bacterial preparation); W = weight of test portion (g); and F = extraction solution dilution factor. H. Controls and Precautions α-amylase is present at high levels in all body fluids. Thus, disposable gloves should be worn. The water used to dissolve the substrate solution, C(f), must be devoid of α-amylase. To ensure this, heat distilled or deionized water to boiling, and cool to <30 C before using to dissolve substrate. The blank absorbance value of freshly prepared substrate should be <0.07. If the absorbance values for a particular assay are >1.2, dilute the enzyme extract with the appropriate buffer and assay again. When the difference between duplicate tests is >3% of the average absorbance value, repeat the assay. The accuracy of the results obtained is dependent on the absorbance accuracy of the spectrophotometer, which should be checked as in D(b). Ref.: J. AOAC Int. 85, (2002) Results and Discussion Collaborators data were evaluated statistically according to AOAC protocols using AOAC-supplied software. Of the 120 pairs of assay results reported, 7 were statistical outliers (17). One outlier result was obtained by each of 7 laboratories. Cochran (repeatability) outliers were reported by Laboratories 2 and 15 (Sample E/G), Laboratory 9 (Sample I/J), and Laboratories 10 and 14 (Sample M/P). Single Grubbs (reproducibility) outliers were reported by Laboratory 8 (Sample I/J) and Laboratory 12 (Sample E/G). No statistically significant outliers were found in results for wheat flour (Samples A/D and B/C), malt flour (Sample F/H), fungal enzyme (Sample K/L), or bacterial enzyme (Sample N/O). Repeatability relative standard deviation (RSD r ) values ranged from 1.4 to 14.4%. The highest value was obtained for wheat flour (Sample B/C) that had been supplemented with a fungal α-amylase preparation. Without this value, the RSD r range was %. Reproducibility relative standard deviation (RSD R ) values ranged from 5.0 to 16.7%. The highest RSD r value, 14.4%, obtained for the white wheat flour sample that was supplemented with fungal α-amylase, can be attributed to problems associated with the homogeneous mixing of the 2 components with very different levels of the analyte. Wheat flours used for bread making contain approximately 0.1 units (Ceralpha) of α-amylase per gram, whereas fungal preparations can contain from 1000 to units per gram. Fungal α-amylase is added to flours that contain very low levels of this enzyme. A certain level is required to ensure that the flour will yield good bread, i.e., there has to be enough α-amylase to hydrolyze starch (from damaged starch granules) to fermentable sugars, required by yeast for gas production. The method specifies that α-amylase in bacterial enzyme preparations be extracted by centrifugation rather than filtration. Centrifugation is used because filtration of some bacterial preparations results in a significant loss of α-amylase activity, possibly as a result of oxidation of the enzyme at the filter surface. This loss of activity due to filtration has been observed only for bacterial preparations and not the other materials. The α-amylase assay procedure described here has the advantage of being specific and simple to use. A major advantage of this procedure is that this is the only method for the measurement of α-amylase in cereal and microbial samples that uses a defined substrate. Collaborators Comments Silja Home (VTT Biotechnology, Espoo, Finland) highlighted the instability of the bacterial α-amylase enzyme to filtration. All collaborators were advised to avoid using filtration for the clarification of bacterial extracts. This problem was not experienced with the enzymes from the other sources as used in this study.

7 1102 MCCLEARY ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 85, NO. 5, 2002 Recommendations The Study Director recommends that the Ceralpha method for the measurement of α-amylase activity in white wheat flour, milled malt, and microbial preparations be adopted First Action. Acknowledgments We thank the following collaborators for their participation in this study: Mae Chin, Deltagen (Australia) Pty. Ltd., Boronia, Victoria, Australia Peter Gothard, Wheat Analytical Group, Monsanto UK, Trumpington, Cambridge, UK Silja Home, VTT Biotechnology, Espoo, Finland Christine Johnson, Enzyme Bio-Systems, Beloit, WI Richard Keeping, Elizabeth Howes, and Juani Recabarren, Heygates Ltd., Bugbrooke Mills, Northampton, UK Meinolf Lindhauer, Ursula Nees, and Heinz Themeier, Federal Center for Cereal, Potato and Lipid Research, Institute for Cereal, Potato and Starch Technology, Detmold, Germany Ron Martin and Charles Gaines, U.S. Department of Agriculture, Agricultural Research Service, Soft Wheat Quality Laboratory, Wooster, OH Dympna Monaghan, Megazyme International Ireland Ltd., Bray Business Park, Bray, County Wicklow, Ireland Ralph Nishwitz and Mark Cormack, Barrett Burston Maltings, Richmond, Victoria, Australia Geoff Palmer, Richard Broadbent, Roberta Marins de Sa, and Maria Koliatsou, Department of Biological Sciences, Heriot-Watt University, Riccarton, Scotland Joe Panozzo and Cassandra Black, Agriculture Victoria Horsham, Victorian Institute for Dryland Agriculture, Horsham, Victoria, Australia Sophie Roumeliotis and Rebecca Fox, Adelaide University Australia, Waite Campus, Department of Plant Sciences, Glen Osmond, Australia Sylvia Seddig, Bundesanstalt fur Zuchtungsforschung an Kultupflanzen, Institut fur Strebphysiologie und Rohstoffqualitat, Grob Lusewitz, Germany Vicky Solah, Curtin University of Western Australia, Perth, Australia Me Sonnet, Belgomalt, Gembloux, Belgium References (1) Manners, D.J. (1973) in Plant Carbohydrate Biochemistry, J.B. Pridham (Ed.), Academic Press, New York, NY, pp (2) MacLeod, A.M. (1979) in Brewing Science, J.R.A. Pollock (Ed.), Vol. I, Academic Press, New York, NY, pp (3) Tipples, K.H. (1969) Baker s Dig. 43, (4) Norman, B.E. (1980) in Enzymes and Food Processing, G.G. Birch, N. Blakebrough, & K.J. Parker (Eds), Applied Science Publishers, London, UK, pp (5) Somogyi, M.J. (1952) J. Biol. Chem. 195, (6) Miller, G.L., Blum, R., Glennon, W.E., & Benton, A.L. (1960) Anal. Biochem. 1, (7) Kaufman, R.A., & Tietz, N.M. (1980) Clin. Chem. 26, (8) Babson, A.L., Tenney, S.A., & Megrew, R.E. (1970) Clin. Chem. 16, (9) McCleary, B.V. (1980) Carbohydr. Chem. 86, (10) Ceska, M., Birath, K., & Brown, B. (1969) Clin. Chim. Acta 26, (11) Farrand, E.A. (1964) Cereal Chem. 41, (12) Sandstedt, R.M., Kneen, E., & Blish, M.J. (1939) Cereal Chem. 16, (13) McCleary, B.V., & Sheehan, H. (1987) J. Cereal Sci. 6, (14) Sheehan H., & McCleary, B.V. (1988) Biotechnol. Tech. 2, (15) Ceralpha: α-amylase Assay Procedure (CER 07/00) Test Kit Booklet, Megazyme International Ireland Ltd., Bray, County Wicklow, Ireland (16) Standard Method 0018 of the Flour Testing Working Group of the Millers and Bakers Association of UK (1997) in Manual of Methods for Wheat and Flour Testing, S. Salmon (Ed.), 2nd Ed., Campden and Chorleywood Food Research Association, Chipping Campden, UK (17) Official Methods of Analysis (2000) 17th Ed., AOAC IN- TERNATIONAL, Gaithersburg, MD, Appendix D