CA2192091A1 - Heat-stable prolylendopeptidase - Google Patents

Abstract

The present invention relates to the field of biotechnology and concerns heat-stable prolylendopeptidase, recombinant DNA coding for heat-stable prolylendopeptidase, and processes for the production of heat-stable prolylendopeptidase and of recombinant DNA coding therefor, a host transformed with said recombinant DNA and a process for the production of said transformed host.

Description

~ W096/00293 2 1 9209 ~

Heat-stable prolylendopept~dase The present invention relatcs to the field of l ~; It~ n~y and concerns heat-stable ~,lUI~l rl.1. ~I.vlJl;~io~P~ n~ .; DNA coding for heat-stable ~-ulyl~ .."1. .l,. I.l;.i~., and ~ processes for the production of heat-stable prolylPn~loprrtiflo~P and of rccnmhinonr DNA
coding therefor, a host ~ l with said . c~ DNA and a process for the production of said m r " ".. ~l host.

BACKGROUND OF TE~ INVENTION _ _ _ E~ulyl~ was first found in human uoerus as a specific Pn~inrcFti~o~P which cleaves a peptide at the carboxyl-terminal side of a proline residue. The unique subsrrate selectivity of the enzyme drew much attention to the study of its physiological functions.
An ~n~inrert~ cp that shows the same substrate specificity as m~mmolion ~lulyb .,.1. 'l" ln;~l ~. was found also in a bacterium, navubd~lcliulll ,.,..,;.,~,.,~.l;......
This finding made prolylendopeptidase commercially available and enabled ils use as a bionhPmi~l reagent for the specific cleavage of peptides. The preparation of ~Ivlyl~ . from F, mPnin~nSP~tirl~m, however, has the following two crucial drawbacks arising from the bacterium. The bacterium is pathogenic and it produces not only ~lulyl~ 7~P but also significant amounts of other specific or non-specific peptidases. The commercial preparations are thus ~ lr.1 with significant amounts of trypsin and ~lllhlu~ Lida~c, which fact severely diminishes the utility of the cnmmPrciol products as specific l,;n. l,...,.;..~l reagent.

Prolyll~n~inrepliriocP catalyzs selective hydrolytic cleavage of peptides at the C-terminal side of a proline residue under physiological conditions. The enzyme can also catalyze the coupling of peptide fragments by cnn~lPn~otinn or n .~ ;nl n depending on reaction conditions and substrates. In the production of ~ l= c ~ lly active peptides~
prolyl. .,ni"~. ~.u,l~P can thus be used to catalyze i) selective cleavage of precursor peptides in order to liberate the ~ ally active peptide, ii) in vitro l l .U l; r;. 1 ;n"
of peptides including amidation of C-termini and iii) coupling of peptides. The term peptide used herein shall not indicate that only short peptides are meant but that the molecules in question are composed of amino acids linked via peptide bonds. Peptides may be short peptides, nlignrpptirips or polypeptides.

Of the three reactions mentioned above, the former two are especially important for dUWl~ lC/~lll processing in production processes of the peptides with rPrnmh;~o~lr DNA
technology. The r~Prnmhin:lnr peptides are often expressed in the form of a precursor or wo 96l00293 r~ c : . ~

2 l 92 09 1 -2-fusion protein, which is then subjected to in vitro processing for the conversion to active or mature forms~ I~uly~ y~ . is for example useful for C-terminal amidation of biologically active peptides such as ACI~I, .,Lul~luki.l, calcitonin, endorphin, insulin, LH-RH, oxytocin and v~uy~ iu, or the like. The f ~ substrate specificity of ylul~ , makes it very useful for the cleavage of the precursors at specific sites and the in vitro . - - .1. .'i~ of peptides without side reactions that are often associated with non-specific peptidases.

However, ylul~ ri. ~y ~ r is quite susceptible to inhibition or i..f~livaliull by conditions usually applied in peptidase catalyzed reactions. It is susceptible to denaturing and/or I ' ' " g agents in buffers used for cleavage of precursor peptides or also to conditions commonly used in peptidase catalyzed coupling reactions in order to make the formation of coupling products favourable over hydrolysis, e.g. the presence of high of organic solvents such as l,~dioxane, DMF or DMSO, extremes of pH
andior high t ~ Therefore, it is desirable to improve the stability of prolyl- .~I..l.~,l,li.l- . in order to make it more versatile as a catalyst for the industrial production of peptides.

OB lECT OF THE II~VENTION
It is an object of the present invention to provide ylulyl ~ , which are more heat-stable than the c- .. ~ r~y~ 1 li e wild type enzymes. Such a ylul~ is hereinafter named "heat-stable ylul,~

A hcat-stable ylul~l~ n.l. ~y~ is y~u li~ukuly useful for an industrial production process of peptides, because its superior stability prolongs the life of the enzyme in the catalytic reactions and thus improves a total turnover of the ~. lu~ It is useful as a stable and selective catalyst for production of the biologically active peptides that contain proline residues. In other words, a heat-stable yl~Jlyl~ ..1. 'lJ~ IJI ;fi~f- can be used in much lower ratio to the substrates than that of the wild-type enzyme, dccreasing costs of the catalyst which is often a c~itical factor for I . ", ,..., . ~1 feasibility of the production process. The high heat-stability of heat-stable ylulyl- ~1 ~lJ ylifl~ of the invention aiso enables the use thereof under the severe conditions that improve . rri- ;. .~; ~ yields or Cul~ ;ù~lS of the catalytic reactions, e.g., higher reaction i . , extreme pH or the presence of a high ~ . . . ~ . ,-l i. ~ of organic solvent. A heat-stable ylul~
created in the present invention is also more resistant to other for ns of protein fiPn:m~r~hf\n i.e. to other physical shress than heat hlaulivalion and more resistant to ~ W096/00293 2 1 q209 t ~-, s .~. ~

tteatment with chemicals. Thus, it is more-stable than the ~ g wild type enzyme in solutions conuining organic solvents, denaturing agents or extreme pH.

Another object is to provide a method for the generation of heat-suble ~)lol,yll . ..1. ~1A 1~ ;.1 A - sUrling from DNA coding for a wild-type enyme and a method for the ihll~lu .. of heat-stability by a "molecular evolution" method c , " multiple cycles of ~ and screening.

A further object of the present invention is to provide ..~ DNA coding forheat-suble ~lvlyl - -~1. q.~ a process for the production of such ' DNA, a host n A .~r.. cl with such ' DNA, and a process for the production of a heat-stable ~-ulJl~ by means of a 1. ~ ~ r ." ~rl host.

DESCRIlYrlON OF THE INYENTION

T~ ~ the term ~,vlyl~ 1~ q~ lJ~ is intended to include any l~ ulyl~ P ~
i.e. any serine protease that catalyzes hydrolytic cleavage of peptides specifically at the carboxyl-terminus side of proline residues. The EC number of such a vly lJ ~ ¦ is 3 .4.2 1 .26.

The preferred meaning of the term is a ~vlyl~ derived from l)l uLuyvL~
preferably from lla~ul,z~L.iu.,~ spec., ~more preferably from F. ., - jr.~ nj ~ most prpfl~n~nnAlly from F. . . ~ r~ l; strain IFO 12535 (ATCC 13253~.

In order to obtain a heat-suble ~)-U~ l q~ of the present invention, a random of a w~ld-ty~e ylulyl~ q~ A - znd subsequent screening for heat-stable ulyl. . l- .IJ~ can be performed. For this purpose, for example, a wild type ~VI~ L ~ gene can be cloned and ~ ' i, .1 A key for a successful , ~.-...g approach is the l I~A~ of gocd libraries of the mutated ~)-UIY1~ q~ genes znd the use of an effcient screening method for the heat stable ~ulyl~ A A For efficient screening for a clone producing a heat-stable yl~ it is essential to use an assay method that can identify a colony of the desired clone among thousands of colonies c~ ;"g a library of the mutants, namely a colony assay method. A colony assay for IJ-ulyl 1 ~ ', however, was not known.
Therefore, in the present invention an example of a sensitive colony assay was developed which can be used for screening of heat-stable ~lulyh,.ldu~ c 3 r~

Prcparation of the startinf~ DNA for ~ ' - r~ " '1' " ~ ._ . _,, , ,. _ ~_ ,, . ." ,,, There are DNA sequences known m thec art which code for ~.ulyl 1 ~ ;fl~, and which can thus be used as starting material for the present invention. However, the skilled worker is not limited to the already known ,ulu~ u~ sequences or to an available DNA ~ It is also possible to prepare starting material for the present invention, i.e. DNA coding for IJ ul.~l lul~ which is used for the IJlc~ iu~l of a mutated DNA coding for a heat-stable JJ ulyl- .A..l.~ ;A_~ of the invention, from any cell or tissue comprising a ~ yl~ gene or, in the case of a eukaryotic cell or rissue, preferably from any cell or tissue expressing the IJ~UIyl~ ~A~ gene. In the case a eukaryûtic DNA, preferably a ~ DNA, for example from uterus or brain, e.g., of human or porcine origin, is chosen, the DNA is preferentially an intron-free cDNA. Preferred starting DNA for the present invention is derived from a bacterium belonging to the genus Il,.v~ in particular F. " ; J .u,~ ul;~ u . most 1" ~ f .., " 1 ;Ally F. strain IFO 12535 (ATCC 13253). The most preferrcd starting DNA is shown im SEQ ID NO. 1.

A starting DNA for the present invention can be prepared by different methods. Since DNA derived from a prokaryote does not include introns, a genomic DNA coding for a ~lulyl. . ,.l~ ,l ;A ~ can be used as starting DNA if a JJ~uh~l yu~iu cell is chosen as source.
In this case, bacteria which contain a ~ul,~ A~ gene, for example F.
r U~ n can be P ~ i and a whole genomiC DNA can be extracted according to a ~,UII v~,~ltiullal procedure. The extracted DNA can then be digested completely or partially with an ~ ' rest~iction enzyme such as Bgl II, Eco RI, Hinc II, Hind m, Pst I or Bam HI. The digestion product can preferably be subjected to preparative ~ ulJllul~,~is with low melting-point agarose gel to enrich DNA fractions of a certain length m order to emich DNA fragments encoding ~lulyl 1~ ;f1A Next, the DNA fragments may be cloned into a suitable cloning vector. The cloning vector may be derived from any vector useful in the art of genetic ... ,~;. . . ;"~, such as from viruses, phages, cosmids, plasmids om,luulllv~ullldl DNA, for example derivatives of SV40, Herpes viruses, Papilloma viruses, Retroviruses, B_.,uluvMl~, phage ~, e.g., NM989 or EMBL4, orphage M13, bacterial plasmids, e.g. pBR322, pUC18, pSF_124, pBR317 or pPLMu., or yeast plasmids, e.g. yeast 2,u plasmid, or also ~ Ulllosulllal DNA c~prising an origin of replication or an: 'y replicating sequence (ARS). Preferably, the cloningvectorisabacterialvectorsuchaspBR322 pUC18,pUCl9,pUC118,pUC119, pKK 223-3 or the like.

~ W096100293 2 1 9~0q ~

~ Al ~ a cDNA library may be prepared from a cell expressing ~Ulyb ~ e.g. from a bacterial cell such as preferably from a bacterium belonging to the genus rl~. t - in particular F. ~ most f~ 'ly F. ~ strain IFO 12535 (ATCC 13253), or from a eukaryotic cell or tissue, e.g. ' cell or tissue, which pt~duces ~-vlyl~ ...1. ,l.,,lJl;.lA -, such as uterus or brain. For example, RNA is extracted from uterus or brain and emiched for mRNA according to a ~..~, ' procedure. Next, a cDNA library may be ~ . .- - ~ . t~ ~l according to a ~:u..~, ~ procedure.

A variety of metbods are known in the alt for the; ~ " of double-stranded cDNA
or genomic DNA into an .~ , - vector. For example, ~ A y l~u~v~ul~
tracts may be added to the double-stranded DNA and the vector DNA by incubation in the presence of the .,Vll~ g ~v~-y ' ' 1 ~ l t, and an enzyme such as terminal ~v~.yll...,l~,vlidyl rrPncfi~rAcr The vector and double-stranded DNA are then joined by base pairing between the ~ A ~ y IIV1llUAJVI~ ;C tails and finallyligated by specific joining enzymes such as ligases. Other p~ ;., are the addition of synthetic linkers to the termini of the double-stranded DNA, or the ;~ of the dvuble-stranded DNA into the vector by blunt- or staggered-end hgation.

Screening of the genomic DNA library or cDNA library is preferably achieved using a DNA hylJl;d;~lLiull probe. Suitable DNA probes are DNAs of known nucleotide sequence consisting of at least 17, ~. 1/.,1 i.l~ ~, for example syntbetic DNAs, cDNAs derived from mRNA coding for ~!~ul,~ , or genomic DNA fragments comprising e.g.
adjacent DNA sequences which are isolated from a natural source or from a genetically engineered ~n;~

To design synthetic DNA probes for screening the above-mentioned genomic DNA library or cDNA library, ~!lulyl l~ l ;AA ~ for which a DNA coding region is to be cloned may be purified, and its partial amino acid sequence is ~lrrrrminrA according to a ~.U..~ iUll procedure. Next, DNA sequences are designed on the basis of the parrial amino acid sequence thus Artr~TninrA Where an exact nucleotide sequence coding for the amino acid sequence is not known, a .~ of nucleotide sequences which partially or totally cover possible nucleotide sequences present due to the ~ ..y of genetic codon may be used. Alternatively, the third nucleotide in a codon may be replaced with inosine WO 96/00293 I ~ ~,5.'1 : .

Synthetic DNA probes can be D,yllL~ DiL~l according to lcnown methods, for example by stepwise c~ ;. ~ using the solid phase l ~ u . ~ , phosphiu triesur or method, e.g., the ~ ~, of l~ 1 ;/lr couplmg units by the method For Lyl~ the DNA probes are labelled, e.g. radioactiveiy labelled by the well Icnown kinase reaction. The hyll~i li~liv~ is performed according to Icnown I ' i.e., in buffer and salt solutions containing adjuncts, e.g. calcium chelators, viscosity regulating c~ ~ ~i v~ prouins, non '~ ln~ -DNA and the like, at ~ - r~
favoring selecive 1.~ ;. ., e.g., between O~C and 80~C, for example between 25~C
and 50~C
~n order to obtain a preferred staning DNA of the present inventio, a DNA libraly of F.
can be used to transform an Al~l v l" ~ ~ r host such as E. coli cells, which are then plaud and cultured on a solid medium to develop colonies, and positive clones can be selected by a colony l~v-iL~Liu-- method using the above-mentioned DNA
probes. The i ' of ;tlJ~JlVIJli.lh, host cells with the DNA library and theselection and r l~ltiplirs~rir n of ", . .~ r ~ host cells are well known in the art.

The nucleotide sequence of DNA selecud as described above can be deurmined by methods known per se; for example, by the Maxam-Gilbert method using end-labelled DNA or by the dideoxy chain ~.. ; - ~ ;r.. method of Sanger.

Once a nucleotide sequence coding for, or an amino acid sequence of, ~ul,yl. 1~
is APtPrrninPA,DNA coding for the enzyme can also be prepared by methods leadingdirectly the desired DNA such as conventional cloning or PCR procedures or e.g. an in vitro synthesis according to ~:v..~.liv..~ methods. In an in vitro synthesis method suitably protected ' ' are linked with one another e.g. by the ~ l;, method, the more efficient l~ method, the phophite triester method or ~IIWDlJllU~ iLlC
method. ~ ;- ., . of the synthesis of the ol;~ U~ and pb~ vLl~D is made possible by the solid phase method, in which the nucleotide chains are bound to a suitable polymer. The known synthesis techniques allow the preparation of pvl~ lwLi~D up to 120 bases im length, in good yield, high purity and in a relatively short time. The actual double-stranded DNA is built up ~ ,.lly from chemically prepared u . . ',, g oli~. . l .1 ;-l- c from both DNA strands, which are held together in the correct A- 1. ,1;.. ~ ''I by base-pairing and are then chemically linked by the enzyrne ~ woss/002s3 21 920~ r~l~

DNA ligase. Another possibility comprises incubating u . _~lrl~Jhlg single ~li ' ' from the two DNA strands in the presence of the four required d~iJ~ ,kODidC i 'I ' , ' ' with a DNA PCil~lln~1UD~ for example DNA pc l~ 1, the Klenow ftagment of pOlyll~rlD~, I or T4DNA pol~ se, or with AMV (avian vitus) reverse i , The two ..l;c. l. .~ are thereby held together in the correct ,, by base-pairing and are ,, ~ ... ,t ~1 with the required ' ' by the enzyme to give a complete double-stranded DNA.

M; ~ ;U and Screenin~
the present imvention an expression plasmid . l _ a ylulyl. . lo~ r.
expression cassette is used for the generation of heat-stable l,lUI,,~ by random _ Moreover, as is explained hereinafter in more detail, the presentinvention also concerns a method c~ e multiple cycles of . ~ ;.... which can beusedforthepreparationofimprovedheat-stablel,-ulyl~ L.l,.l.l;.l ~,. A~ '' '!~.
the present invention concerns a method for the ~ r~ of a rJ ~ DNA
molecule coding for heat-stable l,lulyl. 1. .l . ~ said method ~ , _ (a) ~ of a starting DNA coding for a ~lulyl- l- ~ whereby the starting DNA may by coding for a wild-type ~lui,yl ~ or, if a second or further cycle of . l ~ is desired, a mutant DNA coding for a heat stable "parent"

(b) generation of a library of mutated DNA sequences obtained in (a), and (c) screening the library for a gene coding for a l~lulyl. ~ ;r Rc~- with improved hea~-stability, if compared to the ~ e wild-type enzyme or, after a second or further cycle of v to the c '~ wild-type enzyme or optionally to the ~ r 'T heat-stable "parent" J~lul~ ,I;rl ~ However, if the sequence of a heat-stable i~lulyl ~ is once known or if it is known, which amino acids positions can be mutated in order to improve heat stability, any - method suitable for the generaion of mutations or generation of a desired r. . ~
DNA sequence can be used. Therefore, the present invention is not limited only to DNA coding for heat-stable ~JIul~ IJ~ obtained by the above-mentioned process based on random . I ~ r ~ but includes also DNA
molecules of the invention obtained by aother method.

wo 96/00293 P_./...,~.~~ l .

If a ~,~uly~ gene is desired to be randomly ,. i the conditions for .1~ u . ..\ must be slightly adopted, for example as explained in the followmg, for each wild-type gene in order to obtain a suitable population of mutated clones in the library Usual conditions used for chemical ,, are reported in "A General Method for Saturaion T\,' of Cloned DNA Fragments" [Myers R.M. et al., Science 229:242-247(1985)] as follows:

Chemical mutagen Concentration Reaction time Reaction temp.

Nitrous acid 1 Ma 60 min room temp.
Formic acid 12 M 10 min room temp.
Hydrazine 60% 10 min room temp.

a1 M sodium nitrite in 250 mM sodium acetate, pH 4.3 The variable parameters of the conditions which have to be optimized for each gene or cDNA in order to obtain a suitable population of mutated claones in the library, are of each chemical mutagen, reation time and reaction ~ In principle, the parametcrs should be adjusted so as to introrluce a limitcd number of, i.e., a single or at most a few, base ~ in the ~luly~ gene. In the present invention it was found that the parameters could be adjusted cu~ .l.ly by monitoring a) a number of 1, ~ r .. - .~ clones in the library of mutant IJ~uly 1~ P~ 1~n~ that is obtained from a certain, low amount (e.g. 10 ng to I ~Lg) of the starting DNA and b) the ratio of the number of the clones expressing the active enzyme to the total number of the I ' The modified conditions adopted for the chemical . - - ~ ~L,... ~ ~ of the proly1....-i"l.. 1 .l ;.l ~.. gene are as follows:

Chemical mutagen Concentration Reaction time Reaction temp.

Nitrous acid 1 Ma 15-20 min 20 C
Formic acid 3 M 5-20 min 20 'C

~ W096/00~93 21 9209 I r~ .

_9 Hydrazine 20% 5-20 min 20 C

al M sodium nitrite in 250 mM sodium acetate, pH 4.3 There are several factors thaI are known to influence the fidelity of DNA synthesis by ThermuS aquaticus (Taq) DNA yul~ For each factor, the normal conditions allowing the yolyll~e to fimction with high fidelity and a typical nmge of the condition known to reduce fidelity of DNA synthesis are as follows [taken from Leumg et al., J.
Methods Cell Mol. Bio. 1:11-15(1989) and Eckert et aL, PCR (ed. McPherson et al.), 1991, pp. 225-244, IRL Press Oxford]:

Factor influenclng~ Normal condition Condition for fidelity reduced fidelity Temperature 25 - 80 'C
pH 8.3 5 - 8 Addition of Mn2+ o ~M 500 - 1000 ~M
Addition of DMSO 0 ~ 10 %
Higher conc. of Mg2+ 1.5 mM 2 - 10 mM
[dATP]/[dNTP] 1b 0.02 - 0.2C
[dGTP]/[dNTP] 1b 5 _ lod bConc. of dGTP, dTTP and dCTP are 0.2 mM
CConc. of dGTP, dTTP and dCTP are 1 mM
dConc. of dATP, dTTP and dCTP are 0.01 mM

For example, conditions of chemical treatments for random ~ are in the present irlvention optimized for the ylulyl 1~ ~Y~.1J~ gene cloned in the plasmid pUK-FPEP-b so as to introduce limited numbers of (a single or a few) base ' at random.
rOly chain reaction (PCR) is also used as a method to induce base ~ in the er~yme gene by the ..~ y~ of ' ' The effect of several factors decreasing the fidelity of the pol~ , such as the d~,v~.y~ l v~ pool imbalance and addition of MnCI2 or DMSO, is examined and reaction condiions are optimized for suitable r.~ c of base ,,.~ il. i.. ,c .

wo 96/00293 2 1 9 2 0 9 ~ r~l,~. . ~

In particular, the 2.3 kb EcoRI-Psd DNA fragment of pUK-FPEP-b encoding the wild-type ~,lvl~ gene is amplifled with a paiT of 25, ~ pruners, which are ' y to the regions ~ ~ ly flanking the EcoRI and PstI sites.

To intro~duce a limited number of, i.e., a single or at most a few, base ~ ;n .l;.... ~
efficiently in the .I~lulyl l~ gene, effects of several factors on the f~delity of the , arc examined. The factoTs ~ ~d involve addition of MnCI2 OT DMSO, alteration on MgCl2 . and the deoxy nucleotide pool imbalance. A basic reaction mixtme contains, 10 mM Tris HCI (pEI 8.3), 50 mM KCL 15 mM MgCl2, 0.01 %
(wlv) bovine serum albumin, 0.2 mM each of four dNTP's, 10 nglml of the templateplasmid~ M each primers and 25 units/ml Taq IJVIyl.~ ~;; The ~ of the reaction condition decreasing the fidelity of the pol~ , are as follows.

l) SO-lOO ~LM of MnC12 can be added and the c~ ;- of MgCI2 can be increased to 5 mM.

2) 5-20 9~ of DMSO can be added to the reaction mixture.

3) The ~.-.... n rl ;~ of MgC12 in the reaction mixture can be changed fTom S mM to 10 mM.

4) While the c. . u ,u ;. . ~ of dGTP, dTTP and dCTP are held constant at I mM, the ~ ... l . ,- l;.. of dATP can be lowered from 0.2 mM to 0.1 mM in the presence of S mM
Mga~.

S) While the ( ., ... ,, l ;....c of dATP, dCTP and dTTP are held constant at 0.1 mM, the ~.. ..u.,~;.... of dGTP can be varied from 0.5 mM to 1 mM in the presence of 1.5 - 2 mM
MgCI2.

The PCR can be carried out according to c~ ,.I;UIIal methods, e.g. by using GeneAmpTM PCR system 9600 (Perkin-Elmer). For example, the reactions can be ca~ried out by melting the template DNA at 94~C for 1 min and annealing with the primers at 50~C
for 1 min. Chain extension can for example be initiated at 70~C for 4 min and a total of 25 cycles are performed and, after the last cycle, the polylll.,.i~ iul, at 70~C can be extended for additional 7 min.

WO 96/00293 P~.l. ,~.'~ ' .
~ 2 1 9209 1 The tr6utated structure gene prepared by the chemical treatments or the PCR ~
can then be excised and recloned into the naive expression vector fragment according to ~,UI~V~ methods. With the mass of the ~ expression plasmid which typically consists of several tens of thousands of ~, ' clones, E. coli is ~ to give a library of mutated ~lul~ The ratio of the number of the clones expressing the active enzyme to the total number of the ~, r -~ is used as an in~icaion of ' , of the base However, any other expression vector suitable for the pteparation of a gene library which can be screened after the expression of the gene in quesion can be used for the generation of the gene library. Likewise, the method is not limited to the use of E. coli cells. Vectors and hosts which can be used are t ~ ' ' '1'' 1~' A I ' ' ' ,, A colony assay method for the activity of ~lulyl~ p ~ is used in order to develop the clones expressing heat-stable ~J~ulyl 1- .lJ~1a ,.1~. The assay comprises the steps (i) lysis of colonies on the filter, (ii) blocking of the filter and (iii) active staining of the enzyme. In a preferred ~ L ' t, the filter is a n;~rocr~ cr filter.

A method for the detection of ~lulyl -~ - p ~ , with Z-Gly-Pro-,B-~ l.yLullidc, Fast Garnet GBC and Triton X-100 has been known for a long time. In the present imvention the assay method is modified and changed into a procedure of an active staining by removing the surfactant Triton X-100 from the assay mixture. By this staining method colonies producing active l~lulyl 1- ~ L~ are visualized as red spots on the filter.

A screening system for heat-stable IJlUly~ p' 1~ is built up by the -- ' of the colony assay and heat-treatment. The heat-treatment is performed after the lysis of the colonies on the filter and then the heat-treated filter is subjected to the active staining.
Conditions of the heat-treatment are adjusted to allow selective staining of the colonies that produce mutants ~,r ~ 'y more stable than the wild-type enzyme. The clones selec~ively stained in the first screening can be isolated and re-screened with heat-~reatment at higher i , to be narrowed down to the most promising clone.

In the present invention clones are generated by the chemical ~
process and the PCR ." l ~ sJ~ .illg process. Heat ~to~ irc of two heat-stable ul~ PEP- l S and PEP-227 0btained by the chemical and PCR

~1 920ql ~

C~ ,ly, are ~u~u~u~.ly evaluated and compared with the wild-type enzyme in terms of first-order rate constants of heat u.~ uu" at high ~ The cvaluation has proven significant 1 u. in h~ iry of the mutant enz3~mes as is described in more detail in the Examples.

The whole nucleotide sequences of the structure genes of the heat-stable ul~ PEP-15 and PEP-227 are ~' ' and compared vith that of the wild-type enzyme, revealing point mutations which lead to amino acid residue ,~,1 ., "" ..t~ f for the ~ " I l~;l ly ~ -u-~

For further UII~JI U . _111.,1~1 of heat-stability of proly L l( ~ . a molccular evolutional approach can be taken: It has been clearly proven in the present invention that repeating the cycle of the ~ ~ and the screening is really effective to improve the y of ~,lulyl - L,l~ Thus a mutated gene coding for a heat-satble ~)IU~ JI~ can be isolated from the clone selected in the ~ /screeningprocess and it is again subjected to the ,, to prepare the second generation library of mutated l~lulyl~ The second generation library is then screened for a clone producimg a more heat-stable ~lul~b 1. .1 - ~ - This .~ ,. c:~l g cycle can be repeated to irnprove stability of ~lulyl~ as much as possible, mimicking the proccss of molecular evolution in vivo. The r=epeating I~.IC.~lUllg cycle allows selective ~ 5~ of base ~ (amino acid residue IC. ' ) that ~yu~ lly and/or additively stabilize the enzyme molecule, and it is not necessary to determine the mutation(s) at each generation.

The repeating ~ ..,luh~g cycle also permits _ ' of silent mutations (base ,, I ~ c) which does not alter the amino acid sequence. lt is well recognized that a single base ,~lb~ in a codon does not result in every possible arnino acid rcsidue r ' bccause of the degeneracy of the genetic code. Since double or triple base ~ 1 .~l ;l, .l ;. . - in a single codon hardly take place at the same time under regular conditions of ~ , IAg~ ~ random ~ f ;;~ is limited in the latitude of every arnino acid residue. The ~ of silent mutations in the repeating . ., . ~ s/screening cycles, however, helps to extend this limit of arnino acid . ' in further cycles of the " ~ /~uc~..l;l.g. Therefore, repeating the " 1l - ~;., S/~ ul.g cycle, i.e., a molecnlar evolution approach, is not only ~ur~;h~rul w~d but also more thorough in the latitude of possible arnino acid residue wo 96/00293 P~
~ - 21 92091 More heat-stable ~ A~ lJ~ ', PEP-361 and PEP ~'~07 are isolated from the clones selected in the second and third cycles of the " ~ ihlg~ . Their stabilities are y~auL~ ly evaluated and compared with those of the wild-type enzyme amd tbe heat-stable ~luly~ of the first generation. It is confirmed that the heat-stabi.ities of the heat-stable ~lul~ lJ~ are constantly improved as the' ' 1'~1;'"' -'~J:I~,IC.~ g cyclesproceed,provingtheefficacyofthemolecular.,v. 'approach.

The /1 ~ ;. ~ of the DNA sequences of the mutated genes encoding PEP-361 andPEP407 rovealed the: ' of base ~ . c (and resulting amino acid residue L f~ ) as sho vn in Table 3 (Example 8). It is verified that the obse ved constant ~ ~ inthe'.-- s ~s~ fifyclearlycoTelates vithcollectingmutationsinthe enzyme molecule; a single or double mutations at each cycle. Besides the mutations leading to the amino acid residue, c~ silent mutations during the repeats of the ... ~ ~ ,, cycles are also ~l~ 1 by the sequence analysis.

Like the starting DNA described above, also a DNA molecule coding for a heat-stable UI~ 1A can be prepared directly once the nucleotide or amino acid sequence is l~nowll, e.g. by the methods mentioned above for the production of the starting DNA.

Cbninf~ vectors and exPression vectors Hybrid vectors can be used for the l~ ul~ and - ~ of the starting DNA for . as well as for the cloning and JJ~ ~.l~iu.. of the mutated ~,.ul~l. --1~ ,~,. lJI i~lf ~..
DNA and for the production of heat-stable l,lul~ in a ' ' host.

The hybrid vectors can be derived frDm any vector useful in the art of genetic ~ ..,;;., . . ;..g such as from viruses, phages, cosmids, plasmids or .,1.1~ - I DNA, for example derivatives of SV40, Herpes-viruses, Papilloma viruses, Retroviruses, Baculovirus, phagel.,e.g.NM9890rEMBL4,orphageMl3,bacteriarplasmids,e.g.pBR322,pUCI8, pSF2124, pBR317 or pPLMu., or yeast plasmids, e.g yeast 211 plasmid, or also . 1 . ", "~". ,. . gl DNA li - an origin of replication or an ly replicating sequence (ARS), or a defective virus, phage or plasmid in the presence of a helper virus, phage or plasmid allowing replication of said defective virus, phage or plasmid, e.g.
M13(+)KS vector in presence of e.g. M1 3K07 helper phage The B~ulu vih ~;~c;~ which wo 96/00293 2 1 9 2 Q 9 1 r~

can be used in the present inveniton are, for example, Auto~rapha californica nuclear yolyhvv1u~;~ virus (AcMNPV), Tri~hr~ ci~ ni MNPV, 1~ ' . ' ou MNP~V~ Galleria mellonella MNPV, Bombvx mori nuclear pul~hvLu~i~ virus (BmNPV), and the like. A
kit comprising a ~ ~, h;, ~ , of an Auto~raDha californica nuclear pu~ u~;s virus and I - ~ ,vi. .,~ transfer vectors pAc700, pAc701, pAc702, pVL1392 and pVL1393 is c~ .. ,.. :-lly available from Invitrogen.

Suitable vectors are those which are operable in the microbial host cell chosen for ,i..g the hybrid vector or for the expression of heat-5table ylu~
Suitable vectors thus contain a complete replicon and a marker gene, which renders possible the selection and i.l. - ~ of the ...;~...,. ~I ~A..;~ll.C ~ rl~ by the expression plasmids by means of a phenotype feature. Vectors which by themselves do not fulfill all . for replication and/or expression but need helper plasmids which ~ missing functions may also be suitable Thus, hybrid vectors of the invention provide for replication of a desired y~ul~ lJ~ DNA in a suitable host, either as an i ' ' element or by imtegration in the host ~ ' wuu~uu~v Several possible vcctor systems are available for integraion and expression of the cloned DNA of the invention. In principle, all vectors which replicate and/or comprise a r~r~m~ - gene which can be expressed in the chosen host are suitable. The vector is selected depending on the host cells envisaged for r " ", ~ In general, such host cells may be ~uLuyuLi~, or eukaryotic micro-organisms such as bacterial fungi such as yeasts or 1;1~ fungi, or cells of higher eukaryotic origin such as animal, for example ~' or insect, cells. Suitable host cells will be discussed in detail I ' ' .. . In principle, the hybrid vectors of the inve~tion comprise a DNA encodmg l~luly~ " . an origin of replication or an ly replicating sequence, optionally dominant marker sequences, and, optionally, additional restricion sites.

An origin of replication or an ly replicating sequence (a DNA element which confers -- ~-",""..~ ly replicating r~p:~hiliti~ c to ~ c ~ ~l elements) is providcd either by cuu~l-uvlioll of the vector to include an ~ J"~ origin such as derived from Simiam virus (SV40) or another viral source, or by the host cell ~ .. . h ~ c A hybrid vector of the invention may contain selective markers depending on the host which is to be j r 1, selected and cloned. Any marker gene can be used which ~ W096/00293 21 92091 r~-~

..

facilitates hhe selechion of i ' due to hhe phenotypic expression of the marker.
Suitable markers are ~L_ ~ '.y genes fDm which a ~rvly~ JLide can be expressed which provides resistance against , ~ toxic to the receipt organism or which completeshhe enzyme system of a mutant lackig such an essenhal pvl.y~JLi~, e.g. of an ~uAoh mutant. Suitable marker genes express, for example, anhihiohic resistance, e.g. against ~h~ illC, ampicillin, or ~ . or provide for ,uluLuLlulJh.~ in an auAuLlu~
mutant, for example in a yeast deficient m the ura3. Ieu2. his3 or trPI gene. It is also possible to employ as markets sh-uctural genes which are associated with an: 'y replicahng segment providing that hhe host to be ' ~ is ~ ul~i ' for hhe pDduct expressed by the marker.

Hybrid vectors for the expression ûf a DNA coding for a heat-stable ~Iu~
have in general hhe same features as the hybrid vectors described 1 .,: .l ,. r ~G for ~ -, ~- 'r~n of DNA, and additionally comprise expression contDI sequences allowing h'le produchion and~ ophionally~ the secrehion of heat-stable lJluly~ Thus, hybrid expression vectors of hhe invenhion comprise a promoter region operably linked with a sh-uctural gene encoding heat-stable pDly~ . and, optionally, if desired or needed, a DNA fragment encoding a leader or signal pephide, a u ,. ~
enhancer, a ribosomal binding site, a u, ~ i terminator region and!or further regulatory sequences.

A wide variety of promoter sequences may be employed, depending on the nature of hhe host oell. Promoters hhat are stDng and at hhe same time well regulated are hhe most useful. Sequences for hhe inihahion of hranslation are for example Shine-Dalgarno sequences. Seqnences necessary for the inihiation and of n, ,~ and for stabilizing the mRNA are commonly available fmm hhe noncoding 5'-regions and 3'-regions, ~ ~Liv.,l~, of viral or eukaryotic cDNAs, e.g from hhe expression host Examples of suitable promours are ~PL. ~PR. or ~N. E. coli lac, h-p, tac, or Ipp, yeast TRP1-, ADHI-, ADHII-, PH03-, PH05-, or glycolyhic promoters such as hhe promoter of the enolase, ~;lyl;~"~ldell.~vc-3-phosphate dcl~ ~c, 3 ~ a"~ ~ kinase (PGK), I . pyruvate d~ul/ui~yldsc~ ~hu~l~Lvrlucluki~ " glucose-6-phosphateisomerase,3-L/Lu~llv~;ly~ mutase,pyluvatekinase. u ;~ isomerase, ,' isomerase and ~h - ~ genes, or promoters deTived from eukaryohic vinises, e.g. SV40, Rous sarcoma virus, adenovirus 2, bovine papilloma virus, ~ayuv~vhu~, cytomegalovirus or Ba,uluv uu~, e.g. Auto~rapha califomica nuclear wo 96/00293 2 1 9 2 0 9 1 1 ~

pol~;.~u~;~ virus (AcMNPV), Tri~h~ ' ni MNPV, Rachiplusia ou MNPV, Galleria mellonella MNPV, derived promoters or l cell derived promoters, e.g. of the actin, collagen, myosin, or ~globin gene. A preferred eukaryotic promûter is a polyhedrin gene promoter of a r ~ v u u~ plrfi ~nti f~liy of the Auto~rapha californica nuclear polyh~ALu~ls virus (AcMNPV). The eukaryotic promoters may be combined vith en-hancimg sequences such as the yeast upstream activating sequences (UAS) or viral or cellular enhancers such as the c.~ .uul.,~ v ~- u~ E enhancers, SV40 enhancer, immuno-globulin gene enhancer or others.

Er hancers useful for the expression are ~ . . simulating DNA sequences, e.g.
derivedfromvirusessuchasSimianvirus, C~ ;GhJVUU:~ polyomavirus,bovine papilloma virus or Moloney sarcoma virus, or of genomic origin. An enhancer sequence may also be derived from the ~.ALI~lu~ 1 ribosomal DNA of Phvsarum ~ol~, ' ' . or it may be the upslream activation site from the acid 1 ' , ' PH05gene, or the PH05, tlp, PH05-GAPDH hybrid, or the like promoter.
~ignal sequences which can be used in the present invention may be. for example, a e or secretory leader directing the secretion of the polypeptide, or dhe like.
Signal sequences which can be used in the present invention are known in the literature.
Another suitable signal sequence extends from amino acid I to 19 of dhe amino acid sequence depicted in dhe sequence hsing under SEQ ID No. 1.

A ribosomal binding site (Shine-Dalgarno Sequence) is either naturally linked to dhe promoter used or may be located on a short nucleotide sequence which may be covalendy linked to the 5' end of dhe coding region for heat-stable l)lulyl 1~ Ribosomal binding sites are well known in the art.

A promoter chosen for dhe con~l- UUtiUII of a hybrid expression vector of the invention may be regulated by a regulatory protein and dhe production of heat-stable l~lulyl~ ...1. 'l" l~ ;"' in the u .,"~r ", A host cell dhen may be inducible or ~ lf The gene for the regulatory protein may be located either in the genome of the host strain, on an additional plasmid vector the host strain may be ~.Ul ~ ~ - r ." ". A with, or on dhe hybrid vector of the invention. The selection of a suitable gene for a regulatory prooein depends on dhe promoter used. The conditions for the induction or L~c~ h>n of the production ofheat-stable ~lul,~ ,U.l~ also Lepend on dhe promoter and on the regulatory protein. A regulatory protein which can be used in the present invention is, for example, a ~ wo s6/00293 2 1 9 2 0 9 1 r~

repressor protein, e.g. a product of the trpR, lacI, i cro, or ~cI gene, or a i sensitive mutant thereof.

PreferAred hybrid expression vectors of the invention are expression vectors suitable for the expression in E. coli of heat-stable l,lul.~ derivable from the amino acid sequence shown in SEQ ID No. 1, more preferably expression vectors comprising a signal sequence, preferably the signal sequence of the ~lul~ gene shown under SEQ ID No. 1, o~ dy Iinloed with the gene encoding the heat-stable According to the above, the present invention concems a ' DNA molecule rnmIIricing a DNA sequence coding for heat-stable ~ la I~ which DNA
sequence is derivable from a DNA sequence coding for a wild-type l~lulyl-e.g. by a method _ , g the steps (a) ~ ~ of a starting DNA coding for a ~ IYI~ ,n ~

(b) generation of a library of mutated DNA sequences obtained in (a), and (c) screening the library for a gene coding for a l~iul~ with improved1l~, ~t~l~jlity if compared to the r 1~, wild-type enzyme.

A pre~erred ' DNA molecule is such comprising a DNA sequence coding for a preferred heat-stable ~luly~ u~ of the present invention.

A.~.~nn1ingly, a preferred 1. ' ~ DNA molecule comprises a DNA sequence coding for a heat-stable ~IU~ rL q.~ derivable from a prolyl~ ~ of a prokaryote, preferably from El~vub~l;ulll spec.~ more preferably from F.
ru~ U~ 11, most ~/ler~ciuti~lly from F. strain IF0 12535 (ATCC
13253).

More preferred is a r~~n-~h~ ~t DNA mole, ule comprising a DNA sequence coding for a heat-stable IJIUIY~ - which derivable from a l/1UI~I~ 1- 'l" 1~l ;~l A~ having the sequence with SEQ ID No. l More preferred is a ~ -- I DNA molecule comprising a DNA sequence coding for a W0 96/00293 r_l~ _ C 1~5 2192091 ~

heat-stable IJ~ulyl~ 1. '1'~ .1.1 ;.1,.~. . derivable from a DNA molecule having the sequence with SEQ ID No. 1.

Evcn morc preferred is a ' DNA molecule ~t~mp~cing a DNA sequence coding for a heat-stable ~.~JIyl~ selected from the group of heat-stable ~lulyl~ having SEQ ID No. 1 with amino acid Glu in position 48 ("Glu48"), PheSl, Alal29, Gly633 amdlor Glu477 replaced by another amino acid.

Even more preferred is a ' DNA molecule ~ , ,, a DNA sequence coding for a heat-stable ~,ulyl~ ~. which . ~ DNA molecule is selected from the group of ~ DNA molecules having the sequence of SEQ ID No. 1 with thecodons for ammo acid Glu in position 48, PheSl, Alal29, Gly633 amdlor Glu477 replaced by a codon coding for another amino acid.

More preferred is also a ,. ' DNA molecule comprising a DNA sequence coding for a heat-stable ~-ulyl- 1- ~ selected from the group of heat-stable yl,~ having SEQ ID No. 1 with amino acid Glu48 replaced by Gln, PheS 1 replaced by Leu, Alal29 replaced by Thr, Gly633 replaced by Val andlor Glu477 replaced by Lys.

More preferred is also a Ir~ DNA molecule comprising a DNA sequence codingfor a heat-stable l,.ul~ l_ - selected from the group of DNA molecules having SEQ ID No. I with the codon coding for amino acid Glu48 replaced by a codon coding for Gln, the codon coding for Phe51 replaced by a codon coding for Leu, the codon coding for Alal29 replaced by a codon coding forThr, the codon coding for Gly633 replaced by a codpn coding for Val andlor the codon coding for Glu477 replaced by a codon coding for Lys.

Even morc preferably the invention concerns a ~r~ ..,. l.: - ~ DNA molecule comprising a DNA sequence coding for a heat-stable !~ulyl- l( l ~lJ~ - selected from the group of heat-stable ~.uly l l( ~ having SEQ lD No. I with amino acid (a) Glu48, (b) Glu48, Alal29 and Gly633, (c) Glu48, Alal29, Gly633 and Phe51, and (d) Glu477 rcplaced by another amino acid.

Even more preferably the invention concerns a ' DNA molecule comprising a DNA sequence coding for a heat-stable IJ-ulyl~ r..1. .~.~1,l;~l_~. . selected from the group of ~ W096/00293 21 92091 r~ 9.

DNA molecules having SEQ ID No. 1 with the codons coding for amino acid (a) Glu48, (b) Glu48, Alal29 and Gly633, (c) Glu48, Alal29, Gly633 and PheS1, and (d) Glu477 replaced by codons cooing for another amino acid.

Even more preferred is also a ' DNA molecule c~mpncing a DNA sequence coding for a heat-stable ~lul~ u~ selected from the group of heat-stable ~ul~ having SEQ ID No. 1 with amino acids (a) Glu48 replaced by Gln (PEP-227), (b) Glu48 replaced by Gln, Alal29 replaced by Thr and Gly633 replaced by Val (PEP-361) (c) Glu48 replaced by Gln, Alal29 replaced by Thr, Gly633 replaced by Val and PheS1 replaced by Leu (PEP 4û7), and (d) Glu477 replaced by Lys (PEP-15).

Even more preferred is also a ' DNA molecule cnmrri ring a DNA sequence coding for a heat-stable ~lul~ selected from the group of DNA moleculeshaving SEQ lD No. 1 with (a) the codon coding for amino acid Glu48 replaced by a codon coding for amino acid Gln, (b) the codon coding for amino acid Glu48 replaced by a codon coding for amino acid Gln, the codon coding for amino acid Alal29 replaced by a codon coding fo} amino acid Thr and the codon coding for amino acid Gly633 replaced by a codon coding for amimo acid Val (c) the codon coding for amino acid Glu48 replaced by a codon coding for amino acid Gln, the codon coding for amino acid Alal29 replaced by a codon coding for amino acid Thr, the codon coding for amino acid Gly633 replaced by a codon coding for amino acid Val and the codon coding for amino acid PheS1 replaced by a codon coding for amino acid Leu, and (d) the codon coding for amino acid Glu477 replaced by a codon coding for amino acid Lys.

Moreover, the invention preferably concems also a ~rt~- hir~ DNA molecule ~ a DNA sequence coding for a heat-stable IJ..,lyl ~UI~ selected from the group of DNA molecules having SEQ lD No. 1 with (a) the nucleic acid G in position 458 ("458 G") replaced by C, (b) 458 G replaced by C, 701 G replaced by A, and 2214 G
replaced by T (c) 458 G replaced by C, 701 G replaced by A, and 2214 G replaced by T, and 467 T replaced by C, and (d) 1745 G to A.

Even more preferred is a .r ~ - -' DNA molecule comprising a DNA sequence coding for a heat-stable ~ul~ selected from the group of DNA molecules coding for a heat-stable ~,ul,~ in pUK-FPEP-lS, in pUK-FPEP-227, in pUK-FPEP-361, or in pUK-FPEP-407.

w0 96/00293 2 1 9 2 0 9 1 ~ ~

A ~ DNA molecule according to the present invention may be an isolated DNA
fragment coding for a heat-stable l~-UI,V~ J~ according to the invention, e.g. such consisting only of the coding region or also such being rrolonged by 1.. ~l~," .. - or L~ .UIO~JUS DNA sequences. A prolonged fragment may, for example, contain linkersequences, e.g. for cloning purposes, or may be linked to other fragments containing marker genes or functional elements for replication or gene ~Tpn c~i~,n Such fragments can be used for n r .". . ~ .., of host cells or as ' for the generation ofcloning anrllor expression vectors. A ' DNA molecule according to the present imvention also includes hybrid vectors for the ~JIUIJ..g~l~iUll and ~ ,.. of a DNA
sequence coding for a heat-stable y-ul,~ - ~p~lJ~ "'' of the invention, and expression vectors for the expression thereof in a suitable j ~ l host.

Preferred expression vectors ate those for the expression in a Ba~uluvhuO/i~L cell expression system or in E. coli.

Most rpeferred are expression vectors pUK-FPEP-15, pU~-FPEP-227, pUK-FPEP-361, and pUK-FPEP 407.

The present invention also concerns the preparation of a - ' DNA molecule of the invention, preferably of a hybrid vector or of a hybrid expression vector of the invention.

Tr~nsfr\rm~A hosts and preparation thereo f The invention also concerns a b ' 1 host cell for ~ g a ~ DNA
molecule of the invention or ~uliuuLuly for the production of a heat-stable ~-ulyl~ 1"1~ as well as a process for the lJlc~ iull of such a l l ~ r ~l ll A host cell.

Hosts mentioned herein can also be used for the, ~ of the starting DNA used for ~ and for the generation of a library of mutated ~lul~l- ...l. 'l" I-l;-l ~ genes.

The 1, ". r " ,, ,. A microbial host strains are cultured in a liquid medium containing sources of carbon and nitrogen which can be ~lccirnil-.tPA by the microbial cell, and inorganic salts, applying methods known in the art. The culture of the hosts is carried out in a cu~ v~,.l~iu.l~.l nutrient medium which may be;, J~l ~1 with or d_prived of chemical cr mronn~ic allowing negative or positive selection of the ~ r ~ , i.e. such hosts wogfj/ooz93 2192091 r~

.

containing the desired DNA molecule together with a selection marker, from the non-: ' i.e. such hosts lacking the desired DNA molecule.

Any i ' ' ' hosts useful in the art may be used, e.g. bactefia, such as ~, fungi, such as S ~ . y ~s cerevisiae. T~' v ~ ~u,., y ~ lactis. or _ _ : fungi, such asAsper 2illus SPec., e.g. A. nidulans. A. orvzae. A. calbvllalill~, A. awamori or A. rli~er.
However, the use of suitable hosts which are devoid of or poor im restriction enzymes or ' = cnzymes may be ~lv Examples of such hosts are bacteria, e.g.
Bacill~ssubtilis.Bacillus,t~v~l ., .,.hl~ r~ C. T~ nc.Sl~lJ~V~-and others, and yeasts, for example Sa~,~,h~vl~lY~,~,s cerevisiae. and in particular strains of T~ ,. ..1 - cvli. for example E. coli X1776, E. coli Y1090, E. coli W3110, E. coli HBlOlLM1035, E. coli JA 221, E. coli DHS4 or prf f~ lly E. coli DHSc~F~, lM109, MH1 ~ HB 101, or E. coli K12 str.un. Further suitable hosts are cells of higher organisms, in parlicular ectohlich~A continuous human or animal cell lines, e.g. human embryonic lumg fibroblasts L132, human malign. nt melanoma Bowes cells, HeLa cells, SV40 ViTUS
' kidney cells of African green mûnkey COS-7 or Chinese hamster ovary (CHO) cells. Other suitable host cells are established insect cell lines, for example, Sporlfpt~afru~iperAa~suchassf2lorr c~ 'IySf9(ATCCCRLl711),Mamestra brassicae. Bombvx mori cell systems usin~ Bombvx mori nuclear pul.~h~hv~;~ virus(BmNPV) and the hke.

The present invention also concerns the ~ JalaLiu~l of a hosm, ~ .~ r ., . A with a ' DNA molecule of the invention. The 1" ~ of such i ' ' hosts compfises the treatment of a desired suitable host cell under i ' g conditions with a desired i DNA molecule of the present invention, preferably a hybrid vector or hybrid expression vector of the invention, optionally together with a selection marker gene andoptionally selecting the i ' T ' of L is carried oot accorAing to l,UllV.,.lLiUlldl methods as described in the hterature.

Acco~dingly, the i ' ~ procedune of E. coli cells includes, for example, Ca2+
1!"; of the cells so as to allow DNA uptake, and incubation with the hybtid vector. The subsequent selection of the ~ r ~ " ~f A cells cam be achieved, for example, by 1, r 11~ F ~ I; ~g the cells to a selective growth medium which allows separation of the r.., .,. A cells from the panent cells dependent on the nature of the marker sequence of the vector DNA. Preferably, a growth medium is used which does not allow growth of cells which do not contain the vector. The l . - ~ ~ .. " ~ of yeast comprises, for = = _ _ . .. . .

wo 96/00293 r~
21 q20ql -~

example, steps of enzymatic removal of the yeast cell wall by means of ~1. ,. n~treatment of the obtained ~ ul~ (, with the vector in the presence of pul~ lc.~
glycol and Ca2+ ions, and ~ ~ L' ~ of the cell wall by; ' ~ ' ,, the ~1 ' u~l~L~ into agar. Preferably, the . ~ agar is prepared im a way to allow ~ and selecdon of the 1 A.l r .. . - ~i cells as described above at the same time.

T r ' of cells of higher eukaryotic origin, such as ' cell lines, is preferably achieved by i ~ T ~ is carried out by cull~,llLiu,l~l ~ , such as calcium phosphate l~lcc;lJ;uLLIull, .,.;~ protoplast fusion, ., i.e. ihlLI~ ' of DNA by a short electrical pulse which transiently increases the IJ~ db;lily of the cell membrane, or in the presence of helper s such as u~ h~' ' yldcAL~ dimethyl sulfoxide, glycerol or pol~,uhjh,ll., glycol, and the like. After the i ' procedure, transfected cells are identified and selected e.g. by cultivation in a selective medium chosen depending on the nature of the selection marker, for example standard culture media such as Dulbecco's modified Eagle medium (DMEM), minimum essential medium, RPMI 1640 medium and the like, containing e.g.the r ~' g antibiotic.

The 1 r..~ r ", A host cells are cultured by methods, known in the art in a liquid medium containing assimilable sources of carbon, e.g. c~l/ull.y~ such as glucose or lactose, nitrogen, e.g. amino acids, peptides, proteins or their fir,~ - products such as peptones, ' salts or the like, and inorganic salts, e.g sulfates, phosphates and/or carbonates of sodium, potassium, ., ' and calcium~ The medium ~ ' contains, for example, growth-promoting substances, such as trace elements, for example iron, zinc, manganese and the like.

The medium is preferably so chosen as to exert a selection pressure and prevent the growth of cells which have not been j r ~ or have lost the hybrid vector. Thus, for example, an antibiotic is added to the medium if the hybrid vector contains an antibiotic resistance gene as marker. If, for instance, a host cell is used which is ~ nul~l; in an essential amino acid whereas the hybrid vector contains a gene coding for an enzyme which . . ' the host defect, a minimal medium deficient of the said amino acid is used to culture the, - ~r... ,. A cells.

Cells of higher eukaryotic origin such as " ~ " cells are grown under tissue culture conditions using rc~ y available media, for example Dulbecco's modified Eagle ~ W096/00293 21 92091 r~

~ 23 ~
1_ medium (DMEM), minimum essenial medium, RPMI 1640 medium and the like as menioned above, optionally ~ with growth-promoting substances andlor ' sera. Techniques for cell cultivation under tissue culture condition are well known in the aTt and include 1 ~ suspension culture, e.g. in an airlift reactor or in a continuous stirrer reactor, or ~ 1i7~ or entrapped cell culture, e.g. in hollow fibers, . ~ .y~ , on agarose . -1~ porous glass beads, ceramic cartTidges, or other ~

Culturing is effected by processes which are known in the art. The culture conditions, such as: r , pH value of the medium and ' time, are chosen so that a maximum expression level of the ~.u~ u~ Lid~ or derivative of the invention is obtained.
Thus, an E. coli or yeast strain is preferably cultured under aerobic conditions by submerged culture witb shaking or stirring at a i . of about 20~C to 40~C, preferably at about 37~C, and a pH value of 4 to 8, preferably of about 7, for about 4 to 30 hours, preferably until maximum yields of the pol~ ,~i~ or derivative of the invention are reached.

Heat-stable l~lVIYI~
Accol!dingly, the present invention concerns a heat-stable prolyl~ obtainable by a process comprising the steps (a) ~ ~ of astartingDNAcodingforal,.ul~,l "I..l..l,li,l~

(b) generation of a library of muLated DNA sequences obLained in (a), (c) screening the library for a gene coding for a lJ-ul~ Y'~ IJ~ with improved heat-stability if compared to the ~:u ~,uollLl.g wild-type en~yme, and (d) expression of the gene obtained under (c) and isolating the expression product.

A preferred heat-stable ,ulol,y~ y~ is such derivable from a ~ul~
of a prokaTyote, preferably from n~vuba~ ;ul,l sPec., more preferably from F.
most ~lcr~ '1!/ from F. . ~ strain IFO 12535 (ATCC
13253).

More preferred is a heat-stable ~IUI,~ derivable by random W096/00293 2 l ~20q l T._1.~,5:: ~

from a ~ulyl 1f~y~ I~n ~1AC~: having the sequence with SEQ ID No. 1.

It is sho vn in the present invention that changes in amino acid position 48, 51, 129, 477 and 633 of the sequence with SEQ ID No. 1 increase the I A~ ;liry of the ~ul ,1 ~ Accordingly, the invention concerns even more preferably a heat-stable ~lul~l- 1. ,lJ~ r selected from the group of heat-stable ~lul~ having SEQ ID No. 1 ~vith amimo acid Glu in position 48 ("Glu48"), PheSI, Alal29, Gly633 and/or Glu477 replaced by another amino acid.

More preferred is also a heat-stable lJIulyl~ selected from the group ofheat-stable ~IUI.rl' 1~ JI;~ having SEQ ID No. I with amino acid Glu48 replaced by Gln, PheS I replaced by Leu, Alal29 replaced by Thr, Gly633 replaced by Val andlor Glu477 replaced by Lys.

Even morc preferably the invention concerns a heat-stable l~-uljl~ ' 1~ '1''1'1 Ir1A~ selected from the group of heat-stable ~ulj1 fl ~ having SEQ ID No. 1 v.~ith amino acid (a) Glu48, (b) Glu48, Alal29 and Gly633, (c) Glu48, Alal29, Gly633 and PheS1, and (d) Glu477 replaced by another ammo acid.

Even more preferred is aLso a heat-stahle ~ulyl~ ~-1- '1'~ l~l ;-1- ~ selected from the group of heat-stable ~,.ul;~ c having SEQ lD No. I with amino acids (a) Glu48 replaced by Gln (PEP-227), (b) Glu48 replaced by Gln, Alal29 replaced by Thr andGly633 replaced by Val (PEP-361) tc) Glu48 replaced by Gln, Alal29 replaced by Thr, Gly633 replaced by Val and PheS1 replaced by Leu (PEP-407), and (d) Glu477 replaced by Lys (PEP-15).

Production of heat-stable ~,.ul.,l. .,.1..
The present invention concerns also a method for the production of ~ 1JI;~

For the expression of ~,.ul~ either 1~ul~yu6~, or eukaryotic host cells may be used as indicated above, e.g. E. coli strains defechve in protease genes, e.g. in the lon protease gene, and genes involved in the regulation of heat shock induced protein synthesis, e.g. in the htpR gene.

Preferably, lJ uly'~ r ~ is produced in E. coli. In this case, to i=mprove the expression, a 5'-terminal non-coding region of the cloned DNA is preferably removed ~ woscloo2s3 21 92091 r~ 1 .

wbile ~ ~ ~ the full length of the coding region, preferably the coding region of a mature heat-stable IJ~ulyl~ preferr~d above. The coding region is most preferably functionally linked with a signal sequence allowing the secretion of the heat-stable ~lvlyl .1 .~t;~l Moreover. the structural gene is functionally linked with a promoter region functional in E. ~, either l -~, or preferably hr ~ , to the desired ~,.ul ~ J~ coding region. The linkage is performed according to a .,u.,~. 1 procedn~, forexample, using an al, ~ restriction enzyme site or deletion by digesting with an ' such as E. coli ~ ' m and successive blunting with a nuclease, e.g. mung-bean nuclease.

In a preferred cnnhorlim~nt of the invention, a genomic DNA having a linker sequence ~y upstream of a full length coding region for heat-stable ~JlUIyl iolJ ~ is hnked with a L~ ulobJu~ promoter such as tac promoter in an expression vector, for example, a plasmid based on pUC119 plasmid. More preferably a coding region in the genomic DNA from llavuba~iu~ encodes a pro-form of a desired ~ul~ J~ . of the invention, e.g. such consisting of a matule form of a heat-stable ulyl~ J~ - and a signal peptide, for example amino acid residues -19 to -1 of the sequence shown under SEQ ID No. 1. When such a type of an expression plasmid is used to transform E. coli host, and the i ~ is cultu~d, then heat-stable l~ulyl. ~ - is produced in E. coli cells and secreted into the ~ ,;, region.
In the process of secretion the signal peptide is removed to give a mature form of the enzyme which is not h~,ul~J~ ' in inclusion bodies, and therefore, the produced heat-suble IJ-ulyl~ ~-1- q~ . is easily recovered.

According to another h~l; ~ of the present invention, a DNA coding for the present enzyrne is inserted into a br~uluvilu~ transfer vector to construct a .~ ' baculovirus transfer vector, and the ~ I,~uluv i. .,~ transfer vector is then co-transfected with a b~uluvllu~ DNA to insect cells to carry out a l-. "",~l~,t,.....
r~rt)mhin~ n .
The l~ ,ulovi u~ transter vector is usually a plasmid containing a segment of b~uluv UUD
DNA, which segment comprises a gene not essential for the replication of lu~uluviu~
The gene not essential for the replication of ' ' .vi. u~ is, for example, a polyhedrin gene comprising a polyhedrin structure gene and a promoter thereof. Such baculovirs transfer vectors known in the art are, for example, pAcYMI, pAc311, pAc360, pAc373, pAc380, pAc700, pAc701, pAc702, p~L1392, pVL1393.

wo 96l00293 P~
2 1 f~209 l O

B ~ VUUD~D which can be used in the present invention are, for example, Trichonlusia ni MNPV, RachiPlusia ou MNPV, Galleria mellonella MNPV, and the like. nef~ tii-l1y used is Auto~rapha californica nu~clear polyll~u~;~ virus (AcM_V). A kit comprising a ' ~ of an Auto~raPha californica nuclear~ul.~l.~u,.O virus and I ' vuuO
transfer vectors pAc700, pAc701, pAc702, pVL1392 and pVL1393 is ~ ,v available from Invitrogen Corp., San Diego, CA, USA. Insect cells useful in the present invention are established insect cell lines, for example, Sp~t~rtf ~ fru~iperda, such as ~
Sf21 or ~ r ~Iy Sf9 (ATCC CRL1711), but also Mamestra brassicae and the like, A
Bombvx mori cell system using Bombvx mori nuclear~ulyl.~v~;s virus (BmNPV) can also be used in the present invention.

The l - l ~IhL~ ' ' can be carried out in accordance with a ~.u~v, procedure as described, for example, in "A Manual of Methods for P ' ~ v u UD Vectors and Insect Cell Culture Pr, ' ;, MD. Summers et al., Texas Agricultural Pxr~i~ntStation Bulletin No. 1555". The transfected insect cells can be cultured in accordance with a ~,UIIV~ '' I procedure. Namely, the transfected insect cells may be cultured in any tissue culture medium in which imsect cells can grow, such as Grace's or TC100 medium ! ~ with ' serum, serum-free medium EX-CELIA00, or the like, at a , of 20~C to 30~C, preferably 27~C to 28~C, for example 27~C, for 2 to 10 days, preferably 3 to 5 days.

The expressed prolyl~ . . L .1,. ~ . can be extracted from microbial cells such as E. coli cells or a ~ of a cell culture by ~UIIV~ iUIIal methods, e.g., comprising 1 ., Ø 1:,-l;~.,l of the cells, ulu~ O . ' y such as ion-exchange, I~yLu~hul~;c or size exclusion u lu- - l ~ - - e.g., with ammonium sulfate or acid, preparativecl~L., ,' suchaspol.ya~. ylaunid~gel~lf l~u~ orisoelectric focussing, and the like. Particularly, heat-stable IJ~ulyl~ lul~ . derivable from ElavulJa~ iulll wild type enzyme, which is expressed in E. ~, is easily and selectively extracted from the cells with an osmotic shock method if the enzyme is secreted to the 1, ;I,1 i~ region. The obtained crude enzyme can be further purified with usual methods, e.g. comprising u~ O . ' y such as ion-exchange, hydlu~!llulJ;u or size-exclusion .,Iu~ 1 Aplly~ preparative clc~llu~Lu~o;o such as ulya~,lyLIIIidt gel clc~ ul!hu~r~D;o~ or isoelectric focussing, and the like.

Production of C-terminallY amidated peptides ~ _ _ ~ W0 96100293 2 11 9 2 0 9 1 r~

~ .;

The present invention concerns also a method for the pDduction of C-tenninus amidated peptides under high . conditions by use of ~lulyl~ ~i~ 'l" l" ;'I_'~
P~UI~ catalyzes not only the hydrolytic cleavage of a peptide at the C-terminus side of a pDline residue, but also, folming the peptide bond in the reverse manner of the hydrolysis, the coupling of a peptide fragment to C-terminus of the other fragmcnt which is terminated by a proline residue. UndercontDlled conditions thecoupling reaction is 1 ' and l!lulyl~ ~. is used to catalyze coupling of two peptide fragments (or of amino acid to a peptide fragment). The preferable conditions of the coupling are an excess of one of peptide fragments (or an amino acid), and the presence of an organic solvent, such as glycerol, ethylene glycol, butanediol, ethanol, _-pDpanol~ ~-pDpanol~ sl~.t~nih~ , DMF, and DMSO, in a high . typically more than 50 %.

In prefierable e- .~ of the present invention, biologically active peptides whose C-termini are ~-amidated and have proline residues, preferably, at or near their C-tetmini are prepared with ,~IU~ rl~ ~lJ' 1~n-~ from two precursors thereof, wherein one of the precursors is a precursor peptide formmg N-terminal region of the amidated bioactive peptide and having a proline residue at its C-terminus and another precursor is a precursor peptide or amino acid forming a C-terminal poltion of the amidated bioactive peptide which precursor peptide or amino acid has been amidated at C-terminus. The ~-amidated bioactive peptides prepared with ~luljl. . 1. .l.~,l;.l ~ involve aspartocin, I,. . ., - ..~
calcitonin, CGRP, CGRP II, crustacean ~ OlllU~lIUIC ~ ;"C hormone, cockroach myoactive peptide L color change hormone, glumitocin, ~ ;..-R, isotocin, LH-RH, mesotocin, morphine ...nl l-~--c r,.,u.u~ ,Lidc, c~-MSH, oxytocin, ~ JIC~ SCPA, SCPB, valitocin, ~ r ~ . and vasotocin.

The present invention pDvides a pDCeSs for the production of a l~ ' DNA
molecule comprising a gene coding for ,~lul,~ , comprising the steps ofpreparing cDNA or genomic DNA fDm cells, plrfi rl~nti~lly bacterial cells, capable of producing ~lul~l L .l.. l,l;.l inserting DNA fragments coding for l,lulrl~
into a cloning vector, and selecting a hybrid vector containing the DNA coding for IJlul~l . 1. '1'' 1'l ;~1 ..

The present invention concerns in particular the ~,...~ " discloscd in the examples.

DESCRIPTIQN OF I~HE FIGURES

w096/00293 2 1 9209 1 P~

Fig. I represents restriction maps of the cloned inserts in pFPEPO2 and pFPEP03. The open box represents the open reading frame of the ~lvl.~ -~ gene and the solid box a consensus sequence of the catalytic site of serine protease.

Flg. 2 represents ~ 1y expression plasmid pUK-FPEP-b. The tac promoter is followed by the structure gene of pluly~ A.,~ . (shaded darldy), the multiple cloning site (MCS) and the E. coli rrr~3 T1 i terminator. The plasmid carries the M13 IG region which enables ll-r~ of the single stranded DNA with the aid of a helper l ~

Fig. 3 represents thermal hl~~ .lioll curves of the wild type and heat-stable JlU~ L ~P~ 1A ~ S Each solution of the wild type (filled-in circles) and the heat-stable ulyl~ p~ PEP-227 (circles), PEP-361 (filles-in squares) andPEP 407 (open squares), in 20 mM sodium phosphate (pH 7.0) is incubated at variant; .p. .~ . for 30 min and remaining activity of ~lul~T .~ is measured. The resiudal activities are shown as functions of i . .

The present invention will now be further illustrated by, but is no means limited to, the following examples.

EXA~LES
In the Examples, the following materials and methods are commonly used.

The bacterial strains and plasmids used are listed in Table 1.

~ wos6/00293 2 1 92Q9 1 F~,~/ 11 '1~

Table 1: Strains andplasmids.
Strains o} plasmid Relevant genotype Strains E. coli JM 83 ara,~(lac-proAB), rpsL(=strA), 080r, lacZ AMlS
JM109 recAI, endAl, gyrA96, dhi, hsdR17, supE44, relAI, ~~, ~(lac-proAB), F'[proAB+, lacIq, lacZ ~M15, traD36]
HB 101 F, hsdS20(r~B, m~B), recA13, ara-14, proA2, lacYl, galK2, rpsL20 (Smr), xyl-S, md-l, supE44, ~~, mcrA+, mcrB~
TGI supE, hsd ~5, dli, D(lac-proAB), F'[proAB+, lacIq, lacZ A M15, traD36]
F. ~ - lFO 12535 (ATCC 13253) Plasmids pUCl9 Ampr, lacIq, lacZ' pUC118 Ampr, lacIq, lacZ', M13IG
pUC119 Ampr, lacIq, lacZ', M13IG
pKK223-3 Ampr, P",~, rrnB TIT2 T ' restriction mappmg, ~ )' ' of plasmids, and odher molecular cloning procedures are done by standard medhods. (Sambrook, J. et al. "Molecular cloning: a laboratory manual," 2nd ed. 1989, Cold Spring Harbor Laboratory, Cold Spring Harbor, Silhavy, T.J. et al. "F l~ with gene fusions," 1984, Cold Spring Harbor Laboratoly, Cold Spring Harbor). Restriction enzymes and DNA-modifying enzymes are used according to the . r~ ., - - l - ~ ;. . - - of the ' - Deletion with r ~
m is carried outby use of a Kilo-Sequence Deletion kit (Yanisch-Perron, C. et al. Gene, 1985, 33, 103-119; Henikoff, S. Gene, 1984, 28, 351-359 ). The nucleotide sequences are ~i~t~rmin~l by tbe dideoxy metbod, by using a Sequenase kit. Genomic DNA from F. ~ r,~ is isolated b=y the method of Saito and Miura (Saito, H. et al.

w096/00293 2 1 9209 1 I~

-3a-Biochiun. Biophys. Acta, 1963, 72, 619-629). ()~ u~ are Oy lLL~Oi~ with an Applied Biosystems Model 381A DNA oyl~Lll~,o;~l. After completion of the trityl-on synthesis the ~' ,, ' ' are cleaved from the support and ~IUt~,t~i, follo ving astGndGrd protocol. The syllLLO;~l DNA is then purified by use of the 1 I~irlLG~iull Cattridge according to the protocols of the .-- rA, ~ rl Restriction enzymes, DNA-modifying enzymes, the Kilo-Sequence Deletion kit and tbe MEGALABEL kit are purchased fmm Takara Shuzo Co. Ltd. (Kyoto). The Sequenase Ver.2.0 kit is the product of U.S . R i~ Corp. (Cleveland, Ohio) . E rul ~ l.,nlu~
dase from F. ~ ~ - - -r;- ~ J~ and F - ~, g Asp-N are purchased from Seikagaku Corp. (Tokyo) and Boehringer Mannheim Co. Ltd. (Tokyo), l~ ,ly.

The Ta~i Dye Deoxy Terminator Cycle Sr ~ n F kit and the ~
P~irl"G~ioll Cartridge are the products of Applied Biosystems, Inc. (Foster City, Califomia). The QIAGEN tip-l00 is obtained from DIAGEN GmbH (Dusseldorf, FRG).
The GENECLEAN II kit and the (~lR('T .~.C'lROW medium are the products of BIO 101, Inc. (Vista, Califomia). The r;~r~ lncr- filte} iô purchased from Schleicher & Schuell (Keene, New Hampshire). The YM 30 Ill~filh~inn membrane is obtained from AMICON Inc. (Beverly, M ' ). The CM52 cation exchange resin is the product of Whatman Paper Ltd. (r ~ ' England). The enzyme substrates, Z-Gly-Pro ~-UG~ -YIG IdLl~ and Z-Gly-Pro p- lirir~ are obtained from NU~GbiUL~ II AG (I ' '' ,, Switzerland). Radio isotopes are purchased from Amersham Japan Co. Ltd. (Tokyo). Other hinrhr.mirAlc are purchased from Sigma Chemical Co. (St. Louis, Missouri), Wako Pure Chemical Industries Ltd. (Osaka, Japan) amd Nacalai Tesque Inc. (Kyoto, Japam).

Exatnple l: Preparation of startin~ DNA from F. . ~ r ~ ~
(~nmm~-~rh~lly obtained ~lulyl 1. .~ I ;.i . is purified by reverse phase HPLC on a 4.6 x 35 mm TSKgel Octadecyl NPR column (Tosoh Co. Ltd.). The column is eluted with 0.01 % TFA in water and a 3:1 mixture of CH~CN and -Propanol, at a flow rate of I ml/min.
The gradient from 35-70% of the organic solvent mixture is applied over 40 min and the major peak is collected.

Since N-temminus of the l A~ .Lj,i is blocked, the enzyme must be subjected to proteolytic cleavage to detemmine its partial primary stmcture. The proteases commonly used for the cleavage like trypsin do not give sGLiOrGLLuly resultO. Therefore, proteaseS and ~ W0 96/00293 r~

conditions of the hydrolytic cleavage are :,~ lly investigated and Fn.l. 'l""~ - -Asp-N is found to give the best resulL

The purified enzyme (05 mg) m 10 mM: carbonate, pH 7.9, containing 4mM
urea is hydrolyzed by I ~lg of Endo~llut~ aae Asp-N at 37~C for 24h. The peptidemixture obtained by this digestion is separated by reverse phase HPLC on a 4.6 x 250 mm Vydac C18 column (.C ~p~.~*~nC Group Corp.) with the mobile phase of 0.01% TFA in water and a 3:1 mixture of CH3CN and i-PrOH. The flow rate is I mVmin. The isolated peptides are further purified by ~ .y~-J. The amino acid sequence of thepuriffed fragments are ~ ' by manual Edman ~ ;.- using the methods described by Kobayashi and Tarr (Kobayashi, R et al. T~ , Kaknsan Koso, 1986, 31, 991-1002; Tarr, GE. "Methods in protein sequencing analysis" (ed. Elzinga, M.), 1982, 223-232, Humana Press, New Jersey).

The nucleotide sequences for the probes are not uniquely ~' l from the amino acid sequences because of multiple codon usage. Out of the 23 partial amino acid sequences six which give relatively less ~ of possible nucleotide sequences are chosen to make DNA probes ~Table 2). Preferred codon usage in F. . . has not been known, and two guidelines are adopted in the design of the nucleotide probes. Namely, three of the 6 probes (A-12, 13 and 19) are designed so as to consist of a single nlig, .. 1.. I;.lr sequence, selecting the most probable codon for each amino acid residue on the ~Cc~mrti~n that the genome DNA of F. .: is GC rich. The other three (A-3, 9 and 18) are mLxtures of ,~ O~ s of the possible sequences. Toreduce further the number of the possible sequences in the mixture, inosine (I) is placed at the position which can be one of four bases, A, G, C and T, since inosine forms stable base pairs with all of four.

Table 2 Det~nnin~11parlialaminoacidsequencesofthefragmentsofl,l~,1~1....l..l.~l,li,l_~.,obtained by the F~ Asp N digestion (shown by the amino acid residur No. in SEQ ID No. 1), and, , ' g nucleotide positions in SEQ ID No. I of the probes designed from the amino acid sequences.

Fragment No. Amino acid residue Probe No. (~r "
No in SEQ ID No.1 nucleotide posiion in SEQ ID No. I
;

WO96/002s3 2 ~ 92091 r~ . O

are ~yllth~,si~ with an Applied Biosystems Model 381A DNA
synthesizer. After removal of d;..l~.lllu/~yllilyl group at the end of the synthetic sequence the . .~ ..1. .c are d~,yl~b_l and cleaved from the supports, according to the protocols of the . . ~ . The ~yu~ i~l DNA are then subjected to p~ .lN ~.
.L.~ with8%pol~ly' ' gelin7Murea. Pulified ~" '- ' are extracted from the separated bands and deionized by use of Waters Sep-Pack C-18 columns.

The .,IIIUInU~UIII~I DNA is isolated from F. ~ - -- and digested by 4 kinds of commonly used restriction enzymes ~ ~, u~ ,g 1, ~ ~ 1 U~ f sequence, i.e., Psd, Hindlll, EcoRI and Bgl~.

Q~ r probes are radio-labeled by use of a MEGALABEL kit vith [~-32P~ATP
to give a specific activity of ca 1 x 106 cpm/pmol. The .,1.1, -1 fragments are 3 0n a O.7% agarose gel and transferred to a Millipore lli~l u~ ll ulo~ filter by the method described by Sambrook et al. (Sambrook et al., 1989, supra).

After ~chybli0;~1ion according to a st mdard protocol (Sambrook et al., 1989, supra), hylJIidi~liu~ is carried out in 6x SSC hylJI;di~Liu~ solution with 0.2 pmoVml of the labeled probe at 45~C for 16 h. The f lter is washed with 6x SSC three times for 3 min at room t~ .IOlUlG and then once for 1 min at 45~C. A ' " ~ ' y is performed with aFuji Bio-image anayzer BAS 2000. Only the A-3 probe is found to give a clear andspecific signal with each of the digested DNA.

The molecular weight of l~lulyl~ is found quite large, 76,000 by SDS-poly-- --- ......... ,.. :.... ,,, _ ~

~ w096100293 2 1 9209 ~ P~

acrylamide gel C~ u~l-u~ (Yoshimoto et al., 1980, supra). The size of the enzymeto 2 kb of the coding region in the genome. The larger the cloned DNA
fragment Is, the higher the chance of including the full length of the open reading fnune.
~ Therefore, rather a long fragment but small enough to get a high efficiency in the - is desired and 7 kb of the Bgl~ fragment is selected. Namely, genomic DNA digested by Bglll is subjected to ~lc~ livc cl~Llu~llulc~ with low-melting-point agarose and the fraction of the gel containing 7 kb fragments is cut out. The excised gel piece is dissolved in a ligation mixture and the extracted I ' ulllu~ul.lal fragments are clonedimtoBamHlsiteofpUCI9. BythisligationmixtureE.coliHB101is nAI-e~
to give a genomlc library . , ~ g about 4,000 clones.

The genomic library is screened with the A-3 probe by colony hylll; li~uu~ and 119 positive clones are obtained. Sixteen positive clones are chosen and analyzed further by restriction ~ 1. .., k A~_ digestion and the enzyme assay. One clone with 7 kb insert is found to show a ~,u~ ivcly high l~lulyl "11~p. ~ . activity. The plasmid is named pFPEP02 and further ~ 1 The restricion map of the insert of pFPEP02 is shown in Flg. 1. The entire nucleotide sequence of the 2.6 kb of Hincll - EcoRI (shown in SEQ
ID No. I) L.~ ' ' by the dideoxy method from a series of deletion subclones generated from either end of a larger fragment which includes the 2.6 kb of HinclI - EcoRI (pFPEP03, deposited as FERM BP-3466).

Plasmid pFPEPO2 is digested with Hincll to obtain a 3.1 k bp Hincll fragment containing ~ulyl- ~ gene, which is then subcloned at the SmaI site of pUC118 by blunt end ligaion to give pFPEPO4. After the ligation, the inserion points at the both ends of tbe fragment are cleavable neither by Hincll nor Smal. To delete the Scal and Pvull sites in the coding region the syntheic double-stranded nl;g. -~ ul ;-1~ fragment which 1l ' to the sequence between the Smal and Pvull sites but is mutated at two posiions (with Synthetic Fragment I, see SEQ ID No. 2) is prepared by annealing the lower strand and tbe upper strand only whose S' end has been ~I-U~IJhVIY ' beforehand by T4 ~JJIy. l. L ~ kin~

On the other hand, the plasmid pFPEPO4 is cleaved at the single Smal site existing in the open reading frame, and the mutated Smal - Pvu13 fragment mentioned above is ligated to the linearized plasmid at both terminus Smal sites. The ligaion product is then digested by Sacl and the longer fragment, containing the S' porion of the coding region, is isolated by agarose gel C~ ulJIlu~ . After kinasing the isolated fragment the missing piece wo 96/002s3 r~ s 21 92091 ~~

between PvuII and SacI sites, prepared separately from pFPEPO4, is ligated to construct plasmid pFPEPO4'. Since one nucleotide at the terminus generated by PvuII has been changed in the synthetic fragment, tbe PvulI site is not l~u x ' by this cychzation.

Next, a new EcoRI site is created ~ Iy upstream from the initiation codon of theLllul.~ gene as follow. Synthetic Fragment II (SEQ ~) No.3) is prepared by ligation of the four nlir,~ .. 1P.. 1;.1. ~ Ul, U2, Ll and L2 (see SEQ ID No.4, 5, 6 and 7, ~c~ ly)~ where two of them (U2 and Ll) have been IJhu~hul ~ ~ at their 5' termini. The prepared fragment f ~ ., . r~ from the initiation codon to PvulI site in the upstream coding region and has protruding cohesive 5' terminus ' ly upstream of the initiation site to introduce EcoRI site after ligation. For the following ligation the both 5' ends of the prepared fragment are pllU~ lUl.~' ' ' by T4 PUIJ ' a ' kinase.

The plasmid pFPEPO4', in which one of four PvulI has been deleted, is digested with PvuII, and the fragment containing most of the coding region is isolated by agarose gel rl~ U~ and ligated to the second synthetic fragment described above The product obtained by the ligarion is digested with EcoRI to isolate the complete open reading frame with the 5' protruding cohesive ends at both oermini, which is then subcloned at the EcoRI
site of pUCI 19. The two synthetic regions in the obtained plasmid, pFPEP-EE, are sequenced and the mutated nucleotide sequences are confirmed. The resulting plasmid is pFPEP-EE.

In the next step of the vector construction, the whole coding region, together with a shoh du .. I~D~ aln non-coding region, is cleaved out from pFPEP-EE by EcoRI. The fragment is then insehed into the EcoRI site of the expression vector, pKh223-3, to provide pKK-FPEP in which the n, ~ of the prolyl ~fl~ gene is under the control of the tac promoter.

The replication origin of pKK223-3 originates from pBR322 and the copy number of this expression vector in a single cell is usually low. Because of the higher dose effect of the gene a high copy number plasmid is preferable as an expression vector for a higher expression level. Therefore, a) a set of the promoter and the coding region or b) a set of the promoter, the coding region and the terminator is excised by BamHI or BbilI,,ly, and ~ l ,t ~ into the high copy number plasmid, pUCI 19. Since the peptidase gene is transferred from pKK-FPEP together with the tac promoter, the original lac promoter of pUCl 19 is removed by PvuII in order to avoid the double promoter. In ~ W096100293 21 92091 r~

. ~

between the blunt ends generated by this PvuII digestion either the gene set a) or the set b), which has been blunted by T4 DNA poly~ aD~, is inserted to give pUK-FPEP-a or pUK-FPEP-b, 1~D~ Y (Fig. 2).

ExamPle 2: Random of the stariin~ DNA bY chemical treatment The plasmid pUK-FPEP-b is digested with EcoRI and Pstl (Fig. 2). The 2.3 kb DNA
fragment containing the l~luly~ gene is separated from the major fragment of theplasmid(thevectorfragment)andsubclonedintopUC118OrPUCll9.The single-stranded DNA of the, ~ plasmids is prepared by the aid of an M13 helper phage and treated at 20~C with 1 M sodium nitrite in 0.25 M sodium acetate, pH 4.3, for 15-20 min, 3 M formic acid for 5-20 min or 20 % hydrazine for 5-20 min by the method of Myers et al. (Science, 1985, ~g, 242-247). The chemically ,, ' single-stranded DNA is made into a duplex form by avian Ill y ~ l.~ , virus (AMV) reverse n r ~ . The resulting 2.3 kb fragment is ligated back with the vector fragment to re-construct the expression plasmid, pUK-FPEP-b.

ExamPle3: Screenin~forheat-stable ~lulyl~
With the ligation mixmre obtained in the Example 2, E coli is i ~ ' and plated out on ~ - filters placed on the TY agar plates (1 % Bacto-tryptone, 0.1 % glucose, 0.8 % NaCI, 1.5 % agar, pH 7.0) containing 60 ~Lg/ml ampicilin. These plates areincubated at 37~C and, when colonies appear on the filters, a replica of colonies on each filter is made on another filter. Both original and replica filters are incubated on the agar plates at 37~C for a few hours The otiginals are then stored at 4~C and tbe replicas subjected to cell Iysis.

The cell Iysis is perfonned by soaking the filters at room I , - sequentially in two lysis buffers; first 2 mg/ml Iysozyme in 50 mM Tris-HCI (pH 8.0) for 20 min and secondly 1 % Triton X-100 in 50 mM Tris-HCI (pH 8.0) for 5 min. The bacterial debris is gently scraped from the surface of the filters in 20 mM sodium phosphate (pH 7.0). The filters are then blocked witb 3 % BSA in 20 mM sodium phosphate (pH 7.0) or 5 % skim milk in tbe same buffer to prevent non-specific adsorption on the filters of a substrate peptide for following acùve staining of IJ-ulyl~ ....1. '1" .1.1;.1_~.. After washing out the blocking buffer with 20 mM sodinm phosphate (pH 7.0), the filters are incubated in 20 mM sodium phosphate (pH 7.0) at 50~C for 15 min. To identify clones that carty residual activity of the enzyme after the heat-treatment, the filters are subjected to the acùve staining with 0.1 mM Z-Gly-Ala-Pro-l~ n~ Ll.ylà.llide and 0.01 % Fast Garnet GBC sulfate salt. The W0 96100293 2 1 ~20q 1 P~ .5C . o colonies producing heat-stable l~lulyl lu~ are visualized as red spots on the filters by the formation of an insoluble coupling product of the Fast Garnet and the hydrolysis product of the substrate, ~- , ' ' yl~..,ille. The clones ~ g to the stained colonies are isolated from the original filters separately stored at 4~C.

By this method 24,000 colonies are screened and 134 clones are picked up as candidates producing heat-stablo l~lulyl~ - The isolated clones are then gridded onto a few r;m~ ' filters placed on the agar plates, and replicas of each filter are prepared again amd subjected to the secondary screening. In the secondary screening the replicas are each treated with a gradually increased l untiD~l ~;r.. -I;..,. of the cloneproducing the most heat-stable ~lu~ (PEP-15).

Exam~le 4~ r.~ u,. of the heat-stable UIUIV1 - '1UI~ UI;~
Throughout the p ; r. - l ;. . steps, ~. uly~ activity is assayed as follows. To 0.94 ml of 0.1 M potassium phosphate (pH 7.0), loo llglml bovine serum albumm, I mM
DTT, is added 0.05 ml of 4 mM Z-Gly-Pro-p--.iLIu~.ilidc (Z = N-bv~ylu~y~,~ul)ull~l) in 40 ~o dioxane. After 3 mim l~ ;. ., at 30~C 0.01 ml of a diluted solution of the enzyme is added to the mixture and changes of the adsorbance at 410 nm are followed with a Hitachi ~l~ m)p~ U-3210 at 30~C One unit of the enzyme activity is defined as the amount of the enzyme that releases I llmol of p--~ ' ~ per minute, g to 8.87 OD/min with this standard procedure.

The E. coli clone obt~uned by the screening for a heat-stable l~lulyl. ,~
(Example 3) is grown in 500 ml of a CIRCLEGROW medium containing 60 ,ug/ml ampicillin at 37~C After 12 hr, IPTG (1 mM) is added and the culture is continued for additional 16 hr. The E. coli cells are harvested~by c Il l; r Izr~ l and washed with cold 0.1 M Tris-HCI (pH 8.0). The following 1~ ; r~ procedures are carried out at 4~Cunless otherwise specified. The washed cells are re-suspended in 50 ml of 0.1 M Tris-HCI
(pH 8.0), 0.5 M sucrose, SmM EDTA. Lysozyme is added to a final c of 160 llglml~ and then the same volume of ice-cold water is added. After incubation for 30 min on ice, ~pl u~ l, formed by the osmotic shock and Iysozyme treatment are removed by ' v at 10,000 x g for ZO min. The ~ (the l,. . jpl~ ..i.
fraction) is diluted again with the same volume of ice-cold water and the pH is adjusted to 7 with I N HCI. The diluted solution is applied to a CM 52 column that has been r~p~ with 20 mM sodium phosphate (pH 6.2). The enzyme is eluted by a lineargradient from 0 to 0.25 M NaCI. The active fractions are pooled and r-- ' using an W096/00293 21 920 9 ~ r~ s .

''' iR

Amicon, ~ = ltrg~inn cell with a YM 30 mcmbrane, ard the is dialyzed against 2u mM sodium phosphate (pH 6.2). The ponfied heal-stable ~lulyl .
PEP-15 is judged to be l ~ by SDS ~JU~ ly- ' ' gel . l~ ~ u u ExamPle 5: Heat stabilitv ûf the mutated prolvl 1~" ,~ "li,l ~.
The specific activity of PEP-15 is found to be 87 units/mg proteir., close to the wild type erlzyme (124 units/mg protein). The heat-stability of the heat-stable ~ylulyll 1~ y~ , is compared with the wild-type with regard to rclative rates of losing activity by heat trcatmcnt. Each solution ûf the heat-stable l~lulyl~ IJ ~ amd the wild-type uly1 ~ .u,l~ ~ in 20 mM sodium phosphate, pH 7.0 (I mg/ml) is incubated at 54~C
and aliquots are removed and assayed for residual enzyme activity after various time irltervals over 120 min. When the relative activities of the heat-stable l,lul~l. ,.,i,.l.. l~l;.
and the wild-type ~lulyl -'~l, 'l' l~li~l-~ are plotted as a function of incubation time, both give linear relation in 'o,, ' plots, indicating the loss of the activities obeyed first~der kinetics. From the slopes of the lines, the rate constants of the heat UI~L;~ IU
for the wild-type amd the heat-stable ,IJlUly~ . are estimated to be 2.17 x 10-2 and 6.24 x 10-3 min~~ ,olJ~ ,ly. The rate constants clearly ~ '~ that PEP-15 is stabilized by a factor of 3.5 at 54~C

Example 6: DNA sequencin~ of the mutated ~ene encodin~ for PEP-IS
To confirm a change in the amino acid sequence of PEP-15, nucleotide sequence of the mutated gene is ~- ' The expression plasmid is isolated from the clone producing PEP-15 using a QIAGEN tip-100, according to the procedure ~ru..~ {I by the supplier. To determine whole sequence of the mutated ~lul.~ JIJ~ gene, 22 sequencing primers that cover all the coding region are ~ , and the nucleotide sequence is ~-t~in,~A by an Applied Biosystems Model 307A DNA sequencer using the Taa Dye Deoxy Terminator Cycle S~ n~ng kit. The nucleotide sequence coding for PEP-IS is compared with that of the wild-type to reveal the ' of nucleotide G to A leading to the insertion of Lys instead of Glu at the amino acid position 477.
.
The coding region was recloned in the native vector fragment in order to prepareexpression vectorpUK-FPEP-15.

Example 7: Random ~ r.~ usin~ a modified PCR method In another ~ "h~l; 1 ~ of the present invention poly~ ,.~e chain reaction (PCR) method is adopted to generate random ~ in the structure gene of IJIulyl. 1.~

_ _ _ . . .. ...

wo 96/00293 ~ 5 ~

The PCR ~, ~ on the contrary to the chemical " ~ amplifies a region of the gene to be mutated under conditions that reduce the fidelity of DNA synthesis by Taq p~l~ The 2.3 kb EcoRI-Psd DNA fragment of pUK-FPEP-b encoding the ylul~ is amplified with a pair of 25 ~ J~ c primers, which are ~- rl- y to the regions " Iy flanking the EcoRI and Psd sites.

To introduce a limited number of, i.e., a simgle or at most a few, baseefficiently in the l,lul~ gene, effects of several factors on the fidelity of the pol~ ,.~., are exammed. The factors v~ t~,d involve addition of MnCI2 or DMSO, alteration on MgC12 ~ and the deoxy nucleotide pool imbalance. A basic reaction mixture contains, 10 mM Tris HCI (pH 8.3), 50 mM KCI, 1.5 mM MgCI2, 0.01 llyo (w/v) bovine serum albumin, 0.2 mM each of four dNTP's, 10 ng/ml of the templateplasmid, 1 ~LM each primers and 25 units/ml Taq l~ol~ , and the " ~ of the basic reaction mixture used in the present invention are as follows (. ~l;ri. -~ - are made in five different ~ 'l"; ' '~):

1) 50-100 IlM of MnCI2 is added with an increased ~ (5 mM) of MgCI2.
2) 5-20 % of DMSO is added.
3) The of MgCI2 is changed from 5 mM to 10 mM.
4) The . of dGTP, dTTP and dCTP are held constant at I mM while the . ~ c - ~ of dATP is lowered from O.2 mM to O.l mM in the presence of 5 mM
MgCI2.
5) While the c ~ of dATP, dCTP and dTTP are held constant at O. l mM, the . ~. . .n,n;. .. ~ of dGTP is valied from 0.5 mM to I mM in the presence of 1.5 - 2 mM
MgCI2.

All five l~ ;ri~ ~li-- - of the basic reaction mixture give many clones producing heat stablel,ulJl 1.,1~ Forimprovingheatstability.the. ~lir~ arerepeatedor different of the above ,"~l;l;. l;..,l methods are applied ~ ~,..Lid~

In the present invention, FPEP-227 was prepared from wild type by ., .~l;ri. li.. . 3), FPEP-361 from FPEP-227 by . ~liri~ 1), and FPEP-407 from FPEP-361 by ""~ 3)-The PCR is carried out by using Gene Amp~ PCR system 9600 (Pet~in-Elmer) by melting the template DNA al 94~C for 1 min and annealing with the primers at 50~C for ~ wos6/002s3 2 1 q ~ 0 9 ~ ~

- 39 - :

1 min. Chain extension is initiated at 70~C for 4 min and a total of 25 cycles are performed. After the last cycle, the pol.~ - Liu.. at 70~C is extended for additional 7 min. To construct a randomly mutated DNA library, the amplified DNA fragment is digested with EcoRI and Pstl and pufified by the GENECLEAN II kit. The resultant2.3 kb EcoRI-Pstl fragment is ligated back to the intact vector fragment of pUK-FPEP-b.

The first cycle of the PCR ,. ~ is started from the wild-type gene. About 80,000 clones are generated by the PCR and screened by the same method as described in Example 3 but with a slightly higher ~ r (52~C) of the heat treatment, which yields a clone producing the most heat-stable l~lul.~ in the mutant library, PEP-227. : ~

ExamPle 8: Repeatin~ Tandom usin~ a modified PCR method Since A ~ of stabilizing mutations in a single molecule of the enzyme is expected to rcsult in an even larger illl~J~U .~ el.L in the stability than that obtained with FPEP-15 or FPEP-227, the PCR ~ -g~ - ~/screening cycle is repeated sequentially to mutations.

The isolated clone coding for FPEP-227 (pUK-FPEP-227) is used as a template for the second cycle of the PCR ~ to give a library consisting of about 40,000 clones.
Screening of the second generation library is performed similarly to yield that clone expressing the most heat-stable ~JIU~ A'5/ (PEP-361). In the third cycle the and screening are repeated once again starting from the expression plasmid encoding for PEP-361 (pUK-FPEP-361~ as a template. The: , of the heat-treatment is gradually increased as the cycle of the ,, 1~ , proceeds and, in the third cycle, that clone producing the most heat-stable l,.ulyl. 1.,~, ~.~;.1 .
(PEP-407) is identified among 50,000 clones by the screening with the heat-treatment at 61~C

It should be noted that the clones isolated and ~ 11 in the present examples are not the only clones that are prepared and are identified as coding for heat-stable enzymes.
The clones specified herein are only isolated as examples in order to teach the principle of the method.

Example 9: 1solation and ~ l 5~ tv~ of hi, hlY heat-stable v- ul Yl~ )v~ v~
From the clones producing the heat-stable ~ul,~ u~ PEP-227, PEP-361 and w096/00293 2192091 P~ "' e PEP-407, expression plasmids are isolated and named pUK-FPEP-227, pUK-FPEP-361 and pUK-FPEP-407, .c~ vly. The DNA sequences of the mutated structure genes carried on the plasmids are determined to follow structure changes in the amino acid sequenco of l,.ul~ p. I~ A'- It is found that a single or a few base ' areimtroduced in each random ~ /s.,lcv.-u.v cycle and it is also confumed that the mutations are r ' ' in the gene by the sequential v g cycles (Table 3).

Table 3: Base ~ identified in mutated genes encoding heat-stable ~ulyl 1.. ~ and resulting amino acid ~ .lace pUK-FPEP-227 pUK-FPEP-361 ~ pUK-FPEP-407 458G~C 458G~C 458G~C
~E48Q) (E48Q) ~E48Qsf 701G~A 701G~A
(A129T) (A129Tf 2,214 G ~ T 2,214 G ~ T
~G633V) (G633V) 2,239 T ~- C 2,239 T ~ C
(silent) (silent) 467T ~ C
(F51L) The specific activitles of the purified heat-stable IJlul~ l lul~ -f ~ PEP-227, PEP-361 and PEP-407 are found to be 116, 87 and 69 units/mg protein, lc~ . The specific activities of the heat-stable IJiulyl~ 1up. ~ 1A s are ~,ulu~ blc with that of the wild-type (124 uni~slmg protein) but shows a tendency to decrease as the amino acid . ~
are: ' ' First, the heat-stability of the heat-stable l~lul~f l. L~ 5 are evaluated with values of T50, at which L~ ,lflLulc the enzyrne loses 50 % activity in 30 min. In 20 mM sodium phosphate ~pH 7.0), 0.2 mg/ml of the wild-type or one of the heat-stable ~ul~ n~ is incubated in Gene AmpTM PCR system 9600 ~Perkin-Elmer) for 30 min at various tv.~ ulvS ranging from 46 to 69~C The residual actiYities are measured and plotted as a function of the tVul~Jvldlul~, ~Fg. 3). While the T50 : . , : ~: .... .. . .. .. . ... . . ~ _ _ ~ _ _ _ . .. _ _ _ ..

wo 96/00293 ~ 2 ~ 9 2 0 q I

of the wild-type is estimated to be 535~C from the figure, the Tso of the heat-stable yl~1yl~ n,l: -, PEP-227, PEP-361 and PEP-407 shifts upward to 565, 59.5 and 60.5~C, ~ y~ iv~ . The changes of the Tso, as an indication of heat-stability, clearly the l A~gl ;lity of yl~ y~ is improved as the ~ '~ g cycle proceeds.

To evaluate more ~, v.,l~ the ihllylu.. of the heat-stability, the enzyme solutions are incubated at a constant i A ' and changes in the remaining activities are followed over 120 min. In 20 mM sodium phosphate (pH 7.0), 0.2 mg/ml of eachenzyme solution is incubated at 60~C. At various time intervals, aliquots are removed and residual activities are measured. All h~. v_Lion processes of the wild-type and the three heat-stable YI~ Y ~ obeyed the frst order kinetics, showing lines when rcsidual activities are plotted ' " ' "~, as functions of incubation time. The hI~LiV~ UII rate contants of the wild-type, PEP-227, PEP-361 and PEP407 are estimated from the slopes to be 4.27 x 10-1, 7.08 x 10-2 1.32 x 10-2 and 7.16 x 10-3 min~l, ~o_lJ~ iv~,ly. The rate consrgnts ~~ large hll~ in stabilities of the heat-stable ~ lyll 1 y ~ the , ~ reaches factoprs of 60 with the most heat-stable ~ llyl~ 1. y~- l" ;.1A ~. obtained in the final cycle of the ~ J~

DePositedMi~,~AJ r.~
E. coli TGI/pFPEPO3 was deposited under the Budapest Treaty with the DSM - Deutsche Sammlung von Mihw~ ,.l und 7~ GmbH. MA~h.,.~dol Weg lb, D-38124 B.~ .,h.. g, Fedelal Republic of Germany, as DSM 9250 on June 16, 1994.

WO 96/00293 r~ ~5.'~ l_, O

SEQIJENCE LISTING

(1) GENERAL INFORNaTION:

(i) APPLICANT:
(A) NAME: Japat Ltd (B) STREET: ~lybeckstr. 141 (C) CITY: 3asel (E) COUNTRY: SCHWEIZ
(F) POSTAL CODE (ZIP): 4057 (G) TELEP_ONE: +41 61 69 11 11 (_) TELEFAX: + 41 61 696 79 76 (I) TELEX: 962 991 (ii) TITLE OF INVENTION: l~eat stable enzyme (iii) N~NBER OF ~ S: 7 (iv) COMPUTER READA3LE FORM:
(A) MEDIIJM TYPE: Floppy disk (B) COMP~JTER: IBN PC I ;hl r~
(C) OPERATING SYSTEM: PC-DOS/MS-DOS
(D) SOFTWARE: PatentIn Release #1.0, Version #1.25 (EPO) (2¦ INFORNATION FOR SEQ ID NO: 1:

(i) SEQUENOE CE~RACTERISTICS:
(A) LENGTEI: 2636 base pairs (B) TYPE: nucleic acid (C) STRANn~n~cs: double (D) TOPOLOGY: l;near (ii) MOLECULE T~PE: DNA (genomic) (iv) ORIGINAL SOIJRCE:

~ W096~0293 2 ~ 920 91 r~

(A) ORGA~ISM: Flavobacterium meningosepticum (ix) FEAT~RE:
(A) NAME/KEY: promoter (B) LOCATION: 1..259 (ix) FEAT~RE:
(A) NAME/REY: CDS
(B) LOCATION: 260..316 (D) OT~ER INFORMATION: /function- nsignal sequence"

(ix) FEATURE:
(A) NAME/KEY: CDS
(B) LOCATION: 317..2374 (D) OT~ER INFORMATION: /product "coding region for mature prolylendop-~rt~ Qe"

(ix) FEAT~RE:
(A) NAME/KEY: misc_feature (B) LOCATION: 2375..2377 (D) OT~ER INFORMATION: /function= nStop codon"

~xi) SEQ~ENOE DESCRIPTION: SEQ ID NO: 1:

GTTGACGGTA AA~TA~TATT TACTAAAAAG A~A~ATAA~A GGTCTTACGT ATCTGTAGCA 60 CCAGATGCTT AATTAAAGCA TTTTATAAAA ATTAAAACCT ~AA~AAh~T TGAGGTTTTT 120 A AAA~CTAA~A G~LLl~l~AA ACCTGTTAGG TTTATTGTGT ATAGGGGTTA 180 AGTGATACAT ATTTATACTG TGCTGAAATG CGAATCTGAT TATTCGA~AA Ll~LCC~lAT 240 Met Lys Tyr Asn Lys Leu Ser Val Ala Val Ala W096/00293 21 ~20 q 1 r~ S

Ala Phe Ala Phe Ala Ala Val Ser Ala Gln A5n Ser Asn Val Leu Lys Tyr Pro Glu Thr Lys Lys Val Ser ~is Thr Asp Thr Tyr Phe Gly Thr Gln Val Ser Asp Pro Tyr Arg Trp Leu Glu Aap Asp Arg Ala Glu Asp Thr Lys Ala Trp Val Gln Gln Glu Val Lys Phe Thr Gln Asp Tyr Leu Ala Gln Ile Pro Phe Arg Asp Gln Leu Lys Lys Gln Leu Met A5p Ile TGG AAT TAT GAG AAA ATT ~ Q G Q CCG TTT AAA AAA GGT AAA TAC ACC 580 Trp Asn Tyr Glu Lys Ile Ser Ala Pro Phe Ly5 Lys Gly Lys Tyr Thr Tyr Phe Ser Lys Asn Asp Gly Leu Gln Ala Gln Ser Val Leu Tyr Arg lO0 Lys Asp Ala Ala Gly Lys Thr Glu Val Phe Leu Asp Pro Asn Lys Phe Ser Glu 1ys Gly Thr Thr Ser Leu Ala Ser Val Ser Phe Asn Lys Lys ~ WO9~OOZ93 21 92091 r. ~[

Gly Thr Leu Val Ala Tyr Ser Ile Ser Glu Gly Gly Ser Asp Trp Asn 140 1~5 150 Lys Ile Ile Ile ~eu Asp Ala Glu Thr Ly9 Lys Gln Leu Asp Glu Thr Leu Leu Asp Val Lys Phe Ser Gly Ile Ser Trp Leu Gly Asp Glu Gly Phe Phe Tyr Ser Ser Tyr Asp Lys Pro Lys Glu Gly Ser Val Leu Ser Gly Met Thr Asp Lys Pis Lys Val Tyr Phe qis Lys Leu Gly Thr Lys Gln Ser Gln Asp Glu Leu Ile Ile Gly Gly As~ Lys Phe Pro Arg Arg Tyr Ile Gly Ala Tyr Val Thr Asp Asp Gln Arg Tyr Leu Val Val Ser 235 240 2~5 GCT GCA AAT GCA.ACC AAC GGA AAC GAG CTT TAC ATT AAA GAC CTG AAG 1108 Ala Ala Asn Ala Thr Asn Gly Asn Glu Leu Tyr Ile Ly9 Asp Leu Lys Asn Lys Thr Asp Phe Ile Pro Ile Ile Thr Gly Phe Asp Ser Asn Val W096/~293 r~
21 92091 ~

Asn Val Ala Asp Thr A3p Gly Asp Thr Leu Tyr Leu Phe Thr Asp Lys Asp Ala Pro Asn Lys Arg Leu Val Lys Thr Thr Ile Gln Asn Pro Lys Ala Glu Thr Trp Lys Asp Val Ile Ala Glu Thr Thr Glu Pro Phe Gln ATC AAT ACG GGA GGC GG~ TAT TTC TTT GCT ACT TAT ATG AAA GAT GCA 1348 Ile Asn Thr Gly Gly Gly Tyr Phe Phe Ala Thr Tyr Met Lys Asp Ala Ile Asp Gln Val Lys Gln Tyr Asp Lys Asn Gly Lys Leu Val Arg Ala 345 350 355 . 360 ATA AAA TTA CCG GGA AGT GGT AAT G Q AGC GGT TTT GGG GGT GAA AA~ 1444 Ile Lys Leu Pro Gly Ser Gly Asn Ala Ser Gly Phe Gly Gly Glu Lys Thr Glu Lys Asp Leu Tyr Tyr Ser Phe Thr Asn Tyr Ile Thr Pro Pro ACG ATC TTT AAA.TAT AAT GTA ACA A Q GGT AAT TCT GAA GTT TAC QG 1540 Thr Tle Phe Lys Tyr Asn Val Thr Thr Gly Asn Ser Glu Val Tyr Gln Lys Pro Lys Val Lys Phe Asn Pro Glu Asn Tyr Val Ser Glu Gln Val W096/00293 .~ r- . .
~ 2~ 920~1 Phe Tyr Thr Ser Ser A3p Gly Thr Lys Ile Pro Met Met Ile Ser Tyr AAG AaA GGC CTG AAA AAA GAC GGT AAA AAC CCT A Q ATA TTA TAC AGC 1684 Lys Lys Gly Leu Lys ~y8 Asp Gly Lys Asn Pro Thr Ile Leu Tyr Ser Tyr Gly Gly Phe Asn Ile Ser Leu Gln Pro Ala Phe Ser Val Val Asn 460 465 '470 Ala Ile Trp Met Glu Asn Gly Gly Ile Tyr Ala Val Pro Asn Ile Arg GGT GGT GGA GAA TAT GGT AAG AAA TGG QT GAT GCC GGA ACT AaA ATG 1828 Gly Gly Gly Glu Tyr Gly Lys Lys Trp Eis Asp Ala Gly Thr LYB Met Gln Lys Lys Asn Val Phe Asn Asp Phe Ile Ala Ala Gly Glu Tyr Leu CAG A~A AAC GGT TAT ACA TCT AAG GAA TAT ATG GCG CTT TCC GGA CGT 1924 Gln Lys Asn Gly Tyr Thr Ser Lys Glu Tyr Met Ala Leu Ser Gly Arg TCC AaC GGC GGT.CTT CTT GTA GGG GCT ACG ATG ACA ATG CGC CCT GAT 1972 Ser Asn Gly Gly Leu Leu Val Gly Ala Thr Met Thr Met Arg Pro Asp TTG GCA AaA GTT G Q TTC CCG GGA GTA GGA GTA CTG GAT ATG CTT CGT 2020 ~eu Ala Lys Val Ala Fhe Pro Gly Val Gly Val Leu Asp Met ~eu Arg 555 ~ 560 565 _ _ _ _ _ _ _ _ _ _ _ .. . . .. .. . . _ . _ . . .. . _ _ W096l00293 21 92091 r~ s O

Tyr Asn 1ys Phe Thr Ala Gly Ala Gly Trp Ala Tyr Asp Tyr Gly Thr Ala Glu Asp Ser 1ys Glu Met Phe Glu Tyr Leu 1ys Ser Tyr Ser Pro Val ~is Asn Val Lys Ala Gly Thr Cys Tyr Pro Ser Thr Met Val Ile Thr Ser Asp ~is Asp Asp Arg Val Val Pro Ala ~is Ser Phe 1ys Phe Gly Ser Glu 1eu Gln Ala Lys Gln Ser Cys Lys Asn Pro Ile Leu Ile Arg Ile Glu Thr Asn Ala Gly ~is Gly Ala Gly Arg Ser Thr Glu Gln Val Ual Ala Glu Asn Ala Asp Leu Leu Ser Phe Ala 1eu Tyr Glu Met GGA ATT AAA AGT.TTA AAA TAGATTTCAA AT~TA~T~ T~ r~ c 2404 Gly Ile Lys Ser Leu 1ys A~ GATTTGCCTG TTTTTTTATG ATACTATTGA GTTTGGATTA TGTT~AATAG 2464 ATTAGATCAT GAGATTTATA TCTCAGGAaA TGATTAACTT T~ TCTTATACAA 2524 ~ W096/00293 r~

TGGAaAAT Q TGACATGA Q ACTTTAGTAC AGGTAATGAA TACTTTGAAA AGAAGAGGCG 2584 TGr~r~rA AATCCAGATG ACAGATGATA GGAaATTTAT ACTT Q GAAT TC 2636 ~2) INFORMATION FOR SEQ ID NO: 2:

(i) SEQ~ENCE rR~R~rTFRTcTIcs:
(A) LENGTR: 44 base palrs (B) TYPE: nucleic acid tc) STR~FnNF.~s double ~D) TOPOLOGY: linear (ii) MOLEC~LE TYPE: DNA lgenomic) (ix) FEAT~RE:
(A) NAME/~EY: misc_feature (B) 10 QTION: l..44 tD) OTPER INFORMATION: /function= "deletes ScaI and PvuII
site from prolylendopeptidase gene"
/product- "synthetic oligonucleotide"

(xi) SEQ~ENCE DESCRIPTION: SEQ ID NO: 2:

GGGAGTAGGA GTTCTGGATA T~llC~llA TAATAAGTTT ACTG 44 (2) INFOP~ TION.FOR SEQ ID NO: 3:

(i) SEQUENCE r~RArT~RT~TIcs:
(A) LENGT~: 54 base palrs (B) TYPE: nucleic acid (C) STR~Nn~n ~.S.C: double (D) TOPOLOGY: linear W096/002~3 r~r''c : .

(ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/REY: misc_feature (B) LOCATION: 1..54 (D) OTHER INFORMATION: /function- "creates new EcoRI site in prolylendopeptidase gene"
/product- "synthetic o~gnn~ o~tide"

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 3:

AATTCATGAA GTA~A~rAAA ~LLL~L~L~G CAGTTGCAGC ~lIL~LLLL GCAG 54 (2) INFORMATION FOR SEQ ID NO: 4:

(i) SEQUEN OE ~RAR~C.TRRT~TICS:
(A) LENGT~: 23 base pairs (B) TYPE: nucleic acid (C) STRAND~N~s: single (D) TOPOLOGY: lirear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:
(A) NAME/KEY: misc_feature (B) LOCATION: 1..23 (D) OTEER INFORMATION: /function- "for preparation of olignn--~loot~o SEQ ID No 3'l /product- "synthetic ~1;gonl1~lootide Ul"

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 4:

~ W09~0~293 r~

~1 92091 AATTCATGAA GTArAArAAA CTT 23 (2) I~7FORMATION POR SEQ ID NO: 5:

(i) SEQI.7ENCE rn~Rl~rT~R7~TIcs:

(A) LENGTL: 31 base pairs (B) TYPE: nucleic acid (C) STRANn~.~NFCS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE:

(A) NAME/KEY: misc_feature (B) 10CATION: 1..31 (D) OT~ER INPORMATION: /function= nfor preparation of n7i~nn~rl~o~i~o SEQ ID No. 3"

~product- nsynthetic olignnnnl~Qtide U2"

(xi) SEQUENOE DESCRIPTION: SEQ ID NO: 5:

~G~AG TTG~rC~TT I~ A G 31 (2) INFORMATION FOR SEQ ID NO: 6:

(i) SEQUENCE r.~7ARA~I r K ~ S

(A) 1ENGTU: 25 base pairs (B) TYPE: nucleic acid (C) STRANnF~NF.~.~: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) W096/00293 l~~

(ix) FEAT~E:
(A) NAME/KEY: misc_feature (B) LOCATIoN 1..25 (D) OTHER INFORMATION: /function= nfor preparation of oligonucleotide SEQ ID No. 3"
/product= "synthetic oligonucleotide Ll"

(xi) SEQ~ENCE DESCRIPTION: SEQ ID NO: 6:

r~r~r.~rT TTGTTGT~CT TCATG : 25 (2) INFORMATION FOR SEQ ID NO: 7:

(i) SEQ~ENCE rR~R~rTF.~TCTICS
(A) LENGTH: 25 base pairs (B) TYPE. nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLEC~LE TYPE: DNA (genomic) (ix~ FEATbRE:
(A) NAME/REY: misc ~eature (B~ LOCATION: 1..25 (D) OTHER INFOPI~ATION: /function= nfor preparation of oligonucleotide SEQ ID No. 3"
/product;- "synthetic oligonucleotide L2"

(xi) SEQ~ENCE DESCRIPTION: SEQ ID NO: 7:
.

CTGCAAAAGC AAAGGCTGCA ACTGC ~ 25

Claims (38)

1. A heat-stable prolylendopeptidase obtainable by a process comprising the steps (a) mutagenization of a starting DNA coding for a prolylendopeptidase, (b) generation of a library of mutated DNA sequences obtained in (a), (c) screening the library for a gene coding for a prolylendopeptidase with improved heat-stability, and (d) expression of the gene obtained under (c) and isolating the expression product.

2. A heat-stable prolylendopeptidase according to claim 1 which is derivable from a wild-type prolylendopeptidase of a prokaryote.

3. A heat-stable prolylendopeptidase according to claim 1 which is derivable from a wild-type prolylendopeptidase of Flavobacterium spec.

4. A heat-stable prolylendopeptidase according to claim 1 which is derivable from a wild-type prolylendopeptidase of F.menigosepticum.

5. A heat-stable prolylendopeptidase according to claim 1 which is derivable from a wild-type, prolylendopeptidase of F. meningosepticum strain IFO 12535(ATCC 13253).

6. A heat-stable prolylendopeptidase according to claim 1 which is derivable from the prolylendopeptidase having the sequence with SEQ ID No. 1.

7. A heat-stable prolylendopeptidase according to claim 1 selected from the group of heat-stable prolylendopeptidase having SEQ ID No.1 with amino acid Glu48, Phe51,Ala129, Gly633 and/or Glu477 replaced by another amino acid.

8. A heat-stable prolylendopeptidase according to claim 1 selected from the group of heat-stable prolylendopeptidase having SEQ ID No. 1 with amino acid Glu48 replaced by Gln, Phe51 replaced by Leu, Ala129 replaced by Thr, Gly633 replaced by Val and/or Glu477 replaced by Lys.

9. A heat-stable prolylendopeptidase according to claim 1 selected from the group of heat-stable prolylendopeptidases having SEQ ID No. 1 with amino acid (a) Glu48, (b) Glu48, Ala129 and Gly633, (c) Glu48, Ala129, Gly633 and Phe51, and (d) Glu477 replaced by another amino acid.

10. A heat-stable prolylendopeptidase according to claim 1 selected from the group of heat-stable prolylendopeptidases having SEQ ID No. 1 with amino acids (a) Glu48 replaced by Gln (PEP-227), (b) Glu48 replaced by Gln, Ala129 replaced by Thr andGly633 replaced by Val (PEP-361) (c) Glu48 replaced by Gln, Ala129 replaced by Thr, Gly633 replaced by Val and Phe51 replaced by Leu (PEP-407), and (d) Glu477 replaced by Lys (PEP-15).

11. A recombinant DNA molecule comprising a DNA sequence coding for heat-stable prolylendopetidase, which DNA sequence is derivable from a DNA sequence coding for a wild-type prolylendopeptidase by a method comprising the steps (a) mutagenization of a starting DNA coding for a prolylendopeptidase, (b) generation of a library of mutated DNA sequences obtained in (a), and (c) screening the library for a gene coding for a prolylendopeptidase with improved heat-stability.

12 A recombinant DNA molecule according to claim 11 comprising a DNA sequence coding for a heat-stable prolylendopeptidase derivable from a wild-type prolylendeopeptidase of a prokaryote.

13. A recombinant DNA molecule according to claim 11 comprising a DNA sequence coding for a heat-stable prolylendopeptidase derivable from a wild-type prolylendopeptidase of Flavobacterium spec..

14. A recombinant DNA molecule according to claim 11 comprising a DNA sequence coding for a heat-stable prolylendopeptidase derivable from a wild-type prolylendopeptidase of F. menigosepticum.

15. A recombinant DNA molecule according to claim 11 comprising a DNA sequence coding for a heat-stable prolylendopeptidase derivable from a wild-type prolylendopeptidase of F. meningosepticum strain IFO 12535 (ATCC 13253).

16. A recombinant DNA molecule according to claim 11 comprising a DNA sequence coding for a heat-stable prolylendopeptidase derivable from a wild-type prolylendopeptidase having the sequence with SEQ ID No. 1.

17. A recombinant DNA molecule according to claim 11 comprising a DNA sequence coding for a heat-stable prolylendopeptidase derivable from a DNA molecule having the sequence with SEQ ID No. 1.

18. A recombinant DNA molecule according to claim 11 comprising a DNA sequence coding for a heat-stable prolylendopeptidase selected from the group of heat-stable prolylendopeptidases having SEQ ID No. 1 with amino acid Glu in position 48 ("Glu48"), Phe51, Ala129, Gly633 and/or Glu477 replaced by another ammo acid.

19. A recombinant DNA molecule according to claim 11 comprising a DNA sequence coding for a heat-stable prolylendopeptidase, which recombinant DNA molecule is selected from the group of recombinant DNA molecules having the sequence of SEQ ID
No. 1 with the codons for amino acid Glu in position 48, Phe51, Ala129, Gly633 and/or Glu477 replaced by a codon coding for another amino acid.

20. A recombinant DNA molecule according to claim 11 comprising a DNA sequence coding for a heat-stable prolylendopeptidase selected from the group of heat-stable prolylendopeptidases having SEQ ID No. 1 with amino acid Glu48 replaced by Gln, Phe51 replaced by Leu, Ala129 replaced by Thr, Gly633 replaced by Val and/or Glu477 replaced by Lys.

21. A recombinant DNA molecule according to claim 11 comprising a DNA sequence coding for a heat-stable prolylendopeptidase selected from the group of DNA molecules having SEQ ID No. 1 with the codon coding for amino acid Glu48 replaced by a codon coding for Gln, the codon coding for Phe51 replaced by a codon coding for Leu, the codon coding for Ala129 replaced by a codon coding forThr, the codon coding for Gly633replaced by a codon coding for Val and/or the codon coding for Glu477 replaced by a codon coding for Lys.

22. A recombinant DNA molecule according to claim 11 comprising a DNA sequence coding for a heat-stable prolylendopeptidase selected from the group of heat-stable prolylendopeptidase having SEQ ID No. 1 with amino acid (a) Glu48, (b) Glu48, Ala129 and Gly633, (c) Glu48, Ala129, Gly633 and Phe51, and (d) Glu477 replaced by another amino acid.

23. A recombinant DNA molecule according to claim 11 comprising a DNA sequence coding for a heat-stable prolylendopeptidase selected from the group of DNA molecules having SEQ ID No. 1 with the codons coding for amino acid (a) Glu48, (b) Glu48, Ala129 and Gly633, (c) Glu48, Ala129, Gly633 and Phe51, and (d) Gly477 replaced by codons coding for another amino acid.

24. A recombinant DNA molecule according to claim 11 comprising a DNA sequence coding for a heat-stable prolylendopeptidase selected from the group of heat-stable prolylendopeptidases having SEQ ID No. 1 with amino acids (a) Glu48 replaced by Gln (PEP-227), (b) Glu48 replaced by Gln, Ala129 replaced by Thr and Gly633 replaced by Val (PEP-361) (c) Glu48 replaced by Gln, Ala129 replaced by Thr, Gly633 replaced by Val and Phe51 replaced by Leu (PEP-407), and (d) Glu477 replaced by Lys (PEP-15).

25.A recombinant DNA molecule according to claim 11 comprising a DNA sequence coding for a heat-stable prolylendopeptidase selected from the group of DNA molecules having SEQ ID No. 1 with (a) the codon coding for amino acid Glu48 replaced by a codon coding for amino acid Gln, (b) the codon coding for amino acid Glu48 replaced by a codon coding for amino acid Gln, the codon coding for amino acid Ala129 replaced by a codon coding for amino acid Thr and the codon coding for amino acid Gly633 replaced by a codon coding for amino acid Val (c) the codon coding for amino acid Glu48 replaced by a codon coding for amino acid Gln, the codon coding for amino acid Ala129 replaced by a codon coding for amino acid Thr, the codon coding for amino acid Gly633 replaced by a codon coding for amino acid Val and the codon coding for amino acid Phe51 replaced by a codon coding for amino acid Leu, and (d) the codon coding for amino acid Glu477 replaced by a codon coding for amino acid Lys.

26. A recombinant DNA molecule according to claim 11 comprising a DNA sequence coding for a heat-stable prolylendopeptidase selected from the group of DNA molecules having SEQ ID No. 1 with (a) the nucleic acid G in position 458 ("458 G") replaced by C, (b) 458 G replaced by C, 701 G replaced by A, and 2214 G replaced by T (c) 458 G

replaced by C, 701 G replaced by A, and 2214 G replaced by T, and 467 T replaced by C, and (d) 1745 G to A.

27. A recombinant DNA molecule according to claim 11 comprising a DNA sequence coding for a heat-stable prolylendopeptidase selected from the group of DNA molecules coding for a heat-stable prolylendopeptidase in pUK-FPEP-15, in pUK-FPEP-227, inpUK-FPEP-361, or in pUK-FPEP-407.

28. A hybrid vector comprising a recombinant DNA molecule according to claim 11.

29. A hybrid vector according to claim 28 comprising a recombinant DNA molecule of claim 11 under the control of a promoter for the expression of the encoded heat-stable prolylendopeptidase in a suitable transformed host.

30. A hybrid vector according to claim 29 for the expression of the encoded heat-stable prolylendopeptidase in E.coli.

31. A hybrid vector according to claim 30 selected from the group consisting of pUK-FPEP-15,pUK-FPEP-227,pUK-FPEP-361. and pUK-FPEP-407.

32. A hybrid vector according to claim 29 for the expression of the encoded heat-stable prolylendopeptidase in a Baculovirus/insect cell expression system

33. A transformed host comprising a recombinant DNA molecule according to claim 11.

34. A method for the preparation of a DNA molecule according to claim 11, said method comprising (a) mutagenization of a starting DNA coding for a prolylendopeptidase, (b) generation of a library of mutated DNA sequences obtained in (a), and (c) screening the library for a gene coding for a prolylendopeptidase with improved heat-stability.

35. A method for the production of heat-stable prolylendopeptidase according to claim 1 comprising the steps (a) transformation of a suitable host with an expression vector according to claim 29, (b) isolation of heat-stable prolylendopeptidase.

36. A method according to claim 35 in which the host is E. coli.

37. A method according to claim 35 in which a Baculovirus expression system is used.

38. A method for the preparation of a transformed host according to claim 33 comprising treating a suitable host cell under transforming conditions with a recombinant DNA
molecule according to claim 11.