WO1998026078A1

WO1998026078A1 - H mutant alpha-amylase enzymes

Info

Publication number: WO1998026078A1
Application number: PCT/US1996/019595
Authority: WO
Inventors: Anthony G. Day; Colin Mitchinson; Andrew Shaw
Original assignee: Genencor International Inc
Current assignee: Danisco US Inc
Priority date: 1996-12-09
Filing date: 1996-12-09
Publication date: 1998-06-18
Anticipated expiration: 1999-06-09
Also published as: EP0942994B1; JP2001506496A; ATE293696T1; JP4253041B2; DE69634640D1; CA2274806A1; CA2274806C; DE69634640T2; AU1461497A; EP0942994A1

Abstract

An improved protein comprising an amino acid sequence which has been modified from a precursor amino acid sequence by the substitution or deletion of an amino acid residue which differs from a corresponding amino acid residue in a less stable but homologous protein, wherein said improved protein has improved properties compared to a protein corresponding to the precursor amino acid sequence.

Description

H MUTANT ALPHA-AMYLASE ENZYMES

FIELD OF THE INVENTION

The present invention is directed to improved proteins, and particularly enzymes, having altered stability characteristics through the recruitment of residues from a less stable homolog or related protein. The present invention is particularly directed to novel mutant α- amylase enzymes derived from B. licheniformis having at least one recruitment from a corresponding residue in B. stearothermophilus and/or B. amyloliquefaciens which differs from the residue in B. licheniformis. The resultant α-amylases exhibit improved stability under adverse conditions.

BACKGROUND OF THE INVENTION

Methods of improving the stability of proteins have been utilized extensively for the purpose of increasing the usefulness of the proteins in practice. Such techniques often involve the use of protein engineering and site-specific mutagenesis to isolate and select specific amino acid residues which appear, based on the specific amino acid properties and/or the structure of the protein, to be especially relevant or are shown to be relevant through experimental evidence. For example, Estell et al., U.S. Patent No. 4,760,025 suggest modification of oxidatively unstable enzymes to confer increased stability by altering residues which are particularly susceptible to oxidation, i.e., methionine, lysine, tryptophan and cysteine, among others. Koths et al., U.S. Patent No. 4,752,585, suggest a method of improving the oxidation stability of a therapeutic protein by making a conservative amino acid substitution for each methionyl residue susceptible to chloramine T oxidation. Barr et al., U.S. Patent No. 4,732,973, suggest substituting the active site methionine from human α antitrypsin with an oxidatively stable amino acid.

Additional techniques which have been suggested include homology modeling of proteins against evolutionarily close enzymes to provide a basis for protein engineering. Siezen et al., Protein Engineering, vol. 4, no. 7, pp. 719-737 (1991) disclose a strategy for improving the properties of proteases used in industrial processes by incorporating or duplicating features from more stable enzymes into a target.

Nonetheless, protein engineering based on homology modeling requires a relatively high degree of homology (greater than 70%) to be considered highly successful. It has further been reported that, where the homology falls to below 30%, such modeling becomes far less successful (see e.g., Machius et al., J. Mol. BioL, vol. 246, p. 547 (1995)). Of particular interest to industry, and used herein to exemplify the invention, are the α-amylases (α-1,4-glucan-4-glucanohydrolase, EC 3.2.1.1) for which increased stability and/or activity are desired. α-Amylases hydrolyze internal α-1 ,4-glucosidic linkages in starch, largely at random, to produce smaller molecular weight malto-dextrins. α-Amylases are of considerable commercial value, being used in the initial stages (liquefaction) of starch processing; in alcohol production; as cleaning agents in detergent matrices; and in the textile industry for starch desizing. α-Amylases are produced by a wide variety of bacterial, fungal and plant sources including Bacillus and Aspergiilus, with most commercial amylases used in the starch processing industry being produced from bacterial sources such as Bacillus licheniformis, Bacillus amyioliquefaciens, Bacillus subtilis, or Bacillus stearothermophilus. In recent years, the preferred enzymes in commercial use have been those from Bacillus licheniformis because of their heat stability and performance, at least at neutral and mildly alkaline pH's. Amylases derived from related species such as Bacillus amyioliquefaciens, Bacillus subtilis and Bacillus stearothermophilus are generally believed to be considerably less stable under many conditions.

Variant α-amylases have been suggested in PCT Publication No. WO 95/10603, which have improved laundry or dishwashing performance and comprise a mutation other than a single mutation at position M 197 in Bacillus licheniformis α-amylase. In PCT Publication No. WO 94/02597, a mutant α-amylase having improved oxidative stability is described wherein one or more methionines are replaced by any amino acid except cysteine or methionine. In PCT publication No. WO 94/18314, a mutant α-amylase having improved oxidative stability is described wherein one or more of the methionine, tryptophan, cysteine, histidine or tyrosine residues is replaced with a non-oxidizable amino acid. In PCT Publication No. WO 91/00353, the performance characteristics and problems associated with starch liquefaction with wild type Bacillus licheniformis α-amylase are approached by genetically engineering the α-amylase to include the specific substitutions Ala-111-Thr, His- 133-Tyr and/or Thr-149-lle.

Studies using recombinant DNA techniques to explore which residues are important for the catalytic activity of amylases and/or to explore the effect of modifying certain amino acids within the active site of various amylases and glycosylases have been conducted by various researchers (Vihinen et al., J. Biochem., vol. 107, pp. 267-272 (1990); Holm et al., Protein Engineering, vol. 3, pp. 181-191 (1990); Takase et al., Biochemica et Biophysica Acta, vol. 1120, pp. 281-288 (1992); Matsui et al., Febs Letters, vol. 310, pp. 216-218 (1992); Matsui et al., Biochemistry, vol. 33, pp. 451-458 (1992); Sogaard et al., J. Biol. Chem., vol. 268, pp. 22480-22484 (1993); Sogaard et al., Carbohydrate Polymers, vol. 21, pp. 137-146 (1993); Svensson, Plant Mol. Biol., vol. 25, pp. 141-157 (1994); Svensson et al., J. Biotech, vol. 29, pp. 1-37 (1993)). Researchers have also studied which residues are important for thermal stability (Suzuki et al., J. Biol. Chem. vol. 264, pp. 18933-18938 (1989); Watanabe et al., Eur. J. Biochem. vol. 226, pp. 277-283 (1994)); and one group has used such methods to introduce mutations at various histidine residues in a Bacillus licheniformis amylase, the rationale being that Bacillus licheniformis amylase which is known to be relatively thermostable when compared to other similar Bacillus amylases, has an excess of histidines and, therefore, it was suggested that replacing a histidine could affect the thermostability of the enzyme. This work resulted in the identification of stabilizing mutations at the histidine residue at the +133 position and the alanine residue at position +209 (Declerck et al., J. Biol. Chem., vol. 265, pp. 15481-15488 (1990); FR 2 665 178-A1 ; Joyet et al., Bio/Technology, vol. 10, pp. 1579-1583 (1992)).

Thus, there has been a considerable effort in the field of protein engineering to produce more stable enzymes. Such techniques, including recruitment from more stable enzymes to less stable enzymes, random mutagenesis and sequence alignment and homology modeling have thus far produced many successes. However, there remains a need in the field for additional techniques useful for the production of improved enzymes.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a protein which has increased stability.

It is a further object of the present invention to provide for an protein, for example an α-amylase, having altered stability profiles, such as pH stability, alkaline stability, oxidative stability and/or thermostability. According to the present invention, an improved protein is provided comprising an amino acid sequence which has been altered from a precursor amino acid sequence by the substitution or deletion of an amino acid residue which differs from a corresponding amino acid residue in a less stable but homologous protein. Preferably, the improved protein has improved stability properties compared to a protein corresponding to the precursor amino acid sequence. Also preferably, the substitution or deletion does not occur at a buried residue but instead at a residue on the surface of the molecule.

Also preferably, the protein is an enzyme, most preferably an α-amylase, lipase, cellulase or protease. In a particularly preferred composition embodiment of the invention, the enzyme is an α-amylase derived from Bacillus licheniformis and the less stable but homologous protein comprises either or both of the amylases derived from Bacillus amyioliquefaciens or Bacillus stearothermophilus. In a particularly preferred embodiment of the invention, the precursor protein is α- amylase derived from Bacillus licheniformis and the substituted residues in the improved protein, which differ in both Bacillus stearothermophilus and Bacillus amyioliquefaciens, comprise one or more of A33, A52, N96, H133, S148, A209, A269, A379 and A435, preferably comprising the following specific substitutions which occur naturally in both

Bacillus stearothermophilus and Bacillus amyioliquefaciens, i.e., A33 S, A52 S, N96 Q, H133 Y, S148 N, A209 V, A269 K, A379 S and/or A435 S.

In a method embodiment of the invention, a method for the production of an improved protein based on a precursor protein comprising: (a) aligning the sequence of the precursor protein with a sequence of a less stable but homologous enzyme and identifying one or more residues which differ between the sequence of the precursor protein and the less stable but homologous enzyme; (b) selecting one or more of the residues identified in step (a) for substitution in the precursor protein; and (c) modifying the precursor protein to incorporate a residue for substitution selected in the step (b). Preferably, the substitution occurs at a residue which is on the surface of the protein and not at a buried residue.

An advantage of the present invention is that it is possible to produce more stable protein through a simple technique of analyzing the aligned sequences of two homologous but stability differentiated proteins.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 illustrates mutagenic oligonucleotides useful during directed mutagenesis of Asn188 from Bacillus licheniformis α-amylase. In this and following figures illustrating oiigonucleotide constructs, bold letters indicate base changes introduced by the oligonucleotide and underlining indicates restriction endonuclease sites introduced by the oligonucleotide.

Figure 2 illustrates PCR primers used for PCR processing of mutagenic oligonucleotide templates.

Figure 3 illustrates the DNA sequence of the gene for α-amylase from Bacillus licheniformis (NCIB 8061) (SEQ ID NO:33) and deduced amino acid sequence of the translation product (SEQ ID NO:41) as described by Gray et al., J. Bacteriology, vol. 166, pp. 635-643 (1986).

Figure 4 illustrates the amino acid sequence (SEQ ID NO:34) of the mature α- amylase enzyme from Bacillus licheniformis.

Figure 5 illustrates an alignment of the primary structures of three Bacillus α-amylases. The Bacillus licheniformis α-amylase (Am-Lich) (SEQ ID NO:35) is described by Gray et al., J. Bacteriology, vol. 166, pp. 635-643 (1986); the Bacillus amyioliquefaciens α-amylase (Am-Amylo) (SEQ ID NO:36) is described by Takkinen et al., J. Biol. Chem., vol. 258, pp. 1007-1013 (1983); and the Bacillus stearothermophilus α-amylase (Am-Stearo) (SEQ ID NO:37) is described by lhara et al., J. Biochem., vol. 98, pp. 95-103 (1985). Figure 6 illustrates plasmid pHP13 wherein Cm" refers to chloramphenicol resistance,

Em^R refers to erythromycin resistance and Rep pTA1060 refers to the origin of replication from plasmid pTA1060.

Figure 7 illustrates the pBLapr plasmid wherein BL AA refers to Bacillus licheniformis α-amylase gene; aprE refers to the promoter and signal peptide encoding region of the aprE gene; AmpR refers to the ampicillin resistant gene from pBR322; and CAT refers to the chloramphenicol resistance gene from pC194.

Figure 8 illustrates the pHP.BL plasmid carrying the gene for Bacillus licheniformis α- amylase.

Figure 9 illustrates a schematic of the PCR method used to produce the mutant oligonucleotides corresponding to α-amylase derived from Bacillus licheniformis.

Figure 10 illustrates the signal sequence-mature protein junctions in α-amylase derived from Bacillus licheniformis (SEQ ID NO:38), Bacillus subtilis aprE (SEQ ID NO: 39) and Bacillus licheniformis in pBLapr (SEQ ID NO:40).

DETAILED DESCRIPTION

"Expression vector" means a DNA construct comprising a DNA sequence which is operably linked to a suitable control sequence capable of effecting the expression of said DNA in a suitable host. Such control sequences may include a promoter to effect transcription, an optional operator sequence to control such transcription, a sequence encoding suitable mRNA ribosome-binding sites, and sequences which control termination of transcription and translation. Where expression is desired in a Bacillus host, a preferred promoter is the Bacillus subtilis aprE promoter. The vector may be a plasmid, a phage particle, or simply a potential genomic insert. Once transformed into a suitable host, the vector may replicate and function independently of the host genome, or may, in some instances, integrate into the genome itself. While plasmid and vector are sometimes used interchangeably as the plasmid is the most commonly used form of vector at present, the invention is intended to include such other forms of expression vectors which serve equivalent functions and which are, or become, known in the art.

"Host strain" or "host cell" means a suitable host for an expression vector comprising DNA encoding a protein according to the present invention. Host cells useful in the present invention are generally procaryotic or eucaryotic hosts, including any transformable microorganism in which the expression of a given protein according to the present invention can be achieved. One of skill in the art will be familiar with the appropriate expression and secretion machinery, including the appropriate host cell, for use with a specific protein. For example, host strains of the same species or genus from which the particular protein is derived are suitable, for example with α-amylase derived from Bacillus, a suitable host cell would be a Bacillus strain. In this case, an α-amylase negative Bacillus strain (genes deleted) and/or an α-amylase and protease deleted Bacillus strain (AamyE, Δapr, άnpή is preferably used. Host cells are transformed or transfected with vectors constructed using recombinant DNA techniques. Such transformed host cells are capable of either replicating vectors encoding the protein and its variants (mutants) or expressing the desired protein. "Recombinant protein" means a protein in which the DNA sequence encoding the naturally occurring or precursor protein is modified to produce a mutant DNA sequence which encodes the substitution, insertion or deletion of one or more amino acids in the protein sequence compared to the naturally occurring or precursor protein. As used herein, the term precursor protein (or enzyme) refers not to an immediate precursor in terms of a chemical reaction, but instead to a parent protein from which the modification is modeled. Accordingly, while it is the precursor which defines the modification, the actual alteration or modification will have its basis in an alteration within the DNA which encodes the precursor, which DNA is then transformed, expressed and the protein product secreted which incorporates the modification.

The improved protein according to the present invention comprises an amino acid sequence which is derived from the amino acid sequence of a precursor protein. The precursor protein may be a naturally occurring protein or a recombinant protein. The amino acid sequence of the improved protein is derived from the precursor protein's amino acid sequence by the substitution, deletion or insertion of one or more amino acids of the precursor amino acid sequence. Such modification is generally of the precursor DNA sequence which encodes the amino acid sequence of the precursor proteins rather than manipulation of the precursor protein perse. Suitable methods for such manipulation of the precursor DNA sequence include methods disclosed herein and in commonly owned U.S. Patent Nos. 4,760,025 and 5,185,258, incorporated herein by reference.

According to the present invention, a protein is modified as follows. The protein is referenced to a homologous protein for differences in sequence. For example, it is possible to line up the sequences of two related proteins according to well known sequence alignment techniques and determine conserved (i.e., identical) positions as well as the positions at which the two proteins differ. In the practice of the present invention, it has surprisingly and unexpectedly been found that by referencing the target protein against a less stable protein and identifying and selecting certain residues which differ from the less stable protein for modification in the target protein, advantageous improvements in stability may be obtained. It is necessary that the two proteins have comparable activity or function and be substantially homologous. By comparable activity, the proteins should have similar biological activity, function, catalytic activity or such other criteria as are commonly used to classify a specific protein. "Substantially homologous" means that the proteins have a significant level of conserved, i.e., identical, amino acids such that their sequences can be meaningfully aligned and major structural, functional or catalytic sites defined. Preferably, the two proteins have a sequence identity of at least 60%, more preferably 65% sequence identity, and most preferably 80% sequence identity.

The improved protein of the invention should have stability which is at least 50% of the precursor protein, more preferably at least 70%, and most preferably at least 90% of the precursor protein under any given set of circumstances, yet having improved stability under other conditions. Thus, an improved protein which has a stability of at least 50% of the precursor enzyme under high temperature but which has improved stability under oxidative conditions would be within the scope of the invention. In a particularly preferred embodiment, the stability of the improved protein is improved with respect to the precursor protein in the presence of oxidants and high temperature.

In a preferred embodiment, the protein comprises an enzyme. The enzyme may comprise any enzyme in the five major enzyme classifications of hydrolase, oxidoreductase, transferase, lyase or ligase. Specific examples of enzymes which may benefit from the present invention include amylase, lipase, cellulase, protease, hemicellulase, glucoamylase, esterase, lactase, polygalacturonase, β-galactosidase, ligninase, oxidase, peroxidase, glucose isomerase or any enzyme for which closely related and less stable homologs exist. α-Amylase will be illustrated as exemplary of the inventive process and compositions.

"α-Amylase" as used herein means an enzymatic activity which cleaves or hydrolyzes the α(1-4)glycosidic bond, e.g., that in starch, amylopectin or amylose polymers. α-Amylase includes naturally occurring α-amylases as well as recombinant α-amylases. Preferred α- amylases in the present invention are those derived from Bacillus sp., particularly those from Bacillus licheniformis, Bacillus amyioliquefaciens or Bacillus stearothermophilus, as well as fungal α-amylases such as those derived from Aspergiilus (i.e., A. oryzae and A. nigeή. Accordingly, the α-amylases according to the present invention are derived from a precursor α-amylase. The precursor α-amylase is produced by any source capable of producing α- amylase. Suitable sources of α-amylases are prokaryotic or eukaryotic organisms, including fungi, bacteria, plants or animals. Preferably, the precursor α-amylase is produced by a Bacillus; more preferably, by Bacillus licheniformis, Bacillus amyioliquefaciens or Bacillus stearothermophilus; most preferably, the precursor α-amylase is derived from Bacillus licheniformis.

Homologies have been found between almost all endo-amylases sequenced to date, ranging from plants, mammals, and bacteria (Nakajima et al., Appl. Microbiol. Biotechnol., vol. 23, pp. 355-360 (1986); Rogers, Biochem. Biophys. Res. Co mun., vol. 128, pp. 470- 476 (1985); Janecek, Eur. J. Biochem., vol. 224, pp. 519-524 (1994)). There are four areas of particularly high homology in certain Bacillus amylases, as shown in Figure 5, wherein the underlined sections designate the areas of high homology. Sequence alignments have also been used to map the relationship between Bacillus endo-amylases (Feng et al., J. Molec. Evol., vol. 35, pp. 351-360 (1987)). The relative sequence homology between Bacillus stearothermophilus and Bacillus licheniformis amylase is about 66% and that between Bacillus licheniformis and Bacillus amyioliquefaciens amylases is about 81 %, as determined by Holm et al., Protein Engineering, vol. 3, No. 3, pp. 181-191 (1990). In order to establish homology to primary structure, the amino acid sequence of a precursor α-amylase is directly compared to the Bacillus licheniformis α-amyiase primary sequence and particularly to a set of residues known to be invariant to all α-amylases for which sequences are known (see e.g., Figure 7). It is possible also to determine equivalent residues by tertiary structure analysis of the crystal structures reported for porcine pancreatic α-amylase (Buisson et al., EMBO Journal, vol. 6, pp. 3909-3916 (1987); Qian et al.,

Biochemistry, vol. 33, pp. 6284-6294 (1994); Larson et al., J. Mol. Biol., vol. 235, pp. 1560- 1584 (1994)); Taka-amylase A from Aspergiilus oryzae (Matsuura et al., J. Biochem. (Tokyo), vol. 95, pp. 697-702 (1984)); and an acid α-amylase from A. niger (Boel et al.. Biochemistry, vol. 29, pp. 6244-6249 (1990)), with the former two structures being similar, and for barley α- amylase (Vallee et al., J. Mol. Biol., vol. 236, pp. 368-371(1994); Kadziola, J. Mol. Biol., vol. 239, pp. 104-121 (1994)). Although there have been some preliminary studies published (Suzuki et al, J. Biochem., vol. 108, pp. 379-381 (1990); Lee et al., Arch. Biochem. Biophys, vol. 291, pp. 255-257 (1991); Chang et al, J. Mol. Biol., vol. 229, pp. 235-238 (1993); Mizuno et al., J. Mol. Biol., vol. 234, pp. 1282-1283 (1993)), there is only a published structure for Bacillus licheniformis α-amylase (Machius et al., J. Mol. Biol. vol. 246, pp. 545-549 (1995)). However, several researchers have predicted common super-secondary structures between glucanases (MacGregor et al., Biochem. J., vol. 259, pp. 145-152 (1989)) and within α- amylases and other starch-metabolising enzymes (Jaspersen, J. Prot. Chem. vol. 12, pp. 791-805 (1993); MacGregor, Starke, vol. 45, pp. 232-237 (1993)); and sequence similarities between enzymes with similar super-secondary structures to α-amylases (Janecek, FEBS Letters, vol. 316, pp. 23-26 (1993); Janecek et al., J. Prot. Chem., vol. 12, pp. 509-514 (1993)). A structure for the Bacillus stearothermophilus enzyme has been modeled on that of Taka-amylase A (Holm et al., Protein Engineering, vol. 3, pp. 181-191 (1990)). The four highly conserved regions shown in Figure 7 contain many residues thought to be part of the active-site (Matsuura et al., J. Biochem. (Tokyo), vol. 95, pp. 697-702 (1984); Buisson et al., EMBO Journal, vol. 6, pp. 3909-3916 (1987); Vihinen et al., J. Biochem., vol. 107, pp. 267- 272 (1990)) including His +105; Arg +229; Asp +231 ; His +235; Glu +261 and Asp +328 under the Bacillus licheniformis numbering system. The crystal structure of a bacterial amylase hybrid enzyme has been published in PCT Publication No. WO 96/23874. As described above, the α-amylases from Bacillus licheniformis, Bacillus stearothermophilus, Bacillus amyioliquefaciens and Bacillus subtilis all bear a significant degree of homology. However, it is known that under the conditions of industrial starch liquefaction, e.g., high temperature (excess of 90 ' C), low pH (pH 4-6) and low calcium, α- amylase derived from Bacillus licheniformis provides the most acceptable performance. Nonetheless, even Bacillus licheniformis α-amylase is susceptible to undesirable instability under liquefaction conditions making a more stable alternative desirable. Accordingly, much work has been performed on the Bacillus licheniformis derived molecule for the purpose of introducing into this enzyme properties which are desirable in connection with its use in liquefaction processes. By aligning the sequence of α-amylase from Bacillus licheniformis against that of α-amylase from either Bacillus stearothermophilus or Bacillus amyioliquefaciens it is possible to identify residues which differ between the homologs. According to the present invention, then, the positions at which residues differ in α-amylase from B. stearothermophilus or B. amyioliquefaciens compared to α-amylase from B. licheniformis are selected for substitution in α-amylase from B. licheniformis. Preferably, the substituted residue is actually an identical residue as that which exists in α-amylase from either B. stearothermophilus or B. amyioliquefaciens. More preferably, the substituted residue corresponds to a position in which the same residue exists in both α-amylase from B. stearothermophilus and B. amyioliquefaciens.

Residues specifically identified herein for replacement in Bacillus licheniformis are those which differ from residues in a corresponding position in Bacillus amyioliquefaciens and/or Bacillus stearothermophilus, and particularly A33, A52, S85, N96, H133, S148, A209, A269, A379 and A435. While specific preferred replacements for these residues are selected from those present in both Bacillus amyioliquefaciens and Bacillus stearothermophilus and correspond to A33S, A52S, N96Q, H133Y, S148N, A209V, A269K, A379S and/or A435S. It has also been discovered that the A85D mutation, which is recruited only from Bacillus amyioliquefaciens also provides a stability benefit. The above residues may be altered in combination with other modifications which provide a performance benefit. In a most preferred embodiment according to the invention, the above residues are in combination with mutations at a residue corresponding to any of M15, N188 and/or M197 and particularly at M15T, N188S, and/or M197T in Bacillus licheniformis. Any of the modifications identified herein as in accordance with the invention in combination with a mutant corresponding to M15T/H133Y/N188S/A209V in Bacillus licheniformis will have a particularly improved stability profile. Modifications corresponding to M15T/A33S/H133Y /N188S/A209V; M15T/D28N/A33S/H133Y/N188S/A209V; M15T/A52S /H133Y/N188S/A209V; M15T/A52S/H133Y/N188S/A209V/L230F;

M15T/S85D/H133Y/N188S/A209V; M15T/N96Q/H133Y/N188S/A209V; M15T/N96Q/H133Y/N188S/A209V/I479T; M15T/H133Y/S148N/N188S/A209V; M15T/H133Y/S148N/N188S/A209V/A379S; M15T/H133Y/S148N/N188S/A209V/G433D; M15T/H 133Y/N 188S/A209V/A379S; M 15T/H 133 Y/N 188S/A209 V/A210S/T322A/A379S in Bacillus licheniformis are particularly desired and represent the best mode of carrying out the invention.

The improved proteins according to the present invention exhibit improved performance characteristics which make those proteins particularly useful in various applications for which the proteins are commonly used and for which improved stability is desired. For example, enzymes, including α-amylases, according to the present invention will exhibit improved thermostability, improved pH stability and/or improved oxidative stability. Enhanced thermostability will be useful in extending the shelf life of products which incorporate them and for applications at high temperatures. Enhanced oxidative stability or improved performance is particularly desirable in cleaning products, and for extending the shelf life of the enzyme in the presence of bleach, perborate, percarbonate or peracids used in such cleaning products. An α-amylase of the present invention is especially useful in starch processing and particularly in starch liquefaction wherein oxidative and thermal stability are particularly important.

An additional embodiment of the present invention comprises DNA encoding a protein according to the present invention and expression vectors comprising such DNA. The DNA sequences may be expressed by operably linking them to an expression control sequence in an appropriate expression vector and employing that expression vector to transform an appropriate host according to well known techniques. A wide variety of host/expression vector combinations may be employed in expressing the DNA sequences of this invention. Useful expression vectors, for example, include segments of chromosomal, non-chromosomal and synthetic DNA sequences, such as the various known plasmids and phages useful for this purpose. In addition, any of a wide variety of expression control sequences are generally used in these vectors. For example, with α-amylase derived from Bacillus, Applicants have discovered that a preferred expression control sequence for Bacillus transformants is the aprE signal peptide derived from Bacillus subtilis. A wide variety of host cells are also useful in expressing the DNA sequences of this invention. These hosts may include well known eukaryotic and prokaryotic hosts, such as strains of E. coli, Pseudomonas, Bacillus, Streptomyces, various fungi, yeast and animal cells. Preferably, the host expresses the protein of the present invention extracellularly to facilitate purification and downstream processing. Expression and purification of the improved protein of the invention may be effected through art-recognized means for carrying out such processes.

The following is presented by way of example and is not to be construed as a limitation to the scope of the claims. Abbreviations used herein, particularly three letter or one letter notations for amino acids are described in Dale, J.W., Molecular Genetics of Bacteria, John Wiley & Sons, (1989) Appendix B.

EXAMPLES

EXAMPLE 1 Construction Of Plasmid pHP.BL

The α-amylase gene shown in Figure 3 was cloned from Bacillus licheniformis NCIB8061 (Gray et al., J. Bacteriology, vol. 166, pp. 635-643 (1986)). The 1.72kb Pstl-Sstl fragment, encoding the last three residues of the signal sequence, the entire mature protein and the terminator region, was subcloned into M13mp18. A synthetic terminator was added between the Bell and Sstl sites using a synthetic oligonucleotide cassette of the form:

Bell Sstl

5^•-GATCAAAACATAAAAAACCGGCCTTGGCCCCGCCGGTTTTTTATTATTTTTGAGCT-3^• (SEQ ID N0:1 ) 3' TTTTGTATTTTTTGGCCGGAACCGGGGCGGCCAAAAAATAATAAAAAC 5' (SEQ ID N0:2)

designed to contain the Bacillus amyioliquefaciens subtilisin transcriptional terminator (Wells et a!., Nucleic Acid Research, vol. 11 , pp. 7911-7925 (1983)).

The pBLapr plasmid was constructed carrying the gene for the Bacillus licheniformis α-amylase. As illustrated in Figure 7, pBLapr comprises a 6.1 kb plasmid including the ampicillin resistance gene from pBR322 and the chloramphenicol resistance gene from pC194, the aprE promoter and the gene encoding for the Bacillus licheniformis α-amylase ("BL AA"). The aprE promoter was constructed from a 660bp Hindlll-Pstl fragment encoding for the promoter and signal sequence of the Bacillus subtilis alkaline protease. The Pstl site was removed, and an Sfil site added close to the aprE/BL AA junction. The BL AA gene comprises the 1720 bp Pstl-Sstl fragment described above. In the work described herein, pBLapr was constructed with an Sfil site adjacent to the 5' end of the start of the coding sequence for the mature amylase gene. Specifically, the 5' end of the pBLapr construction was subcloned on an EcoRI-Sstll fragment from pBLapr into M13BM20 (Boehringer Mannheim) to obtain a coding-strand template for the mutagenic oligonucleotide below:

5'- CCC ATT AAG ATT GGC CGC CTG GGC CGA CAT GTT GCT GG - 3' (SEQ ID NO:3)

This primer introduced an Sfil site (indicated by underlining) which allowed correct forms to be screened for by the presence of this unique restriction site. Subcloning the EcoRI-Sstll fragment back into the pBLapr vector gave a version of the plasmid containing an Sfil site.

Plasmid pHP13 (Haima et al., Mol. Gen. Genet., vol. 209, pp. 335-342 (1987)) (Figure 6) was digested with restriction enzymes EcoRI and Hindlll and the resulting vector purified on a polyacrymide gel and then eluted. Plasmid pBLapr was digested with Hindlll, Asp718 and in a separate incubation with Asp718, EcoRI and gel purified. Two bands, Hindlll- Asp718 (1203 bp) and Asp718-EcoRI (1253 bp)were gel purified, eluted from the gel and ligated into the vector by a 3-way ligation, to give plasmid pHP.BL, the plasmid used in expression of the α-amylase (Figure 8).

EXAMPLE 2

Construction Of Plasmid Encoding α-Amylase Comprising Substitutions For Asparaqine 188

A series of mutagenic primers encoding for substitutions of Asn188 ("N188") with each of the naturally occurring amino acids were synthesized and are shown in Figure 1 (SEQ ID NOS:4-22). The α-amylase gene mutations encoding for these changes were made by PCR, according to the procedure summarized in Figure 9, using the PCR primers shown in Figure 2 (SEQ ID NOS:23-32).

Step (1): The mutagenic primers were used as templates for the PCR primers PCR A+ and PCR B- resulting in a lengthened (61 bp) double stranded DNA. Each contained a different amino acid replacement at position 188, and all except N188M contained a different restriction site. Initially the PCR primers were annealed at 35^βC for five minutes followed by a one minute DNA extension with taq polymerase at 75°C. The double stranded DNA was then melted at 95°C for one minute, followed by the annealing and extension steps. Melting, annealing and extension continued for a total of 30 cycles. Step (2): DNA upstream and downstream of position 188 were made in separate PCR reactions. The template was pBLapr, and the PCR primers were LAAfsδ (SEQ ID NO:27) and PCR A- (SEQ ID NO:24) for upstream; and PCR B+ (SEQ ID NO:25) and PCR Cla-Sall (SEQ ID NO:28) for downstream DNA. The DNA was melted at 95°C for one minute, annealed at 45°C for three minutes and elongated at 68°C for 3 minutes. The upstream portion is 290 bp and downstream is 498 bp. This procedure was repeated for 18 cycles using pfu polymerase. The same PCR procedure was used in steps (3) and (4).

Step (3): The upstream portion of DNA described in step (2) was attached to the double stranded mutagenic primers described in step (1). Primers LAAfsδ (SEQ ID NO:27) and PCR B- (SEQ ID NO:26) were used. As the result of primer design there is a 24 bp overlap between these templates allowing for the attachment of the two pieces of DNA.

Step (4): The downstream portions of DNA described in Step (2) and the product of Step (3) were attached to give the final product. A 24 bp overlap between the two PCR products allows for the attachment. Primers used were LAAfsδ (SEQ ID NO:27) and PCR Clal-Sall (SEQ ID NO:28).

Step (5): Unique restriction sites, Asp718 and BssHII, are located upstream and downstream, respectively, of the 188 site. The final PCR product is digested with Asp718 and BssHII, the 333 bp fragment isolated by polyacrylamide gel electrophoresis and subcloned into the pHP.BL vector to obtain pHP.N188X.

Mutations were confirmed by dideoxy sequencing (Sanger et al., Proc. Natl. Acad. Sci. U.S.A., vol. 74, pp. 5463-5467 (1977)).

With reference to the DNA sequence and numbering system used in Figure 3, the codon encoding for the +188 amino acid position is at base pairs 812-814. PCR primers A+ and A- correspond to base pairs 784-807. PCR primers B+ and B- correspond to base pairs 821-844. The 5' end of PCR primer LAAfsδ corresponds to base pair 518. The 5' end of PCR primer PCR Clal-Sall corresponds to base pair 1317. The Asp718 site corresponds to base pair 724. The BssHII site corresponds to base pair 1053.

EXAMPLE 3

Construction Of Plasmid Encoding Mutations At M15 And N188

A pBLapr plasmid having threonine substituted for methionine at amino acid 15 was constructed according to U.S. Patent Application Serial No. 08/194,664 (PCT Publication No. WO 94/18314). This plasmid (pBLaprM15T) was digested with Sfil and Asp718, and the 477 base pair fragment subcloned into pHP.BL to create pHP.M15T. In a manner analogous to that described above, Example 1, pHP.M15T was digested with Asp718 and BssHII, gel purified and eluted from the gel. The 333 base pair fragment comprising Asp718 to BssHII and the fragment from pHP.N188S were then subcloned into pHP.M15T to give plasmid pHP.M15T/N188S. In an analogous manner, starting with plasmids pBL aprM15L and pHP.NI88Y, the plasmid pHP.M15L/N188Y was constructed.

EXAMPLE 4

Transformation Of Plasmids Into Bacillus subtilis, Expression And Purification of Mutant α-Amylase

α-Amylase was expressed in Bacillus subtilis after transformation with the plasmids described in Examples 1-3. pHP13 is a plasmid able to replicate in E. coli and in Bacillus subtilis. Plasmids containing different variants were constructed using E. coli strain MM294, the plasmids isolated and then transformed into Bacillus subtilis as described in Anagnostopoulos et al., J. Bacter., vol. 81 , pp. 741-746 (1961). The Bacillus strain had been deleted for two proteases (Δapr, Δnpr) (see e.g., Ferrari et al., U.S. Patent No. 5,264,366) and for amylase (AamyE) (see e.g., Stahl et al., J. Bacter., vol. 158, pp. 411-418 (1984)). The bacillus strain expressing M15L/N188Y was found to form larger zones of clearing than the strain expressing M15L on agar plates containing 1 % insoluble starch indicating increased amylolytic activity. After transformation, the sacU(Hy) mutation (Henner et al., J. Bacter., vol., 170, pp. 296-300 (1988)) was introduced by PBS-1 mediated transduction (Hoch, J. Bact., vol. 154, pp. 1513-1515 (1983)).

Secreted amylases were routinely recovered from Bacillus subtilis cultures as follows: The culture supernatant was adjusted to 20% saturated ammonium sulfate and stirred for one hr. at 4°C. After centrifugation, the resultant supernatant was adjusted to 70% saturated ammonium sulfate and stirred for one hr. at 4°C. After centrifugation of the supernatant, the resultant pellet was re-dissolved in 50mM sodium acetate, pH 6.0, 5mM calcium chloride, and sterile filtered.

EXAMPLE 5 Assay For Determining α-Amylase Activity

Soluble Substrate Assay: A rate assay was developed based on an end-point assay kit supplied by Megazyme (Aust.) Pty. Ltd. A vial of substrate (p-nitrophenyl maltoheptaoside, BPNPG7) was dissolved in 10ml of sterile water followed by a 1 :4 dilution in assay buffer (50mM maleate buffer, pH 6.7, 5mM calcium chloride, 0.002% Tween20). Assays were performed by adding 10μl of amylase to 790μl of the substrate in a cuvette at 25°C. Rates of hydrolysis were measured as the rate of change of absorbance at 410nm, after a delay of 75 seconds. The assay was linear up to rates of 0.2 absorption units/min. α-Amylase protein concentration was measured using the standard Bio-Rad Assay (Bio-Rad Laboratories) based on the method of Bradford, Anal. Biochem., vol. 72, p. 248 (1976) using bovine serum albumin standards.

Starch Hydrolysis Assay: α-Amyiase activity on starch was determined through an assay which depends on the ability of starch to form a blue colored complex with iodine and the disappearance of this color when starch is hydrolyzed to shorter dextrin molecules. The α-amylase activity was defined in terms of the digestion time required to produce a color change denoting a definite state of dextrination of the starch. Reagents used were as follows:

Phosphate buffer- Potassium dihydrogen phosphate (340 g) and sodium hydroxide (25.3 g) were dissolved in water and diluted to ~ two liters. The buffer was cooled to room temperature and the pH was adjusted to 6.2±0.1. The buffer was diluted to two liters in a volumetric flask.

Starch substrate - Ten grams (dry substance) of soluble lintner starch were suspended in 50 ml of water and washed into ~300 ml of boiling water. The suspension was again brought to boiling and was boiled for five minutes with constant stirring. The starch solution was cooled with constant stirring to room temperature and 125 ml of phosphate buffer was added. The solution was diluted to 500 ml with water. The starch substrate was made fresh daily.

Stock iodine solution - Iodine crystals (5.5 g) and potassium iodide (11.0 g) were dissolved in water and were volumetrically diluted to 250 ml. The solution was kept from light. Dilute iodine solution - Potassium iodide (20 g) and two ml of stock iodine solution were dissolved in water and diluted volumetrically to 500 ml. The solution was made fresh daily.

Enzyme diluting solution - Calcium chloride (11.1 g) was dissolved in four liters of water. Water used for all reagents was either distilled or deionized. An α-amylase sample was diluted to between 10-15 LU/ml (as defined below) with enzyme diluting solution. For many commercial α-amylase preparations a suitable dilution was found to be 2000 fold. Five milliliter aliquots of dilute iodine solution were dispensed into 13 x 100 mm test tubes and 10 ml of starch substrate was placed in a 23 x 200 mm test tube. All tubes were placed in the 30°C water bath. A Hellige comparator equipped with a special α-amylase color disc (catalog number 620-s5) was used to make readings. Five milliliters of diluted enzyme (also at 30°C) were mixed with the starch substrate and timing was begun. At appropriate time intervals, for example one minute intervals early in the reaction and 15 second intervals later in the reaction, one ml aliquots of the enzyme- substrate mixture were transferred to a tube containing the dilute iodine solution. The starch iodine solution was mixed and transferred to a 13 mm precision square tube and the color was compared with the standard α-amylase color disc in the Hellige comparator. When the time of the end point was approached, samples were taken at 0.25 minute intervals.

The time required for the colors of the samples and the color disc to match were recorded and the activity (in liquefons per gram or ml) was calculated according to the formula:

LU/ml or LU/g =

Where: LU = liquefon unit

V = volume of enzyme (5 ml or grams) t = dextrinization time (minutes)

D = dilution factordilution volume divided by ml or g of enzyme diluted.

EXAMPLE 6 Preparation and Testing of Additional Mutant Alpha-Amylases for Thermal Stability

Mutant alpha-amylases were prepared having substitutions at one or more of five positions for which corresponding residues in both Bacillus stearothermophilus and Bacillus amyioliquefaciens were identical: A33S, A52S, N96Q S148N, A379S in combination with M15T/H133Y/N1885/A209V and compared with a mutant comprising only the

M15T/H133Y/N1885/A209V substitutions. Additionally, the mutation S85D which represents a recruitment from the Bacillus amyioliquefaciens. The mutations were prepared according to the procedures provided in Examples 1-4 except that appropriate PCR primers were provided to effect the desired mutations. Thermal inactivation rates for the various mutants were measured according to the following procedure. Amylase stock solutions were dialysed extensively into 20 mM ammonium acetate, 4 mM CaCI₂ pH 6.5. For measurement of stability, this stock was diluted >50fold into a solution designed to induce rapid inactivation of wild type amylase: 50mM ammonium acetate, 5mM CaCI₂, 0.02% Tween 20 and a pH of 4.9, or 4.8 to a final concentration of between 30 and 50 μg/ml. Six 100μl aliquots were put into Eppendorf tubes and placed into a water bath at either 82 ' C or 83 ^' C. The Eppendorf tubes were removed at regular, measured intervals of between 30 seconds and 5 minutes and placed on ice to stop the inactivation. The residual activity was assayed using a soluble substrate as described in Example 5. The natural log of the activity was plotted against time of incubation, and the rate constant for inactivation obtained from the slope of the straight line. The half-life was calculated as fn(2) divided by the rate constant. Results for various mutants are provided in Tables 1-4.

TABLE 1

Claims

An improved protein comprising an amino acid sequence which has been modified from a precursor amino acid sequence by the substitution or deletion of an amino acid residue which differs from a corresponding amino acid residue in a less stable but homologous protein, wherein said improved protein has altered stability compared to a protein corresponding to the precursor amino acid sequence.

The protein according to claim 1 , wherein said protein comprises an enzyme.

The enzyme according to claim 2, wherein said enzyme comprises a hydrolase, an oxidoreductase, a transferase, a lyase or a ligase.

4. The enzyme according to claim 3, wherein said enzyme comprises an amylase, lipase, cellulase, protease, hemicellulase, glucoamylase, esterase, lactase, polygalacturonase, β-galactosidase or ligninase.

5. The α-amylase according to claim 4, wherein said precursor amino acid sequence comprises an amino acid sequence corresponding to α-amylase from Bacillus licheniformis, and said less stable but homologous enzyme comprises Bacillus stearothermophilus, Bacillus subtilis or Bacillus amyioliquefaciens.

6. The α-amylase according to claim 5, wherein said substituted amino acid residue is selected from amino acid residues which differ from corresponding residues in both Bacillus stearothermophilus and Bacillus amyioliquefaciens, and wherein said corresponding residues in both Bacillus stearothermophilus and Bacillus amyioliquefaciens are the same amino acid.

7. The α-amylase according to claim 6, wherein said substituted amino acid residue is substituted with the same amino acid as is present at the corresponding residue in both of said Bacillus stearothermophilus and Bacillus amyioliquefaciens.

8. The α-amylase according to claim 5, wherein said substituted amino acid residues correspond to A33, A52, S85, N96, H133, S148, A209, A269, A379 and A435.

9. The α-amylase according to claim 7, wherein said substituted amino acid residues correspond to A33S, A52S, N96Q, H133Y, S148N, A209V, A269K, A379S and/or A435S.

10. The α-amylase according to claim 7, comprising additional substitutions at M15, N188 and/or M 197.

11. The α-amylase according to claim 10, wherein said substitutions comprise M15T/A33S/H133Y /N188S/A209V; M15T/D28N/A33S/H133Y/N188S/A209V; M15T/A52S /H133Y/N188S/A209V; M15T/A52S/H133Y/N188S/A209V/L230F;

M 15T/S85D/H 133Y/N 188S/A209V; M 15T/N96Q/H 133Y/N 188S/A209V;

M15T/N96Q/H133Y/N188S/A209V/I479T; M15T/H133Y/S148N/N188S/A209V;

M15T/H133Y/S148N/N188S/A209V/A379S;

M 15T/H 133Y/S 148N/ 188S/A209V/G433D; M 15T/H 133 Y/N 188S/A209 V/A379S; M15T/H 133Y/N 188S/A209V/A210S/T322A/A379S.

12. The protein according to claim 1 , wherein said less stable but homologous protein has a sequence identity of at least 60% to said precursor protein.

13. The protein according to claim 12, wherein said less stable but homologous protein has a sequence identity of at least 65% to said precursor protein.

14. The protein according to claim 12, wherein said less stable but homologous protein has a sequence identity of at least 80% to said precursor protein.

15. The protein according to claim 1 , wherein said modified protein has improved stability compared to said precursor protein.

16. A method for the production of an improved protein based on a precursor protein comprising:

(a) aligning the sequence of said precursor protein with a sequence of a less stable but homologous enzyme and identifying one or more residues which differ between the sequence of the precursor protein and said less stable but homologous enzyme; (b) selecting one or more of the residues identified in step (a) for substitution in said precursor protein; (c) modifying said precursor protein to incorporate a residue for substitution selected in said step (b).

17. The method according to claim 16, wherein said protein comprises an enzyme.

18. The method according to claim 17, wherein said enzyme comprises a hydrolase, an oxidoreductase, a transferase, a lyase or a ligase.

19. The method according to claim 17, wherein said enzyme comprises an amylase, lipase, cellulase, protease, hemicellulase, glucoamylase, esterase, lactase, polygalacturonase, β-galactosidase or ligninase.

20. The method according to claim 19, wherein said precursor amino acid sequence comprises an amino acid sequence corresponding to α-amylase from Bacillus licheniformis, and said less stable but homologous enzyme comprises Bacillus stearothermophilus, Bacillus subtilis or Bacillus amyioliquefaciens.

21. The method according to claim 20, wherein said substituted amino acid residue is selected from amino acid residues which differ from corresponding residues in both Bacillus stearothermophilus and Bacillus amyioliquefaciens, and wherein said corresponding residues in both Bacillus stearothermophilus and Bacillus amyioliquefaciens are the same amino acid.

22. The method according to claim 21 , wherein said substituted amino acid residue is substituted with the same amino acid as is present at the corresponding residue in both of said Bacillus stearothermophilus and Bacillus amyioliquefaciens.

23. The method according to claim 16, wherein said less stable but homologous protein has a sequence identity of at least 60% to said precursor protein.

24. The method according to claim 12, wherein said less stable but homologous protein has a sequence identity of at least 65% to said precursor protein.

25. The method according to claim 12, wherein said less stable but homologous protein has a homology of at least 80% to said precursor protein.

26. A method for the liquefaction of starch comprising the steps of:

(a) preparing an aqueous solution of starch; and

(b) contacting said aqueous solution of starch with the α-amylase according to claim 5.

27. A detergent composition comprising an enzyme according to claim 2.

28. A feed additive composition comprising an enzyme according to claim 2.