Streptomyces proteases and methods for improved secretion of recombinantly-expressed proteins

Abstract

A family of proteases endogenous to Streptomyces cells degrades heterologous proteins secreted from Streptomyces host cells. The previously unidentified proteases include (1) tripeptidyl aminopeptidase--Streptomyces ("Tap"), (2) a Streptomyces protease ("Ssp") which displayed significant amino acid sequence homology to Subtilisin BPN' and showed an ability to remove tripeptides from the amino termini of proteins and peptides, and (3) other proteases derived from Streptomyces which degraded certain substrates under certain conditions. Degradation was alleviated by selective inhibition of secreted proteases or by using hosts with impaired capabilities to produce proteases. An irreversible inhibitor was designed based upon the mechanism and substrate specificity of the target protease. Hosts secreting high amounts of proteases were selected. Impaired hosts were produced by deleting or altering the nucleotide sequence for the proteases.

Description

FIELD OF THE INVENTION

This invention relates generally to proteases produced by Streptomyces which degrade products expressed in genetically-engineered Streptomyces as hosts, inhibitors of such proteases, improved hosts with impaired protease systems, hosts selected for high expression of such proteases and the use of such proteases, inhibitors and improved hosts.

BACKGROUND OF THE INVENTION

Production methods employing recombinant technology use genetic expression systems. These systems generally consist of host cells encompassing a genetic system to be expressed, and expression vectors which introduce the genetic expression capabilities into the host cells. Under conditions allowing expression, a product, generally a protein, is made by the host cells.

Problems in commercial use of genetic expression systems arise because host cells have a variety of endogenous proteases, each with a specific action that may degrade the product. Degradation of product may also decrease the shelf lives of the bulk protein product and of the final dosage form of drugs.

Endogenous proteases degrade substrates in different ways. Aminopeptidases have broad substrate specificity, e.g., leucine aminopeptidase (Hanson and Frohne, 1976). However, when a proline residue is reached during degradation, such enzymes are unable to further degrade the peptide. Aminopeptidase P enzymes hydrolyse aminoacyl-proline bonds when proline is in the penultimate position from the amino terminus (X-Pro) of a polypeptide (Yoshimoto et al., 1988). After that action, proline aminopeptidase is capable of removing the exposed amino terminal proline residue.

Dipeptidyl peptidases have been found in many eukaryotic species (Kreil, 1990), but only in a few prokaryotic species (Lloyd et al., 1991; Fukusawa and Harada, 1981). These enzymes can remove N-terminal dipeptides including X-Pro dipeptides.

Tripeptidyl aminopeptidases are capable of degrading a peptide or polypeptide at its amino terminus by removing an amino acid triplet. Tripeptidyl aminopeptidase activity has not previously been reported in prokaryotes; however, serine proteases from human, rat and pig tissues with tripeptidyl aminopeptidase activity have been characterized (McDonald et al., 1985, Balon et al., 1986), and a cDNA sequence has been reported (Tomkinson and Jonsson, 1991).

Endoproteases can also cause rapid degradation of secreted proteins. Serine proteases are widespread throughout the prokaryotes as are metalloproteases. A wide variety of cleavage site specificities have been observed in various microbial species. Enzymes which cleave adjacent to positively charged, negatively charged, and aromatic amino acids have all been reported.

Proteases may be neutralized by various methods including by using inhibitors and by constructing improved strains with impaired proteases. The use of protease inhibitors to prevent the degradation of proteins during their purification is well established for proteins derived from yeast and higher eukaryotes. This approach has also been employed in the isolation and purification of proteins generated as inclusion bodies from E. coli. The general method involves lysing the protein source in the presence of broad spectrum protease inhibitors. Such inhibitors may include leupeptin, EDTA, phenylmethanesulfonylfluoride, or pepstatin.

The application of protease inhibitors in a system involving a living organism is more delicate. EDTA increases the fragility of many microorganisms and can cause cell lysis. Some inhibitors may be taken up by the organism. Such a process may lead to cell death or a disruption of cellular functions. Ideally, a protease inhibitor employed under these conditions should 1) be soluble in the fermentation media, 2) inhibit the target protease as selectively as possible, 3) not inhibit cell growth, and 4) be cost-effective.

Chloromethylketones are known to provide selective inhibition of some proteases. The earliest studied chloromethylketones, tosyllsine chloromethylketone (TLCK) and tosylphenylalanine chloromethylketone (TPCK), selectively inhibit trypsin and chymotrypsin, respectively (Schoellman et al., 1963, Shaw et al., 1965). Longer peptide sequences are needed for the inhibition of certain proteases and improve the specificity of the inhibition in some cases.

The use of improved strains with impaired proteases also can prevent degradation of proteins during production. Improved strains carrying deletional mutations in multiple protease-encoding genes have been made in Bacillus strains (Sloma et al, 1992). International Application Number PCT/US92/01598 of Omnigene, Inc. describes a Bacillus cell containing a mutation in the residual protease III gene resulting in the inhibition of the production by the cell of proteolytically active RP-III. In that case, the inactivation of the major protease allowed detection of other minor proteases which were still present in quantities sufficient to cause degradation of secreted products.

International Application Number PCT/US92/05532 of Amgen Inc. entitled "Isolation and Characterization of a Novel Protease from Streptomyces lividans" describes a protease called "Protease X" of S. lividans, its DNA and amino acid sequence, antibodies raised against such protease and a strain of S. lividans deficient in such protease. Protease X has different DNA and amino acid sequences than the proteases described in this application and cleaves different substrates than those described in this application.

A specific recombinant genetic expression system designated CANGENUS.TM. has been used to ferment and produce a variety of protein products, for example, granulocyte macrophage-colony stimulating factor (GM-CSF), interleukin-3 (IL-3), interleukin-6 (IL-6), and erythropoietin (EPO) (see Canadian Patent Numbers 1,295,563; 1,295,566; and 1,295,567; and U.S. Pat. No. 5,200,327).

Although the CANGENUS.TM. system has been successful in producing exogenous products, some undesirable proteases produced by expression of endogenous genes deleteriously affect the quality, quantity or stability of exogenous products.

Thus, a need exists to impair the action of these Streptomyces proteases. Among strategies which can be employed to meet this need are the use of inhibitors to inhibit the effect of proteases during the production processes and the use of improved strains which lack such proteases or which have impaired proteases.

Isolation of the protease genes could also be useful in the design of vectors directing the expression and secretion of heterologous proteins from Streptomyces. The promoter and signal sequence of such proteases could be used to enhance and direct the export of heterologous proteins from Streptomyces. The proteases themselves could be usefully employed to remove specific amino acid sequences, peptides or polypeptides from a protein. Furthermore, it would be useful if the level of expression of such proteases could be enhanced through mutation, selection or genetic engineering.

SUMMARY OF THE INVENTION

Streptomyces strains secrete a wide variety of heterologous proteins including GM-CSF, IL-3, IL-6, EPO, TNF, SCF, IL-7 and IL-2. These strains are useful in production of these proteins as desired products of commercial manufacturing systems. However, proteases of such Streptomyces strains impair the quality and quantity of secreted proteins. Before this invention, attempts to improve the quality and quantity of such proteins were not successful. This invention meets that goal by (A) inhibiting certain Streptomyces proteases, and (B) providing new Streptomyces strains which lack or have impaired degradative proteases.

To circumvent protein degradation, this invention uses selective inhibitors which are capable of protecting secreted peptides, and polypeptides including heterologous protein biopharmaceuticals from degradation by secreted host proteases.

This invention also uses improved strains that have impaired protease production systems, yet which are capable of expressing desired products.

This invention relates to the selection of Streptomiyces strains with enhanced expression of proteases and the isolation and purification of Streptomyces proteases. An embodiment of a protease is a tripeptidyl aminopeptidase designated Tap. Amino acid sequences of the proteases and substantially equivalent sequences are aspects of the present invention. Promoters and signal sequences of such proteases are further aspects of the present invention.

A signal sequence is typically composed of the amino-terminal portion of the unprocessed polypeptide, extending from the amino terminal residue to the beginning of the mature protein sequence. The signal sequence is typically a small peptide which directs the protein to a particular cellular or extracellular location, or for export from the cell, at which point the signal peptide is preferably cleaved.

This invention also relates to nucleotide sequences encoding impaired proteases and the use of those sequences to increase the quality, quantity or stability of peptides and polypeptides including heterologous proteins secreted from a host transformed with a vector containing the nucleotide sequence for such impaired proteases.

This invention also relates to the use of the isolated and purified proteases to cleave peptides or polypeptides or to cleave amino acids, peptides or polypeptides from a protein.

A further aspect of this invention is the construction of an inhibitor comprising, L-alanyl-L-prolyl-L-alanine chloromethylketone, its salts and analogs. Another aspect of this invention is the use of the inhibitor L-alanyl-L-prolyl-L-alanine chloromethylketone to inhibit a tripeptidyl aminopeptidase derived from Streptomyces.

This invention also relates to a method of increasing the quality, quantity or stability of peptides or polypeptides including heterologous proteins secreted from a host by using an inhibitor comprising, L-alanyl-L-prolyl-L-alanine chloromethylketone.

Another aspect of this invention is the construction of an improved Streptomyces strain having impaired expression of at least one endogenous protease. The strain is capable of expressing an exogenous gene product S. lividans, S. ambofaciens, S. coelicolor, S. alboniger, S. fradiae, S. griseus, S. parvulus and S. rimosus. The impaired expression decreases the activity or quantity of endogenous protease resulting in an increase in quality, quantity or stability of exogenous gene product.

Impaired expression is accomplished by deleting or mutating one or more nucleotides in the sequence encoding for a protease, or by deleting and substituting nucleotides in the sequence encoding for a protease.

A further aspect of this invention is a vector which has a recombinant DNA sequence encoding a Streptomyces protease or an impaired Streptomyces protease and a regulatory sequence for expression of the coding sequence. The regulatory sequence includes a promoter sequence, an operator sequence, a transcriptional-start sequence, a ribosome-binding site sequence, and a signal sequence.

Another aspect of this invention is a method of fermentation using genetically engineered Streptomyces host cells with impaired protease activity, The method includes the steps of: (a) constructing Streptomyces host cells with impaired protease activity and which express a desired exogenous product under suitable conditions; and (b) placing the cells in suitable conditions for expression of the desired product. The method of fermentation can be used to express GM-CSF, IL-3, IL-6, EPO, TNF, SCF, IL-7 and IL-2 or any other desired product.

In another aspect, this invention envisions introducing the DNA sequences encoding such proteases into recombinant vectors which, when transformed into suitable host strains, enable the production of heterologous proteases having the biological activity of the wild type proteases. Both prokaryotic and eucaryotic hosts may serve as hosts for producing such proteases.

Further aspects of this invention are kits containing (a) isolated and purified proteases derived from Streptomyces, or (b) inhibitors of proteases derived from Streptomyces.

A kit for ELISA would consist of:

1) A protease, Tap, covalently linked to biotin or other carrier capable of participating in the formation of an antigen-antibody complex (example: Tap covalently linked to a goat antirabbit IgG); 2) A substrate, APA-pNA or APA-AMC, which would be cleaved by the Tap bound in the antigen-antibody complex thereby generating an increase in light absorbance at 405 nm with APA-pNA as substrate or an increase in fluorescence when an excitation/emission near 380/460 nm is employed with APA-AMC as substrate.

The present invention describes a method for improving the secretion of mature protein from a genetic expression system. The levels of secreted proteins that are increased are those that have amino terminal structures that interfere with the processing of the signal peptide (structural constraints). Secretion of heterologous proteins by a genetic expression system is improved by adding tripeptides (propeptides) to the amino terminal end of the protein which is a precursor to the desired product of the system, the addition occurring immediately adjacent to the signal peptidase cleavage site, allowing the cleavage to occur to form a mature protein, and then removing the tripeptide from the mature protein by use of a protease such as Tap.

The invention also relates various new protease such as SlpD and SlpE that are useful to attach polypeptides to bacteria during processing.

In this application, the following terms have the following meanings:

"Heterologous" or "exogenous" refers to nucleic acids, amino acids, peptides, polypeptides or proteins which do not naturally occur in a particular host cell.

"Host cell" means a prokaryotic or eucaryotic cell, strain, species or genera, suitable for introduction and for expression of heterologous DNA sequences. Such DNA sequences may be modified for expression in a particular host as a DNA sequence containing (i) codons preferably used by the host, or (2) promoters, operators, ribosome binding sites and terminator sequences used by the host.

"Substantially equivalent" in reference to a sequence means a sequence, whether natural or engineered, which has additions, deletions, or substitutions compared to the sequence of another protease described or claimed in this application and which produces a functionally similar protease to the protease described or claimed.

"Wild type" means the activity characteristic of a host cell in which endogenous proteases are not impaired. Illustrative embodiments of impaired proteases include a host strain in which DNA at the chromosomal locus encoding a protease in a Streptomyces strain is deleted. This strain exhibits a significantly reduced level of activity or no activity when compared to a wild type Streptomyces strain.

"Impaired" means that the activity and/or the quantity of protease produced by a nucleotide sequence is impaired compared to a "wild type" nucleotide sequence, that is, a sequence not altered to affect expression as it generally occurs in the host species and strain.

"Endogenous protease" means a protease that is able to cleave one or more of the substrates referred to in this application.

"Selective inhibitor" means an inhibitory molecule that inhibits a secreted protease, or a protease released into the fermentation as a result of cell breakage.

______________________________________ABBREVIATIONS______________________________________3 = protein from which three amino acid residues have been removed from the N- terminus of the protein4 = protein from which four amino acid residues have been removed from the N- terminus of the protein6 = protein from which six amino acid residues have been removed from the N- terminus of the proteinaa = amino acidAAPA-pNA = L-alanyl-L-alanyl-L-prolyl-L-analine p- nitroanilideAA-pNA = L-alanyl-L-alanine p-nitroanilideAMC = 7-amino-4-methylcoumarinAPACMK = L-alanyl-L-prolyl-L-alanine chloromethylketoneAPA-AMC = L-alanyl-L-prolyl-L-alanine 7-amino-4- methylcoumarinAPF-bNA = L-alanyl-L-prolyl-L-phenylalanine beta- naphthylamideAPA-pNA = L-alanyl-L-prolyl-L-alanine p- nitroanilideAPM-pNA = L-alanyl-L-prolyl-L-methione p- nitroanilideA-pNA = L-alanine p-nitroanilideAPS-bNA = L-alanyl-L-prolyl-L-serine beta- naphthylamidebNA = beta-naphthylamideBoc = N-t-butoxycarbonylBoc-AAPA-pNA = N-t-butoxycarbonyl L-alanyl-L-alanyl-L- prolyl-L-alanine p-nitroanilideBoc-APARSPA-bNA = L-alanyl-L-prolyl-L-analyl-L-arginyl-L- seryl-L-prolyl-L-alanine beta- naphthylamideD-FPR-bNA = D-phenylalanyl-L-prolyl-L-arginine beta- naphthylamideDMSO = dimethyl sulphoxideD-PFR-pNA = D-prolyl-L-phenylalanyl-L-arginine p- nitroanilideEDTA = Ethylenediaminetetraacetic acidELISA = enzyme-linked immunosorbent-assayFPLC = fast protein liquid chromatographyGPL-bNA = Glycyl-L-prolyl-L-leucine beta- napthylamideGP-pNA = Glycyl-L-proline p-nitroanilideGPM = Glycly-L-prolyl-L-methionineHEPES = N-2-hydroxyethylpiperazine-N'-2- ethanesulphonic acidHOHD = 2-hydroxy-6-oxohepta-2,4-dienoateL-pNA = L-leucine p-nitroanilideMNNG = N-methyl-N'-nitro-N-nitrosoguanidineN-Ac = N-acetylN-Ac-APA-pNA = N-acetyl-L-alanyl-L-prolyl-L-alanine p- nitroanilideN-Bz = N-benzoylN-Bz-R-pNA = N-benozyl-L-arginineN-Bz-VGR-pNA = N-benzoyl-L-alanyl-glycyl-L-arginine p- nitroanilident = nucleotideORF = open reading framePAGE = polyacylamide gel electrophoresisPMSF = phenylmethanesulfonyl fluoridepNA = p-nitroanilineP-pNA = L-proline p-nitroanilideR-pNA = L-arginine p-nitroanilideSDS = sodium dodecyl sulphateS-bNA = L-serine beta-naphthylamideSPA-bNA = L-seryl-L-prolyl-L-alanine beta- naphthylamideSsp = Streptomyces Subtilisin-like proteinssp = gene encoding SspTap = tripeptidyl aminopeptidase-Stap = gene encoding TapTSB = Trypicase Soya Broth______________________________________

DESCRIPTION OF DRAWINGS

FIGS. 1A-1B. Degradation of GM-CSF and IL-3 by S. lividans fermentation broth.

FIG. 2. Cleavage of synthetic substrates by S. lividans fermentation broth.

FIGS. 3A-3B. Demonstration of purification of Tap.

FIG. 4. Inhibition of IL-3 cleavage by Tap after PMSF treatment.

FIG. 5. Inhibition of Tap by APACMK: GM-CSF assay.

FIG. 6. Inhibition of Tap by APACMK: APA-PNA assay.

FIG. 7. Inhibition of degradation of GM-CSF during fermentation in the presence of APACMK.

FIG. 8. (A) Common restriction map for tap-containing plasmid DNA isolated from clone P3-13 (and P3-5).

FIG. 8. (B) The tap-deletion clones.

FIG. 8. (C) The tap-integration clones.

FIG. 9. Southern hybridization analysis of chromosomal DNA from S. lividans 66 and S. lividans MS7, using DNA from the P3-13 plasmid (0.3 kb BglII) as a probe.

FIG. 10. Profiles of extracellular proteins from S. lividans 66 strains carrying the P3-5 and P3-13 clones; the profiles were generated by SDS-PAGE and the gels stained with Coomassie Brilliant Blue.

FIG. 11. Conversion of the substrate of intact GM-CSF to its "-3 form" upon incubation with fermentation culture supernatants from cells carrying the tap clones.

FIG. 12A-12B. Nucleic acid and encoded amino acid sequences of the tripeptidyl aminopeptidase (tap) gene (SEQ ID NOS 1 and 2).

FIG. 13. Amino acid sequence similarity between Tap (residues 199-228 of SEQ ID NO:2) and HOHD (SEQ ID NO:11) from Pseudomonas putida F1.

FIG. 14. Activity of fermentation culture supernatants from S. lividans MS5 (tap+) and S. lividans MS7 (tap-) strains against chromogenic tripeptide substrates.

FIG. 15. Reduction in the rate of degradation of intact GM-CSF by fermentation supernatants of cultures of the tap mutant.

FIG. 16. PAGE resolution and Coomassie Brilliant Blue staining of fermentation supernatants from cultures of S. lividans 66 and S. lividans MS7 mutant protoplasts transformed with the GM-CSF expression vector pAPO.GMCSF.

FIG. 17. Homologs of tap are present in many Streptomyces strains.

FIG. 18. Common restriction map for P5-4 and P5-15 and their deletion clones.

FIG. 19. SDS-PAGE resolution and silver staining of proteins secreted in a fermentation culture containing the P5-4 plasmid DNA.

FIGS. 20A-20C. Nucleic acid and encoded amino acid sequences of the cloned P5-4 DNA (SEQ ID NOS 3 and 4).

FIG. 21. Comparison of the predicted amino acid sequence encoded by the P5-4 DNA (SEQ ID NO:4) and that of subtilisin BPN (SEQ ID NO:12).

FIG. 22. Proteolytic activity of S. lividans deletion strains using the substrate APA-pNA.

FIG. 23. Homologs of the P5-4 DNA are present in the chromosomal DNA of many Streptomyces strains.

FIG. 24. Common restriction map for P5-6 and P5-15 and their deletion clones.

FIG. 25A-25C. Nucleic acid and predicted amino acid sequence of P5-6 DNA (SEQ ID NOS 7 and 8).

FIG. 26. Comparison of the predicted amino acid sequences for the Tap (SEQ ID NO:2) and P5-6-encoded putative protein (SEQ ID NO:8).

FIG. 27. Restriction map of P5-10 DNA.

FIG. 28. Restriction map of P8-2 and its deletion clone.

FIG. 29A-29C. Nucleic acid and predicted amino acid sequence of P8-2 (SEQ ID NOS 5 and 6).

FIG. 30. Conversion of an intact substrate GM-CSF to its "-3 form" upon incubation with fermentation culture supernatants from cells carrying P5-4, P5-10 and P5-15.

FIG. 31. Conversion of an intact GM-CSF to its "-3 form" upon incubation with fermentation culture supernatants from cells carrying P5-6, P5-10 and P5-17.

FIG. 32A-32B. Demonstration of the use of Tap in ELISA technology by standard calibration curve in hIL-3.

FIG. 33A. The AP6.H vector.

FIG. 33B. The AP6.SlpD vector.

FIG. 33C. The AP6.SlpE vector.

FIG. 34. Protein (SCF) secretion of AP3, AP6, AP6 and APz constructs analyzed by SDS PAGE and visualized by silver staining.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

A previously unknown protease, a tripeptidyl aminopeptidase ("Tap") derived from Streptomyces, has been identified, isolated, and characterized. The enzyme was purified by pH precipitation and chromatography. The proteolytic activity was followed both by assaying the degradation of GM-CSF and by the release of the yellow p-nitroaniline molecule from the specially synthesized substrate L-alanyl-L-prolyl-L-aniline p-nitroanilide (APA-pNA). The pure protease had an apparent molecular weight of 55,000 daltons as determined by SDS-PAGE. The amino terminal sequence of the purified protease was determined by Edman degradation of the protein after purification.

A selective inhibitor of Tap, L-alanyl-L-prolyl-L-alanine chloromethylketone (APACMK), has been designed, synthesized, and applied to inhibit this protease. APACMK stopped the release of p-nitroaniline from APApNA by Tap. APACMK stopped the cleavage of GM-CSF by Tap. In fermentations of GM-CSF, APACMK prevented cleavage of GM-CSF by Tap during fermentation but did not significantly retard the rate of cell growth.

A tripeptidyl aminopeptidase gene (tap) was cloned from S. lividans 66 by screening for overexpression of endogenous enzyme activity using the chromogenic substrate GPL-bNA as a liquid overlayer on colonies of Streptomyces growing on agar medium. When these colonies were selected on the basis of activity they exhibited according to the chromogenic assay disclosed herein, and were grown in liquid culture, a major secreted protein with an estimated apparent molecular weight of 55,000 daltons as determined by SDS-PAGE was identified in the culture supernatant. The appearance of this protein was correlated with elevated levels of Tap activity in liquid assays using GPL-bNA and other substrates, suggesting that Tap presence was causative of the activity detected by the assay.

The amino terminal sequence of the overexpressed protein was determined by various procedures, e.g., by Edman degradation of the protein after purification by SDS-PAGE. The amino terminal sequence of the overexpressed protein matched the amino terminal sequence of Tap isolated from fermentations of the host strain. The tap gene was localized within the cloned DNA fragment by monitoring the Tap activity of strains containing various subclones and deletion clones derived from the original clones.

DNA sequences adjacent to the tap gene were used to construct a subclone in which the tap gene was precisely deleted. This deletion clone was then substituted into the chromosomes of S. lividans 66 strains by homologous recombination to replace the wild type tap locus with a mutant gene which encoded a defective Tap.

Disruption of the chromosomal tap gene in S. lividans resulted in a reduction in Tap activity of at least tenfold, indicating that this enzyme was responsible for the majority of the activity observed in S. lividans strains. Deletional inactivation of the gene encoding a second protease (Ssp) resulted in a further reduction in the ability of cell-free broth to hydrolyse APA-bNA. Strains carrying such chromosomal DNA deletions generally exhibited significantly lower Tap activity (FIG. 22), reducing the degradation of proteins produced by genetically engineered host cells, and enabling higher recovery of secreted proteins from the culture supernatant produced by fermentation of the host strain in liquid medium.

I. Prokaryotic Tripeptidyl Aminopeptidases Tripeptidyl aminopeptidase

Degradation products were found in fermentations producing GM-CSF and IL-3.

FIG. 1 shows the degradation products derived from GM-CSF and IL-3. (A) shows a native gel electrophoresis analysis of GM-CSF degradation. Lane 1 shows intact, full length GM-CSF. Lane 2 shows GM-CSF from S. lividans fermentation. Lane 3 shows degraded isolated GM-CSF(-3). Lane 4 shows a mixture of isolated GM-CSF(-4) and GM-CSF(-6). (B) shows an analysis of IL-3 degradation by electrophoresis on an SDS-urea gel (6M urea in the polyacrylamide gel). A 20-fold concentrated fermentation broth was prepared by subjecting a cell-free fermentation broth to ultrafiltration employing a membrane with a 10 kDa cutoff. Lane 1 shows IL-3 before incubation. Lane 2 shows IL-3 after 2 hours incubation at 32.degree. C.

The major degradation products were isolated and analyzed by amino acid sequencing. This analysis indicated that the major degradation products (FIG. 1A, Lane 3 and FIG. 4, Lane 5) were produced by the removal of the N-terminal tripeptides, APA and APM, from GM-CSF and IL-3, respectively.

Based upon this information, the molecule APA-pNA was synthesized as a potential substrate. This and several commercial substrates were employed in a survey of proteolytic activities in S. lividans fermentation broths.

FIG. 2 is the quantification of proteolytic activities in the fermentation broth as measured with synthetic substrates. The assays were conducted in 50 mM Tris-HCl, pH 8.0 with 0.8 mM substrate incubated at 37.degree. C. The change in absorbance at 405 nm was measured after 1, 2, and 4 hours of incubation. The results are reported as micromoles of p-nitroaniline released in 1 hour by 1.0 ml of fermentation broth. 1=APA-pNA; 2=D-PFR-pNA; 3=L-pNA; 4=R-pNA; 5=P-pNA; 6=AP-pNA; 7=A-pNA; 8=AA-pNA; 9=N-Benzoyl-R-pNA; 10=Boc-AAPA-pNA; 11=N-Acetyl-APA-pNA; 12=N-Benzoyl-Y-pNA.

As shown in FIG. 2, APA-pNA cleaving activity was greater than any other activity measured in the broth. This data suggested that a single protease, a tripeptidyl peptidase, not a group of several enzymes, was responsible for the activity. Additionally, the lack of activity towards the amino-blocked analog, N-Ac-APA-pNA, indicated that the enzyme responsible was an aminopeptidase.

The wild-type protease was purified after cell removal and concentration of the fermentation broth by ultrafiltration. The method of purification is described in Example 1. To purify Tap (FIG. 3A-3B), approximately 20 ug of protein were denatured under reducing conditions and analyzed by SDS-PAGE on 10% polyacrylamide gel. (A) represents purification of wild-type Tap. St=Molecular weight standards; Lane 1=Broth obtained after cell removal and concentration of broth by ultrafiltration through a 10 kDa membrane; Lane 2=Redissolved pH 4.0 precipitate; Lane 3=Q-Sepharose chromatography pool; Lane 4=Phenyl-Sepharose chromatography pool. (B) represents purified Tap from the overproducer strain. St=Molecular weight standards; Lane 1=Tap purified from fermentation of the overexpressor (P3-5) strain.

The pure protease cleaved the N-terminal tripeptide from GM-CSF and cleaved the N-terminal tripeptide from IL-3. When GM-CSF or IL-3 were used as a substrate, the cleaved products produced by the pure Tap were identical to the major degradation products found in Streptomyces fermentations. These assays are described in Example 2.

As described in Example 2, Tap releases p-nitroaniline from APApNA. The enzyme was also active when APM-pNA, APA-AMC, APS-bNA, GPL-bNA, and SPA-bNA were used as substrates. It did not release the reporter group from A-pNA, L-pNA, P-pNA, R-pNA, S-bNA, N-Bz-R-pNA, AA-pNA, GP-pNA, D-PFR-pNA, N-Ac-APA-pNA, N-Bz-VGR-pNA, AAPA-pNA, Boc-AAPA-pNA, and Boc-APARSPA-bNA. The enzyme only released the reporter group from substrates with a free amino terminal. The enzyme cleaved only tripeptide units since no reporter release was seen with mono-, di-, or tetra-amino acid substrates.

The effect of pH on the activity of Tap has been examined. When APA-PNA was used as a substrate, the enzyme was active from between pH 5.0-9.5 with the maximal activity obtained from between 8.0-8.5. The enzyme cleaved GM-CSF from between pH 4.0-10.0 with greatest activity from between 5.0-9.0. The broad maximum for GM-CSF reflected the high sensitivity of this substrate to Tap. The enzyme cleaved IL-3 from between pH 5.0-9.0 with maximal activity attained between 7.0 and 8.5.

An inhibitor survey indicated that tripeptidyl aminopeptidase was a serine protease. Table I shows the inhibition of Tap activity by various protease inhibitors. The protease and inhibitor were preincubated for 15 minutes at 22.degree. C. Substrate was added and the mixture was incubated at 37.degree. C. Activity was measured by monitoring the change in absorbance at .lambda.=405 nm.

TABLE I______________________________________Inhibition of TAP in the APA-pNA AssaySample Concentration Residual Activity______________________________________Enzyme only -- 100PMSF 1.6 mM 7HgCl.sub.1 0.1 mM 99 1.0 mM 93CaCl.sub.2 1.0 mM 96 10 mM 97CoCl.sub.2 1.0 mM 98 10 mM 97EDTA 1.0 mM 95 10 mM 95IDA 1.0 mM 82DTT 1 mM 86DTT + EDTA 1 mM + 10 mM 97 (respectively)Elastatinal 0.1 mM 97Chymostatin 0.1 mM 98Pepstatin 0.1 mM 95Benzamidine 10 mM 94______________________________________

The enzyme was inhibited by the serine protease inhibitor, phenylmethanesulfonyl fluoride (PMSF). Treatment of Tap with PMSF inhibited cleavage of GM-CSF, IL-3, and APA-PNA.

The inhibition of IL-3 cleavage is demonstrated in FIG. 4 and the inactivation protocol is described in Example 3. Lanes 1-4 show the incubation of IL-3 with TAP-S that has been treated with PMSF. Lane 1=4 hrs; Lane 2=2 hours; Lane 3=1 hr.; Lane 4=0 hours. Lanes 5-8 show the incubation of IL-3 with uninhibited Tap. Lane 5=4 hrs.; Lane 6=2 hrs.; Lane 7=1 hrs.; Lane 8=0 hrs. Lane 9 is a human carbonic anhydrase marker, pI=7.4. Lane 10 contains pI markers. As can be seen in Lanes 5-8 of FIG. 4, the IL-3 (pI=7.4) is completely converted to the -3 form (pI=7.1) by Tap within 2 hours. Lanes 1-4 show that with PMSF treatment, intact IL-3 is clearly detected after 4 hours. The enzyme is not affected by sulfhydryl reagents, chelators or aspartyl protease inhibitors (Table I).

Table II shows the N-terminal sequence of the isolated wild-type Tap. The sequence data was obtained as described in Example 4.

TABLE II______________________________________N-Terminal Sequence of Isolated TapCycle Amino Acid, Wild-Type______________________________________1 Asp2 Gly3 His4 Gly5 His6 Gly7 Arg8 Ser9 Trp10 Asp11 Arg12 Glu13 Ala14 Arg15 Gly______________________________________ II. L-Alanyl-L-Prolyl-L-Alanine Chloromethylketone (APACMK)

The synthesis of APACMK is described in Example 5.

APACMK inactivated Tap at very low concentrations when residual activity was assayed with GM-CSF or APA-pNA respectively (FIGS. 5 and 6).

FIG. 5 shows the titration of Tap with APACMK as assayed with GM-CSF. The assay was performed as described in Example 7. The Tap concentration in the assays was 5 nM. Lane 1=GM-CSF standard; Lane 2=GM-CSF after digestion with Tap in the absence of APACMK; Lanes 3 and 4=150 uM APACMK; Lanes 5 and 6=15 uM APACMK; Lanes 7 and 8=1.5 uM APACMK; Lanes 9 and 10=150 nM APACMK; Lanes 11 and 12=15 nM APACMK; Lanes 13 and 14=1.5 nM APACMK; Lanes 15 and 16=150 pM APACMK.

FIG. 6 shows the inactivation of Tap by various APACMK concentrations when assayed with APA-pNA as substrate. The concentration of Tap in the inactivations was 1.0 uM. The inactivation and assay were conducted as described in Example 8. In FIG. 6, (O)=2.70 uM APACMK; (.DELTA.)=2.16 uM APACMK; (.quadrature.)=1.73 uM APACMK; (*)=1.38 uM APACMK; (+)=No APACMK.

The inhibitor APACMK yielded K.sub.i =3.3 uM and k.sub.inact =0.14 min.sup.-1 with >99% inactivation within 6 minutes at 0.degree. C. at an inhibitor concentration of 2.7 uM and an inhibitor/enzyme molar ratio of 2.7 (FIG. 6). The methods employed are described in Examples 6, 7, and 8.

FIG. 7 demonstrates the inhibition of Tap by APACMK during the fermentation of S. lividans grown in the presence and absence of 10 uM APACMK as described in Example 9. When APACMK and GM-CSF were added to the protease-containing broth from S. lividans fermentations, the formation of the GM-CSF(-3) degradation product was inhibited. Lane 1=Standard containing GM-CSF and GM-CSF(-3). Lanes 2-6 show a fermentation in the presence of 10 uM APACMK. Lane 2=25 hours growth; Lane 3=27 hours growth; Lane 4=29 hours growth; Lane 5=31 hours growth; Lane 6=48 hours growth. Lanes 7-11 show a fermentation without APACMK. Lane 7=25 hours growth; Lane 8=27 hours growth; Lane 9=29 hours growth; Lane 10=31 hours growth; Lane 11=48 hours growth. GM-CSF degradation was analyzed by native gel electrophoresis.

III. Nucleotide Sequence Encoding Streptomyces Proteases and Amino Acid Sequence of Such Proteases

Methods of identifying and isolating the DNA encoding Tap are described in Example 10.

FIG. 8 is a restriction enzyme site map of cloned tap DNA. FIG. 8(A) The location and direction of potential protein encoding regions is shown by arrows, of which the larger represents the tap gene. Phenotype in the GPL-bNA hydrolysis agar plate assay is shown qualitatively as the number of + signs judging red color developed on the colonies. The EcoRI site shown in parentheses was present in the pSS12 vector adjacent to the BamHI cloning site. FIG. 8(B) None of the three deletion clones shown produced any more red color in colonies than did the pSS12 control plasmid and they were scored as "+" due to the background level of hydrolysis from the chromosomally-encoded tap gene in the S. lividans 66 host. (C) The DNA fragments shown were subcloned into the integration plasmid and used to transform protoplasts of S. lividans 66 to thiostrepton resistance. Clone numbers 1, 2, 4 and 5 all produced thiostrepton-resistant transformants, whereas clone 3 did not presumably due to the small size of the homologous DNA fragment in this clone.

FIG. 9 is a Southern hybridization analysis of the chromosomal tap locus in Streptomyces lividans 66 and deletion mutant strains. The DNA was digested with BamHI or StuI and transferred to a nylon membrane (Hybond, Amersham). Using a .sup.32 P-labelled probe for the BglII fragment internal to the tap gene resulted in a strong band of hybridization at approximately 1.8 kbp in the BamHI digests (lanes 2 and 5) and two bands in the StuI digests (lanes 6 and 9) for both the S. lividans control and colony #3 indicating that this DNA fragment was present in both strains. However, no hybridizing bands were observed for colonies 2 and 3 (lanes 3, 4, 7 and 8) confirming the loss of the 0.3 kbp BglII fragment. Lanes 1 and 10 show a Lambda / HindIII molecular weight marker.

FIG. 10 is an SDS-PAGE analysis of cell-free broth supernatants from cultures of S. lividans 66 carrying the P3-13 or P3-5 plasmids. Cultures were sampled at 23 or 29 hours after inoculation into TSB medium.

FIG. 11 is a conversion of exogenously added, purified full length GM-CSF degraded to the -3 form upon incubation with fermentation culture supernatants from culture samples carrying the tap clones.

The nucleic acid sequence (SEQ ID NO:1) for the S. lividans tap gene is shown in FIGS. 12A-12C. The deduced amino acid sequence (SEQ ID NO:2) is shown for each codon.

Serine proteases possess a nucleophilic serine which attacks the carbonyl of the peptide bond to catalyze hydrolysis (White, Handler and Smith, 1973). Although the nucleophilic serine modified by PMSF has not been isolated, a homology study of the DNA sequence can identify potential candidates. The protease is encoded by the DNA sequence (SEQ ID NO:1) shown in FIG. 12A-12C. The amino acid sequence (SEQ ID NO:2) derived from the DNA sequence is also shown.

The most likely active site serine residue was identified by its homology with that described for a serine esterase enzyme characterized in a Pseudomonas species by the conserved amino acid sequence motif G X S X G (SEQ ID NO:9) (Menn et al., 1989). The homologous sequence in Tap would be GVSYG (residues 204-208 of SEQ ID NO:2).

FIG. 13 is the amino acid sequence similarity between Tap (residues 199-228 of SEQ ID NO:2) and the HOHD from Pseudomonas putida F1. The amino acid sequences were compared using the BLAST (Altschul et al) program to screen the protein sequence databases.

The first 15 residues of the N-terminal of the isolated wild-type protease (Table II) have been determined and identically matched amino acids 40-54 derived from the DNA sequence (FIG. 12A-12C) SEQ ID NOS 1 and 2. Residues -39 to -4 appear to be a signal peptide. An autolytic tripeptide cleavage removing APA after signal peptide removal would yield the N-terminal found for the secreted protease.

Table III shows the amino acid composition of the wild-type Tap. The amino acid composition derived from the corresponding portion of the tap SEQ ID NOS 1 and 2 gene DNA sequence (FIG. 12) is shown for comparison. The composition data was obtained as described in Example 4.

The small differences in composition may be attributable to low level impurities in the enzyme sample. The method of analysis for the wild type enzyme is described in Example 4.

The N-terminal of the protease from the overproducer (P3-5) (Example 13) matches the sequence of the N-terminal of the wild-type enzyme. Both the isolated wild-type and isolated overproducer proteases had an apparent molecular weight of 55,000 daltons as determined by SDS-PAGE (FIG. 3). These factors indicated that the wild-type protease and the P3-5 overproduced protease were the same enzyme.

A further embodiment of this invention relates to the use of strains, containing specific impairments in their capability to

TABLE III______________________________________Mole percentageAmino Acid Protein DNA______________________________________Asp + Asn 13.6 12.4Glu + Gln 10.9 7.6Ser 4.7 4.7Gly 10.0 8.9His 2.2 2.3Arg 7.4 7.4Thr 6.3 6.3Ala 14.3 14.3Pro 7.2 7.2Tyr 3.9 3.8Val 6.4 7.6Met 1.2 1.3Ile 2.3 3.0Leu 5.6 6.3Phe 1.7 2.5Lys 2.5 4.4______________________________________ produce secreted proteases, and the isolation and purification of other proteases which cleave substrates such as APAbNA and which also exist in the wild type strain but are expressed at much lower levels than Tap. Methods are described in Examples 20-23 to identify the genes encoding such minor proteolytic activities. It would be extremely difficult to purify such proteases from the wild type strain whereas the methods described here are rapid and simple. One protease (designated Ssp) having significant amino acid sequence homology with the B.subtilis protein Subtilisin BPN was identified by virtue of its ability to cleave APA-bNA using the agar plate assay screening method. Furthermore, deletion of this gene from the S. lividans chromosome in a strain in which the tap gene had already been inactivated resulted in an incremental reduction in the APA-pNA hydrolytic capability of the strain.

Another protease gene was identified and shown to encode a protease which catalyzed the hydrolysis of APA-pNA and also showed a significant amino acid sequence homology to that of the Tap. Particularly strong sequence conservation was noticed around the putative active site serine residue of the Tap.

IV. Methods of Preparing Nucleic Acid Sequences Capable of Coding For the Impaired Proteases

Methods of preparing nucleic acid sequences capable of coding for the impaired proteases include: site specific mutagenesis to alter the sequence coding for an essential component of the activity and/or the expression of the protease; and deletion or mutation of the wild type gene by exposure to mutagens. Generally, the deletion of a wild type gene together with the insertion of an impaired gene, would be preferred.

Example 15 describes production of DNA clones with various deletions and mutations resulting in the identification of DNA sequences the removal of which lead to inactivation of the tap gene.

V. Methods of Producing Host Cells with Impaired Protease Activity

Vectors were prepared according to Section III.

Recombinant vectors and isolated segments may therefore variously include the basic protease active site encoding region in an inactive form, coding regions bearing selected alterations or modifications in the basic coding regions, or larger proteins which include the basic coding region. An example is shown in FIG. 8B. In any event, it should be appreciated that due to codon redundancy, this aspect of the invention is not limited to alteration of the particular DNA sequences shown in FIGS. 8B or 8C.

Recombinant vectors such as the foregoing are useful both as a means for preparing quantities of the protease-encoding DNA itself, or as a means of producing defective proteases for use in transforming recombinant host cells for use in fermentation processes to produce various peptides and proteins.

Example 16 describes the use of the deletion clones of the tap gene for integrational mutation into the S. lividans 66 chromosome resulting in inactivation of the wild type tap gene. Loss of the wild type tap gene occurred by homologous recombination with the integrated mutant DNA sequence using the natural ability of the S. lividans host cell to resolve such regions of chromosomal DNA containing directly repeated nucleotide sequences. Resolution occurred apparently at random to produce strains carrying either the wild type parental tap gene or the exchanged mutant tap gene. Mutant strains were identified by their inability to hydrolyse the chromogenic substrate GPL-bNA.

Example 14 describes the use of chemical mutagenic treatment of spores of the S. lividans 66 strain to produce mutant strains in which the Tap encoding DNA is defective, resulting in reduced or abolished expression of Tap.

EXAMPLES

Example 1

Purification of Wild-Type Tripeptidyl Aminopeptidase

S. lividans 66 was grown in 11 liters of minimal media (minimal media=12 g Difco Soytone, 10.6 g K.sub.2 HPO.sub.4, 5.3 g KH.sub.2 PO.sub.4, 2.5 g (NH.sub.4).sub.2 SO.sub.4, and 1.0 g MgSO.sub.4 -7H.sub.2 O per liter) for 24 hrs at 32.degree. C. with stirring at 300 rpm in a Chemap fermenter. Cells were removed from the media by ultrafiltration with a 0.45 um filter (Pellicon System, Millipore). Proteins in the filtrate were concentrated by ultrafiltration employing a membrane with a 10 kDa cutoff (Millipore). The protease activity was followed by assaying with APApNA and GMCSF as described in Example 2 . The protease was precipitated at 4.degree. C. by lowering the pH to 4.0 with 0.1M HCl. The precipitate was collected by centrifugation (Model J2-21, Beckman) at 10,000 g at 4.degree.-10.degree. C. and was redissolved in 50 ml 10 mM Tris-HCl, pH 8.0. After dialysis against 4 liters of the Tris buffer at 4.degree. C., the protease was loaded at ambient temperature onto a 1.6.times.10 cm anion exchange column (Q-Sepharose Fast Flow, Pharmacia) equilibrated with the Tris buffer. After washing with equilibration buffer, the bound protease was eluted with a 200 ml gradient from 0 to 500 mM NaCl at a flow rate of 2 ml/minute. The active fractions were pooled and made 2M in ammonium sulfate. This material was loaded at ambient temperature onto a 1.6.times.10 cm hydrophobic interaction column (Phenyl-Sepharose Fast Flow, Pharmacia) equilibrated in 10 mM Tris-HCl, pH 8.0, 2M ammonium sulfate. After washing with equilibration buffer, the column was eluted with a 200 ml gradient from to 2 to 0M ammonium sulfate at a flow rate of 2 ml/minute. The active fractions were assayed for purity by SDS-PAGE.

Example 2

Assays of Tap Activity

Aliquots of Tap column fractions were diluted 100-fold with 20 mM Tris-HCl, pH 8.0.

GM-CSF as Substrate

To 10 ul of rhGM-CSF (10 ug, Cangene) and 20 ul 20 mM Tris-HCl, pH 8.0, 20 ul of Tap were added. The assays were incubated at 37.degree. C. for 2 hrs. 20 ul of 125 mM Tris-HCl, pH 6.8, 0.1% bromophenol blue in 50% aqueous glycerol were added. Products were separated by native gel electrophoresis at constant current on a 17% polyacrylamide gel by a modification of the method of Davies (Davies, 1964) in which the pH of all buffers was modified with H.sub.2 SO.sub.4. Products were visualized by staining with Coomassie Blue G-250 (see FIG. 1A).

IL-3 as Substrate

To 50 ul 20 mM Tris-HCl, pH 8.0, 40 ul rhIL-3 (2.5 ug/ul, Cangene) was added followed by 10 ul Tap. The assays were incubated at 37.degree. C.. 25 ul aliquots were withdrawn at the desired time points and frozen on crushed dry ice. The products were separated by isoelectric focusing from pH 3-10 using Pharmalyte 3-10 (Pharmacia) ampholytes (FIG. 4). Products were visualized by staining with Coomassie Blue G-250. Intact IL-3 had a pI=7.4. The -3 form demonstrates a pI=7.1.

APA-DNA as Substrate

The assay was conducted in a 96 well microtiter plate. To each well in the assay, 50 ul 100 mM Tris-HCl, pH 8.0, were added followed by 25 ul 3.2 mM APA-pNA. 25 ul of Tap were added to the wells and the absorbance was read at 405 nm. The assays were incubated at 37.degree. C. for 2 hours. The absorbance was read at 405 nm. The activity (release of p-nitroaniline) was calculated from the change in absorbance.

Example 3

Inactivation of Tap with PMSF: Assayed with IL-3

Tap stock (Example 1) was diluted 100-fold with 20 mM Tris-HCl, pH 8.0. A fresh solution of 80 mM PMSF was prepared in isopropanol (iPrOH). A Stock Buffer of 20 mM Tris-HCl, pH 8.0 was prepared. Four preincubations were prepared as follows.

iPrOH=58 ul Stock Buffer+2 ul iPrOH PMSF=58 ul Stock Buffer+2 ul PMSF/iPrOH Tap+iPrOH=18 ul Stock Buffer+40 ul Tapgw 2 ul iPrOH Tap+PMSF=18 ul Stock Buffer+40 ul Tap+2 ul PMSF/iPrOH These were incubated at 22.degree. C. for 30 minutes. When the preincubation was complete, 40 ul rhIL-3 (2.5 ug/ul, Cangene) were added and incubation was initiated at 37.degree. C.. Aliquots of 25 ul were removed at 0, 1, 2, and 4 hours. These aliquots were immediately frozen on dry ice. When the sampling process was complete, the products were analyzed by isoelectric focusing from pH 3-10 (Example 2).

Example 4

Amino Acid Sequencing of Tap

Tap was purified as described in Example I and was desalted by size exclusion chromatography. An Immobilon PVDF membrane (Millipore) was solvated according to the manufacturers instructions. Tap was adsorbed to the membrane by filtration employing a slot blot assembly. Protein bound to the membrane was visualized with Amido Black. The sample was excised and subjected to automated Edman degradation for 15 cycles.

Example 5

Synthesis of APACMK

21.3 g (70 mmol) Boc-Ala-Pro (Bachem Biosciences) dissolved in 175 ml anhydrous dimethylformamide (DMF) were activated by adding 7.8 ml (70.7 mmol) 4-methylmorpholine followed by 9.3 ml (70.7 mmol) isobutylchloroformate at -20.degree. C. with stirring. After 15 minutes, 15.1 g A-OBz in 175 ml anhydrous DMF were added. The solution was stirred for 1 hour at -20.degree. C. and then for 17 hours at ambient temperature. The DMF was remove by vacuum rotary evaporation. The residue was taken up in 175 ml ethyl acetate and extracted each with citric acid, saturated sodium bicarbonate, water, and brine. The organic layer was dried over anhydrous sodium sulfate for 1 hour. The sodium sulfate was remove by filtration.

2.5 g 5% pd on activated carbon were added and the suspension was agitated under a hydrogen atmosphere for 2 hours. At that time, the starting material had been completely converted to product. The hydrogenation catalyst was removed by filtration through Celite. The solvent was removed by vacuum rotary evaporation.

The resulting 23.7 g (66.3 mmol) of Boc-APA were dissolved in 140 ml anhydrous ethyl acetate and reacted with 7.8 ml (70 mmol) of 4-methylmorpholine followed by 9.2 ml (70 mmol) of isobutylchloroformate at -20.degree. C. with stirring. After 15 minutes, a solution of diazomethane in anhydrous ether prepared from 100 mmol N-methyl-N-nitroso-p-toluenesulfonamide (Aldrich) was added. After 1 hour at ambient temperature, the solution was extracted twice with 140 ml portions of water. The organic layer was dried over 2 g anhydrous sodium sulfate powder for 1 hour. The solution was removed by decantation. Deblocking of the N-terminal and generation of the chloromethylketone group were achieved simultaneously by adding 100 ml of HCl(g) saturated ethyl acetate. The resulting solution was allowed to stand at ambient temperature for 30 minutes. The product was removed from the organic solvent by extraction into 400 ml of water. The aqueous pool was frozen and lyophilized to yield the product, APACMK, as its hydrochloride salt.

Example 6

Inactivation of Tap by APACMK: Assayed with APA-pNA

A stock solution of 10 nM Tap in 100mM Tris-HCl, pH 8.0 was prepared. Serial dilutions of 210 uM, 21 uM, 2.1 uM, 210 nM, 21 nM, and 2.1 nM APACMK (Example 5) were prepared. To the microtiter well, 25 ul of Tap followed by 25 ul of an APACMK dilution or distilled water, for an uninhibited control, were added. The assays were incubated for 20 minutes at 22.degree. C. 50 ul 1.6 mM APA-pNA were added to each well. The absorbance was read at 405 nm then incubated at 37.degree. C. The change in absorbance at 405 nm was read after 15 and 60 minutes of incubation.

Example 7

Inactivation of Tap by APACMK: Assayed with GM-CSF

A stock solution of 10 nM Tap in 20 mM Tris-HCl, pH 8.0 was prepared. Serial dilutions of 210 uM, 21 uM, 2.1 uM, 210 nM, 21 nM, and 2.1 nM APACMK (Example 5) were prepared. To 20 ul Tap, 20 ul of an APACMK dilution (or water for an uninhibited enzyme control) were added and incubated at 22.degree. C. for 30 minutes. 10 ul of GM-CSF (1 ug/ul, Cangene) were added and incubated at 37.degree. C. for 2 hours. Products were analyzed by native gel electrophoresis as described in Example 2.

Example 8

Inactivation of Tap by APACMK--Determination of Kinetic Constants

A stock solution of 1.1 uM Tap in 50 mM Tris-HCl, pH 8.0 was prepared. APACMK stock solutions of 11 uM, 13.8 uM, 17.3 uM, 21.7 uM, 27.0 uM, 54.0 uM, 108 uM, and 1.08 mM were prepared. The Substrate Solution was 50 mM Tris-HCl, pH 8.0, 0.8 mM APA-pNA. The inactivation was performed by placing 90 ul of Tap (1 nanomole) in a 1.5 ml Eppendorf tube on ice and adding 10 ul of water (uninhibited control) or 10 ul of APACMK. A 10 ul aliquot was removed immediately and was assayed by adding it to a cuvette containing 390 ul Substrate Solution at 22.degree. C. The initial velocity was obtained from the change in absorbance at 405 nm during the first 10 seconds of the assay. Additional aliquots were removed at time points and assayed by the same method. At APACMK concentrations greater than 5.0 uM in the incubation, it was not possible to remove an aliquot from the incubation before 90% inactivation occurred.

Example 9

Application of APACMK in Fermentation

100 ml of media was inoculated in 500 ml baffle-bottom flasks with 100 ul of S. lividans 66 working seed bank material. The cultures were grown in a New Brunswick gyratory incubator at 32.degree. C. and 240 rpm. The cultures were sampled at 25, 27, 29, 31, and 48 hours post-inoculation and analyzed by native gel electrophoresis (see FIG. 7). Following removal of the 25 hour sample, 100 mM APACMK in sterile water were added to yield a final concentration of 10 uM. A control flask without APACMK was retained. The addition of APACMK significantly reduced formation of GM-CSF(-3) but did not inhibit cell growth.

Example 10

Construction and Screening of a S. lividans Genomic Library

A S. lividans 66 (Hopwood et al., 1983) genomic library was made using size fractionated (3-12 kbp) fragments of chromosomal DNA partially digested with Sau 3AI and ligated into the BamHI site of the bifunctional cloning vector, pSS12 (Butler et al., 1992). The ligated DNA was used to transform competent cells of E coli HB101 and pooled plasmid DNA was isolated from a mixture of approximately 30,000 transformed colonies grown in SOB medium (Maniatis et al., 1982) containing ampicillin (Sigma). This DNA was used for transformation of S. lividans 66 protoplasts yielding 15,000 transformant colonies resistant to thiostrepton (E. R. Squibb). Two days later the colonies were screened by overlaying with substrate mixture (containing 5 ml phosphate buffer (50 mM, pH 7.0), 25 .mu.l GPLbNA (20 mg.ml.sup.-1 in DMSO), 0.1 ml Fast Garnet GBC �10 mg.ml.sup.-1 in water!). The plates were incubated for three minutes at room temperature and washed three times with saline solution (Atlan et al., 1989, Alvarez et al., 1985). Positive colonies stained intensely orange against a background for pale orange colonies.

Two colonies reproducibly showed strong color. Plasmid DNA was isolated from each of these two colonies and the phenotype was retained when the DNA was used to transform protoplasts of S. lividans 66.

The plasmid DNA from each of these clones (P3-5 and P3-13) was investigated by restriction enzyme analysis. The data indicated that P3-5 and P3-13 were identical (presumably siblings) and the common restriction map is shown in FIG. 8A-8C. Southern hybridization analysis of chromosomal DNA, using the plasmid P3-13 as a probe (FIG. 9), suggested that the DNA contained in P3-13 had not been rearranged during cloning.

Example 11

Tap Activity of S. lividans 66 Strains Carrying the P3-5 and P3-13 Clones

The S. lividans 66 strains carrying the P3-5 and P3-13 clone or pSS12 were grown in TSB (containing 1% glucose, 0.1M MOPS and 20 .mu.g ml.sup.-1 thiostrepton). Aliquots (40 ml) of each culture were removed at 23 and 29 hours, and the supernatant and mycelium fractions were separated by centrifugation. Aliquots of the supernatant fractions were added to reactions (100 .mu.l) containing various tripeptide-bNA substrates(8 nmol) in microtiter wells. After incubation at 37.degree. C. for 4 hours, a solution (50 .mu.l) containing Fast Garnet GBC dye was added and the A.sub.540 was measured in a microtiter plate reader. The results are shown in Table IV.

TABLE IV______________________________________Tripeptidyl Aminopeptidase Activity(A.sub.540 above background)SampleSupernatants GPL-bNA GPM-bNA APF-bNA D-FPR-bNA______________________________________P3-5/23 HRS Max Max Max 0.02P3-5/29 HRS Max Max Max 0.08SS12/23 HRS 0.19 0.28 0.63 0.02SS12/29 HRS 1.38 2.46 Max 0.17______________________________________ ("Max" indicates a A.sub.540 reading of >3.0)

At as early as 23 hours of culture, a 1 .mu.l aliquot of the supernatant from S. lividans carrying the P3-5 clone was showing strong activity against the GPL-, GPM- and APF-bNA substrates. At the same time point, a 25- .mu.l aliquot of the control culture had at least 15 to 20 fold lower activity with the same substrates. However, against the D-FPR- and APF-bNA substrates, the Tap overproducer had little activity over the control. An aliquot (1 .mu.l) of each supernatant (which was harvested after 23 hours of growth) was added to a reaction containing 4 .mu.g of purified intact GM-CSF. Following a 2.5-min. incubation at 37.degree. C., the proteins were analyzed by native PAGE and stained with Coomassie Brilliant Blue. The full-length GM-CSF (lane 1 of FIG. 11) was rapidly converted to the -3 form upon incubation with culture supernatants from cells carrying the tap clones. By contrast, no significant degradation was observed when GM-CSF was incubated with the control culture due to the small volumes of culture supernatant and short time of incubation used compared to those described in Example 2.

Example 12

Analysis of Extracellular Proteins From S. lividans 66 Strains Carrying the p3-5 and p3-13 Clones

The S. lividans 66 carrying the P3-5 and P3-13 clones were grown in liquid culture, and supernatant fractions were collected following the teaching of Example 11. As described by Laemmli (1970), samples were prepared from aliquots (200 .mu.l) of the supernatant fractions, and SDS-10% polyacrylamide gels were run at 100 v for 5 to 6 hours. The profile of separated proteins was then visualized by staining with Coomassie Brilliant Blue (FIG. 10). An abundant protein with an apparent molecular weight of 55,000 daltons was present among the extracellular proteins from S. lividans 66 carrying either P3-5 (lanes 2 to 7) or P3-13 (lanes 8 to 13). From 23 to 29 h of culture, the level of Tap increased to approximately 0.1 mg/ml, relative to the BSA standards (lanes 14 to 19). Lanes 1 and 20 show molecular weight markers.

Example 13

Amino Terminal Amino Acid Sequence of the Tap Protein Purified From an S. lividans 66 Strain Carrying the P3-13 Clone

The S. lividans 66 strain carrying the P3-13 clone was grown in liquid culture and supernatant fractions were collected, following Example 11. The extracellular proteins were separated by SDS-PAGE, following the teaching of Example 12, and transferred onto Immobilon PVDF (Millipore) membranes as directed by the supplier. After briefly staining the filters with Coomassie Brilliant Blue, the bands containing the major protein (apparent molecular weight 55,000 daltons) were excised from the filter, and subjected to automated Edman degradation analysis. The N-terminal amino acid sequence (SEQ ID NO 10) determined was: Asp-Gly-His-Gly-His-Arg-Ser-Gln (or Ser)-Asp-Ala.

Example 14

Production of Mutant Strains of S. lividans Defective in Protease Activities Using Chemical Mutagenesis

S. lividans 66 spores were treated with N-methyl-N-nitro-N-nitrosoguanidine (MNNG) according to the method of Hopwood et al., (1985). Briefly, a suspension containing 2.5.times.10.sup.12 spores in 3 mls of Tris/maleic acid buffer was incubated at 30.degree. C. in a preweighed vial containing 10 mgs of MNNG (which had been solubilized in 0.5 ml DMSO immediately prior to the addition of the spore suspension). 1 ml aliquots were removed from the mixture at 30 minute intervals and washed twice by centrifugation to remove the MNNG. Serial dilutions of the treated spores were plated on agar medium to determine the effectiveness of the mutagenic treatment in terms of the proportion of viable surviving colony forming units remaining compared to untreated spores. Survival rates of 2.8.times.10.sup.-3 %, 1.2.times.10.sup.-4 % and 9.times.10.sup.-6 % were observed after 30, 60 and 90 minutes, respectively.

Two hundred surviving colonies from each of the three treatment times were purified and examined for their ability to grow on minimal media. Colonies which were unable to grow were classified as auxotrophic mutants of which 1, 4 and 2 were observed at the 30, 60 and 90 minute treatment times, respectively.

Spores from the 60 minute treatment were, therefore, examined for the presence of strains carrying mutations which inactivated specific proteolytic phenotypes. A direct agar plate screening technique was used in which the colonies were overlayed with substrate mixture (containing 0.1 ml of GPL-bNA (Bachem Inc., 1 mg dissolved in DMSO), 0.1 ml Fast Garnet GBC (Sigma) dye (10 mg.ml.sup.-1 in water), 6 ml of 50 mM phosphate buffer, pH 7.0 and 0.2 ml DMSO. The plates were incubated for twenty minutes at room temperature and washed three times with saline solution (Atlan et al., 1989, Alvarez et al., 1985).

Screening of 2,700 colonies using GPL-bNA revealed two colonies which did not turn red. Testing supernatants from liquid cultures of one of these colonies (12-5 or 12-8), with various chromogenic tripeptide substrates (Table V), confirmed that this specific hydrolytic ability had been either eliminated or at least very substantially reduced compared to the original untreated S. lividans strain.

TABLE V______________________________________Tripeptidyl Aminopeptidase Activity(A.sub.540 above background)SampleSupernatants GPL-bNA GPM-bNA APF-bNA D-FPR-bNA______________________________________12-5/T2 0.01 0.01 0 012-5/T4 0.10 0.10 0.05 0.0612-8/T2 0.02 0.02 0 012-8/T4 0.13 0.12 0.12 0.081-5/T2 0.01 0.01 0.01 0.021-5/T4 2.55 Max Max 0.09______________________________________ ("Max" indicates a A.sub.540 reading of >3.0)

In a similar experiment to that described above, a L-bNA substrate was used, resulting in the isolation of one mutant (lap.sup.-) strain (1-5) from 1500 colonies screened. By comparison, the Tap activity of this mutant strain was unchanged from that of wild type S. lividans 66.

Aliquots of each culture supernatant were added to reactions containing 2.5 .mu.g GM-CSF and incubated at 32.degree. C. for 2 minutes. The proteins were separated by SDS-PAGE and visualized by Western blotting, using an antiserum raised against the amino terminal 35 amino acids of GM-CSF. At 40 h (T3), the cultures from the tap mutants, #11 and #12 contained less activity for converting GM-CSF to the -3 from than those from the S. lividans, MS2 and the lap mutant, #1.

Protoplasts were prepared from the various S. lividans 66 mutants, and were transformed using the GM-CSF expression vector pAPO.GMCSF (as described in Canadian patent number 1,295,567 and U.S. Pat. No. 5,200,327). The transformed cells were grown in liquid culture and the supernatant fractions were collected following the teaching of Example 11. Aliquots of each culture supernatant were analyzed by SDS-PAGE. The transformants with the tap mutants, 12-5 and 12-8 generally showed more intact GM-CSF at later time points in the culture than the S. lividans, MS2. However, the formation of the -3 form of GM-CSF was not completely eliminated with the tap mutants.

Example 15

Construction of A Deletion Subclone From the tap Clone

Specific deletions were made in the tap clone to localize the gene and enable chromosomal disruption. A 1.2 -kbp DNA fragment was removed between BamHI (1100) and BglII (2300) (see FIG. 8B) to construct the deletion clone .DELTA.1. P3-5 DNA was digested by means of EcoRI and BglII, and the vector fragment was isolated; and P3-5 was digested with EcoRI and BamHI and the 1.1 -kbp insert fragment was isolated. The vector and insert fragments were ligated, using T4 DNA ligase, and used to transform E.coli. The plasmids were screened by restriction analysis and the correct plasmid, .DELTA.1, used to transform protoplasts of S. lividans 66. The S. lividans 66 carrying the .DELTA.1 deletion clone was screened with a plate assay using GPL-bNA. A transformant was grown in liquid culture, and the level of Tap activity was determined in a liquid assay using tripeptide-.beta.napthylamide substrates. The S. lividans 66 carrying the .DELTA.1 deletion subclone had a similar Tap activity to that of the untransformed host strain.

Deletion clone .DELTA.2 was constructed by subcloning the EcoRI-BglII fragment into the vector pSS12 which had previously been digested with EcoRI and BamHI. .DELTA.3 was made by digestion of P3-5 DNA with BglII, followed by relegation, resulting in the loss of the 300 nt BglII fragment around the centre of the tap gene. The high level of Tap activity associated with the P3-5 plasmid was not observed with .DELTA.2 or .DELTA.3, confirming that the deletions resulted in loss of enzyme activity.

Example 16

Deletion Clones Used for Integrational Mutation of tap into the S. lividans 66 Chromosome

Subcloning of the DNA insert sequences from the deletion clones was not straightforward due to the presence of multiple BamHI sites. A partial BamHI digestion of P3-5 DNA was followed by a complete EcoRI digestion. The 3.1 kbp tap-encoding fragment was isolated from an agarose gel and subcloned into the E.coli vector pT7T3 which had previously been digested with BamHI and EcoRI. Appropriate transformants were identified and the DNA insert was used to create further subclones in the pINT vector as follows. .DELTA.1int was produced by a three way ligation of the EcoRI-BamHI, BglII-HindIII (in the polylinker of the pT7T3 vector) fragments from the pT7T3 subclone and the EcoRI-HindIII fragment produced by digestion of pINT. .DELTA.2int was the result of a direct subcloning of the EcoRI-BglII fragment from the pT7T3 subclone into pINT digested with EcoRI and BamHI. A3int involved the BglII-HindIII fragment from the pT7T3 subclone and BamHI plus HindIII digested pINT. .DELTA.4int was a direct subcloning of the whole inserted fragment in the pT7T3 subclone (EcoRI+HindIII) into the same sites in pINT. .DELTA.5int was made from A4int by digestion with BglII and relegation. The DNA contained within the various .DELTA.int clones is shown in FIG. 8C.

Plasmid DNA was isolated from the E.coli transformed strains and used to transform protoplasts of S. lividans MS5 (a strain derived from S. lividans 66 by deletion of DNA fragments at the slpA and slpC (Butler el al., 1992) loci; in addition the pepP gene (Butler et al., 1993) and a second PepP-encoding gene (Butler et al., J. Ind. Microbiol., in the press) were also subjected to specific chromosomal DNA deletion events, each of which reduced the PepP activity of the S. lividans strains). Integrative transformants resistant to thiostrepton were purified and allowed to grow in the absence of thiostrepton to allow recombinational resolution to occur. Strains which had undergone excision events were easily identified by screening for the loss of the ability to hydrolyse GPL-bNA. The results obtained were somewhat unexpected. .DELTA.1int did not produce any integrative thiostrepton-resistant transformants in three independent experiments. .DELTA.2int did lead to integrative transformants, indicating that there was no practical impediment to recombination events at this locus on the S. lividans chromosome. .DELTA.3int failed to produce integrative transformants, possibly due to the relatively small length of DNA (900 nt) available for homologous recombination to occur. .DELTA.4int yielded tranformants as did .DELTA.5int. Subsequent experiments using .DELTA.1int were successful using S. lividans 66 protoplasts (to make a strain designated MS9 which was defective only at the tap locus) suggesting that the earlier failure in the MS5 experiment was due to the lower transformation capability of that particular batch of MS5 protoplasts.

Integrative transformants from .DELTA.5int were grown in the absence of the thiostrepton selection on agar medium. After sporulation had occurred the spores were harvested and replated onto fresh agar plates. Colonies were screened using the .beta.naphthylamide substrate assay for tap activity. The frequency of excision events which led to loss of the activity was very low (approximately 1 in 1000). Three colonies were obtained with reduced Tap activity. Chromosomal DNA was isolated and Southern hybridization analysis (FIG. 9) confirmed that one colony (#2) had lost the 300 nt BglII fragment (lanes 3 and 7 compared to the S. lividans 66 control lanes 2 and 6). Similar experiments with a 3.3 kbp DNA probe revealed a complex hybridizing band pattern in colony 1 chromosomal DNA whereas colony 2 DNA showed only the expected bands with a reduction in size of one band consistent with the desired specific chromosomal deletion. Colony 2 was designated Streptomyces lividans MS7. Another strain was constructed using .DELTA.5int and S. lividans 66 protoplasts. This strain was designated MS8 and shown to have properties indistinguishable from those of MS9.

Example 17

The S. lividans MS7 Strain Shows a Substantial Reduction in its Ability to Hydrolyse Tripeptide bNA Substrates and GM-CSF in vitro

The S. lividans MS7 strain was grown in liquid culture (TSB medium) and supernatants collected by centrifugation to remove the mycelial material. Aliquots (50 .mu.l) of the supernatants were added to each of the tripeptide substrates (8 nmol) in a final volume of 100 .mu.l. After incubation at 37.degree. C. for 45 minutes, 50 .mu.l of a solution of Fast Garnet GBC dye was added and the A.sub.540 measured using a microtiter plate reader.

FIG. 14 shows the activity of S. lividans MS5 (tap+) and MS7 (tap) strains against chromogenic tripeptide substrates. Cell-free broth from the strains was isolated at various times of fermentation (without thiostrepton) and incubated with either APA-bNA or GPL-bNA.

The symbols represent the following combinations:

MS7+APAbNA (-.quadrature.-) MS7+GPLbNA (-*-) MS5+APAbNA (-.DELTA.-) MS5+APLbNA (-O-)

The results are summarized in FIG. 14 and indicate that under these assay conditions, the supernatants derived from the MS7 culture were (within experimental error) devoid of any significant hydrolytic ability against these substrates, whereas the supernatant derived from S. lividans MS5 showed the ability to rapidly degrade both substrates.

FIG. 15 shows the degradation of full-length GM-CSF by cell-free broth from S. lividans MS5 and MS7. Cell-free broth was isolated from cultures grown without thiostrepton for 25 hours. Degradation was significantly slower for MS7 than MS5.

When the same supernatant samples were analyzed for the ability to degrade GM-CSF in vitro (according to the teaching of Example 15), it was clear that the rate of degradation of GM-CSF for the MS7 samples (FIG. 15, lanes 4-6) was much slower than for the MS5 samples (FIG. 15, lanes 1-3).

Example 18

Production of Undegraded GM-CSF by the S. lividans MS7 Strain

The GM-CSF expression plasmid vector pAPO.GMCSF was used to transform protoplasts of the S. lividans MS7 strain. Following the teaching of Example 11, liquid cultures were prepared from the transformed strain as well as transformants from the S. lividans MS5 strain.

FIG. 16 illustrates production of GM-CSF by S. lividans 66 and the deletion mutant strain MS7. Cell-free broth from the strains was harvested after fermentation for the times shown and analyzed by native PAGE.

Native PAGE analysis of the culture supernatants revealed that while degradation of the secreted GM-CSF occurred in both strains, it was only evident in the MS7 supernatant material (FIG. 16, lanes 5-8) at later times of growth compared to the MS5 samples (FIG. 16, lanes 1-4). This property of the new S. lividans MS7 strain allowed it to be used to produce a higher yield of undegraded GM-CSF than was possible using the wild-type S. lividans 66 strain.

Example 19

Tap Activity is Present in a Wide Variety of Streptomyces Species

Genomic DNA was isolated from the following Streptomyces strains. S. alboniger 504 (P. Redshaw, Austin College, Texas, USA) , (S. coelicolor M130 (John Innes Institute), S.fradiae ATCC 14544, S. griseus IMRU 3499, S.griseus ATCC 10137, S.parvulus 2283 (John Innes Institute) S.rimosus ATCC 10970. 10 .mu.g of each DNA were digested in 100 .mu.l of appropriate buffer for the restriction enzymes BamHI and PstI respectively. 30 units of each enzyme were added together with 1 .mu.l of RNAse A (10 mg/ml, Sigma). The reactions were incubated at 37.degree. C. for 3 hours. A further 15 units of enzyme were added and the samples incubated overnight at 37.degree. C. Digestions were terminated by the addition of 11 .mu.l of stop buffer (Orange G, 0.08%; glycerol, 50%; EDTA, 67 mM; pH8). Approximately 3 .mu.g of each digested DNA sample were loaded onto a 1% agarose horizontal gel and electrophoresed at 100V for 4 hours. A molecular weight marker was included (Lambda DNA digested with HindIII, Bethesda Research Laboratories to calibrate the gel. After electrophoresis the gel was soaked in 0.25M HCl, followed by 0.5M NaOH, 1.5M NaCl and rinsed in water. The DNA was transferred to a Nylon membrane (Boehringer Mannheim) using a Vacublot (Pharmacia) apparatus with 20.times.SSC buffer for 1 hour at 50 mbars pressure. After transfer the membrane was washed in 2.times.SSC and baked for 1.5 hours at 80.degree. C.

The DNA insert fragment from the EcoRI site to the right-most BamHI site was isolated by partial BamHI and complete EcoRI digestions of the P3-13 DNA. The fragment was subcloned into the E.coli plasmid vector pT7T3 (Pharmacia). From this clone it was possible to isolate larger quantities of the same DNA fragment by digestion with EcoRI and HindIII. 0.5 .mu.g of this 3.3 kbp- fragment were labelled according to the manufacturers's recommendations (Boehringer Mannheim) to produce a digoxigenin-labelled probe. 25 ng of probe were used per ml of hybridization solution. Lambda DNA was labelled in the same way to allow visualization of the molecular weight marker fragments. Hybridization was carried out at 68.degree. C. overnight using 2.5 ml of hybridization solution per 100 cm.sup.2 of nylon membrane. The hybridization solution contained; 5.times.SSC; blocking reagent, 1% (w/v); N-lauroylsarcosine, 0.1% (w/v); sodium dodecyl sulphate, 0.02% (w/v). Filters were prehybridized for 1 hour at 68.degree. C. Probes were boiled for 10 minutes, quick chilled on an ice/NaCl bath, diluted with 100 .mu.l hybridization solution and added to the prehybridized membrane in a stoppered glass bottle. Hybridization and prehybrization were carried out using a Hybaid mini-hybridization oven. Membranes were washed twice at 68.degree. C. for 30 minutes in 5.times.SSC, 0.1% SDS (50 ml/100 cm.sup.2 membrane). The membranes were then transferred to plastic containers and processed according to the manufacturer's instructions.

Finally, membranes were transferred to plastic bags, sealed and incubated at 37.degree. C. for 30 minutes. Membranes were then exposed to X-ray film for 10 minutes. After development of the X-ray film the autoradiogram shown in FIG. 17 was obtained.

The autoradiogram showed hybridizing bands in all lanes except those containing S. fradiae DNA. Lanes 1 and 18 contained Lambda /HindIII molecular weight markers. In FIG. 17, lanes 2 and 10, S. alboniger; lanes 3 and 11, S. coelicolor; lanes 4 and 12, S. fradiae; lanes 5 and 13, S. griseus IMRU 3499; lanes 6 and 14, S. griseus ATCC 10137; lanes 7 and 15, S. lividans 66; lanes 8 and 16, S. parvulus; lanes 9 and 17, S. rimosus.

Identical hybridizing bands were observed with S. lividans and S. coelicolor with a common band in both S. griseus strains as well as the S. parvulus DNA. S. rimosus and S. alboniger produced hybridizing bands at different molecular weights suggesting restriction fragment length differences in these species. No strong band was observed for the S. fradiae DNA. Taken overall the results suggested that the Tap-encoding DNA sequence occurs widely throughout the Streptomyces strains examined.

In a similar experiment using S. ambofaciens ATCC 23877 DNA, strongly hybridizing bands were observed after digestion with BamHI, PstI, SacI, and SalI. This indicated the likely presence of a tap gene in S. ambofaciens which would be expected to be detrimental to product yield when expression of secreted proteins is desired in this strain.

The following examples relate to proteases, other than Tap, derived from Streptomyces, their DNA sequences and amino acid sequences. These proteases degrade certain substrates under certain conditions. Example 20 describes one such protease, which displayed a significant amino acid sequence homology with the Bacillus subtilis protease BPN' (using the BLAST program �Altschul et al! to screen the protein sequence databases) and was therefore designated Ssp (Subtilisin-like-protein). An improved strain of Streptomyces in which this protease is impaired, was created. Southern blot hybridization indicated that Ssp is present in many Streptomyces species. Three other proteases, the DNA sequences and deduced amino acid sequences for two of them, are described in Examples 21, 23 and the n-terminal amino acid sequence of the third protease is indicated in Example 22.

Example 20

Characterization of P5-4 and P5-15.

Following the teaching of Example 10, the S. lividans 66 genomic library was used to transform protoplasts of the MS7 mutant strain. Transformant colonies were screened with the substrate APA-bNA. Among the thirteen thousand colonies screened, two clones were isolated by virtue of the plasmid-encoded phenotype (colonies appeared red against a background of pale colonies). Plasmid DNA was isolated from these colonies and used to transform E.coli competent cells from which larger quantities of plasmid DNA were isolated.

Restriction enzyme site mapping established that two clones (designated P5-4 and P5-15) were shown to represent overlapping fragments of S. lividans chromosomal DNA containing the Ssp-encoding gene. FIG. 18 shows the restriction enzyme sites present in the P5-4 and P5-15 DNA. K=KpnI, B=BamHI, M=MluI. The hydrolytic capabilities of strains containing the cloned DNA (or deletions thereof) was measured visually using the agar plate assay method. Southern hybridization against chromosomal DNA showed the expected pattern of hybridizing bands indicating that no major DNA rearrangements had occurred during the isolation of these clones.

Following the teaching of Example 15 the region of DNA encoding the proteolytic activity was defined within the deletion clones P5-4-1 and P5-4-3 (FIG. 18). Specifically, the larger of the two Ncol fragments deleted in P5-4-2, P5-4-4 and P5-4-5 appears to be correlated with the proteolytic activity.

FIG. 19 shows SDS-PAGE analysis of protein secreted by strains carrying the P5-4 DNA (Lanes 1 and 2) or the P5-4-4 deletion clone (Lanes 3 and 4). Lanes 1 and 3 contained 30 .mu.l of cell-free broth. Lanes 2 and 4 contained approximately 2 .mu.g protein derived from the cell-free broth samples by ammonium sulphate precipitation. The positions of molecular weight marker are shown by arrows. A major protein species was observed at a position consistent with a molecular weight of approximately 45,000. Preparative SDS-PAGE followed by electrotransfer to PVDF membrane (as described in Example 13) allowed direct automated Edman degradation to be carried out to yield the amino acid sequence (residues 1-30 of SEQ ID NO:4) NH.sub.2 -Asp-Thr-Gly-Ala-Pro.sup.5 -Gln-Val-Leu-Gly-Gly-.sup.10 -Glu-Asp-Leu-Ala-Ala-.sup.15 -Ala-Lys-Ala-Ala-Ser.sup.20 -Ala-Lys-Ala-Glu-Gly.sup.25 -Gln-Asp-Pro-Leu-Glu.sup.30.

DNA sequence analysis (shown in FIG. 20A-20C (SEQ ID NOS 3 and 4)) of the P5-4 DNA revealed a potential protein coding region located within the region of DNA defined by the two Ncol fragments in FIG. 18. This was consistent with the respective activities of the plasmid deletion clones P5-4-1, P5-4-2, P5-4-3, P5-4-4 and P5-4-5 . Inspection of the predicted protein sequence reveals the exactly matching, experimentally determined amino terminal amino acid sequence noted above. Furthermore, the predicted amino acid sequence also shows a putative signal sequence at the amino terminus, followed by a putative pro region defined by the amino terminal end of the experimentally determined mature amino acid sequence.

FIG. 21 shows a comparison of the amino acid sequence of the proteins predicted from the P5-4 DNA sequence (SEQ ID NO:4) with that of the Bacillus protein subtilisin BPN' (SEQ ID NO:12). 1 designates the S. lividans sequence while 2 designates the Bacillus sequence. GM-CSF degradation assays according to methods used in Example 2 using cell-free broth from S. lividans MS7 strain carrying the P5-4 plasmid DNA culture in TSB medium demonstrated that the overproduced Ssp also degraded GM-CSF. In FIG. 30, lane 1 shows such GM-CSF degradation by a P5-4-containing MS7 strain; lane 3 shows a similar result with a P5-15-containing MS7 strain. In contrast, lane 2 shows broth from a P5-10 culture which shows only slight degradation to the "-3 form". The same results were obtained with samples from cultures carrying only the PSS12 plasmid.

Deletion of the Ssp-encoding DNA from the S. lividans chromosome was accomplished following the teaching of Example 16. Specifically, the DNA from plasmid deletion clone P5-4-4 (FIG. 18) was subcloned into pT7T3 using the EcoRI site immediately adjacent to the leftward side of the DNA insert (shown in FIG. 18). Since there was no convenient restriction enzyme site to the rightward side of the DNA insert this was excised using the XhoI site (in the replication origin of the plasmid vector, pSS12) which was subsequently ligated to the SalI-digested pT7T3. Hence, overall the EcoRI-XhoI fragment was inserted in EcoRI and SalI digested T7T3 DNA. The fragment was subsequently excised by digestion with EcoRI and HindIII and inserted into the integration vector, pINT using the same restriction enzyme sites. The pT7T3 intermediate step was required because the SalI site in the multiple cloning site of pINT was not unique and, therefore, not convenient for subcloning purposes.

This integration clone was used to create strains containing the specific deletion at the ssp locus in two S. lividans host strains. Firstly, the MS7 host strain was used to create a new strain designated MS11 (pepP.sup.-, pepP2.sup.-, slpA.sup.-, slpC.sup.-, tap.sup.-, ssp.sup.-). Secondly, another tap-deleted strain (MS9) was used to create MS12 (tap.sup.-, ssp.sup.-). The deletion strains MS7, 9, 11 and 12 were cultured in TSB/PPG liquid medium for 22 hours and examined for the ability of cell-free broth to hydrolyse APA-PNA.

FIG. 22 shows the activity of cell-free broth samples derived from S. lividans 66 (-O-) , MS7 (-.DELTA.-), MS9 (-.quadrature.-), MS11 (-*-) and MS12 (-+-) strains against the APA-bNA substrate according to the teaching of Example 2.

The results (FIG. 22) showed a reduction in hydrolytic capability with the MS12 strain showing the lowest activity. All the strains displayed a significantly reduced hydrolytic capability compared to S. lividans 66 but the MS9 strain showed a lower level than the MS7 strain. (This was shown in a separate experiment not to be due to the different integration clones used, since MS8 used the same integration clone as MS7 but was derived from S. lividans 66 protoplasts and showed indistinguishable properties to MS9).

Southern hybridization experiments detected DNA sequences homologous to the ssp DNA in many Streptomyces species. FIG. 23 shows a Southern blot hybridization experiment using the 2.25 kb BamHI-KpnI DNA fragment which had been subcloned into pT7T3.18.mu.. Lanes 1 and 18 are lambda/HindIII molecular weight markers. Lanes 2 to 9 represent chromosomal DNA digested with NcoI while lanes 10 to 17 show DNA digested with SphI. Lanes 2 and 10, S. alboniger; Lanes 3 and 11, S. ambofaciens; Lanes 4 and 12, S. coelicolor; Lanes 5 and 13, S. fradiae; Lanes 6 and 14, S. griseus; Lanes 7 and 15, S. lividans 66; Lanes 8 and 16, S. parvulus; Lanes 9 and 17, S. rimosus.

It should be noted that the same library of clones was screened as in Example 10. Presumably, the lower background level of APA-bNA-hydrolysing activity in MS7 (compared to S. lividans) allowed the P5-4 and P5-15 clones to be identified. This has been noticed by other workers particularly relating to neutral protease activities in B.subtilis (Sloma et al., 1990).

Example 21

A Protease Encoding Gene, P5-6 and a Predicted Protein

Following the teaching of Example 21 yet another protease-encoding gene was isolated from the same library screening experiment. Two clones were identified as being different (in terms of restriction enzyme sites) from the tap or ssp clones described in this application. Clone numbers P5-6 and P5-17 were shown to represent overlapping fragments of chromosomal DNA (FIG. 24).

FIG. 24 shows the common restriction enzyme site map of the P5-6 and P5-17 DNA and deletion clones derived from P5-17.

Activity against APA-bNA is shown by the number of asterisks adjacent to each plasmid and was estimated using the agar plate assay method described in Example 10.

Although these clones encoded significant hydrolytic capability against the APA-bNA substrate in the agar plate assay, no activity above background was observed in cell-free broth derived from cultures containing these plasmids grown in TSB media. Neither was it possible to experimentally identify the protein product of this locus. When cultured in liquid medium resembling the agar medium composition (i.e. R2 without added phosphate or agar and containing 0.25% yeast extract--instead of the usual 0.5%) APA-bNA -degrading activity was observed in the cell-free broth. However, in contrast to the Tap and Ssp proteins, this activity was unable to hydrolyse GPL-bNA in R2, although it did show degradation of full-length GM-CSF according to the methods described in Example 2 (FIG. 31, lanes 3 and 7).

DNA sequence analysis of the P5-6 DNA (FIG. 25 SEQ ID NOS 7 and 8) revealed a potential coding region. The predicted protein once again displayed a putative secretion signal peptide, followed by a predicted protein of 492 amino acid residues (FIGS. 25A-25C). Furthermore, when the amino acid sequence was compared to that of the Tap (FIG. 26 SEQ ID NOS 8 and 2) a strong homology was obvious around the region encoding the putative active site serine residue.

Plasmid deletion clones were constructed from P5-17 and shown to encode no activity above background in the agar plate assay.

Example 22

Characterization of P5-10.

Another cloned DNA fragment was isolated from the same APA-bNA screening experiment described in Examples 20 and 21. This DNA species was designated P5-10 and showed a different pattern of characteristic restriction enzyme sites (FIG. 27) than those observed for the other clones described above. A significant protein band was observed by SDS-PAGE analysis of supernatants of strains carrying this plasmid. Its molecular weight is approximately 50,000 daltons. Amino terminal amino acid sequence analysis was carried out according to the teaching of Example 13 yielding the following sequence (SEQ ID NO:13): Ala-Glu-Pro-Xaa-Ala.sup.5 -Val-Asp-Ile-Asp-Arg.sup.10 -Leu. The activity of supernatant material containing this protein from MS7 host cultures, grown in TSB medium, was very low against APA-bNA and GPL-bNA. However, when cultured in R2YE liquid medium a high level of activity was observed against APA-bNA but not GPL-bNA. Furthermore, degradation of full-length GM-CSF according to the methods described in Example 2, was also detectable in samples grown in R2YE but not TSB (FIG. 31, lane 5).

Example 23

Characterized of P8-1, 2 and 3.

A chromogenic substrate was designed to model the amino terminal region of GM-CSF except that the amino terminal residue was modified by the addition of a Boc-group (or other similar moieties such as Fmoc), such that proteases whose activity requires a free NH.sub.2 -group would be unable to act directly on this substrate. However, any endoprotease present in the S. lividans host having a recognition sequence compatible with that of the substrate (specifically Boc-APARSPA-bNA) would be able to cleave and remove the Boc-group in addition to some portion of the peptide. Such cleavage would generate a smaller peptide-linked bNA moiety which o now contains a free NH.sub.2 -group at the N-terminus and can be acted upon to release the chromogenic bNA moiety which can subsequently be visualized by reaction with Fast Garnet GBC dye.

This strategy was used to screen the S. lividans 66 genomic DNA library after transformation into the MS5 host strain (tap+). After screening of eight thousand colonies, six clones were confirmed to encode the ability to degrade the substrate significantly faster than the host strain alone. Two clones proved on restriction enzyme site analysis to be identical to P5-6 described in Example 21. Another clone was similarly shown to be the same as P5-17. Three other clones (P8-1, 2 and 3) were isolated and shown to represent the same region of chromosomal DNA (by Southern hybridization experiments). P8-3 contained a larger DNA fragment which was probably derived from the co-cloning of noncontiguous Sau3AI fragments in the construction of the library. P8-1 contained an inserted DNA fragment of approximately 8 kbp, while P8-2 had a smaller insert (3.6 kbp). Deletion mapping and DNA sequence analysis revealed a potential protein coding region in the central part of the cloned DNA (FIG. 28.) Comparison of the predicted protein sequence derived from the DNA sequence (FIG. 29) with those encoded by the tap and P5-6 clones showed a significant homology between the proteins encoded by P8-2 and P5-6. A smaller but still significant homology was detectable with the Tap protein. Specifically of interest is the conservation of amino acid sequences around the putative active site serine residues of these proteins as follows:

__________________________________________________________________________Tap (residues 204-225 of SEQ ID NO:2) - G V S Y G T Y L G A V Y G T L F P D H V R RP5-6 (residues 199-220 of SEQ ID NO:8) - G A S Y G T F L G A T Y A G L F P D R T G RP8-2 (residues 231-252 of SEQ ID NO:6) - G I S Y G T E L G G V Y A H L F P E H V G__________________________________________________________________________ R

Example 24

An Immunoassay Using Tap

Tap as a unique protease with a well established assay using a synthetic substrate for determination of its activity (described in this patent application) may be applied as a useful tool for immunoassay.

The uses of high performance immunoassay have increased greatly in the last decade, extending to almost every discipline in the life sciences. In the majority of applications, antibodies are labelled with enzymes, biotin or fluorochromes, and serve as components of a signal generating/amplifying system. This technology has a broad applicability and can be used in a wide variety of laboratory techniques including enzyme-linked immunosorbent-assay (ELISA), immunoblotting, immunohisto/cytochemistry and immuno-electrophoresis. In the following example we will show how one can use Tap in the most widely used technique--microwell ELISA.

In microwell ELISA, antigens are immobilized in a microwell and probed by labelled antibody (conjugate). The enzyme-labelled reagents are detected with the appropriate substrate, which is converted to a visible colored product at the reaction site. The intensity of color produced is proportional to the amount of measured antigen.

To date, the most common enzymes used for generating color are alkaline phosphatase or horseradish peroxidase. In this example, those enzymes are replaced with Tap and using the synthetic substrate, developed and described in this patent application, such as APA-PNA for visible color and APA-AMC for fluorescence technology detection.

To demonstrate this idea, IL-3 was used as an example for antigen quantitation. Rabbit anti-IL-3 antisera (Cangene Corporation, Canada) was used as the first antibody. The second antibody, goat anti-rabbit IgG linked to biotin (Sigma, St. Louis, U.S.A.), and streptoavidin (Boehringer Mannheim GmbH) were used as the amplification system. Tap linked to biotin was used as the enzyme. The Tap was purified as described in Example 1 and 9.0 mL of the Tap (approximately 0.3 mg/mL) were biotinylated with D-Biotinyl-E-aminocaproic acid N-hydroxysuccinimide ester as described in Biochemia Bulletin of Boehringer Mannheim (1989, Antibodies and Reagents for Immunochemistry, p.115). Serial dilutions of recombinant hIL-3 (Cangene Corporation, Canada) were applied to the microplate wells (100 .mu.L/well), and then incubated at 4.degree. C. for over 16 hours. The wells were then washed and 5% BSA (bovine serum albumin) was added as a blocker. After 1 hour incubation, the wells were washed and rabbit anti hIL-3 sera (Cangene Corporation, Canada) was added at a dilution of 1/2000. Incubation was performed at 37.degree. C. for 1 hour. The wells were then washed and the second antibody, goat anti-rabbit IgG-Biotin (Sigma), was added at a dilution of 1/2000 for 1 hour at 37.degree. C. After washing, a mixture of Streptoavidin and Biotin-Tap was added. This mixture was prepared previously as follows: 40 .mu.L of Streptoavidin (Boehringer Mannheim, 1 mg/mL) and 35 .mu.L of Biotin-Tap were added to 5 mL Tris buffer pH 8.0 containing 1% BSA. The mixture was pre-incubated for 45 minutes before being added to the microplate assay. The mixture was washed from the microplate after incubation for 45 minutes at room temperature. Then 100 .mu.L of the enzyme substrate (0.8 mM) were added. For color developing, APA-pNA was used as a substrate and the assay was read after 2 and 16 hours incubation by absorbance at 405 nm. For faster analysis, APA-AMC was used as a fluorescent substrate, where the incubation was performed for 30 minutes and the assay was analyzed at exitation/emission of 400/450 nm by the multiwell plate scanning fluorescent system.

FIG. 32 shows a hIL-3 calibration curve using ELISA technology with Tap as the enzyme and APA-pNA as the substrate for color forming (Panel A) incubated for either 2 hours (O- - -O) or 16 hours (.DELTA.- - -.DELTA.), and APA-AMC as a fluorescent substrate (Panel B) incubated 30 minutes.

There are some advantages to using Tap in the ELISA system compared to the common enzymes. The substrates for Tap are much more stable and simple. The reaction can be incubated much longer and can be measured anytime without stopping the reaction. If necessary, the reaction can be stopped specifically by APA-CMK. Tap activity is not affected by peroxidases, catalases, phospatases, chelators, or sodium azide which may interfere with common ELISA enzymes. Using Tap in ELISA does not compromise the sensitivity and may even increase sensitivity by using fluorescent substrate.

Example 25

Secretion of Soluble Forms of the Enzymes Encoded by P5-6 and P8-2.

No extracellular hydrolytic activity could be observed in liquid cultures of strains carrying the cloned P8-2 DNA sequence of FIG. 28 even when modified R2 liquid medium was used. Moreover, SDS PAGE analysis with silver staining could not detect extracellular proteins of the anticipated sizes in modified R2 liquid cultures of S.lividans MS7 carrying the cloned DNA sequences of FIG. 25 (eg. P5-6) or FIG. 29 (eg. P8-2). Although the strains carrying these cloned DNA sequences clearly exhibited hydrolytic activities against their respective substrates on modified R2 agar plates, significant levels of these activities could not be localized to either the intracellular or extracellular fractions.

Consistent with these observations, the amino termini of the potential coding regions of P5-6 and P8-2, unlike conventional signal peptides, contain sequences which match well with the signal peptidase II consensus sequence characteristic of lipoproteins. As predicted by von Heijne (1989), the signal peptidase II processing would precede the cysteines in the sequence LATACSAGGAS of P5-6 (FIG. 25 residues 21 to 9 of SEQ ID NO:8) band LTAGCSGGSS of P8-2 (FIG. 29 residues 17-26 of SEQ ID NO: 6) Each sequence shows a striking clustering of turn-producing amino acids following the cysteine, consistent with the amino termini of lipoproteins. The highly positively charged amino terminus of the potential coding region of P5-6, with 7 arginines and a single aspartate, is commonly found on other Gram positive signal peptides. Overall, the amino-terminal sequences for the potential coding regions of P5-6 and P8-2 are consistent with membrane bound forms of each enzyme, designated SlpD and SlpE, respectively.

In order to allow biochemical purification of the predicted proteins from culture supernates, to examine their hydrolytic capabilities and to confirm that the predicted proteins are directly responsible for these activities, the nucleotides encoding both the putative promoter region and the lipoprotein signal peptide including the +1 cysteine were replaced by sequences encoding the aminoglycoside phosphotransferase (aph) promoter and the protease B signal peptide (Henderson et al., 1987). Also, a small leader peptide was added to the C-terminus of the protease B signal peptide which preceded the sequences coding for the SlpD and SlpE proteins. This was accomplished by the use of oligonucleotides to adapt the protease B signal peptide at its C-terminal coding region with the leader and a cloning site, and to adapt the SlpD and SlpE proteins at their N-termini with appropriate cloning sites.

To adapt the C-terminus of the protease B signal peptide, a Streptomyces expression vector (APO.H) containing the aph promoter followed by the protease B signal peptide (Garven and Malek, U.S. Pat. No. 5,200,327), was used. It contained an NsiI cloning site at the 3' end of the protease B signal, with an internal GCA codon encoding the -1 Ala of the protease B signal. A HindIII site was located adjacent to the NsiI site. Oligonucleotides encoding a smaller leader peptide were inserted at the 3' end of the protease B signal peptide by the digestion of APO.H with NsiI and HindIII, then insertion of this pair of oligonucleotides with complementary base extensions to the NsiI and HindIII sites. These oligonucleotides encoded six amino acids (APAAPA SEQ ID NO:14), with an internal PstI cloning site containing a GCA codon at the last Ala of the leader to allow subsequent insertion of the N-termini of the sequences encoding sIpD and sIpE downstream of this leader. This modified vector was designated AP6.H (See FIG. 33A).

To adapt the N-terminus of the SlpD protein, oligonucleotides encoding the 11 amino acids of SlpD immediately downstream of the SPase II +1 cysteine were synthesized. An EcoRl cloning site at the 5' end allowed for ligation of the oligonucleotides into the EcoRI site contained within the polylinker of a T7T318U based subclone (#4) of SlpD clone p5-6. This subclone also contained a HindIII site from the polylinker located 380 nucleotides downstream of the SlpD stop codon. The oligonucleotides also contained at their C-terminus a BamH I site, which joins to a natural BamHI site within the SlpD encoding sequence, located 30 nucleotides downstream from the SPase II +1 cysteine.

A subclone containing these oligonucleotides was subjected to DNA sequence analysis, a routine procedure employed to confirm the fidelity of the cloned oligonucleotide sequence, and the sequence was found to be correct. An NsiI cloning site contained within the N-terminus of the oligonucleotides allowed for ligation to the Pst I site of AP6.H and subsequent joining of the protease B signal plus leader directly to the SlpD at the serine residue immediately adjacent to the SPase II cysteine. The 1920 NsiI to HindIII fragment encoding SlpD was subsequently cloned into AP6.H to produce AP6.SlpD (See FIG. 33B).

An analogous strategy was used to adapt the N-terminus of the SlpE protein with oligonucleotides encoding the 35 amino acids of SIpE immediately downstream of the SPase II +1 cysteine. A PstI compatible site located at the 5' end allowed for ligation of the oligonucleotides into the PstI site located within the polylinker of a T7T318U based subclone (#5) of SlpE clone p8-2. The oligonucleotides also contained at their 3' end a PflMI site which joins to a natural PflMI site within the SlpE encoding sequence, located 100 nucleotides downstream from the SPase II +1 cysteine. At the 3' end of one of the oligonucleotides creating the PflMI site, there is a potential secondary structure which could potentially have caused difficulties in cloning by forming a relatively stable hairpin, thus providing the PflMI sticky end from participating in the ligation. The sequence of this oligonucleotide and its complement were modified to abolish the hairpin structure, while still encoding the correct amino acid sequence for SlpE.

DNA sequence analysis of two of the three pT7T3.18U subclones containing these oligonucleotides showed that their 5' ends did indeed contain the nucleotide sequences from the oligonucleotides (i.e. they contained an NsiI site), but surprisingly, the sequences at their 3' ends upstream of the PflMI cloning site where the nucleotides should have been substituted to abolish the potential hairpin structure, contained wild type nucleotides. The SlpE encoding sequence remained completely intact and in the correct reading frame, and sequences past the PflMI site were also intact and in the correct frame.

An NSiI cloning site contained within the N-terminus allowed for the subsequent ligation in the correct reading frame into the PstI site of AP6.H and the joining of the protease B signal plus leader directly to the SlpE at the serine residue immediately adjacent to the SPase II +1 cysteine. A SacI site located 238 nucleotides downstream of the SlpE stop codon was used in conjunction with a HindIII-SacI 8mer adapter (AGCTAGCT SEQ ID NO:15) to join the 3' end of the SlpE clone to the HindIII site in the AP6.H expression plasmid. The 1820 bp NsiI to SacI fragment encoding SlpE was then used along with the HindIII-SacI adapter in a three way ligation into AP6.H to produce AP6.SlpE (See FIG. 33C).

When these plasmids were used to transform protoplasts of MS11, secreted proteins for both AP6.SlpD and AP6.SlpE were observed at approximate molecular weights of 55kDa and 56kDa, respectively. Direct automated N-terminal Edman degradation analysis of the secreted proteins produced the following amino acid sequences (SEQ ID NOS 16 and 17) SAGGASTXAG for SlpD and APAAPASGGSSDEDK for SlpE. For SLPD, culture supernatants showed a dramatic increase in the ability to hydrolyse APA-BNA.

TABLE VI______________________________________Soluble Protease Substrate AssaysTransformant Timepoint A.sub.405 A.sub.540______________________________________SS12 18 0.144 0.100 23 0.132 0.038 41 0.126 0.018p5-6 17.5 1.147 0.246 23 0.990 0.278 41 0.105 0.000p8-2 17.5 0.115 0.084 23 0.111 0.015 41 0.108 0.036______________________________________ The A.sub.405 values reflect the APA-.delta.NA assay on 20 .mu.l CFB from Tap deleted S.lividans 66 cultures. The A.sub.540 values reflect the Boc-APARSPA-.beta.NA assay on 20 .mu.l CFB from S.lividans 66 cultures. There is no adjustment for dry weights. This correlates with the N-terminal sequence data on SlpD which shows that it is lacking the leader peptide APAAPA (SEQ ID NO:14), which may have been cleaved due to autocatalytic activity of the SlpD itself. In contrast, SlpE culture supernatants showed no ability to hydrolyse APABNa, correlating with the presence of an intact P6 leader at the N-terminus of the secreted protein.

Example 26

Use of Tap to improve secretion of heterologous proteins

Heterologous protein secretion in bacterial cells is facilitated by the inclusion of propeptides between the signal peptide (signal sequence) and the amino acid sequence of the actual heterologous protein. These propeptides are useful for stabilizing the secreted protein against hydrolytic activities and enhancing secretion of the protein by providing a homologous signal peptidase processing site. The use of propeptides for the secretion of heterologous proteins in Streptomyces has been described using signals and propeptides from B-galactosidase for interleukin-1B (Lichenstein et al., 1988) and thaumatin (Illingworth et al., 1989); from tendamistat for proinsulin (Koller et al., 1989), interleukin-2 (Bender et al., 1990a) and hirudin (Bender et al., 1990b); and from serine protease inhibitor for domains of immunoglobulin G (Yoshikata et al., 1993) and CD4 (Ueda et al., 1993).

Although the most common mechanism for the secretion of proteins across biological membranes involves the proteolytic removal of an amino terminal signal peptide with a signal peptidase, certain amino acids of protein structures at or near the amino terminus of the mature protein may block or greatly reduce the efficiency of the signal peptidase, leading to lower secretion of the protein. Some proteins are secreted at low levels using the previously described CANGENUS.TM. expression vector APO.H (see Canadian Patent Numbers 1,295,563; 1,295,566; and 1,295,567; and U.S. Pat. No. 5,200,327 and U.S. patent application Ser. No. 07/397,681). Some of these proteins contain structural constraints located very close to the amino terminus of the mature protein, such as cysteine residues which are involved in a disulfide bond. This may cause steric hindrance to the signal peptidase, thereby preventing cleavage and subsequent release of the mature protein. In such a case, the efficiency of signal peptide removal may be enhanced by insertion at the signal peptidase processing site of amino acids which would provide a more flexible structure between the signal peptide and the amino terminus of the mature protein. The additional amino acids could be removed from the amino terminus of the mature protein. The additional amino acids could be removed from the amino terminus of the secreted protein by an aminopeptidase. The action of the aminopeptidase would be stopped by the amino acid or protein structure at the amino terminus of mature protein. The aminopeptidase may be present in the culture medium into which the protein is being secreted, or may be subsequently added to the secreted protein during the downstream processing.

The present invention describes a process for increasing the level of secreted proteins which have amino terminal structures that interfere with the processing of the signal peptide.

In illustrative embodiments, suitable proteins are interleukin-7 (IL-7), stem cell factor (SCF) and erythropoietin (EPO), which have disulfide bonds involving the amino terminal second, fourth and seventh amino acids, respectively. A signal peptide which is suitable for use for the secretion of IL-7, SCF and EPO is the 37 amino acid signal peptide from the Streptomyces griseus protease B precursor.

The present invention further describes the use of short propeptides, that are multiples of three amino acids in length which, when placed between the signal peptide and the heterologous protein, can increase the level of secreted protein. A peptide leader of either three (APA) or six (APAAPA(SEQ ID NO:14)) amino acids is placed between the protease B signal peptide and the mature protein.

The present invention further describes the secretion of a correctly processed protein secreted from Streptomyces lividans by the successive actions of a signal peptidase to remove the protease B signal peptide, and a tripeptidyl aminopeptidase (Tap) to remove the amino terminal peptide leader. The action of Tap can remove peptides from the propeptide, but not from the heterologous protein, due to an amino-terminal structure, such as a disulfide bond, that prevents further degradation activity.

The present invention further describes the use of Tap for the removal of a propeptide from the amino terminus of a fusion protein comprising a heterologous protein. In a process for the production of a heterologous protein by the secretion of said fusion protein into the growth medium, Tap may be initially present in the growth medium, secreted into the medium during growth, or added after growth to a preparation of said fusion protein.

Two tripeptide leaders that were used were Ala-Pro-Ala (designated AP3) and (Ala-Pro-Ala) -.sub.2 which (SEQ ID NO:14) was designated AP6. Oligonucleotides were designed to encode these amino acids and to create a Pst I site which was then used to introduce DNA fragments encoding proteins to be secreted. The pairs of oligonucleotides when annealed formed sticky ends complementary to Nsil and Hind III. The oligonucleotides APA.1 (SEQ ID NO:18) (GCGCCTGCAGCCTA) and APA.2 (SEQ ID NO:19) (AGCTTAGGCTGCAGGCGCTGCA) were used to make the pAP3.H vector by direct ligation to the Nsil-Hind III vector fragment of pAP0.H, containing the aph promoter and encoding the protease B signal peptide. Similarly, APA2.1 (SEQ ID NO:20) (GCGCCGGCGGCGCCTGCAGCCTA) and APA2.2 (SEQ ID NO:21) (AGCTTAGGCTGCAGGCGCCGCCGGCGCTGCA) were used to make the pAP6.H vector.

PstI-Hind III DNA fragments encoding SCF, IL7 and EPO were ligated to the Pstl-Hind III vector fragments of pAP3.H and pAP6.H respectively. DNA from each of the resulting plasmids was used to transform protoplasts of S.lividans 66. Single transformant colonies were grown in 15 ml LB (containing 5 .mu.g/ml thiostrepton) seed medium for 3 days. After homogenization the cultures were inoculated into 1 liter flasks containing 200 ml TSB. Aliquots were removed after 18, 24 and 30 hours of growth at 30.degree. C. The proteins secreted into the culture supernatant fractions (15 .mu.l aliquots) were analyzed by SDS PAGE and visualized by silver staining. The results for the SCF experiments show (FIG. 34) significantly greater protein secretion by the AP3 and AP6 constructs than those of APO and APz. The inclusion of the peptide leader increased the secretion of SCF approximately 20 fold, IL-7 approximately 10 fold and EPO approximately 5 fold relative to control vectors lacking the propeptides. Each protein was initially secreted with an amino terminal tripeptide or hexapeptide leader. At a later time in the same culture this initial form of each protein was processed to the mature form with the correct amino terminus by the action of the Tap which was secreted into the medium. The amino terminal structure of each of the proteins prevented the Tap from removing any tripeptides from the amino terminus of each mature protein. This invention is applicable to proteins having an amino terminal structure which would prevent Tap digestion and efficient signal peptidase processing.

The present invention has been described in terms of particular embodiments found or proposed by the present inventors to comprise preferred modes for the practice of the invention. It will be appreciated by those of skill in the art that, in light of the present disclosure, numerous modifications and changes can be made in the particular embodiments exemplified without departing from the intended scope of the invention. For example, due to codon redundancy, changes can be made in the underlying DNA sequence without affecting the protein sequence. Moreover, due to biological functional equivalency considerations, changes can be made in protein structure without affecting the biological action in kind or amount. All such modifications are intended to be included within the scope of the appended claims.

REFERENCES

The references listed below are incorporated herein by reference to the extent that they supplement, explain, provide a background for, or teach methodology, techniques and/or compositions employed herein.

Atlan, D., P. Laloi and R. Portalier. 1989. Isolation and characterization of aminopeptidase-deficient Lactobacillus bulgaricus mutants. Appl. Env. Microbiol. 55:1717-1723. Alvarez, N. G., C. Bordallo, S. Gascon and P. S. Rendueles. 1985. Purification and characterization of a thermosensitive X-prolyl dipeptidyl aminopeptidase from S.cersvisiae. BBA 832:119-125. Aretz, W., K-P. Koller and G. Riess. 1989. Proteolytic enzymes from recombinant Streptomyces lividans TK24. FEMS Microbiol. Lett. 65:31-36. Balon, R-M., Tomkinson, B., Ragnorsson, U. and Zetterqvist, O. 1986. J. Biol. Chem. Purification, Substrate Specificity and Classification of Tripeptidyl Peptidase II. 261 (5) 2409-2417. Bender, E., K-P. Koller and J. W. Engels. 1990a. Secretory synthesis of human interleukin-2 by Streptomyces lividans. Gene 86:227-232. Bender, E., Vogel, R., Koller, K. P. and J. W. Engels. 1990b. Synthesis and Secretion of Hirudin by Streptomyces lividans. Appl. Microbiol. Biotechnol. 34:203-207. Bibb, M. J., M. J. Bibb, J. M. Ward and S. N. Cohen. 1985. Nucleotide sequences encoding and promoting expression of three antibiotic resistance genes indigenous to Streptomyces. Mol. Gen. Genet. 199:26-36. Bibb, M. J., P. R. Findlay and M. W. Johnson. 1984. The relationship between base composition and codon usage in bacterial genes and its use for the simple and reliable identification of protein-coding sequences. Gene 30:157-166. Brawner, M., D. Taylor and J. Fornwald. 1990. Expression of the soluble CD-4 receptor in Streptomyces. J. Cell. Biochem., supplement 14A p103. Butler, M. J., C. C. Davey, P. Krygsman, E. Walczyk, and L. T. Malek. 1992. Cloning of genetic loci involved in endoprotease activity in S. lividans 66: a novel neutral protease gene with an adjacent divergent putative regulatory gene. Can. J. Microbiol., in the press. Davies, B. J. 1964. Ann, N.Y. Acad. Sci. 121, 404 Doggette, P. B., and F. R. Blattner. 1986. Personal access of sequence databases on personal computers. Nucleic Acids Res. 14:611-619. Fornwald, J. A., Donovan, J. J., Gerber, R., Keller, J., Taylor, D. P., Arcuri, E. J. and Brawner, M. E. 1993. Soluble forms of the human T cell receptor CD4 are efficiently expressed by Streptomyces lividans. Bio/Technology 11:1031-1036. Fukusawa, K. M. and M. Harada. 1981. Purification and properties of dipeptidyl peptidase IV from Streptococcus mitis ATCC 9811. Arch. Biochem. Biophys. 210:230-237. Hanson, H. and M. Frohne. 1976. Crystalline leucine aminopeptidase from lens in proteolytic enzymes (Ed., L. Lorand) Methods Enzymol. 45:504-521. Henderson, G., P. Krygsman, C. J. Lui, C. C. Davey and L. T. Malek. 1987. Characterization and structure of genes for proteases A and B from Streptomyces griseus. J. Bacterial. 169:3778-3784. Hopwood, D. A., M. J. Bibb, K. F. Chater, T. Kieser, C. J. Bruton, H. M. Kieser, D. J. Lydiate, C. J. Thompson, C. P. Smith, J. M. Ward and H. Schrempf. 1985. Genetic manipulation of Streptomyces, a laboratory manual. The John Innes Foundation, Norwich, U.K. Illingworth, C., Larson, G. and Hellekant, G.. 1989. Secretion of the sweet-tasting plant protein thaumatin by Streptomyces lividans. J. of Industrial Microbiology 4:37-42. Ingram, C., M. Brawner, P. Youngman and J. Westphaling. 1989. xylE functions as an efficient reporter gene in Streptomyces spp.: Use for the study of gal P1, a catabolite-controlled promoter. J. Bacteriol. 177:6617-6624. Koller, K. P., Riess, G., Sauber, K., Uhlmann, E. and Wallmeier, H.. 1989. Recombinant Streptomyces lividans secretes a fusion protein of tendamistat and proinsulin. Biotechnology 7:1055-1059. Kreil, G. 1990. Processing of precursors by dipeptidyl aminopeptidases: a case of molecular ticketing. TIBS. 15:23-26. Lichenstein, H., Brawner, M. F., Miles, L. M., Meyers, C. A., Young, P. R., Simon, P. L. and Eckhardt, T. 1988. Secretion of interleukin-1B and Escherichia coli galactokinase by Streptomyces lividans. Gene. 129:129-134. Lloyd, R. J. and G. G. Pritchard. 1991. Characterization of X-prolyl dipeptidyl aminopeptidase from Lactococcus lactis subsp. lactis. J. Gen. Microbiol. 137:49-55. Malek, L. T., G. Soostmeyer, C. C. Davey, P. Krygsman, J. Compton, J. Gray, T. Zimny and D. Stewart. 1990. Secretion of Granulocyte Macrophase Colony Stimulating Factor (GM-CSF) in Streptomyces lividans. J. Cell. Biochem., supplement 14A, p127. Maniatis, T., E. F. Fritsch and J. Sambrook. 1982. Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. McDonald J. K., Hoisington, A. R. and Eisenhauer, D. A. 1985. Partial Purification and Characterization of an Ovarian Tripeptidyl Peptidase: A lysosomal exopeptidase that sequentially releases Collagen-related (Gly-Pro-x) Triplets. 126:63-71. Menn, F-M., Zylstra, G. J. & Gibson, D. T. 1991. Location and sequence of the tool F gene encoding 2-hydroxy -6-oxohepta-2, 4-dienoate hydrolase in Pseudom . . . putide F1. Gene 104:91-94. Pearson, W. R. and D. J. Lipman. 1988. Improved tools for biological sequence comparison. Proc. Natl. Acad. Sci. USA. 85:2444-2448. Schoellmann, G. and Shaw, E. 1963. Direct evidence for the presence of histidine in the active center of chymotrypsin. Biochemistry 2:252. Shaw, E., Mares-Guia, M., and Cohen, W. 1975. Evidence for an active center histidine in trypsin through the use of a specific reagent, TLCK, the chloromethyl ketone derived from N-tosyl-lysine. Biochemistry 4:2219. Sloma, A., Rufo, G. A., Jr. and Pero, J. Residual protease III WO, A, 92/16642 (see page enclosed) Tagakuchi, S., I. Kumagai, J. Nakayama, A. Suzuki and K. Miura. 1989. Efficient extracelluar expression of a foreign protein in Streptomyces using secretory protease inhibitor (SSI) gene fusions. Biotechnology 7:1063-66. Tinoco, I., Jr., P. N. Borer, B. Dengler, M. D. Levine, O. C. Uhlenbech D. M. Crothers and J. Gralla. 1973. Improved estimation of secondary structure in ribonucleic acid. Nature New Biol. 246:40-41. Tomkinson, B. and Jonsson, A-K. 1991. Characterization of cDNA for Human Tripeptidyl Peptidase II: The N-Terminal Part of the Enzyme is Similar to Subtilising. Biochemistry 30:168-174. Ueda, Y, Tsumoto, K., Watanabe, K. and Kumagai, I. 1993. Synthesis and expression of a DNA encoding the Fv domain of an anti-lysozyme monoclonal antibody, HyHEL10, in Streptomyces lividans. Gene. 120:129-134. von Heijne, G. 1989. The structure of signal peptides from bacterial lipoproteins. Protein Engineering 2:531-534. White, Handler, and Smith. 1973. Wilbur, W. J. and D. J. Lipman. 1983. Rapid Similarity searches of nucleic acid and protein data banks. Proc. Natl. Acad. Sci. USA. 80:726-730. Yoshimoto, T., N. Murayama, T. Honda, H. Tone, and D. Tsuru. 1988. Cloning and expression of aminopeptidase P gene from E.coli HB101 and characterization of expressed enzyme. J. Biochem. 104:93-97. Yoshimoto, T., H. Tone, T. Honda, K. Osatomi, R. Kobayashi, and D. Tsuru. 1989. Sequencing and high expression of aminopeptidase P gene from E.coli HB101. J. Biochem. 105:412-416. Canadian Patent No. 1,295,563. Canadian Patent No. 1,295,566. Canadian Patent No. 1,295,567. U.S. Pat. No. 5,200,327. U.S. patent application, Ser. No. 07/397,681. __________________________________________________________________________SEQUENCE LISTING(1) GENERAL INFORMATION:(iii) NUMBER OF SEQUENCES: 21(2) INFORMATION FOR SEQ ID NO:1:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 1908 base pairs(B) TYPE: nucleic acid(C) STRANDEDNESS: double(D) TOPOLOGY: linear(ix) FEATURE:(A) NAME/KEY: CDS(B) LOCATION: 146..1759(ix) FEATURE:(A) NAME/KEY: misc.sub.-- feature(B) LOCATION: 146..148(D) OTHER INFORMATION: /note= "Met at position -39represents fMet"(ix) FEATURE:(A) NAME/KEY: sig.sub.-- peptide(B) LOCATION: 146..262(ix) FEATURE:(A) NAME/KEY: mat.sub.-- peptide(B) LOCATION: 263..1756(xi) SEQUENCE DESCRIPTION: SEQ ID NO:1:GGCGGGGACCGGCCGACGGCCCCGCCGAACGAACGCCCTTCTCCGTTTATCGGATTGGCA60AAGAAGTAGCACTGGCCCTGTTCTCAGGAAACCCACAGCGGCGAGGATCCCCGTACTTGT120CGCGAACACGTACGGGGAGGGCCACTTGAGGAAGAGCAGCATACGGCGGAGG172MetArgLysSerSerIleArgArgArg39-35GCGACCGCCTTCGGCACGGCCGGAGCACTGGTCACCGCCACGCTGATC220AlaThrAlaPheGlyThrAlaGlyAlaLeuValThrAlaThrLeuIle30-25-20-15GCCGGCGCCGTCTCGGCACCCGCCGCGAGCGCCGCCCCGGCCGACGGC268AlaGlyAlaValSerAlaProAlaAlaSerAlaAlaProAlaAspGly10-51CACGGGCACGGGCGGAGCTGGGACCGGGAGGCGCGCGGTGCCGCCATC316HisGlyHisGlyArgSerTrpAspArgGluAlaArgGlyAlaAlaIle51015GCCGCCGCCCGCGCCGCCCGGGCGGGCATCGACTGGGAGGACTGCGCA364AlaAlaAlaArgAlaAlaArgAlaGlyIleAspTrpGluAspCysAla202530GCCGACTGGAACCTGCCCAAGCCCATCCAGTGCGGCTACGTCACGGTG412AlaAspTrpAsnLeuProLysProIleGlnCysGlyTyrValThrVal35404550CCGATGGACTACGCCAAGCCGTACGGCAAGCAGATCAGGCTCGCCGTC460ProMetAspTyrAlaLysProTyrGlyLysGlnIleArgLeuAlaVal556065GACCGCATCGGCAACACCGGAACCAGGAGCGAGCGCCAGGGCGCCCTG508AspArgIleGlyAsnThrGlyThrArgSerGluArgGlnGlyAlaLeu707580ATCTACAACCCCGGCGGTCCCGGCGGCTCCGGCCTGCGTTTCCCGGCC556IleTyrAsnProGlyGlyProGlyGlySerGlyLeuArgPheProAla859095CGCGTCACGAACAAGAGCGCGGTCTGGGCCAACACGGCCAAGGCCTAC604ArgValThrAsnLysSerAlaValTrpAlaAsnThrAlaLysAlaTyr100105110GACTTCGTCGGCTTCGACCCGCGCGGCGTCGGCCACTCCGCGCCCATC652AspPheValGlyPheAspProArgGlyValGlyHisSerAlaProIle115120125130TCCTGCGTCGACCCGCAGGAGTTCGTCAAGGCACCCAAGGCCGACCCC700SerCysValAspProGlnGluPheValLysAlaProLysAlaAspPro135140145GTGCCCGGCTCCGAGGCCGACAAGCGCGCCCAGCGCAAGCTCGCCCGC748ValProGlySerGluAlaAspLysArgAlaGlnArgLysLeuAlaArg150155160GAGTACGCCGAGGGCTGCTTCGAGCGCAGCGGCGAGATGCTCCCGCAC796GluTyrAlaGluGlyCysPheGluArgSerGlyGluMetLeuProHis165170175ATGACCACGCCGAACACCGCGCGCGACCTCGACGTCATCCGCGCCGCC844MetThrThrProAsnThrAlaArgAspLeuAspValIleArgAlaAla180185190CTCGGCGAGAAGAAGCTCAACTACCTCGGCGTCTCCTACGGCACCTAC892LeuGlyGluLysLysLeuAsnTyrLeuGlyValSerTyrGlyThrTyr195200205210CTCGGCGCCGTCTACGGCACCCTCTTCCCGGACCACGTCCGCCGCATG940LeuGlyAlaValTyrGlyThrLeuPheProAspHisValArgArgMet215220225GTCGTCGACAGCGTCGTCAACCCGTCCCGCGACAAGATCTGGTACCAG988ValValAspSerValValAsnProSerArgAspLysIleTrpTyrGln230235240GCCAACCTGGACCAGGACGTCGCCTTCGAGGGCCGCTGGAAGGACTGG1036AlaAsnLeuAspGlnAspValAlaPheGluGlyArgTrpLysAspTrp245250255CAGGACTGGGTCGCCGCGAACGACGCCGCCTACCACCTCGGCGACACC1084GlnAspTrpValAlaAlaAsnAspAlaAlaTyrHisLeuGlyAspThr260265270CGCGCCGAGGTCCAGGACCAGTGGCTGAAGCTGCGCGCCGCCGCCGCG1132ArgAlaGluValGlnAspGlnTrpLeuLysLeuArgAlaAlaAlaAla275280285290AAGAAGCCGCTGGGCGGCGTCGTCGGACCGGCCGAGCTGATCTCCTTC1180LysLysProLeuGlyGlyValValGlyProAlaGluLeuIleSerPhe295300305TTCCAGAGCGCCCCGTACTACGACTCCGCCTGGGCGCCGACCGCGGAG1228PheGlnSerAlaProTyrTyrAspSerAlaTrpAlaProThrAlaGlu310315320ATCTTCAGCAAGTACGTCGCCGGCGACACCCAGGCGCTCGTCGACGCC1276IlePheSerLysTyrValAlaGlyAspThrGlnAlaLeuValAspAla325330335GCCGCACCCGACCTGTCCGACACCGCGGGCAACGCCTCCGCGGAGAAC1324AlaAlaProAspLeuSerAspThrAlaGlyAsnAlaSerAlaGluAsn340345350GGCAACGCCGTCTACACGGCCGTCGAGTGCACCGACGCCAAGTGGCCC1372GlyAsnAlaValTyrThrAlaValGluCysThrAspAlaLysTrpPro355360365370GCCAACTGGCGCACCTGGGACCGGGACAACACCCGGCTCCACCGCGAC1420AlaAsnTrpArgThrTrpAspArgAspAsnThrArgLeuHisArgAsp375380385CACCCGTTCATGACCTGGGCCAACGCCTGGATGAACCTGCCCTGTGCC1468HisProPheMetThrTrpAlaAsnAlaTrpMetAsnLeuProCysAla390395400ACCTGGCCGGTCAAGCAGCAGACCCCGCTGAACGTGAAGACCGGCAAG1516ThrTrpProValLysGlnGlnThrProLeuAsnValLysThrGlyLys405410415GGACTTCCGCCGGTGCTGATCGTCCAGTCCGAGCGTGACGCCGCCACC1564GlyLeuProProValLeuIleValGlnSerGluArgAspAlaAlaThr420425430CCGTACGAGGGCGCCGTCGAACTGCACCAGCGGTTCCGGGGATCCCGC1612ProTyrGluGlyAlaValGluLeuHisGlnArgPheArgGlySerArg435440445450CTGATCACCGAGCGGGACGCCGGCTCCCACGGCGTCACCGGCCTGGTC1660LeuIleThrGluArgAspAlaGlySerHisGlyValThrGlyLeuVal455460465AACCCGTGCATCAACGACCGGGTCGACACCTACCTGCTCACCGGCAGG1708AsnProCysIleAsnAspArgValAspThrTyrLeuLeuThrGlyArg470475480ACGGACGCCCGCGACGTGACCTGCGCGCCGCACGCCACGCCCAGGCCG1756ThrAspAlaArgAspValThrCysAlaProHisAlaThrProArgPro485490495TAACCCGGGCTCAGGCCAAGCGGGGGGAGGGGGCGACCGGTCCGACCGGCCGCCCCCTCC1816CCCCACCTGTCGCTACCGTCCCTCGGCCCAGGCGTCCTCCGCCGCGTAGTCGAAGAGGTC1876GCCGTACGCCTTGAACATCTTCGGGTAGGCCT1908(2) INFORMATION FOR SEQ ID NO:2:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 537 amino acids(B) TYPE: amino acid(D) TOPOLOGY: linear(ii) MOLECULE TYPE: protein(xi) SEQUENCE DESCRIPTION: SEQ ID NO:2:MetArgLysSerSerIleArgArgArgAlaThrAlaPheGlyThrAla39- 35-30-25GlyAlaLeuValThrAlaThrLeuIleAlaGlyAlaValSerAlaPro20-15- 10AlaAlaSerAlaAlaProAlaAspGlyHisGlyHisGlyArgSerTrp515AspArgGluAlaArgGlyAlaAlaIleAlaAlaAlaArgAlaAlaArg10152025AlaGlyIleAspTrpGluAspCysAlaAlaAspTrpAsnLeuProLys303540ProIleGlnCysGlyTyrValThrValProMetAspTyrAlaLysPro455055TyrGlyLysGlnIleArgLeuAlaValAspArgIleGlyAsnThrGly606570ThrArgSerGluArgGlnGlyAlaLeuIleTyrAsnProGlyGlyPro758085GlyGlySerGlyLeuArgPheProAlaArgValThrAsnLysSerAla9095100105ValTrpAlaAsnThrAlaLysAlaTyrAspPheValGlyPheAspPro110115120ArgGlyValGlyHisSerAlaProIleSerCysValAspProGlnGlu125130135PheValLysAlaProLysAlaAspProValProGlySerGluAlaAsp140145150LysArgAlaGlnArgLysLeuAlaArgGluTyrAlaGluGlyCysPhe155160165GluArgSerGlyGluMetLeuProHisMetThrThrProAsnThrAla170175180185ArgAspLeuAspValIleArgAlaAlaLeuGlyGluLysLysLeuAsn190195200TyrLeuGlyValSerTyrGlyThrTyrLeuGlyAlaValTyrGlyThr205210215LeuPheProAspHisValArgArgMetValValAspSerValValAsn220225230ProSerArgAspLysIleTrpTyrGlnAlaAsnLeuAspGlnAspVal235240245AlaPheGluGlyArgTrpLysAspTrpGlnAspTrpValAlaAlaAsn250255260265AspAlaAlaTyrHisLeuGlyAspThrArgAlaGluValGlnAspGln270275280TrpLeuLysLeuArgAlaAlaAlaAlaLysLysProLeuGlyGlyVal285290295ValGlyProAlaGluLeuIleSerPhePheGlnSerAlaProTyrTyr300305310AspSerAlaTrpAlaProThrAlaGluIlePheSerLysTyrValAla315320325GlyAspThrGlnAlaLeuValAspAlaAlaAlaProAspLeuSerAsp330335340345ThrAlaGlyAsnAlaSerAlaGluAsnGlyAsnAlaValTyrThrAla350355360ValGluCysThrAspAlaLysTrpProAlaAsnTrpArgThrTrpAsp365370375ArgAspAsnThrArgLeuHisArgAspHisProPheMetThrTrpAla380385390AsnAlaTrpMetAsnLeuProCysAlaThrTrpProValLysGlnGln395400405ThrProLeuAsnValLysThrGlyLysGlyLeuProProValLeuIle410415420425ValGlnSerGluArgAspAlaAlaThrProTyrGluGlyAlaValGlu430435440LeuHisGlnArgPheArgGlySerArgLeuIleThrGluArgAspAla445450455GlySerHisGlyValThrGlyLeuValAsnProCysIleAsnAspArg460465470ValAspThrTyrLeuLeuThrGlyArgThrAspAlaArgAspValThr475480485CysAlaProHisAlaThrProArgPro490495(2) INFORMATION FOR SEQ ID NO:3:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 2185 base pairs(B) TYPE: nucleic acid(C) STRANDEDNESS: double(D) TOPOLOGY: linear(ix) FEATURE:(A) NAME/KEY: CDS(B) LOCATION: 531..2069(ix) FEATURE:(A) NAME/KEY: sig.sub.-- peptide(B) LOCATION: 531..902(ix) FEATURE:(A) NAME/KEY: mat.sub.-- peptide(B) LOCATION: 903..2069(ix) FEATURE:(A) NAME/KEY: misc.sub.-- feature(B) LOCATION: 531..533(D) OTHER INFORMATION: /note= "Met at position -124represents fMet"(xi) SEQUENCE DESCRIPTION: SEQ ID NO:3:GGTACCAGGCGACGAAGGCGACGGTCAGCGGGAACGCGAAGGAACGGAAGGAGCGGCGCA60GTTCGGCGAACTCGGCGCTCTGCTGCACTTCGGAGAACTCCTCGGCGGAGGGGAGGCGGT120GCTCCTCTTGCGAGGGGGGCTCCTCTTTGGAGGGGGGCGGTGCGTCGGGTGGCCACGGAG180TCTCCTCGTACGACGGACATGACGGCTTGGACCTCGGTGTTCTCGCAGGGGGCTGATCGT240GCTCGGGCTCCCTGTCCAACGACACGGCGCCCCGCGGGGCCCGGTTCAACACCCGTGGCA300CTTTCCGAAGTCGTCCTCGGCGGGTCATTGCTGGCCAGGGACTTCGGGGGATAGCTTCAC360CCTGCACCACTACGTCATGTACCTGCCCGGCCCGTTTCACCCGTGCCCGGGCAGGTGCTG420TTTGCCGGATGATGTGGAGACCCCATGGATCATCTGCGCTTCCCGCGCGACCCGCGCTCC480AGACGCGGGCTCGTTTCCCGAGCTTTCCCGACGGACTGGAGACATCACGCATGACC536MetThr124GCTCCCCTCTCGCGTCACCGCCGTGCCCTCGCGATTCCGGCGGGCCTG584AlaProLeuSerArgHisArgArgAlaLeuAlaIleProAlaGlyLeu120-115-110GCCGTGGCCGCGTCGCTCGCGTTCCTGCCGGGCACCCCGGCCGCCGCG632AlaValAlaAlaSerLeuAlaPheLeuProGlyThrProAlaAlaAla105-100-95ACCCCCGCGGCCGAGGCCGCGCCCTCGACGGCGGCGGACGCGACCTCG680ThrProAlaAlaGluAlaAlaProSerThrAlaAlaAspAlaThrSer90-85-80-75CTCAGCTACGTCGTCAACGTCGCCTCCGGGCACCGTCCTTCGGCCACC728LeuSerTyrValValAsnValAlaSerGlyHisArgProSerAlaThr70-65-60GTGCGGCGGGCGATAGCCAAGGCGGGCGGCACGATCGTCACGTCGTAC776ValArgArgAlaIleAlaLysAlaGlyGlyThrIleValThrSerTyr55-50- 45GACCGGATCGGCGTGATCGTCGTCCACTCCGCCAACCCCGACTTCGCC824AspArgIleGlyValIleValValHisSerAlaAsnProAspPheAla40-35-30AAGACCGTGCGCAAGGTGCGCGGCGTGCAGTCGGCCGGTGCCACCCGC872LysThrValArgLysValArgGlyValGlnSerAlaGlyAlaThrArg25-20-15ACCGCGCCACTGCCCTCGGCCGCCACCACCGACACGGGCGCGCCGCAG920ThrAlaProLeuProSerAlaAlaThrThrAspThrGlyAlaProGln10-515GTGCTCGGCGGCGAGGACCTGGCCGCCGCCAAGGCCGCCTCCGCGAAG968ValLeuGlyGlyGluAspLeuAlaAlaAlaLysAlaAlaSerAlaLys101520GCCGAGGGCCAGGACCCGCTGGAGTCGCTCCAGTGGGACCTGCCCGCC1016AlaGluGlyGlnAspProLeuGluSerLeuGlnTrpAspLeuProAla253035ATCAAGGCGGACAAGGCGCACGAGAAGTCGCTGGGCAGCAGGAAGGTG1064IleLysAlaAspLysAlaHisGluLysSerLeuGlySerArgLysVal404550ACCGTCGCCGTCATCGACACCGGCGTCGACGACACCCACCCGGACATC1112ThrValAlaValIleAspThrGlyValAspAspThrHisProAspIle55606570GCCCCGAACTTCGACCGGCAGGCGTCCGTCAACTGTGTGGCGGGCAAG1160AlaProAsnPheAspArgGlnAlaSerValAsnCysValAlaGlyLys758085CCGGACACCGCCGACGGGGCCTGGCGGCCGAGCGCGGCGGAGAGCCCG1208ProAspThrAlaAspGlyAlaTrpArgProSerAlaAlaGluSerPro9095100CACGGCACCCACGTGGCCGGGGAGATAGCCGCCGCCAAGAACGGCGTC1256HisGlyThrHisValAlaGlyGluIleAlaAlaAlaLysAsnGlyVal105110115GGCATGACCGGCGTGGCACCCGGGGTGAAGGTGGCCGGCATCAAGGTC1304GlyMetThrGlyValAlaProGlyValLysValAlaGlyIleLysVal120125130TCCAACCCCGACGGCTTCTTCTACACCGAGGCCGTGGTCTGCGGCTTC1352SerAsnProAspGlyPhePheTyrThrGluAlaValValCysGlyPhe135140145150ATGTGGGCGGCCGAGCACGGCGTCGACGTGACCAACAACAGCTATTAC1400MetTrpAlaAlaGluHisGlyValAspValThrAsnAsnSerTyrTyr155160165ACCGACCCGTGGTACTTCAACTGCAAGGACGACCCCGACCAGAAGGCG1448ThrAspProTrpTyrPheAsnCysLysAspAspProAspGlnLysAla170175180CTCGTCGAGGCCGTCTCGCGGGCCTCCCGGTACGCGGAGAAGAAGGGC1496LeuValGluAlaValSerArgAlaSerArgTyrAlaGluLysLysGly185190195GCGGTCAACGTCGCCGCGGCCGGCAACGAGAACTACGACCTCACCTCC1544AlaValAsnValAlaAlaAlaGlyAsnGluAsnTyrAspLeuThrSer200205210GACGAGATCACCGACCCGTCCTCGCCCAACGACACCACGCCCGGCGAC1592AspGluIleThrAspProSerSerProAsnAspThrThrProGlyAsp215220225230CGGACCGTCGACCCGTCGAAGTGCCTGGACATCCCGACCCAGCTGCCG1640ArgThrValAspProSerLysCysLeuAspIleProThrGlnLeuPro235240245GGTGTCGTGACGGTCGCGGCGACCGGTGCGAAGGGCCTCAAGTCGTCC1688GlyValValThrValAlaAlaThrGlyAlaLysGlyLeuLysSerSer250255260TTCTCCAACCACGGGCTGGGCGTCATCGACATCGCCGCGCCCGGCGGC1736PheSerAsnHisGlyLeuGlyValIleAspIleAlaAlaProGlyGly265270275GACTCGACGGCCTACCAGACCCCGGAGCCGCCCGCCACGAGCGGCCTG1784AspSerThrAlaTyrGlnThrProGluProProAlaThrSerGlyLeu280285290ATCCTGGGCACGCTGCCCGGCGGCAAGTGGGGCTACATGGCCGGTACG1832IleLeuGlyThrLeuProGlyGlyLysTrpGlyTyrMetAlaGlyThr295300305310TCCATGGCCTCCCCGCACGTCGCGGGCGTCGCCGCCCTCATCAAGTCG1880SerMetAlaSerProHisValAlaGlyValAlaAlaLeuIleLysSer315320325ACGCACCCGCACGCCTCCCCCGCCATGGTGAAGGCGCTGCTGTACGCC1928ThrHisProHisAlaSerProAlaMetValLysAlaLeuLeuTyrAla330335340GAGGCCGACGCCACGGCGTGCACCAAGCCGTACGACATCGACGGCGAC1976GluAlaAspAlaThrAlaCysThrLysProTyrAspIleAspGlyAsp345350355GGCAAGGTCGACGCGGTGTGCGAGGGCCCGAAGAACCGCAACGGCTTC2024GlyLysValAspAlaValCysGluGlyProLysAsnArgAsnGlyPhe360365370TACGGCTGGGGCATGGCCGACGCGCTGGACGCGGTGACCTGGTAGCCGGTAC2076TyrGlyTrpGlyMetAlaAspAlaLeuAspAlaValThrTrp375380385GCGTACCCGTGCGTGAGGCGGGGGCGGCGGTCCGGTTCCCGTCCGGTCCGCCGCCCCCGT2136CGTCGTCGTCGTACGACAGTATCTTCGCCATGGACACTTACGAGGATCC2185(2) INFORMATION FOR SEQ ID NO:4:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 512 amino acids(B) TYPE: amino acid(D) TOPOLOGY: linear(ii) MOLECULE TYPE: protein(xi) SEQUENCE DESCRIPTION: SEQ ID NO:4:MetThrAlaProLeuSerArgHisArgArgAlaLeuAlaIleProAla124- 120-115-110GlyLeuAlaValAlaAlaSerLeuAlaPheLeuProGlyThrProAla105-100- 95AlaAlaThrProAlaAlaGluAlaAlaProSerThrAlaAlaAspAla90-85-80ThrSerLeuSerTyrValValAsnValAlaSerGlyHisArgProSer75-70-65AlaThrValArgArgAlaIleAlaLysAlaGlyGlyThrIleValThr60-55-50-45SerTyrAspArgIleGlyValIleValValHisSerAlaAsnProAsp40-35-30PheAlaLysThrValArgLysValArgGlyValGlnSerAlaGlyAla25-20- 15ThrArgThrAlaProLeuProSerAlaAlaThrThrAspThrGlyAla10-51ProGlnValLeuGlyGlyGluAspLeuAlaAlaAlaLysAlaAlaSer5101520AlaLysAlaGluGlyGlnAspProLeuGluSerLeuGlnTrpAspLeu253035ProAlaIleLysAlaAspLysAlaHisGluLysSerLeuGlySerArg404550LysValThrValAlaValIleAspThrGlyValAspAspThrHisPro556065AspIleAlaProAsnPheAspArgGlnAlaSerValAsnCysValAla707580GlyLysProAspThrAlaAspGlyAlaTrpArgProSerAlaAlaGlu859095100SerProHisGlyThrHisValAlaGlyGluIleAlaAlaAlaLysAsn105110115GlyValGlyMetThrGlyValAlaProGlyValLysValAlaGlyIle120125130LysValSerAsnProAspGlyPhePheTyrThrGluAlaValValCys135140145GlyPheMetTrpAlaAlaGluHisGlyValAspValThrAsnAsnSer150155160TyrTyrThrAspProTrpTyrPheAsnCysLysAspAspProAspGln165170175180LysAlaLeuValGluAlaValSerArgAlaSerArgTyrAlaGluLys185190195LysGlyAlaValAsnValAlaAlaAlaGlyAsnGluAsnTyrAspLeu200205210ThrSerAspGluIleThrAspProSerSerProAsnAspThrThrPro215220225GlyAspArgThrValAspProSerLysCysLeuAspIleProThrGln230235240LeuProGlyValValThrValAlaAlaThrGlyAlaLysGlyLeuLys245250255260SerSerPheSerAsnHisGlyLeuGlyValIleAspIleAlaAlaPro265270275GlyGlyAspSerThrAlaTyrGlnThrProGluProProAlaThrSer280285290GlyLeuIleLeuGlyThrLeuProGlyGlyLysTrpGlyTyrMetAla295300305GlyThrSerMetAlaSerProHisValAlaGlyValAlaAlaLeuIle310315320LysSerThrHisProHisAlaSerProAlaMetValLysAlaLeuLeu325330335340TyrAlaGluAlaAspAlaThrAlaCysThrLysProTyrAspIleAsp345350355GlyAspGlyLysValAspAlaValCysGluGlyProLysAsnArgAsn360365370GlyPheTyrGlyTrpGlyMetAlaAspAlaLeuAspAlaValThrTrp375380385(2) INFORMATION FOR SEQ ID NO:5:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 1777 base pairs(B) TYPE: nucleic acid(C) STRANDEDNESS: double(D) TOPOLOGY: linear(ix) FEATURE:(A) NAME/KEY: CDS(B) LOCATION: 190..1731(ix) FEATURE:(A) NAME/KEY: misc.sub.-- feature(B) LOCATION: 190..192(D) OTHER INFORMATION: /note= "Met at position 1represents fMet"(xi) SEQUENCE DESCRIPTION: SEQ ID NO:5:GGTACCGGCGGCCAAGACCGTGTGCTCCTGACCGCGGACGCCACCACAGGTCGGCAGAAG60CAGCAGATCGACAGAAGTAGCAGGTCAGAGCGTTATCCACAGGCGTCGGCGGGTGCTGCC120CCCGCCACCTACCATGGCAGGAACGCCATCCGCCGCACGGCGCGGACGGCTTGCCAGGGG180GGAGAGGACATGGCGCGTCTCGTCCGGTGGACGGCTCTGACGGCCGCC228MetAlaArgLeuValArgTrpThrAlaLeuThrAlaAla1510GCCGCACTGCTGACGGCGGGCTGCAGCGGCGGCTCGTCCGACGAGGAC276AlaAlaLeuLeuThrAlaGlyCysSerGlyGlySerSerAspGluAsp152025AAGGACGACGGGGGCAGGAGCAGCGCGGGACCTTCGGCGGCGGCACCC324LysAspAspGlyGlyArgSerSerAlaGlyProSerAlaAlaAlaPro30354045TCCGGGGTGCCGGAGGCACTGGCGTCCCAGACGCTGGACTGGGCCCGA372SerGlyValProGluAlaLeuAlaSerGlnThrLeuAspTrpAlaArg505560TGCGAGGGCAGCGACGATGCCCCGGCGCCGGACGGCGACTGGCGGTGC420CysGluGlySerAspAspAlaProAlaProAspGlyAspTrpArgCys657075GCCACGCTGAAGGCACCGCTGGACTGGTCCGACCCCGACGGCGAGACG468AlaThrLeuLysAlaProLeuAspTrpSerAspProAspGlyGluThr808590ATCGATCTCGCGCTGATCCGGTCCCGGGCGAGCGGGGACGACCGCATC516IleAspLeuAlaLeuIleArgSerArgAlaSerGlyAspAspArgIle95100105GGCTCCCTGCTGTTCAACTTCGGCGGCCCGGGCGCCTCCGGCGTCTCC564GlySerLeuLeuPheAsnPheGlyGlyProGlyAlaSerGlyValSer110115120125ACGATGCCGTCCTACGCCGACACCGTCTCCTCCCTGCACGAGCGGTAC612ThrMetProSerTyrAlaAspThrValSerSerLeuHisGluArgTyr130135140GACCTGGTGAGCTGGGACCCGCGCGGGGTGGCCGCCAGCGAGGGCGTC660AspLeuValSerTrpAspProArgGlyValAlaAlaSerGluGlyVal145150155CGCTGCCGCACCGACGAGGCGATCGAGGCCGCCGAGTCGGTGGACTCC708ArgCysArgThrAspGluAlaIleGluAlaAlaGluSerValAspSer160165170ACGCCGGACTCCCCGGCCGAGGAGCAGGCCTACCTGAAGGACGCCGCC756ThrProAspSerProAlaGluGluGlnAlaTyrLeuLysAspAlaAla175180185GACTTCGGCAGGGGCTGCGAGAAGGCCGCCGGCAAGCTCATGGAACAC804AspPheGlyArgGlyCysGluLysAlaAlaGlyLysLeuMetGluHis190195200205GTCTCGACCACGGACACGGCCCGCGACATGGACCTGATGCGGCACGTC852ValSerThrThrAspThrAlaArgAspMetAspLeuMetArgHisVal210215220CTGGGCGACGAGAGGATGCACTACTTCGGCATCTCCTACGGCACCGAA900LeuGlyAspGluArgMetHisTyrPheGlyIleSerTyrGlyThrGlu225230235CTCGGCGGCGTCTACGCCCATCTGTTCCCCGAGCACGTGGGCCGCGTG948LeuGlyGlyValTyrAlaHisLeuPheProGluHisValGlyArgVal240245250ATCCTCGACGCGGTGGTGGACCCGGGCGCCGACACGATGGGCCACGCC996IleLeuAspAlaValValAspProGlyAlaAspThrMetGlyHisAla255260265GAGAACCAGGCCAGGGGTTTCCAGCGCGCGCTGGACGACTACCTGGAG1044GluAsnGlnAlaArgGlyPheGlnArgAlaLeuAspAspTyrLeuGlu270275280285TCGACCGGCCAGGAACCCGAACAGGGGTCGCGGAAGATCGCCGGCCTG1092SerThrGlyGlnGluProGluGlnGlySerArgLysIleAlaGlyLeu290295300CTGGAGCGGCTGGACGCCGAGCCACTGCCCACGTCCTCGCCGGGGCGG1140LeuGluArgLeuAspAlaGluProLeuProThrSerSerProGlyArg305310315GAGCTGACGCAGACCCTCGCGTTCACCGGCATCGTGCTGCCGCTGTAC1188GluLeuThrGlnThrLeuAlaPheThrGlyIleValLeuProLeuTyr320325330AGCGAGAGCGGCTGGCCGGCCCTGACCAGTGCGCTGAAGGCGGCCGAG1236SerGluSerGlyTrpProAlaLeuThrSerAlaLeuLysAlaAlaGlu335340345GAGGGCGACGGCTCGGAGTTGCTGGCCCTCGCCGACGGCTACAACGAG1284GluGlyAspGlySerGluLeuLeuAlaLeuAlaAspGlyTyrAsnGlu350355360365CGTGATCCCTCGGGGCGCTACGGCACGACGACCCACTCGCAAAGGGTC1332ArgAspProSerGlyArgTyrGlyThrThrThrHisSerGlnArgVal370375380ATATCGTGCCTGGACGACAAGCAGAGGCCGACCGTGGAGGAGACGAAG1380IleSerCysLeuAspAspLysGlnArgProThrValGluGluThrLys385390395AAGCTGCTGCCGAGGTTCGAGAAGGTCTCTCCCGTCTTCGGCGCCTTC1428LysLeuLeuProArgPheGluLysValSerProValPheGlyAlaPhe400405410CTCGGCTGGGACACGGCCGGGTGGTGCCACGACTGGCCGGTGGCCGGT1476LeuGlyTrpAspThrAlaGlyTrpCysHisAspTrpProValAlaGly415420425CAGCACGAGACCGCGGAGGTGAGCGCGCCCGACGCGGCCCCGGTCCTG1524GlnHisGluThrAlaGluValSerAlaProAspAlaAlaProValLeu430435440445GTGGTCGGCAACACGGGCGACCCGGCCACGCCCTACGAGGGCGCCCGC1572ValValGlyAsnThrGlyAspProAlaThrProTyrGluGlyAlaArg450455460AGGATGGCGGACGAGCTGGGCAAGGACGTCGGCGTGGTGCTGACCTGG1620ArgMetAlaAspGluLeuGlyLysAspValGlyValValLeuThrTrp465470475CAGGGCGAGGGACACGGTGCCTACGGGAACGGAAGCGACTGTGTCGAC1668GlnGlyGluGlyHisGlyAlaTyrGlyAsnGlySerAspCysValAsp480485490TCCGCGGTGGACGCCTACCTGTTGAAGGGGACGGTGCCGAAGGACGGC1716SerAlaValAspAlaTyrLeuLeuLysGlyThrValProLysAspGly495500505AAGGTCTGCTCATGACGGCGGCGGGGGCTTCGGGCACCTGCGGTGCGCGAAA1768LysValCysSer510CCCCCGCCG1777(2) INFORMATION FOR SEQ ID NO:6:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 513 amino acids(B) TYPE: amino acid(D) TOPOLOGY: linear(ii) MOLECULE TYPE: protein(xi) SEQUENCE DESCRIPTION: SEQ ID NO:6:MetAlaArgLeuValArgTrpThrAlaLeuThrAlaAlaAlaAlaLeu151015LeuThrAlaGlyCysSerGlyGlySerSerAspGluAspLysAspAsp202530GlyGlyArgSerSerAlaGlyProSerAlaAlaAlaProSerGlyVal354045ProGluAlaLeuAlaSerGlnThrLeuAspTrpAlaArgCysGluGly505560SerAspAspAlaProAlaProAspGlyAspTrpArgCysAlaThrLeu65707580LysAlaProLeuAspTrpSerAspProAspGlyGluThrIleAspLeu859095AlaLeuIleArgSerArgAlaSerGlyAspAspArgIleGlySerLeu100105110LeuPheAsnPheGlyGlyProGlyAlaSerGlyValSerThrMetPro115120125SerTyrAlaAspThrValSerSerLeuHisGluArgTyrAspLeuVal130135140SerTrpAspProArgGlyValAlaAlaSerGluGlyValArgCysArg145150155160ThrAspGluAlaIleGluAlaAlaGluSerValAspSerThrProAsp165170175SerProAlaGluGluGlnAlaTyrLeuLysAspAlaAlaAspPheGly180185190ArgGlyCysGluLysAlaAlaGlyLysLeuMetGluHisValSerThr195200205ThrAspThrAlaArgAspMetAspLeuMetArgHisValLeuGlyAsp210215220GluArgMetHisTyrPheGlyIleSerTyrGlyThrGluLeuGlyGly225230235240ValTyrAlaHisLeuPheProGluHisValGlyArgValIleLeuAsp245250255AlaValValAspProGlyAlaAspThrMetGlyHisAlaGluAsnGln260265270AlaArgGlyPheGlnArgAlaLeuAspAspTyrLeuGluSerThrGly275280285GlnGluProGluGlnGlySerArgLysIleAlaGlyLeuLeuGluArg290295300LeuAspAlaGluProLeuProThrSerSerProGlyArgGluLeuThr305310315320GlnThrLeuAlaPheThrGlyIleValLeuProLeuTyrSerGluSer325330335GlyTrpProAlaLeuThrSerAlaLeuLysAlaAlaGluGluGlyAsp340345350GlySerGluLeuLeuAlaLeuAlaAspGlyTyrAsnGluArgAspPro355360365SerGlyArgTyrGlyThrThrThrHisSerGlnArgValIleSerCys370375380LeuAspAspLysGlnArgProThrValGluGluThrLysLysLeuLeu385390395400ProArgPheGluLysValSerProValPheGlyAlaPheLeuGlyTrp405410415AspThrAlaGlyTrpCysHisAspTrpProValAlaGlyGlnHisGlu420425430ThrAlaGluValSerAlaProAspAlaAlaProValLeuValValGly435440445AsnThrGlyAspProAlaThrProTyrGluGlyAlaArgArgMetAla450455460AspGluLeuGlyLysAspValGlyValValLeuThrTrpGlnGlyGlu465470475480GlyHisGlyAlaTyrGlyAsnGlySerAspCysValAspSerAlaVal485490495AspAlaTyrLeuLeuLysGlyThrValProLysAspGlyLysValCys500505510Ser(2) INFORMATION FOR SEQ ID NO:7:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 1821 base pairs(B) TYPE: nucleic acid(C) STRANDEDNESS: double(D) TOPOLOGY: linear(ix) FEATURE:(A) NAME/KEY: CDS(B) LOCATION: 104..1720(ix) FEATURE:(A) NAME/KEY: sig.sub.-- peptide(B) LOCATION: 104..244(ix) FEATURE:(A) NAME/KEY: mat.sub.-- peptide(B) LOCATION: 245..1720(xi) SEQUENCE DESCRIPTION: SEQ ID NO:7:CCCGGGCCCGCGTCGGAGTCATGACCGGTTGACGCCGTAACACGTACGGGGCACGCGCAC60CACGCACCGCAACTGCTTCGTCGCGGAGAGTTACGCTCGCTGAATGGACACAAGG115MetAspThrArg47-45CGCACTCACCGCAGGACCCGCACCGGCGGCACCCGTTTCCGGGCCACG163ArgThrHisArgArgThrArgThrGlyGlyThrArgPheArgAlaThr40-35- 30CTGCTCACCGCCGCGCTGCTCGCCACCGCCTGCTCGGCCGGGGGCGCG211LeuLeuThrAlaAlaLeuLeuAlaThrAlaCysSerAlaGlyGlyAla25-20-15TCGACGTCCGCCGGATCCCCCGCGGCCAAGGCGGCCGGCGCGACGGAG259SerThrSerAlaGlySerProAlaAlaLysAlaAlaGlyAlaThrGlu10- 515GCGGCCACGGCGACCCTGACCCCCCTGCCGAAGGCCACGCCCGCCGAG307AlaAlaThrAlaThrLeuThrProLeuProLysAlaThrProAlaGlu101520CTGTCCCCGTACTACGAGCAGAAGCTCGGCTGGCGCGACTGCGGCGTC355LeuSerProTyrTyrGluGlnLysLeuGlyTrpArgAspCysGlyVal253035CCGGGCTTCCAGTGCGCCACCATGAAGGCCCCGCTCGACTACGCCAAG403ProGlyPheGlnCysAlaThrMetLysAlaProLeuAspTyrAlaLys404550CCCGCCGACGGCGACGTCCGGCTCGCGGTGGCCCGCAAGAAGGCCACG451ProAlaAspGlyAspValArgLeuAlaValAlaArgLysLysAlaThr556065GGGCCGGGCAAGCGCCTCGGCTCGCTGCTGGTCAACCCGGGCGGACCG499GlyProGlyLysArgLeuGlySerLeuLeuValAsnProGlyGlyPro70758085GGCGGCTCGGCGATCGGCTACCTCCAGCAGTACGCGGGCATCGGCTAC547GlyGlySerAlaIleGlyTyrLeuGlnGlnTyrAlaGlyIleGlyTyr9095100CCGGCGAAGGTCCGCGCCCAGTACGACATGGTGGCGGTCGACCCCCGG595ProAlaLysValArgAlaGlnTyrAspMetValAlaValAspProArg105110115GGCGTGGCCCGCAGTGAACCCGTCGAGTGCCTGGACGGGCGCGAGATG643GlyValAlaArgSerGluProValGluCysLeuAspGlyArgGluMet120125130GACGCGTACACGCGCACCGACGTCACCCCGGACGACGCGGGCGAGACG691AspAlaTyrThrArgThrAspValThrProAspAspAlaGlyGluThr135140145GACGAGCTGGTCGACGCCTACAAGGAGTTCGCCGAGGGCTGCGGGGCG739AspGluLeuValAspAlaTyrLysGluPheAlaGluGlyCysGlyAla150155160165GACGCGCCGAAGCTGCTGCGCCACGTCTCCACGGTCGAGGCGGCACGC787AspAlaProLysLeuLeuArgHisValSerThrValGluAlaAlaArg170175180GACATGGACGTCCTGCGCGCGGTGCTGGGCGACGAGAAGCTGACCTAC835AspMetAspValLeuArgAlaValLeuGlyAspGluLysLeuThrTyr185190195GTGGGAGCGTCGTACGGCACCTTCCTGGGCGCGACCTACGCCGGTCTG883ValGlyAlaSerTyrGlyThrPheLeuGlyAlaThrTyrAlaGlyLeu200205210TTCCCCGACCGGACGGGCCGCCTGGTCCTGGACGGCGCGATGGACCCC931PheProAspArgThrGlyArgLeuValLeuAspGlyAlaMetAspPro215220225TCGCTGCCCGCCCGCCGCCTGAACCTGGAGCAGACGGAGGGCTTCGAG979SerLeuProAlaArgArgLeuAsnLeuGluGlnThrGluGlyPheGlu230235240245ACGGCGTTCCAGTCCTTCGCGAAGGACTGCGTGAAGCAGCCGGACTGC1027ThrAlaPheGlnSerPheAlaLysAspCysValLysGlnProAspCys250255260CCCCTCGGCGACAAGGACACCACCCCCGACCAGGTCGGCAAGAACCTC1075ProLeuGlyAspLysAspThrThrProAspGlnValGlyLysAsnLeu265270275AAGTCCTTCTTCGACGACCTGGACGCGAAGCCCCTGCCCGCCGGCGAC1123LysSerPhePheAspAspLeuAspAlaLysProLeuProAlaGlyAsp280285290GCCGACGGCCGCAAGCTCACCGAATCCCTCGCCACCACCGGCGTGATC1171AlaAspGlyArgLysLeuThrGluSerLeuAlaThrThrGlyValIle295300305GCCGCGATGTACGACGAGGGCGCCTGGCAGCAGCTGCGCGAGTCCCTC1219AlaAlaMetTyrAspGluGlyAlaTrpGlnGlnLeuArgGluSerLeu310315320325ACCTCGGCGATCAAGGAGAAGGACGGTGCGGGCCTGCTGATCCTCTCC1267ThrSerAlaIleLysGluLysAspGlyAlaGlyLeuLeuIleLeuSer330335340GACAGCTACTACGAGCGCGAGGCCGACGGCGGCTACAGCAACCTGATG1315AspSerTyrTyrGluArgGluAlaAspGlyGlyTyrSerAsnLeuMet345350355TTCGCCAACGCCGCCGTGAACTGCCTCGACCTCCCCGCCGCCTTCTCC1363PheAlaAsnAlaAlaValAsnCysLeuAspLeuProAlaAlaPheSer360365370TCCCCGGACGAGGTGCGCGACGCCCTCCCCGACTTCGAGAAGGCGTCC1411SerProAspGluValArgAspAlaLeuProAspPheGluLysAlaSer375380385CCGGTCTTCGGCGAGGGCCTCGCCTGGTCCTCCCTGAACTGCGCGTAC1459ProValPheGlyGluGlyLeuAlaTrpSerSerLeuAsnCysAlaTyr390395400405TGGCCGGTGAAGCCCACGGGGGAGCCGCACCGCATCGAGGCGGCCGGC1507TrpProValLysProThrGlyGluProHisArgIleGluAlaAlaGly410415420GCCACCCCGATCGTCGTGGTCGGCACCACCCGCGACCCGGCCACCCCC1555AlaThrProIleValValValGlyThrThrArgAspProAlaThrPro425430435TACCGCTGGGCCGAGGCCCTCTCCGACCAGCTCACCTCCGGCCACCTC1603TyrArgTrpAlaGluAlaLeuSerAspGlnLeuThrSerGlyHisLeu440445450CTCACCTACGAGGGAGACGGCCACACCGCGTACGGCCGCGGCAGCTCC1651LeuThrTyrGluGlyAspGlyHisThrAlaTyrGlyArgGlySerSer455460465TGCATCGACTCCGCGATCAACACGTACCTGCTGACCGGCACCGCCCCG1699CysIleAspSerAlaIleAsnThrTyrLeuLeuThrGlyThrAlaPro470475480485GAGGACGGCAAGCGCTGCTCGTAACCCCCGCCTGCCCGCCCCGGGACCCAC1750GluAspGlyLysArgCysSer490GCCTCCGGGGGCGGGTTCGGAGCACCCCGGGAAACTGTGTAGACTTGCCGACGTTGCTGA1810TCGCACCATGG1821(2) INFORMATION FOR SEQ ID NO:8:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 539 amino acids(B) TYPE: amino acid(D) TOPOLOGY: linear(ii) MOLECULE TYPE: protein(xi) SEQUENCE DESCRIPTION: SEQ ID NO:8:MetAspThrArgArgThrHisArgArgThrArgThrGlyGlyThrArg47-45-40-35PheArgAlaThrLeuLeuThrAlaAlaLeuLeuAlaThrAlaCysSer30-25-20AlaGlyGlyAlaSerThrSerAlaGlySerProAlaAlaLysAlaAla15-10-51GlyAlaThrGluAlaAlaThrAlaThrLeuThrProLeuProLysAla51015ThrProAlaGluLeuSerProTyrTyrGluGlnLysLeuGlyTrpArg202530AspCysGlyValProGlyPheGlnCysAlaThrMetLysAlaProLeu354045AspTyrAlaLysProAlaAspGlyAspValArgLeuAlaValAlaArg50556065LysLysAlaThrGlyProGlyLysArgLeuGlySerLeuLeuValAsn707580ProGlyGlyProGlyGlySerAlaIleGlyTyrLeuGlnGlnTyrAla859095GlyIleGlyTyrProAlaLysValArgAlaGlnTyrAspMetValAla100105110ValAspProArgGlyValAlaArgSerGluProValGluCysLeuAsp115120125GlyArgGluMetAspAlaTyrThrArgThrAspValThrProAspAsp130135140145AlaGlyGluThrAspGluLeuValAspAlaTyrLysGluPheAlaGlu150155160GlyCysGlyAlaAspAlaProLysLeuLeuArgHisValSerThrVal165170175GluAlaAlaArgAspMetAspValLeuArgAlaValLeuGlyAspGlu180185190LysLeuThrTyrValGlyAlaSerTyrGlyThrPheLeuGlyAlaThr195200205TyrAlaGlyLeuPheProAspArgThrGlyArgLeuValLeuAspGly210215220225AlaMetAspProSerLeuProAlaArgArgLeuAsnLeuGluGlnThr230235240GluGlyPheGluThrAlaPheGlnSerPheAlaLysAspCysValLys245250255GlnProAspCysProLeuGlyAspLysAspThrThrProAspGlnVal260265270GlyLysAsnLeuLysSerPhePheAspAspLeuAspAlaLysProLeu275280285ProAlaGlyAspAlaAspGlyArgLysLeuThrGluSerLeuAlaThr290295300305ThrGlyValIleAlaAlaMetTyrAspGluGlyAlaTrpGlnGlnLeu310315320ArgGluSerLeuThrSerAlaIleLysGluLysAspGlyAlaGlyLeu325330335LeuIleLeuSerAspSerTyrTyrGluArgGluAlaAspGlyGlyTyr340345350SerAsnLeuMetPheAlaAsnAlaAlaValAsnCysLeuAspLeuPro355360365AlaAlaPheSerSerProAspGluValArgAspAlaLeuProAspPhe370375380385GluLysAlaSerProValPheGlyGluGlyLeuAlaTrpSerSerLeu390395400AsnCysAlaTyrTrpProValLysProThrGlyGluProHisArgIle405410415GluAlaAlaGlyAlaThrProIleValValValGlyThrThrArgAsp420425430ProAlaThrProTyrArgTrpAlaGluAlaLeuSerAspGlnLeuThr435440445SerGlyHisLeuLeuThrTyrGluGlyAspGlyHisThrAlaTyrGly450455460465ArgGlySerSerCysIleAspSerAlaIleAsnThrTyrLeuLeuThr470475480GlyThrAlaProGluAspGlyLysArgCysSer485490(2) INFORMATION FOR SEQ ID NO:9:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 5 amino acids(B) TYPE: amino acid(D) TOPOLOGY: linear(xi) SEQUENCE DESCRIPTION: SEQ ID NO:9:GlyXaaSerXaaGly15(2) INFORMATION FOR SEQ ID NO:10:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 10 amino acids(B) TYPE: amino acid(D) TOPOLOGY: linear(ix) FEATURE:(A) NAME/KEY: Modified-site(B) LOCATION: 8(D) OTHER INFORMATION: /note= "Xaa at position 8represents Gln or Ser"(xi) SEQUENCE DESCRIPTION: SEQ ID NO:10:AspGlyHisGlyHisArgSerXaaAspAla1510(2) INFORMATION FOR SEQ ID NO:11:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 30 amino acids(B) TYPE: amino acid(D) TOPOLOGY: linear(xi) SEQUENCE DESCRIPTION: SEQ ID NO:11:ArgValAspLeuValGlyAsnSerPheGlyGlyAlaLeuSerLeuAla151015PheAlaIleArgPheProHisArgValArgArgLeuValLeu202530(2) INFORMATION FOR SEQ ID NO:12:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 381 amino acids(B) TYPE: amino acid(D) TOPOLOGY: linear(xi) SEQUENCE DESCRIPTION: SEQ ID NO:12:MetArgGlyLysLysValTrpIleSerLeuLeuPheAlaLeuAlaLeu151015IlePheThrMetAlaPheGlySerThrSerSerAlaGlnAlaAlaGly202530LysSerAsnGlyGluLysLysTyrIleValGlyPheLysGlnThrMet354045SerThrMetSerAlaAlaLysLysLysAspValIleSerGluLysGly505560GlyLysValGlnLysGlnPheLysTyrValAspAlaAlaSerAlaThr65707580LeuAsnGluLysAlaValLysGluLeuLysLysAspProSerValAla859095TyrValGluGluAspHisValAlaHisAlaTyrAlaGlnSerValPro100105110TyrGlyValSerGlnIleLysAlaProAlaLeuHisSerGlnGlyTyr115120125ThrGlySerAsnValLysValAlaValIleAspSerGlyIleAspSer130135140SerHisProAspLeuLysValAlaGlyGlyAlaSerMetValProSer145150155160GluThrAsnProPheGlnAspAsnAsnSerHisGlyThrHisValAla165170175GlyThrValAlaAlaLeuAsnAsnSerIleGlyValLeuGlyValAla180185190ProSerAlaSerLeuTyrAlaValLysValLeuGlyAlaAspGlySer195200205GlyGlnTyrSerTrpIleIleAsnGlyIleGluTrpAlaIleAlaAsn210215220AsnMetAspValIleAsnMetSerLeuGlyGlyProSerGlySerAla225230235240AlaLeuLysAlaAlaValAspLysAlaValAlaSerGlyValValVal245250255ValAlaAlaAlaGlyAsnGluGlyThrSerGlySerSerSerThrVal260265270GlyTyrProGlyLysTyrProSerValIleAlaValGlyAlaValAsp275280285SerSerAsnArgAlaSerPheSerSerValGlyProGluLeuAspVal290295300MetAlaProGlyValSerIleGlnSerThrLeuProGlyAsnLysTyr305310315320GlyAlaTyrAsnGlyThrSerMetAlaSerProHisValAlaGlyAla325330335AlaAlaLeuIleLeuSerLysHisProAsnTrpThrAsnThrGlnVal340345350ArgSerSerLeuGluAsnThrThrThrLysLeuGlyAspSerPheTyr355360365TyrGlyLysGlyLeuIleAsnValGlnAlaAlaAlaGln370375380(2) INFORMATION FOR SEQ ID NO:13:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 11 amino acids(B) TYPE: amino acid(D) TOPOLOGY: linear(xi) SEQUENCE DESCRIPTION: SEQ ID NO:13:AlaGluProXaaAlaValAspIleAspArgLeu1510(2) INFORMATION FOR SEQ ID NO:14:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 6 amino acids(B) TYPE: amino acid(D) TOPOLOGY: linear(xi) SEQUENCE DESCRIPTION: SEQ ID NO:14:AlaProAlaAlaProAla15(2) INFORMATION FOR SEQ ID NO:15:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 8 base pairs(B) TYPE: nucleic acid(C) STRANDEDNESS: single(D) TOPOLOGY: linear(xi) SEQUENCE DESCRIPTION: SEQ ID NO:15:AGCTAGCT8(2) INFORMATION FOR SEQ ID NO:16:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 10 amino acids(B) TYPE: amino acid(D) TOPOLOGY: linear(xi) SEQUENCE DESCRIPTION: SEQ ID NO:16:SerAlaGlyGlyAlaSerThrXaaAlaGly1510(2) INFORMATION FOR SEQ ID NO:17:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 15 amino acids(B) TYPE: amino acid(D) TOPOLOGY: linear(xi) SEQUENCE DESCRIPTION: SEQ ID NO:17:AlaProAlaAlaProAlaSerGlyGlySerSerAspGluAspLys151015(2) INFORMATION FOR SEQ ID NO:18:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 14 base pairs(B) TYPE: nucleic acid(C) STRANDEDNESS: single(D) TOPOLOGY: linear(xi) SEQUENCE DESCRIPTION: SEQ ID NO:18:GCGCCTGCAGCCTA14(2) INFORMATION FOR SEQ ID NO:19:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 22 base pairs(B) TYPE: nucleic acid(C) STRANDEDNESS: single(D) TOPOLOGY: linear(xi) SEQUENCE DESCRIPTION: SEQ ID NO:19:AGCTTAGGCTGCAGGCGCTGCA22(2) INFORMATION FOR SEQ ID NO:20:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 23 base pairs(B) TYPE: nucleic acid(C) STRANDEDNESS: single(D) TOPOLOGY: linear(xi) SEQUENCE DESCRIPTION: SEQ ID NO:20:GCGCCGGCGGCGCCTGCAGCCTA23(2) INFORMATION FOR SEQ ID NO:21:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 31 base pairs(B) TYPE: nucleic acid(C) STRANDEDNESS: single(D) TOPOLOGY: linear(xi) SEQUENCE DESCRIPTION: SEQ ID NO:21:AGCTTAGGCTGCAGGCGCCGCCGGCGCTGCA31__________________________________________________________________________

Claims

1. An isolated and purified tripeptidyl aminopeptidase of Streptomyces, wherein (A) said aminopeptidase is endogenous to Streptomyces, (B) said aminopeptidase cleaves an amino terminal tripeptide from a polypeptide, and (C) said terminal peptide has the sequence: X-Proline-Y, where X denotes an aliphatic or hydroxy amino acid and Y denotes an aliphatic, hydroxy, or sulfur-containing amino acid.

2. The aminopeptidase of claim 1, wherein said aminopeptidase has the amino acid sequence of SEQ ID NO:2.

3. The aminopeptidase of claim 1, wherein said aminopeptidase has the amino acid sequence of SEQ ID NO:4.

4. The aminopeptidase of claim 1, wherein said aminopeptidase has the amino acid sequence of SEQ ID NO:8.

5. Isolated and purified SlpD protein that has an approximate molecular weight of 55 kDa and comprises the amino acid sequence of SEQ ID NO:8 from amino acid 17 to amino acid 492.

6. An isolated and purified protease, wherein said protease has the amino acid sequence of SEQ ID NO:6.

7. Isolated and purified SlpE protein that has an approximate molecular weight of 56 kDa and comprises the amino acid sequence of SEQ ID NO:6 from amino acid 22 to amino acid 512.

8. An aminopeptidase according to claim 1, wherein said aminopeptidase cleaves an amino terminal tripeptide from a polypeptide, and wherein said terminal peptide is selected from the group consisting of APA, APM, APS, GPL, and SPA.

9. An aminopeptidase according to claim 1, wherein said aminopeptidase cleaves the amino terminal tripeptide APA from a polypeptide.

10. An aminopeptidase according to claim 1, wherein said aminopeptidase cleaves the amino terminal tripeptide GPL from a polypeptide.

11. A polypeptide comprising, in the order of translation, a signal peptide, a propeptide comprising at least one tripeptide, and the amino acid sequence of a mature protein, wherein said amino acid sequence has a structural constraint that would inhibit cleavage of said signal peptide by a signal peptidase in the absence of said propeptide, and wherein said tripeptide has the sequence: X-Proline-Y, and wherein X denotes an aliphatic or hydroxy amino acid and wherein Y denotes an aliphatic, hydroxy, or sulfur-containing amino acid.

12. A polypeptide according to claim 11, wherein said tripeptide is selected from the group consisting of APA, APM, APS, GPL, and SPA.

13. A polypeptide according to claim 11, wherein said tripeptide is APA.

14. A polypeptide according to claim 11, wherein said tripeptide is (APA).sub.2.

15. A polypeptide according to claim 11, wherein said tripeptide is GPL.

16. A method for improving the secretion of a mature protein by a transformed host cell, comprising the steps of:

(a) providing a host cell transformed with a nucleic acid expression construct which encodes a first polypeptide that comprises, in the N-terminal to C-terminal direction:

(i) a host-specific signal peptide,

(ii) at least one tripeptide, wherein said terminal peptide has the sequence: X-Proline-Y, and wherein X denotes an aliphatic or hydroxy amino acid and wherein Y denotes an aliphatic, hydroxy, or sulfur-containing amino acid, and

(iii) a second polypeptide sequence encoding said mature protein;

(b) culturing the transformed host cell of step (b) under conditions suitable for the expression of said first polypeptide and for the secretion of a third polypeptide which comprises at least one tripeptide of step (a) (ii) and the second polypeptide of step (a)(iii); and

(c) exposing said third secreted polypeptide to a Streptomyces tripeptidyl aminopeptidase, whereby said peptidase cleaves said tripeptide from said third polypeptide to produce the mature protein, and wherein said aminopeptidase is endogenous to Streptomyces.

17. A method of claim 16, wherein the tripeptide is Ala-Pro-Ala.

18. A method of claim 16, wherein the tripeptide is (Ala-Pro-Ala)-.sub.2 (SEQ ID NO:14).

19. A method according to claim 16, wherein said tripeptide is selected from the group consisting of APA, APM, APS, GPL, and SPA.

20. A method according t o claim 16, wherein said tripeptide is GPL.