CA2156425C - Method and system for enhanced production of commercially important exoproteins in gram-positive bacteria - Google Patents

Abstract

The invention provides a method and expression system for enhancing secretion of hyperproduced homologous and heterologous exoproteins in gram-positive bacteria such as Bacillus sp. The method and system comprise overproduction of PrsA protein in a grampositive bacterial host capable of also overproducing at least one exoprotein of interest. Use of the method and system of the invention results in greatly enhanced secretion of the synthesized exoproteins into the growth medium. Once in the growth medium these secreted exoproteins can be recovered and purified in a straightforward manner.

Description

~~~ ~4 z 5 Method and System for Enhanced Production of Commercially Important Exoproteins in Gram-Positive Bacteria Field of the Invention This invention relates to a method and expression system for enhanced production of industrially and medically important exoproteins in gram-positive bacteria, especially species of the genus Bacillus.
Background In gram-positive bacteria secreted proteins are exported across a cell membrane and a cell wall, and then are subsequently released into the external medium. On the other hand, gram-negative bacteria are surrounded by two cell (or surface) membranes; they have no cell wall. In gram-negative bacteria, most exported proteins are not released from the cell but stay in the inter-membrane periplasmic space and in the cuter membrane.
Two types of components of the secretion machinery have been identified in E. coli: soluble cytoplasmic proteins and membrane associated proteins (see for review, Wickner et al., (1991 ) Annu. Rev. Biochem., 60:101-124). Soluble cytoplasmic proteins, including SecB and heat shock proteins, all prevent the folding of precursors of secreted proteins into a conformation incompatible with secretion. The set of membrane-associated proteins includes the peripheral membrane protein SecA, integral membrane proteins Sect, SecE, SecD, SecF and the signal peptidases Lep and Lsp (reviewed in Hayashi, S. and Wu, H.C. (1990) J. Bioenerg. Biomembr., 22:451-471; Dalbey, R.E. (1991 ) Mol. MicrobioL, 5:2855-2860). These membrane-associated proteins are involved in binding of the precursor and in its translocation across the cytoplasmic membrane, followed by cleavage of the signal peptide and release of the protein.
Knowledge on the secretion machinery of gram-positive bacteria is more limited. The available data on B. subtilis, the genetically and physiologically well characterized model organism of the genus, suggest an SUBSTITUTE SI-IEET

~1~ 6 4~5 overall similarity with that of E, coli, but also differences in the structure and specificity of individual components, possibly reflecting demands set by the very different composition and architecture of the respective cell envelopes.
Gram-positive bacteria such as B. subtilis, B. amyloliquefaciens, B. licheniformis have a very high capacity for~secreting proteins, and indeed, many bacillar extracellular enzymes are utilized industrially. Since secreted proteins in gram-positive bacteria are so important commercially, and since the gram-positive secreted proteins traverse through a cell envelope with a very different structure from that of E. coli, the molecular mechanisms of protein secretion in gram-positive bacteria is of considerable academic and industrial importance.
In this regard we recently discovered a novel component, the PrsA protein, of the secretion machinery of B, subtilis (Kontinen, V.P. and Sarvas, M., (1988) J. Gen. Microbiol., 134:2333-2344; Kontinen, V.P., et al., (1991 ) Mol. Microbiol. 5:1273-1283). The prsA gene, which encodes the PrsA protein, was initially defined by nonlethal mutations that decreased the secretion of several exoproteins (Kontinen, V.P. and Sarvas, M., (1988) J.
Gen. Microbiol., 134:2333-2344). Based on the DNA sequence of the cloned prsA gene and our subsequent work with this gene and protein, we assert that prsA encodes a protein (PrsA) that acts as a chaperone, and is translocated across the cytoplasmic membrane (for the initial work, see Kontinen, V.P., et al., ( ~ 991 ) Mol. Microbiol. 5:1273-1283). The PrsA
protein has been found to possess a limited amount of sequence homology (about 30%) with the PrtM protein of Lactococcus lactis, a protein proposed to assist the maturation of an exported serine protease (Haandrikman, A.J., et al, (1989) J. BacterioL, 171:2789-2794; Vos, P., et al., (1989) J. BacterioL, 171:2795-2802). A similar function has not been associated with other known proteins of the secretion machinery of bacteria, suggesting that PrsA protein is a novel type of component in the pathway of protein secretion facilitating the release and probably folding of secreted proteins after their translocation across the cytoplasmic membrane in gram-positive bacteria.
SUBSTITUTE SHEET

WO 94/19471 _ PCT/FI94/00072 It is advantageous to produce proteins of interest in bacteria in secreted form, since exported proteins usually maintain their native conformation, in contrast to intracellular production which, in many cases, results in aggregation of the produced protein. Another advantage of producing industrially and medically important proteins in bacteria in secreted form is that secretion facilitates purification of the protein product.
Additionally, unlike E. coli, gram-positive bacteria such as Bacillus sp. do not contain toxic compounds like lipopolysaccharide, making them especially appropriate hosts for production of medical and pharmaceutical proteins.
Increased yield of secreted proteins would be of great significance for improving the gram-positive bacillar strains used in the industrial production of a number of exoenzymes, such as alpha-amylases, proteases and lipases. The strategy thus far has been to overexpress the appropriate gene. There are known and readily available methods for doing this, such as increasing gene expression by using multicopy plasmids or enhancing the activity of the gene by modifying its regulatory elements, e.g., by using strong promoters, or multiple promoters. Dramatic increases of the synthesis of exoproteins have been achieved this way, up to a level at which increasing the synthesis is of no further benefit because of bottlenecks in the secretion machinery. It would be desirable to increase the capacity of secretion in parallel with increased synthesis. However, to date this has not been possible.
It is an object of the present invention to alleviate the bottleneck of the secretion mechanism in gram-positive bacteria, and to provide a method and a system whereby the levels of proteins normally secreted from gram-positive bacteria such as Bacillus can be enhanced when the expression of a given homologous or heterologous protein of interest has been elevated over the amount normally produced in unmodified or wild type organisms.
It is a further object of the present invention to describe bacterial hosts and plasmids which can be used to enhance the production of a variety of commercially important exoproteins.
SUBSTITUTE SHEET

Summary The invention provides a method and expression system for enhancing the levels of homologous or hete,~ologous proteins) normally secreted from gram-positive bacteria (such as Bacillus sp.) when expression 5 of the homologous or heterologous proteins) has been elevated over <.
unmodified or wild type amounts produced'by unmodified or wild type organisms.
The method and system of our invention comprise overproduction of PrsA protein, or a functional homologue thereof, in a gram 10 positive bacterial host capable of also overproducing at least one homologous or heterologous exoprotein of interest. According to the teaching of the invention, overproduction means an amount greater than wild-type, i.e., more than the amount of the protein (PrsA or a functional homologue thereof, or exoprotein of interest) normally produced by wild type bacteria. Also 15 according to the invention, overproduction is accomplished by standard means known to the art, e.g., use of multicopy plasmids, multiple copies of the genes encoding PrsA, or a functional homologues) thereof, and/or the exoprotein of interest, in the chromosome of the host, combined with altering the regulatory elements to increase expression, e.g., use of strong 20 promoter(s), use of multiple promoters, use of enhancers, and so forth. Use of the method and system of the invention results in greatly enhanced secretion, e.g., as much as five to ten fold over controls, of synthesized exoproteins into the growth medium. Once in the growth medium these secreted exoproteins can be recovered and purified in a straightforward 25 manner.
The expression system of the invention comprises a host gram-positive bacteria, e.g., species of Bacillus, capable of expressing greater than wild-type amounts of PrsA protein, or a functional homologue thereof, and greater than wild-type amounts of an exoprotein of interest, e.g., alpha amylase, subtilisin, pneumolysin, lipases, or other exoproteases of commercial interest. The method of the invention comprises using this expression system to enhance production of commercially important SUBSTITUTE SHEET

WO 94/19471 _ PCT/FI94/00072 exoproteins in gram-positive bacteria. According to the method, at least one exoprotein of interest is overexpressed in a host gram-positive bacteria which also overexpresses (i.e., expresses greater than the amounts produced by wild type bacteria) PrsA protein, or a functional homologue thereof.
5 According to the teaching of the invention, a functional homologue of PrsA
protein is a protein which when overexpressed is capable of enhancing the secretory capability of a gram-positive bacteria with respect to secretion of an exoprotein of interest. Also according to the teaching of the invention, a functional homologue of PrsA can be identified by several means including sequence homology to prsA or PrsA, immunological reaction with anti-PrsA
antibodies) of high titer, and/or functionally, i.e., as a protein which when overexpressed, is capable of enhancing the secretory capability of a gram-positive bacteria with respect to secretion of an exoprotein of interest.
A preferred means for transforming host gram-positive bacteria, such as species of Bacilli, so they produce greater than wild-type amounts of PrsA protein is to transform the host with plasmid pKTH277 which carries the prsA gene from Bacillus subtilis. Comparable plasmids can be constructed to carry genes which encode functional homologues of the PrsA protein. These plasmids can be used to transform host gram-positive bacteria so they overproduce the functional homologues of PrsA. Once engineered to overproduce PrsA homologues (which can also be referred to as PrsA-like proteins), these host gram-positive strains can be used, according to the teaching of the invention, for enhanced secretion of hyperproduced exoproteins of interest.
The present invention also discloses, and includes, methods and constructs related to our discovery that secretion in gram-positive bacteria can be enhanced by increasing the amount of cellular PrsA protein, or functional homologues) thereof, in gram-positive hosts that express greater than wild-type amounts of exoproteins of interest.
In one aspect our invention includes an expression system for enhancing secretion of exoproteins in gram-positive bacteria comprising a gram-positive bacteria capable of expressing greater than wild-type amounts SUBSTITUTE SHEET

21~ 6 X25 of PrsA protein, or functional homologues) thereof, and capable of expressing greater than wild-type amounts of at least one exoprotein of interest.
In another aspect our invention includes a gram-positive bacteria able to express greater than wild-type amounts of at least one exoprotein of interest further comprising pKTH277.
In yet another aspect, our invention includes a gram-positive bacteria able to express greater than wild-type amounts of at least one exoprotein of interest and further comprising at least one of the following:
at least two copies of the prsA gene from Bacillus subfilis, or a functional homologue thereof; the prsA gene from Bacillus subtilis, or a functional homologue thereof, operatively linked to strong regulatory sequences which result in overexpression of said prsA gene, or functional homologue thereof.
Our invention also includes a DNA construct comprising the prsA gene from Bacillus subtilis, or a functional homologue thereof, under the control of expression signals which cause overexpression of said prsA gene, or functional homologue thereof, plus a vector further comprising the prsA
gene from Bacillus subtilis, or a functional homologue thereof, under the control of expression signals which cause overexpression of said prsA gene, or functional homologue thereof.
In yet another aspect our invention includes a method for enhancing secretion of an exoprotein of interest in a gram-positive bacteria comprising expressing greater than wild type amounts of PrsA protein from Bacillus subtilis, or a functional homologue thereof in the gram-positive bacteria, wherein the gram-positive bacteria is also capable of expressing greater than wild type amounts of the exoprotein.
Still further the invention includes a method for creating an improved non-Bacillus subtilis gram-positive host organism useful for enhanced secretion of an exoprotein of interest that is overexpressed in the host organism, the method comprising (a) identifying a gene from the host organism that encodes a functional homologue of PrsA protein from Bacillus subtilis, and (b) enhancing the expression of the gene identified in step (a) by SUBSTITUTE SHEET

WO 94/19471 ~ ~ ~ PCT/FI94/00072 at least one of the following: introducing into the host organism at least one additional copy of the gene; introducing into the host organism the gene operatively linked to expression sequences which result in overexpression of the gene.
The invention also includes a method for identifying a gene which encodes a functional homologue of PrsA from Bacillus subtilis, the method comprising identifying, by means of Southern blotting, DNA which hybridizes with DNA probes) from the prsA gene from Bacillus subtilis, and demonstrating that the gene encodes a protein which when overexpressed, is capable of enhancing the secretory capability of a gram-positive bacteria with respect to secretion of an exoprotein of interest.
Still further, the invention includes a method for identifiying a gene which encodes a functional homologue of PrsA from Bacillus subtilis, this method comprising identifying protein that -reacts with anti-PrsA
antibodies) of high titer, and demonstrating that when the protein is present in greater than wild-type amounts in a gram-positive bacteria, the protein is capable of enhancing the secretory capability of the gram-positive bacteria with respect to secretion of an exoprotein of interest.
These and other features, aspects and advantages of the invention will become better understood with reference to the following description, examples, methods and materials, and appended claims.
Description When the expression level (synthesis) of an exported protein is high in gram-positive bacteria such as Bacillus sp., the capacity of the secretion apparatus is a limiting factor in protein secretion and production of these proteins in secreted form. Our invention provides a system and method for overcoming this limitation or bottleneck.
Our invention is based on our initial surprising discovery that secretion in gram-positive bacteria such as species of Bacillus can be enhanced by increasing the amount of only one component of the bacillar export machinery, i.e., the amount of cellular PrsA protein, or functional SUBSTITUTE SHEET

~lr~ ~ X25 homologues thereof, in gram-positive bacterial hosts that express greater than wild-type amounts of exoproteins of interest. The method and system of the invention are useful regardless of how the proteins of interest are overproduced in the gram-positive bacterial host. Thus the method and system can be used to improve a variety of overproducing commercial strains now used in industrial applications.
The method and system of the invention can be used with any gram-positive bacteria. Bacteria of the genus Bacillus are preferred.
Especially preferred are Bacillus alkalophilus, Bacillus amyloliquefaciens, Bacillus brevis, Bacillus circulans, Bacillus coagulans, Bacillus lautus, Bacillus lentus, Bacillus licheniformis, Bacillus stearothermophilus, Bacillus subtilis, and Bacillus thuringiensis.
The method and system of the invention can also be used with any desired exoprotein of interest. Examples of exoproteins of interest that 75 may be produced according to the method and system of the invention are listed below, where the exemplary exoproteins are presented by general categories.
Antigenic proteins of microbes and protozoa: Capsule, outer membrane and fimbria proteins from any gram negative bacteria, but especially those from: Bacteroides fragilis, Fusobacterium spp., Bordetella pertussis, Haemophilus influenzae, Yersinia entercolitica, Yersinia pestis, Branhamellla catarrhalis, Escherichia coli, Klebsiella pneumonia, Vibrio cholerae, Proteus mirabilis, Pseudomonas aeruginosa, Serratia marcescens, Legionella pneumophila, Neisseria gonorrhoeae, Neisseria meningitides, Salmonella typhimurium, Salmonella typhi, Salmonella pararyphi 8.
Mycobacterium tyberculosis, Chlamydia trachomatis, and Shigella spp.
Protein toxins or secreted proteins from any bacteria, but especially those from: Staphylococcus aureus, Pseudomonas aeruginosa, Clostridium spp., Escherichia coli, Yersinia pestis, Vibrio cholerae, Bordetella pertussis, M-Protein of the Streptococcus pyogenes bacterium, Excreted enzymes of Stretococcus mutans.
SUBSTITUTE SHEET

2'15642 WO 94119471 ' PCT/FI94/00072 Surface proteins of any microorganism, but especially those from the following microorganisms (in all phases of development):
Plasmodium spp., Toxoplasma spp., Leishmania spp., Schistosoma spp., Trypanosama spp. Adhesion protein of Streptococcus sp., and adhesion protein of Staphylococcus aureus.
Antigen proteins or viruses: HA and NA proteins of myxoviruses (influenza A H1-H12, influenza B, influenza C): HN and F proteins of paramyxoviruses (parainfluenze 1-4, Newcastle disease virus, Measles virus, Respiratory syncytial virus, Parotitis virus, Distemper virus): G protein of Rabies virus; E1 and E2 proteins of alfaviruses (Chikungunya, Western, Easter, Venezuelan equine encephalitis virus, O'nyong-nyong virus, Semliki Forest virsu, Sindbis virus); V1 and V3 proteins of flavin viruses (Dengue 1-4, Japanese encephalitis virus, Mite encephalitis viruses, Murray Valley encephalitis virus, Kyasanur Forest disease virus, Looping ill virus, Omsk hemorrhagic fever virus); surface proteins of German measles virus; surface proteins of Hog Cholera virus; surface proteins of Equine arthritisvirus; G1 and G2 proteins of Bunya viruses (Rift Valley fever virus, Crimean hemorrahagic fever virus, California encephalitis virus, Phlebotomus fever virus); G1 and G2 proteins of arena viruses (Lassa fever virus, Lymphocytic chorion meningitis virus); proteins V1-V4 of picorna viruses (polio 1-3, Coxsackie A viruses 1-24, Coxsackie B viruses 1-6, ECHO viruses 1-8, 11-34, hepatitis A virus, hepatitis B virus, hepatitis C virus, Human rhino viruses 1-113); surface proteins of rota viruses; surface proteins of herpes viruses (HSV 1, 2, Cytomegalo virus, Epstein-Barr virus, Equine abortion virus); VP1-VP3 proteins of papova viruses (BK virus, Human wart virus); p proteins of parvo viruses (mink enteritis virus, Bovine panro virus, Feline parvo virus, Procine parvo virus); structure proteins of Human hepatitis B virus; surface proteins of Ebola and Marburg viruses; and Hexone, pentone and fiber proteins of adeno viruses, (Human adeno viruses 1-33).
Industrial enzymes: With regard to industrially important enzymes, such enzymes may be amylolytic, lipolytic and proteolytic enzymes, carbohydrases, transferases, isomerases, peroxidases, oxidoreductases, SUBSTITUTE SHEET

215f~~2~5 _.
,o oxidases etc. More specifically, the enzyme of interest may be a protease, a lipase, a cutinase, an amylase, a galactosidase, a pullulanase, a cellulase, a glucose isomerase, a protein disuphide isomerase, a CGT'ase (cyclodextrin gluconotransferase), a phytase, a glucose oxidase, a glucosyl transferase, laccase, bilirubin oxidase, or a xylanase. Examples include, but are not limited to: alpha-amylase, amino acid acylase, amyloglucosidase, bromelain, phisine, beta-galactosidease, beta-gulcanase, glucose-ismorase, glucoseoxidase, hemicellulase, invertase, catalase, collagenase, xsylanase, lactase, lipase, naringinase, pancreatin, papain, pectinase, penicillinamidase, pepsin, protease, pullulanase, isoamylase, rennin, ribonuclease, cellulase, streptokinase and trypsin.
Exoproteins of medical interest can also be produced. Such proteins include diagnostic antigens, proteins that can be used as vaccines, and pharmaceuticals.
According to the teaching of the invention, the exoproteins of interest need not be native exoproteins, but instead can be novel proteins that have been designed and created to be exoproteins using genetic engineering techniques. For example, a normally non-secreted protein from one species (or an engineered non-native protein) can be engineered to be an "exo"
protein by adding a signal sequence to the sequence encoding the structural protein. This engineered exoprotein can be expressed in a gram-positive bacteria such as a species of Bacillus, which overexpresses PrsA protein, or a functional homologue thereof. In this way the method of the invention can be used to enhance secretion of these non-native or engineered proteins of interest.
Turning now to aspects of our invention, to illustrate one embodiment of our invention we show the effect overexpression of the prsA
gene from Bacillus subtilis has on the secretion of the two important industrial exoenzymes of Bacilli, alpha-amylase and subtilisin. For these studies, a 5.3 kB insert, containing the entire prsA gene from Bacillus subfilis, was cloned into a low copy number shuttle plasmid (pICTH277), which was then used to introduce additional copies of prsA into Bacillus subtilis. (The DNA and SUBSTITUTE SHEET

deduced amino acid sequences of the prsA gene from Bacillus subtilis appear in the EMBUGenBank/DDBJ Nucleotide Sequence Data Libraries under the accession number X57271.) (pKTH277 was obtained by ligating the 5.3 kB EcoRl-BamHl fragment from pKTH268 with low copy number shuttle plasmid pHPl3 linearized by digestion with respective restriction enzymes and transforming into the E, colt strain TG1. The sizes of pKTH268 and pKTH277 are 8.5 kB and 10.2 k8, respectively. See also Kontinen, et al., (1991 ) Moi: Microbiol., 5:1273-1283).
The presence of pKTH277 in 8. subtilis increased the amount of the protein corresponding to the PrsA protein by approximately 10-fold over the wild type. When the genes for different secreted proteins are expressed in strains of Bacillus containing these increased levels of PrsA protein, the level of protein secreted into the culture medium is increased substantially. For example, the secretion of alpha-amylase of Bacillus amyloliquefaciens was found to be increased by 2.5-fold in this system, the secreted level of the thermoresistant alpha-amylase of Bacillus licheniformis was elevated by six-fold, and subtilisin (alkaline protease) from Bacillus licheniformis was secreted at two times the control level.
In these studies the exoenzymes were overexpressed in host strains in amounts likely to saturate the secretion machinery, either by placing the gene which encoded the exoprotein on a multicopy plasmid or inserting it in the chromosome of the host. (In these studies, alt multicopy plasmids coding for the exoenzymes were derivatives of pUB 110, which belongs to a different incompatibility group than the shuttle plasmid pKTH277, allowing their replication in the same host cell. The stability of these piasmids was further increased in most cases by using a recE4 host strain, which prevents efficient recombination between homologous sequences (Dubnau et ai., 1973; Keggins et al., 1978).
The first exoenzyme studied to illustrate this aspect of our invention was alpha-amylase of B. amyloliquefaciens (AmyE) encoded by pKTH 10 (Palva, I. ( 1982a) Gene, 19:81-87; Palva, I., et al., { 1982b) Proc.
NatL Acad. Sci. USA, 79:5582-5586). We found that in the wild-type strain WO 94119471 " PCTIFI94/00072 hyperproducing this alpha amylase, the presence of pKTH277 indeed enhanced the secretion of alpha-amylase throughout the stationary phase of growth, about 2.5-3 fold over the level of the control strain which did not overexpressing PrsA. The highest concentration.of alpha amylase in the culture supernatant (about 3400 microgramslml) was found after the growth of 24 hours. In the absence of pKTH277, the strain secreted only 1200 micrograms/ml. 4ualitatively similar results were obtained when alpha-amylase was expressed from one copy of the amyE gene, which was inserted in the chromosome and transcribed at high level due to modified regulation.
The second exoprotein we tested was the thermoresistant alpha-amylase of B. licheniformis (Amyl), the major liquefying alpha-amylase of industrial importance (Ortlepp et al., 1983; Diderichsen, B., et al., (1991 ) Res. Microbiol., 142,. 793-796). Secretion of this enzyme in B. subtilis at amounts comparable with those of the alpha amylase of 8. amyloliquefaciens was achieved by expressing the appropriate gene from the secretion vector based on the promoter and signal sequence of the gene of the latter enzyme (Palva, I. (1982) Gene, 19:81-87; Palva, I., etal., (1982) Proc. Natl. Acad.
Sci.
USA, 79:5582-5586); Sibakov, M. (1986) Eur. J. Biochem., lsS, 577-581 ).
Introduction of pKTH277 into one such strain (to result in IH6760) increased the amount of alpha-amylase in the culture medium about six fold, with the same difference seen from late exponential stage to cultures of 45 hours.
Alkaline protease, subtilisin, is a different type of exoprotein, whose precursor contains in addition to the signal sequence a further extension, the prosequence (Wells, et al., (1983) Nucleic Acids Res., 11:7911-7925; Wong, S.-L., etal., (1984) Proc. NatL Acad. Sci. USA, 81:1884-11188). The effect of an increased amount of PrsA on this secretion was studied by comparing two strains, one with increased level of PrsA
(IH6789), another one with the wild type level. Both secreted the heterologous subtilisin of 8. licheniformis (SubC, which is used as a laundry powder and is an important industrial product) coded by the multicopy plasmid pMJ57 (Hastrup, S. and Jacobs, M.F. (1990) In Zukowski, M.M., et SUBSTITUTE SHEET

WO 94/19471 2 ~. ~ 6 4 2 5 PCT/FI94/00072 aG, (eds.), Lethal phenotype conferred by xylose-induced overproduction of apr-IacZ fusion protein, vol. 3. Academic Press, Inc., San Diego, California, pp. 33-41 ). In this plasmid the subC gene is under the control of a xylose inducible promoter. Comparison of the secretion of subtilisin from the two strains, when fully induced; showed that its amount in the culture supernatant of IH6789 (increased amount of PrsA) was about twofold higher than that of the control IH6788 at all time points assayed.
We also studied the effect of pKTH277 on the natural low level secretion of endogenous exoenzymes in a strain devoid of any plasmid causing hypersecretion. The amount of secreted alpha amylase and total proteases in the late exponential phase of growth or in overnight cultures was the same in strains carrying pKTH277 or the cloning vector pHPl3. Based on these results it appears that the increased amount of PrsA protein enhances secretion of hyperproduced exoenzymes only.
In order to confirm the role of PrsA in the enhancement caused by the 5.3 kb fragment in the above plasmids, we inactivated the prsA gene in the plasmid pKTH277 by insertions in the EcoRV site of its prsA gene (at the nucleotide 382). In pKTH3261 the insert was a 560 by fragment of the blaP
gene of 8. licheniformis flanked by translational stop codons, and in pKTH3262 a 4.6 kb EcoRV fragment of phage lambda. SDS-PAGE analysis of whole cell proteins of E. coli carrying these plasmids showed no full-size PrsA protein expressed by either plasmid, and a putative truncated PrsA of the expected size (14 kDa) expressed by pKTH3261 (data not shown). As a control, we constructed pKTH3253 in which the 5.3 kbp fragment was truncated for 1.9 kbp downstream of prsA, leaving this gene intact. In B.
subtilis (IH6624 carrying pKTHlO) the two plasmids with an insertion in prsA
did not enhance the secretion of alpha amylase, while pKTH3253 did.
Enhanced secretion obtained by overproduction of prsA is of obvious advantage for large scale industrial production of exoenzymes. In such applications it is sometimes desirable to avoid the use of potentially unstable multicopy plasmids. One strategy is to insert one or few copies of the structural gene of the exoenzyme in the chromosome, combined with SUBSTITUTE SHEET

altering its regulatory elements to increase expression. We therefore tested the effect of PrsA overproduction on the secretion of alpha amylase in such a system, where one copy of amyE was inserted in the chromosome fused to the target sequence of the regulatory protein DegQ i~ a strain overexpressing DegQ. Also in this strain the high level of alpha-amylase secretion was enhanced abouttthree fold by increasing the amount of PrsA protein (Table 1, strains BRB764 and IH67703). This indicates that the enhancement of secretion is achieved when the starting level of expression of exoprotein is high, regardless of the way the increased expression of the target gene has been obtained.
Turning now to the presence of PrsA in gram-positive bacteria other than Bacillus subtilis, we have confirmed the presence of PrsA or PrsA
homologues in other species, e.g., Bacillus amyloliquefaciens and Bacillus licheniformis. The amount of PrsA protein in Bacillus amyloliquefaciens is similar to the amount in Bacillus subtilis cells, while the amount of PrsA
protein in Bacillus licheniformis cells appears to be less. In addition, the components of the secretion machinery in these gram-positive bacillar strains are similar to that of Bacillus subtilis. Bacillus amyloliquefaciens and Bacillus licheniformis are two of the most widely used species of Bacillus in large scale industrial processes for the production of secreted proteases and amylases.
The method of the invention using overproduction of PrsA protein to increase secretion of homologous and heterologous exoproteins of interest is especially useful in these strains.
As part of our invention we also teach how PrsA protein and/or the prsA gene can be identified in other gram-positive bacteria, and that functional homologues of the PrsA protein from Bacillus subtilis exist and can be used in the method and system of the invention. Functional homologues of the PrsA protein from Bacillus subtilis can be identified in other gram-positive species by using anti-PrsA antibody of high titer. Alternatively, the prsA gene or homologous prsA-like genes which encode functional homologues of the PrsA protein from Bacillus subtilis can be identified by Southern blotting using probes) from the prsA gene, or a prsA-gene r-.-fragment. If the PrsA-like protein is found to exist, but there is insufficient homology to the prsA gene or PrsA protein to be detected unequivocally with antibodies specific for the PrsA protein, or DNA probes containing sequences homologous to the prsA gene, the homologous gene can be located and 5 cloned, so unequivocal identification can be made.
However, in most cases homologous proteins and genes can be found using antibodies specific for the PrsA protein, or DNA probes containing sequences homologous to the prsA gene. For immunological identification, immunoblotting (Western blotting) can be used to detect the 10 PrsA protein with anti-PrsA antibody of high titer. The antiserum is produced by immunizing an appropriate animal (e.g., rabbit) with PrsA protein of B.
subtilis, or preferably with a PrsA protein homologue of another species more closely related to the species of interest than B. subtilis. To identify the PrsA, the bacterium of interest can be grown on a number growth media, but 15 preferably, the bacterium is grown on a medium where there is minimal induction of proteases. Bacterial cells are collected, again preferably at a growth phase with minimal amount of proteases present, and broken with a method appropriate for the species (usually a combination of enzymatic treatment and mechanical disrupture with sonication, French pressure cell, or shearing with glass beads). Samples of various sizes of broken cells and particulate fraction of the disrupted whole cells, prepared with ultracentrifugation, are electrophoresed in SDS-PAGE with standard methods, proteins are transferred to membrane filters and detected with anti-PrsA antiserum and the labeled second antibody. (Smaller amounts of PrsA
protein can be detected if the particulate fraction is prepared). In all these steps standard methods and commercial reagents can be used.
To identify the prsA gene, or a prsA-like gene that encodes a functional homologue of PrsA in a species of interest, Southern blotting can be used. In this method, appropriate DNA probes from the prsA gene are hybridized to/with appropriately fragmented and electrophoresed chromosomal DNA of the species of interest according to the standard method of Southern hybridization. The hybridization probe may be any SUBSTITUTE SHEET

WO 94/19471 ~~ ~ PCT/FI94/00072 fragment of DNA containing the prsA gene of B. subtilis, or a fragment of this gene, or a DNA fragment containing the prsA gene homologue of another species or a fragment of that gene. Once identified, the prsA gene homologue can be sequenced to further confirm;.ifs identity.
The teaching of the present invention includes not only overexpression of the PrsA protein of Bacillus subtilis, but also overexpression of a functional homologue of the PrsA protein from Bacillus subtilis in a gram-positive species of interest. According to the teaching of the invention, either the prsA gene of B. subtilis or the prsA gene homologue from another species, including the host species, is introduced into the host species. The prsA gene or its homologue is brought under the control of expression signals which are active in the species of interest in order to result in the high level (but not lethal) expression of the prsA gene. This can be accomplished in a variety of ways, including:
(1 ) The transfer of the plasmid pKTH277 to the species of interest. The transfer can take place with any method of transformation applicable to that species, like transformation, transduction, protoplast transformation, electroporation or conjugation. pKTH277 is maintained as multiple copy plasmid in many other gram-positive species other than Bacillus, and the expression signals of the prsA gene in that plasmid are active in many gram-positive species.
(2) Inserting the 5.3 kb Sad fragment of pKTH277 into any other plasmid compatible with the species of interest and maintaining it at suitable copy number for high, but not lethal level of expression of prsA, relative to the activity of the prsA gene of 8. subtilis in that species. The plasmid is then transferred to that species using any method of transformation applicable to that species. Alternatively, the fragment of DNA inserted into the plasmid can contain a prsA gene homologue of another species with its expression signals.
(3) Inserting the DNA fragment of pKTH277 encoding the signal sequence and the mature part of the PrsA protein to an expression vector suitable for the species of interest, under the control of expression signals in SUBSTITUTE SHEET

WO 94/19471 " PCT/FI94/00072 that plasmid to achieve high level expression of prsA. As above, the appropriate fragment may derive from a prsA gene homologue of another species.
(4) The DNA constructions of paragraphs (2) and (3) above can be inserted into the chromosome of the species of interest instead of a plasmid. In that case, expression signals have to be chosen which are active enough to ensure high level expression of PrsA although there is only one copy of the gene per the genome.
As inventors who are also basic scientists, by design and of necessity much of our work is done under laboratory or simulated industrial co~~ditions. However, with the help of industrial collaborators, it has been shown that the method and system of our invention work very well with commercially useful bacterial stains, under industrial fermentation conditions.
Methods and materials used in our studies, and examples of our invention are included below to further aid those skilled in the art in practicing the method and system of our invention.
Materials and Methods Bacterial strains and plasmids prs mutants of Bacillus subtilis Marburg 168 and their parent strains are shown in Table 1. Listed are also 8. subfilis and E. coli strains overexpressing PrsA protein and B. subtilis strains with enhanced secretion of exoenzymes due to increased cellular amount of PrsA protein and their appropriate control strain. E. coli strains used as cloning hosts with plasmid vectors were HB101, TG1 and DHScc (Sambrook J., et al., (1989) Molecular cloning. A laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York) and with lambda, Kw25 1 (Promega, Madison, Wisconsin).
pHPl3 (Haima, P., et al., (1987) Mol. Gen. Genet., 209:335-342, pJH101 (Ferrari, F.A., et al., (1983) J. Bacteriol., 154:1513-1515), pGEM3zf(+) (Promega) and pDR540 (Pharmacia, Upsala, Sweden) were used as cloning vectors for prsA gene and its fragments.
SUSSTiTUTE SHEET

WO 94/19471 ~~ ~ ~ PCT/FI94/00072 ,.

Properties of these plasmid vectors and constructed derivative plasmids carrying the prsA gene with a 5.3 kb (pKTH277 and pKTH268) or 3.4 kb (pKTH3253) insert are shown in Table 2. The prsA gene in pKTH3253 was disrupted by inserting a fragment either from B. licheniformis blaP gene (0.5 kb) or bacteriophage lambda genome (4.6 kb) in th~_unique EcoRV site in the ORF of prsA (the resulted plasmids were pKTH~3261 and pKTH3262).
pKTHlO (Palva, I. (1982a) Gene, 19:81-87; Palva, I., etal., (1982b) Proc.
Natl. Acad. Sci. USA, 79:5582-5586) and pMJ57 (Hastrup, S. and Jacobs, M.F. (1990) In Zukowski, M.M., et al., (eds.), Lethal phenotype conferred by xylose-induced overproduction of apr-IacZ fusion protein, vol. 3. Academic Press, Inc., San Diego, California, pp. 33-41 ) are multicopy plasmids producing large amounts of B. amyloliquefaciens alpha-amylase (AmyE) and B. licheniformis subtilisin (SubC) secreted into the external medium, respectively. pKTH1582 has been constructed cloning the amyl gene from B. licheniformis on a secretion vector system of B, subtilis (BRB360 in Sibakov, M. (1986) Eur. J. Biochem., lsS, 577-581; Palva, I. (1982a) Gene, 19:81-87; Palva, I., et al., (1982b) Proc. Natl. Acad. Sci. USA, 79:5582-5586).
Growth media and culture conditions Bacteria were grown in modified L-broth (1 % tryptone, 0.5%
yeast extract, 0.5% NaCI) with shaking at 37 degrees C, or on L-plates containing 1.5% agar (Difco, Detroit, Michigan) with appropriate antibiotics at +37 degrees C (Kontinen, V.P., et al., (1991 ) Mol. Microbiol. 5:1273-1283);
plates were modified for alpha amylase (5% starch and 2.5% agar) and subtilisin overexpressing strains (0.2% xylose and 1 % milk powder). L-broth was supplemented with 2% soluble starch (Merck, Darmstadt, Germany) or used as 2-fold concentrated medium for production of exoenzymes. Strains producing subtilisin were grown on L-plates containing 1 % milk powder. The production of exoenzymes was studied in two-fold concentrated L-broth with vigorous shaking. The growth was indicated by turbidity of the culture measured with KIettSummerson colorimeter (Klett Manufacturing Co., Inc.
N.Y.) using a no. 66 filter.
SU~3STITUTE SHEET

WO 94119471 ~ ~ PCT/FI94/00072 Enzyme assays Alpha amylase was assayed with Phadebas tablets (Pharmacia) as described in Kontinen, V.P. and Sarvas, M., (1988) J. Gen. Microbiol., 134:2333-2344. For the plate assay bacteria were streaked on L-plate containing 5% of starch, and the halo around colonies was measured after incubation of the plates at 4 degrees C. Typically, there was no zone around wild-type colonies producing endogenous alpha amylase, while that of strains carrying pKTHlO was more than 2 mm depending on the age of plates.
Subtilisin of B. licheniformis and chromosomally-encoded proteases were assayed in 1 ml of 0.1 M Tris-0.0 1 M CaCl2 (pH 8.0) with a chromogenic peptide substrate succinyl-AIaAIa-Pro-Phe-p-nitroanilide (Del Mar, E.G., et al., (1979) Anal. Biochem., 99:316-320). The rate of hydrolysis was measured on a Hewlett Packard diode array spectrophotometer at 410 nm. To determine the specific activities of alpha-amylases and subtilisin their amount in a sample of culture supernatant was estimated either as in ((Kontinen, V.P. and Sarvas, M., (1988) J. Gen. Microbiol., 134:2333-2344) or with SDS-PAGE stained with Coomassie Blue. The enzyme amounts were expressed as micrograms/ml.
Table 1. Bacterial strains Strain Relevant genotype and Parent strain, properties source or reference B.subtilis IH6064 metB5 sacA32l (Sibakov et al., 1983) IH6090 his metBS sac32) IH6064 IH6157 IH6090 (pKTHlO) IH6090 IH6160 IH6064 (pKTHlO) IH6064 IH6480 prs-3 metBS sacA321 (pKTHlO) IH6157 IH8482 prs-29 metBS sacA321 (pKTH10) IH6157 IH6483 prs-33 metBS sacA321 (pKTH10) IH6 157 SUBSTITUTE SHEET

WO 94119471 ~ ~~ ~ ~ PCT/FI94/00072 IH6484 prs-40 metBS sacA32i (pKTH10}IH6157 IH6485 prs-3 metBS sacA321 IH6480 IH6487 prs40 metB5 sacA321 IH6484 IH6489 prs-II mefBS sacA321 (pKTHlO)' ~ IH6160 ~

5 IH6491 prs-13 metBS sacA321 (pKTH:~~O)IH6160 IH6494 prs-33 metB5 sacA321 IH6483 IH6497 prs-26 metBS sacA321 (pKTH10)IH6157 IH6498 prs-13 metBS sacA321 IH6491 IH6501 prs-26 metB5 sacA321 IH6497 10 IH6504 prs-II metBS sacA321 IH6489 IH6513 QB917 (pKTHlO) QB917 IH6521 prs-13 hisAl trpC2 (pKTHlO) IH6513 IH6523 prs-Il metBS sacA32J (pKTHlO)IH6513 IH6622 IH6624 ,(pKTH277) IH6624 15 IH6623 IH6624 (pHPl3) IH6624 IH6624 PSLI (pKTHlO) PSLI

IH6654 prs-29 metBS sacA321 IH6482 IH6752 PSLI (pMJ57) PSLI

IH6755 PSLI (pHPl3) PSLI

20 IH6757 PSLI (pKTH1582) PSLI

IH6759 IH6755 (pHPl3) IH6755 IH6760 IH6757 (pKTH277) IH6757 IH6770 BRB764 (pKTH277) BRB764 IH6774 IH6160 (pKTH277) IH6160 IH6788 IH6752 (pKTH3229) IH6752 IH6789 IH6752 (pKTH3230) IH6752 PSLI arg(GH) 15, IeuAB, rm-, recE4,IA510 in BGSC

stp, thrA

QB917 hisAl thr-5 trpC2 IAIO in BGSC

BRB764 ::~('PS"bE'-amyE) (pKTH1743} A. Palva, University of Helsinki SUBSTITUTE SHEET

_ 215 6 425 B.subtilis IH6559 prsA29, recE4, trpC2 (::pKTH1601Kontinen et al, ) 1991 IH6799 IH6064 (::pKTH3200) IH6064 IH6811 IH6624 (pKTH3253) IH6624 IH6812 IH6624 (pKTH3261 ) IH6624 IH6813 IH6624 (pKTH3262) IH6624 E. coli EH1568 HBIOI (pKTH268) HBIOI

EH1581 DH5a (pKTH3101) DHSa EH1631 TGI (pKTH3180) TGI

EH1639 TGI (pGEM3zf(+)) TGI

EH 1640 TGI (pDR540) TGI

EH1674 TGI (pKTH277) TGI

EH1675 TGI (pKTH3253) TGI

EH1678 TGI (pKTH3261) TG1 EH 1679 TGI (pKTH3262) TGI

1 ) pKTH1743 is a dreivative of pUB110 carrying a 0.3 kbp insert with an ORF for B. subtilis deco gene 2) The Bacillus Genetic Stock Center, The Ohio State University, Ohio Table 2. Plasmids Plasmid Cloned genes and Derivative of number resistance markers (Source and/or references) pKTH10 amyE neo pUB110 (Palva, 1982;
Gryczan et al., 1978) pKTH268 prsA bia pGEM3zf(+) (Promega;
SUBSTITUTE SHEET

WO 94/19471 ~~~ ~ ~ PCT/FI94/00072 ..

Sambrook et al., 1989) pKTH277 prsA cat ermC pHPl3 (Haima et al., 1987) pKTH1582 amyl neo pUB110 (I. ~Palva:

Sibakov; 1986) pKTH1743 degQ neo pUB110 (A. Palva) pKTH1786 bla cat neo pAM~il (M. Simonen;

Leblanc and Lee, 1984) pKTH3101 'prsA bla pKTH268 pKTH3180 OIP~e~-'prsA bia pDR540 (Pharmacia) pKTH3229 ermC pHPl3 pKTH3230 prsA ermC pKTH3229 pKTH3253 prsA cat ermC pHPl3 pKTH3261 0.5 kb insert in pKTH3253 prsA

cat ermC

pKTH3262 4.5 kb insert in pKTH3253 prsA

cat ermC

pJH101 bla cat tet (Ferrari et al., 1983) pMJ57 OIPX~,-'subC cat pUB110 (Hastrup and Jacobs, 1990 Examples Example 1 Enhancement of alpha amylase secretion in B. subtilis when there is an increased amount of PrsA protein The following table (Table 3) illustrates enhancement of a-amylase secretion in Bacillus subtilis under various conditions when the host gram-positive bacteria also overexpresses PrsA protein.
SUESTITUTE SHEET

WO 94/19471 _ PCT/FI94/00072 Table 3. Enhancement of a-amylase secretion in B. subtilis by overexpression of PrsA protein The plasmid PrsA was expressed from a-Strain No. a-amylase amylase Plasmid No. prsA gene in expressed from secreted a the plasmid gene in pg~ml Multicopy plasmide~

IH6774 KTH277 Intact 3100 Chromosome~

IH6770 KTH277 Intact 2000 IH6811 pKTH3253 Intact 3700 IH6812 Multicopy plasmid~pKTH3261 Disrupted 1300 IH6813 KTH3262 Disru ted 1400 a) pKTH10 with amyE of 8. amyloliquefaciens b) One copy of amyE under a promoter of increased activity c) The a-amylase activity of culture supernatant was assayed in the late stationary phase of growth SUBSTITUTE SHEET

WO 94/19471 ~ ~~ ~ PCT/FI94/00072 Example 2 Enhancement of alpha amylase secretion in B:' amyloiiquefaciens when there is hyperexpression of-PrsA protein This example demonstrates the_effect of overproduction of PrsA
protein of B. subtilis in B. amyloliquefaciens when the host gram-positive bacteria also hyperexpresses PrsA protein.
ALK02732 is a derivative of ALK02100 and contains the multicopy plasmid pKTHlO encoding the a-amylase of B. amyloliquefaciens.
The strain thus contains tens of copies of a-amylase gene, and secretes about 20 fold more a-amylase than the wild type strain ALK02100 (Vehmaanpera et al., J. Biotechnol. 1991,19,221-240). Immunoblotting of cells of ALK02732 with anti PrsA antiserum showed that the amount of PrsA
protein was similarly.small in ALK02732 as in ALK02100, which is referred to below. Thus PrsA is rate limiting for protein secretion in ALK02732.
The plasmid pKTH277 was transferred with electroporation to the B. amyioliquefaciens strain ALK02732 to make ALK02732(pKTH277).
The amount of PrsA in ALK02732(pKTH277) was many folds higher than in ALK02732, as determined with immunoblotting. This is in good agreement with the similar increase of PrsA proteins in B. subtilis strains transformed with PKTH277.
The amount of a-amylase secreted to the growth medium (Luria broth of double strength with 2% of soluble starch) by ALK02732 and ALK02732(pKTH277) were determined during logarithmic and early stationary phases of growth (shake flask cultures). The results are shown in Tables 4-1 (Exp. 1 ) and 4-2 (Exp. 2). It can be seen that in two separate experiments the amount of a-amylase in the culture medium of ALK02732(pKTH277) was 1.5 to 2.5 fold the amount in the growth medium of ALK02732.
SU BSTITUTE SHEET

~'VO 94/19471 ~ PCT/FI94/00072 Tables 4-1 and 4-2. The Effect of pKTH277 on a-amylase secretion of B.
amyloliquefaciens (ALK02732).
5 Table 4-1, Experiment 1.
ALK02732 ALK02732 (pKTH277) TIME(h) GROWTH a-AMYLASE GROWTH' a-AMYLASE

(mg/I) (mg/I) 10 3 106 1,1 100 1,1 4 241 8,8 240 8,8 SUBSTITUTE SHEET

_ 215 6 425 Table 4-2, Experiment 2.
ALK02732 ALK02732(pKTH277) GROWTH a-AMYLASE GROWTH a-AMYLASE
~

TIME(h) r (mg~l) (mg~l) 3 100 0,9 112 1,5 4,5 350 24 363 38 6,5 545 190 577 450 12,5 700 900 The density of the culture as determined with the Klett-Summerson colorimeter using filter no. 66, indicated with Klett units.
Materials and Methods Used in this Example Bacterial strains and plasmids: B. amyloliquefaciens strain ALK02732 (described by Vehmaanpera et al., 1991 ) was used in transformation and growth experiments. ALK02732 (pKTH277) was made by transforming plasmid pKTH277 into ALK02732 (this study) and used in growth experiments. Plasmid pKTH277 carrying prsA gene is described by Kontinen et al. (1991 ).
Transformation of pKTll277 into ALK02732 by electroporation:
Plasmid pKTH277 was isolated using the alkaline lysis method and methylated with BamHl methylase (New England Biolabs) according to manufacturers instructions. About 0.5 Ng of methylated plasmid DNA was SUBSTITUTE SHEET

T 'JVO 94/19471 _ ~ PCTIFI94/00072 used for electroporation. Electroporation was done as described by Vehmaanpera (1989). Cells were pulsed in 0.2cm sample cuvettes (Bio-Rad Laboratories) with Gene PuIserT"~ apparatus (Bio-Rad Laboratories) set at 1.5 kV, 25NF and 400i~. Transformants were screened for chloramphenicol resistance on Luria-Kanamycin(lONg/ml)-Chloramphenicol(5Ng/ml) plates.
(Kanamycin was also on the plates to avoid loss of pKTHlO, since ALK02732 contains pKTHlO, conferring kanamycin resistance).
Growth experiments and sample collection: First ALK02732 and ALK02732 (pKTH277) were grown over night on Luria plates. From the plates bacteria was added to 10 ml Luria and they were grown to logarithmic phase (Klett 100). Then 1 ml Glycerol (1/10 of cultivation volume) was added and cell suspension was frozen and stored in -70C. Growth experiments were started by diluting Klett 100 cells 1:100 or 1:200 in 2x Luria + 2%
starch.
The Luria used in growth experiments contained no salt. Cultivation volume was 20 ml and growth was in bottles in 37C with vigorous shaking. For both strains Kanamycin (lONg/ml) and for ALK02732 (pKTH277) also Chloramphenicol (5Ng/ml) was supplemented to growth medium. The growth was indicated by measurements with Klett-Summerson colorimeter (Klett Manufacturing Co., Inc. N.Y.) using a number 66 filter. 0.5 ml samples were taken during growth, samples were centrifugated and culture supernatants were stored in -20C for a-amylase assays.
a-amylase assays: a-amylases in culture supernatants were determined using Phadebas tablets (Pharmacia). Samples were incubated for 1 h at 37C in 1 ml buffer (50mM MES pH 6.8, 50mM NaCI, 1 OONM CaCl2 containing 1/4 dispersed phadebas tablet, after which 50N1 5M NaOH was added to stop the reaction. After filtration through lAlhatman no.1 filter paper, absorbency of the filtrate was measured using 616-624nm as an analytical wave length range and 800-804nm as a reference wave length range.
Commercially available a-amylase of B. amyloliquefaciens (Sigma) was used as a standard and results were expressed as mg enzyme per I.
SUBSTITUTE SHEET

N
. 28 References: Kontinen V., Saris P. and Sarvas M. (1991 ): A
gene (prsA) of Bacillus subtilis involved in a novel, late stage of protein export. MoL Microbiol. 5:1273-1283. Vehmaanpera, J. (1989):
Transformation of Bacillus Amyloliquefacien~ by electroporation. FEMS
Microbiol. Lett 61:165-170. Vehmaanpe:rav;J., Steinborn G. and Hofemeister J.(1991 ): Genetic manipulation of Bacillus amyloliquefaciens. J.
Biotechnol.i 9:221-240.
Example 3 Enhancement of secretion of subtilis from B. lentus when there is hyperexpression of PrsA protein This example demonstrates the method and system of the invention in enhancing secretion of overexpressed exoproteins in B. lentus when the host gram-positive bacteria also hyperexpresses PrsA protein.
The effect of hyperexpression of prsA has been tested with respect to the secretion of subtilisin from B. lentus, commercially available as ExperaseT"' and described in WO 89/06279 (in the name of Novo Nordisk A/S). The subtilisin is transcribed from the plasmid pPL1800 which is based on the expression vector pPL1759 (Hansen, C., Thesis, 1992, The Technical University of Denmark) with a pUB110 origin and the promoter and signal peptide from the alpha-amylase of B. licheniformis (amyL). The plasmid pSX94 is described in WO 89; 06279. The B, subtilis strain SHa273 used for production is a protease weak derivative of DN1885 (Jorgensen, P.L. et aL, (1991 ) FEMS Microbiol. Lett., 77:271-276), in which two additional proteases apr and npr have been inactivated. The secretion of the B. lentus subtilisin was measured from strains either with (MOL253) or without (MOL252) the prsA plasmid pKTH277, and growth was performed at 30°C in soya broth BPX supplemented with kanamycin and chloramphenicol.
Measurements of subtilisin levels from the two strains after five days show that the strain with the prsA plasmid has a four fold higher secretion of the 8. lentus subtilisin (160 microgramslml) compared to the strain without this plasmid (40 micrograms/ml).
SUBSTITUTE SHEET

2~~~~25 WO 94/19471 PCTIFi94100072 The BPX medium used has the following composition:
BPX: Potato starch 100 g/I
Barley flour 50 g/I
BAN 5000 SKB 0.1 g/I
Sodium caseinate 10 g/I
Soy Bean Meal 20 g/I
Na2HP04, 12 H20 9 g/I
Pluronic 0.1 g/I
Strain List:
B. subtilis Genotype and properties Parent strain DN 1885 amyE, amyR2 RUB200 PL1801 amyE, amyR2, apr-, npr- DN1885 MOL252 PL1801 (pPL1800) PL1801 MOL253 PL1801 (pPL1800, pKTH277) PL1801 The RUB 200 strain is described by Yoneda et al., 1979, Biochem Biophys. Res. Common. Vol. 91, 1556-64.
Example 4 The presence of PrsA protein in Bacillus amyloliquefaciens and Bacillus subtilis The presence of PrsA protein has been demonstrated in two strains of 8. amyloliquefaciens, strain ALK089 and strain ALK02100. Strain ALK089 is an industrial strain used for the production a-amylase. Strain ALK089 is an overproducer described in (Bailey, M.J. and Markkanen P.H. J.
Appl. Chem. Biotechnol. 1975,25,73-79). Strain ALK02100 is a derivative of ATCC23843 (J. Vehmaanpera, FEMS Microbiology Letters 49(1988) 101-105).
The presence of PrsA protein has also been demonstrated in two strains of Bacillus licheniformis, strain 749/C (Pollock, M.R. (1965) Biochem. J. 94,666-6/5), and ATCC 14580.
The PrsA protein in these non Bacillus subtilis gram positive bacteria was identified using immunoblotting techniques. More specifically, SUBSTITUTE SHEET

2'~~ ~ 4~~
cells collected at late exponential phase of growth (to minimize the amount of proteases - PrsA protein is protease sensitive) were immunoblotted, treated with lysozyme shortly (to make cell wall leaky, but again avoiding long treatment to minimize proteolysis), and solubilized at 100°C with sample 5 buffer containing 2% SDS. PrsA protein was detected with rabbit antiserum (KH1283) raised against PrsA protein of B~ subfilis produced in Escherichia coli (thus there was minimal antigenic doss reaction with any bacillar protein except PrsA). The antiserum detects nanogram amounts of B. subtilis like PrsA in the immunoblots.
10 In addition, all four strains were found to contain a protein of the size of PrsA of B. subtilis, and specifically identified by the above antiserum.
The intensity of the staining of the band was approximately similar in both B.
amyloliquefaciens strains and similar to that of PrsA protein in wild type B.
subfilis. Coomassie,Blue staining of parallel SDS-PAGE of cellular proteins of 15 B. amyloliquefaciens showed only a very weak band at the position of PrsA
protein like in the case of B. subtilis and consistent with PrsA protein of 8.
amyloliquefaciens being a minor cellular protein as it is in 8. subtilis. The intensity of staining of the PrsA protein of the two B. licheniformis strains was weaker than in B. subtilis, suggesting a somewhat smaller amount of PrsA
20 protein. However, it cannot be excluded that the weaker staining is due to less efficient binding of the antiserum of PrsA protein of B. licheniformis than to that of B. subtilis. Table 5 is a summary of the roughly estimated amount of PrsA in the different strains.
25 Table 5. Approximate amounts of PrsA protein found in cells of Bacillus strains during early stationary growth phase (at a cell density of Klett 400). Estimates are based on Western blots and are preliminary.
Strain PrsA in cells (arbitrary units) 30 B. amyloliquefaciens RH2079 4'~
SUBSTITUTE SHEET

215 6 4 ~ ~ PCT/FI94/00072 B, licheniformis ' RH 305 1 B. subfilis IH67742~ 100' 1 ) Cells were collected from mid-logarithmic phase (at a density of Klett 100) of growth and the PrsA values estimated to correspond a density of Klett 400.
2) This is an overproducer of PrsA, due to the content of pKTH277.
Materials and Methods Used in this Example Strains: Bacillus amyloliquefaciens RH2078 = ALKOB9 =
VTTI97 = E18. This is an a-amylase overproducer. A gift from J.Vehmaanpera, ALKO. Bacillus amyloliquefaciens RH2079 = ALK02100, derived from ALK02099 (pE194/pC194). A gift from J. Vehmaanpera, ALKO.
Bacillus licheniformis RH2080 = BRAS = ATCC14580. This is a producer of thermoresistent a-amylase. A gift from P.Saris, BI, H:ki. Bacillus licheniformis RH305 = 749/c. This is a penicillinase constitutive strain, originally derived from J.O.Lampen. Bacillus subtilis IH6074. This is metB5 sacA321.Ref. M.Sibakov et.al.1983. Bacillus subtilis IH6774. This is derived from IH6064, contains plasmids pKTHlO (carrying the a-amylase gene) and pKTH277(carrying the gene for coding PrsA).
Growth media: Luria-agar plates (L-plates); twice concentrated Luria-broth (2x L) Purified PrsA protein: PrsA was purified from pKTH277 containing B. subtilis by M. Lauraeus.
Immune serum: E. coli-produced PrsA of B. subtilis, which was run in and cut out from a SDS-gel.
Chemicals: Phenylmethylsulfonylfluoride, Sigma P-7626.
100mM in ethanol at -20°C. EDTA, TitriplexRlll p.a. Merck 8418. As a 0.5 M
solution, pH 8. Lysozyme, Sigma L-6876. This was used as 1 mg/ml in the SUBSTITUTE SHEET

~,5~4~~

following solution: 20mM potassium phosphate pH 7,15 mM MgCl2, 20%
sucrose. TCA 100% BCAt Protein Assay Reagent by Pierce.
SDS-PAGE and Western Blot equipment and chemicals according to BioRad.
Blots were stained with 4-chloro-1-naphtol.
Culture conditions and sample preparation for gel electrophoresis: Bacteria were grown on L-.plates overnight at 37°C.
Colonies were picked with a glass rod into a preweighed Eppendorf tube, and weighed. Sample buffer was added to get either 10 or 100 mg cells(ww)/ml.
Samples were heated 10 min at 100°C.
Bacteria were grown in 2xL broth with agitation at 37°C. To minimize the protease effect bacteria (from -20°C) were first grown to Klett 100 (corresponding to about 1-2 mg cells ww/ml, or about 1 O9cells/ml). This was used as an inoculum at 10'2 dilution. 20 ml of bacteria were grown in Klett flasks and 4 ml samples were taken at Klett 100, at Klett 100+2h (Klett appr.400), and at Klett 100+4h (Klett appr.550).
Samples were immediately transferred into an ice bath, PMSF
was added to 1 mM, and EDTA to 10 mM. Cells were separated from culture supernatant by centrifugation at 12 OOOxg 10 min., and treated with lysozyme, 15 min at 37°C in a 1 /20 volume. An equal volume of sample buffer was addad. Culture supernatant was precipitated in 10% TCA at 4°C and concentrated 20-fold in sample buffer.
The samples were run in 12% SDS-PA gels, stained with Coomassie Brilliant Blue R, or blotted onto PVDF filters according to BioRad.
PrsA was detected with the specific anti-PrsA rabbit antiserum KH 1283.
Example 5 Enhanced Secretion of Lipase from Pseudomonas mendocina in Bacillus subtilis that Overproduces PrsA Protein Using the prsA gene in pKTH277, scientists at Genencor International, South San Francisco, CA, USA, have shown that when Bacillus subtilis overexpresses both PrsA protein and lipase (from Pseudomonas SUBSTITUTE SHEET

~ ~ 5 s 4 z ~ PCT/F194/00072 mendocina, a gram-negative bacteria), the amount of lipase secreted into the medium is about 3.5 times greater than it is in controls that do not overexpress the prsA gene. The Genencor International scientists used industrial strains of Bacillus, and industrial fermentation conditions. (Data not shown).
Conclusion Thus it can be seen that the present invention discloses a method and system for enhancing the production of industrially and medically important exoproteins in gram-positive bacteria. Without departing from the spirit and scope of this invention, one of ordinary skill can make various changes and modifications to the invention to adapt it to various usages and conditions. As such, these changes and modifications are properly, equitably, and intended to be, within the full range of equivalence of the following claims. Various features of the invention are also evident from the following claims.
SUBSTITUTE SHEET

Claims (11)

Claims

1. A transformed Gram-positive bacterium, which secretes an exoprotein of interest and which comprises (a) multiple genes encoding PrsA protein in the chromosome;
(b) a wildtype gene encoding PrsA protein in the chromosome and a gene encoding PrsA protein on a multicopy plasmid; and/or (c) altered regulatory elements to increase expression of PrsA protein.

2. The bacterium of claim 1, which is a species of the Bacillus genus.

3. The bacterium of claim 2, which is of a species chosen from the group consisting of Bacillus alkalophilus, Bacillus amyloliquefaciens, Bacillus brevis, Bacillus circulans, Bacillus coagulans, Bacillus lautus, Bacillus lentus, Bacillus licheniformis, Bacillus stearothermophilus, Bacillus subtilis, and Bacillus thuringiensis.

4. The bacterium of any one of claims 1 to 3, wherein said exoprotein is a protease, a lipase, a cutinase, an amylase, a galactosidase, a pullulanase, a cellulase, a glucose isomerase, a protein disuphide isomerase, a cyclodextrin gluconotransferase, a phytase, a glucose oxidase, a glucosyl transferase, lactase, bilirubin oxidase, a xylanase, an antigenic microbial or protozoan protein; a bacterial protein toxin, a microbial surface protein, or a viral protein.

5. The bacterium of claim 4, wherein said exoprotein is an amylase.

6. A method for producing an exoprotein of interest, the method comprising expressing said exoprotein in a Gram-positive bacterium according to any of claims 1 to 5.

7. A method for improving the secretion of an exoprotein in a Gram-positive bacterium, said method comprising transforming a Gram-positive bacterium, which secretes an exoprotein of interest, with a suitable polynucleotide construct resulting in a transformed bacterium, which comprises (a) multiple genes encoding PrsA protein in the chromosome;

(b) a wildtype gene encoding PrsA protein in the chromosome and a gene encoding PrsA protein on a plasmid; and/or (c) altered regulatory elements to increase expression of PrsA protein.

8. The method of claim 7, wherein the bacterium is a species of the Bacillus genus.

9. The method of claim 8, wherein the Bacillus bacterium is of a species chosen from the group consisting of Bacillus alkalophilus, Bacillus amyloliquefaciens, Bacillus brevis, Bacillus circulans, Bacillus coagulans, Bacillus lautus, Bacillus lentus, Bacillus licheniformis, Bacillus stearothermophilus, Bacillus subtilis, and Bacillus thuringiensis

10. The method of any one of claims 7 to 9, wherein said exoprotein of interest is a protease, a lipase, a cutinase, an amylase, a galactosidase, a pullulanase, a cellulase, a glucose isomerase, a protein disuphide isomerase, a cyclodextrin gluconotransferase, a phytase, a glucose oxidase, a glucosyl transferase, lactase, bilirubin oxidase, a xylanase, an antigenic microbial or protozoan protein, a bacterial protein toxin, a microbial surface protein, or a viral protein.

11. The method of claim 10, wherein said exoprotein of interest is an amylase.