WO2006032281A2

WO2006032281A2 - Method of preparing a dough-based product

Info

Publication number: WO2006032281A2
Application number: PCT/DK2005/000602
Authority: WO
Inventors: Lars Beier; Esben Peter Friis; Henrik Lundquist
Original assignee: Novozymes A/S
Priority date: 2004-09-24
Filing date: 2005-09-23
Publication date: 2006-03-30
Also published as: US20110305795A1; US20160081357A1; DK1794291T3; ES2399341T3; US20130136823A1; EP1794291B1; CA2610683A1; WO2006032281A3; AU2005287737A1; CA2610683C; US9439442B2; US9226510B2; EP1794291A2; AU2005287737B2; US20080003341A1

Abstract

Dough with a high sucrose content (such as cake dough) tends to inhibit the activity of an anti-staling amylase such as Novamyl, making it less effective to prevent the staling of dough-based products with high sucrose content such as cakes. A good anti-staling effect in cakes can be achieved by using a carefully selected anti-staling amylase with certain properties. Analysis of a 3D structure of Novamyl shows that sucrose may inhibit by binding in the active site. Sucrose docks into the active site of Novamyl differently from the substrate or inhibitor in published models 1QHO and 1QHP. This finding is used to design sucrose-tolerant variants.

Description

METHOD OF PREPARING A DOUGH-BASED PRODUCT

SEQUENCE LISTING AND DEPOSITED MICROORGANISMS

Sequence listing The present invention comprises a sequence listing.

Deposit of biological material

None.

FIELD OF THE INVENTION

The present invention relates to the use of anti-staling amylases in the preparation of dough or dough-based edible products with a high sucrose content.

BACKGROUND OF THE INVENTION

US 3026205 describes a process of producing baked confections and the products re¬ sulting therefrom by alpha-amylase.

WO 9104669 describes the use of a maltogenic alpha-amylase to retard the staling of baked products such as bread; the maltogenic alpha-amylase described therein is commercially available under the tradename Novamyl^® (product of Novozymes A/S). US 6162628 describes Novamyl variants and their use for the same purpose. Three-dimensional structures of Novamyl are published in US 6162628 and in the Protein Data Bank (available at http://www.rcsb.org/pdb/) with identifiers 1QHO and 1QHP.

SUMMARY OF THE INVENTION

The inventors have found that a high sucrose content dough (such as cake dough) tends to inhibit the activity of an anti-staling amylases such as Novamyl, making it less effective to prevent the staling of dough-based products with high sucrose content such as cakes. They have found that a good anti-staling effect in cakes can be achieved by using a carefully selected anti-staling amylase with certain properties, and they have identified such amylases.

By analyzing a 3D structure of Novamyl, the inventors further found that sucrose may inhibit by binding in the active site. They have found that sucrose docks into the active site of Novamyl differently from the substrate or inhibitor in published models 1QHO and 1QHP, and they have used this finding to design sucrose-tolerant variants. Accordingly, the invention provides a method of preparing dough or a dough-based edible product (e.g. a baked product) by adding a sucrose-tolerant anti-staling amylase. It also provides novel sucrose tolerant variants of a maltogenic alpha-amylase.

DETAILED DESCRIPTION OF THE INVENTION

Maltogenic alpha-amylase and sucrose docking

A maltogenic alpha-amylase (EC 3.2.1.133) having more than 70 % identity (particu¬ larly more than 80 % or 90%, such as at least 95% or 96% or 97% or 98% or 99%) with the No- vamyl sequence shown as SEQ ID NO: 1 may be used as the parent enzyme for designing su¬ crose tolerant variants. Amino acid identity may be calculated as described in US 6162628. For Novamyl (SEQ ID NO: 1), a 3D structure including a substrate or inhibitor as de¬ scribed in US 6162628 or in the Protein Data Bank with the identifier 1QHO or 1QHP may be used. Alternatively, a Novamyl variant may be used, such as a variant described in US 6162628 or in this specification, e.g. the variant F188L +D261 G +T288P. A 3D structure of a variant may be developed from the Novamyl structure by known methods, e.g. as described in T.L. Blundell et al., Nature, vol. 326, p. 347 ff (26 March 1987); J. Greer, Proteins: Structure, Function and Genetics, 7:317-334 (1990); or Example 1 of WO 9623874.

The inventors found that sucrose may inhibit Novamyl by binding in the active site. Docking of sucrose into the active site of Novamyl (using the software GOLD version 2.1.2, Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge, CB2 1EZ, UK and the protein part of the x-ray structure IQHO.pdb) reveals a specific binding configuration as unique to sucrose. The cartesian coordinates for the sucrose atoms in this binding configuration, using the coordinate system of the x-ray structure IQHO.pdb are given in Fig. 1.

Maltogenic alpha-amylase assay

The activity of a maltogenic alpha-amylase may be determined using an activity assay such as the MANU method. One MANU (Maltogenic Amylase Novo Unit) is defined as the amount of enzyme required to release one micro-mole of maltose per minute at a concentration of 10 mg of maltotriose substrate per ml in 0.1 M citrate buffer at pH 5.0, 37°C for 30 minutes.

Amino acid alterations

The amino acid sequence of a maltogenic alpha-amylase may be altered to decrease the sucrose inhibition. The inventors found that the alteration may be made at an amino acid residue having at least one atom within 4 Angstroms from any of the sucrose atoms when the sucrose molecule is docked in the 3D structure of the maltogenic alpha-amylase. Using the No¬ vamyl structure 1QHO and the sucrose docking in Fig. 1 , the following Novamyl residues are within 4 A: K44, N86, Y89, H90, Y92, W93, F188, T189, D190, P191 , A192, F194, D372, P373, R376.

Further the following positions have been identified as relevant: 115, R81 , T87, G88, L196, N371 or N375 of SEQ ID NO: 1. The alteration may be a substitution or deletion of one or more of the selected resi¬ dues, or one or more residues (particularly 1-4 residues or 5-6 residues) can be inserted adja¬ cent to a selected residue.

The substitution may be with a smaller or larger residue. A substitution to increase the size of the residue may diminish the space obtained by the docked sucrose molecule thereby preventing the binding of sucrose. Amino acid residues are ranked as follows from smallest to largest: (an equal sign indicates residues with sizes that are practically indistinguishable): G < A=S=C < V=T < P < L=I=N=D=M < E=Q < K < H < R < F < Y < W The substitution may also be such as to eliminate contacts with the sucrose molecule, in particular by moving or removing potential sites of hydrogen bonding or Van der Waals inter- actions.

The substitution may particularly be with another residue of the same type where the type is negative, positive, hydrophobic or hydrophilic. The negative residues are D₁E, the posi¬ tive residues are K/R, the hydrophobic residues are A₁C, F, G, I, L₁M, P₁V₁W, Y, and the hydrophilic residues are H, N, Q₁S₁T. Some particular examples of substitutions are I15T/S/V/L, R18K, K44R/S/T/Q/N,

N86Q/S/T, T87N/Q/S, G88A/S/T, Y89W/F/H, H90W/F/Y/R/K/N/Q/M, W93Y/F/M/E/G/V/T/S, F 188H/L/I/T/G/V, D190E/Q/G, A192S/T, F194S/L/Y, L196F, N371 K/R/FΛ7Q, D372E/Q/S/T/A and N375S/T/D/E/Q.

Examples of deletions are deletion of residue 191 or 192. An example of an insertion is Ala inserted between 192 and 193.

The polypeptide may include other alterations compared to Novamyl (SEQ ID NO: 1), e.g. alterations to increase the thermostability as described in US 6162628.

Nomenclature for amino acid alterations

In this specification, an amino acid substitution is described by use of one-letter codes, e.g. K44R. Slashes are used to indicate alternatives, e.g. K44R/S/T/Q/N to indicate substitution of K44 with R or S etc. P191* indicates a deletion of P191. *192aA indicates insertion of one Ala after A192. Commas are used to indicate multiple alterations in the sequence, e.g.

F 188L.D261 G.T288P to indicate a variant with three substitutions. Properties of anti-staling amylase for use with sucrose

The amylase for use in high-sucrose dough may be selected so as to have mainly exo- amylase activity. More specifically, the amylase hydrolyzes amylose so that the average mo¬ lecular weight of the amylose after 0.4-4 % hydrolysis is more than 50 % (particularly more than 75 %) of the molecular weight before the hydrolysis.

Thus, the amylase may hydrolyze amylose (e.g. wheat amylose or synthetic amylose) so that the average molecular weight of the amylose after 0.4-4 % hydrolysis (i.e. between 0.4-4 % hydrolysis of the total number of bonds) is more than 50 % (particularly more than 75 %) of the value before the hydrolysis. The hydrolysis can be conducted in a 1.7 % amylose solution by weight at suitable conditions (e.g. 10 minutes at 6O⁰C, pH 5.5), and the molecular weight distri¬ bution before and after the hydrolysis can be determined by HPLC. The test may be carried out as described in C. Christophersen et al., Starch 50 (1), 39-45 (1998).

An exo-amylase for use in high-sucrose dough may have a specified sugar tolerance. Compared to its activity in the absence of sucrose, the amylase may have more than 20 % ac- tivity at 10 % sugar, more than 10 % activity at 20 % sucrose, or more than 4 % activity at 40 % sucrose. The sugar tolerance may be determined as described in the examples.

The exo-amylase may have optimum activity in the pH range 4.5-8.5. It may have suffi¬ cient thermostability to retain at least 20 % (particularly at least 40 %) activity after 30 minutes incubation at 85°C at pH 5.7 (50 mM Na-acetate, 1 mM CaCI₂) without substrate. The exo-amylase may be added to the dough in an amount corresponding to 1-100 mg enzyme protein per kg of flour, particularly 5-50 mg per kg.

The exo-amylase may be non-liquefying. This can be determined by letting the exo- amylase act on a 1 % wheat starch solution until the reaction is complete, i.e. addition of fresh enzyme causes no further degradation, and analyzing the reaction products, e.g. by HPLC. Typical reaction conditions are e.g. 0.01 mg enzyme per ml starch solution for 48 hours. The exo-arnylase is considered non-liquefying if the amount of residual starch after the reaction is at least 20 % of the initial amount of starch.

The exo-amylase may have maltogenic alpha-amylase activity (EC 3.2.1.133). The exo-arnylase may be the amylase described in DK PA 2004 00021 , or it may be a Novamyl variant described in this specification.

Dough and dough-based edible product

The dough may have a sucrose content above 10 % by weight, particularly above 20 % or 30 %, e.g. 30-40 %. The flour content is typically 25-35 % by weight of total ingredients. The dough may be made by a conventional cake recipe, typically with cake flour, sugar, fat/oil and eggs as the major ingredients. It may include other conventional ingredients such as emulsifiers, humectants, gums, starch and baking powder. It generally contains such ingredients as soft wheat flour, milk or other liquids, sugar, eggs, chemical leaveners, flavor extracts and spices, as well as others that may or may not include shortening.

The dough is generally heat treated, e.g. by baking or deep frying to prepare an edible product such as cakes including pound cake, yellow and white layer cakes, cakes containing chocolate and cocoa products, sponge cakes, angel food cake, fruit cakes and foam-type cakes and doughnuts.

EXAMPLES

Example 1 : Sucrose tolerance of Novamyl variants

The amylase activity of a number of polypeptides were tested by incubation with Phadebas tablets (product of Pharmacia^®) for 15 minutes at 6O⁰C in the presence of sucrose at various concentrations (in % by weight). The results are expressed in % of the result without sugar:

Example 2: Sucrose tolerance of Novamyl variants A number of polypeptides were tested as in Example 1. The results are expressed as activity with 10 % sucrose in % of the activity without sucrose:

The following variants are also considered of interest in the context of the present in¬ vention:

Alterations compared to SEQ ID NO: 1

115T₁ N86K, P191S, D261G, T288P

I15T, P191S, D261G. T288P

115T₁ P191 S, Y258F, D261G, T288P, N375S, Y549C, Q648H

115T₁ G153R, P191S, D261 G. T288P, N371 K, K645R Example 3: Sucrose tolerance and thermostabϊ lity of amylases

The following amylases were tested for thermostability and sugar tolerance: bacterial alpha-amylase from B. amyloliquefaciens (BAN™, product of Novozymes AIS), fungal alpha- amylase from A. oryzae (Fungamyl^®, product of Novozymes A/S), maltogenic alpha-amylase having the sequence of SEQ ID NO: 1 (Novamyl^®, product of Novozymes A/S), a Novamyl variant having SEQ ID NO: 1 with the substitutions F188L +-D261 G +T288P, and bacterial alpha-amylase from B. licheniformis (Termamyl^®, product of Novozymes A/S).

Exo-amylase activity

The five amylases were tested for exo-amylase activity as described above. The re- suits show that Novamyl and the Novamyl variant had exo-amylase activity by this test, and the other three did not.

Thermostability

Each amylase was incubated at 85°C at pH 5.7 (50 mM Na-acetate, 1 mM CaCI₂) with¬ out substrate, and the amylase activity was measured after 0, 15, 30 and 60 minutes heat treatment. The results are expressed as residual activity in % of the initial activity:

0 15 30 60

BAN 100 3 1 0

Fungamyl 100 0 0 0

Novamyl 100 51 29 13

Novamyl variant 100 64 48 54

Termamvl 100 100 71 85

The results show that the Novamyl variant and Termamyl were not deactivated by the heat-treatment. BAN and Fungamyl lose all their activity after 15 min while Novamyl loses it gradually with heat-treatment time.

Sucrose tolerance

The experiment was repeated in 10 % sucrose solution. The results are expressed as residual activity in % of the initial activity without sucrose:

0 15 30 60

BAN 93 2 1 0

Fungamyl 31 0 0 0

Novamyl 7 6 1 3

Novamyl variant 21 19 14 16

Termamyl 116 112 97 82 The results show that BAN and Termamyl were not inhibited by sugar while Fungamyl and the Novamyl variant were somewhat inhibited, and Novamyl was heavily inhibited by sugar.

The combination of sugar and heat-treatment shows that the Novamyl variant and Termamyl could be active during baking of cakes. Termamyl and the Novamyl variant fulfill the criterion for thermostability and sugar tolerance used in this invention.

Example 4: Preparation of sponge cake with amylase

Sponge cakes were made with addition of amylase as follows: BAN (0.83. 8.3 or 83 mg/kg flour), Novamyl (1.3 or 13 mg/kg flour) or the Novamyl variant used in Example 1 (1 , 10 or 100 mg/kg flour). A control cake was made without amylase.

The cakes were baked according to the High Ratio Sponge Sandwich Cake (HRSSC) method. After baking, the cakes were cooled down for 60-120 minutes, and the cakes were stored at room temperature in sealed plastic bags filled with nitrogen until analysis. The cakes were evaluated on day 1 , 3, 7 or 23. Texture profile analysis (TPA) was performed as described in Bourne M. C. (2002) 2. ed., Food Texture and Viscosity: Concept and Measurement. Avcademic Press. The results showed that the increase in hardness was slower with increasing dosage of the Novamyl vari¬ ant. The addition of BAN or Novamyl had only a slight effect, and only at the highest dosage.

The cohesiveness of the cakes decreased with storage time. The addition of the No- vamyl variant delayed this decrease. The addition of BAN or Novamyl had a slight effect, and only at the highest dosage.

Water mobility was characterized by low field NMR. The addition of the Novamyl vari¬ ant and BAN increased the mobility, indicating that the two amylases were able to keep the cakes more moist. Novamyl had virtually no effect. A small sensory evaluation of softness and moistness was performed on day 13 for the

3 cakes with the Novamyl variant and the control cake. The cakes were evaluated regarding three parameters; Firmness, Moistness and preferability. The control was the firmest, driest and least preferred. The higher dosage of the Novamyl variant, the less firm (softer), moister and better liked. A large panel sensory evaluation was performed on day 13. It was a paired comparison test where a control cake was compare to the cake with the Novamyl variant at the highest dos¬ age. A 30-member panel was asked two questions (1 ) Which cake is moister and (2) which cake is fresher. All panel members agreed on that the cake with the Novamyl variant was moister and fresher. The preference was significant at a significance level above 99.999 %. To summarize, the data show that the Novamyl variant had anti-staling properties and was able to improve moistness perception and moistness measured by NMR. The two other amylases had only a slight effect.

Example 5: High-ratio unit cakes Cakes were made with addition of amylase as follows: BAN (0.83. 8.3 or 83 mg/kg flour) or the Novamyl variant used in Example 1 (1 , 10 or 100 mg/kg flour). A control cake was made without amylase.

Cakes were baked according to the High ratio unit cake (HRUC) method. After baking, the cakes were cooled down for 60-120 minutes, and the cakes were stored at room tempera- ture in sealed plastic bags filled with Nitrogen until analysis. The cakes Λ/vere evaluated on day 7, 20 and 34 by the same methods as in the previous example.

The increase in hardness was slower with the Novamyl variant at the highest dosage. The addition of BAN to the cake resulted in a low volume and a doughy cake which gave poor results in hardness measurements. The addition of the Novamyl variant delayed the decrease in cohesiveness while BAN did not influence it at all.

The Novamyl variant and BAN were able to keep the cake more moist than the control. This increase in mobility of the free water could partly be explained by the cakes with BAN and the Novamyl variant being able to retain the moisture content. A small sensory evaluation on day 34 showed that the cake with the Novamyl variant at the highest dosage was clearly better than the control cake; it was more moist and it was less crumbly.

Over-all, there was an anti-staling effect of the Novamyl variant at the high dosage, similar to the effect on sponge cakes in the previous example. The staling of HRUC cakes was slower than Sponge cakes but it was still evident that the Novamyl variant had an anti-staling effect. The anti-staling effect was seen with texture analysis, NMR and sensory evaluation. BAN showed anti-staling effects in HRUC but it was sensitive to over-dosage which resulted in cake collapse and a doughy cake.

Example 6: Sponge cake Sponge cakes were made with addition of the amylase of DK PA 2004 00021 at dos¬ ages 0.5, 1 , 2, 5 and 20 mg/kg flour and a control cake without amylase.

Texture and NMR was measured on day 1 , 7 and 13. The addition of the amylase re¬ duced the increase in firmness, especially at the highest dosage. The amylase also had a bene¬ ficial effect on the mobility of water which was correlated with the moistness of the cake. A blind sensory ranking evaluation performed on day 14 showed a ranking according to the dosage, the higher dosage the more soft and moist cake. The most preferred cake was the one with the highest dosage.

Example 7: Baking procedure Tegral Allegro cake Recipe

The following recipe was used:

*commercially available from Puratos NV/SA, Groot-Bijgaarden, Belgium Procedure

The ingredients were scaled into a mixing bowl and mixed using an industrial mixer (e.g. Bjørn AR 5 A Varimixer) with a suitable paddle speed. 300 g of the dough was poured into forms. The cakes are baked in a suitable oven (e.g. Sveba Dahlin deck oven) for 45 min. at 180 ⁰C. The cakes were allowed to cool down at room temperature for 1 hour.

The volume of the cakes was determined when the cakes had cooled down using the rape seed displacement method. The cakes were packed under nitrogen in sealed plastic bags and stored at room temperature until analysis.

The cakes were evaluated on day 1 , 7 and 14, two cakes were used at each occa¬ sions.

The cohesiveness and hardness of the cakes was evaluated with Texture analyser and the water mobility was characterized by low field NMR.

The Texture profile analysis (TPA) was performed as described in Bourne M. C (2002) 2. ed., Food Texture and Viscosity: Concept and Measurement. Academic Press.

The mobility of free water was determined as described by P. L. Chen, Z. Long, R. Ruan and T. P. Labuza, Nuclear Magnetic Resonance Studies of water Mobility in Bread during Storage. Lebensmittel Wissenschaft und Technologie 30, 178-183 (1997). The mobility of free water has been described in literature to correlate to moistness of bread crumb. Result

Compared to cakes with no addition of enzymes the volume of the cakes is not affected by the addition of the reference enzyme (SEQ ID NO.: 1) nor by the addition of variants hereof, i.e. the cakes did not collapse upon addition of enzyme.

The cohesiveness of the cakes decreased with storage time. The addition of variants of SEQ ID NO: 1 delayed this decrease as can be seen in Table 1.

Table 1 : Change in Cohesiveness fas/gsi with storage time of cakes with 25 mg protein enzyme per kg flour

The free water mobility is correlated with the moist perception of the cake crumb, it de¬ creases with time. The addition of the Novamyl variants increased the mobility compared to the control, indicating that the amylases were able to keep the cakes more moist. Results are listed in Table 2.

Table 2: Change in free water mobility [micros! with storage time of cakes with 25 mg protein enzyme per kg flour

The hardness of the cakes increased with storage time. The addition of variants of SEQ ID NO: 1 delayed this increase in hardness as can be seen in Table 3.

Table 3: Change in hardness Fαl with storage time of cakes with 25 mg protein enzyme per kg flour

Claims

1. A method of preparing a dough or a dough-based edible product, comprising adding a poly¬ peptide to the dough, wherein the dough comprises at least 10 % sucrose by weight, and the polypeptide: a) has an amino acid sequence which is at least 70 % identical to SEQ ID NO: 1 , and b) compared to SEQ ID NO: 1 comprises an amino acid alteration which is substitution or deletion of or insertion adjacent to 115, R18, K44, N86, T87, G88, Y89, H90, Y92, W93, F188, T189, D190, P191 , A192, F194, L196, D329, N371 , D372, P373, N375 or R376.

2. The method of claim 1 wherein the alteration is substitution with a larger or smaller amino acid residue.

3. The method of claim 1 wherein the alteration is insertion of 1-4 amino acid residues at the N- or C-side of the specified residue.

4. The method of claim 1 wherein the polypeptide comprises a substitution I15T/S/V/L, R18K, K44R/S/T/Q/N, N86Q/S/T, T87N/Q/S, G88A/S/T, Y89W/F/H, H90W/F/Y/R/K/N/Q/M,

W93Y/F/M/E/G/V/T/S, F188H/L/I/T/G/V, D190E/Q/G, A192G/S/T/Q/R, F194S/L/Y, L196F, N371K/R/FΛ7Q or D372E/Q/S/T/A, a deletion of 191 or 192 or an insertion of Ala after 192.

5. The method of claim 1 wherein the polypeptide has the amino acid sequence of SEQ ID NO: 1 with one of the following sets of alterations:

6. A polypeptide which: a) has amylase activity which is less inhibited by sucrose than the amylase activity of SEQ ID NO: 1 , b) has an amino acid sequence which is at least 70 % identical to SEQ ID NO: 1 , and c) compared to SEQ ID NO: 1 comprises an amino acid alteration which is substitution or deletion of or insertion adjacent to K44, Y89, H90, Y92, W93, D329, D372, P373 or R376 or an alteration N86T/VG/K, T87N/Q, H90W/F/Y/R/M, F1881/L/T/GΛ/, D190G, P191*. A192G/Q/R, F194S/L/Y, L196F, L225S, F252L, D261G, T288P, I290V, N371 K/R/FΛ7Q, D372E/Q/S/T/A/V, N375S, S446A, G469R, S578G or P642Q.

7. The polypeptide of claim 6 wherein the alteration is substitution with a larger or smaller amino acid residue.

8. The polypeptide of claim 6 which comprises insertion of 1-4 amino acid residues at the N- or C-side of the specified residue.

9. The polypeptide of claim 6 which comprises a substitution K44R/S/T/Q/N, Y89W/F/H, W93Y/F/M/E/G/V/T/S, D261G, T288P.

10. The polypeptide of claim 6 which has the amino acid sequence of SEQ ID NO: 1 with one of the following sets of alterations:

1 1. A method of preparing a polypeptide, comprising a) providing a parent polypeptide having an amino acid sequence, and having malto- genic alpha-amylase activity, b) selecting an amino acid residue in the sequence corresponding to 115, R18, K44, N86, T87, G88, Y89, H90, Y92, W93, F188, T189, D190, P191 , A192, F194, L196,

D329, N371 D372, P373, N375 or R376 in SEQ ID NO: 1 , c) substituting or deleting the selected residue or inserting one or more residues adja¬ cent to the selected residue to obtain an altered amino acid sequence, d) preparing an altered polypeptide having the altered amino acid sequence, e) testing the amylase activity and the sugar tolerance of the altered polypeptide, and f) selecting a polypeptide which has amylase activity and has higher sucrose tolerance than the parent polypeptide.

12. A method of constructing a polypeptide, comprising: a) providing a parent maltogenic alpha-amylase having an amino acid sequence and a three-dimensional structure, b) docking a sucrose molecule in the structure, c) selecting an amino acid residue having a C-alpha atom located <10 A from an atom in the docked sucrose molecule, d) substituting or deleting the selected residue or inserting one or more residues adja- cent to the selected residue to obtain an altered amino acid sequence, e) docking a sucrose molecule in a structure of the polypeptide having the altered se¬ quence and calculating the binding energy for the sucrose, and f) selecting a polypeptide wherein the binding energy and/or binding configuration is changed.

13. A method of preparing dough or a dough-based edible product, comprising adding an exo- amylase to dough wherein the dough comprises at least 10 % sucrose by weight, and the exo- amylase has amylase activity in the presence of 10 % sucrose by weight which is more than 20 % of the activity without sucrose.

14. The method of the preceding claim wherein the exo-amylase retains at least 20 % (particu- larly at least 40 %) activity after 30 minutes incubation at 85⁰C at pH 5.7 (50 mM Na-acetate, 1 mM CaCI₂) without substrate.