NO307304B1 - Transformed microbial host cell and DNA construct suitable for the production of a lipolytic enzyme and method for the production thereof - Google Patents

Abstract

Novel microbial host strains are provided which are transformed by a vector molecule comprising a DNA fragment encoding a lipolytic enzyme and a marker for selection, capable of producing active lipase. Said DNA fragment is preferably derived from a Pseudomonas species.

Description

The present invention relates to the production by recombinant DNA technology of an enzyme for the enzymatic breakdown of fatty materials, in particular a lipolytic enzyme which has certain characteristics which make it suitable for use as additives for detergents.

A particular problem associated with washing clothes is the removal of stains with a greasy character. Fatty dirt is currently emulsified and removed using a combination of elevated temperature and high alkalinity. There is, however, a strong tendency towards the use of relatively low washing temperatures, i.e. approximately 40°C or lower, conditions which are particularly unsuitable for removing greasy stains. It is therefore necessary to have additives for detergents which are effective at lower washing temperatures, stable in highly alkaline detergent solutions and stable under storage conditions in both solid and liquid detergent compositions. A group of enzymes that hydrolyze triglycerides are lipases (E.C. 3.1.1.3). Lipases are present in many places, in many different prokaryotic organisms as well as eukaryotic cells. Depending on the enzyme source, the substrate specificity as well as other characteristics including stability under different conditions vary greatly. Lipases have been used in detergent compositions, but those that were used had low washing efficiency under the washing conditions and, in addition, they did not meet the stability requirements for using the detergent.

Lipolytic enzymes that can exhibit lipase activity under modern washing conditions, i.e. that they are stable and effective at high detergent concentrations at high pH and at low washing temperatures, are produced by certain strains belonging to the species Pseudomonas pseudoalcaligenes, Pseudomonas stutzeri and Acinetobacter calcoaceticus (see European patent application EP -A-0218272). These species, on the other hand, are potentially pathogenic for plants and animals, and little data is available on the conditions for fermentation that are necessary for an efficient production process for lipase when used for these microorganisms.

It is therefore desirable to develop an efficient and safe way to produce lipolytic enzymes with the desired characteristics for detergent additives. On the other hand, it is desirable that the lipases are secreted by the host organism so that the enzyme can be isolated directly from the extracellular fluid of the fermentation mixture.

For lipases produced by Pseudomonas species, it is known that the culture conditions strongly influence the final location of these enzymes (Sugiura, In: "Lipases", eds. B. Borgstrom and H.L. Brockman (1984) pp. 505-523. Elsevier , Amsterdam). Problems often arise in providing efficient expression of heterologous genes in microorganisms, including incorrect folding of the proteins that are formed, protein degradation and mislocalization of the products. Harris, in: "Genetic Engineering", Vol. 4 (1983), Academic Press, New York. In E.coli, the use of secretion cloning vectors generally enables the transport of heterologous gene products into the periplasmic area and the products, on the other hand, are only rarely found in the culture medium; Lunn et al., Current Topics in Microbiol. and Immunol. 125 (1986) 59-74. Using E.coli as a host cell, cloned microbial lipases, described by Gotz et al., Nucleic Acids Res. 13 (1985) 5895-5906, Kugimiya et al., Biochem. Biophys. Res. Commun. 141 (1986) 185-190 and Odera et al., J. Ferment. Technol. 64 (1986) 363-371, excreted in a bad way in the culture medium.

Wohlfarth and Vinkler, J. Gen. Microbiol, 134 (1988) 433-440, reports on the physiological characterization of newly isolated lipase-deficient mutants from Pseudomonas aeruginosa strain PAO 2302 and on the chromosomal mapping and cloning of the corresponding gene. Bacillus species, particularly Bacillus subtilis strains, have been used with varying degrees of success as host strains for expression of both foreign and endogenous genes and for secretion of the encoded protein products. For an overview, see e.g. Sarvas, Current Topics in Microbiology and Immunology 125 (1986) 103-125, H.C. Wu and P.C. Tai, eds Springer Verlag; see also Himeno et al., F.E.M.S. Microbiol. Letters 35 (1986) 17-21.

US Patent No. 3,950,277 and British Patent Specification No. 1,442,418 describe lipase enzymes combined with an activator and calcium and/or magnesium ions, respectively, which are used to soak soiled fabrics and to remove triglyceride stains and dirt from polyester and polyester/cotton factory blends, respectively. Suitable microbial lipases disclosed include those derived from Pseudomonas, Aspergillus, Pneumococcus, Staphylococcus, Mycobacterium tuberculosis, Mycotorula lipolytica and Sclerotinia.

British Patent Specification No. 1-372,034 describes a detergent composition comprising a bacterial lipase produced by Pseudomonas stutzeri strain ATCC 19154. The patent further describes that the preferred lipolytic enzymes should have a pE optimum between 6 and 10 and should be active in said range, preferably between 7 and 9. (This putative Pseudomonas stutzeri strain has been reclassified as Pseudomonas aeruginosa).

European patent application (EP-A-0130064) describes an enzymatic detergent additive comprising a lipase isolated from Fusarium oxysporum which has a higher lipolytic cleaning efficiency than conventional lipases. Lipolytic detergent additives are also described in, for example, British Patent Specification No. 1,293,613 and Canadian patent No. 835,343.

The European patent applications EP-A-0205208 and EP-A-0206396 describe the use of Pseudomonoas and Cromobacter lipases in detergents. For a comprehensive review article on lipases as detergent additives, see Andree et al., J. Appl. Biochem, 2 (1980) 218-229.

New compositions comprising transformed microbial cells and methods for their production have been provided and produce lipolytic enzyme suitable for use in detergent compositions. Host microbial cells are transformed using expression cassettes comprising a DNA sequence encoding a lipolytic enzyme that is active at alkaline pH and stable under laundry conditions. Methods for producing lipolytic enzyme include cloning and expression in microbial systems and screening on the basis of DNA homology.

A transformed microbial host cell containing a DNA sequence encoding a lipolytic enzyme acquired from a prokaryotic microorganism is provided.

Of particular interest for the production of lipase, the lipase genes are derived from certain Pseudomonas species.

Brief description of the drawings

Figure 1: Strategy for the lipolytic enzyme gene cloning. For symbols see the text in Figure 2.

Figure 2:; Restriction map of pATl. A number of restriction enzyme recognition sites have been determined in plasmid pAT1.

pUN121 vector DNA

■■■i: Thai IV 17-1 DNA Deposit.

The symbols used are:

- Ori E.coli: E.coli origin of replication

- Ap<r>: pUN121 gene coding for ampicillin resistance; - Tc<r>: the pUN121 gene encoding tetracycline resistance; - cl: the bacteriophage lambda gene coding for the cl repressor.

The position where partially Sau3A-cleaved chromosomal DNA from Pseudomonas stutzeri Thai IV 17-1 (CBS 461.85) was ligated into pUN121 is indicated by BclI/Sau3A.

This position of the gene that codes for lipolytic activity is indicated by a dashed line. Figure 3: Restriction map of pAT3. The symbols are the same as those used in Figure 2. Figure 4: Restriction map of pET1. The symbols are the same as used in Figure 2. Figure 5: Restriction map of pET3. The symbols are the same as those used in Figure 2. Figure 6: Restriction map of pAM1. The symbols are the same as those used in Figure 2. Figure 7: Restriction map of pM6-5. The symbols are the same as those used in Figure 2. Figure 8: Restriction map of pP5-4. The symbols are the same as those used in Figure 2. The position in which partially EcoRI<*>digested chromosomal DNA of Acinetobacter calcoaceticus GR V-39 (CBS 460.85) was ligated to pUN121 is indicated by EcoRI/EcoRI<*.>Figure 9 : A SDS 10-15$ gradient fixed gel (Pharmacia). Figure 9A: After staining with Coomassie Brilliant Blue. Figure 9B: After staining with p-naphthyl acetate/Fast Blue

BB.

Column 1; Lysate from E. coli JM101 hsdS recA strain containing pUN121 heated with SDS for 10 min. at 95°C. Column 2: Lysate from E. coli JM101 hshS recA strain containing pET3 heated with SDS for 10 min. at 95°C. Column 3: Culture supernatant from P. stutzeri Thai IV 17-1 strain heated with SDS for 10 min at 95°C.

Column 4: Same sample as in column 2, but heated with SDS for 5 min. at 95°C.

Column 5: Same sample as in column 3, but heated with SDS for 5 min at 95°C.

Column 6: Same sample as in column 2, but incubated with SDS for 10 min. at room temperature.

Column 7: Same sample as in sample 3, but incubated with SDS for 10 min. at room temperature.

Column 8: Low molecular weight protein markers from Pharmacia.

Figure 10: SDS 13$ polyacrylamide gel after staining with Coomassie Brilliant Blue.

Column 1: Purified M-1 lipase

Column 2: Low molecular weight protein markers (Pharmacia).

Figure 11: Restriction map of pTMPvl8A.

I: pTZ18R vector DNA

■HBh M-l DNA Deposit

The symbols used are:

-Ori E.coli: E. coli origin of replication derived from pBR322 vector. -fl ori: origin of replication from filamentous bacteriophage fl -Pt7: promoter of bacteriophage T7 for preparation of in vitro transcripts.

-Ap<r>: Gene coding for ampicillin resistance

-Lip: Gene coding for M-1 lipase.

Figure 12: shows the nucleotide sequence (ie the first 942 nucleotides) of the M-1 lipase gene and the deduced amino acid sequence of M-1 lipase. The termination codon TGA is indicated by an asterisk. The A box represents the active center of the lipase protein. The arrow indicates the putative signal peptidase cleavage site. The amino-terminal sequence of the mature lipase protein is underlined.

Figure 13: Restriction map of pSW103

In pUC19 vector DNA

PAO 1 DNA insert

The symbols used are:

Ori E. coli: E. coli origin of replication derived from pBR322 vector.

^lac: promoter of the E. coli lac operon

Ap<r>: gene coding for ampicillin resistance.

- Lip: gene coding for PAO 1 lipase

Figure 14: Partial nucleotide sequence of the PAO lipase gene (ie from the internal Sali site) and derived amino acid sequence of PAO lipase. Termination codon TAG is indicated by an asterisk. The A box represents the active center of the lipase protein. Figure 15: Construction of the plasmids pBHAM1 and pBHCM1. The symbols used are:

- Bacillus ori: Bacillus origin of replication.

- Km<r>: pUBHO gene coding for neomycin resistance.

- Cm: the Tn9 transposon coding for chloramphenicol resistance.

- Pflpall<1>Hpall promoter to plasmid pUBllO.

Other symbols are as in figure 11.

Figure 16 shows construction of plasmid pBHAM1N1. The symbols are the same as used in Figure 15. Figure 17 illustrates construction of plasmid pTZNIM1.

Symbols used are:

- lac Za: N-terminal part of the LacZ gene coding for the α-domain of β-galactosidase.

"^lac: Promoter to the E. coli lac operon.

Other symbols are as in figure 11.

Figure 18 shows construction of plasmid pMCTMI. The symbols used are:

-Cm<r>: gene coding for chloramphenicol resistance.

~<p>tac: hybrid trp-lac E. coli promoter.

Other symbols are as in figures 11 and 15.

Figure 19: Immunoblot detection of the M-1 lipase protein produced by transformed E. coli cells.

Columns A-D contain periplasmic fractions of the E. coli cells containing the following constructs:

Column A: pTZ18RN

Column B: pTZNIMl

Column C: pMCTMI

Column D: purified M-l lipase from Pseudomonas pseudoalcaligenes strain M-l. Molecular mass of the marker proteins (Rainbow™ from Amersham) is indicated in kDa on the right.

Figure 20: Construction of plasmid pBHM1N1.

The symbols are the same as used in figure 15.

Figure 21: Autoradiograph of<35>g labeled proteins synthesized in vitro.

Columns A-D: Immunoprecipitation of in vitro translated samples by monoclonal antibodies against M-1 lipase. Columns E-H: In vitro transcription/translation products of the following plasmids:

Columns A and E: pTZ18RN

Columns B and F: pTMPvl8A

Columns C and G: pMCTMI

Columns D and H: pMCTbliMl

Figures 22A and 22B: Detection of lipase coding sequences in bacterial DNA. Five nanogram amounts of plasmid DNA and five microgram amounts of chromosomal DNA were cut with restriction enzyme as indicated, separated on 0.8% agarose gel blotted on nitrocellulose filters and hybridized with nick-translated insert of pET3 and pTMPvl8A, respectively. Figure 22A shows autoradiography after hybridization with pET3 EcoRI insert. Figure 22B shows autoradiography after hybridization with pTMPvl8A XhoI-lcoRV insert.

Column A: HindIII/SStI cleavage of plasmid pTMPvl8A. Column B: EcoRI digestion of plasmid pET3.

BRL DNA gel marker. MW of 0.5, 1.0, 1.6, 2.0, 3.0, 4.0, 5.0, 6.0, 7, 8, 9, 10, 11, 12 kb

Column C: Sali cutting of P. pseudoalcaligenes M-l

(CBS 473.85)

Column D: Sali cutting of P.<p>seudoalcaligenes IN II-5 (CBS 468.85)

Column E: Sali cutting of P. alcaligenes DSM 50342 Column F: Sali cutting of P. aeruginosa PAC IR (CBS 136.89)

Column G: Sali cutting of P. aeruginosa PA02302 (6-1)

Column H: Sali cutting of P. stutzeri Thai IV 17-1

(CBS 461.85)

Column I: Sali cutting of P. stutzeri PG-I-3 (CBS 137.89)

Column J: Sali cutting of P. stutzeri PG-I-4 (CBS 138.89 )

Column K: Sali cutting of P. stutzeri PG-II-11,1

(CBS 139.89)

Column L: Sali cutting of P. stutzeri PG-II-11,2

(CBS 140.89)

Column M: Sali cutting of P. fragi serm. DB1051

(= Ferm BP 1051)

Column N: Sali cutting of P. gladioli (CBS 176.86)

Column 0: Sali cutting of A. caloaceticus Gr-V-39

(CBS 460.85)

Column P: Sali cutting of S. aureus (ATCC 27661).

The present invention relates to a transformed microbial host cell, characterized in that it comprises an expression cassette which in its 5'-3' direction for transcription includes: (1) a transcription regulatory region and a translation initiation region functional in said host cell; (2) a DNA sequence encoding a lipolytic enzyme acquired from a prokaryotic microorganism, said DNA sequence being a 300 bp nucleic acid sequence of the lipase encoding gene on pTMPvl8A (CBS 142.89) or pET3 (CBS 155.87); and (3) translation and transcription termination regions functional in said host cell,

wherein expression of said DNA sequence is regulated by said transcription and translation regions, and optionally comprises a leader sequence linked in the correct reading frame 5' to a gene that codes for said lipolytic enzyme, and that said expression cassette optionally comprises a marker gene.

Techniques used in isolating the lipase gene are known, and include synthesis, isolation from genomic DNA, preparation from cDNA or combinations thereof. The various techniques for manipulating the gene are well known and include restriction, cleavage, resection, ligation, in vitro mutagenesis, primer repair, use of linkers and adapters and the like. See Maniatis et al., "Molecular Cloning", Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, 1982.

The method generally comprises the preparation of a genomic library from an organism that expresses a lipase with the desired characteristics. Examples of such lipases are those obtained from Pseudomonas and Acinetobacter, and especially from strains belonging to the species Pseudomonas al caligenes, Pseudomonas pseudoalcaligenes, Pseudomonas aeruginosa, Pseudomonas stutzeri and Acinetobacter calcoaceticus. A number of these lipases and strains are more fully described in EP-A-0218272, and the description is incorporated herein by reference. The genome of the donor microorganism is isolated and digested with an appropriate restriction enzyme, such as Sau3A. The fragments provided are linked to a vector molecule that has previously been cut with a compatible restriction enzyme. An example of a suitable vector is plasmid pUN121 from E. coli which can be cut with restriction endonuclease Bell. The amino acid sequence can further be used to construct a probe to screen a cDNA or a genomic library prepared from mRNA or DNA from cells of interest as donor cells for a lipase gene.

By using lipase DNA or a fragment thereof as a hybridization probe, structurally related genes found in other microorganisms can be easily cloned. In particular, isolation of genes from organisms that express lipolytic activity using oligonucleotide probes based on the nucleotide sequence of lipase genes that can be obtained from the organisms described in EP-A-0218272. Alternatively, these oligonucleotides can be derived from amino acid sequences of lipases of interest. Such probes can be considerably shorter than the entire sequence, but should be at least 10. Longer oligonucleotides are also useful up to the full length of the gene. Both RNA and DNA probes can be used.

In use, the probes are usually labeled in a detectable manner (eg with 32P, 35S,<3>H, biotin or avidin) and are incubated with single-stranded DNA and RNA from the organism in which a gene is being sought. The hybridization is detected by means of the marker after the single-stranded and double-stranded (hybridized) DNA (DNA/RNA) has been separated (usually using nitrocellulose paper). Hybridization techniques suitable for use with oligonucleotides are well known to those skilled in the art.

Although probes are usually used with a detectable label that enables easy identification, unlabeled oligonucleotides are also useful, both as precursors to labeled probes and for use in methods that provide direct detection of double-stranded DNA (or DNA/RNA). The term "oligonucleotide probe" refers to both labeled and unlabeled forms.

The present invention relates to a method for producing a lipolytic enzyme, characterized in that said method comprises: (1) cultivation in a nutrient medium of a host cell which comprises an expression cassette which includes in the transcription 5'-3' direction a transcription regulatory region and a translation initiation re -gion functional in said host cell; a DNA sequence encoding said lipolytic enzyme acquired from a prokaryotic microorganism; of which said DNA sequence is a 300 bp nucleic acid sequence from the lipase-encoding gene on pTMPvl8A (CBS 142.89) or pET3 (CBS 155.87) and translation and translation termination regions functional in said host cell, wherein expression of said DNA sequence is under regulatory control of said transcription - and translational regions; and (2) isolating said lipolytic enzyme from the nutrient medium.

The invention further relates to a recombinant DNA construct, characterized in that it comprises: (1) in the direction of transcription, a transcription regulatory region which is functional in a microbial host cell; (2) a DNA sequence encoding a lipolytic enzyme obtained from a prokaryotic microorganism, said DNA sequence being a 300 bp micleic acid sequence of the lipase encoding gene on pTMPvl8A (CBS 142.89) or pET3 (CBS 155.87); and (3) a transcription termination regulatory region which is functional in said host cell, and that said recombinant DNA construct optionally comprises at least one of a selection marker gene and one secretory leader sequence.

The lipase enzymes of P. aeruginosa and P. stutzeri were produced and tested for purification properties in the SLM test. It was surprisingly found that all the enzymes exhibiting high levels of homology with the M-1 lipase gene also showed superior stability, efficiency and performance under the conditions simulating a modern washing process. According to Ausubel et al., Current Protocols in Molecular Biology, 1987-1988, the Southern hybridization technique can demonstrate homology at this level. The homologies found were not observed with P. gladioli described in EP-A-0205208 and EP-A-0206390 or Humicola languinosa lipase, described in EP-A 03052216.

Clones containing the inserted DNA fragment can be identified using a direct or positive selection procedure such as that developed for E. coli (Kuhn et al., Gene 44 (1986) 253-263) and for B. subtilis (Gryczan and Dubnau, Gene 20 (1982) 459-469). An example of such a positive selection vector for E. coli. is pUN121 (Nilsson et al., Nucleic Acids Res. 11 (1983) 8019-8030).

In addition, clones expressing the lipolytic enzymes can be identified, for example by using a suitable indicator plate assay, such as agar medium containing tributerin or olive oil in combination with rhodamine B (Kouker and Jaeger, Appl. Encv. Microbiol. 53 (1987) 211) . Furthermore, replicate colonies can be screened using an adapted soft agar technique based on the procedure described for the detection of esterase activity (Hilgerd and Spizizen, J. Bacteriol. 114 (1978) 1184). Alternatively, clones expressing lipolytic enzymes can be identified using genetic complementation in a suitable lipase-negative recipient strain, as described by Wohlfarth and Winkler, J. Gen. Microbiol. 134 (1988) 433-440.

Once a complete gene has been identified, either as cDNA or chromosomal DNA, it can be variously manipulated to provide expression. Microbial hosts can be used, which for example include bacteria, yeast and fungi, such as E. coli. Kluyveromvces. Aspergillus. Bacillus and Pseudomonas species. When the gene is to be expressed in a host that recognizes the wild-type transcriptional and translational regulatory regions of the lipase, the entire gene with its wild-type 5' and 3' regulatory regions can be carried inside an appropriate expression vector. Various expression vectors exist that utilize the replication system from prokaryotic cells (see, for example, Pouwels et al., "Cloning Vector, A Laboratory Manual", Elsevier, 1985. These replication systems have been developed to provide markers that enable selection of transformants and for providing appropriate restriction sites where the genes can be inserted.

When the gene is to be expressed in a host that does not recognize the naturally occurring wild-type transcriptional and translational regulatory regions, further manipulation will be necessary. The various 3' transcriptional regulatory regions are known and can be inserted downstream of the stop codons. The non-coding 5' region upstream from the structural gene can be removed by restriction endonuclease, Bal31 resection or the like. Alternatively, when an appropriate restriction site is present near the 5'-terminal end of the structural gene, the structural gene can be cleaved and an adapter used to link the structural gene to a promoter region where the adapter provides lost nucleotides to the structural gene.

Different strategies can be used for providing an expression cassette, which in the 5'-3' direction of the transcription has a transcription regulatory region and a translation initiation region, which can also include regulatory sequences that allow induction of the regulation gene; an open reading frame encoding a lipolytic enzyme, preferably comprising a secretory leader sequence recognized by the proposed host cell; and translation and transcription termination regions. The expressing cassette may additionally comprise at least one marker gene. The initiation and termination regions are functional in the host cell and can be either homologous (derived from the original host) or heterologous (derived from a foreign source or from the synthetic DNA sequences). The expression cassette can thus be either wholly or partially derived from natural sources, and either wholly or partially derived from sources that are homologous to the host cell or heterologous to the host cell. The various DNA constructions (DNA sequence, vectors, plasmids, expression cassettes) according to the invention are isolated and/or purified, or synthesized and are thus not "naturally occurring".

Selection of appropriate regulatory sequences will take the following factors that affect expression into consideration. With regard to transcription regulation, the amount and stability of messenger RNA are important factors that affect the expression of the gene products. The amount of RNA is determined by the copy number of the particular gene, relative efficiency to its promoter and factors that regulate the promoter, such as enhancers or repressors. The stability of the mRNA is governed by the susceptibility of the mRNA to the ribonuclease enzymes. In general, exonuclease cleavage is inhibited by the presence of structural motifs at the ends of mRNA; palindromic structures, altered nucleotides or specific nucleotide sequences. Endonuclease cleavage is thought to occur at specific recognition sites within mRNA and stable mRNA, or lack such sites. It also suggests that mRNA with high translation levels is also protected from degradation by the presence of ribosomes on the mRNA.

With regard to translational regulation in the presence of mRNA, expression can be regulated by influencing the degree of initiation (ribosome binding to mRNA), the degree of elongation (translocation of the ribosome along the mRNA), degree of post-translational modification and the stability of the gene product. The degree of elongation is probably influenced by which codons are used, and the use of codons for rare tRNAs can reduce the degree of translation. Initiation is believed to occur in the region just upstream of the beginning of the coding sequence. In prokaryotes, this region in most cases contains a consensus nucleotide sequence AGGA, designated the Shine-Dalgarno sequence. Because this sequence characterizes the ribosomal binding site, it is clear that both upstream and downstream sequences can influence successful initiation.

It also suggests that the presence of nucleotide sequences within the coding region can affect ribosome binding, possibly by forming structural parts through which the ribosome recognizes the initiation site. The position of the AGGA sequence with respect to the initiating ATG codon can affect expression. It is thus the interaction of all these factors that determines a specific degree of expression. The expressed genes, on the other hand, have developed a combination of all these factors to provide a particular level of expression. Designing an expression system to provide high levels of the gene product must take into account not only the particular regions that have been determined to affect expression, but also how these regions (and thus their sequences) interact with each other. Illustrative transcriptional regulatory regions or promoters include, for example, sequences derived from genes that are overexpressed in industrial production strains.

The transcriptional regulatory region may additionally comprise regulatory sequences that allow expression of the structural gene to be modulated, for example by the presence or absence of nutrients or expression products in the growth medium, temperature, etc. For example, expression of the structural gene in pro-karyotes may cells are regulated by temperature using a regulatory sequence comprising the bacteriophage lambda P1 promoter together with the bacteriophage lambda O1 operator and a temperature-sensitive repressor. Regulation of the promoter is achieved by interaction between repressor and operator. Of particular interest are expression cassettes capable of expressing lipolytic enzymes using the regulatory sequences of Bacillus amylase and protease genes. The structural gene of interest is linked downstream of the ribosomal binding site so that it is under the regulatory control of the transcriptional regulatory region and the translational initiation region.

In addition, a linked gene can be prepared by providing a 5' sequence to the structural gene encoding a secretory leader and a processing signal. If it is functional in the chosen host cell, the signal sequence of the lipase gene itself can also be used. Illustrative heterologous secretory leaders include the secretory leaders of penicillinase, amylase, protease and yeast alpha factor. By fusion in the correct reading frame with a secretory leader with structural genes of interest, the mature lipolytic enzyme can be washed out into the culture medium.

The expression cassette can be contained within a replication system for episomal maintenance in a suitable host organism or can be provided without a replication system where it can be integrated into the host genome. The method of transformation of the host microorganism with different DNA constructs is not critical in the present invention. DNA can be introduced into the host according to known techniques such as transformation using calcium phosphate-precipitated DNA, conjugation, electroporation, transfection by contacting the cells with a virus, micro-injection of DNA into the cells or the like. The host cells can be whole cells or protoplasts.

As a host organism, any microorganism suitable for the production and extraction of a lipolytic enzyme can be used; preferably the host organism can also excrete the enzyme to be produced, the enzyme can be isolated from the cell-free fermentation liquid. The host organism is also preferably a non-pathogenic organism. Examples of host organisms that meet the above criteria include

E. coli, Pseudomonas putida and Bacillus strains, especially B. subtllis and B. licheniformis, Streptomyces strains and fungi and yeast strains such as Aspergillus and Kluyveromyces, respectively.

The host strains can be laboratory strains or can include industrial microorganism strains. Industrial strains are characterized by their resistance to genetic exchange, such as phage infection or transformation. The strains are stable and may or may not be able to form spores. They are prototropic and modified to provide high yields of endogenous protein products, such as the enzymes alpha-amylase and various proteases. The yield of an endogenous protein product provided in an industrial production process can amount to at least 5 g/l (0.5$ w/v). Industrial strains also secrete DNases, which result in degradation of DNA in the medium and provide protection against genetic exchange.

Once the structural gene has been introduced into the appropriate host, the host cell can be cultured to express the structural gene. The production levels of lipolytic activity may be comparable to, or higher than, the original strains from which the genes are derived. The host cell can be grown to high density in an appropriate medium to form a nutrient-rich broth. When the promoter is inducible, permissive conditions will then be applied, for example temperature change, depletion or excess of a metabolic product or nutrient, etc.

Once secretion is achieved, the expression product can be isolated from the growth medium in conventional manner. Release of the produced lipolytic enzyme can be enhanced by dilute surfactant solutions. When secretion is not provided, the host cells can be harvested and lysed according to conventional conditions. The desired product is then isolated and purified according to known techniques such as chromatography, electrophoresis, solvent extraction, phase separation, etc.

The particular compositions can be used in many different ways. The transformed host cell organisms can be used for increased production of lipase which has the characteristic that makes it useful in detergent compositions. The cloned lipase genes can also be used in screening for lipase genes, including the identification of a lipolytic gene as a lipase instead of an esterase.

They can also be used for enzyme reengineering using known techniques of random or site-directed mutagenesis resulting in lipase with the required altered characteristics.

The lipolytic enzyme compositions can be used in the washing compositions together with a detergent and possibly other ingredients which are usually used in the washing compositions. These ingredients may include at least one of the following; detergents, water softeners such as complex phosphates, alkali metal silicates and bicarbonates; fillers such as alkali metal sulfate; other enzymes such as proteases and amylase; bleaching agents; as well as disparate compounds such as perfumes, optical brighteners, etc.

In the enzymatic cleaning composition, the lipase activity is in the range of from 1 to 20,000 TLU/g of the composition, while the protolytic enzyme activity is in the range of from 50 to 10,000 Delft units/g of cleaning composition. A TLU (true lipase unit) is defined as titratable fatty acids equivalent to the amount of 1 jjmol NaOH/min released from olive oil/gum arabic emulsion at pE 8.0, 25°C. The Delft units are defined in J. Amer. Oil Chem. Soc. 60 (1983) 1672.

The cleaning compositions can be prepared in the usual way, for example by mixing the components together or by preparing an original premix, which is then finished by mixing with other ingredients. According to one possible preparation route, one or more lipase preparations are mixed with one or more of the other compounds to provide a concentrate with a predetermined enzymatic activity and the concentrate can then be mixed with the other desired components.

The lipolytic enzymes are suitable for an enzymatic detergent sub-additive f. This additive may also contain one or more other enzymes, for example a protease and/or an amylase, which can be used in modern detergent compositions and one or more other components which are usually used within the range, for example a non-ionic salt, stabilizing agent and/or bleaching agent. The enzymatic detergent additive may comprise, in addition to lipase, a protease and optionally an alpha-amylase. The protolytic enzymes are compatible with the lipolytic enzymes in this formulation. The enzymatic cleaning additives are usually mixed with one or more cleaning agents and other components known in the art to form laundry compositions. The enzymatic detergent additive is usually used in the range from 10<2> and 10<7>TLU/g of additive, while the possibly present proteolytic activity is in the range of from 5x10<4> to 10<6>Delft units/g.

The enzymatic detergent additives can be in the form of, for example, granules or pellets, produced according to the methods that are usually used in the field. See, for example, British Patent Nos. 1,324,116 and 1,362,365 and US Patent Nos. 3,519,570, 4,106,991 and 4,242,219.

The enzymatic detergent additive may be in liquid form with an enzyme stabilizer, for example propylene glycol. They may also be in the form of organic or inorganic slurries, emulsions or encapsulations, immobilized on a soluble or insoluble carrier or in an aqueous or anhydrous solution in the presence of one or more stabilizers. Such additives are preferably used in liquid detergent compositions.

EXPERIMENTAL

General cloning techniques were used as described by Maniatis et al., "Molecular Cloning: A Laboratory Manual", Cold Spring Harbor Laboratory, 1982, CSH, New York. All DNA-modifying enzymes were obtained from commercial vendors. They were used according to the dealer's instructions. Materials and equipment for DNA purification and separation were used according to the instructions from the retailer.

EXAMPLE 1

Molecular cloning of triacylglycerol acyl hydrolases

A. DNA template and selection vector.

EP-A-0218272 describes several bacterial strains which produce lipases suitable for use in detergents. Out of these, Acinetobacter calcoaceticus Gr V-39 (CBS 460.85), Pseudomonas stutzeri Thai IV 17-1 (CBS 461.85), Pseudomonas pseudoalcaligenes IN II-5 (CBS 468.85) and Pseudomonas pseudoalcaligenes M-l (CBS 473.85) were chosen as a source for lipolytic genes.

Plasmid vector pUN121 (Nilsson et al., Nucleic Acids Research 11 (1983) 8019) carrying an ampicillin resistance gene, a tetracycline resistance gene and the cl repressor gene of bacteriophage lambda was obtained from Dr. M. Uhlen, Royal Institute of Technology, Department of Biochemistry, Teknikringen 10, S-10044 Stockholm, Sweden. Transcription of the tetracycline gene is prevented by the cl repressor. Insertion of foreign DNA into the unique restriction sites (Bell. Smal, HindIII and EcoRI) results in activation of the tetracycline gene. This enables direct (positive) selection of recombinant transformants on Luria medium agar plates containing 8 µg/ml tetracycline and 50 µg/ml ampicillin.

B. Gene library preparation (see Figure 1)

Plasmid and chromosomal DNA are isolated as described by Andreoli, Mol. Gen. The gene. 199 (1985) 372-380. Chromosomal DNA isolated from Thai IV 17-1 and M-1, respectively, was partially Sau3A cut. DNA was then ligated with T4 DNA ligase to Bcll-cut pUN121 DNA as described by Maniatis et al., (above) and transformed into competent cells E. coli strain JM101 hsdS recA (Dagert and Ehrlich, Gene 6 (1979) 23- 28). E. coil JM101 hsdS recA was obtained from the Phabagen collection (Accession number PC2495), Utrecht, The Netherlands. Transformants resistant to tetracycline at 8 µg/ml in Luria medium agar plates were selected for.

C. Screening for transformants

The gene library prepared as described above was replica plated and screened for lipolytic activity using the following two procedures. In the first procedure, the replicate colonies were screened for lipolytic activity on peptone agar media containing tributyrin. Lipolytic activity was detected as a clear zone (ring) around a colony caused by breakdown of the turbid lipid emulsion. In the second method, replicate colonies were screened for esterase activity using a soft agar overlay technique. The method was based on that of Hilgerd and Spizizen, J. Bacteriol. 114 (1987). A mixture consisting of 0.4$ low-melting agarose, 0.5 M potassium phosphate (pH 7.5), 0.5 mg/l 3-naphthyl acetate, dissolved in acetone and 0.5 mg/l Fast Blue BB (see example 2) got all over the transformants. Within a few minutes, colonies with esterase or lipase activity were stained violet.

Of the 1200 tetracycline-resistant transformants provided from the Thai IV 17-l/pUN121 genebank, three produced rings on tributyrin agar plates. Out of 12,000 recombinant transformants examined from the M-1/pUN121 gene bank, only one of the tested clones showed weak lipolytic activity.

Tributyrin-positive clones were grown overnight in 2TY medium (16 g/l Bacto-tryptone, 10 g/l Bacto-yeast extract, 5 g/l NaCl, pH 7.0) medium and analyzed for both plasmid content (see under ID) and the ability to convert various 3-naphthyl substrates, an indication of lipolytic activity (see Example 2S).

D. Plasmid isolation

D.l Containing Thai IV 17- 1 lipase gene

The plasmids isolated from the Thai IV 17-1/pUN121 transformants were designated pAT1 and pAT3. Their structure is indicated in Figures 2 and 3, respectively. A third clone, designated pAT2, contained a plasmid identical to the pAT1 construct. The gene encoding the lipolytic activity was located within a 2.7 kb EcoRI fragment of pAT1 (Figure 2, dashed line) and within a 3.2 kb EcoRI fragment of pAT3 (Figure 3, dotted line). The two EcoRI fragments were subcloned into appropriate vectors, both for DNA sequencing and to provide higher yields of lipolytic activity.

Cloning of the 2.7 kb EcoRI fragment from pAT1 into the pUN121 vector produced the recombinant plasmid pET1 (Fig. 4). A sample of E. coli JM101 hsdS recA containing plasmid pET1 was deposited with CBS on February 5, 1987, under No. CBS 157.87.

Cloning of the 3.2 kb EcoRI fragment from pAT3 in the pUN121 vector produced the recombinant plasmid pET3 (Figure 5). A sample of E. coli JM101 hsdS recA containing plasmid pET3 was deposited with CBS on February 5, 1987, under No. CBS 155.87.

D.2 Containing the M-l lipase gene

The recombinant plasmid isolated from the M-1/pUN121 transformant, designated pAM1, was isolated and characterized (Figure 6). Biochemical characterization of the lipolytic activity of pAM1 shows, on the other hand, that this plasmid did not code for the expected lipase (see also examples 2 and 3). Therefore other strategies had to be developed and these are illustrated below.

The plasmid pAM1 in E. coli JM101 hsdS recA was deposited with CBS February 5, 1987 under No. 154.87.

D.3 Containing the IN II-5 lipase gene

The partial Sau3A cut Pseudomonas pseudoalcaligenes IN II-5 DNA fragments were cloned into E. coli K12 DH1 strain (ATCC 33849) using plasmid vector pUN121. After ligation and transformation into competent DH1 cells prepared as described by Hanahan, J. Mol. Biol. 166 (1983) 557-580, approximately 1500 transformants resistant to 50 µg/ml ampicillin and 8 µg/ml tetracycline were obtained. Transformants with the ability to hydrolyze tributyrin and p-naphthyl acetate were selected for as described under 1C. Starting from a positive clone, plasmid DNA was isolated and characterized by determining several restriction enzyme recognition positions. The physical map of this plasmid, designated pM6-5, is shown in Figure 7.

The activity of E. coli DH1 (pM6-5) to p-naphthyl esters was determined (Table 1). A sample of E. coli DH1 containing plasmid pM6-5 was deposited with CBS on February 5, 1987, under No. 153.87.

D.4 Containing the Gr V-39 lipase gene

Partially EcoRI<*>cut (conditions according to Gardner et al., DNA 1 (/1982) 109-114) Acinetobacter calcoaceticus Gr V-39 DNA was mixed with EcoRI linearized pUN121 DNA. After recycling using T4 polynucleotide ligase, the DNA mixture was introduced into J. coli DH1 (ATCC 33849) using the transformation procedure described earlier in this example. All 1800 tetracycline-resistant transformants obtained were screened for lipolytic activity as described under 1C.

Three clones from this Gr V-39/pUN121 gene library produced a lipolytic enzyme. Plasmid DNA from one of these clones, designated pP5-4, was isolated and characterized by restriction endonuclease. The physical map of this plasmid pP5-4 is shown in Figure 8. The hydrolysis of p<->naphthyl esters with crude enzyme preparations from E_j. coli DH1 (pP5-4) was then determined (Table 1). Plasmid pP5-4 in E. coli DH1 was deposited at CBS February 5, 1987, under No. 151.87.

EXAMPLE 2

Characterization of cloned lipolytic enzyme preparations

A. Determination of lipolytic activity

Tributyrin-positive JL. coli colonies were inoculated into 500 ml 100 ml 2TY medium containing ampicillin and tetracycline in a 500 ml conical flask. E. coli cultures were shaken for 40 h at 30°C in an orbital shaker at 250 rpm. Pseudomonas and Acinetobacter strains were grown at 30°C. After 40 h, the optical density at 575 nm was measured and the force was centrifuged in a Sorvall RC5B centrifuge in a GSA rotor at 6,000 rpm for 10 min. The supernatant was stored at 4°C until enzyme analysis was performed.

The cells were resuspended in 4 ml of lysis buffer (25% sucrose, 50 mM Tris-HCl pH 7.5). Lysozyme was added, and after incubation for 30 min. at 21°C, DNase (20 pg/ml) was added and incubation was continued for 30 min at 37°C. Triton-X100 (0.1% v/v) was added and the cell suspensions were sonicated on ice with a Labsonic 1510 kit (5 pulses for 30 sec at 100 watts at 1 min intervals). The cell debris was then removed by centrifugation for 15 min. at 12,000 rpm in a Hettich Mikro Rapid/K centrifuge. The resulting supernatant was then analyzed for lipolytic activity. The analysis is based on lipolytic enzymes hydrolyzing p<->naphthyl esters. p<->naphthyl which is liberated reacts with diazonium salt Fast Blue BB producing an azo dye absorbing at 540 nm. The method is essentially that of McKellar, J. Dairy Res. 53 (1986) 117-127, and was carried out as follows.

The reaction tube containing in a final volume of 2.0 ml: 1.8 ml of 55 mM TES (N-tris(hydroxymethyl)methyl-2-aminoethanesulfonic acid, Sigma); 0.02 ml of 100 mM p<->naphthyl ester dissolved in dimethyl sulfoxide (DMSO, Merck) or methyl cellusolv<R>acetate (Merck), 0.1 ml of 120 mM NaTC (Na-taurocholate, Sigma) and 0.1 ml of an enzyme preparation.

Controls without enzyme and p<->naphthyl (Sigma) standards without enzyme and substrate were also used.

Corning centrifuge tubes (15 ml) containing the reaction mixture were incubated at 37°C or specified temperature for 30 min. A 0.02 ml 100 mM FB solution (Fast Blue BB salt (Sigma), dissolved in DMSO) was added, and the incubation was continued for 10 min. The reaction was terminated with 0.2 ml of 0.72 N TCA (trichloroacetic acid, Riedel-De Haen) and the colored complex was extracted by vigorous mixing with 2.5 ml of 1-butanol (Merck). The layers were separated by centrifugation at 5,000 rpm for 5 min. in a Heraeus Christ minifuge RF. The absorbance of the top layer was measured at 540 nm using an LKB ultraspec II spectrophotometer. After subtraction of the controls, the readings were converted to TLU (True Lipase Units) using Candida cylindracea lipase (L1754, Sigma) as a standard. A TLU is defined as the titratable fatty acid equivalent of the amount of 1 jjmol NaOH/min (see also EP-A-0218272). The results are shown in Table 1 below.

Comparison of hydrolysis data of the p<->naphthyl esters, varying acyl chain length from C4 to C^g, indicates that: 1°. the clones containing the pAT3 and pET3 plasmids produce true lipases; and 2°. the clones containing the pAT1, pET1, pAM1, pP5-4 and pM6-5 plasmids produce enzymes with substantial esterase activities. Most of the lipolytic enzymes synthesized by E. coli containing recombinant plasmids were found in the cell lysates.

B. Further characterization of cloned lipolytics

preparations

The cloned lipolytic enzyme preparations were characterized using SDS gel electrophoresis on a Phastgel system (Pharmasia) using a 10-15% Phastgel gradient according to the manufacturer's instructions. Cell-free extracts from the E. coli clones pET3 and the PHI strain with the pTJN121 vector were compared with partially purified enzyme from the donor strain Thai IV 17-1 (Stuer et al., J. Bacteriol. 168 (1986) 1070-1079).

Sample<p>repair. Four volumes of an appropriate dilution of the enzyme preparations were mixed with a portion of sample buffer containing 10% SDS, 10% p<->mercaptoethanol in 0.5 M Tris-HCl, pH 6.8. This resolution was divided into three equal parts. A portion was not subjected to further treatment, but was kept at room temperature until the gel electrophoresis had taken place. The second and third portions were heated for 5 and 10 min. at 95°C, respectively, followed by cooling on ice and then kept at room temperature until run on gel electrophoresis.

Electrophoresis. The treated samples, in duplicate, were electrophoresed on the Phastgel system at 65 volts/hour. A gel was stained for protein with Coomassie Brillant Blue according to "Pharmacia Development technique file no. 200". Figure 9A shows the polypeptide patterns after staining with Coomassie Brilliant Blue. The second gel was washed with 50 mM Tris-HCl pH 7.5, 0.1% Triton X-100 to remove SDS and to reactivate the enzyme activity, presence of lipolytic activity in the washed gel was visualized by a soft agar overlay technique based on the p<->naphthyl acetate/Fast Blue BB salt method, described in Example 1C. After incubation for 30 min, at 30°C, violet bands became visible against a clear background. As shown in Figure 9B, the lipase from the E. coliPET3 clone and native P. stutzeri Thai IV 17-1 lipase (MW 40 kDA) had identical mobilities on SDS gel electrophoresis. Both enzymes show a so-called heat modifiability. A similar heat modifiability was described for the outer membrane protein (OmpA gene product) of E. coli K12, Freudl et al., J.Biol. Chem. 261 (1986) 11355-11361.

EXAMPLE 3

Construction of a Pseudomonas pseudoalcaligenes gene library in broad-spectrum host vectors.

As an alternative strategy for cloning the lipase gene from the P. pseudoalcaligenes M-1 strain, a binary broad-spectrum host cloning system was used which enabled the genebank to be screened directly by complementation against different mutant strains of Pseudomonaceae. Two broad-spectrum host vectors pLAFR1 (Friedman et al., Gene 18 (1982) 289-296) and pKT248 (Bagdasarian et al., gene 16 (1981) 237-247) were used. Plasmid pLAFR1 is an RH2-derived cosmid with a wide host range of tetracycline resistance and is mobilizable but not self-transmissible. Plasmid pKT248 is a mobilizable R300B-derived plasmid with a broad host range with streptomycin and chloramphenicol resistance. Transfer (mobilization) of these vectors from E^coli to Pseudomonas was carried out using plasmid pRK2013, which contains RK2 transfer functions (Ditta et al., Proe. Nati. Acad,. Sei. USA 77

(1980) 7347-7351) according to the triparental mating procedure of Friedman et al., (Gene 18 (1982) 289-296). A lipase-negative mutant 6-1 of P. aeruginosa strain PAO 2302 (provided from S. Wohlfarth and U.K. winkler, Ruhr Universitåt, Bochum, FRG) was used as a Pseudomonas recipient (J. Gen. Microbiol. 134 (1988) 433- 440).

The preparation of the in vitro lambda phage packaging extracts and packaging of pLAFRl DNA was performed essentially as described by Ish-Horowicz and Burke (Nucleivc Acids Res. 9 (1981 (2989-2998). Total P. pseudoalcaligenes M-l DNA was partially digested with EcoRI or SalI and was ligated to either EcoRI digested pLAFRl DNA or SalI digested pKT248 DNA. The ratio of insert to vector was 5:1 to reduce the possibility of vector-to-vector ligation. Ligated M-l/pLAFl DNA was packaged in vitro in lambda phage -heads and injected into E_j_ coli DH1 (Maniatis et al., supra 1982).

Approximately 2,500 tetracycline-resistant transductants were provided per jjg P. pseudoalcali<g>enes M-l. If it is assumed that P. pseudoalcaligenes has a genome size of 5,000 kb and that at least 50$ of the tetracycline-resistant transductants contain a 20-kb insert, 2,300 independent clones were necessary to ensure a 99$ probability of finding a particular DNA sequence ( Clark and Carbon, Cell 9 (1976) 91). Since the gene library contained more than 8,000 different recombinant colonies, it was likely that it contained the entire P. pseudoalcaligenes genome.

Ligated M-1/pKT248 DNA was transformed into competent E. coli cells as described in Example 1. The transformants of E. coli JM101 hsdS recA were selected for streptomycin resistance (Sm<R>) and counter-selected for chloramphenicol sensitivity (Cm< s>).5,000Sm<R>Cm<s->clones were provided. The two M-1 gene libraries provided in the E_j_ coli host were replica plated and screened for lipolytic activity as described in Example 1. None of the 13,000 recombinant transformants examined showed lipase activity.

Transfer of the clone from E. coli to P. aeruginosa PAO 2302 (6-1) was therefore carried out as follows. Recombinant plasmids were transferred to Pseudomonas recipients by replica plating donor strains on a blanket of the recipient strain (PA02302/6-1) and the helper strain (E . coli MC1061 or DH1 containing pRK2013 plasmid). After overnight growth of donor, recipient and the helper strain, on a Heart Infusion agar dish, exconjugates were selected for by replica plating on minimal agar medium containing 0.2% citrate, methionine (10 µg/ml) and streptomycin or tetracycline.

Citrate is not metabolized by E^coli. Restoration of the lip phenotype of lipase-defective mutant 6-1 with recombinant plasmids was investigated by replica plating P. aeruginosa exconjugates on nutrient medium agar plates containing trioleoylglycerol and the fluorescent dye rhodamine B as described by Kouker and Jaeger, Appl. Approx. Microbiol. 53

(1987) 211-213.

Four of the 13,000 exconjugates screened showed lipase activity as shown by the development of orange fluorescence rings visible at 360 nm, around the bacterial colonies after 40 h of incubation at 37°C. One of these positive clones, pALM5, was selected for further characterization.

To confirm that the lipase produced by the PAO 2302/6-1 exconjugant containing pALM5 has the desired characteristics conveyed by the lipase from the P. pseudoalcaligenes M-1 strain, enzyme samples were prepared and subjected to both biochemical assays (see Example 2) and SLM -test (as described below in Example 10). The results obtained in these tests indicated that the lipolytic activity of the enzyme produced by the pALM5 clone had similar characteristics to those of the enzyme obtained from the parental M-1 strain.

EXAMPLE 4

Molecular cloning of the Pseudomonas pseudoalcaligenes M-l lipase gene

A. Protein purification and sequence analysis

Fermentation and preparation of a freeze-dried supernatant of the Pseudomonas pseudoalcaligenes M-1 strain is described in EP-A-0218272. The lipolytic enzyme was purified from this supernatant substantially according to Wingerder et al., Appl. Microbiol. Biotechnol. 27 (1987) 139-145. After purification, the protein preparation was more than 80% pure according to SDS-polyacrylamide gel electrophoresis with subsequent Coomassie Brilliant Blue staining (see Figure 10).

N-terminal sequence analysis was performed after SDS gel electrophoresis and electroblotting on Immobilon transfer membrane

(Millipore) according to Matsudaira (J. Biol. Chem. 262 (1987) 10035-10038). This analysis yielded the following sequence (using the conventional single-letter amino acid code):

B. Cloning of the M-1 lipase gene

Two synthetic 32-mer oligonucleotides:

was derived from amino acids 6 to 15 (TGYTKTKYP I) of the N-terminal sequence of the mature lipase protein as described above, after the difference in the codons for Pseudomonaceae and the degeneracy of the genetic code are taken into account. For use as a hybridization probe, the oligonucleotides were end-labeled using T4 polynucleotide kinase.

Chromosomal DNA is isolated from the Pseudomonas pseudoalcaligenes M-1 strain (CBS 473.85) according to Andreoli (Mol. Gen. Genet. 199 (1985) 372-380), cut with several restriction endonucleases and separated on a 0.8% agarose gel. Southern blots of these gels using radiolabeled 32-mer oligonucleotide as a probe showed the following unique hybridizing DNA bands: 1.8 kb Bell. 2.0 kb PvuII and 1.7 SalI. Therefore, the M-1 chromosomal DNA was cut separately with these three restriction endonucleases, fractionated by 0.8$ agarose gel electrophoresis and the hybridizing fractions described above were isolated by electroelution in a bio-trap BT 1000 apparatus from Schlewicher and Schull.

The 1.8 kb Bell-cleaved fraction was ligated into the BamEI-cut dephosphorylated vector pTZ18R/19E (sold from

Pharmacia, Woerden, The Netherlands). The 2.0 kb PvuII cleaved fraction was ligated into the SmaI-cut dephosphorylated vector pTZ18R/19R. The 1.7 kb SalI-digested fraction was ligated into the SalI-cut dephosphorylated vector pZT18R/19R. All three ligations were transformed into competent coli JM101 hsdS recA cells (as described by Andreoli, above) and plated on Luria medium agar plates containing ampicillin, X-gal (5-bromo-4-chloro-3-indolyl-g <->D-galactopyrano side) and IPTG (isopropyl-β-D-thiogalactoside).

Approximately 4x10<3>white colonies were picked on fresh agar plates and screened by colony hybridization to radiolabeled 32-mer probe. Using this method, 12 positive clones were obtained. From each of these positive clones, a plasmid mini-prep was prepared by the alkaline lysis method and cut with the appropriate restriction endonucleases according to the manufacturer's instructions. Six plasmids contained the expected hybridizing inserts. One of the plasmids contained only the 2.0 kb PvuII fragment. designated pTMPvl8A, was chosen for detailed analysis. A sample of E_;_ coli JM101 hsdS recA containing plasmid pTMPvl8A was deposited with CBS March 8, 1989, under No. CBS 142.89.

E. coli pTZ18R/19R recombinant transformants obtained as described above were replica-plated and screened for lipolytic activity as described in Example 2. None of the 5x10<4>coli transformants examined showed lipolytic activity. Several reasons for this can be mentioned: a) the gene expression initiation signals of the M-1 lipase genes were either not recognized by E^. coli transcription/translation system (see, e.g., Jeenes et al., Mol. Gen. Genet. 203 (1986) 421-429), b) a regulatory sequence or protein is required to activate this M-1 lipase gene, c) proper folding or secretion of M-l lipase in E. coli may be insufficient. C. Characterization and sequencing of the M-1 lipase gene Plasmid pTMPvl8A was cut with different restriction enzymes that have a 6 bp recognition sequence. Analyzes of the fragment sizes resulting from these experiments allowed the creation of a preliminary restriction endonuclease cutting map of the 2.0 kb PvuII insert of pTMPvl8A. This map is shown in figure 11.

The pTZ18R/19R vectors used provide a versatile "all-in-one system" enabling DNA cloning, dideoxy DNA sequencing, in vitro mutagenesis and in vitro transcription (Mead et al., Protein Engineering 1 (1986) 67-74 ). The double-stranded plasmids were converted to single-stranded DNA by superinfection with a helper phage M13K07, obtained from Pharmacia.

A DNA sequence analysis of a 0.94 kb DNA fragment between Xhol and EcoRV of pTMPvl8A is shown in Figure 12. This DNA sequence, after anyone sequenced in all possible reading frames, showed a large open reading frame encompassing the NH2-terminal amino acid residues to the lipase protein as determined by direct amino acid sequencing (residues 1 - 24), see this example under A. The methionine at position -24 is the initiation codon of a preprotein due to the methionine preceding a number of amino acids common to bacterial signal peptides (see Von Heijne , J. Mol. Biol. 192 (1986) 287-290). This 24 amino acid signal peptide is cleaved off during the secretion process after the A E A (-3 to -1) signal peptidase recognition site. It is noteworthy that the region around the cysteine residue of this signal sequence is very similar to a lipoprotein consensus signal peptide (see Von Eeyne, ibid. ).

The amino acid sequence determined from the DNA sequence indicates that mature M-1 lipase consists of a 289 amino acid protein that terminates with the TGA stop codon (see Figure 12). The predicted molecular weight of this mature protein is 30,323 which is consistent with the molecular weight determined for M-1 lipase (MW 31,500) by SDS-polyacrylamide gel electrophoresis (see Figure 10).

EXAMPLE 5

Molecular cloning of pseudomonas

aeruginosa PAO lipase gene

The Pseudomonas aeruginose lipase is a lipid hydrolyzing enzyme (EC 3.1.1.3), with a MW of approximately 29,000, which is washed into the medium during the late exponential growth phase (see Stuer et al., J. Bacteriol. 168 (1986) 1070-1074 ).

A. Cloning and characterization

To clone the lipase gene from the Pseudomonas aeruginosa PAO 1 strain (ATCC 15692), a broad-spectrum host cloning system was used which enabled the gene bank to be screened directly by complementation against different mutant strains of Pseudomonaceae. The broad-spectrum host vector pKT248 (Bagdasarian et al., Gene 16 (1981( 237-247)) was used, and is a transferable R300B-derived plasmid with a broad host range with resistance to streptomycin and chloramphenicol.

Pseudomonas aeroginosa PAO 1 DNA was partially cut with restriction endonuclease Sal I and ligated into the single SalI site of the pKT248 vector. The insert:vector ratio was 5:1 to reduce the possibility of vector-to-vector ligation. Ligated PA0/pKT248 DNA was transformed into competent E_j_ coli SK1108 (Donovan and Kushner, Gene 25

(1983) 39-48) as described in Example IB. The transformants of E_j_ coli SK1108 were selected for streptomycin resistance (Sm<R>) and counterselected for chloramphenicol sensitivity (Cm^).

Six thousand (6,000) Sm<R>Cm<s> clones were provided and screened for lipolytic activity as described in Example 2. None of the clones showed such activity, and this was probably caused by poor recognition of the Pseudomonas promoters in E^.coli (Jeenes et al., supra).

Transfer of the clones from E.coli to Pseudomonas aeruginosa PAO 2302 lipase-negative mutant 6-1 (Wohlfarth and Vinkler, J. Gen. Microbiol. 134 (1988) 433-440) was therefore carried out as follows; 6,000 clones were divided into 120 portions, each consisting of 50 clones. The plasmid preparation was provided from each portion to transform competent cells of the PAO 2302 (6-1) lip" mutant. Competent Pseudomonas cells were prepared according to Olsen et al., J. Bacteriol. 150 (1982) 60-69. Selection was performed on calcium triolein (CT) agar plates (Wohlfarth and Vinkler, supra), supplemented with streptomycin (50 µg/ml). Ten of the 5,000 PAO transformants screened showed lipase activity and were visible as white crystals on top of the colonies One of these positive PAO transformants, pSW1, was selected for further characterization.

The plasmid contained therein was characterized by restriction analysis followed by electrophoresis on agarose gels and was found to contain a 3.1 kb SalI insert consisting of a 1.3 kb, a 0.97 kb and a 0.76 kb SalI subfragment. These inserts were subcloned into appropriate vectors for DNA sequencing and to provide expression of the lipase gene.

Plasmid DNA from a pUC19-derived subclone, designated pSV103 (containing the 1.3 and 0.97 kb inserts), was isolated and characterized with restriction endonucleases. The physical map of this plasmid, pSW103, is shown in Figure 13. A sample of E.coli JM101 hsdS recA containing plasmid pSW103 was deposited with CBS on March 8, 1989, under No. 141.89.

B. Sequencing and characterization of Pseudomonas

aeruginosa PAO 1 llpase gene

The 2.3 kb Sali insert of pSW103 containing the PAO 1 lipase gene was sequenced using two methods: both a "forced" cloning sequencing and a "shot" cloning sequencing method of Deininger (Anal. Biochem. 129

(1983) 216-223) on the M13 bacteriophage derivatives mpl8 and mpl9 (Yanisch-Perron et al., Gene 33 (1985) 103-119) the dideoxy chain termination technique of Mizusawa et al., Nucleic Acids Res. 14 (1986) 1319-1324.

In the "forced" cloning sequencing procedure, the SalI/Pstl and SalI/EcoRI fragments of the 2.3 kb inserts of pSW103 were purified and ligated into an appropriate M13mpl8 vector. The entire sequence was obtained by joining together the collection of the provided pieces of the DNA sequence. Most of the DNA sequence has been determined for both strands. This DNA sequence, when translated in all possible reading frames, shows that only the 0.97 kb Sal I fragment of pSW103 (see Figure 14) encodes the mature PAO 1 lipase.

The 1.3 kb SalI fragment of the pSW103 subclone did not encode, when translated in all possible reading frames, the 24 N-terminal amino acid residues of the PAO 1 lipase protein as determined by direct amino acid sequencing.

The amino acid sequence provided from the DNA sequence of the 0.97 kb SalI fragment shows an interesting domain (designated A in Figure 14).

Domain A encodes an amino acid sequence G-H-S-H-G already defined as the active center of both eukaryotic lipases (Wion et al., Science 235 (1987) 1638-1641; Bodmer et al., Biochem. Biophys. Acta 909 (1987) 237- 244) and the prokaryotic lipases (Kugimiya et al., Biochem. Biophys. Res. Commun. 141 (1986) 185-190).

EXAMPLE 6

A. Expression of cloned M-1 lyase in bacteria

To improve the expression level of M-1 lipase in heterologous hosts, the 2.0 kb Knp_I -Hindi 11 fragment of pTMPvl8A containing the M-1 lipase gene was ligated into Kpn1 and HindIII cleaved pBHA/C1 vectors. The nucleotide sequence of the pBHAl vector is described in EP-A-0275598. The pBHCl vector is similar to the pBHCl vector except for the difference in the expression of the following antibiotic resistance genes: the chloramphenicol acetyltransferase gene (Cm) is expressed in E>_ coli strain WK6 containing the pBHCl vector and the beta-lactamase gene (Ap) is expressed in E^. coli strain WK6 containing the pBHAl vector. The E. coli WK6 strain is described by R. Zell and H.J. Fritz, EMBO J. 6 (1987) 1809.

After transformation of WK6 and analysis of either the provided ampicillin-resistant colonies (in the case of pBHAl) or the provided chloramphenicol-resistant colonies (in the case of pBHCl), the respective plasmids pBHAM-1 and pBHCM-1 were found, see Figure 15 .

The next step was the introduction of an NdeI restriction endonuclease site at the ATG initiation codon of the M-1 preprotein. Site-directed mutagenesis on the pBHA/CM-1 plasmids was performed as described by Stanssens et al., in "Protein Engineering and Site-Directed Mutagenesis", 24th Harder Conference (1985), Ed. YEAR. Fersht and G. Winter.

After mutagenesis, potential mutants were screened for the relevant mutation by both restriction enzyme analysis and sequence analysis using the dioxy method of Mizusawa et al., see Example 5. After these analyses, the correct plasmid, designated pBHAMlNl, was found (see Figure 16). To achieve regulated expression of M-1 lipase in E^. coli. the 1.7 kb NdeI-HindiII fragment of pBHAM1N1 containing the M-1 lipase structural gene was isolated and ligated into two expression vectors; - pTZ18RN, a pTZ18R derivative containing a unique Ndel site on the ATG initiation codon of beta-galactosidase (see Mead et al., Protein Engineering 1 (1986) 67-74);

pMCTN, a pMC derivative (see Stanssens et al., supra) containing a unique Ndel site after a tac promoter and ribosomal binding site.

After transformation of the two ligations into competent E. coli JM-101 hsdS recA cells. the provided transformants were screened for lipolytic activity on tributyrin agar plates containing 0.5 mM IPTG (isopropyl-β<->D-thiogalactoside). After incubation of these tributyrin dishes at 30°C for 48 h followed by storage of the plates in a refrigerator for several days, some colonies formed a faint ring showing lipolytic activity. The plasmids of these tributyrin positive clones were characterized by restriction enzyme analysis. The two plasmids sought were found: pTZNIM1 containing the M-1 lipase gene after the lac control sequences (see Figure 17); and - pMCTMI containing the M-1 lipase gene after the tac control sequences (see Figure 18).

To identify the lipolytic activity produced by the E. coli transformants containing plasmids pTZNlM-1 and pMCTM-1, respectively, these clones were grown overnight in 100 ml 2 TY medium (16 g/l Bacto tryptone, 10 g/l Bacto- yeast extract, 5 g/l NaCl, pH 7.0) at 30°C, followed by incubation for 3 h with 0.5 mM IPTG at 30°C. ]L_ coli strain JM-101 hsdS recA containing pTZ18RN was used as a negative control. The cells were separated from the culture supernatant by centrifugation and then distributed into periplasm and membrane/cytoplasmic components according to Tetsuaki et al., Appl. Environmental Microbiol. 50 (1985) 298-303. Each fraction was analyzed on SDS-polyacrylamide gels According to Laemmli, Nature 227 (1970) 680-685, consisting of a 13% separating gel and a 5% collecting gel. The gels were run at 60 mA until the Bromophenol Blue (BPB) marker reached the bottom of the gel. Sample preparation and protein blotting procedure were performed as described in EP-A-0253455. The Western blot was analyzed using polyvalent rabbit antisera against purified lipase from P. pseudoalcaligenes strain M-1. From the results shown in Figure 19, it can be concluded that the E_j_ col_i strains containing pTZNIMl and pMCTMI, respectively, can synthesize and flush out in the periplasmic area an M-l lipase-specific 31.5 kDa polypeptide (see Figure 19, columns B and C) . The lipolytic activity of this 31.5 kDa polypeptide was confirmed by the soft agar plating technique based on the p-naphthyl acetate/Fast Blue BB salt method, described in Example 1C.

EP-A-0275598 and EP-A-0253455 describe a method for the efficient transfer into Bacillus of a vessel containing a gene of interest and resulting in Bacillus strains which efficiently secrete the desired polypeptide products. This procedure was used for both the Thai IV 17-1 and M-1 lipase genes.

In the case of the M-1 lipase gene, the pBHAM1N1 plasmid (see above) was digested with NdeI and religated. The ligation mixture was transformed into Bacillus licheniformis T9 protoplasts. A number of neomycin-resistant tributyrin-positive clones were analyzed and the correct plasmid was provided. The plasmid was designated pBHM1N1 (see Figure 20) with the M-1 protein after the H<p>alI<p>romoter of pUB110 (see Zyprian and Matsubara, DNA 5 (1986) 219-225). The T9 transformants containing the pBHMlNl plasmid were tested for their ability to hydrolyze β-naphthyl esters after fermentation at industrial power according to the procedure described in Example 2. The results indicate that the lipolytic activity of the enzyme produced by the T9 clones has similar characteristics to that of the lipase provided from the parent Pseudomonas<p>seudoal-caligenes M-1 strain.

For expression in Pseudomonas hosts, the constitutive promoter of the p78 gene from the Pseudomonas specific bacteriophage Pf3 was used (see R.G.M. Luiten, "The Filamentous Bacteriophage Pf 3: A Molecular Genetlc Study", PhD Thesis 1987, Catholic University of Nijmegen, The Netherlands) . A synthetic DNA fragment was prepared encoding this Pf3 promoter:<*35>*10

This fragment was ligated into vector pTZ18R cleaved with EcoRI and KpnI yielding plasmid pTZPf31A.

Plasmids pTZPf31A and pMCTMI were cut with both BamHI and HindIII, ligated and transformed into competent E. coli JM-101 hsdS recA cells. The correct plasmid with the M-1 lipase gene placed under the control of the Pf3 promoter was provided and designated pTZPf3Ml.

To achieve expression of the cloned M-l lipase gene in pseudomonads, the M-l cassettes present in plasmids pTMPvl8A, pMCTMI and pTZPf3M1 were inserted into broad-spectrum host plasmids pKT231 (Bagdasarion et al., Gene 16

(1981) 237-247, pLAFR3 (Staskawisz et al., J. Bacteriol. 169 (1987 (5789-5794) and pJRD215 (Davison et al., Gene 51 (1987) 275-280). The provided plasmids with broad-spectrum hosts containing the M-1 expression cassettes, were transferred according to the triparental mating procedure of Friedman et al., Gene 18 (10' 982) 289-296) into the following Pseudomonas strains: the lipase negative mutant 6-1 into the P. aeruginose strain PAO 2302 ( Wohlfarth and Vinkler, supra.) the lipase negative P. putlda strain KT 2442 (Zeyer et al., Applied Environmental Microbiol. 50 (1985) 1409-1413) the P. pseudoalcaligenes strain M-1 (CBS 473.85)

P. pseudoalcaligenes strain IN II-5 (CBS 468.85).

As an alternative method for transferring the broad-spectrum host plasmids from Ej_coil to Pseudomonas, the electric field-mediated transformation ("electroporation") procedure was used according to the manual of the Gene Pulser (Bio-rod Laboratories).

The provided Psedomonas transformants were tested for their lipase production after fermentation in olive oil-based media as described by Odera et al., J. Ferment. Technol. 64

(1986) 363-371.

In the following table 2, the lipolytic productivity of the different strains is shown.

The lipases produced by these Pseudomonas transformants were analyzed by SDS gel electrophoresis and migrated like the lipase produced from the original P. pseudoalcaligenes strain M-1.

The improvement achieved by introducing multiple copies of the M-lek expression cassettes into the P. pseudoalcaligenes strains is 2-4 fold compared to the level of lipase produced by the donor strain. It can further be concluded that the original gene expression initiation signal or promoter of the M-1 lipase gene is active in the Psedomonas pseudoalcaligenes strains relative to the PAO 1 and KT2442 strains.

B. In vitro expression of the cloned M-1 lipase gene

In vitro expression of M-1 llpase-containing clones was performed using a prokaryotic DNA-directed translation kit (Amersham International). This system enables in vitro expression of genes contained in a bacterial plasmid if the relevant control signals are present. The following four bacterial plasmids were analyzed: B.l. Plasmid pTMPvl8A (see Figure 11) coding for both the M-1 lipase and the β-lactamase gene products and confers ampicillin (Ap) resistance. pTMPvl8A contains its own regulatory signals, promoter, Shine-Dalgarno and leader sequence.

B.2. Plasmid pMCTMI (see Figure 18) contains the M-1 lipase gene and the chloroamphenicol resistance gene (Cm). In this construct, the lipase promoter is exchanged with a strong tac promoter.

B.3. Plasmid pMCTbliM1 also contains the M-1 lipase gene and the chloroamphenicol resistance gene. In this construct, the lipase signal sequence was exchanged with the α-amylase signal sequence (see EP-A-0224294). The promoter was the same as in construct pMCTMI.

B.4. Plasmid pTZ18RN (see Figure 17) was used as a negative control.

DNA (0.5 pg) of the mentioned plasmids was transcribed in vitro. This reaction was performed by adding 0.5 µl of 10xTB/10xNTP mix (a mixture of equal volumes of 20xTB and 20xNTP mix; 20xTB contains 800 mM Tris HC1 pH 7.5, 120 mM MgCl2 and 40 mM spermidine; 20xNTP mix containing 10 mM ATP, 10 mM CTP, 10 mM GTP and 10 mM UTP), 0.5 µl of 0.IM DTT, 0.5 µl of RNasin (40 u/µl, Promega) and 0.5 µl of T7 RNA polymerase (15 µl, Promega) or 1 µl of E. coli RNA polymerase (1 µl, Boehringer). The reaction mixture was incubated in oxygen at 39.5°C.

In vitro translation of the RNA transcripts was performed according to the manufacturer's instructions. M-1 lipase was immunoprecipitated as described by Van Mourik (J. Biol. Chem. 260 (1985) 11300-11306) using monoclonal antibodies against M-1 lipase.

As a negative control, pTZ18RN was used (Figure 21, columns A and E). Immunoprecipitation of pTMPvl8A (column B) shows an unprocessed M-1 lipase of 34 kDa. Immunoprecipitation of pMCTM-1 (column C) shows an Unprocessed M-1 llpase of 34 kDa and the mature M-1 lipase of 31.5 kDa, while Immunoprecipitation of pMCTbliMl (column D) shows the mature M-1 lipase of 31, 5 kDa. From the in vitro translation experiments, it can be concluded that the M-1 lipase gene can be expressed in S-30 extracts of the E. coli cells. Data provided by in vitro transcription/translation of pTMPvl8A suggest an absence of lipolytic activity on the tributyrin dishes (see Example 4B) because there is no processing by M-1 lipase.

EXAMPLE 7

Molecular enzyme screening using characterized lipase genes as probes.

To further demonstrate the general applicability, we used the probes described in this application to search for lipase genes with comparable characteristics, and originating from other microorganisms. We demonstrated that we can select the lipase genes that code for lipases that have equal or comparable detergent applicability.

As an example, we describe the use of the DNA inserts of the plasmids pTMPvl8A and pET3, respectively, as hybridization probes to show that homologous genes from both are present in most of the analyzed microorganisms. There is no cross-hybridization between the two lipase genes (column A and column B) and this suggests that each of these lipase coding sequences comes from different ancestral genes.

In Example 8, we also demonstrate that this hybridization technique enables us to clone homologous lipase genes from other microorganisms.

To achieve this, chromosomal DNA from the following strains, Pseudomonas pseudoalcaligenes M-l (CBS 473.85), P. pseudoalcaligenes IN II-5 (CBS 468.85), P. alcallgenes (DSM 50342) P. aeruglnose (PAC IR (CBS 136.85), P. aeruginosa PAO (ATCC 15692) (Wohlfarth and Vinkler, 1988, J. Gen. Microbiol. 134, 440). P. stutzeri Thai IV 17-1 (CBS 461.85); P. stutzeri PG-1-3 (CBS 137.89 ), P. stutzeri PG-I-4 (CBS 138.89), P. stutzeri PG-II-11,1 (CBS 139.89), P. stutzeri PG-II-11,2 (CBS 140.89), P. fragi DB1051, P. gladioli (CBS 176.86), Acinetobacter calcoaceticus Gr-V-39 (CBS 460.85) and Staphylococcus aureus (ATCC 27661) were isolated, and 5 pg were digested and analyzed by Southern blot technique (Maniatis et al., supra). was transferred to a nitrocellulose filter (Schléicher & Schuell, BA85; 0.45 mM).The filters were prehybridized in 6xSSC, 5xDenhardt, 0.5% SDS and 100 pg/ml denatured calf thymus DNA.

After two hours of preincubation, <32>P labeled inserts of pTMPvl8A or pET3 respectively were added, and hybridization was performed at 55°C for 16 hours.

DNA inserts were isolated by digestion of pTMPvl8A with Xhol and EcoRV and digestion of pET3 with EcoRI, followed by separation of the fragments by 0.8% agarose gel electrophoresis and isolation of the fragments by the glass milk procedure (Gene Clean™). The insert of pTMPvl8A, a 1 kb XhoI/EcoRV fragment and the insert of pET3, a 3.2 kb EcoRI fragment were labeled in vitro to high specific activity (2-5 10<8>cpm/jjg) with oc32P ATP at nick translation (Feinberg and Vogelsteln, Anal. Biochem. 132, 1983, 6-13). It was shown that the probe fragments have an average length varying from 300-800 bases.

The filters were then washed twice for 30' at 55°C in 6xSSC, 0.1% SDS. The filters were dried at room temperature and autoradiographed for 1-3 days by exposure to KODAK X-omat AR films or Cronex 4 NIF 100 X-ray film (Dupont) with intensifying screens at -70°C (Cronex light plus GH 220020).

The following equation has been obtained from analyzing the effect of various factors on hybrid stability: Tm = 81 + 16.6 (loglO Ci) + 0.4 ($ G + C) - 600/n -1.5$ mispairing ( Current protocols in molecular biology 1987-1988), by Ausubel et al.)

n = length of the shortest chain of the probe Ci = ionic strength (M)

G+C = base composition

was used to determine homolog! which could be demonstrated in our experiments. If a probe length was 300 bases, it was possible to detect a homologous gene that shows at least 67% homology with a fragment of 300 bases or more. In determining percent homology, the GC content of Pseudomonas was assumed to be 65% (Normore, 1973, in Laskin and Lechevalier (ed), Handbook of microbiology vol. II, CRC press, Inc. Boca Raton. Fla).

Figure 22A shows the hybridization pattern of the SalI cut

chromosomal DNA when hybridized with EcoRI insert to pET3. Most Pseudomonas strains contain homologous genes to our pET3 clone. In P. f ragi DB 1051 (column M), A. calcoacetlcus Gr-V-39 (column 0) and S. aureus (column P) no homologous genes were detected. Moderate hybridization signals were observed in P. alcali<g>enes (column E), P. pseudoalcaligenes (column C and column D), P. aeruginosa (column F and column G) while weak hybridization was observed in P.<g> ladioli (column N). Very strong hybridization was seen in the P. stutzeri strains (columns H to L), which is not surprising because pET3 was originally derived from P. stutzeri Thai IV 17-1.

Figure 22B shows the hybridization pattern with EcoRV/XhoI insert to pTMPvl8A.

Strong hybridization signals were obtained from P.<p>seudoalcaligenes chromosomal DNA. (columns C and D), P. alcaligenes (column E) and P. stutzeri strains (column H to column L). Weaker hybridization was detected in chromosomal DNA from P. aeru<g>inosa (columns F and G) and no hybridization was found with chromosomal DNA from P. f ragi DB1051 (column M), P. gladioli (column N), A . calcoaceticus Gr-V-39 (column 0) and S. aureus (column P).

EXAMPLE 8

Cloning of Homologous Lipase Genes from Other Microorganisms To demonstrate its applicability, the use of plasmid PTMPvl8A to clone a homologous gene originating from P. alcaligenes (DSM 50342) (see Figure 22B, column E) is described.

In Figure 22B, column E, it can be seen that P. alcaligenes clearly displayed the 5.0 kb Sal I fragment and a weak 0.5 kb Sal I fragment, hybridizing with the probe. This shows that P. alcaligenes contains a gene that has at least 67% homology in a fragment of 300 bp or more in relation to the XhoI/EcoRV inserted into the pTMPvl8A plasmid.

To determine whether the 5.0 kb Sal I fragment of P. alcali<g>enes represents a gene encoding a lipase, the Sal I fragments were cloned. A Sali gene library was constructed using vector pUC19 and transformed into E. coil JM109 (Yanish-Perron et al., Gene 33 (1985) 103-119). Replicate filters of the gene library were hybridized with α<32>P-labeled insert of pTMPvl8A using the conditions described in Example 7. Three positive clones all containing a 5.0 kb SalI fragment were isolated from the library: pCH1, pCH2 and pCH3 . The 5.0 kb Sal I fragment of pCH1 was recloned into vector pKT248 (Bagdasarian et al., Gene 16 (1981) 237-247) using the SalI site. which is located in the chloramphenicol resistance gene. Streptomycin-resistant chloramphenicol-sensitive transformants were selected in E. coli JM109. One of these transformants, pCH101, containing the putative 5.0 kb SalI insert was transformed into Pseudomonas aeru<g>inosa 2302 (6-1) lip- (see Example 5).

Colonies were grown on NB calcium triolein agar. After growth, the plates were stored at 10°C. After several days, a calcium precipitate was visible around the transformed colonies and not around the non-transformed P. aeruginosa 2302 (6-1) lip-, which were grown on similar agar plates without antibiotics. We conclude that the cloned 5.0 kg SalI fragment complements the lip phenotype of P. aeruginosa 2302 (6-1) lip and therefore codes for an extracellular lipase, which can be produced in a suitable host.

On the basis of DNA hybridization experiments, it can be concluded that this lipase shows homology with M-1 lipase and thus with other lipases, which hybridize with the pTMPvl8A insert used as a probe. The description of cloning of P. alcali<g>enes lipase represents a generally applicable method for cloning lipase genes showing homology to M-1 lipase.

EXAMPLE 9

Comparison of the nucleotide sequence of lipase obtained from Pseudomonas pseudoalcaligenes M-l with other lipases The nucleotide sequence of Psedomonas pseudoalcaligenes M-l lipase (Figure 12) was compared with that of Pseudomonas fragi (IFO-3458) lipase (Kugimiya et al., Biochem. Boiophys. Res. Commun. 141 (1986) 185-190), Staphvlococcus hvicus lipase (Gotz et al., Nucleic Acids Res. 13, (1985) 1891-1903) and Psedomonas aeru<g>inosa PA01 lipase (Figure 14). A homology of 81$ was found between M-1 lipase and P. aeru<g>inosa PA01 lipase and a homology of 62$ was found between M-1 lipase and P. fragi (IFO-3458). the sequence of P. fra<g>i lipase is, on the other hand, considerably shorter than the sequence of the other two Pseudomonoas lipases. No homology was found between M-l lipase and Staphylococcus hyicus lipase.

These results confirm the data provided in Example 7, where no hybridization was detected with (chromosomal DNA) derived from Pseudomonas fragi and Staphylococcus aureus. when pTMPvl8A was used as a probe.

The amino acid sequence provided from the nucleotide sequence to

P. pseudoalcaligenes M-1 lipase was also compared with other lipases. The homology between M-l lipase and P. aeruginosa PA01 lipase was found to be (78$) and between M-l lipase P. fragi lipase 48$. Again, no homology was found between the amino acid sequence of M-l lipase and Sta<p>h<y>lococcus hyicus lipase. However, close homology exists between the four lipases in the region of essential serine 87 in the mature M-1 lipase. Kugimiya et al., (supra), described that the sequence of this region G-H-S-H-G is the active center of the lipase enzymes.

EXAMPLE 10

The compatibility of detergent to homologous lipases

in SLM test

This example illustrates the performance of the lipase enzymes produced by P. aeru<g>iuosa. The P. stutzeri and P. alcaligenes strains showing a high degree of DNA sequence homology with the M-1 gene, in a washing process according to the modified SLM test where the compatibility of these enzymes in modern laundry composition was tested.

The detergent compositions used were a powder detergent composition (ALL<R->base, described in EP-A-0218272) and a liquid detergent formulation (Liquid TIDER, also described in EP-A-0218272, but without protease inactivation). The lipase enzymes were prepared by culturing the bacteria according to the following procedure: an inoculum culture was prepared by culturing the bacteria in 100 ml brain-heart infusion (BHI) medium in a rotary shaker at 30"C for 24 hours. A lab fermenter (2 liter) containing 1.0 liter of a medium of the composition mentioned below was then inoculated.

Media composition

The fermentation was carried out at 30°C. Sixteen hours after inoculation, feed of soybean oil was started at a rate of 1 g/h and continued for the remainder of the fermentation. The air speed was 60 l/h and the agitation speed was 700 rpm. The fermentation was run for a total of 64 hours.

After fermentation, the bacteria were removed by centrifugation. The supernatant was rapidly mixed with stirring with 2.5 volumes of acetone at room temperature. The mixture was then stirred for 10 minutes, allowed to settle and filtered through a glass filter under suction. The filter cake was then washed with 70$ acetone followed by 100$ acetone and then dried under vacuum.

The lipase enzyme preparations thus obtained were then examined in the SLM procedure. The procedure for the SLM test is described in EP-A-0218272. This procedure was modified as follows: A volume of 20 µl containing 10 mg of olive oil dissolved in acetone (25$) was applied to a piece of polyester (3x3 cm) and air-dried at room temperature. A washing solution consisting of 10 mL of SHW (standard hardness water: 0.75 mM CaCl2, 0.25 mM MgCl2) or detergent dissolved in SHW was placed in an Erlenmeyer flask (25 mL) with a stopper, and kept in a shaking water bath at 40°C. The detergents tested were a powder detergent composition (ALL<R->base) and a liquid formulation (TIDER<->liquid). The concentration of the ALL base wash solution was 4 g/l (pH 9.5). The concentration of liquid TIDE was 1.5 g/l (pH 7.5). The solutions were buffered with 0.1 M Tris/HCl. The washing process was started by adding the enzyme preparation and then transferring the dirty piece to the ERlenmeyer flask and shaking for 40 min. or longer at 40°C. The final lipase concentration was 2 TLU/ml.

After washing, the piece was cleaned with SHW and then dried at 55°C for one hour. The dried pieces were washed again in the second wash cycle using a fresh detergent and enzyme solution for 40 minutes. After the second wash cycle, the piece was treated with 0.01N HC1 for 5 min., rinsed and dried at room temperature overnight. The dried pieces were extracted by rotation in a glass tube containing 5 ml of a solvent (n-hexane/isopropyl alcohol/formic acid: 975:25:2.5 (v/v), 1 ml/min). Residual triglyceride, diglyceride and free fatty acid formed were determined by HPLC.

HPLC equipment and conditions:

Column: CP Microspher-Si (Chrompack), 100x4.6 mm Injection system: V/isp (Millipore) 10 pl

Pump: Model 2150 (LKB)

Detection: Refractive index monitor (Jobin Jvon) Integrator: SP 4270 (Spectra Physis)

Eluent: n-hexane/isopropyl alcohol/formic acid: 975:25:2.5 (v/v), 1 ml/min.

Temperature: Room

The retention time to triolein was 1.2 min., to 1.3 diglycerides was 2.5 min., to 1.2 diglycerides was 3.6 min. and to oleic acid was 1.6 min. The peak area, or peak height, was determined as an indication of the yield of the triglyceride and the fatty acid. The yield of the triglyceride after extraction from the unwashed piece was determined to be $100. The ratio of the refractive index responses between triolein and oleic acid was found to be 1.0 on the basis of the peak height.

The results of these SLM tests are shown in the following tables. In these tables, the yields of triglycerides and total lipids are shown. Total lipid yield is the triglyceride plus 1,2- and 1,3-diglycerides plus free fatty acids. The difference between total lipid yield and triglyceride yield is a measure of lipase activity and performance. The difference between total lipid yield and control (without lipase enzyme) shows the removal of oily stain from the fabric and demonstrates that these enzymes are stable and effective under realistic washing conditions, as simulated in the SLM test.

Claims (6)

1. Transformed microbial host cell, characterized in that it comprises an expression cassette which in its 5*-3' direction for transcription includes: (1) a transcription regulatory region and a translation initiation region functional in said host cell; (2) a DNA sequence encoding a lipolytic enzyme acquired from a prokaryotic microorganism, said DNA sequence being a 300 bp nucleic acid sequence of the lipase encoding gene on pTMPvl8A (CBS 142.89) or pET3 (CBS 155.87); and (3) translation and transcription termination regions functional in said host cell, wherein expression of said DNA sequence is regulated by said transcription and translation regions, and optionally comprises a leader sequence linked in the correct reading frame 5' to a gene that codes for said lipolytic enzyme, and that said expression cassette optionally comprises a marker gene.; 2. Transformed host cell according to claim 1, characterized in that the host cell is a prokaryotic cell.; 3. Transformed host cell according to claim 2, characterized in that the prokaryotic cell is a Bacillus, E. coli or Pseudomonas cell.; 4. Transformed host cell according to claim 1, characterized in that said DNA sequence is acquired from P. stutzeri Thai IV 17-1 (CBS 461.85) or P. pseudoalcaligenes M-1 (CBS 473.85).; 5. Method for the production of a lipolytic enzyme, characterized in that said method comprises: (1) cultivation in a nutrient medium of a host cell comprising an expression cassette which includes in the transcription 5*-3' direction a transcriptional regulatory region and a translation initiation region functional in said host cell; a DNA sequence encoding said lipolytic enzyme acquired from a prokaryotic microorganism; of which said DNA sequence is a 300 bp nucleic acid sequence from the lipase-encoding gene on pTMPvl8A (CBS 142.89) or pET3 (CBS 155.87) and translation and translation termination regions functional in said host cell, wherein expression of said DNA sequence is under regulatory control of said transcription - and translational regions; and (2) isolating said lipolytic enzyme from the nutrient medium. 6. Recombinant DNA construct, characterized in that it comprises: (1) in the direction of transcription, a transcriptional regulatory region which is functional in a microbial host cell; (2) a DNA sequence encoding a lipolytic enzyme obtained from a prokaryotic microorganism, said DNA sequence being a 300 bp nucleic acid sequence of the lipase encoding gene on pTMPvl8A (CBS 142.89) or pET3 (CBS 155.87); and (3) a transcription termination regulatory region which is functional in said host cell, and that said recombinant DNA construct optionally comprises at least one of a selection marker gene and one secretory leader sequence.