GB2267148A - OVINE FOOTROT: Proteases,antibodies,vaccines and diagnostic assays. - Google Patents

Abstract

There is detailed a variety of acidic and basic proteases produced by Dichelobacter nodosus and antibodies thereto. These proteases comprise epitopes which can be used to distinguish virulent and/or intermediate strains of D. nodosus from benign strains. The proteases can be produced by recombinant means. There is also disclosed: 1. A DNA molecule which comprises the signal peptide derived from a gene which encodes either Bacillus subtilis neutral or alkaline protease or D. nodosus serine protease. 2. A hybrid protease molecule wherein the protease portion is derived from D. nodosus or Xanthomonas campestris. 3. A peptide having an isoelectric point of 9.0-9.5 in microcystalline form. 4. The use of subtilisin-like proteases in detergents. 5. The use of D. nodosus proteases to catalyse the synthesis of proteins or peptides.

Description

FOOTROT ANTIGENS, VACCINES AND DIAGNOSTIC ASSAYS This invention relates to isolated nucleic acids encoding antigens of ovine footrot, antigens produced by recombinant techniques and to vaccines comprising such antigens.

Further, the invention relates to assays for the diagnosis of ovine footrot and to antibodies and probes for use in such assays.

Ovine footrot is a contagious disease of the sheep's foot which is characterised by separation of the hoof from the living epidermal tissues due to a spreading infection in the interdigital skin and beneath the horn. The essential causative agent of ovine foot rot infection is the anaerobic Gram-negative bacterium Dichelobacter nodosus (formerly Bacteroides nodosus). The disease is of considerable economic significance throughout the temperate areas of the world where sheep are raised. In Australia alone the estimated losses due to footrot amount to millions of dollars annually (Stewart D.J., 1989 In: Footrot and Foot Abscess of Ruminants, pp5-45, Eds J.R. Egerton, W.K. Yong and G.G. Riffkin, Boca Raton, CRC Press).

Footrot is a notifiable and quarantinable disease in South Australia, Western Australia and parts of New South Wales and Victoria. Major emphasis is being placed on programs for substantially reducing the prevalence of virulent footrot in all Australian States following the recent implementation of Strategic Control Plans for regulatory control in the remainder of New South Wales and Victoria. An element of the program in Australia, has been the recommendation that the cut-off point, below which eradication is not economically advisable, should be standardised to a point between virulent/intermediate footrot and benign footrot.

Standardisation can only be achieved by introduction of appropriate laboratory tests which obtain information on the virulence of D. nodosus, because the expression of the disease results from the interaction between the spectrum of virulence exhibited by D. nodosus, host resistance and environment.

Several laboratory tests for discriminating the virulence of D. nodosus based on the protease characteristics are currently available. These include the protease thermostability (gelatin gel), elastase and zymogram tests.

The protease thermostability test compares the proteolytic activity of heated and unheated broth cultures (Kortt et al, 1982, Aust. J. Biol. Sci., 35:481-489; Stewart et al., 1982, Aust. Adv. Vet. Sci., pp 219-222). Strains causing virulent/intermediate footrot produce thermostable protease whereas strains from benign footrot have thermolabile proteases which are rapidly degraded upon heating. A result can be obtained within 8-14 days from the time of sampling when the protease thermostability test is used.

The elastase test (Stewart, 1979, Res. Vet. Sci., 27:99-105; Stewart et al, 1986, Aust. Vet. J., 63:317-326. has also been routinely used for differentiating the more virulent strains from benign, although the test is slow, requiring a minimum of 28 days for a result. The major advantage of this test is that it can be used to differentiate strains of intermediate virulence (Stewart et al, 1986, supra).

Zymogram analysis of D. nodosus culture supernatants following non-denaturing polyacrylamide gel electrophoresis (PAGE), has demonstrated the presence of protease isoenzymes or acidic proteases which can be separated into banding patterns unique to either virulent or benign strains on the basis of differences in the ionic charge (Kortt et al, 1983, Res. Vet.

Sci. 35:171-174). These acidic protease isoenzyme patterns can be used for differentiating virulent and benign strains.

However, results for zymogram analysis also take 8-14 days from the time of sampling.

Thus, it may be appreciated, that there exists a need for new rapid diagnostic tests able to provide information on the virulence of D. nodosus.

D. nodosus produces both acidic (Kortt et al, 1982, supra; Kortt et al, 1986, In: Footrot in Ruminants : Proceedincrs of a Workshop, Melbourne, 1985, pp 237-243, Glebe, CSIRO Division of Animal Health and Australian Wool Corporation) and basic proteases the latter only recently found to be present in cultures (Lilley petal, 1992, Eur. J. Biochem. 210:13-21), (Australian Patent No. 614754). The basic protease is serologically distinct from the acidic proteases and, like the latter, possesses common antigenic determinants shared by strains from the different serogroups (Stewart et al, 1990, Adv. Vet. Dermatoloqy 1:359-369, London, Balliere Tindall).

Anti-basic protease monoclonal antibodies react with broth cultures from 98% virulent strains, 68% intermediate strains and 5% benign strains (see Table 1) and may be useful in discriminating virulent/intermediate strains from benign strains (Australian Patent No. 614754).

In zymograms, virulent isolates of D. nodosus possess 4 bands whereas benign strains have a pattern of 5 bands. A characteristic feature of the pattern of benign strains is the presence of the two lower mobility bands 1 and 2 which are not present in any of the virulent strains. However, in some benign strains bands 1 and 2 are difficult to detect. A second feature is that bands 3 and 4 of benign strains consistently differ in mobility compared with bands 1, 2 and 3 of virulent strains and this difference can be used for differentiation in the absence of benign bands 1 and 2 (Kortt et al., 1983, supra).

Virulent strains having the typical virulent zymogram can be divided into type 1 and 2 on the basis of variation in the virulent zymogram pattern. All virulent strains contain bands 1, 2 and 3 but compared with type 1, the type 2 strains lack band 5 but have an additional band (4) migrating between bands 3 and 5.

Band 5 from benign and virulent strains have similar electrophoretic mobilities and the acidic proteases corresponding to these bands are structurally and antigenically similar. Band 5 is present in all benign and type 1 virulent strains examined to date (Kortt et al., 1983, supra). Hereinafter band 5 shall be referred to as V5/B5.

Hereinafter the isoenzymes corresponding to typical virulent zymogram bands 1, 2, 3 and 4 shall be referred to as V1, V2, V3 and V4 respectively. Similarly, the isoenzymes corresponding to typical benign zymogram bands 1, 2, 3 and 4 shall be referred to as B1, B2, B3 and B4 respectively.

The apparent high degree of relatedness between the benign and virulent Band 5 proteases has now been confirmed at the gene level where 99% homology in the DNA coding sequences was observed with deduced amino acid sequences having only two amino acid changes. In addition, the coding sequence for the B2 fragment elucidated (Figures 2 and 13) was found to be 99% homologous to that of V2. V2 has been found to be structurally and antigenically similar to V1 and V3, whilst B2 has been found to be antigenically similar to B1, B3 and B4.

However, despite the apparent high similarity between benign and virulent acidic proteases, it has been surprisingly found that acidic proteases from virulent D. nodosus strains possess at least one epitope that is not present in the acidic protease isoenzymes of benign strains. Further, the acidic proteases from benign strains possess at least one epitope that is not present in the acidic protease isoenzymes of virulent strains.

These observations permit, for the first time, the development of diagnostic assays able to positively discriminate between virulent and benign D. nodosus strains. It further permits the production of improved or alternative vaccines effective in the treatment of ovine footrot.

Accordingly, in one aspect the present invention provides a diagnostic assay for D. nodosus infection comprising the detection in a sample of one or more of the following: (i) an epitope present in virulent strains and/or at least a portion of intermediate strains, but not present in benign strains; (ii) an epitope present in benign strains and/or at least a portion of intermediate strains, but not present in virulent strains; (iii) an epitope present in some intermediate strains but not present in virulent strains or the majority of benign strains.

(iv) antibodies specific to an epitope present in virulent strains and/or at least a portion of intermediate strains, but not present in benign strains; (v) antibodies specific to an epitope present in benign strains and/or at least a portion of some intermediate strains, but not present in virulent strains; and/or (vi) antibodies specific to an epitope present in some intermediate strains but not present in virulent strains or the majority of benign strains.

Where an epitope or epitopes are to be detected, the sample assayed may be an extract from a swab of lesion exudate or a portion of broth culture. Monoclonal or polyclonal antibodies may be used to detect the epitopes listed at (i), (ii) and (iii) above. A double antibody capture and detection system for improving the sensitivity and specificity of the protease diagnostic tests may also be used. Where antibodies (iv),(v) and (vi) are to be detected, the sample assayed may be sheep serum. This assay could involve the use of native or recombinant antigens.

Any of the methods commonly used in the art for carrying out such immunoassays are suitable. The methods include, for example, the well-known ELISA and radioimmunoassays. However, where antibodies in serum are to be detected, it is preferable that the antigens are not directly bound to microtitre plates as they tend to lose their epitopes.

Preferably, the epitope present in virulent strains and/or at least a portion of intermediate strains but not present in benign strains, is an epitope present on an acidic protease.

More preferably, it is an epitope present on V2.

Alternatively, it is an epitope present in an acidic protease fraction of virulent strains 198, 334 and intermediate strain 450 from the Sephadex G100 peak 1 activity peaks (hereinafter referred to as G100P1 acidic protease) that remains in the broth supernatant after isolation of the basic protease by the method described in Australian Patent No. 614754; that is this acidic protease fraction is that which does not bind to the SP-Sephadex C-25 ion-exchanger at pH8.6 used to isolate the basic protease as described in Australian Patent No.614754.

Preferably the epitope present in benign strains and/or at least a portion of intermediate strains but not present in virulent strains, is an epitope present on an acidic protease.

More preferably, it is an epitope present on B2.

Alternatively, it is an epitope present in a basic protease (pI-8.6) of benign strains that elutes from SP-Sephadex C-25 at 0.1M NaCl and is distinct from the basic protease (pI-9.5) of virulent strains that elutes from SP-Sephadex C-25 at 0.2 M NalC.

Preferably the epitope present in some intermediate strains but not present in virulent strains or the majority of benign strains, is an epitope present on an acidic protease, particularly V1.

Preferably the antibodies specific to an epitope present in virulent strains and/or at least a portion of intermediate strains but not present in benign strains, are specific to an epitope present on an acidic protease, particularly V2 or the G100P1 acidic protease fraction.

Preferably the antibodies specific to an epitope present in benign strains and/or at least a portion of intermediate strains but not present in virulent strains, are specific to an epitope present in an acidic protease, particularly B2, or the basic protease present in benign strains as defined above.

Preferably the antibodies specific to an epitope present in some intermediate strains but not virulent strains or the majority of benign strains, are specific to an epitope present in an acidic protease, particularly V1.

In a second aspect the invention provides a monoclonal antibody specific to a D. nodosus epitope present in virulent strains and/or at least a portion of intermediate strains, but not present in benign strains. Preferably the monoclonal antibody is raised against the Gl00P1 acidic protease fraction, or V2, but it is to be noted that V1 and V3 may also be suitable as they are structurally and antigenically similar to V2.

In a third aspect the invention provides a second monoclonal antibody specific to a D. nodosus epitope present in benign strains and/or at least a portion of intermediate strains, but not present in virulent strains. Preferably this monoclonal antibody is raised against B2, but it is to be noted that B1, B3 and B4 may also be suitable as they are antigenically similar since these bands form complexes with the benign strain specific B2 monoclonal antibody.

In a fourth aspect the invention provides monoclonal antibodies capable of binding to epitopes present in all strains of D. nodosus. Preferably, these monoclonals are raised against purified or recombinant V5/B5 or V2/B2 (Table 1). They may be useful in assays performed in parallel to those according to the invention. They may also be useful as the tagged secondary antibody in double antibody capture and detection assays. Alternatively, V5/BS or V2/B2 proteases or antigens may be used in assays.

As mentioned above, an acidic protease containing fraction (G100P1 acidic protease) from virulent strains has been found.

This acidic protease fraction is present in the minor peak of proteolytic activity (peak one) eluted near the void volume of the Sephadex G100 column and does not bind to sulfo-propyl (SP)-Sephadex C-25 equilibrated with 0.01 M Tris HC1-5 mM CaC1 pH8.6. The latter column is used to isolate the basic protease by elution with 0.2 M NaCl. These acidic proteases, which appear to be distinct from the other acidic proteases, may provide the basis for assays discriminating between virulent and intermediate strains. Virulent strain-specific monoclonal antibodies may be raised against this acidic protease fraction.

Also, at least two benign strains (305, 309) have a basic protease (pl - 8.6) which elutes from SP-Sephadex C-25 at 0.1 M NaCl in contrast to the basic protease from virulent/intermediate strains which elutes at 0.2 M NaCl. A monoclonal antibody specific for virulent D. nodosus basic protease (Australian Patent No. 614754) has been observed to cross react with a protease in this fraction, but it may also have unique epitopes which permit the production of assays for discriminating benign strains from intermediate strains.

Benign strain-specific monoclonal antibodies may be raised against this benign basic protease.

Further, an additional anti-protease MAb has been produced which reacts with a subcategory of low intermediate strains (e.g. 312,478,639,718,and 733) and is also recognised by the benign anti-protease MAb (Table 2). The former MAb was raised against acidic protease V1, derived from the intermediate strain 312. The other strains (e.g. 478,639,718, and 733) are also classified as intermediate strains based on clinical and laboratory characteristics (elastase, gelatin gel, zymogram tests and colony morphology characteristics). This intermediate virulence anti-protease MAb raised against Strain 312 does not react with strains 305 and 309 recognised by the benign anti-protease MAb nor does it react with strains 198, 408, 450, 466 recognised by the virulent anti-protease MAb.

The observation that the D. nodosus virulent acidic proteases (V1, V2, V3 other than V5) are antigenically distinct from the benign acidic proteases (sl, B2, B3, B4) also suggests that the virulent and benign acidic proteases would be beneficially included within the basic protease vaccines described in Australian Patent No. 614754. Moreover it may be further beneficial to include V5 or B5 in the vaccine. Vaccines comprising the G100P1 acidic protease fraction may also be effective in the treatment of ovine footrot. Accordingly, it is to be understood that the present invention extends to these vaccines.

Further, the genes encoding B2, B5, V2, V5 and benign basic protease, have been cloned from a B2, V2 cosmid bank and subsequently sequenced.

Thus in a further aspect, the invention relates to a nucleotide sequence encoding at least one of B2, B5, V2, V5 and benign basic protease.

The term nucleotide sequence includes RNA, DNA, cDNA and nucleotide sequences complementary thereto. Such nucleotide sequences also include single or double stranded nucleic acid molecules and linear covalently closed circular molecules.

The nucleic acid sequences may be the same as the sequences as herein described or may contain single or multiple nucleotide substitutions and/or deletions and/or additions thereto. The term nucleotide also includes sequences with sufficient homology to hybridize with the nucleotide sequence under low, preferably medium and most preferably high stringency conditions (Sambrook et al. 1989 Molecular Cloning: a laboratory manual, Cold Spring Laboratories Press) and to nucleotide sequences encoding functionally equivalent sequences.

In still a more preferred embodiment, the invention comprises the nucleotide sequences substantially as shown in Figures 2, 3, 4, 10 or 11 hereinafter or parts thereof, mutants or derivatives.

The present invention also further extends to oligonucleotide primers for the above sequences, homologues and analogues of said primers, antisense sequences and nucleotide probes for the above sequences.

It will be appreciated that the invention also provides a method for producing V2, V5, B2, B5 and benign basic protease or portions thereof, by introducing an expression vector including a nucleotide sequence coding for the desired protease into an appropriate host cell, expressing the protease, and isolating the protease thus produced.

In this regard it has also been surprisingly found that extracellular proteins from Gram-negative D. nodosus may be expressed and secreted from B. subtilis using the native signal peptide.

Thus, in a still further aspect, the present invention provides a method for the production of a desired protein or peptide in a Gram-positive or Gram-negative host, comprising introducing to the host an expression vector including a sequence encoding the desired protein or peptide and a sequence coding for, or substantially homologous to, a signal peptide derived from a gene encoding a serine protease from D.

nodosus.

The host may be, for example, E. coli or B. subtilis and the sequence coding for the signal peptide may be derived from the basic protease gene from D. nodosus. More preferably the signal peptide is: MNLSNISAVKVLTLWSAAIA GQVCA A 1. 2.

1. Predicted cleavage site in E. coli (21 residues) 2. Predicted cleavage site in B. subtilis (26 residues) The invention should also be understood to extend to hosts and recombinant DNA molecules for use in the process.

Further, the proteases of D. nodosus, X. campestris and possibly other microorganisms contain a unique loop region (32 amino acid residues for D. nodosus proteases and 28 amino acid residues for X. campestris protease) at approx. amino acid 73104 which is absent from most other members of the subtilisinlike family of proteases (Figs. 5 and 9). The functional significance of this loop structure is unknown, but as most subtilisin-like proteases do not have this region and also that the size and sequence of the loops from the D. nodosus and X. campestris proteases are not identical, it is suggested that this region is not essential for protease activity.

The non-essential and flexible properties of the loop structure may permit these subtilisin-like proteases to be utilised as carriers for well-defined foreign epitopes. In this regard, it is to be noted that the highly hydrophilic nature of the loop sequence would indicate that the loop region is located at the surface of the protein molecule - an ideal location for presentation of foreign epitopes. Welldefined, small, linear epitopes can easily be inserted into the loop region by PCR-based mutagenesis. This kind of hybrid protease molecule could then be used in vaccine applications and/or as novel diagnostic reagents for a serological test of footrot or other diagnostic tests.

Accordingly, the invention further extends to hybrid proteases (and genes encoding them), including a heterogenous epitope within the loop region located at about amino acid 73-104.

Using a protease as an epitope carrier has several advantages: 1) the protein is secreted and produced at high level and is easy to purify which will lower the overall cost; 2) the protease activity is very easy to monitor. (This could provide a convenient way to ensure the hybrid molecule is presented in the correct conformation); and 3) due to the destructive nature of protease, such molecules are expected to be immunogenic.

The invention will now be further described with reference to the accompanying drawings and the following non-limiting example.

Brief DescriDtion of Drawings Figure 1 provides the amino acid sequence for the B2 (strain 309) V2 and V5 (strain 198) isoenzymes.

Homologies with D. nodosus basic protease (strain 198) are indicated.

Figure 2 provides the partial gene sequence and deduced amino acid sequence for a fragment of the B2 isoenzyme gene.

Figure 3 provides the nucleotide sequence and deduced amino acid sequence for the V5 isoenzyme gene.

Figure 4 provides the nucleotide sequence and deduced amino acid sequence for basic protease from benign strain 305.

Figure 5 compares the amino acid sequence of the D.

nodosus basic protease (virulent strain 198) against other subtilisin-like proteases: Xanthomonas campestris (xc), Saccharomyces cerevisiae protease B (scPB), Bacillus subtilis DY subtilisin (DY), thermitase from Thermoactinomyces vulgaris (TH), proteinase K from Tritirachium album (PK) and aqualysin from Thermus acquaticus YT-1 (AQ). Conserved residues between D. nodosus basic protease and the other proteases are boxed and the active site residues aspartic acid, histidine and serine are marked by an asterisk. The N-terminal and C-terminal residues are marked by arrow heads. The sequences of the prepro-regions and the C terminal extensions of the precursors of D.

nodosus protease and X. campestris protease and the C-terminal extension of S. cerevisiae protease are also shown.

Figure 6 provides a histogram showing the proteolytic activity of an unconcentrated culture supernatant of protease deficient Bacillus subtilis strain DB403 containing the cloning/expression vector pNC3kb with an inserted gene encoding the extracellular basic protease of Dichelobacter nodosus in comparison to an unconcentrated culture supernatant of B.

subtilis containing a negative control plasmid pNC-CTL and an unconcentrated culture supernatant of D. nodosus strain 198 as a positive control. Panel A is a protease assay using Hide Powder Azure. Panel B is an ELISA using a monoclonal antibody (MAb) to the D.

nodosus basic protease.

Figure 7 provides protein contents analyses using SDS PAGE.

A. Coomassie staining of total proteins separated on a 12% (w/v) polyacrylamide gel (Laemmli, 1970, Nature, 227:680-685). Lanes 1 and 2 contained supernatant samples of DB403(pNC-CTL) and DB403{pNC3kb} respectively.

Lanes 3 and 4 were the pellet fractions from DB403(pNC-CTL) and DE403(pNC3kb} cells. The protein samples were separated into supernatant and cell pellet by 1 min centrifugation in a microcentrifuge. The supernatant was concentrated by precipitation with 4 volumes of cold acetone, followed by vacuum drying, and finally resuspended in 60 yl of Laemmli sample loading buffer. The cell pellet was resuspended in 30y1 freshly made lysis buffer (10 mM Tris HCl pH 7.6, lmM EDTA, 10 mM NaC1, and 5 mg/ml lysozyme) and incubated at 370C for 10 min, which was then lysed by mixing with 30 yl of 2x Laemmli sample loading buffer.Both supernatant and pellet fractions were boiled for 2 min before loading 10 pl in each lane.

B. Western blot analysis. The protein samples in each lane were the same as in panel A. After SDS-PAGE, the samples were blotted to a nitrocellulose membrane using standard Western blot procedure as described before (Wang and Doi, 1987b). The primary antibody was polyclonal sheep anti-basic protease antibody (see page 18) which was used at a 1:1000 dilution. The secondary antibody was an alkaline phospatase-conjugated donkey anti-sheep IgG (Silenus Laboratories, Australia) used at 1:500 dilution. The colour development was carried out using the ProtoBlotR Western Blot AP System (Promega, Madison). The numbers given at the left side are molecular weight markers in kDa, determined from the prestained molecular weight markers from Bio-Rad.The open and solid arrows on the right indicate the predicted size for the precursor of basic protease (prepro-mature-C terminal extension) and mature basic protease molecules (Lilley et al., supra), respectively.

Figure 8 provides a plasmid map for pNC3kb showing the B.

subtilis neutral protease promoter (Pnpr) which was used to drive the transcription of the D. nodosus basic protease gene (bprV). The origin of replication (ori) and the kanamycin resistant gene (neo) derived from pUB110 are also shown in the map. Key restriction sites of the vector and insert are given on the outside of the map.

Figure 9 provides a comparison of the amino acid sequences of D. nodosus protease V5 (V5), D. nodosus basic protease (BP), Xanthomonas campestris protease (XC) and B. subtilis subtilisin Carlsberg (CA). Residues conserved between D. nodosus protease V5 and the other proteases are boxed. The active site residues aspartic acid (41), histidine (107) and serine (279) are marked by an asterisk; the residues identified in subtilisin (and proteinase K) as being involved in substrate binding are marked by #O Gaps (-) have been introduced to maximise the alignment of X. campestris protease and subtilisin Carlsberg.

Figure 10 provides the nucleotide sequence and deduced amino acid sequence for the B5 (strain 305) isoenzyme.

Figure 11 compares the nucleotide sequence of B5 with the nucleotide sequence of V5.

Figure 12 provides the nucleotide sequence and deduced amino acid sequence for V2 (strain 198) isoenzyme.

Figure 13 compares the nucleotide sequence of B2 with the nucleotide sequence of V2.

Figure ld compares the amino acid sequences including the C-terminal extension of the protease genes of D.

nodosus (basic protease, V5 and partial V2) and Xanthomonas campestris (XC).

Figure 15. provides the partial sequence of the bprV gene (see Lilley et al., 1992 supra for full sequence) showing the translational initiation site and boundary region between signal peptide and pro-peptide. The amino acid sequence was numbered from the first M (methionine) residue while the DNA sequence from the A residue of the initiation codon ATG. The RBS site was underlined, which has a calculated free energy of interaction of -12.2 kcal/mole. Two putative signal peptide cleavage sites are marked by the upward arrows as C1 and C2.

~amDle 1: Specificity of Anti-protease Monoclonal Antibodies Materials and Methods Preparation of Monoclonal Antibodies D. nodosus acidic proteases were purified from strains 305 and 198 by the method described in Australian Patent No. 580825 with the following modifications: 1. Chromatography on Sephadex G100 was performed with 0.01M MES-5 mM CaCl buffer pH 6.5 2. Ion-exchange chromatography was performed on a column of DEAE-Sepharose CL-6B with 0.01M Tris HC1-SmM CaCl2 buffer, pH 8.6 and a linear gradient of 0-0.08M NaCl over 800 ml at 40C was used to elute the bound proteases.

Samples of the B2 isoenzyme from strain 305 and the V2 isoenzyme from strain 198 that were judged to be homogeneous by PAGE and SDS-PAGE analysis (i.e. migrated as single protein band), were selected and inactivated with PMSF (lmM). For preparation of monoclonal antibodies, Balb/c mice were injected with one of the inactivated protease preparations according to the regimes below.

B2 (6 week old Balb/c mice) 1. initial: 50 g CFA, I/P 2. 3 weeks: 50 g IFA, I/P 3. 5 weeks: 25 g IFA, I/P 4. 8 weeks: 25 yg IFA, I/P 5. 4 weeks: 25 g Saline, I/V V2 (12 week old Balb/c mice) 1. initial: 50 yg CFA, I/P 2. 4 weeks: 40 yg IFA, I/P 3. 4 weeks: 40 yg IFA, I/P 4. 40 weeks: 40 yg IFA, I/P 5. 4 weeks: 25 yg Saline, I/V CFA = complete Freunds adjuvant (Difco) IFA = incomplete Freunds adjuvant (Difco) I/P = intraperitoneal I/V = intravenous Three days after the final booster, the spleens were removed for fusion. Spleen cells were fused with an appropriate mouse myeloma cell line (NS-1) using polyethyleneglycol.The details of the fusion protocol have been described elsewhere (Kohler and Milstein, 1975, Nature, 256:495-497). Fused cells (hybridomas) were separated by dilution and antibody-secreting clones were subcloned by limiting dilution.

As proteases tend to denature (lose epitopes) on direct binding to ELISA plates, screening was carried out using native protease antigen in solution and with the putative mouse immunoglobulin (anti-protease monoclonal antibody) immobilised with a capture antibody (rabbit anti-mouse Ig).

Monoclonal antibodies specific for an epitope present in acidic proteases from all D. nodosus strains were also raised using V2/B2 and V5/B5 in a similar manner.

Preparation of Polyclonal Antibodies The purified basic protease and acidic protease isoenzyme fractions were used for the production of polyclonal antibodies in sheep vaccinated with a minimum of two doses 28 days apart of 2 ml subcutaneously on opposite sides of the neck followed by bleeding for collection of serum three to four weeks later. The vaccines contained 100 yg of the protease fraction in the aqueous phase incorporated in an equal volume of Alhydrogel (Superfos Biosector a/s, Denmark) then emulsified in incomplete Freund's adjuvant (Difco) in the ratio 1:2.

Collection of Foot Swabs Foot swabs were collected with cotton-tipped swabs from active D. nodosus lesions by vigorously rubbing and rolling the swab in the necrotic exudate and/or hyperkeratotic lesion material of the interdigital skin or the underrun horn of the hoof in order to obtain as much lesion material as possible (a good indicator was that flecks of tissue could be seen on the swab). The swab samples were transported to the laboratory in plastic or glass containers sealed with screw caps or stoppers. When the swabs in their containers reached the laboratory they were stored frozen at -200C unless processed immediately.

Enzvme Immunoassay Method for Detection of Basic Protease Microtitre (Nunc-immuno Maxisorp II F96) plates are coated overnight at room temperature, with 100y1/well of a 1:1000 dilution (in antigen binding carbonate buffer, pH 9.6) of polyclonal anti-basic protease capture antibody raised in sheep (Australian Patent No. 614754).

Plates are washed three times with 0.01M phosphate buffered saline (PBS) pH 7.2 washing solution and then blocked with 0.2W Coffee-mate (Carnation Australia Pty. Limited) for 30 minutes at 370C.

A swab extract is prepared by cutting the cotton wool from the swab stick (using a scalpel blade) and placing it in the barrel of a 2ml syringe. The plunger is then used to flush 0.5 ml of PBS-0.05% v/v Tween 20 back and forth (X12) through the cotton wool into a 5ml plastic tube.

After plates are washed (x3), add 20y1 of 2% Coffee-mate to each sample well before adding 50pl of swab extract to duplicate wells. For positive and negative control wells, add dilutions of B. nodosus broth culture in 1% Coffee-mate, and for blank wells, use diluent only. Plates are incubated for lh at 370C.

Broth cultures can be assayed as described above for positive and negative control wells.

After plates are washed (x5), add 100well of a 1:1000 dilution of the second antibody (anti-basic. protease monoclonal antibody) in diluting solution. Plates are incubated for 30-60 minutes at 370C.

After three washes, add 100y1/well of a 1:1000 dilution of horseradish peroxidase conjugated sheep anti-mouse immunoglobulin antiserum (Silenus Laboratories Pty. Ltd.) in diluting solution. Incubate for 30-60 minutes at 370C.

Plates are washed (x3) before adding to each well 100 l of substrate/chromogen (e.g. HO/5-aminosalicylic acid). Prepare solution immediately before use by adding 12y1 of 30% w/w HO to 20 ml of 5-aminosalicylic acid (5-AS) warmed to 370C.

Following incubation for 1-2 h, the optical density of each well is read on an ELISA spectrophotometer at a wavelength of 450 nm. A reading ) O.d O.D. is considered positive.

Enzvme Immunoassav Method for Detection of Acidic Protease Microtitre plates are coated overnight at room temperature, with 100 1/well of a 1:1000 dilution (in antigen binding buffer) of anti-isoenzyme (V2, B2, V2/B2, V5/B5) monoclonal (capture) antibodies.

Plates are washed three times with PBS washing solution and then blocked with 0.5t BLOTTO (skim milk powder) for 30 minutes at 370C.

A swab extract is prepared by cutting the cotton wool from the swab stick (using a scalpel blade) and placing it in the barrel of a 2ml syringe. The plunger is then used to flush 0.5 ml of PBS-Tween back and forth (X12) through the cotton wool into a 5ml plastic tube.

After plates are washed (x3), add 20y1 of 10% BLOTTO to each sample well before adding 50y1 of swab extract to duplicate wells. For positive and negative control wells, add dilutions of B. nodosus broth culture in 2.5k BLOTTO, and for blank wells, use diluent only. Plates are incubated for lh at 370C.

Broth cultures can be assayed as described above for positive and negative control wells.

After plates are washed (x5), add 100well of a 1:500 dilution of an anti-isoenzyme (polyclonal) second antibody in 5.0% BLOTTO. Plates are incubated for 30-60 minutes at 37 OC.

After three washes, add 100y1/well of a 1:1000 dilution of horseradish peroxidase conjugated donkey anti-sheep immunoglobulin antiserum (Silenus Laboratories Pty. Ltd.) in 0.5W BLOTTO. Incubate for 30-60 minutes at 370C.

Plates are washed (x3) before adding to each well 100 l of substrate/chromogen i.e. H202/5-aminosalicylic acid, (5-AS).

Prepare solution immediately before use by adding 12 y1 of 30% w/w H202 to 20 ml of 5-AS warmed to 370C. Following incubation for 1-2 h, the optical density of each well is read on an ELISA spectrophotometer at a wavelength of 450 nm. A reading ) 0.4 O.D. is considered positive.

Other more sensitive substrate/chromogen solutions that can be used include H202/2,2 'Azino-di-[3 ethylbenzthiazoline sulphonic acid] (ABTS) or H202/tetramethylbenzidine (TMB). Add H202 (final concentration 0.002%) to 0.55mg/ml ABTS in phosphate/citrate buffer pH4.3 or to 0.1 mg/ml TMB in 0.1 M acetate buffer pH6.0. For ABTS stop enzyme reaction with 0.01k w/v sodium azide in 0.1 M citric acid. With TMB the reaction is stopped after 30 mins with 50y1 of 0.5 M H2SO4.

For further improving the sensitivity of the MAb ELISA by reducing the background created by the polyclonal antibody the V2/B2 MAb is labelled with horse radish peroxidase Wilson and Nakane, 1978, In: Immunofluorescence and Related Staining Techniques, W. Knapp, K. Holubar and G. Wick (eds.) Elsevier/North Holland Biomedical Press, Amsterdam, pp. 15-4.

and is then used in place of the anti-isoenzyme polyclonal antibodies and conjugated anti-sheep immunoglobulin antibodies.

Results Monoclonal antibodies raised against V2, B2, V2/B2, V5/B5 and basic protease were tested using the abovementioned assay methods for their ability to discriminate Trypticase-arginineserine (TAS) broth (Skerman, 1975, F. Gen. Microbiol., 87:107119) cultures with Lab-Lemco powder (Oxoid Ltd.) (Stewart et al., 1986, supra) of representative virulent, intermediate and benign strains. The results are summarised at Tables 1, 3 and 4.

TABLE 1: Specificity of anti-protease monoclonal antibodies (MAbs) in discriminating broth cultures of representative virulent, intermediate and benign strains.

1. Basic protease MAb" 98% Virulent strains 68% Intermediate strains 5% Benign strains 2. Benign (B2) protease MAb" 0W Virulent strains 45% Intermediate strainsC 100% Benign strains 3. Virulent (V2) protease MAbb 100% Virulent strains 56% Intermediate strainsd 0% Benign strains 4. Total (V2/B2, V5/B5) protease MAb" 100% Virulent strains 100% Intermediate strains 100% Benign strains a - tested on 105 Virulent, 130 Intermediate and 76 Benign strains b - tested on 55 Virulent, 41 Intermediate and 30 Benign strains.

low ~ low to medium virulence intermediate strains.

d - medium to high virulence intermediate strains.

The results obtained with the benign (B2) protease MAb (Table 1) indicate that there is a unique epitope in benign protease isoenzymes and approximately 45% of low to medium intermediate strains (hereinafter referred to as low intermediate) which is absent from proteases on virulent and 56% of medium to high virulence intermediate strains (hereinafter referred to as high intermediate). The virulent (V2) MAb recognises a virulent protease isoenzyme but does not react with protease from benign strains, so that there is no background signal with the latter in contrast to the basic protease MAb (Table 1). This MAb also reacts with proteases from the remainder (56%) of the intermediate strains (high intermediate) not recognised by its "mirror image", the benign protease MAb (Table 1).The virulent and benign anti-protease MAbs are mutually exclusive (Tables 1 and 2)o Table 2: Specificity of low intermediate MAb in comparison to virulent and benign MAbs in descriminating D. nodosus isolates grown in TAS broth culture.

Virulence Category * MAb Virulent Benign Benign Low High Low Intermediate Intermediate Intermediate Virulent 69169** 0159 0/41 0118 Benign 0/69 59/59*** 41/41 8/18*** Low Intermediate 0/69 23/59*** 6/41 17/18*** * Isolates of D. nodosus classified according to virulence parameters in the colony morphology, elastase test and protease thermostability tests (Table 3).

* * Includes isolates classified as virulent, isolates classified as high intermediate virulence and isolates classified as medium intermediate virulence.

* * * Includes two isolates classified as medium intermediate virulence.

Table 3: Virulence parameters of D. nodosus assessed by conventional laboratory tests.

Colony Elastase Protease thermostabiligy Laboratory Morphology (21 days) (gelatin gel) diagnosis V ++++ 70-90% V It,V ++ 60-70% It Im + 50-60% Im lb, B < + 10-50 /O Ib B - 0-trace B V Virulent Ib Intermediate bottom of range It Intermediate top of range B Benign Im Intermediate mid-range The results obtained with the low intermediate MAb (Table 2) demonstrated that there is an epitope on protease isoenzymes of the low intermediate strains and some benign isolates that is absent from virulent strains (Table 2). The benign MAb also recognises an epitope on the protease isoenzymes of the low intermediate category of D. nodosus. However, the virulent MAb does not recognise the low intermediate category and conversely the low intermediate MAb does not recognise the virulent/high intermediate category.

Thus, it is now possible to positively identify either benign or virulent strains in lesion exudate as described below.

Table 4: Comparison of two capture/detection ELISA systems incorporating virulent and benign specific MAbs for detecting pro teases of virulent, intermediate and benign strains of D.

nodosus directly in swabs of lesion material.

Virulence CapturetDeteotion ELISA System Flock Strain Serogroup Clinical Laboratory Double MAb** MAb/Polyclonal Ab*** (Conventional Tests) (Substrate TMB) (Substrate 5-AS) FIELD: 1. 734,735 61 V V 3/4+ (V) 1/4* (V) 2. 758 F1 V V 517* (V) 1/7* (V) 3. 198 A1 V V 28/34* (V) 14/34* (V) 4. 759,760,761 B/E I It 8/8* (V) 0/8* # 5. 334 B2 V V 9/22* (V) 1/22* (V) 6. 762, 763, 64,765 F1 V V 4/6 (V) ND 7. 709 E V V ND 1/3* (V) 8. 708 F2 V V ND 1/2* (V) 9. 726,727 Cl B B ND 1/3+ (B) 10. 736 G B B 2/7+ (B) one TOTAL (%+ve) 57/81 (70.4%) 21/90 (23.3%) PEN: 1. 465 F3 V V 4/6* (V) 3/6* (V) 2. 198 Al V V 6/8+ (V) 5/8+ (V) TOTAL (%+ve) 10/14 (71.4%) 8/14 (57.1%) 3. 690 B1 B B 1/2+ (B) 0/2# 4. 720 E I B 4/4+ (B) 0/4# 5. 463 A1 B B 0/2# 0/2# 6. 337 1 B B one 0/7# 7. 309 A2 B B 3/5+ (B) 0/5# 8. 687 H - B 3/5+ (B) 0/5# TOTAL (%+ve) 11/25 (44%) 0125 (0%) Experimental foot abscess (Fusobacterium necrophorum) 0/10# 0/10# Sheep faeces 0/5t 0/5t TOTAL (%+ve) 0115 (0%) 0115 (0%) + Number of swabs positive to virulent (V2) MAb. All swabs negative to benign MAb.

MAb virulence diagnosis in brackets.

t Number of swabs positive to benign MAb. All Swabs negative to V2 MAb.

MAb virulence diagnosis in brackets.

t No reaction with V2 and benign MAb.

++ Cut-off at OD450 = 0.6.

Cut-off at OD460 = 0.4.

The ELISA system using the double MAb for Capture/Detection for D. nodosus proteases and TMB for substrate was much more sensitive than the single MAb/polyclonal antibody method and 5-AS substrate in diagnosis of virulent and benign footrot from swab lesion material. The double MAb/TMB system was able to obtain a true positive diagnosis from all foot score categories from virulent footrot (1, 2, 3a, 3b, 3c, 4) and benign footrot (1, 3a, 3b). (Table 4).

Although both the virulent and benign MAbs were highly specific for diagnosis of virulent and benign footrot, the benign MAb was less sensitive than the virulent MAb in detecting their respective virulence specific proteases (Table 4).

Table 5: Spectrum of reactions of swab lesion exudates from virulent, intermediate and benign footrot with basic protease (BP), virulent (V2), benign (B), and total (T) MAbs using the MAb/polyclonal antibody and 5-AS substrate system.

MAbs Diagnosis Flock Strain Serogroup ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~ BP V2 B T Clinical MAb FIELD: 1. 735,735 B1 + + - + V V - - + ? 2. 758 F1 - + - + V V + - + (V?) + - - (V?) 3. 198 Al + + . + V V + - (V?) + - - + (V?) + + + V + + - + V + + - + + V + - - - (V?) + + - + V + - + + V + + - + V + + + V + + + V + - - - (V?) + + + V + 7 - + - + - + V + + + V + + - + V + + - + V + - - + (V?) 4. 759, 760, 761 B/E + - - V (V?) 5. 334 B2 - + + V V 7. 709 E + + - (+) V V + + - - V + - - - (V?) 8. 708 F2 - + - + V V 9 726, 727 C1 . - + + B B + MAbs Diagnosis Flock Strain Serogroup BP V2 B T Clinical MAb PEN: 1. 465 F3 - + - + V V + + - + V + + - + V - + - + 2. 198 Al + + - + V + + - + V + + - + V + + - + V - - - + ? + + - + V 4. 720 E - - - + I ? + - - - 7 + ? + 7 + - - + 7 When the spectrum of reactions of the basic protease (BP), virulent (V2), benign (B) and total (T) MAbs were compared, more swabs were positive to the T MAb (38/47) than the BP MAb (32/47) and the V2 MAb (28/47) (Table 5) . When single reactions with the BP MAb were compared with the T MAb there were 9/47 positive swabs with the former and 7/64 positive swabs with the latter. All swabs reacting with the V2 MAb also reacted with the BP MAb and/or the T MAb.However, there were 2/47 swabs that reacted with both the BP MAb and the V2 MAb only and 7/47 reacting only with both the V2 MAb and T MAb.

Further studies showed that V1, V2 and V3 are antigenically similar and all react with the V2 MAb in an indirect capture ELISA (see enzyme immunoassay for acidic proteases) and in immunoblots.

Non-dissociating, non-reducing PAGE was carried out in 5% polyacrylamide separating gels using the discontinuous buffer system of Davis (1964) as descibed by Kortt et al., supra).

Proteolytic activity was detected after electrophoresis by a gelatin/agarose substrate plate assay (Kortt et al., supra).

The slab gel was then electro-blotted on to nitrocellulose paper (Schleicher and Schuell) according to the method of Burnette, 1981, Anal. Biochem., 112:195-203. Strips were incubated sequentially for 1 h at 250C with MAb, then goat anti-mouse Ig-HRP, with three washes in PBS between each step.

A substrate solution containing 0.05% (w/v) 4-chloro-l-napthol and 0.01t (v/v) H202 in 10 mmol/L Tris-saline, pH 7.4, was added and the strips were incubated for 15 min before the reaction was terminated by washing with water.

In ELISA and immunoblots the V2 MAb does not react with acidic protease V5 and the V5/B5 MAb does not react with V1, V2 or V3. The V5/B5 MAb reacts with V5 and V4 acidic proteases.

Also, B1, B2, B3 and B4 are antigenically similar since these bands form complexes with the benign strain specific (B2) MAb and are displaced to the top of the gel when examined by nondenaturing PAGE and zymogram analysis. B1, B2, B3 and B4 all react with the B2 MAb in an indirect capture ELISA and immunoblots. The B2 MAb does not react with acidic protease B5 and the V5/B5 MAb does not react with B1, B2, B3 and B4.

The V2/B2 MAb reacts with all acidic proteases except V4, V5 and B5. Purified proteases V1, V2 (Table 6) and V3 (data not shown) are structurally similar as indicated by their amino acid composition and markedly different from V5 and B5 (Table 6). Peptide profiles of V1, V2 and V3 were also very similar (data not shown) Table 6. Amino acid compositions of purified D. nodosus acidic proteases.

Amino acid Protease V5 V1 V2a V2b B2 B3 V5 B5b sequence LYs 9.8 8.4 7.4 8.8 9.3 8.4 8.1 8 His 6.6 7.2 7.0 7.6 8.9 9.0 7.9 9 Arg 16.1 21.9 20.2 17.4 17.8 13.3 15.2 14 Asp 48.6 51.6 49.0 50.6 48.1 44.3 44.6 28 Asn 26 Thr 31.8 21.9 20.0 33.0 34.8 21.2 21.5 21 Ser 33.2 27.9 28.9 36.5 37.3 22.4 23.1 26 Glu 22.6 20.4 16.9 20.4 17.7 17.1 16.1 7 Gln 7 Pro 21.1 18.4 16.8 19.7 20.1 21.7 21.0 18 Gly 33.7 40.0 40.0 31.0 31.5 43.0 36.8 42 Ala 39.3 41.9 41.9 37.9 38.2 41.0 41.0 41 1/2Cys 1.2 1.7 2.8 2.6 2.5 2.3 4.4 4 Val 30.7 32.2 30.5 28.7 28.8 32.3 30.6 37 Met 5.0 4.2 4.7 9.5 9.2 4.2 0.8 5 Ile 13.8 13.8 13.2 13.8 13.5 18.3 17.9 21 Leu 19.2 19.5 19.0 13.8 18.3 10.8 11.2 10 Tyr 4.5 7.0 9.1 6.3 7.1 9.7 8.1 14 Phe 7.7 8.4 7.4 7.4 7.1 4.2 5.1 4 Trp n.d n.d n.d n.d n.d n.d 5 Total 347 Values are given as mol/mol protein after 24h hydrolysis in 6M HCl and are not corrected. n.d, not determined.The compositions of V1, V2a, V2b, B2, and B3 were calculated assuming 345 residues/mol.

Example 2: Cloning and Sequencing of D. nodosus genes encoding acidic proteases and benign strain basic pro tease.

The complete amino acid sequence of protease V5 and partial sequences of proteases V2, B2 and B5 were performed according to Lilley et al., 1992 supra and Kortt et al., 1993, Biochem and Molecular Biol. Int. 29: 989-998. The cloning and gene sequencing was performed as described below.

Materials and Methods Isolation of strain 305 basic protease The basic protease from strain 305 was isolated using the same oligonucleotide probes as used for the isolation of the virulent basic protease (Lilley, et al., supra).

The strain 305 gene, from the initiating MET codon through to the transcription terminator, was isolated by PCR with oligonucleotides N498 (forward) and N928 (reverse) (Table 7).

After restriction with EcoR1 and BamH1, the fragment was cloned into the Amersham expression vector, pTTQ18.

The 5' non-coding region of the 305 gene was isolated by oligonucleotide N562 (forward) and N190 (reverse), the latter specific for a sequence internal in the translated region of the gene (Table 7). The fragment was cloned without restriction into Bluescript for sequencing.

The sequence of the 305 basic protease was compiled from the two sequence regions.

Table 7. Oligonucleotide primers used in isolating and sequencing the benign strain 305 basic protease Primer Sequence N190 TTT TTG ATA CGC TAA ACG- > 3' (antisense, reverse complement of gene sense sequence) N498 GGT GAA AGC ATG AAT TCA TCG AAC ATT TCT- > 3' (forward or sense seq.) N562 CGG TTT CGA ATT CGC AGG CGT TG- > (sense) N928 CCC GGA TCC AAG CTT CGT TAA ATT AAC CCC- > (reverse) N1010 CAATTAAAAACAACCAAA Forward N1011 GCCCGAGATTTAGATCAA Forward N1012 GGTTATGATTCTGATATT Forward N1013 AACGATCAAGCACGCGTT Forward N1014 ACTTTGAAAGTAACCGAT Forward N1015 AACAGGTCCTGCTTG Reverse N308 AAAACTTTAACACCGTCA Forward V5/B5 Protease Clowning and Secuencing Cosmid Bank Construction Chromosomal DNA of D. nodosus strain 198 was isolated from an 11 litre culture grown to a density of 2.6x105 cells per ml at 300 C. The DNA was extracted by washing the bacterial pellet in TES buffer (50mM Tris-HCl, pH8.0, 5mM EDTA, 50mM NaC1), centrifuging to re-pellet, and resuspending in 2ml 25% sucrose in 50mM Tris-HCl, pH8.0. One ml lysozyme (lOmg/ml in 0.25M EDTA) was added and incubated on ice for 20 minutes. Added to the bacterial suspension was 0.75ml TE buffer, 0.25ml lysis solution (5k w/v sarkosyl, 50mM Tris-HCl, 0.25M EDTA, pH8.0) and lOmg of pronase, followed by a 60 minute incubation at 560C. The DNA was then purified by repeated phenol extractions until the aqueous phase cleared.The DNA-containing aqueous phase was dialysed extensively against lxTE at 40C, and stored as a dilute DNA solution at -200 C. Prior to restriction of the chromosomal DNA for Southern blot analysis or gene bank construction, the DNA was ethanol precipitated and adjusted to a concentration of 500 ug/ul.

The chromosomal DNA was partially digested with the restriction endonuclease Sau3Al (10 ug DNA/ 0.25 units Sau3Al) over a period of 30 to 60 seconds, and electrophoresed on a 1% agarose gel. DNA corresponding to the 35-45 Kb range was purified from the gel and adjusted to a concentration of 500 ug/ml after ethanol precipitation. The chromosomal DNA (40ug in 80 ul) was ligated with the same molar amount of cosmid pHC79 cleaved with BamH1.

In vitro packaging was performed with the Promega Packagene lambda DNA packaging system, following the protocol supplied with the kit. A gene bank of approximately 20,000 colonies was generated by this method. To detect proteolytic colonies, the gene bank was plated directly onto 1% milk agar plates. The cosmids isolated from proteolytic colonies were used to retransform E. coli strain DHSa. Those which retained the proteolytic phenotype were used for further studies.

Screening for Protease V5 and B5 Expressing Clones The cosmid library in host strain DHSa was screened for proteolytically active clones by direct plating on LB agar with 100,tg/ml ampicillin and 1% skim milk powder. Colonies displaying zones of clearing were selected, replated onto milk plates and incubated at 370C for 2-3 days, until significant clearing occurred. The cells were removed and the agar containing the recombinant protease was spun in Eppendorf tubes at 14K rpm at room temperature for 30 minutes.The liquid supernatant was removed and tested specifically for the presence of protease V5 by the zymogram technique (Kortt et al., 1983) and by ELISA, using monoclonal antibodies specific for V5 protease.

Determination of Nucleotide Sequence Sequencing grade cosmid DNA was obtained from protease positive clones using Qiagen-tip 100(Diagen) or CsCl purification. Oligonucleotide primers were designed from the mature protein sequence (Kortt et al., 1993, Biochem. and Molecular Biol. Int., 29:989-998) and synthesised on a Pharmacia Gene Assembler Plus. The oligonucleotides used are shown in Table 8.

DNA sequence was obtained in two ways. Cosmid clones were sequenced directly by utilizing the BRL dsDNA Cycle Sequencing system. Oligonucleotides designed from the mature V5 protein sequence( Kortt et al., supra) were used to sequence the portion of the gene coding for the mature protein. Reverse primers were used to sequence outwards to obtain the entire gene sequence coding for the pre- and pro- regions of the V5 protease. The sequencing reactions were run on a Corbett FST-IC Thermal Sequencer, employing a standard program of 35 cycles, with an annealing temperature of 50 -55 C.

Secondly, subcloned regions of aprVS were also sequenced by the double stranded DNA method using a T7 DNA polymerase sequencing kit (Pharmacia).

The DNA sequences were electrophoresed on a 6% TBE sequencing gel, fixed in 10% ethanol-10k glacial acetic acid, rinsed in water, and dried at 800C. Dried gels were autoradiographed on Fuji-RX X-ray film at -700C with intensifying screens for 16924 hours and developed in a Fuji RGII X-ray film processor.

Subcloning of the Protease V5 Gene (aprV5) and B5 Gene (aprB5) A 36 mer oligonucleotide primer at region 270bp upstream of the start codon (Primer PCRV5-P20) and a reverse primer at the end of the mature enzyme coding region (Primer PCRVS-P21) were designed and used to amplify a 1.7Kb fragment in the polymerase chain reaction from virulent strain 198 and benign strain 305 genomic DNA. This fragment, representing the protease V5 gene, aprV5, without the carboxy terminal region, was cloned on a BamH1-EcoR1 fragment into vector pBR322, and transformed into DH5a.Positive clones were screened as tetracycline sensitive on LB plates with 100 g/ml ampicillin and 20pg/ml tetracycline, and confirmed by DNA sequence data and Southern Blot analysis, probing with the 32P- labelled 1.7Kb PCR fragment.

Twenty-seven clones expressing proteolytic activity were identified. Half of the clones were found to express the basic protease with the remaining clones expressing protease V5. No clones expressing protease V2 were identified. Clone V5COS-3 displayed stable expression of protease V5 after successive culturing and cosmid isolation. Agarose gel electrophoresis showed the protease V5 gene, aprVS, was present on a 35kb fragment. Two of the cosmid clones showed expression of both the basic protease and V5 isoenzyme, with cosmid mapping and PCR analysis determining that the 2 genes lie approximately lKb apart, with apr V5 upstream of btr V.

The protease V5 1.7Kb clone from strain 198, and the corresponding B5 gene from strain 305 were expressed on milk plates, with the product confirmed as protease V5 and protease B5 by zymogram and ELISA.

Table 8: Oligonucleotide primers used in the V5/B5 study Name Sequence Direction V5-P8 CAAGATGGCGATGGGCGTGATTCT Forward V5-P9 ATCGGGATACTTACCGCACGC Reverse V5-P12 CCATTGGTCGTTATAGAA Reverse V5-P13 TCTGGCGTTGCCTACGAT Forward V5-P14 GGTTATAACTCTGATATTAACG Forward V5-P15 AGCGGATTCGTAATTTACGGC Reverse V5-P16 ACCACTTCTAACGGTACT Forward V5-P17 GCAGGTATTGTTGATGCT Forward V5-P18 AATATTTTCATTTTACTA Reverse V5-P19 CTCGTTTCTCCGAAAGGT Forward V5-P20 TTCTGTTCTGCGGATCCTGCTTTTTGAAGCGCGGCG Forward V5-P21 CGCTTGCGGTTTGAATTCGTCGATCGGACGTTAAAC Reverse V5-P22 ACGCGGTGTTACCGG Forward V5-P23 CCCGTTTTCTCCACCCC Forward V5-P24 CCCGCTGGTGATAATAAGACGGATCCACGTCGAGC Reverse V2/B2 Protease Cloning and Sequenceing Degenerative oligonucleotide primers V2-P1 and V2-P6 (Table 9) were designed from strain 305 B2 protease peptide sequence data, and used in the PCR to amplify an 800bp DNA fragment from benign strain 309. Primers V2-P7 and V2-P8 were subsequently synthesised and used in the PCR to amplify a corresponding 780bp fragment from strain 198. This 780bp fragment was used as a probe to screen a range of 198 genomic DNA single and double restriction endonuclease digests by Southern analysis. A 2.9Kb EcoRl-Ba'iH1 band was identified as containing the V2 protease gene (aprV2). Genomic DNA from strain 198 was extensively digested by EcoR1-BamH1, the fragments separated by agarose gel electrophoresis, and the 2.5Kb-3.OKb region removed and purified by Geneclean. The purified DNA was ligated into the vector pUC118 digested with EcoR1-BamH1 and electroporated into E. coli strain DHSa.

Transformants were patched onto LB plates containing 100ug/ml of ampicillin, incubated, then lifted onto Amersham Hybond N+ for colony blot screening using the method of Buluwela et al., 1989, Nucleic Acids Res., 17:452. Positive colonies were confirmed by Southern analysis of the clones.

Dideoxy chain terminating double stranded DNA sequencing using oligonucleotide primers V2-P7 to V2-P16 inclusive confirmed that the clones contained aprV2, with the EcoR1 site located in the pro-region, 10 amino acids upstream of the mature region cleavage site. The mature protein region of 345 amino acids is followed by a Carboxy-terminal extension of 128 amino acids.

Table 9: Oligonucleotide primers used in the V2/B2 study Name Sequence Direction V2-P1 AATGATCAG CAAiGTATCGAITG Forward V2-P6 AACA/8TG CGGAG CCG CCATT/AGAT/GGTA/GCC Reverse V2-P7 CGGCGTTAAAGCTTATAAAGTTTGGG Forward V2-P8 ACCACGAGAAGTTGTGGATCCAACGG Reverse V2-P9 CCGTTGGTGCCACAACTTCTCGGTGG Forward V2-Pl 1 AATAGCGTATTGCGC Forward V2-Pl 2 TACTACGGATCCTTAATTGCCTTCGTTGCC Reverse V2-P13 GTTTAAATCTCGGTG Reverse V2-P14 CGGTGTGGCGCAACTGGC Forward V2-Pl 5 GTGTTGATCATTTGGATCCGCGGG Reverse V2-Pl 6 TTCGTAGAATTCGACCGTATTGCTCC Forward The amino acid sequences of the mature and precursor forms of V2, B2 and V5 were determined both by a combination of protein and DNA sequencing techniques and are provided at Figure 1 and where they are compared with the sequence of D. nodosus basic protease to show the extent of similarity between these D.

nodosus proteases. The nucleotide sequences of the genes for B2, V2, V5 and the benign strain 305 basic protease (isoelectric point, pl-8.6) are presented at Figures 2, 3 and 4 respectively. The basic protease found in benign strains 305 and 309 differs from the basic protease isolated from virulent/intermediate strains by ten amino acid changes in the sequence of the mature protein resulting in two additional negative charges (2 Asp residues) which lower the pI to -8.6 from pI-9.5. The possession of these genes and then sequences permits the production of these isoenzymes in large amounts by recombinant DNA techniques which allow the transfer of the genes encoding these proteins into a new host. The new host may be another bacterium such as species selected from Bacillus, Xanthomonas, Escherichia and Pseudomonas or any other type of cell in which the genes are capable of being expressed with or without genetic modification of the host genes or additional genetic engineering of the cloned D.

nodosus gene sequences themselves.

An open reading frame was found within the determined V5 sequence of 2371 bases (Fig 3) having the potential to encode a precursor protein of 595 residues. The sequence contained the complete sequence of protease V5 preceded by a putative hydrophobic signal sequence of 21 or 27 residues and a pro-region of 106-112 residues (Fig 3). The mature Protease sequence of 347 residues is followed by a 120 amino acid carboxyl terminal-extension before the stop codon. These structural features of aprVS are similar to those of the D.nodosus basic Protease gene, bprV (Lilley et al., supra). A putative RBS with the sequence 5'-AGGTGA-3' was located 7 bp upstream of a potential translation initiation codon at position +1 (Fig 3) which was followed by a characteristic hydrophobic, prokaryotic signal sequence.A TAA stop codon was located at position 1785 bp of the aprV5 gene, followed by a 25 bp long transcription termination loop (TTL) a further 87 bp downstream (Fig 1). The TTL possesses a free energy of interaction of around -23 k cal/mole.

In the amino terminal region two additional amino acids, serine and glycine, at positions 188 and 189 respectively (DNA sequence S62TCT GGTS) (Figure 3) are present in the mature protein sequence of the V5 and B5 protease, but absent in the basic protease, protease V2 and B2. However, three amino acids present in the pro-region of the basic protease, glutamic acid (E-106), serine (S-105) and isoleucine (I-104) (Figure 14). DNA sequence 82GAA AGT ATC90 in Figure 4 have been deleted as a block from the unprocessed V5 sequence between signal peptide amino acid residues 27 and 28, at one of the proposed positions where the signal sequence is cleaved from the pro-region (Figure 14).The mature V5 sequence also contains an additional valine residue (DNA sequence 1429GTT [Figure 3)) at the end of the protein sequence where the Cterminal region is cleaved. This results in the mature V5 protease being 347 amino acids in length in comparison to the processed basic protease which has 344 amino acid residues.

DNA sequence comparison between the V5 gene and the basic protease gene sequences across the region coding for the complete, unprocessed protein shows a homology of 31%. This low level of homology is consistent with Southern blot analysis of restricted D. nodosus genomic DNA, which under stringent conditions showed no cross-reaction when either the V5 gene or the basic protease gene was used as a probe (data not shown). However, when sequence alignment takes into consideration codon insertions and deletions, the overall homology of the complete genes increases to 64.5W. This DNA homology translates to 58% identity of amino acid sequence for the unprocessed protein (complete precursor molecule), but increases to 64% similarity for the mature protease.

Although the V5 and basic protease genes possess C-terminal extensions of similar size, the DNA homology is much lower in this region than for the rest of the protein. DNA homology for the C-terminal region is 36%, translating to an amino acid homology of > 10 > . Although the amino acid sequence of the Cterminal region has not been conserved hydrophobicity profiles for both unprocessed precursors of protease V5 and basic protease appear very similar indicating that the hydrophilic/hydrophobic character of the C-terminal region has been conserved. Therefore, although the amino acid sequence of the C-terminal region is not highly conserved between the V5 and basic protease, this region displays similar properties and hence a likely similar function in D. nodosus.

The protease expressed by the positive clones was verified as the V5 isoenzyme by the zymogram technique, and by ELISA using monoclonal antibodies specific for the V5 protease. The V5 gene displays a similar structure to the basic protease gene, both containing the pre-, pro- and carboxy terminal regions.

The V5 gene and corresponding B5 gene from strain 305 have been cloned on a 1.7 kb fragment, without the C-terminal extension. A TAA stop codon was introduced at the end of the mature gene sequence, with the subsequent clones expressing an active protease. This indicates that the C-terminal extension is not essential for the protease production or processing in E. coli.

A higher degree of homology in the nucleotide sequence between both D. nodosus serine proteases (V5 and basic protease) is displayed at the DNA level for the signal sequence (58%) proregion (68%) and the mature region (67W). These translate into an amino acid similarity of 63%, 69k and 64% respectively. This conservation suggests there are greater functional constraints upon these sequences resulting in less sequence divergence than for the C-terminal region.

Comparison of the sequence of D. nodosus basic protease and acidic protease V5 with sequences of other bacterial serine Proteases show that they are related to the subtilisin family of proteases (Figures 5 and 9). The sequences around the catalytic residues (Asp, His and Ser) of the subtilisins, which are strongly conserved, were used to align the sequence of the D. nodosus protease with those of other members.

However, the overall similarity is only approximately 25-30W with most of the other subtilisin-like proteases. The mature D. nodosus proteases are approximately 70 residues longer than the typical members of this family from Gram-positive bacteria.

The deduced amino acid sequence of the precursors of D.nodosus protease V5 and basic protease are shown in Fig 14, and are compared with that of the extracellular serine protease of X.

campestris (Liu et al., 1990, Mol. & Gen. Genet., 220:433-440) which is similar in length and has a signal peptide, pro-peptide, mature protease sequence and C-terminal extension of similar length to the D. nodosus protease regions. The protease V5 signal peptide region is of similar length in aprV5 and bprV and contains two potential cleavage sites at either Ala-110-Ser-109 or Alal-Vall. Comparison of the sequence alignment of the three precursor molecules (Fig 14) suggest that Ala4-Val103 may be the cleavage site rather than Ala-110-Ser-109 yielding a signal peptide of 27 residues.

There is 52% identity between this region of the protease V5 and basic protease whereas in the X. campestris precursor only the alanines at the potential signal peptide cleavage sites have been conserved. The only conserved segments in the pro-region of X. campestris protease precursor may be the putative functional segments located at positions -85 to -79 and -34 to -33 (Figure 2) (Lilley et al., supra). Both the D. nodosus proteases show about 50% sequence identity to the X.

campestris protease - Gene probes using PCR (polymerase chain reaction) technology are potentially more sensitive than monoclonal antibodies.

Knowledge of the sequences at Figures 2-4, 10 and 12 enables nucleic acid probes to be designed which may detect the sequences responsible for the epitopes located in the genes of virulent, intermediate and benign acidic proteases and the basic protease or the genes thereof. These probes shall be useful in DNA fingerprinting assays and for the specific detection of virulent, intermediate and benign strains. For example, it has been found that when the basic protease gene (Australian Patent No. 614754) is used as a probe against genomic DNA cut with HindIII, a different restriction pattern between the benign strain 305 and virulent strain 198 is evident. Sequence data suggests that this is due to a single T to C base change 72bp downstream of the start codon which introduces a HindIII site in the benign gene only. This probe correctly identified 21 out of 21 virulent strains and 11 out of 14 benign strains. Further, a single A to G base change has been observed 6 bp upstream of the start codon (in the proposed RBS), of the benign basic protease gene. The detection of this change should also provide a useful diagnostic assay.

Protease V5 and B5 sequences showed 99% similarity at the DNA level, with the deduced amino acid sequences of the pre- proand mature regions displaying 99% similarity, having only 2 amino acid changes. One change occurs in the pro-region at residue 34 (N to K) and one in the mature protease region at residue 216 (H to Q) (Figure 3, 10).

Similar assays could be developed utilising the differences between the nucleotide sequences of the genes encoding V5 and B5 (see Figure 11), particularly at V5 nucleotide sequence no.

469 (T o A), 1015 (C # A) and 1538 (A # G), and V2 and B2 (see Figure 13), particularly at V2 nucleotide sequence no.

79-80 (CG # AA), 84 (T # C), 112-114 (ATG # CGC), 117 (C # A), 123 (T # C), 130-131 (AA # GG), 305-306 (TA # CG) and 715 (G # C) (Figure 13).

The C-terminal extension of the protease gene could also be used to develop probes as this region shows only limited homology between V5/B5, V2/B2 and the basic protease (Figure 14) - Both aprVS and bprV contain A-T rich sequences in the region 300bp upstream of the start codon (G+C= 35e), hence it is possible that these regions are involved in gene regulation. Example 3: Direct Secretion in Bacillus subtilis of a Basic Protease from the Gram-negative Dichelobacter nodosus using its own Signal Pep tide Unlike the Gram-negative E. coli, the Gram-positive bacterium Bacillus subtilis contains only one layer of plasmic membrane.

This was believed to be one of the major factors contributing to the fact that most extracellular proteins from B. subtilis are secreted directly into growth medium rather than into the periplasmic space as in the case of E. coli extracellular proteins (Behnke, 1992 in Biology of Bacilli-Applications to Industry, Butterworth-Heinemann, Boston 143-188). Comparison of various known signal peptide sequences from both organisms indicated an important difference in their structure, i.e. the average length of the Gram-negative signal peptide (22 aa) tends to be shorter than that of the Gram-positive signal peptides (30 aa) (Behnke supra). This may explain why some extracellular proteins of Gram-positive origin could be secreted into growth medium when expressed in E. coli without any modification of their signal peptide (Hinchcliffe, 1984, J. Gen.Microbiol., 1285-1291) while Gram-negative extracellular proteins failed to be directly secreted in B.

subtilis using their native signal peptide (Behnke, supra).

However, in this example, a Gram-negative extracellular protein was successfully secreted in the Gram-positive organism, B. subtilis without any modification of the heterologous gene.

Construction of Expression Plasmid pNC3kb pNC3 is a B. subtilis cloning expression vector (Wu et al., 1991, Gene, 106:103-107) which uses the B. subtilis neutral protease gene as a screen marker and contains seven unique cloning sites. The bprV (basic serine protease) gene from D.

nodosus (Lilley et al., 1992, Eur. J. Biochem., 210:13-21) was inserted into pNC3 by digestion with HindIII and EcoRI and ligated with the 3kb HindIII-EcoRI fragment containing the bprV gene isolated from plasmid pBR3kb to form the expression plasmid pNC3kb (Fig.8). The neutral protease promoter located in the upstream region of the bprV gene in pNC3kb would ensure a proper transcription in the correct orientation. In other words, this forms a transcription fusion, so that the translation initiation has to rely on its own ribosome binding site.

Direct Expression and Secretion in the D. nodosus Basic Protease in B. subtilis pNC3kb was introduced into the protease deficient B. subtilis host strain DB403 (Wang et al., 1989, J. Gen. Appl. Microbiol.

35:487-492) by natural transformation and used to inoculate a 2 ml liquid culture of 2xSG medium containing 1% skim milk and 5 yg/ml kanamycin with the culture incubated at 370C with shaking for 48 hours. The sporulation medium 2xSG was used to maximise the expression of the neutral protease promoter, which is turned on during stationary phase (Wu et al., supra).

A parallel culture of DB403(pNC-CTL) was included as a negative control. pNC-CTL is a control plasmid derived from pNC3 containing a 2.3kb insert derived from pBR322 (nucleotides 2064-4359). After 1-min of centrifugation, the supernatant was taken for quantitative assays as outlined below.

Protease Assay 10 mg Hide Powder was mixed with 1 ml. assay buffer (20mM Tris-HCl, 5mM CaCl2, 0.2M NaCl, pH8.5) in an Eppendorf tube.

The reaction was started by adding the appropriate amount of culture supernatant (40, 20 and 10y1 for pNC-CTL and pNC3kb; 8, 4 and 2 ,tl for S198). The tube was shaken at 370C for 30 mins. The reaction was stopped by centrifugation for 1 min.

The clear supernatant was then taken for absorbency measurement at 595 nm. The results are given in Fig. 6A which clearly shows that using Hide Powder Azure as the substrate (Lilley et all., supra) there was protease activity associated with DB403(pNC3kb), which was absent from the control culture of DB403(pNC-CTL). Furthermore the protease level for S198 is at around 10x higher than that from DB403 (pNC3kb). This would suggest that the basic protease activity only contributes to about 5% of total protease activity in the S198 broth.

In order to eliminate the possibility that the protease activity associated with pNC3kb might be due to the activation of an indigenous B. subtilis protease gene rather than directly derived from the bprV gene product, an ELISA was carried out using a conformation-specific monoclonal antibody which reacts only with active D. nodosus basic protease.

ELISA Assay Supernatant was diluted in coating buffer (1:2, 1:4 and 1:8), and coated directly onto the plate. After blocking and washing, the basic protease MAb was added, followed by horse radish peroxidase-conjugated anti-mouse antibody to detect a signal. This ELISA test demonstrated the existence of specific immunoreactive product in the supernatant of culture DB403(pNC3kb) with nothing reacting from the control DB403 (pNC-CTL) (Fig. 6B). Therefore no Bacillus native protein was reacting with the basic protease MAb. The level of basic protease produced by DB403 (pNC3kb) was 2-3x of that of the S198 supernatant, the latter containing a number of other proteases not recognised by the MAb.

These results indicated that the bprV gene was expressed in B.

subtilis. The fact that the gene product in the culture medium was biologically active suggested that the protease was secreted using its own signal peptide and that the secreted gene product was processed correctly in this heterologous system. In order to confirm that the release of active protease is truly a result of proper protein secretion rather than due to "leaking" from cell lysis, the total protein contents inside the cell and in the growth medium were analysed by SDS-PAGE and Western blot. The results are shown in Fig. 7.

Panel A is a stained SDS-PAGE gel showing the total protein profiles for both DB403{pNC3kb} and DB403fpNC-CTL}. The extracellular protein patterns in lanes 1 and 2 from DE403(pNC-CTL) and DB403(pNC3kb}, respectively, were almost identical indicating that significant cell lysis of DB403(pNC3kb) had not occurred in comparison to the control culture. The Western blot analysis in Panel B revealed several points. Firstly in the culture supernatants (lanes 1 and 2) BprV protease was detected only from culture DB403(pNC3kb), confirming the results obtained from the protease assay and ELISA. Secondly, there was only one band detected in lane 2, which has an apparent molecular weight of about 37 kDa (as indicated by the solid arrow).Thus the recombinant BprV protein migrates at a position the same as that predicted from the authentic protease, this would indicate that the BprV products were processed correctly in B. subtilis. Thirdly, Western blot on intracellular proteins (lanes 3 and 4) further suggested that there is no accumulation of pre-pro-BprV products inside the cell. The open arrow on the right side of the figure points to a position where the predicted pre-pro-BprV will migrate. As one can see there is no obvious signal in this region. There are a few bands in the range of 60 to 90 kDa present in both lanes 3 and 4 which were detected by cross reactive antibodies in the polyclonal antiserum used for the Western Blot analysis.Thus the results demonstrate that at a low level of expression there is no accumulation of precursors inside the cell and that the Gram-negative signal peptide seems to function efficiently in B. subtilis.

Purification and Characterisation of the Recombinant Protease Recombinant basic protease was purified from DB403 (pNC3kb) culture supernatant to establish the identity of the expressed product to the mature enzyme. The supernatant from 1 litre of DB403 (pNC3kb) culture was concentrated in an ultrafiltration cell with a Diaflo YM10 membrane (Amicon) to about 80ml., dialysed extensively against 0.01 M Tris-HCl, 5mM CaCl2 pH 8.6 and then further concentrated to 40ml. The concentrate was chromatographed on a sulfopropyl-(SP)-Sephadex C-25 colum (20 x 2 cm) equilibrated in the above buffer and the bound protein was eluted with a linear gradient of 0-0.3 M NaC1 over 400 ml.

A protein peak with proteolytic activity eluted at about 0.2 M NaCl, as found previously for native D. nodosus basic protease (Lilley et awl., 1992, Eur. J. Biochem., 210:13-21). SDS-PAGE of this protein peak showed a major band (Mr37 kDa) with the same mobility as that of the native basic protease which reacted with a sheep anti-basic protease antibody on Western blot analysis. The lower Mr bands co-eluting with the major 37 kDa protein did not react with this antiserum.

The identification of the 37 kDa protein as the recombinant basic protease was further confirmed by N-terminal sequence analysis. After electroblotting to Selex 20 membrane (Scheicher and Schuell), the protein was subjected to sequencing analysis on an Applied Biosystems 470A sequencer.

N-terminal sequence of the recombinant protein of AAPNDPSYRQQ confirmed that it was the product of the bprV gene and that both signal peptide and propeptide had been correctly cleaved.

The amino acid composition of the intact recombinant protease and the HPLC peptide profile generated on autolysis of the recombinant protease indicated that the 127 aa C-terminal extension peptide (Lilley et al., supra) was also correctly removed to yield a recombinant protease product that was identical to the native enzyme. The final yield of the purified recombinant protease was around 2 mg per litre.

It has been found that purified D. nodosus basic protease from virulent/intermediate strains requires an ionic strength of about 0.2M to remain soluble at pH 8.6 and to exhibit maximum activity during assay (Figure 7). Dialysis against 0.01M Tris HC1-5mM CaC12, pH 8.6 to remove the salt after the SP-Sephadex C-25 chromotography results in the quantitative precipitation of the protein as microcrystals which are subsequently difficult to redissolve. In contrast the acidic proteases such as V5 and V2/B2 are soluble at low ion strength.

Interestingly the basic protease from benign strain 305 is also soluble at low ionic strength i.e. not requiring salt for solubility and therefore does not form microcrystals presumably because of the two additional negative charges (2 Asp residues) resulting in a lower pI for the benign strain basic protease.

The precipitation into microcrystals, which have also been obtained from basic protease expressed in B. subtilis or E.

coli, may influence the choice of a suitable purification method. These microcrystals also have veterinary and medical vaccine applications, in particular, the protease vaccine for footrot since the heterologous modified, native basic protease vaccine which contained microcrystals of native basic protease was found to be as effective as an homologous pili vaccine and induced better protection and 10-fold higher mean antiprotease antibody titres that that induced by unmodified (soluble) native basic protease (Stewart D.J., Kortt A.A., and Lilley G.G. New approaches to footrot vaccination and diagnosis utilising the proteases of Bacteroides nodosus in: Advances in Veterinary Dermatology Vol. 1: Proceedings of the First World Congress of Veterinary Dermatology, Dijon, France, 1989, pp 359-369. (Eds. C. von Tscharner and R.E.W. Halliwell).London Balliere Tindall, 1990). Microcrystals may also be formed with other proteases of other protein protective antigens with high isoelectric points (e.g. pI > 9.0-9.5).

Example 4: Preparation and Testing of Vaccines This following example illustrates the use of E. coli and Bacillus subtilis expressed basic serine protease in the preparation of vaccines and in the use of such vaccines for inducing both protection and cure in sheep against footrot.

The experiment was conducted on irrigated pasture at the CSIRO Werribee Field Station utilising flood irrigation to provide sufficient moisture for pasture growth and transmission of footrot. Eighty four mature Merino wethers were ranked on body weight and randomly distributed into 6 equal groups of 14. Six donors infected with pure cultures of B. nodosus strain 198 (serogroup Al) by the method of Egerton and Roberts (1971) as modified by Stewart, et a. (1985), were introduced into the experimental flock at the time of the first vaccination. The sheep in all groups received two 2ml doses of vaccine subcutaneously in the neck region with the second dose of vaccine being administered to the appropriate group 26 days after the first. Sheep were bled at the first and second vaccination and 21, 43 and 71 days after the second (47, 64 and 97 days after the first). The feet of all sheep were also inspected at these times and the lesions evaluated by the scoring system of Stewart, et al., (1982, 1983b).

Vaccines: 1: Strain 198 (serogroup Al) recombinant pili, 50pg per dose - two doses.

2: Strain 198 native basic protease (crystalline form), 50pg per dose - two doses.

3: E. coli recombinant protease (crystalline form), 50yg per dose - two doses.

4: B. subtilis recombinant protease (crystalline form), 50yg per dose - two doses.

5: Strain 198 native basic protease (soluble form), 50pg per dose - two doses.

6: Adjuvant controls.

Recombinant pili were prepared from Pseudomonas aeruginosa PAK2fS harbouring a chimeric plasmid derived from pVSI (Itoh, et al., 1984, Plasmid, 11:206-220). and the pilin gene of D.

nodosus strain 198 controlled by the strong prokaryotic PR promoter - Strain 198 native basic protease was purified from bulk cultures of virulent strain 198 grown anaerobically at 370C for three days in Trypticase-arginine-serine (TAS) broth containing 0.15% CaCl2 as described in Australian Patent No.614754. After removal of the cells by centrifugation, the culture supernatant was concentrated on a Diaflo spiral-wound ultrafiltration membrane cartridge Model SlYlO (Amicon mol.wt.

cut-off about 10,000) and in an Amicon ultrafiltration cell using a Diaflo YM10 membrane (mol.wt. cut-off about 10,000).

The concentrated supernatant was fractionated by gel filtration on Sephadex G-100, followed by ion-exchange chromatography of the (high molecular weight) proteolytic activity peak eluting near the column void volume on SP Sephadex C-25 using the conditions described earlier. The proteolytic activity, eluted from the SP-Sephadex C-25 column with the salt gradient, contained the basic protease and was homogeneous as judged by PAGE.

The basic protease expressed in E. coli and B. subtilis was purified as described on page 51.

The amount of protein in the protease vaccines prior to crystallisation and in the soluble protease vaccine was estimated by absorbency at 280 nm assuming an extinction coefficient of 10. The extinction coefficient is defined as the absorbency of a 1% solution measured in a lcm spectrophotometer cell. The vaccines were diluted in 0.05M Tris-HCl pH7.8 (vaccines 1 and 6), 0.01M Tris-HCl 5mM CaCl2 pH9.6 (vaccines 2, 3, 4) or 0.01M Tris 5mM CaCl2 0.2M NaCl pH9.6 (vaccine 5). Vaccine 1 (recombinant pili) also contained 0.5t v/v formalin (37% formaldehyde). Four parts of the aqueous phase of the vaccines were mixed with one part by volume of Alhydrogel (2% Al(OH)3,Superfos Biosector a/s, Denmark) and the mixture was then emulsified with incomplete Freund's adjuvant (Difco) in the ratio 1:2.

Results: All of the basic protease vaccines (soluble and crystalline native and crystalline recombinant E. coli and B. subtilis) induced both a marked curative and protective effect against natural field challenge with the homologous strain of D. nodosus (Tables 10,11). This effect was particularly evident after the second dose of vaccine with far fewer feet affected with either interdigital skin or underrunning lesions than the adjuvant controls and comparable with the results obtained with the recombinant pili vaccine. Although the recombinant pili vaccine performed slightly better than the basic protease vaccines after the first dose, the vaccines were of similar high efficacy after the second dose.

All of the basic protease vaccines induced a substantial increase in mean anti-basic protease antibody titres after the first and second dose in contrast to the minimal rise in mean titres observed in the recombinant pili and adjuvant control groups (Table 12) . The crystalline native basic protease vaccine compared to the soluble native basic protease vaccine had higher mean antibody titres 43 days after the second vaccination (Table 12). This higher level of antibody response for the former compared to the latter was also evident 71 days after the second vaccination (mean anti-basic protease titres 10, 366 and 3,315, respectively). This experiment differed from that described in Stewart, Kortt and Lilley, 1990, supra (see page 52). Firstly the pH of the vaccines was 9.6 compare to pH < 9.0 in the latter.Secondly, the aqueous phase of the latter vaccines aalso contained an equal volume of Al hydrogel.

Samples of lesion material from vaccinated, adjuvant control and donor sheep were cultured 21 days after the second vaccination and the serogroup identity of the freshly isolated stains of D. nodosus was checked with typing antisera in agglutination tests (Claxton, et al., 1983). The virulence characteristics of the fresh isolates were checked by colony morphology, elastase, proteinase thermostability and zymogram tests (Stewart, et al., 1986c). The results of these tests on 6 different swabs from 5 sheep confirmed that virulent strain 198 belonging to serogroup Al was the only strain present in the flock.

TABLe :10 Results of a field experiment showing number and percentage of feet affected with footrot (lesion scores 2 to 4 inclusive) in sheep challenged with strain 198 at the time of the first vaccination.

Vaccine 26/8/92 26 days PV1 21 days PV2 43 days PV2 71 DAYS PV2 22/12/92 12/1/93 3/2/93 3/3/93 1. Monovalent recombinant pilus (50 g) 0/56* 50.1# 10/56 3/52# 49.5 3/52 47.4 2/52 49.2 2. D. nodosus native basic protease 0/56 50.5 19/56 8/52# 50.5 5/52 48.7 9/52 51.2 (insoluble) (50 g) 3. E. coli recombinant basic protease 0/56 50.5 21/56 11/56 49.3 5/56 46.6 6/56 51.1 (insoluble) 4. B. subtilis recombinant basic 0/56 49.9 23/56 10/56 47.5 6/56 45.4 4/56 49.4 protease (insoluble) 5. D. nodosus native basic protease 0/56 50.6 27/56 12/56 49.3 8/56 47.2 10/56 48.6 (soluble) (50 g) 6. Adjuvant control 0/56 50.3 35/56 33/56 46.1 26/56 44.5 29/56 54.4 Natural challenge initiated with strain 198 at first vaccination PV1 Post vaccination one PV2 Post vaccination two Second vaccination given 26 days PV1 *No. of feet with footrot ( > 2 lesions) out of 14 sheep in each group #Body weight (Kg) #One sheep in each of vaccine groups 1 and 2 died of natural causes TABLE : 11 Results of a field experiment showing number and percentage of feet affected with severe lesions (#3c) of footrot in sheep challenged with strain 198 at the time of the first vaccination.

Vaccine No. feet with severe footrot ( > c lesions) 26 days PV1 21 days PV2 43 days PV2 71 Days PV2 1. Monovalent recombinant 0/56* 3/52# (5.8)# 2/52 (3. 0/52 (0) pilus (50 g) 8) 2. D. nodosus native basic 2/56 8/52# (1 3/52 (5. 2/52 (3.8) protease (insoluble) 5. 8) (50 g) 4) 3. E. coli recombinant basic 1/56 5/56 (8 2/56 (3. 1/56 (1.8) protease (insoluble) .9 6) ) 4. B. subtilis recombinant 1/56 9/56 (1 6/56 (1 0/56 (0) basic protease (insoluble) 6. 0.

1) 7) 5. D. nodosus native basic 3/56 10/56 1) 7/56 (1 4/56 (7.1) protease (soluble) (50 g) 7. 2.

9) 5) 6. Adjuvant control 6/56 27/56 (48.2) 24/56 4) 20/56 (35.7) 2.

9) PV1 Post vaccination one #% feet PV2 Post vaccination two Natural challenge initiated with strain 198 at first vaccination.

Second vaccination given 26 days PV1 *No. of feet with severe footrot (#3c) out of 14 in each group +One sheep in each of vaccine groups 1 and 2 died of natural causes TABLE :12 Mean ELISA anti-basic protease antibody titres to the Strain 198 basic protease in vaccinated and control adjuvant sheep.

ELISA anti-basic protease antibody titres Vaccine Group 26 days PV1 21 days PV2 43 days PV2 1. Strain 198 recombinant pili 16.5 36.5 88.3 2. D. nodosus native basic protease 895 2,668 7,672 (crystalline) 3. E. coli recombinant basic protease 1,754 9,233 10,370 (crystalline) 4. B. subtilis recombinant basic protease 346 3,613 7,805 (crystalline) 5. D. nodosus native basic protease (soluble) 1,084 6,009 4,659 6. Adjuvant control 16.6 43.3 122 Example 5: Improvement in the production of basic protease using B subtilis regulatory elements Although the native basic protease gene (bprV) was expressed and secreted in B. subtilis using its own signal peptide (Example 3), its level of production was low. To increase the production of BprV in B. subtilis various gene constructs were made using the B. subtilis promoter, ribosomal binding site and signal peptide derived from the genes coding for B.

subtilis neutral protease (nprE) and alkaline protease (aprE).

Materials and Methods 1. PCR primers used in this study are listed in Table 13.

2. Unique restriction enzyme site (PstI) was introduced into the bprV gene at the two putative signal peptide cleavage sites C1 and C2 (Fig. 15) using PCR primers P1 and P2 (see Table 13). By cutting with PstI, the sequence coding for signal peptide was removed forming a pro-BprV gene fragment ready to be fused with heterologous signal peptides - 3. B. subtilis nprE gene was modified by PCR with primer P3 (Table 13) to introduce a unique PstI site at the signal peptide cleavage site. The same was done for B. subtilis aprE gene using primer P4 (see Table 13).

4. Fusion genes were constructed by joining B. subtilis gene elements (including promoter and upstream region, ribosomal binding site and signal peptide) with the pro BprV gene fragment either at C1 or C2 sites utilising the unique PstI site engineered at the ends of all these gene fragments - 5. Modification of bprv gene ribosomal binding site was carried out by PCR using primer P5 (Table 13). The native ribosomal binding site of bprV gene was changed into a synthetic ribosomal binding site which had a maximal sequence homology to the consensus ribosomal binding site sequence derived from published B. subtilis gene sequences.

6. All of the above recombinant genes were cloned in B.

subtilis plasmid pNC3 or pUB18 and expressed in B.

subtilis after transformation.

Results and Discussion 1. Up to date, three fusion genes have been constructed and tested for their protease production. The results are presented in Table 14. All of the fusion genes produced higher level of basic protease than the native gene in pNC3kb. The recombinant gene containing the aprE signal peptide fused to bprV C2 site was the best in terms of protease production and plasmid stability.

2. Modification of ribosomal binding site increased the production of basic protease about 5-fold (Table 14), but was not as effective as the B. subtilis gene elements as presented above.

3. As demonstrated in Table 14, the highest protease production (around 40 mg/litre) was obtained from the pAC2-bprV gene construct. Further enhancement of protease production is anticipated by using other B.

subtilis regulatory genes which are known to be able to enhance protease production and by integrating the fusion gene into B. subtilis chromosome to stabilise the gene construct in vivo.

Table 13 Oligo nucleotide primers used in this study Name Sequence (5'---- > 3') Enzyme Site(s) P1 AGCGGATCCACTG CAGGTCAAGTTTGCGC BamHl,Pst P2 TTTGCG CTG CAGAAAGTATCGTAAA(C,T)TA(C/T) GAATC Pstl P3 AGTTGACGATACTTTCTG CAG CCTGAACACC Pstl P4 ACTTTCTGCAGCCTGCGCAGACAT Pstl P5 AAAGGAGGTGAACATATGAATCTATCGAACATT Ndel Table 14 Comparison of protease production Plasmid RBS SP F-site Vector BprV Yield pNC3kb bprV bprV n/a pNC3 + pNC1-bprV nprE nprE C1 pNC3 ++++ pAC1-bprV aprE aprE C1 pNC3 ++++ pAC2-bprV aprE aprE C2 pUB18 +++ + + pRBS-bprV syn bprV n/a pUB18 8 Abbreviations:RBS, ribosomal binding site; SP, signal peptide; F- site, fusion site; n/a, not applicable; syn, synthetic.

Example 6: Expression of acidic protease aprV5 gene in E.

coli and B. subtilis This example details the expression of aprV5 gene in both E.

coli and B. subtilis.

Materials and Methods 1. PCR primers used in this study are listed in Table 15.

2. Since the complete aprVS gene seemed to be unstable in E.

coli, a shortened gene fragment without the C-terminal extension was obtained by PCR cloning using primers P1 and P2. A stop codon was designed in P2 to terminate the translation of aprV5 gene immediately after the end of mature enzyme sequence.

3. A unique StyI site was introduced into the aprV5 gene in the pro-peptide region by PCR using primer P3. This Styl site was engineered so that it will be in-phase with the unique StyI site present at the same position in bprV gene.

4. Construction of a bprV-aprVS fusion gene was done by ligation of two gene fragments joining at the unique StyI site which was present in the native bprV gene and introduced in the aprV5 gene as described above.

5. Both the native aprV5 gene and the bprV-aprVS fusion gene were cloned in E. coli in plasmid pBR322 and expression was carried out after transformation into strain DH5a.

6. The same gene constructs for E. coli expression were cloned in B. subtilis in plasmid pNC3 and expression was carried out after transformation into strain DB403.

Results and Discussions 1. When the C-terminal extension of the aprVS gene was removed and a translation stop codon was introduced at the end of the mature enzyme sequence, the thus modified gene was able to be cloned in E. coli. Low level of protease expression was detected (Table 16). Exchange of the signal peptide by the bprV signal peptide in pBR-BaprV5 resulted in slight increase in protease production. This indicated that the signal peptides between these two proteases might be interchangeable. Another interesting finding was that although secreted enzyme in the culture medium was detected, the majority of active enzyme was still cell-associated, probably trapped in the periplasmic space or associated with inner or outer membrane.

2. When the aprV5 gene was introduced into B. subtilis strain DB403, no protease activity was detectable using either plate or liquid assay. On the other hand, the fusion gene in pNC-BaprV5 expressed low level of AprV5 protease. This might be explained by the difference in ribosomal binding site or signal peptide between bprV and aprVS.

3. Further analysis of C-terminal extension on the expression and secretion of aprVS may be required before high level expression could be achieved. In addition, the observation that bprV signal peptide could function in the secretion of aprVS protease could mean that the strategies and constructs used for the improved expression of bprV gene in B. subtilis (presented above in Example 4) would also work for the expression of aprVS in B. subtilis.

The results obtained in Example 1 have demonstrated that a new laboratory diagnostic test incorporating monoclonal antibodies can enable the rapid detection and virulence differentiation of D. nodosus directly in hoof lesion material or from broth cultures. The results indicate that it is now possible to positively and rapidly (i.e. within 24 hours of sample acquisition) identify either benign or virulent strains in lesion exudate from infected sheep without culture.

Alternatively, the monoclonal antibody test for detection of virulent and benign proteases can be used on D. nodosus broth cultures. Kits encompassed by the present invention for diagnostic testing may include benign (B2) protease MAb and/or virulent (V2) protease MAb, optionally together with either or all of basic protease MAb, total (V2/B2 or V5/B5) protease MAb and acidic protease fraction G100 Peak 1 MAb (low intermediate MAb) Preferably, kits would include all five monoclonal antibodies. Binding fragments, for example Fab or F(ab1), of these monoclonal antibodies may also be suitable.

The new diagnostic test avoids the need to culture D. nodosus and can also be readily standardised. It may speed the process of decision making on treatment options and quarantine by inspectors with savings in labour because return property visits should not be necessary for confirmation of diagnosis.

Early diagnosis will assist in reducing the spread of virulent footrot and facilitate earlier treatment and eradication thereby reducing the severe financial, managerial and emotional burdens incurred with virulent footrot. There will be applications for rapid diagnosis in accreditation schemes, saleyards, travelling sheep, movement of sheep into Protected Areas, disease surveillance, prevalence surveys and litigation cases.

Further, subtilisin-like proteases when used as an enzyme formula have wide applications as detergent, laundry detergent, soap powder and dishwashing compositions as well as the dairy, food, textile, leather, pharmaceutical, medical and cosmetic industries and veterinary applications. These applications are also to be considered within the scope of the present invention.

The application of proteases to catalyse the synthesis of peptides has been well documented (Kullmann, W. (1986) Enzymatic Peptide Synthesis. CRC Press Inc. Boca Raton, Florida, USA; Whittaker, R.G. & Schober, P.A. (1990), Todav's Life Sci. 2: 60-64) and offers a number of advantages over classical chemical synthesis methods such as stereospecificity of reaction and that reactive side chains of amino acids do not have to be protected. Many of the well known proteases such as papain, trypsin, chymotrypsin, chymopapain, subtilisin, proteinase K, bromelain and elastase have been evaluated for their suitability for peptide synthesis. The most important attribute of a protease for synthetic application is its specificity.The finding that the specificity of D. nodosus proteases V2, V5 and basic protease is distinct from that of subtilisin and proteinase K indicates that the D. nodosus proteases may be useful for catalysing peptide synthesis. This application of the subtilisin-like proteases is also to be considered within the scope of the present invention.

Other aspects of the present invention, and modifications and variations thereto, will become apparent to those skilled in the art on reading this specification, and all such other aspects and modifications and variations are to be considered as included within the scope of the present invention.

Table 15 Oligo nucleotide primers used in this study Name Sequence (5'---- > 3') Enzyme Site(s P1 TTCTGTTCTGCGGATCCTGCTTTTTGAAGCGCGGCG BamHI P2 CGCTTGCGGTTTGAATTCGTCGATCGGACGTTAAAC EcoRI P3 GGCACGCCATCAAGCCAAGGGATGACCAACCGCAG StyI Table 16 Expression of aprv5 gene lacking C-terminal extension Plasmid SP F-site Vector Yield B. coli pBR-aprV5 aprV5 n/a pBR322 ++ pBR-BaprV5 bprV StyI pBR322 B. subtilis pNC-aprV5 aprV5 n/a pNC3 pNC-BaprV5 bprV Styl pNC3 O Abbreviations: SP, signal peptide; F-site, fusion site; n/a, not applicable.

Claims (60)

The claims defining the invention are as follows:

1. A diagnostic assay for Dichelobacter nodosus infection comprising detecting in a sample one or more of the following: (i) an epitope present in virulent strains and/or at least a portion of intermediate strains, but not present in benign strains; (ii) an epitope present in benign strains and/or at least a portion of intermediate strains, but not present in virulent strains; (iii) an epitope present in some intermediate strains but not present in virulent strains or the majority of benign strains; (iv) antibodies specific to an epitope present in virulent strains and/or at least a portion of intermediate strains, but not present in benign strains; (v) antibodies specific to an epitope present in benign strains and/or at least a portion of intermediate strains, but not present in virulent strains;; and/or (vi) antibodies specific to an epitope present in some intermediate strains but not present in virulent strains or the majority of benign strains,

2. A diagnostic assay according to claim 1 wherein the epitope present in virulent strains and/or at least a portion of intermediate strains but not present in benign strains, is an epitope present on an acidic protease.

3. A diagnostic assay according to claim 2 wherein the epitope is present on V2 or G1OOPI acidic protease fraction.

4. A diagnostic assay according to claim 1 wherein the epitope present in some intermediate strains but not present in virulent strains or the majority of benign strains, is an epitope present on an acidic protease.

5. A diagnostic assay according to claim 4 wherein the epitope is present in V1.

6. A diagnostic assay according to claim 1 wherein the epitope present in benign strains and/or at least a portion of intermediate strains but not present in virulent strains, is an epitope present on an acidic protease or a basic protease.

7. A diagnostic assay according to claim 6 wherein the epitope is present on B2.

8. A diagnostic assay according to claim 1 wherein the antibodies specific to an epitope present in virulent strains and/or at least a portion of intermediate strains but not present in benign strains, are specific to an epitope present on an acidic protease.

9. A diagnostic assay according to claim 8 wherein the antibodies are specific to an epitope present on V2 or G100P1 acidic protease fraction.

l0.A diagnostic assay according to claim 1 wherein the antibodies specific to an epitope present in benign strains and/or at least a portion of intermediate strains but not present in virulent strains, are specific to an epitope present on an acidic protease or a basic protease.

1l.A diagnostic assay according to claim 10 wherein the antibodies are specific to an epitope present on B2.

12.A diagnostic assay according to claim 1 wherein the antibodies specific to an epitope present in some intermediate strains but not present in virulent strains or the majority of benign strains, are specific to am epitope present on an acidic protease.

13.A diagnostic assay according to claim 12 wherein the antibodies are specific to an epitope present in V1.

14, An antibody capable of binding to a Dichelobacter nodosus epitope present in virulent strains and/or at least a portion of intermediate strains but not present in benign strains.

15. An antibody according to claim 14 wherein the epitope is present on an acidic protease.

16. An antibody according to claim 15 wherein the epitope is present on G10OP1 acidic protease fraction or V1, V2 and/or V3.

17. An antibody capable of binding to a Dichelobacter nodosus epitope present in benign strains and/or at least a portion of intermediate strains, but not present in virulent strains.

18. An antibody according to claim 17 wherein the epitope is present on an acidic protease or a benign protease.

19. An antibody according to claim 18 wherein the epitope is present on B1, B2, B3 and/or B4.

20. An antibody capable of binding to an epitope present in all strains of Dichelobacter nodosus.

21. An antibody according to claim 20 wherein the epitope is present on V5/B5 and/or V2/B2.

22. Dichelobacter nodosus GlOOP1 acidic protease fraction,

23. An antibody capable of binding to Dichelobacter nodosus G100P1 acidic protease fraction.

24. Dichelobacter nodosus benign basic protease in substantially pure form.

25. An antibody capable of binding to an epitope unique to Dichelobacter nodosus benign basic protease.

26. An isolated DNA molecule encoding an acidic protease or portion thereof of Dichelobacter nodosus.

27. An isolated DNA molecule encoding Dichelobacter nodosus acidic protease V2 or a portion thereof.

28. An isolated DNA molecule encoding Dichelobacter nodosus acidic protease V5 or a portion thereof.

29. An isolated DNA molecule encoding Dichelobacter nodosus acidic protease B2 or a portion thereof.

30. An isolated DNA molecule encoding Dichelobacter nodosus acidic protease B5 or a portion thereof.

31. An isolated DNA molecule encoding Dichelobacter nodosus benign basic protease,

32. An isolated DNA molecule comprising the nucleotide sequence substantially as shown in Figure 12 or a portion thereof.

33. An isolated DNA molecule comprising the nucleotide sequence substantially as shown in Figure 3 or a portion thereof.

34. An isolated DNA molecule comprising the nucleotide sequence substantially as shown in Figure 2 or a portion thereof.

35. An isolated DNA molecule comprising the nucleotide sequence substantially as shown in Figure 10 or a portion thereof -

36. An isolated DNA molecule comprising the nucleotide sequence substantially as shown in Figure 4 or a portion thereof.

37. A host cell containing a DNA according to any one of claims 26 to 36.

38. A method for producing a Dichelobacter nodosus protease or portion thereof comprising introducing an expression vector including a DNA molecule according to any one of claims 26 to 36 or a portion thereof into an appropriate host cell, expressing the protease or portion thereof, and isolating the protease or portion thereof thus produced.

39. A method according to claim 38 wherein the DNA molecule encoding the protease is operably linked to a sequence encoding a heterogenous signal peptide.

40. A method according to claim 39 wherein the sequence encoding a heterogenous signal peptide, encodes the signal peptide from B. subtilis neutral protease or alkaline protease and the host cell is B. subtilis.

41. A protease or portion thereof produced by the method according to any one of claims 38 to 40.

42. A vaccine comprising a protease according to any one of claims 22, 24 or 41, together with a pharmaceutically acceptable carrier therefor.

43. A vaccine according to claim 42 further comprising an adjuvant.

44. A method for producing a desired protein or peptide in a gram-positive or gram-negative bacterial host, comprising introducing to an appropriate host cell and expression vector including a nucleotide sequence encoding the desired protein or peptide and a nucleotide sequence coding for, or substantially homologous to, a signal peptide derived from a gene encoding a serine protease from Dichelobacter nodosus.

45. A method according to claim 44 wherein the appropriate host cell is E. coli or B. subtilis.

46. A method according to claim 44 or 45 wherein the nucleotide sequence coding for the signal peptide may be derived from a basic protease gene from Dichelobacter nodosus.

47. A method according to any one of claims 44 to 46 wherein the nucleotide sequence coding for a signal peptide codes for a signal pep tide of the amino acid sequence in MNLSNISAVKVLTLVVSAAIAGQVCAA or functional equivalents thereof,

48. A method according to any one of claims 44 to 47 wherein the desired protein or peptide is a protease or virulence factor from Dichelobacter nodosus or other pathogenic organism.

49. A hybrid protease molecule characterised by a heterogenous epitope located within a loop region in the protease molecule -

50. A hybrid protease molecule according to claim 49 wherein the protease portion corresponds to a protease from Dichelobacter nodosus or Xanthomonas campestris.

51. A hybrid protease molecule according to claim 50 wherein the heterogenous epitope is located within a loop region at about amino acid 73-104.

52. Recombinant DNA molecules encoding hybrid protease molecules according to claims 49, 50 or 51.

53. A host cell containing a DNA according to claim 52.

54. An antibody capable of binding to an acidic protease of a low intermediate subcategory of Dichelobacter nodosus strains comprising 312, 478, 639, 718 and 733.

55. A kit comprising an antibody according to any one of claims 14 to 21, 23, 25 and 54.

56. Purified peptide or polypeptide having an isoelectric point (pI) > 9.0-9.5 in microcrystal form.

57. Purified peptide or polypeptide according to claim 56 wherein the peptide or polypeptide is a protease or portion thereof according to any one of claims 22, 24 or 41.

58. A vaccine comprising purified peptide or polypeptide according to claim 56 or 57, together with a pharmaceutically acceptable carrier there for.

59. A vaccine according to claim 58 further comprising an adjuvant.

60. The use of subtilisin-like proteases in detergents for household, dairy, food, textile, leather, pharmaceutical, medical, cosmetic and veterinary purposes.

61 The use of Dichelobacter nodosus proteases to catalyse the synthesis of proteins or peptides.