NZ225798A - Procaryotic xylose isomerase muteins and methods of increasing their stability - Google Patents

Description

A 225798 Priority Date(s): JSSvS5. :§?^ ...AA =f Complete Specification Filed Class: ??.?y •••■••••.••■• • <c \ ^W\61 go;.C\^P.^\[ B* V dec" '^O'" PubliceT!2n Date: ^ • • • • • • • • •1 ,\350. • • • • < P.O. Journal, No: NEW ZEALAND,- I PATENTS ACT, 1953 No.: Date: COMPLETE SPECIFICATION PROCARYOTIC XYLOSE ISOMERASE MDTEINS AND METHOD TO INCREASE PROTEIN STABILITY jP/We, CETUS CORPORATION, a corporation of the State of Delaware, USA, of 1400 Fifty-Third Street, Emeryville, California 94608, United States of America, THE STATE OF OREGON ACTING BY AND THROUGH THE OREGON STATE BOARD OF HIGHER EDUCATION ON BEHALF OF THE DNIVERSITY OF OREGON, a corporation of the State of Oregon, of Susan Campbell Hall, University of Oregon Campus, Eugene, Oregon97403-1229, United States of America hereby declare the invention for which we pray that a patent may be granted to rr^/us, and the method by which it is to be performed, to be particularly described in and by the following statement:- (followed by page 1A) I' 225798 PROCARYOTIC XYLOSE ISOMERASE MUTEINS AND METHOD TO INCREASE PROTEIN STABILITY This invention relates to improved mutant forms of an industrially valuable enzyme and to site-specific mutations to direct microbial production of these forms. More specifically, the invention relates to mutated procaryotic xylose isomerases with improved stability and/or catalytic activity, and/or lowered pH optima, and to a method for predicting certain amino acid substitutions that increase the stability of proteins.

The conversion of glucose to fructose by the enzyme xylose isomerase is an important industrial process because fructose is sweeter to human taste than an equivalent amount of glucose or sucrose. Fructose has nutritional advantages over glucose or sucrose as a sweetener because less fructose is needed to impart a desired level of sweetness, and because it does not support the growth of the bacteria responsible for dental plaque as well as does sucrose which is the only economically competitive sweetener. However, the maximum exploitation of these benefits depends on rendering fructose economically competitive with alternative sweeteners, by devising the least expensive process for manufacturing food-grade fructose.

Current industrial practice uses a single-step enzyme-catalyzed isomerization of glucose to an approximately equilibrated mixture of glucose and fructose, known as high-fructose syrup. Using this process, at equilibrium, only approximately 50% of the glucose has been transformed (Tewari et al., Appld. Bioch. and Biotech. 11:17-24 (1985)). Because percentage conversion varies directly with temperature, the fructose yield, and potentially the process economics, benefit from performing industrial glucose isomerization at the highest practical temperature. r*- 225798 Enzymes which catalyze the isomerization of sugars, including glucose, have been isolated from various organisms, including Bacillus subtilis. Escherichia coli. Ampullariella species and several Streptomyces species. The Streptomvces enzyme commonly used for commercial fructose production is most accurately designated xylose isomerase (XI), because it has much higher activity in converting xylose to xylulose than turning glucose into fructose. For industrial use, the purified enzyme is immobilized by adsorption to a solid support packed into a column, or "reactor", through which a concentrated solution of glucose is passed at the highest feasible temperature. The enzyme near the reactor inlet experiences a high concentration of glucose and low concentration of fructose. The enzyme near the reactor outlet is exposed to approximately equal concentrations of glucose and fructose. At any level in the catalytic reactor, the isomerase catalytic rate (V) depends on glucose (S) and fructose (P) concentrations ([ ]) as indicated in the following rate equation: kcatf [S]/KS - kcatr [P]/KP V/lE,° 1 + [S]/KS - [pl'KP In this equation [E]0 is the total enzyme concentration, Kg is the Michael is constant for glucose, Kp is the Michaelis constant for fructose, and V/[E]0 is the enzyme specific activity, an expression of the catalytic effectiveness per enzyme molecule. kcat£ and kcatr report the intrinsic catalytic activities of the glucose-saturated and fructose-saturated enzyme active sites, respectively representing the maximum possible forward (glucose * fructose) and reverse (fructose ■* glucose) values of V/[E] for a given temperature and pH. Ks, Kp, kcatf/ and kcatr' vaiT with temperature, generally increasing with increased temperature below the temperature range where conformational unfolding of the enzyme occurs. Although Ks and Kp do not necessarily equal the respective dissociation constants for glucose and fructose, they •? m,c.« r* 225798 probably approximate the dissociation constants in the case of Streptomyces XI, and therefore are inversely related to the affinities of the enzyme for glucose and fructose substrates.

XI catalytic activity in the industrially relevant (forward) direction is enhanced by environmental or mutational changes which increase kcat or Kp or decrease kcat or Ks, increase the intrinsic forward catalytic efficiency or affinity for glucose or decrease the intrinsic reverse catalytic efficiency or affinity of XI for fructose. Currently used industrial glucose isomerization processes do not produce the maximum possible (equilibrium) percent conversion of glucose to fructose because the reaction slows as equilibrium is approached. Improvements which permit closer approach to equilibrium by weakening the fructose-XI interaction or by strengthening glucose-XI binding can be as valuable as improvements which permit conversion at higher temperature, where the equilibrium percent conversion is greater.

The preceding rate equation implies that there are many ways to change k , k , K or K to get a net increase in V/[E]o* Cd U r C cl t . , b P Detrimental changes in one or more kinetic parameters can be outweighed by beneficial changes in others. Some combinations of changes would reduce net activity. Structural changes affecting activity will alter several or all of the parameters, not all of them favorably, for two reasons: (a) The four kinetic parameters are inescapably linked through the Haldane relationship: Keq = [P]equilibrium/[S]equilibrium = k K /k K CcI"l.£ p Cdt^ o At a given temperature and pH, a change in one parameter must be accompanied by a balancing change in some combination of the others to preserve the value of Keq* equilibrium constant; and r-. :^25798 (b) The relatively few amino acid residues which line the xylose isomerase active site interact with glucose, fructose, and catalytic intermediates. These interactions determine the values of the four kinetic parameters. Changing any one active site residue will strengthen or weaken several of these interactions and therefore modify several parameters.

It is thus difficult to target a simple set of improvements in catalytic activity because a change which improves one parameter may have strongly damaging effects on others. However, atomic resolution i.e. x-ray crystallographic data on the xylose isomerase active site permits the selection of a limited number of protein structural changes to increase net catalytic activity, for example, by strengthening the binding of glucose or by weakening the binding of fructose.

Recently, computer-graphic examination of the active sites of enzymes other than XI has led to successful prediction of structural changes affecting just kcat, Km, and substrate specificity for these enzymes {Wilkinson et al., Nature 307:187-188 (1984); and Craik et al., Science 228:291 (1985)).

In addition to identifying active site mutations that may improve kinetic parameters, computerized graphical examination of the atomic-resolution crystallographic data for XI also permits prediction of amino acid substitutions, insertions, or deletions to stabilize the enzyme toward conformational unfolding or inactivating chemical reactions. Following are several recent examples of structurally stablizing mutations accomplished by site-specific or random mutagenesis.

Replacement of a glycine residue located in an a-helix has conformationally stabilized a neutral proteinase, increasing the thermal melting temperature by several degrees centigrade (Imanaka et al., Nature 324:695 (1986)). * 3 or ^£T!98g/: 225798 Replacement of amino acids in the hydrophobic core of a protein with aromatic residues such as tyrosine, especially at positions near preexisting clusters of aromatic side chains, has been shown to promote resistance to thermal inactivation in kanamycin nucleotidyl transferase (Liao et al., Biochem. 83:576-580 (1986)), and phage Lambda repressor (Hecht et al., Biochem. 81:5685-5689 (1984)).

The introduction of new disulfide bonds to create covalent crosslinks between different parts of a polypeptide has been used to improve the thermal stability of bacteriophage T4 lysozyme (Perry et al., Science 226:555 (1984)), bacteriophage Lambda repressor (Sauer et al., Biochem. 125:5992 (1986)), E. coli dihydrofolate reductase (Villafranca et al., Biochem. 26:2182 (1987)), and subtilisin BPN' (Pantoliano et al., Biochem. 26:2077-2083 (1987)). A recently developed computer program (Pabo et al., Biochem. 25:5987-5991 (1986)) permits efficient scanning of the crystallographically determined three-dimensional structure of a protein to suggest those sites where insertion of two cysteines might lead to disulfide bonds which would not disrupt the larger-scale conformation while stabilizing the local conformation.

Deamidation of an asparagine residue near the inter-subunit interface of a homodimeric protein (triose phosphate isomerase) promotes irreversible thermal denaturation of this enzyme. Replacement of this asparagine with isoleucine enhanced thermal stability (Ahern et al., P.N.A.S. USA 84:675-679 (1987)).

Fusion of the subunits of the homotetrameric enzyme, B -galactosidase, by duplication and in-phase head-to-tail fusion of the structural gene for the enzyme, using a DNA polylinker coding for a number of additional amino acids, resulted in a protein that was more stable toward proteolysis and heat compared jto._ the wild-type enzyme (Kuchinke et al., EMBO J. 4(4) :106 7—^ £ A' ^ (1985)). V C;V ^ 3 OCT1983 ~ r> 225798 Another class of potentially inactivating reactions include oxidation of amino acid residues at or near the active site of an enzyme, leading to a loss or reduction in catalytic activity. For example, oxidation of a key methionine residue in the protein subtilisin has been shown to lead to loss of proteolytic activity (Markland et al., in The Enzymes (P. Boyer, ed.) Vol. 111:561 Academic Press (1971)). Replacement of that methionine by a serine, alanine or leucine residue resulted in an oxidation-resistant mutant protein (Estell et al., J. Biol. Chem. 260:6518-6521 (1985)).

Recent studies also have shown that some amino acid substitutions may have cumulative beneficial effects on thermal stability of the protein subtilisin (Bryan et al., J. Cellular Biochem. Supp. 11C (N305) (1987); Matsumura et al., Nature (Letters) 323:356-358 (1986)).

Xylose isomerase isolated from S. rubiqinosus (strain C3) has a pH optimum at 25°C of 8.8 and at 75°C the pH optimum drops to 7.3. The commercial isomerization reaction is normally run at 55°C to 65°C at a pH of approximately 6.5. Even at 65°C and pH 6.5 there is a significant amount of alkaline degradation of fructose and glucose to colored and bitter byproducts that are undesirable in high-fructose corn syrup. If the pH optimum of glucose isomerase could be dropped to from pH 5.5 to 6.5 there would be less alkaline degradation of fructose or glucose. Additionally, the isomerization reaction could be run at a pH at which the enzyme should have greater activity.

Computer graphic analysis of X-ray crystallographic stucture of a protein also provides the ability to predict amino acid alterations that may reduce the pH optimum of an enzyme. By reducing this parameter, the enzyme xylose isomerase could be used to produce a high fructose corn syrup commercially at a lower pH without any loss in activity, thereby reducing the '3 0CT;?en r: 2257 alkaline formation of undesirable degradation products of glucose and fructose. Such a mutein would also be desirable for use with other enzymes of low pH optima, e.g. glucoamylase, thereby reducing the overall number of steps required to manufacture high-fructose corn syrup, in single batch processes.

Recently, studies have shown that the pKa of enzyme active-site functional groups, and consequently the pH optimum of enzymes can be altered by specific mutations near, but not necessarily at, the active-site[s] of the enzyme subtilisin (Thomas et al., Nature 318:375-376, (1985); Sternberg et al., Nature 330:86-88, (1987)). In these studies, site-specific mutations within 10-15 angstroms of the active-site histidine residue (H1s64) lowered the pKa of subtilisin by 0.18 to 1.00 (for a double mutant) pKa units. Although naturally occurring glucose isomerases may have pH optima that vary from pH 7 to pH 9, none have a pH optimum as low as desired (for example pH 5.5).

Classical mutation of bacteria using radiation or chemicals has been used to produce mutant strains having different properties including altered protein activity. However, selective improvement of the organisms or the proteins has not been realized due to the randomness of the mutation process, which also requires tedious selection and screening steps to identify organisms which may possess the desired characteristics. Furthermore, with random mutagenesis an undesirable property may result along with the characteristic sought in the mutation.

More recently, random mutagenesis has been replaced by site-specific (also known as primer-directed) mutagenesis. Site-specific mutagenesis permits substitution, deletion or insertion of selected nucleotide bases within a DNA sequence encoding a protein of interest using synthetic DNA oligonucleotides having the desired sequence. Recombinant DNA procedures are used to substitute the synthetic sequence for the target sequence to (. i G tj y\, o ~7~ 13 0CTI9SS 225 r introduce the desired mutation. (See Craik et al., Science. 228:291 (1985) for a review of these procedures). Development of the Ml3 bacteriophage vectors (Messing, in Methods in Enzymoloqy 101:20-78 (1983)) permits cloning of DNA fragments into single-stranded circular recombinants capable of autonomous replication. A modification of site-specific mutagenesis, termed gapped circle mutagenesis, provides an improved method for selective mutagenesis using oligonucleotide primers (Kramer et al., Nuc. Acids Res. 12:9441-9456 (1984)). Kits for carrying out site-specific mutagenesis and the gapped circle method are commercially available.

Mutant xylose isomerases having characteristics which vary from native enzyme would be useful. In particular, mutant isomerases having enhanced oxidation and thermal stability would be useful to extend the commercial utility of the enzyme.

Unfortunately, unless proteins share regions of substantial sequence or structural homology, it is not possible to generalize among proteins to predict, based on beneficial mutation of one protein, precisely where the sequence encoding another protein should be changed to improve the performance of that protein. It is therefore generally necessary to undertake an analysis of the precise structural and functional features of the particular protein to be altered in order to determine which amino acids to alter to produce a desired result such as increased thermostability or catalytic activity.

Summary of the Invention The present invention provides mutated forms ("muteins") of enzymatically active procaryotic xylose isomerase. Analysis of the structure of Streptomvces rubiqinosus xylose isomerase (XI), to select alterations encoding the enzyme to enhance stability and/or activity and/or lower the pH optimum of the resulting XI r*> 225798 muteins, was undertaken. Site-specific mutagenesis of the sequence encoding the enzyme was designed to produce the muteins. Regions of structural homology between xylose isomerases from several microorganisms were identified.

Accordingly, the present invention provides muteins containing specific modifications of procaryotic xylose isomerase, and materials and methods useful in producing these proteins, as well as modified microorganisms and cell lines useful in their production. Other aspects of the invention include the expression constructs and products thereof for the modified xylose isomerases as well as cloning vectors containing the DNA encoding the modified xylose isomerases.

The DNA encoding the reference procaryotic xylose isomerase is modified using site-directed gapped circle mutagenesis enabling the generation of a change at a selected site within the coding region of the isomerase. By this method, a change is introduced into isolated DNA encoding procaryotic xylose isomerase which, upon expression of the DNA, results in substitution of at least one amino acid at a predetermined site in the xylose isomerase, or insertion of a polylinker peptide for fusing at least two subunits of the xylose isomerase protein.

The present invention also provides a method of enhancing thermostability in proteins by introducing proline amino acid substitutions into a protein to decrease the entropy of unfolding of the protein.

The modified xylose isomerases of the invention may exhibit improved stability and/or catalytic activity, and may have varied k cat » Ksor Kp. In addition, the muteins may exhibit a lowered pH optimum.

One aspect of the invention is a method for increasing the. stability of a protein by substituting an amino acid at a * m ■57 3 8 preselected substitution site in the protein, the substitution site selected to have phi and psi backbone conformational angles in the range of phi = -40° to -90° when psi = 0° to -60°, or in the range of phi = -40° to -95° when psi = 120° to 180° and the site capable of accommodating the amino acid without disruption of the three-dimensional structure of the protein such that introduction of the amino acid decreases the configurational entropy of unfolding of the protein.

Another aspect of the invention is a method for increasing the stability of a protein comprising substituting a glycine amino acid residue having a negative phi angle with an alanine to decrease the configurational entropy of unfolding of the protein.

Another aspect of the invention is a method for selecting substitution sites suitable for introduction of amino acids in a protein such that introduction of the amino acids increases the stability of the protein by a) determining from the crystallographic structure of a protein the backbone conformational angles phi and psi of said protein; b) screening the phi and psi angles determined in step a) to identify potential substitution sites in the protein having conformational phi and psi angles in the range of phi = -40° to -90° when psi = 0° to -60°, or in the range of phi = -40° to -95° when psi = 120° to 180° for introduction of said amino acids; and c) examining a structural model of the protein to from the potential substitution sites identified in step b) substitution sites that will accomodate substitution of an amino acid without disruption of the three-dimensional structure of the protein, such that substitution of the substitution site results in a decrease in the configurational entropy of unfolding of the protein.

In yet another aspect of the invention Streptomvces rubiqinosus. (S. rubiqinosus), xylose isomerase mutein is provided having a change in at least one position in the native 3.'V -25798 amino acid sequence at a position equivalent to a native amino acid residue selected from the group consisting of Lysinei83, Lysine289r Histidine54, Histidine220» Methionine223* Argininei40, Tryptophan]^, Tryptophan]^, Phenylalanineg4, Glycinei46, Glycinei66/ Glycine;^, Glycine2i9^ Glycine23i, Glycine248, Glycine298/ Glycine305, Glycine3gg, Leucine;^, Alanine2g, Alanine33, Asparaginei07, Arginineiog, Glycinei46, Valineisi, Glycinelog, Leucinei92, Glutamic acid207f Arginine25g, Threonine342# Arginine354, Glycine3gg, Aspartic acid28> Arginine32, Serine^, Valine2i8» Arginine292# Isoleucine252t Aspartic acidg, Glutamine2i, Alanine29, Arginine32, Glutamic acid38r Leucine46, Aspartic acids6, Leuciness, Valinei27, Threoninei33, Alaninei36, Argininei77, Isoleucineiso> Leucineig3, Leucine2H, Asparagine227/ Glutamine234, Alanine238» Leucine246f Arginine284» Arginine308' Leucine3n, Arginine3ig, Leucine335, Valine362' Methionine370, Leucine375, Leucine383, Glutamine2i, Asparagineg2> Asparagineio7, Asparagineiss, Asparagine227/ Glutamine234, Glutamine256/ Asparagine309, Glutamine377, Tryptophan270/ Glycinei46, Phenylalanine320, Histidine382/ Glutamic acid337, Arginineio9/ Glycinei89» Glutamic acidi44, Glycine25l, Glycine225r Alanine366' Valinega, Glutamine249, Glycine2i9r Glutamic acid207» Aspartic acidi63, Aspartic acid57, Glutamic acidi86? Glutamic acidi4i, Glutamic acid221» Aspartic acid2877 Argininei77,* and Aspartic acid345« Figure 1 is drawings depicting the structure of native and mutant T4 lysozymes as described in Example I, infra. 1A: Stereo drawing showing the structure of native T4 lysozyme in the vicinity of Gly77; solid circles = oxygen atoms, half-solid circles = nitrogen atoms and open circles = carbon atoms, broken lines = hydrogen bonds; IB: electron density map showing the difference in density between mutant G77A and native lysozyme (coefficients are (Fmut"FwT) and phases are from the refined model of native lysozyme (Weaver and Matthews, J. Mol. Biol. 193:189-199 (1987)). .. "w o -u- ' 30CT;?33 o Resolution is 1.7 A. Positive contours (solid lines) and negative contours (broken lines) are drawn at levels of + 4a, where o is the root-mean square density throughout the unit cell. The native structure is superimposed. Amino acids are indentified by the one-letter code. The positive peak due to the addition of the 8-carbon at residue 77 is of height 13a). 1C: Superposition of the structures of G77A lysozyme (open bonds) and native lysozyme (solid bonds).

Figure 2 are drawings depicting the comparative structures of A82P lysozyme and native lysozyme as described in Example I, infra. 2A: Electron density difference map for A82P lysozyme minus native lysozyme. (Coefficients, contour levels and conventions are as in Figure IB, supra. The positive peak indicating the addition of the pyrrolidine ring [of proline] is of height 11a; the negative peak due to displacement of the bound solvent molecule W355 is -11a. Part of the side chains of Leu79 an<^ Argso were omitted for clarity). 2B: Superposition of the structures of A82P lysozyme (open bonds) and wild-type lysozyme (solid Figure 3 are graphs showing the kinetics of inactivation of native and mutant lysozymes as described in Example I, infra. 3A: First-order plot, all activities normalized to 1000 units/ug at zero time; 3B: second-order plot, A0/At is the ratio of the initial activity to the activity remaining after time t. (Second-order rate constants: native: 10.4 X 10^ mol-^sec-1 G77A, 6.7 X 10^ mol-1* sec-1; and A82P, 2.1 X 10^ mol-1,sec-1.

Figure 4 is a restriction map of the XI gene and flanking region on the Streptomvces rubiqinosus chromosome; Figure 5 (A and B) shows the DNA sequence and DNA-deduced amino acid sequence of Streptomvces rubiqinosus xylose isomerase used as the reference protein; bonds). f1 ^ 0CTi9$gr 22 5 Figure 6 is a comparison of the amino acid sequence of native reference Streptomvces rubiqinosus XI with the amino acid sequences of native XI from other organisms; Figure 7 is a graph depicting the effect of temperature on the glucose/fructose equilibrium; Figure 8 is a graph of relative activity of Streptomvces rubiqinosus XI as a function of temperature; and Figure 9 is a graph of the pH activity profile of a xylose isomerase.

Def initions As used herein "reference" xylose isomerase ("XI") refers to the xylose isomerase encoded by a DNA sequence obtained from Streptomvces rubiqinosus (S. rubiqinosus) derived from ATCC strain 21,175 as described in U.S. Patent No. 4,410,627, incorporated herein by reference. As used herein, XI is an enzyme having the characteristics of converting glucose to fructose and xylose to xylulose. Enzymes having this activity have an enzyme classification number of E.C.5.3.1.5.

"Mutein" in relation to the "reference" XI, refers to a protein having a related amino acid sequence which has enzymatic activity substantially the same as the reference XI in that the enzyme converts glucose to fructose and xylose to xylulose. However, it contains one or more amino acid substitutions, inversions, deletions or insertions for amino acid residues. These residues have been selected by predicting structural and chemical alterations that will result from particular substitutions at particular locations in the protein using x-ray crystallographic structural data for the reference XI. The term also includes a protein having an amino acid sequence related to the reference XI, but containing fused subunits. 25798 "Expression vector" refers to a DNA construct containing a DNA sequence encoding XI, which is operably linked to a suitable control sequence capable of effecting the expression of said DNA in a suitable host. Such control sequences include a promoter to effect transcription, an optional operator sequence to control such transcription, a sequence encoding a suitable RNA ribosome binding site, and sequences which control termination of transcription and translation. 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. In the present specification, "plasmid" and "vector" are sometimes used interchangeably as the plasmid is the most common form of vector at present. However, 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.

"Recombinant host cells", "host cells", "cells", "cell cultures" and so forth are used interchangeably to designate individual cells, cell lines, cell cultures and harvested cells which have been or are intended to be transformed with the recombinant vectors of the invention. The terms also include the progeny of the cells originally receiving the vector.

"Transformed" refers to any process for altering the DNA content of the host, including in vitro transformation procedures as described below, phage infection, or such other means for effecting controlled DNA uptake as are known in the art.

"Operably linked" as used herein regarding DNA sequences or genes refers to the situation wherein the sequences or genes are juxtaposed in such a manner as to permit their ordinary functionality. For example, a promoter operably linked to a coding sequence refers to those linkages where the promoter is capable of controlling the expression of the sequence. ^ 1 L 'i > ^ c -14- ./V •V' , 130CT/389; r ( "Control sequences" refers to DNA sequences which control the expression of the sequence which encodes the mutein. Examples include promoters for transcription initiation, optionally with an operator, enhancer regions, ribosome binding site sequences and translation signals which initiate and terminate translation of the gene. Such control sequences must be compatible with, i.e., operable in, the host into which they will be inserted, A number of naturally occurring xylose isomerases and their genes may be obtained from a variety of procaryotic organisms, such as Bacillus subtil is. Ampullariella species, both disclosed in U.S. Patent No. 3,826,714, S_j. rubiqinosus (ATCC 21,175 disclosed in U.S. Patent Nos. 3,666,628 and 4,410,627) and E. coli. The foregoing patents are incorporated by reference herein. In addition, naturally occuring mutants of xylose isomerase may be employed as sources for genetic material for mutation.

The DNA sequence encoding the gene for S^ rubiqinosus xylose isomerase may be obtained and cloned in accord with the general method herein. As will be seen from the examples, this method includes determining at least a portion of the amino acid sequence for the enzyme, synthesizing labeled probes having putative sequences encoding sequenced regions of the xylose isomerase, preparing genomic DNA libraries prepared from chromosomal DNA isolated from S^. rubiqinosus cells expressing the isomerase, and screening the library for the gene encoding xylose isomerase by hybridization to the labeled probes. Positively hybridizing clones are then restriction enzyme mapped and sequenced.

Once the xylose isomerase gene has been identified and cloned, a number of modifications are undertaken to modify the gene to encode enzyme muteins with improved characteristics compared to the reference enzyme, for industrial uses. The 1 3 OCT 198B,. c* rs 225798 reference enzyme is the enzyme prior to the modifications as described herein.

Crucial to selection of sites for mutagenesis is procurement of an atomic-resolution x-ray crystal structure of the reference enzyme. Computer graphics analysis of the enzyme's crystal structure allows the identification of specific sites for alteration that may result in muteins possessing improved properties.

To facilitate selection of the desired modifications, a strategy may be devised using a computer-based model-building system, for example using computer assistance such as the Proteus computer program described by Pabo et al., Biochem. 25:5987-5991 (1986), incorporated by reference herein. Generally, such methodology involves analyzing geometric aspects of protein structure revealed, for example, by x-ray diffraction crystallography. Preferably, such a strategy takes into account how the proposed modification will fit with the remaining (unmodified) portion of the protein, taking into consideration the environment of the amino acid residues.

The stability of a protein structure is determined by the net difference in free energy between the folded and unfolded forms of the protein. Both enthalpy and entropy contribute to the free energy terms. The contribution of any given residue to the configurational entropy of unfolding of the polypeptide backbone of a protein can be estimated as SCOnf = R In z where R is the gas constant and z is the number of conformations available to that residue in the unfolded state. (Schellman, C.R.Trav.Lab. Carlsberg Ser. Chim. 29:230-259 (1955).) The value of z is not the same for all amino acids. The pyrrolidine ring of proline restricts this residue to fewer conformations than are available to the other amino acids. As a consequence, the backbone configurational entropy of protein unfolding varies from one amino acid to another; glycine has the largest and proline has the least conformational entropy change of unfolding. r*. -25798 An exact evaluation of ASconf requires a statistical averaging over all conformations. However, an approximate estimate can be obtained by considering the area that is available to a given amino acid in a Ramachandran et al. conformational map (Ramachandran et al., J.Mol.Biol. 7:95-99 (1963)). If yy the relative area in a conformational map accessible to amino acid Y (where y = 1 for the entire map) and yn is the relative area that corresponds to residue Y in the folded structure, then the entropy of unfolding for residue Y is given (Nemethy et al., J. Phvs. Chem. 70:998-1004 (1966)) by ASconf(Y) = R ln(YY/YN> C2] From Eq. 2 and the assumption that changes in yn are negligible (see below), one can estimate the relative entropy of unfolding of a different type of residue, Z, relative to Y, namely, ASzy = ASconf(Z) - ASconf(Y) = R ln(Yz/YY^ 13] On this basis, Nemethy, et al.., supra, estimated that the backbone contribution to the entropy of unfolding of an alanine relative to a glycine is -2.4 cal/deg*mol ("eu") (1 cal = 4.184J). On the same basis, a proline relative to an alanine can be estimated to have a relative configurational entropy of unfolding of about -4 cal/deg*mol ("eu"). For T4 lysozyme at pH 6.5, -4 cal/deg*mol corresponds to a change of ^ 1.4 kcal in the free energy of unfolding and an increase in the melting temperature of about 3.5°C.

The present invention provides a method for predicting substitution by certain amino acids to decrease the configurational entropy of unfolding a protein, thus increasing protein stability. The most effective substitutions contemplated are those in which amino acids in a protein are replaced with prolines.

Another useful substitution contemplated is to replace glycine residues in a protein with alanine. To carry out the method of 13 ocr nag.,. > r-, 225798 the invention, the three-dimensional structure of a protein is obtained, for example by X-ray crystallography. From the three-dimensional coordinates the backbone conformational angles (phi <J> and psi i|j) are calculated and listed. Direct, visual inspection of these angles permits the selection of all possible sites where an amino acid such as proline can be accommodated. The values of $ and at the substitution site must be within one of two regions. In Region 1, 4> = -40° to -90° and ij> = 0° to -60°. In Region 2, $ = -40° to -95° and ^ = 120° to 180°. Sites with <J> and values 10° or more inside these limits are preferred. For proline substitution there is also a restriction on the $ and \|» values of the amino acid in the amino acid sequence of the protein immediately preceding the site of the proposed proline substitution. This restriction is as follows: If the value of the residue preceding the proline substitution site is between 0° and -90° then the substitution site itself must have $ and i|j values in Region 1. If the value of the residue preceding the proline substitution site is not between 0° and -90° then the substitution site itself can have 4> and tJj values in either Region 1 or Region 2. These quoted <J> and values are based on an analysis of the conformations of proline residues in all protein structures refined to a resolution of 1.7 A or better. These $ and i|> values were determined from the coordinates of proteins (x,y,z values) deposited in the Brookhaven Data Bank, Brookhaven, NY (Bernstein et al., J. Mo1. Biol. 112: 535-542 (1977), Coordinate listing of January 1, 1987).

Once the <J> and if values have been used to obtain potential sites for amino acid replacement, each site must be inspected to determine if the substitution can be made without disruption of the three-dimensional structure of the protein. Amino acid substitutions that would cause unfavorable steric interactions with other parts of the protein structure are avoided. Removal of an amino acid that makes favorable interactions with neighboring O •V257£ protein atoms is also avoided. The inspection of the three-dimensional protein structure can be carried out in different ways, for example by using a wire model of the structure, or by displaying a model of the structure on a graphics system with a program such as FRODO (described by Jones In Crystallographic Univ. Press, Oxford, pp. 303-317 (1982), available from Evans and Sutherland, Salt Lake City, UT 84108).

The method of the invention for predicting sites for replacement of glycine with alanine is similar in principle to that used for the proline substitutions. First the <t> and tli values of all glycine residues in the protein are inspected.

Only those glycines with a negative value of 4> are possible candidates for replacement with alanine because a positive value of is known to be unfavorable for residues with a 8-carbon (Ramachandran et al., J. Mol. Biol. 7:95-99 (1963)). A three-dimensional model of the protein is then inspected to determine those glycine to alanine substitutions that can be made without perturbing the three-dimensional structure of the protein.

The enhancement of protein stability based on the difference between backbone configurational entropy of different amino acids is not restricted to replacements involving proline or glycine. Residues such as threonine, valine and isoleucine with branched 0-carbons restrict the backbone conformation more than nonbranched residues. As a consequence, there are many possible amino acid substitutions that alter the backbone configurational entropy of unfolding of a protein and that may be used to increase protein stability.

One specific aspect of the present invention is a method for increasing the stability of a protein by substituting an amino acid at a preselected substitution site in the protein, the substitution site selected to have phi and psi backbone conformational angles in the range of phi = -40° to -90° when psjL_= 19 225758 0° to -60°, or in the range of phi = -40° to -95° when psi = 120° to 180° and the site is capable of accommodating the amino acid without disruption of the three-dimensional structure of the protein such that introduction of the amino acid decreases the con-figurational entropy of unfolding of the protein. The preselected substitution site may be any amino acid residue except proline and the amino acid introduced at the site is proline. The preselected substitution site may be a glycine amino acid residue with a negative phi angle and the amino acid introduced is alanine.

Another aspect of the invention is a method for increasing the stability of a protein comprising substituting a glycine amino acid residue having a negative phi angle with an alanine to decrease the configurational entropy of unfolding of the protein.

Another specific aspect of the invention is a method for selecting substitution sites suitable for introduction of amino acids in a protein such that introduction of the amino acids increases the stability of the protein by a) determining from the crystallographic structure of a protein the backbone conformational angles phi and psi of the protein; b) screening the phi and psi angles determined in step a) to identify potential substitution sites in the protein having conformational phi and psi angles in the range of phi = -40° to -90° when psi = 0° to -60°, or in the range of phi = -40° to -95° when psi = 120° to 180° for introduction of the amino acids; and c) examining a structural model of the protein to determine from the potential substitution sites identified in step b) substitution sites that will accommodate substitution of an amino acid without disruption of the three-dimensional structure of the protein, whereby substitution of the site results in a decrease in the configura-tional entropy of unfolding of the protein. 13 ocr '/2 . n 225793 If the amino acid introduced in the above-described method of the invention is proline, then the step of screening (step b) has the additional substep of determining whether the amino acid residue preceding the potential substitution site identified in step b) has psi angles between 0° and -90°, and if so, then the step c) of examining comprises the substep of determining a substitution site having phi and psi angles in the range phi = -40° to -90° when psi = 0° to -60°.

Another specific aspect of the invention provides Streptomvces rubiqinosus. (S. rubiqinosus), xylose isomerase mutein having a change in at least one position in the native amino acid sequence at a position equivalent to a native amino acid residue selected from the group consisting of Lysinei83/ Lysine289» Histidine54, Histidine220» Methionine223r Argininei4Q, Tryptophan^, Tryptophan]^, Phenylalanineg4, Glycinei4g, Glycines* Glycine]^, Glycine2l9* Glycine231r Glycine248' Glycine298> Glycine305, Glycine36g, Leucine]^, Alanine29, Alanine33, Asparaginei07, Arginine^og, Glycinei46, Valineisi, Glycinei89» Leucineig2, Glutamic acid207» Arginine259, Threonine342, Arginine354, Glycine3gg, Aspartic acid28, Arginine32f Serineg4, Valine2i8, Arginine2g2/ lsoleucine252> Aspartic acidg, Glutamine2i, Alanine2g, Arginine32, Glutamic acid38, Leucine46, Aspartic acid56, Leuciness, Valinei27r Threoninei33, Alaninei30, Argininei77, Isoleucineigo/ Leucineig3, Leucine2H, Asparagine227#■ Glutamine234, Alanine238/ Leucine246, Arginine284r Arginine308» Leucine3n, Arginine3i6, Leucine335, Valine362/ Methionine37o, Leucine375, Leucine383/ Glutamine2i, Asparagineg2, Asparaginei07, Asparaginei85, Asparagine227» Glutamine234, Glutamine256/ Asparagine309, Glutamine377, Tryptophan270» Glycinei46, Phenylalanine320/ Histidine382> Glutamic acid337, Arginineio9, Glycinei89, Glutamic acidi44, Glycine25i, Glycine225f Alanine366, Valinegs, Glutamine24g, Glycine2ig, Glutamic acid207/ Aspartic acidig3, Aspartic acid.-57y n 225708 Glutamic acid i86»* Glutamic acid 141, Glutamic acid22l» Aspartic acid287; Argininei77; and Aspartic acid345.

In another specific aspect of the invention the change is in the lysine amino acid residue equivalent to Lysi83 and the change is substitution by an amino acid selected from the group consisting of Arg, Gin, Asn, Asp, Glu, Ser, Thr, His, Tyr, Ala, Val, Leu and lie; or the change is in the lysine amino acid residue equivalent to lYs289 an<3 the change is substitution by an amino acid selected from the group consisting of Arg, Gin, Asn, Asp, Glu, Ser Thr, His, Tyr, Ala, Val, Leu and lie; or the change is in the histidine amino acid residue equivalent to His54 and the change is substitution by an amino acid selected from the group consisting of Gin, Glu, Asn, Asp, Ser, Thr, Ala, Val, and Tyr; or the change is in the histidine amino acid residue equivalent to His220 anc* the change is substitution by an amino acid selected from the group consisting of Gin, Glu, Asn, Asp, Ser, Thr, Ala, Val, and Tyr; or the change is in the methionine amino acid residue equivalent to Met223 an<* the change is subsitution by an amino acid selected from the group consisting of Gly, Ala, Val, Leu, lie, Phe, Tyr, Gin, and Asn; or the change is in the arginine amino acid residue equivalent to Argi4Q and the change is substitution by an amino acid selected from the group consisting of Gin, Asn, Glu, Asp, lie, Leu, Ala, Val, and Tyr; or the change is in the tryptophan amino acid residue equivalent to Trpi6 and the change is substitution by an amino acj.d n 225758 selected from the group consisting of Asn, Gin, Ser, Thr, Gly, Ala, Val, Leu, lie, Tyr, Phe, and His; or the change is in the tryptophan amino acid residue equivalent to Trpi37 and the change is substitution by an amino acid selected from the group consisting of Asn, Gin, Ser, Thr, Gly, Ala, Val, Leu, lie, Tyr, Phe, and His; or the change is in the phenylalanine amino acid residue equivalent to Pheg4 and the change is substitution by an amino acid selected from the group consisting of Thr, Ser, His, Val, Gly, Ala, lie, Leu, Asn, and Gin; or the change is substitution of the glycine amino acid residue equivalent to Glyx where x is selected from the group consisting of residues 146, 166, 197, 219, 231, 248, 298, 305 and 369, and the Gly substituted with an amino acid other than glycine; or the change is substitution by proline in the amino acid residue equivalent to an amino acid residue selected from the group consisting of Leuis, AsP28' Ala2g, Arg32, Ala33, Serg4, AsniQ7r Argio9/ Glyi46, Valisi, Glyisg, Leuig2, G1U207, Val218» He252f Arg259/ Arg2g2, Thr342, Arg354, Gly36g, Argi77, and ASP345; or the change is double substitutions of cysteine in the amino acid residues equivalent to pairs of amino acid residues selected from the group consisting of Trp270 and Glyi46, Phe320 and His382» GIU337 and Argio9, ^lYlQS and GIU144, Gly251 and Gly225' Ala336 and Valgs, Gln249 and Gly2ig, and/or G1U207 and Aspi63; or the change is substitution by tyrosine in the amino acid residues equivalent to an amino acid residue selected from the group consisting of Aspg, Gln2i, Ala29/ Arg32, GlU3g, Leu46, Asp56, Leusa, Vali27» Thri33, Alai36, Argi77f lienor Leui93» Leu2H, Asn2271 Gln234r Ala238' Leu246/ Arg284' Arg3Q8, Leu3n, Arg3i6/ Leu335, Val362> Met37Q, Leu375 and Leu383J o '3 °CT 1988 f! •-2579 8 the change is substitution by phenylalanine in the amino acid residue equivalent to an amino acid residue selected from the group consisting of Leu46, AsP56' LeU5Q, Thri33, Alai3g, Ilei80» Leuig3, Leu2H, Asn227, Gln234/ Ala238» Leu246> Leu3n, Leu335' Val362' Met37o, Leu375 and Leu303; or the change is substitution by tryptophan in the asparagine amino acid residue equivalent to Asn227/' or the change is substitution by an amino acid residue selected from the group consisting of Ala, Val, Leu, lie, Ser, Thr, His, Tyr, Lys, Arg, Met and Pro in the amino acid residue equivalent to an amino acid residue selected from the group consisting of Gln2i, Asng2, Asnio7» Asnigs, Asn227» Gln234» Gln256' Asn309, and Gln377; or the change is in the aspartic acid amino acid residue equivalent to ASP57 and the change is substitution by an amino acid selected from the group consisting of Lys, Arg, Gly, Ala, Gin, Asn, Thr and Ser; or the change is in the glutamic acid amino acid residue equivalent to Glui86 anc* the change is substitution by an amino acid selected from the group consisting of Lys, Arg, Gly, Ala, Gin, Asn, Thr and Ser; or the change is substitution of the aspartic acid amino acid residue equivalent to ASP57 and the substitution is with an amino acid other than aspartic acid or glutamic acid; or the change is substitution in the glutamic acid amino acid residue equivalent to Glui86 and the change is substitution by an amino acid other than aspartic acid or glutamic acid; or the change is substitution by glutamine in the glutamic acid amino acid residue equivalent to G1U221J or r- 225703 the change is substitution by glutamine in the glutamic acid amino acid residue equivalent to GIU141.

Still another aspect of the invention is a nucleic acid encoding the xylose isomerase of the invention where the nucleic acid is substantially free of nucleic acid that does not encode the xylose isomerase.

Another specific aspect of the invention is an expression vector for mutant procaryotic xylose isomerase which has the nucleic acid of the invention operably linked to control sequences compatible with a host cell.

Another aspect of the invention is a method for enhancing the conversion of glucose to fructose and xylose to xylulose by exposing an effective amount of the xylose isomerase muteins of the invention to glucose and xylose, respectively.

Still another aspect of the invention is an expressed xylose isomerase mutein which exhibits a change in one or more of the characteristics of chemical stability, kcat , ^,at , , K p temperature stability, specific activity 0/ the isomerase and a lowered ph optimum, as compared to a reference xylose isomerase.

After the desired modifications are selected, the DNA sequence encoding the protein is site-specifically mutagenized to substitute nucleotides encoding selected amino acids at the predetermined positions within the sequence.

In the following examples, modifications will be described in two different proteins, xylose isomerase and bacteriophage T4 lysozyme. It is to be understood that the above method of protein stabilization applies to proteins in general and is not restricted to the two examples provided herein.

■/'V o r* 0^ ~~o ■ ( o Site-specific mutagenesis (also known as primer-directed mutagenesis) is a technique which is well-established in the art. A preferred procedure is gapped circle mutagenesis (Kramer et al., Nucl. Acids Res. 12:9441-9456 (1986)). In this method the DNA sequence encoding the gene to be mutagenized is ligated into an M13 vector having amber mutations which prevent its replication. The oligonucleotide primer incorporating the desired nucleotide changes is ultimately joined to a similar M13 vector lacking the mutation. The phage incorporating primer preferentially replicates in a susceptible host, thus enriching for the altered gene.

In general, site-specific mutagenesis is performed by cloning the DNA sequence encoding the reference enzyme into a convenient M13 cloning vector and using an appropriate primer, to convert a residue at an identified position for example, to a conservative amino-acid replacement. A synthetic oligonucleotide complementary, except in areas of limited mismatching to the desired sequence, is used as a primer in the synthesis of a strand complementary to the single-stranded reference isomerase sequence in the phage vector. The resulting double-stranded DNA is transformed into a phage-supporting host bacterium. Cultures of the transformed bacteria are plated in top agar, permitting plaque formation from single cells which harbor the phage. Theoretically, 50% of the plaques will consist of phage containing the mutant form; 50% will have the original sequence. Using the gapped circle method, the plaques will be enriched for phage having the mutant form. The plaques are hybridized with kinased synthetic primer under stringency conditions which permit hybridization only with the desired sequence which will form a perfect match with the probe. Hybridizing plaques are then picked and cultured, and the DNA is recovered.

The mutated cloned xylose isomerase genes may then be ligated into an expression vector (which may also be the cloning 225738 vector) with requisite regions for replication in the host. The vector is transfected into a host for enzyme synthesis, and the recombinant host cells are cultured under conditions favoring enzyme synthesis, usually selection pressure such as is supplied by the presence of an antibiotic, the resistance to which is encoded by the vector. Culture under these conditions results in enzyme yields multifolds greater than the wild type enzyme synthesis of the parent organism, even if it is the parent organism that is transformed.

The mutated cloned xylose isomerases are used to transform a host cell in order to express the mutated isomerase. In the preferred embodiment, the mutated xylose isomerase gene is ligated into a high copy number plasmid. This plasmid replicates in hosts in the sense that it contains the well-known elements necessary for plasmid replication: a promoter operably linked to the gene (which may be the gene's own homologous promoter if it is recognized, i.e., transcribed by the host), a transcription termination and polyadenylation region (necessary for stability of the mRNA transcribed by the host from the xylose isomerase gene) which is exogenous or is supplied by the endogenous terminator region of the isomerase gene and, preferably, a selection gene such as an antibiotic resistance gene that enables continuous growth in antibiotic-containing media. High copy number plasmids also contain an origin of replication compatible with the host, thereby enabling large numbers of plasmids to be generated in the cytoplasm without chromosomal limitations. However, it is within the scope of the invention herein to integrate multiple copies of the isomerase gene into the host genome. This is facilitated by bacterial strains that are particularly susceptible to homologous recombination. The resulting host cells are termed recombinant host cells.

W*" n 'J O rr — r, o •-tCU ( J & Standard Methods Most of the techniques which are used to transform cells, construct vectors, effect hybridization with probe, and the like as well as to perform x-ray crystallography of a protein, are widely practiced in the art, and most practitioners are familiar with the standard resource materials which describe specific conditions and procedures (see for example, Maniatis et al., Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Press, Cold Spring Harbor, NY (1982)). However, for convenience, the following paragraphs may serve as a guideline.

Control Sequences And Corresponding Hosts Procaryotes most frequently are represented by various strains of E. coli. However, other microbial strains may also be used, such as bacilli for example, Bacillus subtilis. Pseudomonas sp., Streptomvces rubiqinosus: various species of fungi or other microorganisms. In such procaryotic systems, plasmid vectors which contain replication sites and control sequences derived from a species compatible with the host are used. For example, E. coli is typically transformed using derivatives of pBR322, a 2:95 (1977). pBR322 contains genes for ampicillin and tetracycline resistance, and thus provides additional markers which can be either retained or destroyed in constructing the desired vector. Commonly used procaryotic control sequences which are defined herein to include promoters for transcription initiation, optionally with an operator, along with ribosome binding site sequences, include such commonly used promoters as the beta-lactamase (penicillinase) and lactose (lac) promoter systems (Chang, et al., Nature (1977) 198:1056), the tryptophan (trp) promoter system (Goeddel, et al., Nucleic Acids Res (1980) 8:4057), and the lambda derived PL promoter and N-gene ribosome binding site (Shimatake, et al., Nature (1981) 292:128), which plasmid derived from an E. coli species by Bolivar, et al., Gene m 225798 has been made useful as a portable control cassette. Also useful is the phosphatase A (phoA) system described by Chang, et al., in European Publication No. 196,864 published October 8, 1986, and assigned to the same assignee and incorporated herein by reference. However, any available promoter system compatible with procaryotes can be used.

In addition to bacteria, eucaryotic microbes, such as yeast, may also be used as hosts. Laboratory strains of Saccharomvces cerevisiae. Baker's yeast, are frequently used, although a number of other strains are commonly available. Many plasmid vectors suitable for yeast expression are known. (See, for example, Stinchcomb, et al., Nature 282:39 (1979), Tschempe, et al., Gene 10:157 (1980) and Clarke, L., et al., Meth. Enzymol. 101:300 (1983)). Control sequences for yeast vectors include promoters for the synthesis of glycolytic enzymes (Hess, et al., J. Adv.

Enzyme. Reg. 7:149 (1968); Holland, et al., Biochemistry 17:4900 (1978)). Additional promoters known in the art include the promoter for 3-phosphoglycerate kinase (Hitzeman, et al., J. Biol.

Chem. 255:2073 (1980)), and those for other glycolytic enzymes, such as glyceraldehyde-3-phosphate dehydrogenase, hexokinase, n- pyruvate decarboxylase, phosphofructokinase, glucose-6-phosphate isomerase, 3 phosphoglycerate mutase, pyruvate kinase, triosepho-sphate isomerase, phosphoglucose isomerase, and glucokinase.

Other promoters, which have the additional advantage of transcription controlled by growth conditions, are the promoter regions for alcohol dehydrogenase 2, isocytochrome C, acid phosphatase, degradative enzymes associated with nitrogen metabolism, and enzymes responsible for maltose and galactose utilization (Holland, ibid). It is also believed that terminator sequences are desirable at the 3' end of the coding sequences.

Such terminators are found in the 3' untranslated region following the coding sequences in yeast-derived genes. Many of the vectors illustrated contain control sequences derived from the _29- / J 3 OCT/?## ' » i r- ^ yvi:; f".- 225798 enolase gene containing plasmid peno 46 (Holland, M. J., et al., J. Biol. Chem. 256:1385 (1981)) or the LEU2 gene obtained from YEpl3 (Broach, J., et al., Gene 8:121 (1978)), however, any vector containing a yeast compatible promoter, origin of replication and other control sequences is suitable.

It is also, of course, possible to express genes encoding polypeptides in eucaryotic host cell cultures derived from multicellular organisms. See, for example, Tissue Culture. Academic Press, Cruz and Patterson, editors (1973). Useful host cell lines include murine myelomas NS1, VERO and HeLa cells, and Chinese hamster ovary (CHO) cells. Expression vectors for such cells ordinarily include promoters and control sequences compatible with mammalian cells such as, for example, the commonly used early and late promoters from Simian Virus 40 (SV 40) (Fiers, et al., Nature 273:113 (1978)), or other viral promoters such as those derived from polyoma, Adenovirus 2, bovine papiloma virus (BPV), or avian sarcoma viruses, or immunoglobulin promoters and heat shock promoters. This system is disclosed in U.S. Patent No. 4,419,446. A modification of this system is described in U.S. Patent No. 4,601,978. General aspects of mammalian cell host system transformations have been described by Axel; U.S. Patent No. 4,399,216 issued August 16, 1983. It now appears, also, that "enhancer" regions are important in optimizing expression; these are, generally, sequences found upstream of the promoter region. Origins of replication may be obtained, if needed, from viral sources. However, integration into the chromosome is a common mechanism for DNA replication in eucaryotes.

Plant cells are also now available as hosts, and control sequences compatible with plant cells such as the nopaline synthase promoter and polyadenylation signal sequences (Depicker, A., et al., J. Mol. Appl. Gen. 1:561 (1982)) are available.

'J r/C'V /-"-V 225798 Recently, in addition, expression systems employing insect cells utilizing the control systems provided by baculovirus vectors have been described (Miller, D.W., et al., in Genetic Engineering. Setlow, J.K., et al., eds., Plenum Publishing, Vol. 8, pp. 277-297 (1986)).

Transformations Depending on the host cell used, transformation is done using standard techniques appropriate to such cells. The calcium treatment employing calcium chloride, as described by Cohen, S. N., Proc. Natl. Acad. Sci. (USA) 69:2110 (1972), is used for procaryotes or other cells which contain substantial cell wall barriers. Infection with Aqrobacterium tumefaciens (Shaw, C. H., et al.. Gene 23:315 (1983)) is used for certain plant cells. For mammalian cells without such cell walls, the calcium phosphate precipitation method of Graham and van der Eb, Virology 52:546 (1978) is preferred. Transformations into yeast are carried out according to the method of Van Solingen, P., et al., J. Bact. 130:946 (1977) and Hsiao, C. L., et al., Proc. Natl. Acad. Sci. (USA) 76:3829 (1979).

Probe of cDNA or Genomic Libraries "Southern Analysis" is a method by which the presence of DNA sequences in a digest or DNA-containing composition is confirmed by hybridization to a known, labeled oligonucleotide or DNA fragment. For the purposes herein, Southern Analysis shall mean separation of digests on 1 percent agarose and depurination as described by G. Wahl et al., PNAS (USA). 76:3683-3687 (1979), transfer to nitrocellulose by the method of E. Southern, J. Mol. Biol. 98:503-517 (1975), and hybridization as described by Maniatis et al., Cell. 15:687-701 (1978). cDNA or genomic libraries are screened using the colony or plaque hybridization procedure. Bacterial colonies, or the 31 225798 plaques for phage are lifted onto duplicate nitrocellulose filter papers (S & S type BA-85). The plaques or colonies are lysed and DNA is fixed to the filter by sequential treatment for 5 min with 500 mM NaOH, 1.5 M NaCl. The filters are washed twice for 5 min each time with 5 x standard saline citrate (SSC) and are air dried and baked at 80°C for 2 hr.

The gels for Southern blot or the duplicate filters for cDNA or genomic screening are prehybridized at 25°-42°C for 6-8 hr with 10 ml per filter of DNA hybridization buffer without probe (0-50% formamide, 5-6 x SSC, pH 7.0, 5x Denhardt's solution (polyvinylpyrrolidine, plus Ficoll and bovine serum albumin; 1 x = 0.02% of each), 20-50 mM sodium phosphate buffer at pH 7.0, 0.2% SDS, 20 yg/ml poly U (when probing cDNA), and 50 ug/ml denatured salmon sperm DNA). The samples are then hybridized by incubation at the appropriate temperature for about 24-36 hours using the hybridization buffer containing kinased probe (for oligomers). Longer cDNA or genomic fragment probes may be labeled by nick translation or by primer extension.

The conditions of both prehybridization and hybridization depend on the stringency desired, and vary, for example, with probe length. Typical conditions for relatively long (e.g., more than 30-50 nucleotide) probes employ a temperature of 42°C and hybridization buffer containing about 20%-50% formamide. For the lower stringencies needed for oligomeric probes of about 15 nucleotides, lower temperatures of about 25°-42°C, and lower formamide concentrations (0%-20%) are employed. For longer probes, the filters may be washed, for example, four times for 30 minutes, each time at 40°-50°C with 2 x SSC, 0.2% SDS and 50 mM sodium phosphate buffer at pH 7, then washed twice with 0.2 x SSC and 0.2% SDS, air dried, and are autoradiographed at -70°C for 2 to 3 days. Washing conditions are somewhat less harsh for shorter probes. 32 / h 2257 Minor variations from these specified hybridization methods are described in the examples below.

Vector Construction Construction of suitable vectors containing the desired coding and control sequences employs standard ligation and restriction techniques which are well understood in the art. Isolated plasmids, DNA sequences, or synthesized oligonucleotides are cleaved, tailored, and religated in the form desired.

Site-specific DNA cleavage is performed by treating the DNA with the suitable restriction endonuclease(s) under conditions which are generally understood in the art, and the particulars of which are specified by the manufacturer of these commercially available restriction enzymes. See, e.g., New England Biolabs, Product Catalog (New England Biolabs, Beverly, MA). In general, about 1 ug of plasmid or DNA sequence is cleaved by one unit of enzyme in about 20 yl of buffer solution. An excess of restriction enzyme is typically used to insure complete digestion of the DNA substrate; however, it may be desirable to carry out partial digestions in which some but not all of the sites of a given restriction enzyme in the DNA are cleaved. Such partial digestions are accomplished by varying the concentration of restriction enzyme or length of time the restriction digestion is carried out. Incubation times of about one hour to two hours at about 37°C are workable, although variations can be tolerated. After each incubation, protein is removed by extraction with phenol/chloroform, and may be followed by ether extraction, and the nucleic acid recovered from aqueous fractions by precipitation with ethanol. If desired, size separation of the cleaved fragments may be performed by polyacrylamide gel or agarose gel electrophoresis using standard techniques. A general description of size separations are found in Methods in Enzvmoloqy (1980) 6^:499-560; Lawn et al., Nucl. Acids Res. 9:6113-6114 (1981) and Goeddel et al., Nucl. Acids Res. 8:4057 (1980)). o 33 A ffs 22579 Restriction cleaved fragments may be blunt ended by treating with the large fragment of E. coli DNA polymerase I (Klenow) in /""N ' the presence of the four deoxynucleotide triphosphates (dNTPs) using incubation times of about 15 to 25 min at 20° to 25°C in 50mM Tris pH 7.6, 50mM NaCl, 6mM MgCl2» 6mM dTT, about 10 U/ul Klenow and 5-lOyM dNTPs. The Klenow fragment fills in at 5' sticky ends but chews back protruding 3' single strands, even ;; • though the four dNTPs are present. If desired, selective repair can be performed by supplying only one of the, or selected, dNTPs within the limitations dictated by the nature of the sticky ends. After treatment with Klenow, the mixture is extracted with phenol/chloroform and ethanol precipitated. Treatment under appropriate conditions with SI nuclease results in hydrolysis of any single-stranded portion.

Ligations are performed in 15-30 yl volumes under the following standard conditions and temperatures: 20mM Tris-HCl, pH 7.5, lOmM MgCl2, lOmM dTT, 33 ug/ml BSA, 10 mM-50 mM NaCl, and either 40pM ATP, 0.01-0.02 (Weiss) units T4 DNA ligase at 0°C (for "sticky end" ligation) or ImM ATP, 0.3-0.6 (Weiss) units T4 DNA ligase at 14°C (for "blunt end" ligation). Intermolecular "T~ "sticky end" ligations are usually performed at 33-100 ug/ml total DNA concentrations (5-100 nM total end concentration). Intermolecular blunt end ligations (usually employing a 10-30 fold molar excess of linkers) are performed at 1 uM total ends concentration.

In vector construction employing "vector fragments", the vector fragment is commonly treated with bacterial alkaline phosphatase (BAP) in order to remove the 5' phosphate and prevent religation of the vector. BAP digestions are conducted at pH 8 in approximately 150 uM Tris, in the presence of Na and Mg using about 1 unit of BAP per ug of vector at 60°C for about one hour. In order to recover the nucleic acid fragments, the preparation is extracted with phenol/chloroform and ethanol JQ / "o ' '98e 225798 precipitated. Alternatively, religation can be prevented in vectors which have been double digested by additional restriction enzyme digestion of the unwanted fragments.

Verification of Construction In the constructions set forth below, correct ligations for plasmid construction are confirmed by first transforming E. coli strain MM294, or other suitable host with the ligation mixture. Successful transformants are selected by ampicillin, tetracycline or other antibiotic resistance or using other markers depending on the mode of plasmid construction, as is understood in the art. Plasmids from the transformants are then prepared according to the method of Clewell, et al., P.N.A.S. (USA) 62:1159 (1969), optionally following chloramphenicol amplification (Clewell, J. Bacteriol. (1972) 110:667). The isolated DNA is analyzed by restriction and/or sequenced by the dideoxy method of Sanger, et al., P.N.A.S. (U.S.A.), 74:5463 (1977) as further described by Messing, et al., Nucleic Acids Res. 9:309 (1981), or by the method of Maxam, et al., Methods in Enzymoloqy 65:499 (1980).

Preparation of Synthetic Oligonucleotides for Modification of DNA Synthetic oligonucleotides may be prepared by the triester method of Matteucci, et al., J. Am. Chem. Soc. 103:3185-3191 (1981), or using automated synthesis methods. Kinasing of single strands prior to annealing or for labeling is achieved using an excess, e.g., approximately 10 units of polynucleotide kinase to 1 nM substrate in the presence of 50 mM Tris, pH 7.6, 10 mM MgCl2• 5 mM dithiothreitol, 1-2 mM ATP. If kinasing is for labeling of probe, the ATP will contain high specific activity 32yP. The synthetic oligonucleotide is used as a primer to direct synthesis of a strand complementary to the phage, and the resulting double-stranded DNA is transformed into a phage-supporting host bacterium. Cultures of the transformed bacteria i * /. o 7 225 are plated in top agar, permitting plaque formation from single cells which harbor the phage.

Hosts Exemplified Host strains that may be used in cloning and expression herein are as follows: For cloning and sequencing, and for expression of the construction under control of most bacterial promoters, E. coli strain MM294 obtained from the American Type Culture Collection, Rockville, MD (ATCC, No. 53,131) is used as the host. This particular strain contains a plasmid, pAW721, and should be plasmid cured prior to use. For expression under control of the trp promoter and trpE translation initiation signal in the expression vector pTRP3, E. coli strain MM294 or DG101, may be used. pTRP3 has been accepted for deposit under the terms of the Budapest Treaty, under accession No. ATCC 39,946.

For M13 phage recombinants E. coli strains susceptible to phage infection, such as E. coli K12 strain DG98 (ATCC No. 39,768) and HB2151 (commercially available from Anglican Biotechnology Ltd., Colchester, Essex, UK) are employed.

Mammalian expression may be performed in COS-7, C0S-A2, CV-1, and murine cells, and insect cell based expression in Spodoptera frugipeida.

The mutant xylose isomerases expressed upon transformation of the suitable hosts have similar enzymatic activity to the reference xylose isomerase and are screened for those that exhibit desired characteristics, for example, kinetic parameters, oxidation stability, thermal stability, lowered pH optima and the like 36 ' - ' ' "I'N-T- ; -■ .; ' ' \ ■ - - , I -i C^- ^ m 2257! f, ' A change in kinetic parameters is defined as an alteration in kcat^, kcat^., Kg and/or Kp. Procaryotic xylose isomerase muteins with increased or diminished k , k , K or K cat/ cat S P values may be obtained as described hereirr. Generally, the objective will be to obtain a mutein having a greater (numerically larger) kcat for the forward reactions (glucose to fructose, and xylose to xylulose), and a reduced (numerically smaller) Kg for the substrates glucose or xylose, thereby enabling the use of the enzyme to more efficiently process its target substrate. kcat and Ks are measured by known procedures, for example by analysis of the progress curve for these known parameters in the enzyme/substrate reaction. The rate of the reaction may be measured as a function of substrate concentration. Data are preferably analyzed by fitting to the Michaelis-Menten equation using a non-linear regression algorithm such as that described by Marquardt, J. Soc. Ind. Appl. Math. 11:431-441(1963). A lowered pH optimum is defined as a shift in the peak of the pH vs. activity profile toward lower pH values. pH vs. activity profiles are measured by known assay procedures under controlled pH conditions. Procaryotic xylose isomerase muteins having a lowered (numerically smaller) pKa may be obtained as described herein. These muteins may possess lowered pH optima.

The deduced amino acid sequence of the recombinant gene for xylose isomerase obtained as described above is shown in Figure 5 (A and B). This sequence was used in conjunction with x-ray crystallographic analysis and molecular modeling using a computer graphics system to display and analyze the three-dimensional structure of xylose isomerase, including the active site. In this manner the effects of replacement, insertion or deletion of one or more key amino acid residues, for example the effects on non-covalent interactions between the active site and the substrate (glucose) or end product (fructose), are determined. , c ^ vV // * . ■ 'V- o n r- 225798 Sites within the DNA sequence for the xylose isomerase of the invention are thus targeted for mutation to improve the activity and stability of the enzyme, for example, to alter the catalytic properties by reducing the Kgand increasing the kcat , increasing Kp, decreasing kcat and/or by increasing the enzyme's stability toward thermal and chemical inactivation. These same mutations may be used at homologous locations within the DNA sequences for other xylose isomerases obtained from other microorganisms, since many of the amino acid residues selected for mutation are conserved between the various isomerases as shown in Figure 6.

The present invention promotes high efficiency of glucose conversion and high yields of fructose, using muteins of procaryotic xylose isomerase which may be used industrially for isomerization of glucose to fructose. The various mutation strategies of the present invention may be grouped as follows: Minimization of Inactivating Chemical Reactions Mutations are directed to removal of certain amino acids at selected positions that contain an amino group capable of reacting with a reducing sugar such as glucose so as to irreversibly inactivate the enzyme. These mutations result in removal of lysine amino acids, the only amino acid containing epsilon amino groups, that can react with a reducing sugar to irreversibly inactivate the enzyme (i.e., undergo Maillard reaction). In these mutations, as in the others, it is preferable to attempt to maintain similar structure and/or chemical properties, for example by introducing amino acids that have similar numbers of atoms, or by conserving salt bridges, hydrophobic interactions or hydrogen bonds, thereby maintaining a conformation like that of the native protein. In addition, oxidation of methionine, histidine or tryptophan residues at or near the active site may lead to a reduction in catalytic activity. Histidine contains an imidazole group, and tryptophan an indole group that may be n 225788 oxidized. Mutations are targeted to replace methionine or histidine residues with amino acids that are not likely to be oxidized, such as glutamine or glycine. Arginine contains a guanido group susceptible to modification by dicarbonyl compounds such as 2,3- butanedione. Similar dicarbonyl compounds may inactivate xylose isomerase. Removal or replacement of arginine may prevent inactivation.

Enhancement of Catalytic Properties Based on the x-ray structure of S. rubiqinosus xylose isomerase, and molecular modeling studies involving substrate docking to the enzyme's active site, tryptophan residues at positions 16 and 137 in the amino acid sequence of the reference enzyme, and a phenylalanine residue at position 94 appear to be critical residues forming the substrate binding site. Because glucose is a larger molecule than xylose (glucose contains an additional -CH, OH), the binding site of xylose isomerase appears to be too small to readily accommodate the larger glucose molecule. Replacement of these key amino acid residues by amino acids possessing small functional groups may reduce the Kg for glucose, k , k . and/or Kp may also be altered.

Cfl Uj. CuL^ Thermostabilization Glycine residues in selected positions, e.g. alpha helices, B-strands or random structures that can accept the increased bulk of the substituted methyl group, are substituted with alanine residues to confer additional stabilization. In addition, proline substitutions are made at selected positions (the polypeptide-backbone torsion angles must accept the atypical proline angle values) to reduce the entropy of the unfolded form of the protein, and stabilize the native conformation. Selection of proline substitution sites may be based on analysis of Phi/Psi (<(>, \Jj) angles computed from the X-ray structure of the reference ? ^ 1 £ //''-?>, \ 39- :* f \e £ ft 225798 protein using the method of the invention, or may be based on analysis of the amino acid sequences of homologous proteins. For example, proline substitutions may be made in the reference XI at positions identified from comparison of the highly homologous and thermostable XI from Ampullariella which contains several proline residues that are not present in the reference XI. Additional stabilizing alterations include the introduction of disulfide bridges at conformationally acceptable positions in the XI structure. Both intersubunit or intrasubunit disulfide bridges in the tetrameric xylose isomerase are contemplated within this invention. Introduction of aromatic amino acid residues such as tyrosine, phenylalanine and tryptophan near aromatic clusters within the enzyme are also within the scope of the invention, to stabilize the enzyme at sites where the additional bulk of aromatic groups will not distort the overall conformation.

To prevent deamidation reactions, selected amino acids (asparagine and glutamine residues) near interfaces between subunits are altered by substitutions with amino acids such as alanine and valine that cannot undergo such reactions.

In addition, or as an alternative to amino acid substitution for increased stability, the structural gene for xylose isomerase is duplicated and the two copies of the gene are fused via a DNA sequence encoding a short peptide segment, between 3 and 10 amino acids long, between the N-terminus of one gene copy and the C-terminus of the other. Preferably oligoglycine or a combination of glycine and additional amino acids such as alanine, serine, threonine or proline is used as the short peptide sequence. Alternatively, the N-terminus of one gene copy can be fused directly to the C-terminus of a second gene copy, or short deletions encoding for 1 to 3 amino acids can be made at either end prior to fusion.

I r*. 225798 Lowered pH Optimum To lower the pH optimum (lower pKa) selected amino acids are mutated to alter the electrostatic potential at the xylose isomerase active site. This may be accomplished by changing negatively charged amino acids to positively charged amino acids near the active-site of the xylose isomerase, while preserving residues directly involved in substrate binding and catalysis (e.g. histidine at position 54 (Hiss4)).

The mutated isomerase proteins, or "muteins", may be more stable than the currently used naturally occurring enzymes at the high temperatures, near 100°C, needed to reach the desired conversion levels (greater than 55% fructose). Some of the stabilizing mutations simply reduce the rate of thermally induced unfolding of the protein conformation. Others prevent covalent modifications of the enzyme which might reduce catalytic activity or conformational stability. The isomerase muteins may have improved catalytic activity for any combination of three reasons: increased intrinsic catalytic activity, increased affinity for substrate glucose, or decreased affinity for product fructose. The muteins may also exhibit lowered pH optima.

These improvements are not completely independent. For example, increasing affinity for substrate can result in increased thermal stability by reducing the fraction of time that an enzyme active site is empty, as it generally is true that binding of substrates or products to an enzyme active site stabilizes the protein conformation.

The improvements contemplated herein are intended to improve the economics of glucose isomerization for several reasons. Increased stability toward conformational unfolding (thermal stabilization) and/or inactivating covalent modification increases the permissible operating temperature and resulting percent oV o 225798 conversion of glucose, or increases the operating lifetime of a given batch of catalyst, thus reducing the cost of catalyst per unit of product. Increased catalytic activity at a given operating temperature allows a given amount of catalyst to bring a mixture of glucose and fructose closer to equilibrium in less time. It also may reduce the amount of enzyme required, again lowering the cost of catalyst per unit of product.

Any number of mutations proposed herein may be combined in a single mutein. Obviously, a particular substitution at one location rules out replacement with another amino acid at that same location in that particular mutein.

The isomerases herein may be obtained as salts. Accordingly, the present invention includes electrically neutral and salt forms of the designated xylose isomerases and the term xylose isomerase refers to the organic structural backbone regardless of ionization state.

The muteins are particularly useful for the food processing industry. The xylose isomerases may be produced by fermentation as described herein and recovered by suitable techniques. (Anstrup, Industrial Aspects of Biochemistry, ed. B. Spencer, pp 23-46 (1974)).

The following examples are presented to illustrate the present invention and to assist one of ordinary skill in making and using the same. The examples are not intended in any way to otherwise limit the scope of the disclosure or the protection granted by Letters Patent hereon.

The method of the invention for predicting amino acid substitutions to reduce the entropy of unfolding was used to alter bacteriophage T4 lysoszyme ("T4L"). Two different types of amino EXAMPLE I / / l 225791 acid substitutions were performed; the first, Glycine to Alanine and the second, Xaa to Proline where Xaa is any amino acid.

There are 11 glycines in T4 lysozyme, of which three (Gly77, Glyno and Glyn3) have $ and conformational angles that are within the allowed range for amino acids with a B-carbon. Inspection of these glycine sites using the graphics display program FRODO, supra. suggested that glycine residues at positions 77 and 113 could potentially accomodate a B-carbon without sterically interfering with neighboring atoms. Gly77 was replaced with alanine as described below. Alanine was chosen to avoid possible secondary effects that might occur with a larger side chain. The mutant protein with a Gly77 to Ala77 substitution was designated G77A.

For the Xaa to Pro substitution, potential sites were screened by inspecting the <J> and i|> conformational angles using the criteria described above. Of the 164 amino acids in T4L, 17 residues met the above criteria. Two of these were already proline residues in the native lysozyme. Inspection with FRODO was used to eliminate sites where a proline side chain would sterically interfere with neighboring atoms. Sites where the side chain of a residue appeared to participate in intramolecular interactions within the native structure were also removed from consideration. This screening procedure left three preferred candidates for proline substitutions: LysgOr Ala82/ and Ala93. The mutant Ala82 to Pro (A82P) was constructed and is described here.

Mutant lysozymes G77A and A82P were obtained by oligonucleotide-directed mutagenesis (Zoller et al., DNA 3:479-488 (1984)). Procedures for mutagenesis, cloning, DNA sequencing, and protein purification were as described elsewhere (Alber et al., jjn Protein Structure, Folding and Design, UCLA Symp. Liss, NY, pp. 307-318 (1986); Alber et al., Methods Enzvmol. 154:511-533 (1987); Grutter et al., J.Mol.Biol.. 43 n. 225798 197:315-329 (1987) and Owen et al., J.Mol.Biol. 165:229-248 (1983), all of which are incorporated by reference herein).

Expression System The primer used to replace alanine with proline at position 82 was 5'-GTTTTAATTTAGGATTTCTC-3'. For position 77 a degenerate oligonucleotide, 5'-CTCAGAATTGNGCGAACAG-3' where N indicates a mixture of C, T and A, was used. This oligonucleotide codes for alanine, serine and threonine replacements of Gly77. Mutants were identified by differential hybridization of plaque lifts and dot blots to the mutagenic primers (Zoller and Smith, supra; and Alber and Matthews, supra). A typical reaction mix contains 0.1 pmole M13 template, 2 pmoles kinased mutagenic oligonucleotide, 0.5 pmole M13 sequencing primer, 0.5X ligase buffer, 0.5X Klenow buffer, 0.5 mM dNTP's, 0.5 mM rATP, 3 U of T4 DNA ligase and 2 U of the large fragment (Klenow) of DNA polymerase. After incubation at 16.C for 8-16 hours, this mixture is used to transform E. coli JM101 in broth culture. This step separates heteroduplex heterozygotes. The number of independent transformants is estimated by plating aliquots of transformed cells. The frequency of mutagenesis is 2-20%. Mutant sequences were then confirmed by DNA sequencing. Only alanine replacements were obtained at position 77 from the degenerate primer. To isolate the proteins, the mutant lysozyme genes were subcloned into an expression system including plasmid pHSe5 provided by Dr. David C. Muchmore (Institute of Molecular Biology, University of Oregon, Eugene, OR). In this system, the lysozyme gene is flanked by tandem lac and tac promoters and the trp terminator. Tight control of expression is provided by the presence of the lacl01 gene on the plasmid. The trp terminator eliminates selection against cells harboring the expression plasmid. The lysozyme was purified from lysates of induced plasmid-bearing cells by column chromatography on CM sepharose (Griffey et al., Biochemistry 24:817-822 (1985); and Alber and Matthews, supra). Cells were grown to 9 x 108/ml with 225738 vigorous aeration. Induction of lysozyme expression was initiated by addition of isopropyl 0-D-thiogalactoside (Calbiochem) to 8 x 10-4 M. The cells were incubated with reduced aeration and stirred for 2 h and then collected by centrifugation at 4000 rpm for 5 min in a Beckman J21 centrifuge using a JA10 rotor. The cells were resuspended in 20 mL of 50 mM tris(hydroxy-methyUaminomethane (Tris) and 1 mM 0-mercaptoethanol buffer, pH 7.4, containing 1 mM CaCl2 and 10 mM MgCl2. To promote cell lysis, EDTA was added to give a concentration of 10 mM and the container was shaken gently. After a few hours, the cells were lysed. DNase was added to a final concentration of 1 yg/mL, and the cellular debris was removed by centrifugation at 12,000 rpm for 20 min. The supernatant was decanted and dialyzed against 50 mM Tris, 1 mM 0-mercaptoethanol, and 1 mM ethyl-enediaminetetraacetic acid (EDTA) buffer, pH 7.4, overnight at 4°C. This material was loaded on a 2 x 30 cm column of CM-Sepharose equilibrated with the same buffer. The supernatant from the first centrifugation (4000 rpm for 5 min) was dialyzed against H2O and then also added to the column to recover lysozyme lost during incubation due to cell lysis. The column was washed with 6 L of the buffer to remove unwanted proteins and was then eluted with a linear gradient from 0 to 0.3 M NaCl in the buffer. The protein elutes as a single peak. The protein-containing fractions were pooled, dialyzed, and loaded onto another 1-mL column of CM-Sepharose for concentrating. This was eluted with 100 mM sodium phosphate and 0.5 M sodium chloride buffer, pH 5.6. Typical yield of purified T4 lysozyme was about 50 mgm.

Structures of Mutant Lvsozymes Crystals of G77A and A82P were obtained under conditions similar to those used for the native enzyme (Remington et al., J.Mol.Biol. 118:81-98 (1978); and Weaver et al., J.Mol.Biol. 193:189-199 (1987)), incorporated by reference. X-ray diffract tion data to 1.7-A resolution were collected by oscillation? f^T //Cv ft 225798 photography as described by Weaver et al., supra, and Schmid et al., Acta Crystalloqr. Sect. A 37:707-710 (1981), incorporated by reference herein. Each data set consisted of about 14,000 independent reflections with agreement between equivalent intensities of 6-7%.

In the map showing the difference in electron density between G77A and native T4L (Fig. IB), the strongest positive feature confirms the addition of the methyl group at residue 77. There are also strong positive and negative features indicating a shift in the nearby side chain of G1ujo8' Refinement of the G77A O structure (R = 15.7% at 1.7 A resolution) with the "TNT" package of programs (Tronrud et al., Acta Crystalloqr. Sect. A. 43:489-501 (1987) available from University of Oregon, Eugene, OR incorporated by reference herein, indicates that the carboxyl oxygen of Gluios closest to Alayy (Fig. 1A) moves 1.3 A relative to its position in native lysozyme (Fig. 1C). In addition, several backbone atoms in the vicinity of the substitution site move O 0.25-0.35 A. These shifts are also indicated in Fig. IB. Otherwise, the G77A structure appears to be essentially identical to the native protein. As judged by inspection of the three-dimensional structures of G77A and native lysozyme, the movement of Gluj.08 does not suggest any structural basis for differences in energy of the respective folded proteins. Although the car- O boxyl of Gluio8 nioves 1 A, it is not in close contact with the O a-carbon of Gly77 of native lysozyme (closest approach, 3.6 A) or with the B-carbon of Ala77 of G77A lysozyme (closest approach, O 3.9 A). The only apparent interactions of the carboxyl group of G1Y108 that contribute to the stability of native lysozyme are two hydrogen bonds, one from the side chain of Asngi (distance, O 2.5 A) and the other from a bound water molecule (distance, O 2.7 A) (Fig. 1A). Both of these hydrogen bonds are retained with O 0 respective distances 2.6 A and 2.7 A in the mutant structure (Fig. 1C). ■flatsw \ rs 225798 0 C 1 * Vr^4* In the case of mutant A82P, the difference map (Fig. 2A) shows the expected positive density corresponding to the addition of the pyrrolidine ring. An adjacent negative feature clearly indicates the displacement of a water molecule (W355) bound to the peptide nitrogen of Alag2 in wild-type lysozyme. The water molecule is bound to native lysozyme and presumably remains bound in the unfolded state. This bound water molecule should contribute approximately equally to the free energy of the two forms. Similarly, in the mutant protein the water molecule cannot bind to either the folded or the unfolded form of the protein and, again, should have no net effect. Refinement of the A82P structure to an R value of 15.8% at 1.7-A resolution shows that it is virtually identical with native lysozyme (Fig. 2B). The refined crystal structures of G77A and A82P provide experimental justification of the assumption made in obtaining Eq. 3, supra, that the backbone configuration of the native state is not changed by the mutations.

Thermal Stability The specific activity of both mutant lysozymes was measured at 22°C in the standard turbidity assay described by Tsugita et al., J.Mol.Chem. 243:391-397 (1968); and Tsugita, in The Enzymes, ed. Boyer, Academic Press, NY, 3rd Ed., Vol. 5, pp. 343-411 (1971)), incorporated by reference herein. The specific activity of both mutants was very close to that of native lysozyme. Stability toward irreversible inactivation was assessed by dissolving the proteins at 30 ul/ml in 100 mM KH2PC>4/100mM KCl/lmM EDTA, pH 6.5, and 300 ul aliquots were equilibrated at 65°C. After incubation at 65°C, samples were removed, plunged into ice and then diluted. The specific activity remaining as a function of time was measured (Perry et al., Science 226:555-557 (1980); Tsugita, 1971 and 1968, supra. incorporated by reference herein). When displayed as a first-order plot (Fig. 3A), (Perry et al., Science 226; 555-557 (1984); and Perry et al., Protein Eng.

C ft *25793 1:101-105 (1987), incorporated by reference herein), the loss of activity of wild-type losozyme appeared to be approximately biphasic. However, the thermal inactivation is better described as being second order in protein concentration (Fig. 3B). This result suggests that irreversible loss of activity is due to a bimolecular process such as intermolecular crosslinking. Whether considered as a first- or second-order process, the loss of activity of mutants G77A and A82P is significantly slower than that of native lysozyme.

Phage T4 lysozyme can be unfolded reversibly under controlled conditions. The transitions were monitored as the change in dichroism at 223 nm, as has been described (Becktel et al., Biopolvmers 26:619-623 (1987) incorporated by reference herein). In practice, great care was taken to ensure that the experimental measurements were reversible and reproducible. Fresh protein samples purified in the absence of chloroform were extensively dialyzed against oxygen-free buffers and reducing agent. Ionic strength was kept above 0.15 with KC1 or NaCl, and pH was adjusted with HC1 (pH 2-3), 10 mM acetate buffer (pH 4-5) or 10 mM phosphate buffer (pH 5.5-7). Protein concentration was kept below 30 ug/ml to avoid irreversible aggregation at high tempera-- ture.

Circular dichroism was monitored using a Jasco J-500C instrument equipped with a Hewlett-Packard 89100A thermoionic controller. The temperature of the sample was changed at a constant rate, typically 1°K per minute, under the control of a Hewlett-Packard 87 XM computer. The temperature and optical signal were digitized for subsequent analysis. Denaturation and renaturation were both monitored to ensure reversibility.

Thermal denaturation of T4 lysozyme was followed by measuring the optical properties of a solution of the purified protein as a function of temperature. Fluorescence emission and circular ■' VrTx n 225798 dichroism provide the most sensitive methods of optical measurement. When the protein unfolds, the optical spectrum (e.g. the molar ellipticity at 223 nm, which is sensitive to protein secondary structure) undergoes a cooperative change. Assuming that the spectrum at any temperature is a linear combination of the spectra of only the native and denatured states of the protein, the fraction of the material that is (un)folded can be calculated at each temperature. The equilibrium measurements of the wild-type and mutant lysozymes were consistent with the two-state assumption. The ratio of the fraction of the protein in the unfolded and folded conformations is the equilibrium constant (K) for the denaturation reaction.

At pH 6.5 for both replacements and at pH 2 for A82P the net free energy change introduced was positive and the mutant proteins were more stable than the native protein (Table I). The enhancement of thermal stability was due to a decrease in entropy rather than changes in enthalpy (Table I).

At pH 6.5 the observed increase in thermodynamic stability of each mutant structure relative to native lysozyme was 50-60% of that expected theoretically (Table I). Considering the simplifications that underlie the theory, the agreement was remarkably good. The theoretical estimate for the change in free energy assumes that this change is solely entropic. Compensation by changes in enthalpy would lessen the stabilization. In the case of A82P, the structure of the mutant lysozyme was virtually identical to native lysozyme (Fig. 2B), and the observed increase in stability can reasonably be attributed to the change in backbone configurational entropy. For Gil A, however, the substitution of alanine for glycine resulted in localized changes in the protein structure (Fig. 1C) that either offset or enhanced the entropic contribution to the free energy of unfolding. It should also be noted that the crystal structures were determined at pH 6.7 and were, therefore, relevant to the enhanced stability of '■or — a- u- 708 G77A at pH 6.5 but did not necessarily reflect the structure of lysozyme at pH 2.0 where GIuiqq was protonated and the mutant protein was less stable than the native protein. o n 225798 TABLE I Thermodynamic parameters for wild-type and mutant lysozymes1 AH AS, Protein Tm, °C AT, °C kcal/mol cal/deg-mol Denaturat ion at pH 2.0 Native 41 .9 ± 0. 4 - 89 ± 5 282 ± 16 G77A 40 .5 ± 0. 7 -1.4 ± 0.8 85 ± 5 270 ± 13 A82P 42 .7 ± 0. 7 0.8 ± 0.4 90 ± 5 283 = 16 Denaturation at pH 6.5 Native 64 .7 ± 0. - 129 ± 9 381 ± 27 G77A 65 .6 ± 0. 2 0.9 ± 0.5 125 ± 9 368 ± 27 A82P 66 .8 ± 0. 2 2.1 ± 0.5 126 ± 9 371 ± 26 Denaturat ion at pH 2.0 AAH, AAS, AAG, -TmSZ/y, Protein kcal/mol cal/deg-mol kcal/mol kcal/mol Wild type - - G77A -1 ± 6 -3 ± 21 -0.4 -0.8 A82P -1 ± 7 -2 ± 23 0.3 1.3 Denaturation at pH 6.5 Wild type - - G77A -6 ± 13 -18 ± 38 0.4 0.8 A82P -7 ± 13 -22 ± 37 0.8 1.4 -51- 1 < J30c ^1988 2257 1/ The thermodynamic parameters were derived from van't Hoff analyses of reversible thermal denaturations of the native and mutant proteins. Equilibrium constants were obtained from the fraction of native protein present under a given set of conditions of sequence, temperature, and pH. Tm is the temperature of denaturation, and AT is the difference in melting temperature. AH is the enthalpy of unfolding, and AAH is the difference in unfolding enthalpy of mutant and wild-type proteins measured at the melting temperature of the wild-type protein. AS is the entropy of unfolding, and AAS is the difference in unfolding entropy of mutant and wild-type proteins. The difference between the free energy of unfolding of mutant and wild-type proteins, AAG, is the observed free energy of stabilization and can be compared with -TmSZfy, which is the backbone entropic stabilization estimated from Eq. 3. The temperature variation of the enthalpy and entropy of denaturation for these T4 lysozymes, ACp, was determined to be 2.0 ± 0.2 kcal/deg-mol.

EXAMPLE II Purification of Xylose Isomerase (XI) from Streptomvces rubiqinosus Strain C3 XI was isolated from Streptomvces rubiqinosis strain C3 obtained from CETUS Corporation, Emeryville, CA using the following method.

The purification scheme for crude extract involves the following steps: filtering the extract to remove insoluble material; precipitating enzyme with alkyldimethylbenzyl ammonium chloride (BTC) to remove impurities not precipitated with the BTC; further filtration with salt; removal of BTC by adsorption resin; desalting and concentration by ultrafiltration to remove low molecular weight impurities; adsorption of the isomerase enzyme on a GDC (granular DEAE cellulose) column to remove unadsorbed impurities; washing and eluting column with sodium r% 225738 chloride (NaCl) to solubilize the isomerase; and gel filtration. Ultrafiltration is used for enzyme desalting and concentration between the various steps and in some cases the preparation may be sufficiently pure for certain applications so that the gel filtration step may be eliminated.

Crude isomerase extract was prepared by fermentation of Streptomvces strain C3 which was derived from ATCC 21,175. At the completion of fermentation, i.e., when isomerase activity was at a maximum, the intracellular enzyme was extracted from the mycelia by treatment with lysozyme and cationic surfactant (BTC). The extraction procedure was rapid and efficient with complete isomerase solubilization occurring within 2-4 hours. After extraction, insoluble materials, consisting mostly of disrupted cell debris, were removed by precoat filtration. The resulting soluble extract had an isomerase potency of 35-50 international xylose isomerase units ("U")/ml. The specific activity (U/mg protein) of the crude extract was difficult to estimate because protein determination is limited by interference from various components of the extract. 2-3 U/mg was expected.

Any turbidity in the crude extract was removed by laboratory filtration through a precoat of filter aid.

The optimum concentration of BTC to be added to the extract was determined in a preliminary experiment. This was accomplished by taking several aliquots of the extract and adding various amounts of BTC. The resulting precipitates were removed by centrifugation and aliquots of the supernates taken for isomerase assay as described below. The lowest BTC concentration at which complete or nearly complete isomerase precipitation takes place was the optimum concentration for the larger scale precipitation with the crude extract. Generally a BTC concentration of 1000-2000 ppm should be sufficient for complete isomerase precip-itat ion. \ 1 L h n ■;r c i 2 tj t JO For BTC precipitation the pH of the extract was adjusted to pH 7.0 to 7.3 and the BTC solution (100 mg/ml) added slowly with vigorous stirring. After additions of BTC the suspension was stirred for 20-30 minutes. A small aliquot was taken and centri-fuged, and the supernates were assayed for isomerase activity to test for completeness of isomerase precipitation.

When precipitation was complete, filter aid (approximately 5g HyFlo SuperCel/liter of suspension) was admixed and the suspension filtered using Whatman 3 paper and a laboratory vacuum. The resulting filter cake was washed with several portions of water to remove entrained solubles.

To solublize the precipitated BTC-isomerase complex, the filter cake was suspended in a minimum volume of 0.5 N NaCl, pH 7.0 (100-200 ml per liter of original extract) and stirred for thirty minutes. The suspension was then filtered using a vacuum and the filter cake washed with several small portions of salt solutions without vacuum. The filtrate and washings were well mixed and samples taken for determination of protein, BTC, and isomerase activity.

Protein Concentration Protein concentration was determined by measuring ultraviolet absorbance at 280 nm. An Extinction Coefficient of 15.4 (1 mg/ml = 1.54 A280K determined based on amino acid composition of the protein was used to convert absorbance to isomerase protein concentration. Samples for protein determination were diluted to an A28O of 0.2-1.0. Turbid samples should be filtered or centrifuged before dilution. Absorbance was measured in 1 cm quartz cuvettes using a suitable blank. The absorbance was scanned from approximately 320 nm to approximately 240 nm. Purified isomerase has a characteristic absorbance spectrum with a distinctive maximum at 278 nm and a minimum at 250 nm. An 54- 225798 a280/a260 ratio of approximately 1.6 was typical of highly purified isomerase preparations.

BTC Estimation Soluble BTC-835 (alkyldimethylbenzyl ammonium chloride, Onyx Chemical Co.) concentration was estimated by measuring ultraviolet absorbance over the 290-240 nm range. BTC has three distinct absorbance peaks at 269, 262 and 256 nm with very little absorbance at 280 nm. To estimate BTC concentration the absorbance at 262 nm was measured and corrected for protein absorbance at this wavelength. An extinction coefficient of 9.5 was used to convert absorbance to concentration (1 mg/ml = 0.95 A262>• Ultraf iltrat ion Depending upon the volume of sample to be handled, ultrafiltration was accomplished with either an Amicon CH4 hollow fiber concentrator or with Amicon 401 or 201 stirred cells using the appropriate Amicon cartridges or membranes. Monitoring for enzyme retention was accomplished by periodic UV scan of the ultrafiltrates. Where enzyme leakage was indicated by UV absorbance, samples were checked by the isomerase described below.

Most of the procedures described were carried out at room temperature. To minimize chances of microbial contamination enzyme solutions were filtered through 0.45 or 0.2 Millipore membranes, and were stored in the cold between purification steps.

BTC removal was effected by treatment with a strong acid cation exchange resin in the sodium form. Resins such as AG-50 (BioRad Laboratories, Richmond, CA) will adsorb BTC in the presence of 0.5N NaCl without affecting isomerase. Other procedures for removal of BTC, include treatment with activated carbon or ultraf iltration-diaf iltration. k'C 225798 The AG-50 resin was added directly to the BTC-isomerase (approximately 1 g dry base resin per 100 ml) solution and the pH was adjusted to 7.0 after a brief period of gentle stirring. The suspension was stirred gently for about 20 minutes and the pH readjusted to 7 when necessary. The resin was allowed to settle by gravity and a sample of the supernatant was taken for UV scan to test for BTC removal. If BTC removal was not complete, additional resin was added until no BTC remained. An additional test for BTC removal can be carried out by diluting a portion of the resin supernate 1 to 5 with water. The presence of residual BTC will be indicated by the formation of a precipitate due to the insolubility of the BTC-isomerase at lower salt concentration.

After removal of BTC, the resin was removed by filtration, and the filtrate desalted and concentrated by ultrafiltration with an Amicon CH4 hollow fiber cartridge. The starting solution for ultrafiltration was optimally free of any insoluble material, and was filtered through a 0.45 micron filter when necessary. Ultrafiltration was carried out until the retentate volume was reduced to a minimum consistent with reasonable flux rate. The retentate was then diluted with 5-10 volumes of water and the pH readjusted to 7. Ultrafiltration was continued. This dilution-diafiltration was repeated two more times. The final retentate had a specific activity of 30-35 U/ml. Recovery of activity based on the starting crude extract was 85-90%.

To prepare the enzyme for GDC adsorption sufficient 1M Tris buffer, pH 7.0, and 1M MgSO was added so that the concentration of each was lOmM in the enzyme solution. Microbial contamination, when the enzyme was to be stored for any period of time, was reduced by filtration through a 0.45 micron Millipore filter.

GDC adsorption-desorption was carried out with a column of granular DEAE-cellulose (Whatman Ltd., Clifton, N.J.). To prepare the column, 300 g GDC was equilibrated in lOmM Tris buffer.

^ O pr-'i 225798 This suspension was poured into a 5 cm x 20 cm chromatography column to form a uniform bed. The column was then washed using two liters of lOmM Tris at a flow rate of approximately 10 ml/min. Washing with buffer was continued until the effluent pH was between 6.8 and 7.2.

The enzyme solution (ultrafilter retentate) was applied directly to the column at a flow rate of approximately 5 ml/min. A total of 2x10 5-3 x 105 U of enzyme can be applied. During enzyme application, and in subsequent washing and elution steps, the effluent from the column was monitored for UV absorbance and periodic samples were assayed using the isomerase assay. After the enzyme was applied, the column was washed with 3-4 liters of 0.15N NaCl at a flow rate of approximately 20 ml/min. This washing step removed weakly adsorbed impurities, as evidenced by the yellow color and UV absorbance of the effluent. Near the end of the washing step the effluent was nearly colorless and contained very little UV absorbing material.

Elution of the isomerase was accomplished by washing the column with 0.45N NaCl in lOmM Tris, pH 7, at a flow rate of 10 ml/min. The effluent from the elution step was collected in 200 ml fractions which were monitored for UV absorbance and isomerase activity. Isomerase activity began to elute immediately after a void volume of 800-900 ml of eluate was collected. More than 90% of the total activity eluted in the first five 200 ml fractions of eluate. Fractions with isomerase activity of 20 U/ml and specific activity of 40 U/mg were pooled for desalting and concentration by ultrafiltration.

The pooled GDC column fractions were desalted and concentrated using either the CH4 concentrator or the 401 stirred cell (30,000 molecular weight cutoff). The CH4 unit was used to reduce the volume to 200 ml, and two or three 5 volume diafiltra-tions were conducted with water to remove salt. The CH4 } 225793 retentate was then further concentrated with the stirred ultrafiltration cell. If the enzyme was to be further purified by gel filtration, diafiltration with 20mM phosphate buffer, pH 7.0, was used to adjust the buffer concentration.

Recovery of activity from the GDC step was greater than 90% of the activity applied to the column, yielding an overall recovery of about 80% based on the starting extract. The specific activity was 40-45 U/mg, indicating that the enzyme was 90-95% pure on a protein basis.

Gel filtration was carried out with a column of Fractogel TSK HW-55 (Pierce Chemical Co. , Rockford, IL) capable of separating proteins in the 50,000-500,000 MW range. The gel was equilibrated in 20mM sodium phosphate buffer according to the manufacturer's recommendation, and used to prepare a 2.5 cm x 90 cm uniformly packed bed. The column was then equilibrated with phosphate buffer at a flow rate of 0.6-0.7 ml/min for at least 16 hours before use.

Total sample volume applied to the column was less than 20 ml, with smaller volumes being more desirable. The sample was applied carefully to the top of the column without disturbing the gel bed, and allowed to flow into the column by gravity. Application of the sample was followed by two 1 ml buffer applications to assure that the sample was completely washed into the bed. The column was then eluted with 20mM phosphate buffer at a flow rate of 0.6-0.7 ml/min. The column effluent was continuously monitored for absorbance at 280 nm, and fractions (10 ml each) were collected automatically. The fractions were analyzed for protein (A28O) and isomerase activity as described elsewhere.

Typically, for the isomerase purified by BTC precipitation followed by DGC adsorption-desorption, enzyme elution was preceded by elution of a small peak of U.V.-absorbing material which n 225798 probably represents some higher molecular weight protein contaminant. The specific activities of the first and last fractions from isomerase elution were generally lower than those of the middle fractions. These first and last fractions were discarded, since enzyme purity was considered to be more important than recovery. The pooled active fractions were ultrafiltered with a 100,000 molecular weight cutoff membrane using the 201 stirred cell. Very little u.v. absorbing material was found in the ultrafiltrate from this step, indicating that the enzyme was free of lower molecular weight protein contaminants. The final diafiltered retentate was filtered through a 0.2 Millipore filter to eliminate microbial contamination during storage.

The final specific activity was 46 u/mg with an overall recovery of about 70% based on starting extract.

The results of each step as set forth above were tabulated in terms of total activity, specific activity, and recovery as shown in Table II. ? m fc /v ^ n 225798 Table II ISOMERASE PURIFICATION Total Specific Recovery Procedure Frac- Vol. Potency Activity Protein Activity % of tion (ml) (U/ml) (U) (mg/ml) (U/mg) Start BTC-Precipitation I 3500 103 3.59xl05 89.2 UF YM 30 II 2000 179 3.48xl05 .70 31.2 86.4 GDC Column (2 batches) III 2400 137 3.29x10s 3.21 42.7 81.6 UF YM 30 IV 200 1565 3.13xlOS .9 43.6 77.6 Gel Filtration1 V 90 252 2.27x10* .5 45.9 70.51 UF XM-100 VI 904 2.26x10* 19.7 46.0 70.21 ^2.5x10* U of Fraction IV was used for gel filtration step.

Alternatively XI protein is isolated from Streptomvces rubiqinosus strain C3 derived from S^ rubiqinosus ATCC 21,175 using the method described in U.S. Patent No. 4,410,627, which is incorporated herein by reference. The strain is grown by submerged aerobic fermentation on a medium with the following composition (by percent weight) dextrose 9.0%, corn steep liquor (solids) 0.06%, diammonium phosphate 0.008%, magnesium sulfate 0.06%, antifoam (pluronic PL-61) 0.003%. The medium is sterilized at 121°C for 45 min, cooled and adjusted to pH 6.8-7.0. The medium is inoculated with 14% (v/v) of an inoculm comprising the contents of a seed fermenter prepared with the S. rubiqinosus strain. XI protein is extracted from the S. rubiqinosus strain by adding 0.35% Maquat MC 1412 (Mason Chemical Co.) and 10 ppm of hen's egg lysozyme and agitating for 5 hr at 40°C., pH 6.3 to 6.6. The mixture is then . ,3ocwii 225798 filtered to provide a solution of unpurified xylose isomerase. The crude isomerase is purified by adsorption on DEAE-cellulose, filtering and washing the adsorbed product with 0.1M NaCl solution to remove impurities, and then desorbing by contacting with 0.45M NaCl solution. The pH of all solutions is maintained at 7.5 during the purification step. The solution of partially purified isomerase obtained thereby is mixed with 3 volumes of 95% ethanol at 0°C to precipitate the isomerase. Perlite filter aid is added, the solids recovered by filtration, and air dried to provide a soluble XI preparation containing 2500 U/g. Specific activity of the preparation thus prepared is 40 U/mg of protein.

Purification following these procedures results in an enzyme having greater than 90% purity based on SDS-PAGE electrophoresis.

Xylose Isomerase Assay Xylose isomerase activity was measured by incubating the protein sample with a buffered solution of glucose for a fixed period of time, quenching the reaction, and then quantitating the amount of product (fructose) made by high performance liquid chromatography (HPLC) analysis. 1 unit of activity is that amount of enzyme that produces 1 pmole fructose/min under the defined reaction conditions.

A 20 ul sample of enzyme (0-3 units of activity) was mixed with 1 ml of substrate mixture (3 M in D-glucose, 25 mM maleic acid (adjusted to pH 6.5 at 60°C with NaOH), 10 mM MgS04 and 1 mM C0CI2) (previously equilibrated at 60°C) to initiate the reaction: The enzyme plus substrate mixture was incubated for 20 minutes at 60°C in a closed tube. At the end of this a 225798 incubation, 0.5 ml of IN HC1 was added to stop the reaction. Precipitated protein was removed by centrifugation, and an aliquot of the supernatant solution was removed for quantitation of fructose by HPLC analysis.

The separation of fructose from unreacted glucose was accomplished using a Beckman liquid chromatograph equipped with a Waters Assoc. (Waters Assoc., Milford, MA), WISP 710B autoinjector, Waters Assoc. differential refTactometer (Model R401) and a Shimadzu C-R3A integrator (Shimadzu Corp., Kyoto, Japan). Carbohydrates were separated using an Applied Science carbohydrate analysis column (amine phase, 250 mm X 4.6 mm) using isocratic sample elution with an acetronitrile/water (80%/20%) solvent flowing at 1.3 ml/min. Integration of peak areas for the resolved fructose peaks from standard fructose solutions or from test samples, allowed quantitation of fructose production for the test samples during the 20 minute incubation.

EXAMPLE III N-terminal Sequencing of S. rubiqinosus Xylose Isomerase o Purified XI is subjected to further analysis to determine the amino ("NH2n)-terminal end of the mature protein.

Edman degradation determination of XI amino acid sequence Sequence analysis by automated Edman degradation was performed using a Beckman Model 890C sequencer (Beckman Instruments, Palo Alto, CA) following standard methodology. In some instances, in order to reduce background and improve signal to noise ratio, orthopthalaldehyde was used to block non-proline residues according to the method of Bauer et al., Anal. Biochem. 137:134 (1984), incorporated by reference herein. t 3 Qr-y m 225798 j | Results for the first 20 residues of the amino-terminal sequence obtained for XI are shown in Table III.

G l. /y ^ -63- i tZOCTI98s; !l 7 aicaiss 225798 Table III C-3 Xylose Isomerase Amino Acid/Yield (nmol) Major Sequence Minor Sequence 1 Met .47 — 2 Asn .44 Met 0.84 3 Tyr .21 Asn 2.73 4 Gin 11.23 Tyr 1.70 Pro 6.78 Gin 4.07 6 Thr 2.11 Pro 1.78 7 Pro .88 Thr 0.77 8 Glu 9.79 Pro 2.56 9 Asp 7.83 Glu 3.74 Arg .37 Asp 3.43 11 Phe 9.26 Arg 3.23 12 Thr 2.14 Phe 4.39 13 Phe 8.68 Thr 0.91 14 Gly 7.03 Phe 4.45 Leu .09 Gly 3.45 16 Trp 4.25 Leu 6.25 17 Thr .67 Trp 1. 37 18 Val 8.52 Thr 1.32 19 Gly .87 Val 4.76 Trp .01 Gly 3.96 EXAMPLE IV Construction of oligodeoxynucleotide probes for detection of the N-terminal region of XI from strain C3 Oligodeoxynucleotide probes were made using conventional methods. Using polynucleotide kinase, the probes were labeled with [32yP]-ATP having a specific activity of 3000 Ci/mole, supplied by New England Nuclear Labs (Boston, MA). The labeled probes were purified by gel filtration on a Biogel P-4 gel (BioRad Laboratories, Richmond, CA). Two pools of four probes were made. Pool 1 consisted of probes having the following sequences GGTTG(A/G)TA(A/G)TTCAT and pool 2 consisted of probes having the following sequences GGCTG(A/G)TA(A/G)TTCAT, wherein the nucleotides in the n Cycle No. j * •>25798 parentheses are alternate nucleotide bases. The two pools were constructed to cover all possible nucleotide ambiguities in the XI gene in the region coding for the NH2"terminal region.

EXAMPLE V Cloning of the XI gene of strain C3 in Plasmid pBR322 Plasmid preparation Plasmid pBR322 DNA was isolated and purified essentially as described in Birnboim et al., Nuc. Acids Res. 7:1513(1979), incorporated by reference herein. After purification of the plasmid in CsCl, the DNA preparation was further digested with RNase at a concentration of 40 Mg/ml at 37°C for 30 minutes and subsequently extracted with phenol and ether. The RNA-free plasmid DNA was then completely digested with Bam HI and dephosphorylated with calf intestinal alkaline phosphatase.

Preparation of S. rubiqinosus Strain C3 DNA High molecular weight chromosomal DNA for S. rubiqinosus strain C3, a derivative of ATCC 21,175, was isolated according to the methods of Chater et al., Current Topics in Microbiology and Immunology, 96:69 (1982), incorporated by reference herein. The DNA was then partially digested with restriction enzyme Sau 3A1 (New England Biolabs, Beverly, MA) under the conditions suggested by the manufacturer. The 4 to 8 kb fragments from the partially digested chromosomal DNA were isolated by sucrose density gradient centrifugation and were concentrated by DEAE ion exchange chromotography. o ' 3 OCT, 1988, c 225798 Ligation of S. rubiqinosus DNA into OBR322 to form a gene bank and Transformation of E. coli with the resulting vector Two hundred ug of the Bam HI digested cloning vector (pBR322) were mixed at a 1:2 molar ratio with the partially digested S. rubiqinosus DNA in ligation buffer under sticky end conditions. After ligation, an aliquot of the reaction mixture containing approximately 100 ug of the cloning vector was used to transform CaCl2~treated competent E. coli strain MM294. The transformed E. coli were diluted ten-fold with 2 x L-broth by volume and were incubated for 90 min at 37°C. The culture was then further diluted 25-fold with 2 x L-broth containing 100 ug/ml ampicillin. The dilute culture was then incubated at 37°C with shaking, overnight. After incubation, the concentration of glycerol in the culture was adjusted to 15% and the mixture was stored at 70°C.

Identification of xylose isomerase clones The transformant gene bank prepared as described above was thawed and plated on L-agar plates containing 40 ug/ml of ampicillin to obtain approximately 400 individual colonies per plate. Colonies were then transferred to nitrocellulose filters as described in Maniatis et al., Molecular Cloning, supra. Filters were prehybridized by the method described in Woo, Methods in Enzymoloqy 68:389 (1979), incorporated by reference herein. Processed filters were then hybridized with [32-yP]-labeled oligonucleotide pool 2(106 cpm/filter) in hybridization buffer (5XSSC, 5X Denhardt's solution, 50 mM sodium phosphate pH 7.0, 100 ug/ml sheared calf thymus DNA, and 1% SDS) at 35°C overnight. Filters were subsequently washed with 5XSSC, 2XSSC, 2XSSC and IXSSC containing 0.1% SDS at hybridization temperature (35°C) for 15 min. each. * 3 or-.- 225798 Fifty percent of the transformants were both ampicillin and tetracyline resistant indicating that about half of the transformants carried inserted S. rubiqinosus DNA. To confirm this conclusion, ten ampicillin resistant transformants were randomly picked, the plasmid DNA extracted and purified, and restriction enzyme analysis with Eco RI was carried out. Agarose gel electophoresis of the digested plasmid DNA showed that 50% of the DNA was about 4 to 8 kb larger than pBR322. Based on these results, approximately 20,000 transformants were obtained using 20 ug of the cloning vector. Based on the reported size of the Streptomvces genome, a complete gene library of S. rubiqinosus was obtained. Of 20,000 colonies screened, 15 colonies hybridized to the mixture of oligonucleotides in pool 2.

The plasmids of each positive colony were isolated as described above and characterized by restriction enzyme fragment analysis using Pst I, Bgl III and Sma I. Three types of clones were distinguished. Two of the representative clones, pTWl and pTW2, carried 4.3 and 7.5 kb Sau 3A1 inserts, respectively. The third representative clone, pTW3, carried a 12 kb insert which was believed to arise by linkage of two Sau 3A1 fragments of the S. rubiqinosus digest.

Of the pool 2 primers, the oligodeoxyribonucleotide designated CS26 having the sequence 5'-GGCTGGTAGTTCAT-3', was found to hybridize strongly with the S. rubiqinosus C3 DNA, and in particular, hybridized 10 times more strongly with the transformant designated pTWl which carried a 4.3 kb insert. An oligonucleotide complementary to C26, designated HWG3, was constructed for further use.

Plasmids pTWl, pTW2 and pTW3 were analysed with a number of restriction enzymes. The 1.35 kb Sal I, 2.3 kb Pst I and 1.8 kb Sma I restriction fragments from the plasmid inserts 225798 hybridized with probe CS26. S. rubiqinosus genomic DNA was digested with the same restriction endonucleases, and fragments of the proper molecular weight hybridized under stringent conditions with CS26, confirming the location of the translation start site of the gene. To determine the transla-tional and transcriptional orientation of the gene, the Nru I-Pst I restriction fragment carrying the 5' end of the gene was further subcloned into the Sma I-Pst I sites of either M13mpl0 or M13mpll replicative form (RF) DNA (obtained from Bethesda Research Laboratories, Bethesda, md). Single-stranded DNA was isolated, purified as described in Messing, Methods in Enzymoloqy, 101:20 (1983), incorporated by reference herein, and hybridized to probes CS26 and HW03. The results shown in Table III indicated that the transcriptional direction of the gene is from left to right in the restriction map of the gene shown in Figure 4. n ^ 225798 Table IV Hybridization of Oligodeoxyribonucleotide probes CS26 and HW03 with the Single-Stranded Recombinant Phage DNAs carrying the 5'-end of the Glucose Isomerase Gene in Two Different Orientations Insert Cloning Hybridizat ion^ SS Phaae DNA Orientation1 Sites CS26 HW03 M13mpl0 (5') -Smal-PstI + M13mpll + (5') -PstI-Smal - + *(+) orientation indicates that the sense-strand of the insert is in the phage, (-) indicates that the antisense-strand is in the phage. 2 one yg of single-stranded phage DNA was used for each hybridization.

EXAMPLE VI Sequencing of the xylose isomerase gene The DNA sequence of the entire xylose isomerase gene was determined based upon the restriction map of Figure 1 and the determination of the transcription orientation. The complete DNA sequence and deduced amino acid sequence is shown in Figure 5. Comparison of the entire sequence for S. rubiqinosus XI with published sequences for other known native procaryotic isomerases (Figure 6) reveals substantial sequence identity between the XI of these organisms. rv 22571 For the xylose isomerases with limited amino acid sequence identity to S. rubiqinosus xylose isomerase, (i.e. less than 50%), for example enzyme obtained from E. coli and B. subtil is. the conserved amino acid residues are primarily those located at or near the active site of the enzyme. It is contemplated that the below-described alterations in amino acids at the active site of S. rubiqinosus XI are likely to produce muteins of these enzymes with similar characteristics to the XI muteins resulting from similar changes in S. rubiqinosus XI. In enzymes that are more closely homologous to S. rubiqinosus XI, such as Ampullariella sp., similar alterations in amino acid residues located in other regions of the protein, as well as in the active site, are likely to result in comparable changes in the stability and activity of these enzymes.

EXAMPLE VII Expression of Xylose Isomerase Muteins Construction of the Expression Vector Construction of an expression vector plasmid pTWll for expression of the XI muteins in E. coli was as follows. The 1.4 kb Nru I-Sma I restriction fragment carrying the entire coding sequence of the glucose isomerase gene from pTWl was isolated and subcloned into the Sma I site of M13mplQ RF DNA. The orientation was such that the ATG initiation codon of the gene was approximately 220 bp from the Eco RI site of the phage, which was designated phage <J>TW23. Ligation of the XI DNA sequence into the phage destroyed the translation termination codon of the gene, and a new one was created by site-specific mutagenesis as described in Zoller and Smith, Methods in Enzvmoloqy 100:468 (1983), incorporated by reference herein, using a synthetic oligodeoxyribonucleotide with the sequence 5'-CGACTCTAGATCATCCCCGGGCG-3'. The new phage having 225798 the desired insert was screened by hybridization with the mutagenesis primer labeled using polynucleotide kinase and [32yP]ATP (3000 Ci/mmole, New England Nuclear) as follows: prehybridization was carried out by the procedure of Woo, Methods in Enzymology 68:389 (1979), incorporated by reference herein. Processed filters were then hybridized with the [32yP] — labeled oligodeoxyribonucleotide (106 cpm/filter) in 20 ml hybridization buffer (5XSSC, 5X Denhardt's solution, 50mM sodium phosphate, pH 7.0, 100 ug/ml sheared calf thymus DNA, 1.0% SDS) at 68°C, overnight. Filters were subsequently washed (15 min each) with 5XSSC, 2XSSC and IXSSC containing 0.1% SDS at hybridization temperature (68°C). The phage having the proper sequence, designated phage TW31, was confirmed by sequencing.

The same procedure was used to insert a Hind III site preceding the translation initiation codon of the xylose isomerase gene in phage <J>TW31 except that the synthetic oligodeoxyribonucleotide had the sequence 5'-GTACTTCATAACTCTTCGCGGCTC-3' and the hybridization and washes were carried out at 65°C. The phage carrying the xylose isomerase gene with the introduced translation termination codon and two Hind III sites bordering the gene was designated phage (|>TW32. The xylose isomerase gene was isolated from phage $TW32 by digestion with Hind III and was ligated into the Hind III site of E. coli expression vector pTRP3 (ATCC No. 39,946 deposited Dec. 18, 1984, described by Goeddel et al. Nuc. Acids. Res. 8:4052 (1980)), incorporated by reference herein. The resulting plasmid having the expression of xylose isomerase under the control of the trp operon promotor and trpE translation initiation signal, was designated pTWll. The plasmid having the xylose isomerase gene in the opposite orientation was designated pTW12. 225798 Preparation of E. coli Strain Lacking XI for Screening The E. coli strain DG101 (thi-1, endAl, hsdR17, SupE44, lacl8, lacZM15) was mutagenized using Nitrosoguanidine (NTG) at a concentration of approximately 200 ug/ml of medium for approximately 30 minutes. The bacteria were pelleted, washed in minimum salts medium, resuspended in minimal salts medium containing 0.5% xylose, and were grown for approximately 30 minutes at 37°C. D-cycloserine was added to a concentration of 100 ug/ml and the culture was incubated at 37°C for approximately 30 minutes. The cells were centrifuged, washed in minimal salts medium, and then grown in rich L-broth for approximately 30 minutes at 37°C. The culture was plated on McConkey medium containing 1% xylose, and white colonies were selected. White colonies were transformed with plasmid pTWll plated on MacConkey medium without xylose. Red colonies, in which the absence of xylose isomerase was complemented by the plasmid, were picked. E. coli strain DG101 xyl- transformed with pTWll were deposited into applicants' depository under accession number CMCC 2210. This strain was deposited in the ATCC on August 5, 1987 under accession number 67,489.

Recovery of XI Muteins The transformed E. coli are cultured in LB media with limiting concentrations of tryptophan. After the mutein is induced enzymatically active xylose isomerase mutein is recovered essentially as described in Example I and in U.S. Patent No. 4,410,627, incorporated by reference herein. n n n 225798 EXAMPLE VIII Site-Specific Mutagenesis of the S. rubiqinosus xylose isomerase gene to Minimize Inactivating Reactions The specific locations for alteration of the DNA sequence encoding XI were selected based on computer-assisted analyses (PS 340, Evans and Sutherland, Salt Lake City, Utah, using MOGLI and Proteus software) of the x-ray crystal structure of xylose isomerase obtained by standard x-ray crystallography methodology.

Oligonucleotide primers, as described below, are synthesized complementary to the DNA sequence of the reference xylose isomerase gene fragment, except for regions of limited nucleotide mismatching to accomplish the desired mutation. Gapped circle site-specific mutagenesis as described by Kramer et al., supra, is used to convert the amino acid at the selected position to a different amino acid. Towards this end, plasmid pTWll and phage ml3mpl0 carrying amber mutations are digested completely with Eco RI and Bam HI. The small fragment of pTWll and the large fragment of M13mpl0 are isolated and ligated together. The phage having the small Eco RI- Bam HI fragment from pTWll and large fragment of M13mpl0 is designated TVW8. To form the gapped circle DNA for use as a template for the oligonucleotide-directed mutagenesis single-stranded TVW8 DNA is mixed with Eco RI-Bam HI digested M13mpl9 RF DNA and the two DNAs are melted together at 100°C and reannealed at 67°C for 30 minutes. A "gapped circle" in which the DNA sequence to be mutagenized remains single-stranded and the remaining DNA is double-stranded is formed in which the single-stranded region includes the XI gene.

The oligonucleotide primers described below are hybridized to the gapped circle DNA (phage TVW8) under hybridization conditions, for example, in a mixture containing c • t 225798 lOOmM NaCl, 20mM Tris-HCl, pH 7.9, 20mM MgCl2 and 20mM B-mercaptoethanol by heating at 67°C for five minutes and 42°C for 25 minutes. Primer extension is carried out using DNA polymerase large fragments in the presence of dNTPs. The ends of the extended primer are ligated using T4 ligase and ATP. The reactions are terminated by heating to 80°C. The mixture is then used to transform competent E. coli strain HB 2151, which are plated onto agar plates and incubated overnight to obtain phage plaques, and grown under conditions suitable for inducing the phage. The plaques are probed using the same [32yP]-labeled primer using kinase at standard prehybrid-ization and hybridization conditions at high stringency (e.g. 42°C for 8 hours). Plaques which hybridize to probes are lifted and are confirmed by sequencing. The phage DNA containing the coding sequence for the mutagenized xylose isomerase gene are isolated. The DNA segment comprising the mutagenized XI gene is removed by Hind III digestion. The small Hind III fragment is isolated, purified, and ligated into plasmid pTRP3, previously digested with Hind III.

To produce XI genes containing more than one modification, successive rounds of mutagenesis, each using the appropriate primer, are carried out.

Mute in-Encoding Primer Sequences The following oligonucleotide primers are used for site-specific mutagenesis to obtain muteins of xylose isomerase resistant to chemical inactivation in E. coli; to convert Lys289 to Arg289 to obtain the Arg289XI mute i n, 5'-CGGTCCGCGGCGGGCGGAAGTCGAAGTGC-3; to convert Lys289 to Gln289 to obtain the Gln289XI mute in, 5'-CGGTCCGCGGCGGCTGGAAGTCGAAGTGC-3'; 225798 1 to convert Lys289 to Asn289 to obtain the Asn289XI mute in, 5'-CGGTCCGCGGCGGGTTGAAGTCGAAGTGC-3'; to convert Lys289 to Asp289 to obtain the Asp289xI mut e i n, 5'-CGGTCCGCGGCGGGTCGAAGTCGAAGTGC-3'' to convert Lys289 to G1U289 to obtain the GlU289XI mutein, 5'-CGGTCCGCGGCGGCTCGAAGTCGAAGTGC-3'; to convert Lys289 to Ser289 to obtain the Ser289xI mutein, 5'-CGGTCCGCGGCGGGGAGAAGTCGAAGTGC-3'; to convert Lys289 to Thr289 to obtain the Thr28gXI mut e i n, 5'-CGGTCCGCGGCGGGGTGAAGTCGAAGTGC-3'? to convert Lys289 to His289 to obtain the His289XI mute in, 5'-CGGTCCGCGGCGGGTGGAAGTCGAAGTGC-3 *; to convert Lys289 to Tyr289 to obtain the Tyr289xI mu t e i n, 5'-CGGTCCGCGGCGGGTAGAAGTCGAAGTGC-3'; to convert Lys289 to Ala289 to obtain the Ala289XI mutein, 5'-CGGTCCGCGGCGGGGCGAAGTCGAAGTGC-3'; to convert Lys289 to Val289 to obtain the Val289XI mu t e i n, 5'CGGTCCGCGGCGGGACGAAGTCGAAGTGC-3' ; to convert Lys289 to Leu289 to obtain the Leu289XI mutein, 5•-CGGTCCGCGGCGGGAGGAAGTCGAAGTGC-3'; to convert Lys289 to Ile289 to obtain the Ile289xI mutein, 5'-CGGTCCGCGGCGGGATGAAGTCGAAGTGC-3'; to convert Lysi83 to Argi83 to obtain the Argia3XI mutein, 5'-GCGGCTCGTTCGGGCGGGGCTCGATGGC-31; to convert Lysi83 to Glni83 to obtain the Glni83XI mutein, 5'-GCGGCTCGTTCGGCTGGGGCTCGATGGC-3'; -75 2257 to convert Lysi83 to Asni83 to obtain the Asni83Xl mutein, 51-GCGGCTCGTTCGGGTTGGGCTCGATGGC-3'; to convert Lysi83 to Aspi83 to obtain the Aspi83XI mutein, 5•-GCGGCTCGTTCGGGTCGGGCTCGATGGC-3'; to convert Lysx83 to Glui83 to obtain the Glui83XI mutein, 5'-GCGGCTCGTTCGGCTCGGGCTCGATGGC-3'; to convert Lysi83 to Seri83 to obtain the Seri83XI mutein, 5'-GCGGCTCGTTCGGGGAGGGCTCGATGGC-3'; to convert Lysi83 to Thri83 to obtain the Thri83XI mute i n, 5'-GCGGCTCGTTCGGGGTGGGCTCGATGGC-3'; to convert Lysi83 to Hisi83 to obtain the Hisi83XI mutein, 5'-GCGGCTCGTTCGGGTGGGGCTCGATGGC-3'; to convert Lysi83 to Tyr^83 to obtain the Tyri83XI mutein, 51-GCGGCTCGTTCGGGTAGGGCTCGATGGC-3'; to convert Lysi83 to Alai83 to obtain the Alai83XI mutein, 51-GCGGCTCGTTCGGGGCGGGCTCGATGGC-3'; to convert Lysi83 to Vali83 to obtain the Vali83XI mute in, 5'-GCGGCTCGTTCGGGACGGGCTCGATGGC-3'; to convert Lysi83 to Leui83 to obtain the Leui83XI mute in, 5'-GCGGCTCGTTCGGGAGGGGCTCGATGGC-3'; to convert Lysi83 to Ilei83 to obtain the Ilei83XI mute in, 5'-GCGGCTCGTTCGGGATGGGCTCGATGGC-3'; to convert Hiss4 to Gln54 to obtain the Gln54XI mutein, 5'-GAGGTCGTCGTCCTGGAACGTGACGCC-3'; to convert Hiss4 to GIU54 to obtain the GIU54XI mutein, 5'-GAGGTCGTCGTCCTCGAACGTGACGCC-3'; n 2257 to convert His54 to Asns4 to obtain the Asn54XI mut e i n, 5'-GAGGTCGTCGTCGTTGAACGTGACGCC-3'; to convert His54 to ASP54 to obtain the ASP54XI mutein, 5'-GAGGTCGTCGTCGTCGAACGTGACGCC-31; to convert His54 to Ser54 to obtain the Sers4XI mute in, 5'-GAGGTCGTCGTCGGAGAACGTGACGCC-3'; to convert His54 to Thrs4 to obtain the Thrs4XI mute in, 51-GAGGTCGTCGTCGGTGAACGTGACGCC-3'; to convert His54 to Ala54 to obtain the Ala54XI mutein, 5'-GAGGTCGTCGTCGGCGAACGTGACGCC-3'; to convert His54 to Val54 to obtain the Val54XI mutein, 5'-GAGGTCGTCGTCGACGAACGTGACGCC-3'; to convert His54 to Tyr54 to obtain the Tyr54XI mut e i n, 5'-GAGGTCGTCGTCGTAGAACGTGACGCC-3'; to convert His220 to G^-n220 to obtain the Gln220XI mute in, 5'GGCCATCTGCTCCTGGCCGACCTCGGG-3'; to convert His220 to GIU22O to obtain the Glu220XI mutein, 5'-GGCCATCTGCTCCTCGCCGACCTCGGG-3'; to convert His220 to Asn220 to obtain the Asn220XI mutein, 5'-GGCCATCTGCTCGTTGCCGACCTCGGG-3'; to convert His220 to AsP220 t0 obtain the Asp220XI mutein, 5'-GGCCATCTGCTCGTCGCCGACCTCGGG-3'; to convert His220 to Ser220 t0 obtain the Ser220XI mut e i n, 5'-GGCCATCTGCTCGGAGCCGACCTCGGG-3'; to convert His220 to 'I'hr220 to obtain the Thr220XI mut e i n, 51-GGCCATCTGCTCGGTGCCGACCTCGGG-3'; "'v : ^'v> n -77- v' o ♦•N 13°ct;?s3i n 2257 to convert His220 to Ala220 to obtain the Ala220XI mute in, 5'-GGCCATCTGCTCGGCGCCGACCTCGGG-3'; to convert His220 to Val220 t0 obtain the Val220XI mute in, 5'-GGCCATCTGCTCGACGCCGACCTCGGG-3'; to convert His220 to Tyr220 t0 obtain the Tyr220XI mutein, 5'-GGCCATCTGCTCGTAGCCGACCTCGGG-3'; to convert Met223 to G^Y223 to obtain the Gly223xI mu t e i n, 5'-GTTCAGCCCGGCCCCCTGCTCGTGGCC-3 *; to convert Met223 to Ala223 to obtain the Ala223xI mute in, 5'-GTTCAGCCCGGCCGCCTGCTCGTGGCC-3 *; to convert Met223 to Val223 to obtain the Val223xI mutein, 5'-GTTCAGCCCGGCCACCTGCTCGTGGCC-3'; to convert Met223 to Leu223 to obtain the Leu223XI mute i n, 5 *-GTTCAGCCCGGCCAGCTGCTCGTGGCC-3'; to convert Met223 to Ile223 to obtain the IXe223XI mute in, 5'-GTTCAGCCCGGCGATCTGCTCGTGGCC-3•; to convert Met223 to Phe223 to obtain the Phe223XI mute in, 5'-GTTCAGCCCGGCGAACTGCTCGTGGCC-3'; to convert Met223 to Tyr223 to obtain the Tyr223xI mute in, 5 *-GTTCAGCCCGGCGTACTGCTCGTGGCC-3 *; to convert Met223 to Gln223 to obtain the Gln223xI mute in, 5'-GTTCAGCCCGGCCTGCTGCTCGTGGCC-31; to convert Met223 to Asn223 to obtain the Asn223XI mu t e i n, 5'-GTTCAGCCCGGCGTTCTGCTCGTGGCC-3'; to convert Argi40 to Glni4Q to obtain the Glni4QXI mut e i n, 51-CTCGGCACCCTCCTGGCCGCCCCAGGC-3 ' ; 22 r «j to convert Argi4o to Asn^o to obtain the Asni4oXI mute in, 5'-CTCGGCACCCTCGTTGCCGCCCCAGGC-3'; to convert Argi4Q to GIU140 to obtain the GIU140XI Mute in, 5'-CTCGGCACCCTCCTCGCCGCCCAGGC-3'; to convert Argi4Q to Aspi4o to obtain the Aspi4QXI mutein, 5'-CTCGGCACCCTCGTCGCCGCCCCAGGC-3'; to convert Argi4Q to Ilei40 to obtain the Ilei4oXI mut e i n, 5'-CTCGGCACCCTCGATGCCGCCCCAGGC-3'; to convert Argi4Q to Leui40 to obtain the Leui4QXI mut e i n, 5•-CTCGGCACCCTCGAGGCCGCCCCAGGC-3'; to convert Argi4Q to Alai4Q to obtain the Alai4oXI mute in, 5'-CTCGGCACCCTCGGCGCCGCCCCAGGC-3'; to convert Argi4o to Vali4o to obtain the Vali4oXI mutein, 5'-CTCGGCACCCTCGACGCCGCCCCAGGC-3'; and/or to convert Argi4o to Tyri4o to obtain the Tyri4oXI mut e i n, 5 *-CTCGGCACCCTCGTAGCCGCCCCAGGC-3'. example ix Site-Specific Mutagenesis of the Xylose Isomerase Gene to Produce Muteins Having Altered Catalytic Properties The procedure of Example VIII is followed in substantial detail to produce xylose isomerase muteins having altered cat alytic properties. Second strand synthesis and recovery of the desired XI muteins uses the following oligonucleotide primers: to convert Trpi6 to Asni6 to obtain the AsnigXI mute in, 5'-CCAGCCGACGGTGTTCAGTCCGAAGGTG-3'; -• v v n ' ^ 225738 n D Vii*i to convert Trp^g to Glnig to obtain the GlnigXI mutein, 5'-CCAGCCGACGGTCTGCAGTCCGAAGGTG-3'; to convert Trpig to Serig to obtain the SerigXI mute in, 5 * -CCAGCCGACGGTCGACAGTCCGAAGGTG-3'; to convert Trpi6 to Thrig to obtain the Thrig XI mut e i n, 5'-CCAGCCGACGGTCGTCAGTCCGAAGGTG-3'; to convert Trpig to Glyi6 to obtain the GlyigXI mute in, 5'-CCAGCCGACGGTCCCCAGTCCGAAGGTG-3'; to convert Trpi6 to Ala16 to obtain the AlaigXI mute in, 5'-CCAGCCGACGGTCGCCAGTCCGAAGGTG-3'; to convert Trpig to Vali6 to obtain the ValigXI mutein, 5'-CCAGCCGACGGTCACCAGTCCGAAGGTG-3'; to convert Trpi6 to Leui6 to obtain the LeuigXI mutein, 5*-CCAGCCGACGGTCAGCAGTCCGAAGGTG-31; to convert Trpig to Ilei6 t0 obtain the IleigXI mutein, 5'-CCAGCCGACGGTGATCAGTCCGAAGGTG-3'; to convert Trpi6 to Tyr16 t0 obtain the Tyri6XI mute in, 5'-CCAGCCGACGGTGTACAGTCCGAAGGTG-3'; to convert Trp^g to Pheig to obtain the Phei6 XI mutein, 51-CCAGCCGACGGTGAACAGTCCGAAGGTG-3'; to convert Trpig to Hisi6 to obtain the HisigXI mute in, 5'-CCAGCCGACGGTGTGCAGTCCGAAGGTG-3'; to convert Trpi37 to Asni37 to obtain the Asni37XI mute in, 5'-CTCGCGGCCGCCGTTGGCCACATAGGTC-3'; to convert Trpi37 to Glni37 to obtain the Glni37XI mute in, 5'-CTCGCGGCCGCCCTGGGCCACATAGGTC-31; n 225798 to convert Trpi37 to Seri37 to obtain the Seri37XI mut e i n, 5'-CTCGCGGCCGCCCGAGGCCACATAGGTC-3' ; to convert Trpj.37 to Thri37 to obtain the Thri37XI mutein, 5'-CTCGCGGCCGCCCGTGGCCACATAGGTC-3*; to convert Trpi37 to Glyi37 to obtain the Glyi37XI mutein, 5'-CTCGCGGCCGCCCCCGGCCACATAGGTC-3'? to convert Trpi^7 to Alai37 to obtain the Alai37XI mutein, 5'-CTCGCGGCCGCCCGCGGCCACATAGGTC-3'; to convert Trp^37 to Vali37 to obtain the Vali37XI mutein, 5'-CTCGCGGCCGCCCACGGCCACATAGGTC-3'; to convert Trp]_37 to Leui37 to obtain the Leui37XI mut e i n, 5'-CTCGCGGCCGCCCAGGGCCACATAGGTC-3•; to convert Trpi37 to Ilei37 to obtain the Ilei37XI mut e i n, 5'-CTCGCGGCCGCCGATGGCCACATAGGTC-3'; to convert Trpi37 to Tyrj.37 to obtain the Tyri37XI mutein, 5'-CTCGCGGCCGCCGTAGGCCACATAGGTC-3'; to convert Trpi37 to Phei37 to obtain the Phei37XI mute i n, 5'-CTCGCGGCCGCCGAAGGCCACATAGGTC-3'; to convert Trpi37 to Hisi37 to obtain the Hisi37XI mutein, 5'-CTCGCGGCCGCCGTGGGCCACATAGGTC-3'; to convert Pheg4 to Thrg4 to obtain the Thrg4XI mutein, 5'-CACCGGGTGGGTGGTCAGGTTGGTGGTG-3'; to convert Pheg4 to Serg4 to obtain the Serg4XI mutein, 51-CACCGGGTGGGTGGACAGGTTGGTGGTG-31; to convert Pheg4 to Hisg4 to obtain the Hisg4XI mutein, 51-CACCGGGTGGGTGTGCAGGTTGGTGGTG-3'; "81" ,3°ct:?ss; to convert Pheg4 to Valg4 to obtain the Valg4XI mute in, 5'-CACCGGGTGGGTGACCAGGTTGGTGGTG-3'; to convert Pheg4 to Glyg4 to obtain the Glyg4XI mute in, 5'-CACCGGGTGGGTGCCCAGGTTGGTGGTG-3'; to convert Pheg4 to Alag4 to obtain the Alag4XI mute in, 5'-CACCGGGTGGGTGGCCAGGTTGGTGGTG-3'; to convert Pheg4 to Ileg4 to obtain the Ileg4XI mu t e i n, 5'-CACCGGGTGGGTGATCAGGTTGGTGGTG-3'; to convert Pheg4 to Leug4 to obtain the Leug4XI mu t e i n, 5' -CACCGGGTGGGTGAGCAGGTTGGTGGTG-31; to convert Pheg4 to Asng4 to obtain the Asng4XI mute in, 5'-CACCGGGTGGGTGTTCAGGTTGGTGGTG-3'; and/or to convert Pheg4 to Glng4 to obtain the Glng4XI mut e i n, 5'-CACCGGGTGGGTCTGCAGGTTGGTGGTG-3'.

EXAMPLE X Site-Specific Mutagenesis of the Xylose Isomerase Gene To Produce Muteins Having Increased Stability The procedure of Example IX is followed in substantial detail, except that the mutagenesis primers differ. Second strand synthesis and recovery of the desired XI muteins uses the following oligonucleotide primers to change alpha-helical Glycine residues to Alanine residues: to convert Glyi4g to Alai46 t0 obtain the Alai46XI mut e i n, 5'-GCACGTCCTTGGCGCCGGCCGACTCGGCACCC-3'; to convert Glyi6g to Alai66 to obtain the Alai66XI mute in, 5'-GGTGACGTACTCGGCGACCAGGTCGAAG-3'; ^ Ql •.ssijftswt-' a '-25798 /""n v.j> to convert Glyig7 to Alaig7 to obtain the Alaig7XI mute in, 5'-CGCCAGGGCGTGGGCGACGGTGGGGAGC-3 *; to convert Gly2l9 to Ala2ig to obtain the Ala2igXI mutein, 5'-CCATCTGCTCGTGGGCGACCTCGGGGTTC-3'; to convert Gly23i to Ala23i t0 obtain the Ala23iXI mute in, 5 *-GCGCCTGCGCGATCGCGTGCGGGAAGTTC-3'; to convert Gly248 to Ala248 to obtain the Ala248*I mute in, 5'-GATGCCGTTCTGGGCGTTGAGGTCG-3'; to convert Gly2g8 to Ala298 to obtain the Ala298XI mu t e i n, 5' -GAGGCCCACACCGCGTCGAAGTCCTC-3'; to convert Gly305 to Ala3Q5 to obtain the Ala3Q5XI mute in, 51-GTTGCGCATGCAGGCGGCCGCCGAGG-3'? and/or to convert Gly36g to Ala369 to obtain the Ala3ggXI mu t e i n, 5'-GCTCGAAGGCCATCGCACGGGCCGCCGCC-3'.

EXAMPLE XI The method of the present invention as described above and in Example I herein, was used to predict amino acid residue positions in the reference xylose isomerase structure that could accomodate proline substitutions. Positions for mutation were initially selected by calculating the phi/psi angles for the various amino acids in the reference XI, and then identifying those residues that met the phi/psi angle criteria set forth in Example I. This led to the selection of twenty-five candidates for proline substitution. The x-ray structure for each mutation candidate was examined using an interactive computer graphics system. A proline residue was substituted for the amino acid that occurs in the reference XI, and the computer-generated model for the potential mutant XI was visually examined for unfavorable steric contact £>* v c: -83- 'i3ocr.%. • .o 22579? (interpenetration of the Van der Waals surfaces for any proline vs. XI atom). This analysis allowed the rejection of many candidate proline-substitution mutations.

The following primers are used to introduce proline residues to produce thermostable xylose isomerase muteins: to convert Leuj.5 to Prois to obtain the ProisXI X, mutein, 5'-CCGACGGTCCACGGTCCGAAGGTG-3*; to convert Asp28 to Pro28 to obtain the Pro28XI mu t e i n, 51 -GCCGCGTGGCGGGACCGAAGGGGTCC- 3' ; to convert Ala29 to Pro2g to obtain the Pro2gXI mutein, 5'-GCGCCGCGTGGGGTCACCGAAGGGG-3 • ; to convert Arg32 to Pro32 to obtain the Pro32XI mutein, 5'-CCGGGTCGAGGGCCGGCCGCGTGGCGTC-3'; to convert Ala33 to Pro33 to obtain the Pro33XI mutein, 5'-CCGGGTCGAGGGGGCGCCGCGTGG-3'; to convert Serg4 to Prog4 to obtain the Pro64XI mutein, 5'-GCTCGCTGTCGGGGGAGCCGAAGGGG-3'; v '' o to convert Asni07 to Pro107 to obtain the Proio7XI mutein, 5'-CGTCGCGGTCGGGGGCGGTGAAGC-3'; to convert Arg^og to Prolog to obtain the ProiogXI mute in, 5•-GTAGCGGCGCACGTCGGGGTCGTTGGCGGTG-31; to convert Glyi46 to Proi4g to obtain the Proi46XI mute i n, 5'-GCACGTCCTTGGCGCCGGGCGACTCGGCACCC-3'; to convert Valisi to Proisi to obtain the ProisiXI mute in, 5'-GAGGGCGTCCCGCGGGTCCTTGGCGCC-3'; ft r*. 22579 D -fn to convert Glyisg to Pro^gg to obtain the ProiggXI mute in, 5'-GAGCAGGATGTCGGGGCGCGGCTCGTTC-3'; to convert Leuig2 to Pro192 to obtain the Pro]_g2XI mutein, 5'-CGGTGGGGAGCGGGATGTCGCCGCG-3'; to convert G1U207 to Pro207 to obtain the Pro207x* mute in, 5'-CAGCTCCGGTCGCGGCAGGCGCTCGATG-3'; to convert Val2l8 to Pro2i8 to obtain the Pro2i8xl mutein, 5'-GCTCGTGGCCGGGCTCGGGGTTCACGC-3'; to convert Arg259 to Pro259 to obtain the Pro25gXI mutein, 5'-CCCGCGCCGAAGGGGAGGTCCTGGTC-3'; to convert Arg2g2 to Pro2g2 to obtain the Pro292xI mutein, 51-GAAGTCCTCGGTCGGCGGCGGCTTGAAG-3'; to convert Thr342 to Pr0342 to obtain the Pro342xI mutein, 5'-CGTAGGCCGCCGGGGGCCGGGCCAG-3'; to convert Arg354 to Pr0354 to obtain the Pro354XI mute in, 5'-CGAAGGCGGACGGGTCGTCGAGCAGG-3'; to convert Gly36g to Pro36g to obtain the Pro3ggXI mute in, 5'-GCTCGAAGGCCATGGGACGGGCCGCCGCC-3'.

In addition, locations in the reference xylose isomerase for proline substitution are selected by comparing the amino acid sequence of Ampullariella xylose isomerase with that of the reference XI to select residues in the reference xylose isomerase for proline substitution. to convert Ile252 to Pro252 to obtain the Pro252xI mute in, 5'-GGTCGTACTTGGGGCCGTTCTGGCCG-3'; to convert Argi77 to Proi77 to obtain the Pro]_77XI mutein; 5'-CTCGATGGCGAAGGGGATGTCGTAGCC-3'; and/or \ 13cct/963 r\ 225 to convert ASP345 to Pro345 to obtain the Pro345XI mut e i n; 5»-GGCCTGCAGACCGGGGGCCGCCGTGGG-3 *.

The following primers are used to introduce aromatic amino acid residues to produce thermostable xylose isomerase muteins: to convert Aspg to Tyrg to obtain the TyrgXI mutein, 5'-GGTGAACCTGTACTCGGGGGTGGGC-3'; to convert Gln2i to Tyr2i to obtain the Tyr2iXI mutein, 5'-GGGGTCCCGTCCGTACCAGCCGACGGTCC-3*; to convert Ala29 to Tyr29 to obtain the Tyr2gXI mutein, 5'-GGCGCGCCGCGTGTAGTCACCGAAGGGG-3'; to convert Arg32 to Tyr32 to obtain the Tyr32XI mut e i n, 5'-GGGTCGAGGGCGTACCGCGTGGCGTCAC-3'; to convert GIU38 to Tyr3Q to obtain the Tyr38XI mute in, 5 *-CCGCCGCACCGAGTAGACCGGGTCGAGGG-3 ' ? to convert Leu46 to Phe4g to obtain the Phe4gXI mute i n, 51-GCCGTGGGCGCCGAACTCGGCCAGCCGCC-3 ' ; to convert Leu46 to Tyr4g to obtain the Tyr46XI mute in, 5'-GCCGTGGGCGCCGTACTCGGCCAGCCGCC-3'; to convert Asps6 to Phe56 to obtain the PhesgXI mu t e i n, 5'-GGGGATGAGGTCGAAGTCGTGGAACGTG-3'; to convert Asp56 to Tyr56 to obtain the TyrsgXI mu t e i n, 5'-GGGGATGAGGTCGTAGTCGTGGAACGTG-3'; to convert Leuss to Phess to obtain the PhessXI mutein, 5'-GCCGAAGGGGATGAAGTCGTCGTCGTGG-3'; SV "V' ,t n r\ O ,J25 79 to convert Leuss to Tyrss to obtain the TyrsaXI mut e i n, 5'-GCCGAAGGGGATGTAGTCGTCGTCGTGG-3'; to convert Vali27 to Tyrj.27 to obtain the Tyri27XI mutein, 51-GGCGCCGAGCTCGTACGCGAGGTCGATG-3'; to convert Thri.33 to Phei33 to obtain the Phei33XI mu t e i n, 5' -CCAGGCCACATAGAACTCGGCGCCGAGCTC-3' ; to convert Thri33 to Tyr133 t0 obtain the Tyri33XI mute i n, 5'-CCAGGCCACATAGTACTCGGCGCCGAGCTC-3 ' to convert Alai36 to phe136 to obtain the Phei36XI mutein, 5'-GGCCGCCCCAGAACACATAGGTCTCGG-3 ' to convert Alai36 to TYr136 to obtain the Tyr^gXI mut e i n, 5'-GGCCGCCCCAGTACACATAGGTCTCGG-3 ' to convert Argi77 to Tyrj.77 to obtain the Tyrj77XI mute in, 5'-GCTCGATGGCGAAGTAGATGTCGTAGCCC-31 ; to convert Ileiso to phei80 to obtain the Phei80XI mute i n, 5' -CGGC.TTGGGCTCGAAGGCGAAGCGGATG- 3 ' ; to convert Ileiso to Tyr180 to obtain the Tyri80XI mute in, 5'-CGGCTTGGGCTCGTAGGCGAAGCGGATG-3'; to convert Leuig3 to Phei93 to obta mute in, 5'-CCGACGGTGGGGAACAGGATGTCGCC-3' to convert Leuig3 to Tyr193 to obta mute in, 5'-CCGACGGTGGGGTACAGGATGTCGCC-3' n the Phei93XI n the Tyr].g3XI to convert Leu2n to Phe2H to obtain the Phe2nXI mut e i n, 5'-GGTTCACGCCGTAGAACTCCGGTCGCTCC-3'; to convert Leu2H t0 Tyr211 to obtain the Tyr2nXI mute in, 5'-GGTTCACGCCGTAGTACTCCGGTCGCTCC-3' ; I 3 OCT .'988,; v* //' a 225798 to convert Asn227 to Phe227 to obtain the Phe227XI mutein, 5'-GCCGTGCGGGAAGAACAGCCCGGCCATC-3'; **0 to convert Asn227 to Tyr227 to obtain the Tyr227XI mutein, 5'-GCCGTGCGGGAAGTACAGCCCGGCCATC-3'; to convert Asn227 to Trp227 to obtain the Trp227XI mu t e i n, 5 ' -GCCGTGCGGGAACCACAGCCCGGCCATC-3'; to convert Gln234 to Phe234 to obtain the Phe234XI mutein, 51-CCGCCCACAGCGCGAACGCGATGCCGTG-3'; to convert Gln234 to Tyr234 to obtain the Tyr234XI mutein, 51 -CCGCCCACAGCGCGTACGCGATGCCGTG-3'; to convert Ala238 to Phe238 to obtain the Phe238xI mutein, 5'-GGAACAGCTTGCCGAACCACAGCGCCTGC-3'; to convert Ala238 to Tyr238 to obtain the Tyr238xI mutein, 5'-GGAACAGCTTGCCGTACCACAGCGCCTGC-3' to convert Leu246 to Phe246 to obtain the Phe246XI mutein, 5'-GTTCTGGCCGTTGAAGTCGATGTGGAAC-31; to convert Leu246 to Tyr246 to obtain the Tyr246XI mutein, 5'-GTTCTGGCCGTTGTAGTCGATGTGGAAC-3'; to convert Arg284 to Phe284 to obtain the Phe284xI mutein, 51-GAAGTCGAAGTGGTACGGGCCGCTGTAGC-3'; to convert Arg308 to Tyr3Q8 to obtain the Ty^ogXI mute in, 5' -GGATCAGGTAGTTGTACATGCAGCCGG3; to convert Leu3n to Phe3n to obtain the Phe3nXI mutein, 5' -CGCTCCTTGAGGATGAAGTAGTTGCGCATGC-3'; to convert Leu3n to Tyr3n to obtain the Tyr3nXI mutein, 5' -CGCTCCTTGAGGATGTAGTAGTTGCGCATGC-3'; -88- / : (o: fuocm* /-"■v r 225798 to convert Arg3i6 to Tyr3i5 to obtain the Tyr3igXI mute in, 5'-GAAGGCCGCCGCGTACTCCTTGAGGATC-3»; to convert Leu335 to Phe33j to obtain the Phe33sXI mutein, 5'-GCCAGCTCGTCGAAACGGGACGCGC-3'; to convert Leu335 to Tyr335 to obtain the Tyr335XI mute in, 5'-GCCAGCTCGTCGTAACGGGACGCGC-3' ; to convert Val362 to Phe362 to obtain the Phe362XI mut e i n, 5'-GCCGCCGCGTCGAAGTCGAACTCCTCG-3'; to convert Val362 to Tyr362 to obtain the Tyr362x* mut e i n, 51-GCCGCCGCGTCGTAGTCGAACTCCTCG-3' ; to convert Met370 to Phe370 to obtain the Phe37oXI mute in, 5' -CGCTCGAAGGCGAACCCACGGGCCG-3'; to convert Met370 to Tyr37o to obtain the Tyr37QXI mute in, 5'-CGCTCGAAGGCGTACCCACGGGCCG-3'; to convert Leu375 to Phe375 to obtain the Phe375XI mutein, 5'-CGCCAGCTGGTCGAAGCGCTCGAAGGC-3'; to convert Leu375 to Tyr375 to obtain the Tyr37sXI mu t e i n, 5'-CGCCAGCTGGTCGTAGCGCTCGAAGGC-3' ; to convert Leu383 to Phe383 to obtain the Phe383XI mutein, 51-CGGGCGCCCAGGAAGTGGTCCATCGC-3'; and/or to convert Leu383 to Tyr383 to obtain the Tyr383XI mutei n, 5'-CGGGCGCCCAGGTAGTGGTCCATCGC-3'.

The following primers are used to substitute amino acid residues for residues that are located near the interface of subunits of the xylose isomerase protein that may undergo deamidation, to produce xylose isomerase muteins more stable toward irreversible thermal inactivation: -89- / ^ . 1 3 0ctj'985,r .vr" & r^ ^ 225798 ! to convert Gln2i to Ala2i to obtain the Ala2iXI | mute in, 5'-GGGGTCCCGTCCGGCCCAGCCGACGGTCC-3'; to convert Gln2i to Val2i to obtain the Val2iXI : mutein, 5'-GGGGTCCCGTCCGACCCAGCCGACGGTCC-3'; I to convert Gln2i to Leu2i to obtain the Leu2iXI mute in, 5'-GGGGTCCCGTCCGAGCCAGCCGACGGTCC-3'; : p' to convert Gln2i to Ile2i to obtain the Ile2i ! mute in, 5'-GGGGTCCCGTCCGATCCAGCCGACGGTCC-3'? to convert Gln2i to Ser2i to obtain the Ser2i ; mutein, 5'-GGGGTCCCGTCCGCTCCAGCCGACGGTCC-3'; j to convert Gln2i to Thr2i to obtain the Thr2i * mute in, 5 *-GGGGTCCCGTCCGGTCCAGCCGACGGTCC-3'; \ to convert Gln2i to His2i to obtain the His2i | mutein, 5'-GGGGTCCCGTCCGTGCCAGCCGACGGTCC-3'; i | to convert Gln2i to Tyr2i to obtain the Tyr2i 1 mutein, 5'-GGGGTCCCGTCCGTACCAGCCGACGGTCC-31; 2 to convert Gln2i to Lys2i to obtain the Lys2i ^ mutein, 5'-GGGGTCCCGTCCCTTCCAGCCGACGGTCC-3'; to convert Gln2i to Arg2i to obtain the Arg2i mute i n, 5'-GGGGTCCCGTCCGCGCCAGCCGACGGTCC-3 *; to convert Gln2i to Met2i to obtain the Met2i mut e i n, 5'-GGGGTCCCGTCCCATCCAGCCGACGGTCC-3'; to convert Gln2i to Pro2i to obtain the Pro2i mutein, 5'-GGGGTCCCGTCCGGGCCAGCCGACGGTCC-3'; to convert Asng2 to Alag2 to obtain the Alag2 mutein, 5'-GGTGGGTGAACAGGGCGGTGGTGGCCATCG-3' t30cr;;« 225798 to convert Asng2 to Val.92 to obtain the Valg2 mutein, 5'-GGTGGGTGAACAGGACGGTGGTGGCCATCG-3'; to convert Asng2 to Leug2 to obtain the Leug2 mutein, 51-GGTGGGTGAACAGGAGGGTGGTGGCCATCG-31; to convert Asng2 to Ileg2 to obtain the Ileg2 mut e i n, 5'-GGTGGGTGAACAGGATGGTGGTGGCCATCG-3'; to convert Asng2 to Serg2 to obtain the Serg2XI mut e i n, 5'-GGTGGGTGAACAGGCTGGTGGTGGCCATCG-3'; to convert Asng2 to Thrg2 to obtain the Thrg2XI mut e i n, 5'-GGTGGGTGAACAGGGTGGTGGTGGCCATCG-3 *; to convert Asng2 to Hisg2 to obtain the Hisg2XI mut e i n, 5'-GGTGGGTGAACAGGTGGGTGGTGGCCATCG-31; to convert Asng2 to Tyrg2 to obtain the Tyrg2XI mutein, 5'-GGTGGGTGAACAGGTAGGTGGTGGCCATCG-3 *; to convert Asng2 to Lysg2 to obtain the Lysg2XI mutein, 5'-GGTGGGTGAACAGCTTGGTGGTGGCCATCG-3'; to convert Asng2 to Argg2 to obtain the Argg2XI mutein, 5'-GGTGGGTGAACAGGCGGGTGGTGGCCATCG-3'; to convert Asng2 to Metg2 to obtain the Metg2XI mutein, 51-GGTGGGTGAACAGCATGGTGGTGGCCATCG-3'; to convert Asng2 to Prog2 to obtain the Prog2XI mute i n, 5'-GGTGGGTGAACAGGGGGGTGGTGGCCATCG-3'; to convert Asni07 to Alai07 to obtain the Alaio7XI mu t e i n, 5' -GCACGTCGCGGTCGGCGGCGGTGAAGCCG-3'; to convert Asni07 to Valio7 to obtain the Val].07XI mutein, 5'-GCACGTCGCGGTCGACGGCGGTGAAGCCG-31; 1 3 octjpsa^j 2257 to convert Asnio7 to Leui07 to obtain the Leu]_07XI mutein, 5'-GCACGTCGCGGTCGAGGGCGGTGAAGCCG-3'; to convert Asnio7 to Ileio7 to obtain the Ileio7XI mut e i n, 5'-GCACGTCGCGGTCGATGGCGGTGAAGCCG-3'; to convert Asnio7 to Serio7 to obtain the Serio7XI mutein, 5'-GCACGTCGCGGTCGCTGGCGGTGAAGCCG-3'; to convert Asni07 to Thri07 to obtain the Thri07XI mutein, 5'-GCACGTCGCGGTCGGTGGCGGTGAAGCCG-3'; to convert AsniQ7 to HisiQ7 to obtain the Hisio7xI mutein, 51-GCACGTCGCGGTCGTGGGCGGTGAAGCCG-3'; to convert AsniQ7 to Tyrio7 to obtain the TyriQ7XI mutein, 5'-GCACGTCGCGGTCGTAGGCGGTGAAGCCG-3 *; to convert Asnio7 to Lysi07 to obtain the Lysi07XI mutein, 5'-GCACGTCGCGGTCCTTGGCGGTGAAGCCG-3' to convert Asn]_07 to Argio7 to obtain the ArgiQ7XI mutein, 5'-GCACGTCGCGGTCGCGGGCGGTGAAGCCG-3'; to convert Asnio7 to Metio7 to obtain the Metio7XI mu t e i n, 51 -GCACGTCGCGGTCCATGGCGGTGAAGCCG-3 ' ; to convert AsniQ7 to ProiQ7 to obtain the Proio7xI mut e i n, 5'-GCACGTCGCGGTCGGGGGCGGTGAAGCCG-3 ' ; to convert Asn^gS to Alai85 to obtain the AlaissXI mutein, 5'-CGCCGCGCGGCTCGGCCGGCTTGGGCTCG-3 ' ; to convert Asn^gs to Vali.95 to obtain the ValissXI mute i n, 5'-CGCCGCGCGGCTCGACCGGCTTGGGCTCG-3 ' to convert Asni85 to Leuiss to obtain the LeuigsXI mutein, 5'-CGCCGCGCGGCTCGAGCGGCTTGGGCTCG-3 ' ; 3 OfT' to convert Asnies to Hexes to obtain the Ilei85XI mute i n, 5'-CGCCGCGCGGCTCGATCGGCTTGGGCTCG-3' to convert Asni85 to Series to obtain the SerxssXI mutein, 5'-CGCCGCGCGGCTCGCTCGGCTTGGGCTCG-31 ; to convert Asniss to Thrx85 to obtain the Thri85XI mutein, 5'-CGCCGCGCGGCTCGGTCGGCTTGGGCTCG-3'; to convert Asniss to Hisx85 to obtain the Hisx85x* mut e i n, 5 *-CGCCGCGCGGCTCGTGCGGCTTGGGCTCG-3'; to convert Asnxss to Tyrx85 to obtain the TyrissXI mutein, 5'-CGCCGCGCGGCTCGTACGGCTTGGGCTCG3; to convert Asniss to Lysx85 to obtain the Lysx85XI mut e i n, T•CGCCGCGCGGCTCCTTCGGCTTGGGCTCG-3'; to convert Asni85 to Argxss to obtain the ArgissXI mut e i n, 5'-CGCCGCGCGGCTCGCGCGGCTTGGGCTCG-3' ; to convert Asniss to Meti85 to obtain the MetissXI mutein, 5■-CGCCGCGCGGCTCCATCGGCTTGGGCTCG-3' ; to convert Asniss to Proi85 to obtain the Proi85XI mutein, 5'-CGCCGCGCGGCTCGGGCGGCTTGGGCTCG-3"; to convert Asn227 to Ala227 to obtain the Ala227XI mutein, 5'-GCCGTGCGGGAAGGCCAGCCCGGCCATC-3'; to convert Asn227 to Val227 to obtain the Val227XI mutein, 5'-GCCGTGCGGGAAGACCAGCCCGGCCATC-3'; to convert Asn227 to Leu227 to obtain the Leu227XI mute in, 5'-GCCGTGCGGGAAGAGCAGCCCGGCCATC-3 *; to convert Asn227 to Ile227 to obtain the Ile227XI mute in, 5'-GCCGTGCGGGAAGATCAGCCCGGCCATC3 1 3 CCTi988<;:[ .sssssif.'" n ^ 225798 to convert Asn227 to Ser227 to obtain the Ser227XI mute in, 5'-GCCGTGCGGGAAGCTCAGCCCGGCCATC-3'; to convert Asn227 to Thr227 to obtain the Thr227XI mute in, 5'-GCCGTGCGGGAAGGTCAGCCCGGCCATC-3'; to convert Asn227 to His227 to obtain the His227XI mu t e i n, 5'-GCCGTGCGGGAAGTGCAGCCCGGCCATC- 3' ; to convert Asn227 to Tyr227 to obtain the Tyr227x* mut e i n, 5'-GCCGTGCGGGAAGTACAGCCCGGCCATC-3'; to convert Asn227 to Lys227 to obtain the Lys227XI mutein, 5'-GCCGTGCGGGAACTTCAGCCCGGCCATC-31; to convert Asn227 to Arg227 to obtain the Arg227XI mutein, 5*-GCCGTGCGGGAAGCGCAGCCCGGCCATC-31; to convert Asn227 to Met227 to obtain the Met227XI mutein, 5'-GCCGTGCGGGAACATCAGCCCGGCCATC-31; to convert Asn227 to Pro227 to obtain the Pro227XI mut e i n, 5'-GCCGTGCGGGAAGGGCAGCCCGGCCATC-3'; to convert Gln234 to Ala234 to obtain the Ala234xI mut e i n, 5'-CGCCCACAGCGCGGCCGCGATGCCGTGCG-3'; to convert Gln234 to Val234 to obtain the Val234xI mutein, 5'-CGCCCACAGCGCGACCGCGATGCCGTGCG-31; (^) to convert Gln234 to Leu234 to obtain the Leu234xI mut e i n, 5'-CGCCCACAGCGCGAGCGCGATGCCGTGCG-3'; to convert Gln234 to Ile234 to obtain the Ile234xI mut e i n, 5'-CGCCCACAGCGCGATCGCGATGCCGTGCG-3'; to convert Gln234 to Ser234 to obtain the Ser234XI mut e i n, 5'-CGCCCACAGCGCGCTCGCGATGCCGTGCG-3'; 130 o ui i :o8s 2257 to convert Gln234 to Thr234 to obtain the Thr234XI mute in, 5' -CGCCCACAGCGCGGTCGCGATGCCGTGCG-3'; to convert Gln234 to His234 to obtain the His234XI mutein, 5' -CGCCCACAGCGCGTGCGCGATGCCGTGCG-3'; to convert Gln234 to Tyr234 to obtain the Tyr234XI mute in, 5 ' -CGCCCACAGCGCGTACGCGATGCCGTGCG-3'; to convert Gln234 to Lys234 to obtain the Lys234XI mutein, 5'-CGCCCACAGCGCCTTCGCGATGCCGTGCG-3'; to convert Gln234 to Arg234 to obtain the Arg234XI mute i n, 5'-CGCCCACAGCGCGCGCGCGATGCCGTGCG-3' to convert Gln234 to Met234 to obtain the Met234XI mute in, 5•-CGCCCACAGCGCCATCGCGATGCCGTGCG-3'; to convert Gln234 to Pro234 to obtain the Pro234XI mutein, 51-CGCCCACAGCGCGGGCGCGATGCCGTGCG-3'; to convert Gln256 to Ala256 to obtain the Ala256x* mut e i n, 5'-CGAAGCGGAGGTCGGCGTCGTACTTGATGC-3'; to convert Gln256 to Val256 to obtain the Val256XI mutein, 5'-CGAAGCGGAGGTCGACGTCGTACTTGATGC-3'; to convert Gln256 to Leu256 to obtain the Leu256XI mutein, 5'-CGAAGCGGAGGTCGAGGTCGTACTTGATGC-3'; to convert Gln256 to Ile256 to obtain the Ile256XI mute i n, 5'-CGAAGCGGAGGTCGATGTCGTACTTGATGC-3'; to convert Gln256 to Ser256 to obtain the Ser256XI mute in, 51-CGAAGCGGAGGTCGCTGTCGTACTTGATGC-3'; to convert Gln256 to Thr256 to obtain the Thr256xI mutein, 5'-CGAAGCGGAGGTCGGTGTCGTACTTGATGC-3'; / c 1 ^ CCTM81 225798 to convert Gln256 to His256 to obtain the His256XI mut e i n, 5'-CGAAGCGGAGGTCGTGGTCGTACTTGATGC-3'; to convert Gln256 to Tyr256 to obtain the Tyr256XI mutein, 5'-CGAAGCGGAGGTCGTAGTCGTACTTGATGC-31; to convert Gln256 to Lys256 to obtain the Lys256xI mutein, 5'-CGAAGCGGAGGTCCTTGTCGTACTTGATGC-3'; to convert Gln256 to Arg256 to obtain the Arg256x* mute in, 5'-CGAAGCGGAGGTCGCGGTCGTACTTGATGC-3 ' ; to convert Gln256 to Met256 to obtain the Met256XI mutein, 5'-CGAAGCGGAGGTCCATGTCGTACTTGATGC-3'; to convert Gln256 to Pro25g to obtain the Pro256XI mute in, 5'-CGAAGCGGAGGTCGGGGTCGTACTTGATGC-3'; to convert Asn309 to Ala3og to obtain the Ala309XI mute i n, 5'-GAGGATCAGGTAGGCGCGCATGCAGCCGGC-3 ' ; to convert Asn309 to Val3og to obtain the Val309XI mut e i n, 5'-GAGGATCAGGTAGACGCGCATGCAGCCGGC-3'; to convert Asn309 to Leu309 to obtain the Leu309XI mut e i n, 5'-GAGGATCAGGTAGAGGCGCATGCAGCCGGC-3 ' ; to convert Asn309 to Ile3og to obtain the Ile309XI mut e i n, 5'-GAGGATCAGGTAGATGCGCATGCAGCCGGC-3 *; to convert Asn309 to Ser3Q9 to obtain the Ser309XI mute in, 5'-GAGGATCAGGTAGCTGCGCATGCAGCCGGC-3' ; to convert Asn309 to Thr3og to obtain the Thr309XI mut e i n, 5'-GAGGATCAGGTAGGTGCGCATGCAGCCGGC-3 ' ; to convert Asn309 to His309 to obtain the His309XI mutein, 5'-GAGGATCAGGTAGTGGCGCATGCAGCCGGC-3' ; • — • i s 13 OCT 1988 to convert Asn3og to Tyr3Q9 to obtain the Tyr3ggXI mute in, 5'-GAGGATCAGGTAGTAGCGCATGCAGCCGGC-3 »; to convert Asn3Qg to Lys3og to obtain the Lys3ogXI mu t e i n, 5'-GAGGATCAGGTACTTGCGCATGCAGCCGGC-3'; to convert Asn3og to Arg3Q9 to obtain the ArgsogXI mute in, 5'-GAGGATCAGGTAGCGGCGCATGCAGCCGGC-3'; to convert Asn3og to Met3og to obtain the Met3ogXI mutein, 5'-GAGGATCAGGTACATGCGCATGCAGCCGGC-3'; to convert Asn3Qg to Pro3Q9 to obtain the Pro3Q9XI mute in, 5'-GAGGATCAGGTAGGGGCGCATGCAGCCGGC-3'; to convert Asn377 to Ala377 to obtain the Ala377XI mute in, 5'-GGTCCATCGCCAGGGCGTCCAGGCGCTCG-3'; to convert Asn377 to Val377 to obtain the Val377XI mutein, 5'-GGTCCATCGCCAGGACGTCCAGGCGCTCG-3'; to convert Asn377 to Leu377 to obtain the Leu377XI mute in, 5'-GGTCCATCGCCAGGAGGTCCAGGCGCTCG-3'; to convert Asn377 to Ile377 to obtain the Ile377XI mutein, 5'-GGTCCATCGCCAGGATGTCCAGGCGCTCG-3*; to convert Asn377 to Ser377 to obtain the Ser377XI mute i n, 5'-GGTCCATCGCCAGGCTGTCCAGGCGCTCG-3 *; to convert Asn377 to Thr377 to obtain the Thr377XI mut e i n, 5'-GGTCCATCGCCAGGGTGTCCAGGCGCTCG-3'; to convert Asn377 to His377 to obtain the His377XI mut e i n, 5'-GGTCCATCGCCAGGTGGTCCAGGCGCTCG-3'; to convert Asn377 to Tyr377 to obtain the Tyr377XI mut e i n, 5'-GGTCCATCGCCAGGTAGTCCAGGCGCTCG-3'; f ' t " f 3 OCT/ r> 225798 to convert Asn377 to Lys377 to obtain the Lys377XI mute in, 5'-GGTCCATCGCCAGCTTGTCCAGGCGCTCG-31; to convert Asn377 to Arg377 to obtain the Arg377XI mut e i n, 5'-GGTCCATCGCCAGGCGGTCCAGGCGCTCG-3'; to convert Asn377 to Met377 to obtain the Met377XI mute in, 5'-GGTCCATCGCCAGCATGTCCAGGCGCTCG-3'; and/or to convert Asn377 to Pro377 to obtain the Pro377XI mutein, 5'-GGTCCATCGCCAGGGGGTCCAGGCGCTCG-31.

The following primers are used to introduce two cysteine residues to produce disulfide bridges in the xylose isomerase protein to create thermostable xylose isomerase muteins: to convert Trp270 to Cys270f and G1yi46 to cys146» 5*-CAGGTCCACCAGGCAGAACGCGGCCCGC-31, and 5'-GTCCTTGGCGCCGCACGACTCGGCACCC-31, to obtain the Cys270cYs146XI lutein; to convert Phe320 to Cys320r an<^ H*-s382 to CYS382' 51-GGTCGGCGCGGCAGGCCGCCGCACGC-3', and '-CGCCCAGCAGGCAGTCCATCGCCAGC-3', to obtain the Cys32OCys302XI mutein; to convert GIU337 to Cys337, and Argio9 to cys109' 5'-GGGCCGGGCCAGGCAGTCCAGACGGGAC-3', and 5'-CGGCGCACGTCGCAGTCGTTGGCGGTG-31, to obtain the CyS337CysiQ9XI mutein; to convert Gly^gg to Cys^sg, and GIU144 to Cysi44, 5'-GCAGGATGTCGCAGCGCGGCTCGTTC-3', and 51-GGCGCCACCCGAGCAGGCACCCTCGCGG-3', to obtain the Cysi89Cysi44XI mutein; ^' ' : c .

[I 1 3 nrriooa o!} 98 n 225798 to convert Gly25i to Cys25i, and Gly225 to Cys225» 5'-TCGTACTTGATGCAGTTCTGGCCGTTG-3', and 5'-GCGGGAAGTTCAGGCAGGCCATCTGCTCG-3', to obtain the CyS251cYs225XI mutein; to convert Ala366 to Cys366» and Valgs to Cysgs, 5'-CCATCCCACGGGCGCACGCCGCGTCGACG-31, and 5'-CGCCGTCCTTGAAGCACGGGTGGGTGAAC-3', to obtain the CyS366Cysg8XI mutein; to convert Gln249 to Cys249, and Gly2ig to Cys2i9, 5'-GTACTTGATGCCGTTGCAGCCGTTGAGGTCG-3'; and 5'-CATCTGCTCGTGGCAGACCTCGGGGTTC-3', to obtain the Cys249Cys2igXI mutein; and/or to convert G1U207 to Cys207r and Aspi63 to CYS163' 5'-CAGCTCCGGTCGGCACAGGCGCTCGATG-3', and 5'-CTCGCCGAGCAGGCAGAAGGCCTCCTTC-3', to obtain the CyS207Cysi63XI mutein.

EXAMPLE XII Site-Specific Mutagenesis of the Xylose Isomerase Gene To produce Muteins Having Lowered pH Optima The procedure of Example IX is followed in substantial detail, except that the mutagenesis primers differ. Second strand synthesis and recovery of the desired XI muteins uses the following oligonucleotide primers to alter amino acids within 15 angstroms of the enzyme active site to eliminate negative charges or introduce positive charges to produce xylose isomerase muteins with lowered pH optima as follows: to convert Asp57 to Lys 57 to obtain the Asp 57 XI mutein; 5'-GCCGAAGGGGATGAGCTTGTCGTCGTGGAACGT-3' 1 3 OCT/?, 2257 to convert ASP57 to Arg 57 to obtain the Arg 57 XI mutein; 5'-GCCGAAGGGGATGAGGCGGTCGTCGTGGAACGT-3' to convert ASP57 to Gly 57 to obtain the Gly 57 XI mutein; 5'-GCCGAAGGGGATGAGGCCGTCGTCGTGGAACGT-3' to convert ASP57 to Ala 57 to obtain the Ala 57 XI mute in; 5 *-GCCGAAGGGGATGAGGGCGTCGTCGTGGAACGT-3' to convert ASP57 to Gin 57 to obtain the Gin 57 XI mute in; 5'-GCCGAAGGGGATGAGCTGGTCGTCGTGGAACGT-3' to convert ASP57 to Asn 57 to obtain the Asn 57 XI mut e i n; 5'-GCCGAAGGGGATGAGGTTGTCGTCGTGGAACGT-3' to convert ASP57 to Thr 57 to obtain the Thr 57 XI mutein; 5'-GCCGAAGGGGATGAGGGTGTCGTCGTGGAACGT-3' to convert ASP57 to Ser 57 to obtain the Ser 57 XI mutein; 5'-GCCGAAGGGGATGAGGGAGTCGTCGTGGAACGT-3' to convert Glui86 to LYS 186 to obtain the Lys 186 XI mutein; 5'-GATGTCGCCGCGCGGCTTGTTCGGCTTGGGCTC-3' to convert Glui86 to Ar9 186 to obtain the Arg 186 XI mut e i n; 5'-GATGTCGCCGCGCGGGCGGTTCGGCTTGGGCTC-3' to convert Glui86 to G1Y 186 to obtain the Gly 186 XI mut e i n; 5'-GATGTCGCCGCGCGGGCCGTTCGGCTTGGGCTC-3' to convert Glui86 to Ala 186 to obtain the Ala 3.86 XI mute in; 5'-GATGTCGCCGCGCGGGGCGTTCGGCTTGGGCTC-3• to convert Glui86 to Gin 186 to obtain the Gin 136 XI mut e i n; 5'-GATGTCGCCGCGCGGCTGGTTCGGCTTGGGCTC-3' to convert Glui86 to Asn 186 to obtain the Asn 186 XI mut e i n; 5'-GATGTCGCCGCGCGGGTTGTTCGGCTTGGGCTT-3' to convert Gluigg to Thr igg to obtain the Thr igg XI mutein; 5'-GATGTCGCCGCGCGGCGTGTTCGGCTTGGGCTC-3' to convert Gluigg to Ser ig6 to obtain the Ser igg XI mu t e i n; 5 *-GATGTCGCCGCGCGGCGAGTTCGGCTTGGGCTC-3' to convert G1U221 to Gin 221 to obtain the Gin 221 XI mutein; 5'-CAGCCCGGCCATCTGCTGGTGGCCGACCTCGGG-3' to convert GIU141 to Gin 141 to obtain the Gin 141 XI mutein; 5'-ACCCGACTCGGCACCCTGGCGGCCGCCCCAGGC-3'; and/or to convert Asp287 to Asn 287 to obtain the Asn 287 XI mutein; 5'-CCGCGGCGGCTTGAAGTTGAAGTGCCGCGGGCC-3'.

EXAMPLE XIII Xylose-Isomerase Muteins Exhibiting Modified Kinetics The isomerase activity of the reference and xylose isomerase muteins obtained as described above, is assayed using the substrates glucose, fructose, xylose and xylulose. Kinetic measurements are taken of the Kp , Kg, kcat , and kcat for both equilibrium reactions using the HPLC assay previously-described. Kinetic parameters are obtained by analysis of the progress curves of the reactions, using a program that carries out a weighted linear or nonlinear least-squares regression analysis of data by using the Lineweaver-Burk or Michaelis-Menten equations, respectively, such as that described by Roberts, in Enzyme Kinetics, Cambridge Univ. Press, Cambridge p. 299-306 (1977), incorporated by reference herein. Data is examined for modified enzymes showing a changed specificity, relative to the reference enzyme, toward either glucose or xylose substrate. '3 oct;9ss: 225798 EXAMPLE XIV Determination of Thermostability of Xylose Isomerase Muteins Reference XI or XI mutein is produced and purified as described in Example II. The purified protein is adjusted to an average activity of 1.8-2.0 u/ml as determined by HPLC assay, and after precooling in a salt and ice bath, is mixed with glucose solution also precooled (400 g/1 glucose, 25 mM maleic acid, lOmM MgS04, pH 6.5) in a 1:1 ratio at 1°C and distributed into 100 pi thin-walled glass micropipettes (Fisher Scientific, Pittsburgh, PA) which are flame-sealed and incubated in heating baths at different isomerization temperatures for 40 minutes. Temperatures of the heating baths are set at 1°C intervals over the range at which the enzyme shows the full range of inactivation (85°C to 100°C for the reference XI). One bath is set at the reference temperature approximately 10°C below the temperature of half maximum enzyme activity. For the blank samples, the buffer solution is mixed with the glucose substrate in a 1:1 ratio and distributed into 100 yl thin-walled glass micropipettes which are sealed and incubated in water baths at the same temperatures and incubation time as used for the enzyme test samples. The reaction is terminated by immersing the micropipettes in a salt-ice bath, and 65 yl of isomerase is removed from each micropipette. 2 yl of IN HC1 is added to stop the isomerase reaction. The isomerase is assayed for fructose and glucose by HPLC using a Beckman Liquid Chromatograph as described in Example II. 225798 The HPLC results are calculated as the degree of isomerization (It) at each temperature as follows: IT ■ F F + G F = (% fructose in sample - % fructose in blank) G = (% glucose in sample) The percent dry basis fructose data is used to calculate the temperature at which the half-life is 20 minutes [T20] as a measure of thermostability, as follows.

The activity function (L-t) at each temperature is calculated as follows: LT = In ( Ie ) where Ie = equilibrium degree of isomerization at each temperature (Figure 7).

The relative activity (Ar) at each of the test temperature is calculated as follows: Aj. = (Lfp/L^) X (k£^./k£iji) L^ = activity function at test temperature Lr = activity function at reference temperature The best reference temperature is about 10°C below the temperature at which 50% of maximum activity is displayed. k£T = isomerization rate constant calculated at the test temperature (see formula below) K&SmRr-"-' :1 "• rvr-;„ >r - ■••. /•~n f • Ci 2257H8 = isomerization rate constant calculated at the reference temperature (see formula below) r^, kf ■ exp (-6654/(T + 273) + 15.957) where T is test (or reference) temperature in °C Relative activity is plotted vs. temperature as shown in ^ Figure 8. Relative activity is related to heating time and to enzyme half-life according to the following relationship: Ar = (1 - 0.5 0.693n where n = t T t = heating time (40 minutes) T = enzyme half-life (minutes) Graphs of relative activity vs. temperature supply information on half-life. Thus, interpolation as shown on the graph in Figure 8 identifies the temperature at which the half-life is 4, 20 or 40 minutes.

T40 - 96 • 0 0 n T20 = 97 .0°C T4 = 99 .0°C T2q is reported to the nearest 0.1°C as a standard expression of thermostability.

The T20 value, the temperature at which the reference or the mutein has a 20 minute half-life, is a sensitive measure of an enzyme's thermostability. Relative to the reference XI thermostabilized muteins should retain more catalytic activity at elevated temperatures. Consequently, a thermostabilized -104- x o\ '' •V : 1 ^ OCT1988£ .... h mutein will have a larger T20 va^ue (it will demonstrate a 20 minute half-life at a temperature at least 1°C higher than the reference enzyme). Quantitation of enzymatic activity to assess thermostability has the advantage of testing both the reversible (conformational) and irreversible (conformational plus chemical) thermoinactivat ion mechanisms. Many point mutations have been shown to result in muteins that are 2°C-5°C more stable than the parent enzyme. Because the precision of the T20 test can be as low as + 0.1°C, thermostable xylose isomerase muteins are clearly identified by the T2q method.

EXAMPLE XV Determination of pH Optima of Xylose Isomerase Muteins The pH optima of the xylose isomerase muteins produced as described in Example XII above, is determined by studying activity of the enzymes under various pH conditions as follows.

The reference XI or XI mutein is produced and purified as described in Example II. The purified protein is dialyzed against distilled water after which the protein concentration is adjusted to 0.3 mg/ml.

Substrate solutions containing 2 mM magnesium chloride, 40% glucose and 20 mM buffer are adjusted to the desired pH at 60°C. Depending upon the pH range to be studied, an appropriate buffer is selected from the group; sodium phosphate (pH 6.0-8.0), sodium bisulfite (pH 6.0-8.0), N,N-bis(2-hydroxy-ethyl) glycine (pH 7.3-9.3), 3-(N-morpholino) propanesulfonic acid (pH 6.2-8.2) .

To determine the XI activity at the selected pH, a 50 yl aliquot of dialyzed enzyme is mixed with a 50 pi aliquot of l3ocrj% 225798 buffered substrate, the mixture is placed in a 250 yl tube, the tube is sealed and incubated at 60°C for 40 minutes. Reactions are stopped by the addition of 5 ul of 1 N hydrochloric acid. Samples are then assayed for fructose production by HPLC assay as described in Example II.

The HPLC results are expressed as percent relative activity. That reaction pH which gives the greatest conversion of glucose to fructose is arbitrarily assigned as having 100% relative activity. The activity observed at all other pH values is expressed as a percent of the maximum activity.

Graphs of relative activity vs. pH are referred to as pH vs. activity profiles and indicate the pH optimum, under defined reaction conditions (e.g. ionic strength and temperature), for the enzyme being tested. For the reference XI or XI muteins, relative activity plotted vs. pH indicates pH optima obtained for the reference XI as shown in Figure 9.

Relative to the reference XI, the muteins should display greater catalytic activity at a lowered pH.

The recombinantly produced S. rubiqinosus xylose isomerase and muteins set forth herein may be used to convert glucose to fructose or xylose to xylulose in various industrial processes. The various muteins may be resistant to various inactivation reactions and more stable, under extreme conditions of temperature and pH, than native XI. In addition, K catmay be increased, K^nay be decreased, Kcat may be decreased and/or Kpmay be increased. Furthermore, the pH optimum of the muteins may be reduced.

Deposits On August 5, 1987, Applicants deposited with the American Type Culture Collection, Rockville, MD, USA (ATCC) the mutein -106- f . cTi98S,. p 225798 expression vector pTWll in EL coli DG101 XI- accession no. 67,489. This deposit was made under the provisions of the Budapest Treaty on the International Recognition of the Deposit of Microorganisms for the Purpose of Patent Procedure and the Regulations thereunder (Budapest Treaty). This assures maintenance of a viable culture for 30 years from the date of deposit. The deposit will be made available by ATCC under the terms of the Budapest Treaty, and subject to an agreement between Applicants and ATCC which assures permanent and unrestricted availability upon issuance of the pertinent US patent. The Assignee herein agrees that if the culture on deposit die or is lost or destroyed when cultivated under suitable conditions, it will be promptly replaced upon notification with a viable specimen of the same culture. Availability of the deposit is not to be construed as a license to practice under the authority of any government in accordance with its patent laws.

This deposit was made for the convenience of the relevant public and does not constitute an admission that a written description would not be sufficient to permit practice of the invention or an intention to limit the invention to these specific constructs. Set forth hereinabove is a complete written description enabling a practitioner of ordinary skill to duplicate the construct deposited and to construct alternative forms of DNA, or organisms containing it, which permit the practice of the invention as claimed.

As will be apparent to those skilled in the art in which the invention is addressed, the present invention may be embodied in forms other than those specifically disclosed above without departing from the spirit or essential characteristics of the invention. The particular embodiments of the present invention described above, are, therefore, to be considered in all respects as illustrative and not restrictive. rs ' 225798 The scope of the present invention is as set forth in the appended claims rather than being limited to the examples contained in the foregoing description. ■1 -i \ j j < ' ' . o a * 3 OCTi988 >;v?.

Claims (18)

^25798 (a WHAT l/WE CLAM IS:

1. A method for increasing the stability of a protein comprising substituting an amino acid at a preselected substitution site in the protein, said substitution site having phi and psi backbone conformational angles in the range of phi = -40° to -90° when psi = 0° to -60°, or in the range of phi = -40° to -95° when psi = 120° to 180° and capable of accomodating said amino acid without disruption of the three-~j " dimensional structure of the protein such that introduction of | said amino acid decreases the configurational entropy of j unfolding of said protein. j

2. The method of Claim 1 wherein said preselected substitu- I tion site is any amino acid residue except proline and the i I amino acid introduced at said site is proline, and said method | further comprises the step of determining the phi and psi | values of the amino acid residue in the amino acid sequence of I the protein immediately preceding the side of said proline | substitution, such that if the psi value of the preceding I amino acid residue is between 0° and -90° then the substitu- | tion site must have phi and psi values in the range of phi | ^ = -40° to -90° when psi = 0° to -60°, but if the psi value of j u the preceding amino acid residue is not between 0° and -90° S then the substitution site may have phi and psi values either | in the range of phi = -40° to -90° when psi = 0° to -60°, or j in the range of phi = -40° to -95° when psi = 120° to 180°. \ j

| 3. The method of Claim 1 wherein said preselected substitu- I tion site is a glycine amino acid residue and the amino acid I introduced is any amino acid having a 6 carbon atom or I branched 8 carbon atom.

| 4. A method for increasing the stability of a protein i comprising substituting a glycine amino acid residue having a -109- 225791 negative phi angle with an alanine to decrease the configura-tional entropy of unfolding of the protein.

5. A method for selecting substitution sites suitable for introduction of amino acids in a protein such that introduction of said amino acids increases the stability of the protein, comprising the steps of: a) determining from the crystallographic structure of a protein the backbone conformational angles phi and psi of said protein; b) screening said phi and psi angles determined in step a) to identify potential substitution sites in said protein having conformational phi and psi angles in the range of phi = -40° to -90° when psi = 0° to -60°, or in the range of phi = -40° to -95° when psi = 120° to 180° for introduction of said amino acids; and c) examining a structural model of the protein to determine from the potential substitution sites identified in step b) substitution sites that will accomodate substitution of an amino acid without disruption of the three-dimensional structure of the protein, whereby substitution of said substitution site results in a decrease in the configurational entropy of unfolding of the protein.

6. The method of Claim 5 wherein the amino acid to be substituted into said substitution site is proline, and the step of screening of step b) comprises the additional substep of determining whether the amino acid residue preceding the potential substitution site identified in step b) has psi angles between 0° and -90°, and if so then the step c) of examining comprises the substep of determining a substitution site having phi and psi angles in the range phi = -40° to -90° when psi = 0° to -60°. -110- 1 OCT 1983 r\ 2257,98

7. Streptomvces rubiqinosus, (S. rubiqinosus). xylose isomerase mutein having a change in at least one position in the native amino acid sequence at a position equivalent to a native amino acid residue selected from the group consisting of Lysinel83' Lysine28g, Histidine54, Histidine22Q, Methionine223» Arginine140, Tryptophanlg, Tryptophan137, Phenylalanineg4, Glycine^g, Glycine^gg, Glycine^^, Glycine219, Glycine231, Glycine248, Glycine2gg, G1ycine305' Glycine^gg, Leucine-^, Alanine2g, Alanine33, Asparag ine-^-^, Arginine2_ggr Valine^g^/ Glycine^gg, Leucine^g2, Glutamic acid2Q7, Arginine2gg, Threonine342, Arginine354, Aspartic acid28, Arginine32, Serine64, Valine218, Arginine2g2, Isoleucine252, Aspartic acidg, Glutamine^, Glutamic acid38, Leucine4g, Aspartic acidgg, Leucine58, Valine127, Threonine]^, Alanine^g, Argininei^y, Isoleucinei8Q, Leucine^^, Leucine2^1, Asparagine227, Glutamine234, Alanine238, Leucine246, Arginine284, Arginine3Q8, Leucine3^, Arginine3, Leucine33j, valine3g2, methionine37Q, leucine37g, leucine383, Asparagineg2, Asparagine185, Glutamine2gg, Asparagine3Qg, Glutamine377, Tryptophan27Q, Phenylalanine32o> Histidine382, Glutamic acid337, Glutamic acid^44, Glycine2gi, Glycine22g, Alanine3gg, Valineg8, Glutamine24g, Aspartic acid-^3, Aspartic acid 57, Glutamic acid ^gg; Glutamic acid 14^, Glutamic acid221, Aspartic acid287; and Aspartic acid34gr the mutations being introduced to increase the stability of a protein.

8. The S. rubiqinosus xylose isomerase mutein of Claim 7 wherein the change is in the lysine amino acid residue equivalent to Lys183 and said change is substitution by an amino acid selected from the group consisting of Arg, Gin, Asn, Asp, Glu, Ser, Thr, His, Tyr, Ala, Val, J I j fetT and lie; or 2257 the change is in the lysine amino acid residue equivalent to Lysine289 an<* sa^ change is substitution by an amino acid selected from the group consisting of Arg, Gin, Asn, Asp, Glu, Ser Thr, His, Tyr, Ala, Val, Leu and He; or the change is in the histidine amino acid residue equivalent to His54 and said change is substitution by an amino acid selected from the group consisting of Gin, Glu, Asn, Asp, Ser, Thr, Ala, Val, and Tyr; or the change is in the histidine amino acid residue equivalent to His220 an^ said change is substitution by an amino acid selected from the group consisting of Gin, Glu, Asn, Asp, Ser, Thr, Ala, Val, and Tyr; or the change is in the methionine amino acid residue equivalent to Met223 and said change is subsitut ion by an amino acid selected from the group consisting of Gly, Ala, Val, Leu, lie, Phe, Tyr, Gin, and Asn; or the change is in the arginine amino acid residue equivalent to Arg^Q and said change is substitution by an amino acid selected from the group consisting of Gin, Asn, Glu, Asp, lie, Leu, Ala, Val, and Tyr; or the change is in the tryptophan amino acid residue equivalent to Trpig and said change is substitution by an amino acid selected from the group consisting of Asn, Gin, Ser, Thr, Gly, Ala, Val, Leu, lie, Tyr, Phe, and His; or the change is in the tryptophan amino acid residue equivalent to Trpjgy and said change is substitution by an amino acid selected from the group consisting of Asn, Gin, Ser, Thr, Gly, Ala, Val, Leu, lie, Tyr, Phe, and His; or -112- 0 2 j 7 ? 8 the change is in the phenylalanine amino acid residue equivalent to Phe94 and said change is substitution by an amino acid selected from the group consisting of Thr, Ser, f*) His, Val, Gly, Ala, lie, Leu, Asn, and Gin; or the change is substitution of the glycine amino acid residue equivalent to Glyx where x is selected from the group consisting of residues 146, 166, 197, 219, 231, 248, 298, 305 and 369, and said Gly substituted with an amino acid other than glycine; or the change is substitution by proline in the amino acid residue equivalent to an amino acid residue selected from the group consisting of Leu^g, Asp2g, Ala2g, Arg^/ A^a33# Serg4, Asn107' Ar9lQ9' ^^-^146' Va^151' ^^189' Leu192' G^u207' Val2ig, Ile252' Ar9259' Ar9292' Thr342' Ar9354' G1y369' Argi77, and Asp345; or the change is double substitutions of cysteine in the amino acid residues equivalent to pairs of amino acid residues selected from the group consisting of Trp27Q and Gly-^g, Phe320 and ^i^2Q2f G^*^337 and Arg^Qg, G^yi8g ®nd GIU244, Gly251 and Gly225» Ala366 and Val98' Gln249 and G1y219' and/or ^ GIU207 and Aspig3; or the change is substitution by tyrosine in the amino acid residues equivalent to an amino acid residue selected from the group consisting of Aspg, Gln2^, Ala2g, Arg32» Glu3g, Leu^g, ^ AsP56' Leu58' Val127' Thr133' Ala136' Ar9i77/ Ile180' Leu193' ^ Leu211' Asn227' Gln234' Ala238' Leu246' Ar^284' Ar9308' Leu311' Ar9316' Leu335' Val362' Met370' Leu375 and Leu383J or the change is substitution by phenylalanine in the amino acid residue equivalent to an amino acid residue selected from the group consisting of Leu^g, Aspgg, Leugg, Thr^33, Alai3g/ Ile180' Leu193' Leu211' Asn227' Gln234' Ala/23S.' Leu246' Leu311' Leu335' Val362' Met370' Leu375 and Leu383; or -113- the change is substitution by tryptophan in the amino acid residue equivalent to Asn227 ; or the change is substitution by an amino acid residue selected from the group consisting of Ala, Val, Leu, lie, Ser, Thr, His, Tyr, Lys, Arg, Met and Pro in the amino acid residue equivalent to an amino acid residue selected from the group consisting of Gln2i, Asng2» ^sn^Qy, Asn-^gg, Asn227» G^n234» Gln25g, Asn30g, and Gln377; or the change is in the aspartic acid amino acid residue equivalent to Asp^y and said change is substitution by an amino acid selected from the group consisting of Lys, Arg, Gly, Ala, Gin, Asn, Thr and Ser; or the change is in the glutamic acid amino acid residue equivalent to Glu186 and said change is substitution by an amino acid selected from the group consisting of Lys, Arg, Gly, Ala, Gin, Asn, Thr and Ser; or the change is substitution of the aspartic acid amino acid residue equivalent to Asp^y and said substitution is with an amino acid other than aspartic acid or glutamic acid; or the change is substitution in the glutamic acid amino acid residue equivalent to Glu18g and said change is substitution by an amino acid other than aspartic acid or glutamic acid; or the change is substitution by glutamine in the glutamic acid amino acid residue equivalent to GIU221; or the change is substitution by glutamine in the glutamic acid amino acid residue equivalent to Glu^i.

9. A nucleic acid encoding the xylose isomerase of Claim 7 or 8 said nucleic acid being substantially free of nucleic acid that does not encode the xylose isomerase of Claim 7 or 8.

10. An expression vector for mutant procaryotic xylose isomerase which comprises the nucleic acid of Claim 9 operably linked to control sequences compatible with a host cell.

11. A method for enhancing the conversion of glucose to fructose and xylose to xylulose which comprises exposing an effective amount of the xylose isomerase mutein of Claim 7 or 8 to glucose and xylose, respectively.

12. The xylose isomerase mutein of Claim 7 wherein the expressed xylose isomerase exhibits a change in one or more of the characteristics of chemical stability, k__.c, k , Kc, KD, CSlj Cut O Jr temperature stability, specific activity and a lowered pH optimum of the isomerase, as compared to the reference xylose isomerase.

13. A method for increasing the stability of a protein as claimed in any one of claims 1 to 4 substantially as herein described with reference to the Examples and/or the accompanying drawings. I

14. A method for selecting substitution sites suitable for introduction of amino acids in a protein such that introduction of said amino acids increases the stability of the protein as claimed in either claim 5 or claim 6 substantially as herein described with reference to the Examples and/or the accompanying drawings.

15. Streptomvces rubiqinosus, (S. rubiqinosus), xylose isomerase mutein as claimed in any one of claims 7, 8 or 12 substantially as herein described with reference to the Examples and/or the accompanying drawings.

16. A nucleic acid as claimed in claim 9 substantially as herein described with reference to the Examples and/or the accompanying drawings. '0 22579? -116-

17. An expression vector for mutant procaryotic xylose isomerase as claimed in claim 10 substantially as herein described with reference to o the Examples and/or the accompanying drawings.

18. A method for enhancing the conversion of glucose to fructose and xylose to xylulose as claimed in claim 11 substantially as herein described with reference to the Examples and/or the accompanying drawings. o jated this u octoe>e<? A. J. PARK & SON per: a <-0-3 AGENTS FOR THE APPLICANTS <o