CA1341040C - Pasteurella vaccine - Google Patents

Abstract

A vaccine for immunizing animals, including human beings, against diseases caused by microorganisms producing an osteolytic toxin, the vaccine comprising an immunologically effective amount of a recombinant, immunogenic, detoxified Pasteurella multocida toxin or toxin analogue. The vaccine is encoded by a DNA fragment comprising a nucleotide sequence coding for a Pasteurella multocida toxin or toxin analogue, the nucleotide sequence being derived from a Pasteurella multocida genome, from a bacteriophage infectious for Pasteurella multocida, or from a plasmid or being a synthetic sequence which is inserted in an expression vector capable of replicating in a suitable host microorganism in which the said DNA fragment may be expressed in terms of production by the host microorganism of Pasteurella multocida toxin or toxin analogue when it is cultivated under suitable conditions, the produced toxin or toxin analogue optionally being subjected to posttransscriptional modifications to provide the detoxified toxin or toxin analogue. Alternatively, the nucleotide sequence coding for the toxin or toxin analogue may be modified prior to insertion into the expression vector so as to provide the expressed toxin or toxin analogue in a detoxified form. A monoclonal antibody has been raised against a Pasteurella multocida toxin which may be useful as a diagnostic agent for determining the presence of a Pasteurella multocida toxin or toxin analogue in a sample.
Further-more, the said monoclonal antibody may be useful for isolating a Pasteurella multocida toxin or toxin analogue from a biological material. A further scope for a useful diagnostic agent which is provided for is the application of a labelled DNA sequence homologous with a DNA sequence coding for a Pasteurella multocida toxin or toxin analogue for the detection of Pasterualla multocida toxin-producing microorganisms. The toxin or toxin analogue per se may be usefully applied for determining the presence of antibodies against a Pasteruealla multocida toxin or toxin analogue in a sample and may also be useful as a reference standard in analytical procedures aiming at quantitative or qualitative determination of the said toxin or toxin analogue in a biological sample or intermediate or final vaccine preparations.

Description

A PASTEURELLA VACCINE ~ 3 4 1 0 4 0 The present invention relates to a vaccine for immunizing animals against diseases caused by microorganisms producing an osteolytic toxin, a DNA sequence encoding a Pasteurella multocida toxin useful for producing tlhe toxin and as a diagnostic agent, methods of pro-ducing and isolating a .P. multocida toxin, use of a P. multocida toxin, a monoclonal antibody directed against a P. multocida toxin, a diagnostic agent comprising said monoclonal antibody and the use of said monoclonal ,3ntibod;y for a variety of diagnostic and other purpo-ses.
Atrophic rhiniti:~ is a disease which profoundly affects the bone structure of the porcine snout. The etiological agent which is cur-rently considered to be the cause of growth retarding progressive atrophic rhiniti:~ is toxigenic (toxin-producing) strains of P. multo-cida which colonize the nasal cavity of pigs (Pedersen and Barfod, 1981, (ref,. 1), Mutter and Rojas, 1982, (ref. 2), Elling and Pedersen, 1985, (ref. 3), Pedersen et al. 1988 (ref. 4). It has been shown that the nasal mucosa are more easily colonized by P. multocida when the resistance to infection is lower such as when the pigs are concomitantly in:Eected with Bordetella broachiseptica or when the nasal mucosa are exposed to a mild chemical irritant (cf. Pedersen and Elling, 1984, (ref. 5).
The pathological effects of P. multocida infection may be ascribed to a toxin produced by this bacterium. The toxin which has an apparent molecular weight of 143 kd and an actual molecular weight of 146.5 kd induces bone res~~rption (osteolysis) of the nasal turbinates and other bone structures in the nasal cavity by stimulating osteoclast activity in porcine turbinate bones, and causes impaired osteoblastic bone formation.
_, The disease is of mayor economic importance to pig breeders all over the world, sincca apart from the pathological effects on the nasal (and occasional:Ly facial) bones noted above, it causes a slower growth rate of t:he infected pigs and consequently higher production costs. Attempts have therefore been made to reduce the occurrence and the significance: of P. multocida infection, for instance by the establishment of SPF (specific pathogen free) pigs via cesarean sec-tion, or by anti:bioti:c treatment of infected animals or prophylactic vaccination.
Known vaccines ~~or the immunization of animals, principally pigs, against disease:. ascribable to P. multocida infection, especially atrophic rhiniti.s, camprise killed P. multocida cells optionally com-bined with killed Bordetella bronchiseptica cells (cf. EP 85 469) and/or an inactivated (usually by heat treatment or addition of formaldehyde) toxin-containing extract of toxi.genic P. multocida.
Vaccines of the latter type are commercially available from Northern Drugs & Chemicals Ltd., Copenhagen, Denmark, under the trademark Atrinord~, as well as from Intervet International BV, Boxmeer, Holland, under t:he trademark Nobi-vacART~.
The present inventors contemplate that an improved immunogenic effect relative to the know-ra ;-accine preparations may be obtained by using a purified and suitably modified toxin preparation for vaccination pur-poses either to replace: the conventional vaccines or as a constituent thereof.
The purification of P. multocida toxin has previously been described.
Thus, Foged et s~l., 1987, (ref. 6) disclose the purification of the toxin by chromatography and polyacrylamide gel electrophoresis. The purified toxin i.s used solely for studying its toxic and pathological effects. Kamp et: al., 1.987, (ref. 7) also disclose the purification of the P. multoc~ida toxin for the purpose of clinical studies. They suggest that the: purified toxin may be used as an antigen to raise specific antibodies useful for serological tests. Nakai et al., 1984, (ref. 8) disclose a method of purifying the P. multocida toxin by chromatograpt~~y and polyacrylamide gel electrophoresis. They further disclose: the production of polyclonal antibodies directed x against the purified toxin which they use to analyse the purity of the purified toxin. It is suggested that the antibodies may be used to further study the role of the toxin in atrophic rhinitis.
None of these publications suggest the use of a purified toxin as a component of a vaccine for immunizing animals against Pasteurella infection, and this is believed to be a novel concept.
Accordingly, in one aspect the present invention relates to a vaccine for immunizing an animal, including a human being, against diseases caused by microorganisms producing an osteolytic toxin, the vaccine comprising an immunogenically effective amount of a recombinant, immunogenic, detoxified P. multocida toxin or toxin analogue together with an immunologically acceptable carrier or vehicle.
For the preparation of the known vaccines, a toxigenic Pasteurella strain is cultivated and the toxin is isolated from the culture medium or from a bacterial extract followed by detoxification by, for instance, thermal or chemical treatment. Compared to this procedure, the production of the toxin or toxin analogue by recombinant DNA
techniques has a number of advantages: it is possible to produce the toxin or toxin analogue by growing a non-pathogenic organism, the toxin or toxin analogue may be produced in higher quantities than those produced by wild-type P. multocida strains, for instance by using a strong promoter to induce a high level of expression of the toxin gene or by using a. high copy number vector for cloning the toxin gene, and it is possible to produce the toxin or toxin analogue in a detoxified form, e.g, by subjecting the gene encoding the toxin to treatment with a mutagen, or by deleting a part of the nucleotide sequence coding for the toxin or toxin analogue, substituting one or more nucleotides in the sequence, etc. The recombinant toxin or toxin analogue may be used in substantially pure form in the vaccine of the invention but may also be employed as a crude or partially purified preparation.
azssasei.oos~nsyHS~ame~ni.M~o~.oa.~ses In the present context, the term "substantially pure" is understood to mean that the vaccine is substantially free from other immunogeni-sally active components the presence of which might give rise to ad-verse immune reactions in the animals immunized with the vaccine and, most importantly, that no other components of the microorganisms pro-ducing the toxin or toxin analogue, such as cell debris or cellular proteins apart from the toxin or toxin analogue itself or a protein or polypeptide to which t:he toxin or toxin analogue is fused (vide below) are present: in t:he>. vaccine preparation. A high purity of the detoxified toxin or toxin analogue is believed to result in a high antitoxin responses on immunization with the vaccine of the invention and a lower dosage of t:he toxin or toxin analogue may consequently be required for immunization purposes than? that used in crude or partially purified vaccine preparations. A substantially pure toxin or toxin analogue has t:he: added advantage that the exact concentra-Lion thereof' in a given vaccine preparation is known so that an exact dosage may be admi.nistere:d to the animal in question.
The microorganism producing an osteolytic toxin (i.e. a toxin direct-ly or indirectly involved in bone resorption) against which the vac-sine confers immunity is preferably P. multocida. Other microorga-nisms which have :shown osteolytic effects or regulation of specific markers of bone me~tabol.ism are e.g. Actinomyces viscosus and Bordetella pertus~:is (7.'rummel et al., 1979, (ref. 9) and Price (ref.
10).
Due to the toxic activity of the P. multocida toxin, it is not pos-Bible to use the native toxin in a vaccine of the invention. Rather, it must be present: in detoxified form. The term "detoxified" should be understood to mean that the toxic activity has been removed from at least a sufficient number, but not necessarily all, of the toxin molecules present in the vaccine preparation so that the vaccine, when administered to an animal to be immunized, will not produce any adverse toxic effects i.n the animal in question, while still giving rise to a satisfacaory immune response.
The detoxification of t:he P. multocida toxin or toxin analogue may be carried out in a variety of ways. Thus, it is possible to subject the 426585B1.003/A31 /HS/ALM/~~LM/07.C14.1989 ~3~10~ 0 toxin or toxin analogue to thermal treatment, the toxin being known to be heat labile and t:o be inactivated (i.e. detoxified) at 70°C.
Furthermore, the toxin or toxin analogue may be subjected to treat-ment with a chemical, such as formaldehyde, glutaraldehyde or a 5 suitable proteolyt:ic enzyme, e.g. trypsin. Detoxification may also be brought about by mutagenizing the gene coding for the P. multocida toxin or toxin analogue> by means of, for instance, ultraviolet radia-tion, ionizing radiation or a chemical mutagen such as mitomycin C, 5-bromouraci.l, met:hylme>.thane sulphonate, nitrogen mustard or a nitro-furan. Furthermore:, the: toxin may be detoxified by substitution, deletion, addition or insertion of one or more amino acids in the toxin or toxin analogue, or by substitution, addition, deletion or insertion of one or more base pairs in the nucleotide sequence coding for the toxin or i:oxin analogue, or a combination of these measures.
In contrast to detoxification by thermal or chemical treatment, the genetic procedure as the obvious advantage of resulting in a uniform population of equally detoxified molecules.
It should be noted that the terms "substitution, deletion, addition or insertion" should be :interpreted with reference to the full-length toxin protein. Thus, "substitution" is intended to mean the replace-ment of any one or more amino acids or nucleotides in the full amino acid or nucleotide: sequence with one or more others, "addition" is understood to mean the addition of one or more amino acids or nucleo-tides at either end of the full amino acid or nucleotide sequence, "insertion" is intended to mean the introduction of one or more amino acids or nu<:leotides within the full amino acid or nucleotide sequence, and "de:Letion" is intended to indicate that one or more amino acids or nu<:leotides have been deleted from the full amino acid or nucleotide sequence whether at either end of the sequence or at any suitable point= within it. It should be understood that the detoxification of the toxin or toxin analogue may also be brought about by a combination of two or more of these procedures.
The term "toxin analogue" is used in the present context to indicate a protein or polyi~eptide of a similar amino acid composition or se-quence as the P. tm,iltocida toxin, allowing for variations which do 42658581.003/A31 /HS/ALM/ALM/07.o4.1989 not have an adverse effects on the immunogenicity of the analogue.
The analogous pol5~pepti.de or protein may be derived from a microorga-nism of another species than P. multocida or may be partially or completely of synthetic: origin. The analogous polypeptide or protein may also be one which comprises at least one epitope reactive with anti-P. multocida toxin antibodies found in samples from individuals with atrophic rhinitis and/or which elicits antibodies reactive with native P. multocida toxin. The term is further intended to mean any immunogenic subsequence, functional equivalent or derivative of the toxin.
The term "immunoge:nic subsequence" is intended to indicate a sequence of the full-length toxin which from the outset is produced in a truncated form relative t:o the full-length toxin protein or which subsequent to production of the full-length protein is generated, for instance by proteolytic: c:leavage thereof or by expression of a nucleotide sequence short:er than the full nucleotide sequence encoding P. multocida toxin. The minimum subsequence is one which at least comprises a relevant epitope of the toxin, i.e. an epitope which gives rise t:o a reT.evant immune response in an animal immunized with the vaccine of the invention.
The term "functional equivalent" is intended to include all immunoge-nically active substances with the ability to evoke an immune respon-se in animals to which a vaccine containing the equivalent has been administered which is similar to the immune response evoked by the detoxified P. muli:ocida t:oxin, in that it is able to confer immunity to diseases caused by mic:roorganisms producing an osteolytic toxin.
The functional equivalent: may be derived from a microorganism of another species than P. multocida or may partially or completely be of synthetic. origin. It: should be understood that the similarities between the P. mW tocida toxin and the functional equivalent are qualitative rather than quantitative, relating to the nature rather than the level of activity of the functional equivalent.
The term "derivati.ve" is understood to mean a modification of the toxin such as one produced by substitution, insertion, addition or deletion of one oz- more amino acids or nucleotides or a combination 426585BL003/A31 /HS/ALM/A~LM/07.04.1989 of these measures, ;~s defined above, or by fusion with another polypeptide In a further aspect, the present invention relates to a DNA fragment comprising a nucleotide sequence: coding for a P. multocida toxin or toxin analogue, as defined above. The DNA fragment may for instance be used in a method of preparing the toxin or toxin a~:~alogue by recombinant DNA techniques or as a diagnostic agent (i.e. a DNA 1>robe),.
In a still further aspect, the ;present invention relates to a monoclonal antibody which is directed against or reactive with a P. multocida toxin or a toxin analogue as defined above, or a fragment of said antibody. It should be noted that the antibody may be re~~ctive with both the toxic and detoxified toxin, thus making it useful for a variety of diagnostic, immunization and isolation purposes as will be described in further detail below.
The invention is further disclosed in the following with reference to the drawings in which Fig. 1 is a graph showing the titration of PMT in a quantitative sandwich ELISA.
The absorbance at 4692 nm obtained in the ELISA is plotted against the PMT
concentration. The minimum detectable concentration of PMT is about 1 ng/ml corresponding to about 50 pg or 0.35 final.
Fig. 2 shows an SD~S-PAGE of fractions from the affinity chromatography described in Example 3. Lane A: the culture supernatant applied on the column, lane 0: the effluent from the column, lane E: the eluted purified PMT, and lane M: molecular weight marker proteins, ~ 341 04 0 Fig. 3 is a Western blot showing PMT production by the S positive recombinant E. coli clones detected in the screening procedure. Lane 1: SPE 301; lanes 2 and 3: SPE 308; lane 4: SPE 315; lanes 5 and 6:
SPE 312; lane 7: SPE 311; lane 8: purified PMT.
Fig. 4 is a restriction enzyme cleavage map of the plasmid pSPE 308 with a length of 21.5 kb (kilobase pairs). The hatched area denotes P. multocida DNA, the shaded area denotes the pmt gene and the verti-cally hatched area denotes plasmid pUN121 DNA.
Fig. 5 is a restriction enzyme cleavage map of the plasmid pSPE 312 with a length of 13.8 kb. The hatched area denotes P. multocida DNA, the shaded area denotes the pmt gene and the vertically hatched area denotes plasmid pUN121 DNA.
Fig. 6 is a restriction enzyme cleavage map of plasmids constructed by enzymatic cleavage of the plasmid pSPE 308. The hatched area deno-tes P. multocida DNA, the shaded area denotes the pmt gene, and the vertically hatched area denotes pUN121 DNA.
Fig. 7 is a restriction enzyme cleavage map of plasmids constructed by enzymatic cleavage of pSPE 312. The hatched area denotes P. multo-cida DNA, the vertically hatched area denotes pUN121 DNA, and the shaded area denotes the pmt gene.
Fig. 8 is a Western blot showing PMT production by derivatives of plasmids pSPE 308 and pSPE 312. Lane 1: purified PMT; lane 2: pSPE
350; lane 3: pSPE 349; lane 4: pSPE 341; lane 5: pSPE 345; lane 6:
pSPE 312; lanes 7 and 8: purified PMT. Plasmid pSPE 349 is identical to plasmid pSPE 347 shown in Fig. 7.
Fig. 9 is a restriction enzyme cleavage map of the pmt gene. The shaded area denotes t:he pmt gene, the vertically hatched area denotes a probable promoter, and the hatched area denotes a probable terminator.
' 30 Fig. 10 (a)-(j) shows the DNA sequence of the pmt gene region and the amino acid sequence deduced on the basis of the DNA sequence. The a34ao4o 7(b) amino acids are identified with single-letter codes according to con-ventional usage. The amino acid sequence has been shown to start at position 213 or 219.
Fig. 11 is a restriction enzyme cleavage map of the plasmid pSPE 525 with a length of 7.7 kb. The hatched area denotes P. multocida DNA, the shaded area denotes the pmt gene, and the vertically hatched area denotes pUN121 DNA.
Fig. 12 is a restriction enzyme cleavage map of the expression vector pSPE 481 with a length of 8.25 kb. The hatched (towards the right) area denotes P. multocida DNA, the shaded area denotes the pmt gene, the hatched (towards the left) area denotes aPL DNA, the crosshatch-ed area denotes the amp gene, and the vertically hatched area denotes the origin of replication.
Fig. 13 is a restriction enzyme cleavage map of the toxA coding region. Extension of coding region present on each derivative plasmid (pSPE A-R) is indicated. (A-R) by bars. Hatched bars: Coding region in correct reading frame; open bars: Coding region not in frame with the 5' part of the coding region.
Fig. 14 is a Western blot showing the recognition by a mouse anti-PMT
antiserum of PMT derivatives produced by the plasmids pSPE A-L. Lanes 7, 13, 14 and 15: different strains harbouring the entire pmt gene;
lane 1: derivative A; lane 2: derivative I; lane 3: derivative $.;
lane 4: derivative J; lane 5: derivative L; lanes 6 and 9: derivative E; lane 8: derivative C.; lane 10: derivative G; lane 11: derivative H; lane 12: derivative D. Approximate sizes (in kilodalton) of pro-minent full-length derivatives and degradation products are indica-ted.
Fig. 15 is a graph showing the distribution of relative absorbances (A/Ao) by PMT-EL,ISA of extracts of non-cytopathic (hatched bars) and cytopathic (black bars) field isolates of P. multocida diluted 1:1 in PBS-T-BSA.
a X341 0~ 0 ~~~>
Fig. 16 is a graph showing the mean tSD of relative absorbances (A/Ao) of dilut9_ons of extracts of cytopathic (black squares) and non-cytopathic I;open squares) field isolates of P. multocida.
Fig. 17 is a graph showing the presence of anti-PMT-antibodies in serum samples from anti-PMT-antibody-negative, infected and vaccina-ted pigs detected by competive ELISA. The graph shows the 50X block-ing titers at an absorbance of 492 nm.
.... negative ---- infected vaccinated Fig. 18 shows the colony hybridization of P. multocida isolates, testing 17 toxin-positive and 18-toxin-negative strains as determined by the ELISA and EBL cell tests for the presence of the pmt gene.
Fig. 19 shows the determination of toxic activities of cell-free sonicates of recombinant E.coli clones. E.coli strain MT102 with PUN121 had no cytopathic effect on EBL-cells when diluted 1;25 in PBS (a). Sonicates of E.coli SPE312 (b) and toxigenic P.multocida (NCTC 12178) (c) diluted 1/3125 showed significant and identical effects (80 X magnification).
Fig. 20 shows the P.multocida DNA flanking the pmt-gene (black area).
The extension c~f the. inserts of the plasmids pSPE308, pSPE312, pSPE344, pL0A03~ and pLOB03 are indicated. The DNA contained in the probes used for the blotting (slant hatched area), and the fragments which contain t:he two homologous sequences (vertically hatched area) are shown.
Fig. 21 shows a Southern blot of restriction enzyme digested P.multocida DNE~. Probe: 2,4 kb BglII-EcoRI fragment of pLOB03. Lanes 10*-14* is a short time exposure of lanes 10-14.
Lanes 1-4: Tox:Cgenic P.multocida 45/78. Lanes 5-9: Nontoxigenic P.muItocida MH81P8. Lane 10: pSPE308. Lane 11: pL0A03. Lane 12:
w pL0A02. Lane 1:3: pSPE312. Lane 14: pLOB03.

7(d) Restriction enzyme;s used: +HindIII: Lanes 1, 5, 10, 11, 12 and 13.
EcoRI: Lanes 2, 6 and 14. BgII: Lanes 3 and 7. PvuII: Lanes 4 and 8.
PstI: Lane 9.
Fig. 22 shows a dot blot of 24 different P. multocida bacteriophage genomes.
Probe: pLOA03. 'The probe does not hybridize to B2 and C5. A7: pSPE308;
B7: pSPE312, C7: pLOA0:3 and D7: pLOB03 are positive controls.
The toxin produced by P. multocida (in the folowing occasionally abreviated to PMT) which, as noted above, is generally believed to be the causative agent of porcine atrophic rhinitis, has in the prior literature been variously termed "dermonecrotic toy;;in", osteolytic toxin", "turbinate atrophy toxin" and "heat labile exotoxin", but it would appear to be the same toxin as the amino acid composition, isoele;ctric point and biological activities of the variously termed toxins show basic ~~imilarities, although minor variations in the properties of toxins isolated fronn different strains of P. multocida appear to exist. The estimated amino acid composition of PMT (as deduced from the DNA sequence) is as follows:
Ala is 76 times 5.91 found =

Cys is 8 times 0.62 found =

Asp is 71 times= 5.53 found Glu is 100 times 7.
found == 78 %

Phe is 69 times 5 found = .

X

Gly isfound 71 times 5.53X
-His isfound 19 times; 1.48X
-Ile isfound 92 times 7.16X
-Lys isfound 70 times 5.45%
-Leu isfound 127 times 9.88%
-Met isfound 36 times 2.80X
-=

Asn isfound 73 time; 5.68%
-Pro isfound 62 time:; 4.82%
-=

Gln isfound 56 times 4.36%
-Arg isfound 58 times 4.51%
=-Ser isfound 97 time.; 7.55%
_=

Thr isfound 66 times; 5.14%
_-Val isfound 63 time; 4.90X
_-=

Trp isfound 18 times 1.40X
.-Tyr isfound 53 time:; 4.12X
_-The total number of amino acid residues is 1285, and the full-length toxin has a molecular weight of 146.5 kd.
The recombinant tc>xin or toxin analogue used in the vaccine of the invention may mores specifically be one encoded by a DNA sequence sub-stantially as shovm in FiLg. 10 (a)-(j) or a subsequence thereof en-coding an immunoge~nic sut>sequence of the toxin or toxin analogue. It should be noted that the amino acid sequence deduced from the DNA
sequence is also :shown in Fig. 10 (a)-(j) above the DNA sequence. A
suitable analogue may t>e one which has a DNA sequence which differs from that of the native' toxin by one or more base pairs and which may be derived by substituting one or more nucleotides in the toxin DNA
sequence either giving ruse to the same amino acid sequence, but where the nucleotide substitutions make the sequence conform to the codon usage of the: microorganism in which the sequence is inserted, or giving rise to a somewhat different amino acid sequence which, however, is functionally similar to that of the native toxin.
Apart from the toy;in or toxin analogue as defined above, the vaccine of the invention ~ilso comprises an immunologically acceptable carrier or vehicle. This vehicle may be any vehicle usually employed in the preparation of vaccines, e.g. a diluent such as isotonic saline, 426585B1.00.3/A31 /HS/ALM/f~LM/07.(I4.1989 suspending agent: etc. The vaccine may be prepared by mixing an immunogenically effective amount of the toxin or toxin analogue with the vehicle in an amount resulting in the desired concentration of the toxin or town analogue in the vaccine preparation. Although the amount of toxin or toxin analogue per unit dose of the vaccine will differ according to the age of the animals to be immunized (for instance according to whether sows or piglets are to be immunized against P, multocida), the route and form of administration, and the immunogenicity of the particular toxin present in the vaccine, a suitable amount of taxin or toxin analogue is contemplated in the range of 0.1-50C1 ~g per dosage of the vaccine.
The vaccine may further. comprise an adjuvant in order to increase the immunogenicity of the vaccine preparation. The adjuvant may be se-lected from Fret~nd's complete or incomplete adjuvant, aluminium hydroxide, Bordeatella pertussis, a saponin, a muramyl dipeptide, an iscom (immune stimulating complex; cf. for instance EP 109 942) and an oil, such as a vegetable oil, e.g, peanut oil, or a mineral oil, e.g. silicone oj:l.
In some cases it: may be advantageous to couple the toxin or toxin analogue to a c~~rrier, in particular a macromolecular carrier. The carrier is usua7lly a polymer to which the toxin is bound by hydropho-bic non-covalent: interaction, such as a plastic, e.g. polystyrene, or a polymer to which the toxin is covalently bound, such as a polysac-charide, or a polypeptide, e.g. bovine serum albumin, ovalbumin or keyhole limpet hemocyanin. The carrier should preferably be non-toxic and non-allergenic. The toxin or toxin analogue may be multivalently coupled to the macromolecular carrier as this provides an increased immunogenicity ~~f the 'vaccine preparation. It is also contemplated that the toxin ~~r toxin analogue may be presented in multivalent form by polymerizing the toxin or toxin analogue with itself.
In a particular embodiment of the vaccine of the present invention, the toxin or toxin analogue as defined above is fused to ;mother polypeptide. Techniques for preparing fused polypeptides are known. The fusion may be provided by fusing the nucleotide sequence encoding the 13~~ ~4 0 toxin to a nucleotide sequence encoding another polypeptide so that the fused nucleotide sequence, when inserted in an appropriate vec-tor, is expressed as a fusion polypeptide on transformation of the vector to a suitat~le microorganism and growth of the microorganism 5 under conditions favorable to the expression of the fused sequence.
The polypeptide to which the toxin is fused may, for instance, be a carrier polypeptide as suggested above, lysozyme or another im-munogenic peptide such a~: a Ty protein from Saccharomyces cerevisiae, protein A from St~~phylacoccus aureus, Hepatitis B core antigen, etc.
10 It is also contemplated that the vaccine may be in the form of tab-let, granule or capsule intended for oral administration since there is some evidence that i.mmunogens may be absorbed through the intes-tinal wall and stimulate B-lymphocytes which then migrate to local epithelial regions. where they transform into immunoglobulin-secreting plasma cells. An oral vaccine should be provided with an enteric coating in order t:o protect the toxin or toxin analogue from substan-ces present in gastric juice which might be deleterious to the toxin or toxin analogue, such as pepsin. The enteric coating may be selec-ted from shellac, cellulose acetate esters such as cellulose acetate phthalate, hydroxypropylmethyl cellulose esters such as hydroxypro pylmethyl cellulose phthalate, polyvinyl acetate esters such as polyvinyl acetate phthalate, and polymers of methacrylic acid and (meth)acrylic acid ester::. Newly developed methods of encapsulations, based on microsphe;res with a diameter of about 5-15 ~cm are of special interest since such particles containing an immunogenic substance after administration will_ be selectively delivered to Peyers's patches thereby providing immunity on mucosal surfaces. Stimulation of the immune response on respiratory mucosal surfaces may also be obtained through i.ntranasal immunizations. (Mestecky, 1987, (ref.
12).
The DNA fragment of the invention comprising the nucleotide sequence encoding the toxin or toxin analogue may be derived from complemen-tary cDNA obtained by preparing a cDNA library on the basis of mRNA
from a toxin-producing P.. multocida strain by standard methods.
Alternatively and preferably, the nucleotide sequence may be derived from a P. multocida genome, by screening for genomic sequences hybri-426585B1.003/A31 /HS/ALM/p~LM/07.04.1989 dizing to a DNA probe prepared on the basis of the full or partial amino acid sequen~~e of t:he toxin in accordance with established procedures or by establishing a toxin gene library and screening for toxin-producing clones b:y means of a toxin-specific antibody (for a more detailed des~~riptaon of this procedure, see Example 4). In the case of PMT, it i;s not possible to prepare a DNA probe on the basis of its N-terminal amino acid sequence since PMT is blocked in the N-terminal and therefore is not degraded by procedures for the sequencing of amino acids.
Another routine screening method which has proven to be difficult in the case of PMT i;s screening for toxin-producing clones by means of an anti-PMT serum. When using serum from a rabbit repeatedly immunized with PM'T, the present inventors found 5 E.coli clones by the Colony blot method in the gene library described in Example 5.
Further studies o:E the above 5 clones, however, showed that none of them were producing PM'r. These results indicate the importance of performing the screening with anti-PMT monoclonal antibodies as described in Example S.
The nucleotide sequence may also be derived from a bacteriophage in-fectious for P. multoc.id3, i.e. one which has been transferred from one bacterial strain which originally carried the sequence to another strain which did not originally carry the sequence by bacteriophage transfection. Similarly, the nucleotide sequence may be derived from a plasmid or other genetic element transferred from one strain to another by conjugation, 'transformation or the like.
Furthermore, the nucleotide sequence coding for the toxin may be a synthetic sequence, that is, one prepared according to standard pro-cedures, e.g. as described in Matthes et al., 1984, (ref. 13).
Finally, the nucleotide sequence may be a mixed genomic and synthetic or mixed cDNA and synthetic sequence prepared by ligating DNA
fragments of genomic, cDNA or synthetic origin (as appropriate) which DNA fragments each contain part of the nucleotide sequence encoding the toxin, in accordance with established methods.
426585B1.003/A31 /HS/ALM/ALM/07.04. ~ 989 12 ~34i040 In accordance with the explanation given above, the DNA fragment may be one which has teen modified by substitution, addition, insertion or deletion of one or more nucleotides in the sequence with the pur-pose of establishing a sequence which, when expressed, results in the production of a detoxified toxin or toxin analogue.
In particular, the: invention relates to a DNA fragment which comprises a nucleotide sequence substantially as shown in Fig. 10 (a)-(j) or a modii:ication thereof as indicated above. The sequence coding for the ful_1-length toxin starts at position 219 (or 213) of the sequence shown in then figure, while the end of the sequence is at position 4073. The: DNA sequence shown in Fig. 10 (a)-(j) has been established by well-known methods as described in Example 7 below.
The DNA fragment of the invention may further comprise a nucleotide sequence encoding another polypeptide fused to the nucleotide sequen-ce encoding the toxin or toxin analogue with the purpose of producing a fused polypeptide, as explained above. A further purpose of pre-paring a fused po7_ypeptide may be to facilitate purification of the toxin. In this case, the fused sequence may be inserted into an ap-propriate vector which is transformed to a suitable host microorga-nism which i.s groom under conditions ensuring expression of the fused sequence after which the fused polypeptide is recovered from the cul-ture by subjecting the fused polypeptide to affinity chromatography involving an antibody or any other ligand reacting with the second polypeptide. After- purification, the second polypeptide may then be removed, for instance by suitable proteolytic cleavage followed by separation of the two polypeptides.
In a further aspeca, the invention relates to an expression vector which is capable of replicating in a host microorganism and which carries a DNA fragment as described above. The vector may either be one which is capable of autonomous replication, such as a plasmid, or one which is replicated with the host chromosome, such as a bacterio-phage. Specific e~:amples of expression vectors of the invention are the plasmids pSPE A-R described in Example 9 below and shown in the appended Fig. 13.
42658581.003/A31 /HS/ALM/A,LM/07.04.1989 In a still further aspe:ct:, the invention relates to a microorganism which is capable crf expressing a DNA fragment as defined above and which carries a vector as described above. The microorganism is preferably a bacts:rium, especially a gramnegative bacterium such as E. coli.
The invention also relates to a method of producing an immunogenic detoxified P. muli:ocida toxin or toxin analogue, the method compri-sing a) isolating a nuc:leoti.de: sequence coding for the P. multocida toxin or toxin analogue, b) inserting said sequence, optionally in suitably modified form resulting in the expression of the detoxified toxin or toxin analogue or a subsequence c:odin~; for an immunogenic subsequence of the toxin of toxin analogue, in an expression vector, c) transforming a suitable host microorganism with the vector pro-duced in step b), d) cultivating the: microorganism produced in step c) under suitable conditions for ex~>ressing the toxin or toxin analogue, e) harvesting the toxin ar toxin analogue from the culture, and f) optionally subjecting the toxin to posttranslational modifications to produce the detoxified toxin or toxin analogue.
In step a) of the method" the nucleotide sequence may for instance be isolated by estab7.ishing a P. multocida gene library and screening for toxin-positive: clones in accordance with established methods as indicated above a:; well as described in detail in Example 4 below.
In step b) of the methad, the modification of the sequence optionally carried out may be' performed before or after the sequence has been inserted in the vector. The modification may comprise substitution, 426585B1.003/A31 /HS/ALM/A~LM/07.04.1989 ._ 1 34~ 04 0 addition, insertion or deletion of one or more nucleotides in the sequence or a court>ination thereof, as explained above.
The transformation in step c) of the method may be carried out by standard procedures, such as disclosed in Maniatis et al. (ref. 14).
The cultivation of the host microorganism in step d) of the method may be carried out: in a culture medium conventionally used for fer-mentation purpose:;, e.g. Luria Broth medium, and under conditions with respect: to pEi, temperature, aeration, etc. suited to the type of microorganism in question, e.g. as disclosed in Maniatis et al. (ref.
14).
In step e) of the method" the harvesting of the toxin or toxin analo-gue may proceed b51 well-~Cnown methods such as by precipitation, gel filtration, ion e~:change or HPLC reverse phase chromatography or immunoaffinity chromatography.
If the nucleotide sequence coding for the toxin or toxin analogue has not been modified in step b) of the method to result in expression of the detoxified to~:in oz- toxin analogue, the toxin or toxin analogue should be subjected to posttranslational modifications in step f) of the method, for instance thermal treatment, treatment with a chemical such as formaldehyde, glutaraldehyde or a suitable proteolytic en-zyme, e.g. t:rypsin, or substitution, addition, insertion or deletion of one or more amino acids in the toxin or toxin analogue.
In a still further aspect, the invention concerns a method of produc-ing a vaccine for immuni::ing an animal, including a human being, against diseases c:ausecf by microorganisms producing an osteolytic toxin, the method comprising formulating the toxin or toxin analogue produced by recombinant 1)NA techniques or by peptide synthesis as described above with an immunologically acceptable carrier or ve-hicle, such as those indicated above.
In a further, interesting aspect, the present invention relates to a non-pathogenic microorganism which carries and is capable of express-ing an inserted nucleotide sequence coding for an immunogenic de-42658581.003/A31 /HS/ALM/f~LM/07.c14.1989 toxified P. multoc~ida toxin or toxin analogue for use as a live vaccine for the immunizat:ion of an animal against diseases caused by microorganisms producing an osteolytic toxin. The use of a live vaccine might be advantageous since there is some indication that 5 vaccines based on living organisms show an excellent immunogenicity, often conferring a lifelong immunity against the disease in question.
Live vaccines also tend t:o be less expensive to produce than those based on a purified protein, no purification step being required.
In order to provide expression of the toxin or toxin analogue in 10 detoxified form, t:he nucleotide sequence coding for the toxin or toxin analogue may be suit=ably modified, either before or after introduction into the rfost microorganism, by substitution, addition, insertion or deletion of one or more nucleotides in the sequence or a combination of these measoures, as explained above.
15 In a particularly advantageous embodiment of the live vaccine of the invention, the nucleotide: sequence coding for the toxin or toxin ana-logue is expressed on the' outer surface of the host cell. This pro-vides a favorable present:ai:ion of the toxin epitope(s) which will be recognized by the immune defense mechanisms of the animal to which the live vaccine i.s administered, thus provoking an appropriate immune response. Clne way of providing the expression of the toxin or toxin analogue on the c:el.l surface is to fuse the nucleotide sequence encoding the toxin or t:o};in analogue to another nucleotide sequence encoding a surfaces protein or a subsequence thereof (e. g. a signal peptide) which cause the toxin or toxin analogue to be expressed on the outer surface of the host cell, optionally as a fused polypepti-de. Examples of u=oeful surface proteins are adhesins, fimbrial pro-teins, e.g. the E. coli 1';88 or Type 1 fimbrial protein, or the Lama protein of E. coli.
The invention also relatea to the use of a recombinant, detoxified immunogenic P. multocida toxin or toxin analogue for preparing a vaccine for the immunization of an animal, including a human being, against diseases caused t>y microorganisms producing an osteolytic toxin. The toxin or toxin analogue used for immunization may be one 426585B1.00.3/A31 /HS/AlM/A~LM/07.04.1989 encoded by the DN~~ sequence shown in Fig. 10 (a)-(j) or a modifica-tion thereof as e:cplained above .
Similarly, the present invention relates to a method of immunizing an animal, including a human being, against diseases caused by micro-s organisms producing an osteolytic toxin, the method comprising ad-ministering to the anima:L an immunogenically effective amount of a recombinant detoxified immunogenic P. multocida toxin or toxin analogue, such as the one encoded by the DNA sequence shown in Fig.
(a)-(j) ar a modification thereof. The toxin or toxin analogue 10 may be administer<:d intravenously, intramuscularly, subcutaneously, intraperitoneally, orally or intranasally. It is contemplated that a suitable dosage range wi:Ll be 0.1-500 ~cg, dependent on the age and condition of the animal :in question, the route and form of ad-ministration and l:he immunogenicity of the toxin or toxin analogue.
In a preferred embodiment, the monoclonal antibody of the present invention is one raised against the P. multocida toxin produced by P.
multocida ssp. mu:(tocida 45/78, which is publicly available from the National Collection of Type Cultures (NCTC), Central Public Health Laboratory, London, England, with the accession number NCTC 12178.
In connection with the research leading to the present invention, several different monoclonal antibodies to the toxin produced by this Pasteurella strain have been prepared (vide Example 1 below), repre-sentative example:: of which are the ones produced by the hybridoma cell lines P3F37 Find P3F.'il. Samples of these cell lines were deposi-ted in accordance with the provisions of the Budapest Treaty on the International Recognition of the Deposit of Microorganisms for the Purpose of Patent Procedure, on 3 December, 1987, in the European Collection of Animal Cel~L Cultures (ECACC), Centre for Applied Microbiology and F;esearch, Porton Down, Salisbury, Wiltshire, Great Britain, with the accession numbers ECACC 87120301 and ECACC
87120302, respectively.
The monoclonal antibody of the invention may be prepared by a method comprising:
426585B1.003/A31 /HS/ALM/A~L.M/07.04.1989 ~34~0~0 a) Immunizing a suitable animal or animal cell with an immunogenic P.
multocida toxin o:r tox:in analogue to obtain cells producing an antibody to said 'toxin or toxin analogue, b) fusing cells p:roduc:ing the antibody with cells of a suitable myeloma cell line, and selecting and cloning the resulting hybridoma cells producing said antibody, or c) immortalizing ;an unfused cell line producing said antibody, e.g.
by viral transformation, followed by d) growing the ce:Lls of ;step b) or c) in a suitable medium to produce said antibody and harvesting the antibody from the growth medium.
The initial immunization of the animals in step a) of the method re-quires a modificavtion of the conventional method of producing mono-clonal antibodies disc:Losed by Kohler and Milstein, Nature 256, 1975, p. 495, since, PM'C, even when it is administered to mice in sublethal doses, will cause an atrophy of the spleen which seriously complica-tes the hybridoma technique. The animals must therefore initially be immunized with a detoxified toxin preparation as it has been found that any subsequent booster immunizations may be carried out with the native toxin which need not be detoxified before use. The immuniza-tion may otherwise be carried out according to conventional pro-cedures, e.g. by ,3dmin:istering a solution or suspension of the (detoxified) toxin or toxin analogue in a suitable solvent such as isotonic saline o:r phosphate buffered saline optionally containing an adjuvant such as one of those indicated above. Animals suited for immunization purposes are, for instance, mice, rabbits, goats, sheep, guinea pigs, etc. The bleeding of the animals and recovery of the polyclonal antibodies ma;y be performed by conventional procedures.
The antibody-producing cells used for fusion to the myeloma cells are preferably spleen or lymph cells (e. g. lymphoblasts, B-lymphocytes, plasma cells or related .cells isolated from spleen, lymph nodes or functionally related tissues). The fusion of antibody-producing cells and myeloma cells may be performed substantially as described by Kohler and Milste:in, supra, that is, preferably in the presence of a 426585B1.003/A31 /HS/ALM/AI.M/07.o4.'1989 _ ~ 341 04 0 fusion promoter such as polyethylene glycol. A ratio of about 3 anti-body-producing cells per myeloma cell is preferred. The myeloma cell line employed is F~referably of a type unable to survive in selective medium and unable to produce immunoglobulins by itself; one type of cell line frequently used for cell fusions is one which lacks the enzyme hypoxanthine-guanine phosphoribosyltransferase and which is consequently unable to grow in a medium containing hypoxanthine, aminopterin and thymidine (HAT medium).
The selection of r~ybridoma cells which produce an antitoxin-antibody may then be carried out: try culturing unfused antibody-producing cells, unfused myeloma cells and supposedly fused cells in a selec-tive medium (such as HAT) in which the unfused myeloma cells cannot grow and eventually die out. The unfused antibody-producing cells can only survive for s~ limi.te:d period of time after which they also die out. On the other hand, successfully fused cells continue to grow as they have inherited permanent growth properties from the parent myeloma cells and the ability to survive in the selective medium from the parent antibody-producing cells.
Alternatively to t:he production of hybridomas, immortalized unfused antibody-producing; cells may be produced either by transferring the genes responsible for production of immunoglobulin from the hybridoma to another (more viable) cell line or by viral transformation as de-scribed by K.lein ea al., (ref. 16).
The resulting antibody-producing cells (whether hybridomas or unfused cells) may be groGm in v~:tro after cloning, e.g. as described in Example 1 below, i.n a suiaable medium, such as RPMI 1640. This re-sults in the production of monoclonal antibodies of a very high purity as these are secreated into the culture supernatant by the cells. The antibodies may then be isolated by conventional methods such as centrifugation, filtration, precipitation, chromatography, or a combination thereof.
In an alternative method" the monoclonal antibodies may also be pro-duced in the body cavity of an animal such as a mouse. In this embo-diment, the antibody-producing cell is injected into an animal such 426585B1.003/A31 /HS/ALM/p,LM/07.C14.1989 as a mouse resulting in 'the formation of an ascites tumour which re-leases high concentrations of the antibody in the ascites of the animal. Although r_he animals will also produce normal antibodies, these will only amount to a minor percentage of the monoclonal anti-s bodies which may lie puri:Eied from ascites by standard purification procedures such a:a centr:ifugation, filtration, precipitation, chroma-tography or a combination thereof.
In a further aspect, the invention relates to a diagnostic agent which comprises a monoclonal antibody as defined above.
Although in some eases such as when the diagnostic agent is to be em-ployed in an agglutination assay in which solid particles to which the antibody is coupled agglutinate in the presence of a P. multocida toxin in the samp:Le subjected to testing, no labelling of the mo-noclonal antibody is necessary, it is preferred for most purposes to provide the antibody with a label in order to detect bound antibody.
In a double antibody ("sandwich") assay, at least one of the anti-bodies may be provided with a label. Substances useful as labels in the present content may be selected from enzymes, fluorescers, ra-dioactive isotopes and complexing agents such as biotin.
Examples of enzymes which may be used as label substances are peroxi-dases, e.g. horseradish peroxidase, or phosphatases. e.g, alkaline phosphatases. As Enzymes are not directly detectable, they must be combined with a substrate: to form a detectable reaction product which may, for instance;, be fluorescent or coloured. Examples of useful substrates are H202/o-phenylene diamine, H202/azinodiethylbenzthiazo-line sulphonic acid and p-nitrophenylphosphate.
Examples of fluore~.scers useful as label substances are H202/p-hy-droxyphenylacetic acid and methylumbelliferyl phosphate. Such sub-stances may be detected by means of a fluorescence spectrophotometer in a manner known per se.
Examples of radioactive isotopes useful as label substances are I-125, S-35 and P-32. I'he radioactivity emitted by these isotopes may 426585B1.003/A31 /HS/ALM/~~LM/07.04.1989 be measured in a ~~amma-counter or a scintillation counter in a manner known per se.
In a favoured embodiment, the diagnostic agent comprises at least one antibody covalent:ly or non-covalently coupled to a solid support.
5 This may be used :in a double antibody assay in which case the antibody coupled ~to the solid support is not labelled. The solid support may be composed of a polymer or may comprise a matrix on which the polymer is applied. The solid support may be selected from a plastic, e.g, l~itex, polystyrene, polyvinylchloride, nylon, polyvi-10 nylidene difluoride, cellulose, e.g. nitrocellulose and magnetic carrier particles (e. g. :iron particles coated with polystyrene).
For use in a diagnostic assay, the solid support may have any conve-nient shape. Thus, it may be in the form of a plate, e.g. a thin layer or, preferably, microtiter plate, a strip, film, paper or solid 15 particles such as latex beads or the like.
Rather than being coupled directly to the solid support, the mono-clonal antibody m;3y be coupled to a ligand immobilized on a solid support. Examples of l:igands include Protein A, or an immunoglobulin-specific antibody.
20 It should be noted that practically all methods or applications based on intact monoclonal or polyclonal antibodies could instead be performed using f:ragment;s of the monoclonal or polyclonal antibody, e.g. F(ab')2 or F;ab fragments (cf. Goding, 1978, (ref. 17).
For use in a sandwich assay, the diagnostic agent may additionally comprise a polyclonal antibody. This antibody may be labelled and/or coupled to a solid support as described above in connection with the monoclonal antibody.
The monoclonal anvtibody of the invention may be used in a method of determining the presence of a P. multocida toxin or toxin analogue in a sample, the method comprising incubating the sample with a monoclo-nal antibody as described above and detecting the presence of bound toxin or toxin an~3logue :resulting from said incubation. The antibody 426585B1.003/A3 t /HS/ALM/ALM/07.o4. ~ 989 may be provided with a label as explained above and/or may be bound to a solid support: as e:xe:mplified above.
In a favoured embodiment of the method, the sample is incubated with a first monoclona7_ antibody coupled to a solid support and subse-quently with a second monoclonal or polyclonal antibody provided with a label. An examp7.e of this embodiment is the sandwich ELISA (enzyme linked immunosorbe:nt assay) assay described in Example 2 below.
In an alternative embodiment (a so-called competitive binding assay), the sample may be incubated with a monoclonal antibody coupled to a solid support and simultaneously or subsequently with a labelled P.
multocida toxin ox' toxin analogue the latter competing for binding sites on the antitrody wit:h any toxin or toxin analogue present in the sample.
The sample subjected to t:he present method may be any sample suspect-ed of containing a P. multocida toxin or toxin analogue. Thus, the sample may be selected from bacterial suspensions, bacterial ex-tracts, culture supernatants, animal body fluids (e. g. serum, colo-strum or nasal mucous) and intermediate or final vaccine products.
Apart from the diagnostic: use of the monoclonal antibody of the in-vention, it is contemplated to utilize the well-known ability of cer-taro monoclonal antibodies to inhibit or block the activity of biolo-gically active antigens by incorporating the monoclonal antibody in a composition for the passive immunization of an animal, including a human being, against diseases caused by microorganisms producing an osteolytic toxin, which comprises a monoclonal antibody as described above and a suitable carrier or vehicle. The composition may be pre-pared by combining an effective immunizing amount of the antibody or fragment thereof with a suitable carrier or vehicle. Examples of suitable carriers and vehicles may be the ones discussed above in connection with the vaccine of the invention. The composition may further comprise an adjuvant such as one of those indicated above.
Based on experiments with mice (cf. Example 11 below) where the mono-clonal antibody induced passive immunity against PMT, it is contem-a2ssasai.ooa~na t ~HS~mnn~aiM~o~.oa. t sas plated that a PMT-specific antibody may be used for prophylactic or therapeutic treatment of atrophic rhinitis in pigs. It may be admini-stered intravenously, subcutaneously or intramuscularly as well as orally in suitably prot:ec:ted form or by means of an intranasal aero-sol.
A further use of t:he manoclonal antibody of the invention is in a method of isolating a P. multocida toxin or toxin analogue, the method comprising adsorbing a biological material containing said toxin or toxin analogue to a matrix comprising an immobilized mono-clonal antibody aso described above, eluting said toxin or toxin analogue from said matrix and recovering said toxin or toxin analogue from the eluate.
The matrix may be composed of any suitable material usually employed for affinity chromatographic purposes such as agarose, dextran, con-trolled pore glas:c, DEAE cellulose, optionally activated by means of CNBr, divinylsulphone, et:c. in a manner known per se.
The present invent:ion further relates to a diagnostic agent for the detection of PMT-producing microorganisms, which comprises a labelled DNA sequence homol.ogouse with a DNA sequence coding for a Pasterurella multocida toxin or toxin analogue. In this context, the term "homologous with" is intended to indicate that the DNA sequence comprises at least: one st:retch of deoxyribonucleotides of at least 15 bases with 80% homol.o~;y to a part of the shown sequence or to a part of the sequence encoding a toxin analogue.
In a method employing the: diagnostic agent, probe DNA is labelled, and the DNA is denatured to separate the strands in both probe and sample DNA; the DrfAs are mixed and the strands are left to reform the double helical structure, but in case of homology, some of the probe DNA will have combined wiah the sample DNA. This is known as hybridi-zation and is described by for instance Southern, 1980, (ref. 18).
The DNA used as the probe' should have a unique nucleotide sequence of a certain length i.n order to be sufficiently specific as a diagnostic agent. The probe DNA may advantageously be labelled with a radioac-tive isotope such as H-3,. I-125, S-35 or P-32 as described e.g. by 426585B1.003/A31 /HS/ALM/A,LM/07.o4.1989 Rigby et al., 1977, (ref. 19); a complexing agent such as biotin (Gebeyechu et al., 1987, (ref. 20); or with digoxygenin-dUTP
according to the method described by the manufacturer of the reagent, Boehringer, Mannheim.
In a particular embodiment of the invention, detection of the presence of Pasteurella multocida toxin producing microorganisms in a sample is performed by use of a DNA probe in the polymerase chain reaction procedures described by Saiki et al., 1985, (ref. 21). The pclymerase chain reaction (PCR) is a procedure used for the amplific-a.::i_on of DNA present in a sample. The procedure involves the use of t::~c; oligonucleotide pri.me~rs which flank the DNA segment to be ~~::plified. The oli.gonucle~otide primers may e.g. comprise the regions of the gene codin~; for Pasteurella multocida toxin or toxin analogue and may thus be u:ced to amplify the said gene or part of it present in a sample. The oligonuc:leotide primers hybridize to opposite strands of the DNE~ sequence to be amplified, and the primers are extended by using DNA polymerase, e.g. the Klenow fragment of E.coli DNA polymerase I or another useful DNA polymerase such as the Taq DNA
polymerase, so as to synthesize a DNA sequence which is complementary to the DNA sequence to which the primers are annealed. Subsequent to the synthesis of these complementary sequences, the DNA synthesized is denatured, e.g, by heating, from the "parent DNA strings", and the parent strings as well as the newly synthesized DNA strings are subjected to a new PCR amplification cycle. In this manner, it is possible to obtain a substantial amplification of specific DNA
sequences which are present in a sample. By use of the PCR amplifica-tion method, it may be possible to amplify and then detect originally very small and undetectable amounts of DNA sequences present in a sample.
In a still further aspect, the present invention relates to a method of determining thc: presence of antibodies against a P. multocida toxin or toxin analogue :in a sample, the method comprising incubating the sample with a P. multocida toxin or toxin analogue and detecting the presence of bound antibody resulting from said incubation.
426585B1.003/A31 /HS/ALM/I1LM/07.p4. ~ 989 A diagnostic agent comprising the toxin or toxin analogue used in this method may otherwise exhibit any of the features described above for diagnostic agents comprising the monoclonal antibody and be used in similar detection methods although these will detect bound anti-s body rather than bound toxin as such. The diagnostic agent may be useful, for instance as a reference standard or to detect anti-toxin antibodies in body fluids, e.g. serum, colostrum or nasal mucous, from animals exposed to the toxin or toxin analogue.
A still further use of a P.multocida toxin or toxin analogue is for the preparation of a toxin reference standard which may be useful as a standard of comparison in qualitative or quantitative analytical procedures. In .a qualitative procedure, the standard toxin in a known concentration may be reacted with a monoclonal or a polyclonal antibody raised against the toxin or toxin analogue, a positive reaction indicating the specificity of the antibodies. In another aspect, the reference .standard preparation is applied in a quantita-tive analytical procedure by which different concentrations of the preparation is :reacted with a monoclonal or a polyclonal antibody in order to provide a calibration curve which may allow the precise amount of toxin or toxin analogue in a sample to be estimated.
x Preparation of monoclonal antibodies against P. multocida toxin Immunization P. multocida toxin (PMT) was purified as described by Foged et al.
(ref. 6), i.e., by 50% (NH4)2504-precipitation of a cell-free extract of the toxigenic type D strain of P. multocida ssp. multocida 45/78 (refs. 3, 22) followed by DEAE-Sephacel~ chromatography and prepara-tive polyacrylamide gel electrophoresis (PAGE) in a manner known per se.
A suspension of the (NH4)2504-precipitate prepared as described above and containing approx. 25 ~g/ml of PMT was detoxified by incubation with 0.37% formaldehyde at 37°C for 1 h. Female BALB/c mice (8-12 weeks old) were immunized subcutaneously on day 0 and 14 with 300 pl of a 1:1 dilution of the crude preparation of detoxified P. multocida toxin and Freund's incomplete adjuvant (day 0) or PBS (day 14). On day 30 and 45 1 ,ug of native PMT in 200 ~1 PBS was injected subcuta-neously and on d.ay 60 the mice were boosted intravenously with 0.5 ~g of PMT in 100 ~1 PBS. Three days after the booster injection the mice were sacrified and their spleens removed far fusion.
Production of hybridoma cell lines and monoclonal antibodies According to procedures described by Fazetas et al. (ref. 23) and Gebeychu et al. (ref. 20), the spleen cells and P3-X63-Ag8.653 myeloma cells were fused using 50X PEG 4000 GK (Merck) and the resulting hybridoma cells were grown selectively in hypoxanthine/ami-nopterin/thymine (HAT) -supplemented RPMI 1640 medium containing 15%
fetal calf serum (FCS) and 4X human endothelial cell supernatant (Costar, The Net:herlands).
Hybridoma cell lines were selected by analyzing their respective monoclonal antibodies by ELISA and immunoblotting.

ELISA-for detection and titration of monoclonal antibodies Microtitre plates (96-well Immuno Plate II, Nunc, Denmark) were coat-ed with 50 ~1/well of a 0.75 ~g/ml solution of purified PMT in PBS at 4°C for 16 h, and at 20°C for 1 h. The wells were emptied and blocked with 200 ~1 PBS-T-BSA (PBS containing 0.05X (v/v) Tween~20 and 1X
(w/v) bovine serum albumin) per well at 20°C for 1 h., then washed 3 times with PBS-T. Fifty ~1/well of hybridoma culture supernatant was applied at 20°C for 1 h., and the plates were washed as described above. The anti-PMT antibody activity was measured colorimetrically after incubating at 20°C for 1 h. with 50 ~1/well of sheep anti-mouse immunoglobulin conjugated with horseradish peroxidase (Amersham In-ternational, U.K.) diluted 1:1,500 in PBS-T-BSA and (after 3 further PBS-T washes as above) with 50 ~1 of an o-phenylene dia~nine (OPD)-H202 substrate solution. The reaction was stopped with 150 ~cl of 2 M
H2S04 after 5 min. and absorbance was determined in a Kontron SLT-210 photometer (SLT Lab-instr., Zurich, Switzerland) at 492 nm (ref. 620 nm).
The mean absorbance at the saturation level of the titration curve (Asat) and the mean concentration of the MAb that resulted in SOX of the Asat (C50X) was determined by ELISA as described above, except that serial dilutions of the protein-A purified MAb in PBS-T-BSA was used. All 'results were based on at least duplicate determinations.
Immunoblotting To determine the specificity of the monoclonal antibodies, the pro-teins contained in a crude cell-free extract of P. multocida 45/78, were separated by SDS-PAGE before transfer to a nitrocellulose mem-brane and immunological. detection. Polyacrylamide gels (total acryl-amide: 10%, relative bi.s-acrylamide: 3X) and an electrophoresis buf-fer were prepared according to Laemmli (ref. 24). Electrophoresis was performed vertically at: 10°C at a constant voltage of 60 V for 16 h.
or 250 V for 4 h.. Pr°te:in-bands on gels were either visualized by silver staining with a detection limit of less than 1 ng of protein per band (8) or transferred to a nitrocellulose membrane (0.45 Vim) i using a semidry elect:roblotter (Ancos, 0lstykke, Denmark (9)). The proteins on the nitrocellulose membrane were either detected by a colloidal gold salver enhancement staining method (detection limit:
approx. 0.5 ng oj_ prote:Ln per band) (ref. 25) or immunologically by a modification of t:he method previously described by Bjerrum et al.
(ref. 26). A positive reaction in immunoblotting was recorded as + or (+), when an intense (or weak) staining of the PMT-band but no other protein band was observed. Staining of other bands or no reaction was recorded as -.
The molecular weight of PMT was estimated by comparison with known markers: ovalbuman (43.0 kd), BSA (66.3 kd), phosphorylase B (97.4 kd), ~-galactosidase (116.2 kd), RNA-polymerise ~ (150.6 kd) and x'(155.2 kd) and myosin (approx. 200 kd).
ELISA for estimating ep:itope specificity Estimation of apparent E:pitope specificity of anti-PMT MAbs was done by a competitive ELISA similar to a method described by Anderson et al. (ref. 27). Mi:crotitre plates were coated with PMT and blocked as described above. Fifty E~1 of the competing MAb diluted to 10 ~g/ml in PBS-T-BSA was added and incubated for 1 h. at 20°C. Without aspira-tion of the well:: 25 ~cl biotinylated monoclonal antibody was added ?.0 and the mixture was incubated for 20 min. st 20°C. After washings ~cl of a 1:2,500 dilution of horseradish peroxidase-conjugated avidin (Kem-En-Tec, Denmark) w:as added and the plates incubated for 45 min.
at 20°C. The substrate, reaction time and determination of absorbance were as described above.
The biotinylated MAb was used at a working dilution resulting in approx. 75X of the absorbance at the saturation level on the titra-tion curve. This curve was obtained by using a diluent instead of the competing MAb an<i seria:L dilutions of the biotinylated MAb. The extent of blocking by a competitive MAb was calculated according to the formula (1-A,iAo) x :LOOX, where A is the mean of the absorbance for three wells with the competing MAb and A° is the mean of the ab-sorbance of eight wells containing diluent instead of the competing MAb.

~34io4o The data of 10 :represe;ntative monoclonal antibodies (MAbs), all of the IgGl subclass, out of 92 ELISA-positive supernatants are shown in Tables 1 and 2.

Ch~iracter:ization of 10 representative MAbs Hybridoma Rep:resen- Asat C50X Immunoblotting group No . tat:ive MAb ng/ml 1 P3F'.il 1.2 110 +

2 P3F64 0.4 250 +

3 P3F:37 0.7 30 (+) 4 P4F.'i8 0.7 110 +

5 P3F:?2 0.6 35 +

6 P4F~i6 1.3 55 +

7 P4F:38 1.9 40 +

8 P4F.'i5 1.3 33 +

9 P3F.'i0 1..8 315 +

10 P3F.'i3 0.9 300 (+) a) Asat is the mean ab;sorbar_ce at 492 nm at the saturation level in the ELISA titration.
X

Extent of blocking by 1C1 representative MAbs in the competitive ELISA
Competing Biotinylat:ed detector MAb (X decrease in Aoa) MAb (hybri- _____________________________________________ doma group No. 1 2 3 4 5 6 7 8 9 10 P3F5i (1) 92 b P3F64 (2) 95 44 P3F37 (3) 63 95 78 P4F58 (4) 91 96 73 P3F22 (5) 71 88 P4F46c (6) 92 90 93 16 P4F38c (7) 93 92 95 27 P4F55c (8) 92 92 95 15 P3F50 (9) 16 24 13 84 91 P3F53 (10) 56 83 Ao 1.43 0.19 0.53 0.80 0.64 0.64 0.85 1.01 0.26 0.52 a) Ao is the mean absorbance with diluent instead of competing MAb b) No blocking (between 12X increase and 9X decrease in Ao) c) The closely related. hybridoma groups 6, 7 and 8 were differen-tiated by a two-site competitive ELISA using a catching MAb (method not described). Results indicated that group 6 was related to groups 3 and 4, group 7 to no other groups and group 8 to group 1.
The selected hybridoma cell lines were then cloned until they were stable. The resulting clones were then grown in "cell factories"
(Nunc, Denmark) at 37°C in RPMI 1640 medium supplemented with lOX FCS
as well as injected about 5x106 cells/mouse into Balb/c mice which after a certain incubation time leads to the formation of a tumour in the peritoneum of the mouse releasing high quantities of antibody in its ascites (about 5-~10 ml containing 5-25 mg/ml).
The hybridoma cell culture supernatants were passed through a protein A agarose column (Kem-En-Tec, Denmark). Bound antibodies were eluted 5 with 0.05M acetic acid, pH 4.0, or 0.03M citric acid, pH 3.0, and immediately neutralized with an appropriate buffer. Purified anti-bodies were biot:inylate~d as described by Guesdon et al., 1979, (ref.
28).
Two hybridoma cell li.nea, P3F37 and P3F51 shown in Table 1 to produce 10 MAb, were deposited on 3 December, 1987 in the European Collection of Animal Cell Cultures, Centre for Applied Microbiology and Research, Porton Down, Salisbury, Wiltshire SP4 OJG, U.K., with the Accession numbers ECACC 87120301 and ECACC 87120302, respectively.

15 Quantification of PMT
Quantification of PMT was carried out by a sandwich ELISA procedure.
The sandwich ELISA was initiated by coating each well of a microtitre plate (96 wells Immuno Plate II, ~:unc, Danmark) with 50 ~cl of 2 ug/ml of the anti-PMT MAb, P3F51 (produced in Example 1) in 0.05 M carbon-20 ate buffer, pH 9.6 for 16 hours at 4°C and 1 hour at 20°C.
Each well was incubated for 1 hour with 200 pI of phosphate-buffered saline containing 0.05X Tween 20 and 1X bovine serum albumine (PBS-T-BSA).
The plates could. be stored for at least 6 months by applying 20 ~1/well of PBS-sorbitol. and sealing with adhesive tape. The analysis 25 was initiated by two PBS-T washings followed by incubation of 50 ~1/well of solutions expected to contain PMT. The solutions were appropriately diluted i.n PBS-T-BSA and incubated for 1 hour at 20°C.
After 3 PBS-T washings each well was incubated with 50 ~1 of 0.5 ~sg/ml of the biotin c.or~jugated MAb, P3F37, for 1 hour at 20°C follo-30 wed by another 3 PBS-T washings and incubation with 50 ~1/well of a 1:2,500 dilution; of horseradish-peroxidase conjugated avidin (Kem-En-Tec, Denmark) for 45 min. at 20°C. Finally, 50 ~1/well of an o-pheny-lene diamine/H2C~2 solution was added. The reaction was stopped with 2M H2S04 after _'i min. and absorbance was determined in an Kontron SLT-210 photometer (aLT Labinstr., Zurich, Switzerland) at 492 nm (ref. 620 nm).
Calibration was performed with a PMT-preparation quantified by amino acid analysis (ref. Ei) and all quantitative data were means of at least dual determinations.
Using a sandwich ELISA, with the MAb P3F51 as catching antibody and biotinylated MAb P3F37 as detecting antibody, it was possible to de-tect less than _'i0 pg of PMT in a 50 ~1 sample. PMT at a concentration of 1 ng/ml resulted in an A492 of approx. 0.1 corresponding to more than 8 times the: background absorbance (cf. Fig. 1).

Affinity purification of PMT
About 100 mg of the pro tein-A purified MAb P3F51 prepared as descri-bed in Example 1. was coupled to 40 ml divinyl sulphone agarose (Mini-Leak; Kem-En-Tec" Denmark) as described by the manufacturer and loaded on a column (2._'ixl0 cm). The supernatant obtained by cultiva-tion of the toxi.genic type D strain P. multocida 45/78 was centri-fuged (12,000 x g for :30 min. at 4°C), filtered (Gelman, 0.45 Vim), mixed with 1/10 vol. oi: 1 M Tris-HC1, pH 7.7 and NaCl was added to 0.5 M before application to the affinity column. Repeated washings before elution of the column were carried out with an 0.1 M Tris-HC1 buffer containing first 1X Triton~ X-100, then 1.5 M NaCl and finally 0.1 M NaCl.. All washing buffers contained O.1X sodium azide and had a pH of 7.8. The fMT was eluted by 0.1 M glycine-HC1, pH 2.8 and im-mediately neutralized with 1 M K2HP04, pH 9Ø
The presence of PMT in the culture supernatant applied to the affini-ty column was indicated by the approx. 143 kd protein band seen by SDS-PAGE (Fig. ;?). The staining pattern of proteins in the material passing through the column (i.e. the effluent) was identical to that seen with mater:Cal before application, except for the 143 kd protein, which was retained on 'the column. Accordingly, the approx. 143 kd '' Trade-mark prote-in band is the anlLy staining seen when the protein composition of the eluted material is visualized by SDS-PAGE (Fig. 2).
Approx. 2.67 mg of the 3.41 mg of PMT applied to the column was eluted in a final volume of 8 ml resulting in a 78X yield by affinity chromatography (Table 3). Nearly all the remaining 22X of PMT applied was eluted in fractions; with PMT-concentrations below 50 ~g/ml.
The specific purity (fig; PMT/mg protein) was 284 times higher in the eluted material than in the culture supernatant (Table 3).
The average minimal dermonecrotic dose of affinity purified PMT in guinea pigs after intra.dermal injection and the average MCD of PMT
for embryonic bovine lung (EBL) cells in the standardized EBL-cell test (ref. 29) was approx. 35 ng and 30 pg, respectively. The LD50 of PMT in BALB/c mice was 20 to 40 ng (corresponding to approx. 1.5 ~sg/kg) and in Wistar rats approx. 100 ng (corresponding to 0.5 ~g/kg) when administered intraperitoneally in a single dose.
X

Purification of PMT by affinity chromatography Vol. Protein PMT Purifica- Recovery of (ml) (mg) (mg) factor PMT, X
Applied culture supernatant from P. multocida 45/78 13,200 970 3.41 1 100 Effluent from affinity column 13,200 970 <0.01 N.D.a <0.35 Eluted material B N.D. 2.67 284b 78 a) N.D. not determined b) Estimated on the assumption of purity of PMT in the eluted material.

Establishment of a P. multocida gene Zibrary in Escherichia coli Donor strain P. multocida strain 45/78. The strain produces a dermonecrotic toxin as described by Foged et al. (ref. 6).
Recipient strain Escherichia coli K-12 strain MT102. Genotype: thi, araD139, (ara-Zeu)A7697, IacAX74, galU, rpsL, hsdR. This strain was constructed .(by Mogens Trier Hansen, Novo Industri A/S, Denmark) as follows:

~3~~040 Pl transduction, donor:
So1386 Selection for tetracycline resistance;
Screening for deo=

P1 transduction, donor:

Selection for de~.m Screening for tetracycline sensitivity and i=
r~Tlo2 x Escherichia coli strain list (all strains are K-12 strains) STRAIN GENOTYPE REFERENCE OR
SOURCE
MC1000 araD139, 0(ara-Leu)7697, ~lacX74, galU, a) galK, rpsL.
S01386 Ieu, 0(deo0-deoC), TnlO-6 (91X co-trans- b) ducible with deo MT85 araD139, ~(ara-leu)7697, alacX74, galU, c) galK, rpsL, TnlO-6, ~(deo0-deoC) X804 met, supF, r-, m+, A(gal-lac) d) b) from Bente My,gind, haboratoriet for Biologisk Kemi B, University of Copenhagen, Denmark.
c) constructed b;y Mogens Trier Hansen as shown above.
a): Casadaban and Cohen, 1980, (ref. 30), d) Wood, 1966 (r~:f. 31). .
Media P. multocida was grown in Tryptic Soy Broth purchased from DIFCO. E.
coli was grown in LB medium (ref. 27).
Restriction enzymes and T4 DNA ligase were obtained from New England Biolabs and used as recommended by the manufacturer.
DNA extraction To isolate chromosomal 1DNA from P. multocida strain 45/78, cells from a 250 ml overnight stationary culture were resuspended in 10 ml' S0 mM
Tris-HC1, pH 8.0, 100 mM EDTA, and incubated with 25 mg lysozyme for 20 minutes at 37°C. 2 mL of lOX (w/v) SDS were added to the mixture which was mixed .and put on ice for 10 minutes. To the solution was then added 15 ml of phemol saturated with TE-buffer (10 mM Tris-HC1, pH 8.0, 1 mM EDT~~), after which it was heated to 65°C, mixed gently 36 ' 3 4' ~ 4 and cooled on i<:e. Aft<~r centrifugation for 30 minutes at 4000 x g the aqueous pha:>e was Extracted with ether and ethanol precipitated, and the pellet resuspended in TE-buffer. The DNA was further purified by banding in a CsCl density gradient (ref. 27). After purification, the DNA was resuspended in 1 ml TE-buffer.
Preparation of clonablea DNA fragments 9-22 kilobase pair (kb) DNA fragments With 5'GATC overhangs were pre-pared in the following way. Chromosomal DNA prepared as described above was digested partially by incubation with the restriction en-donuclease Sau3A,. At certain intervals after the incubation had been initiated, fractions of the incubation mixture were stopped with 1/20 volume 0.25 M E~~TA. A sample of each fraction was run in a 1X agarose gel in TAE buffer as described in (ref. 27), and a fraction contain-ing 4-22 kb fragments was identified. This fraction was further fractionated on a 8 ml sucrose gradient (40-10X) by layering the DNA
on top of the gradient prior to ultracentrifugation at 41000 rpm for 7.5 hours. 0.5 m.l subfractions were extracted, diluted with 1 volume TE-buffer, ethanol-precipitated and resuspended in TE-buffer. Two of these, containing 9-16 and 15-22 kb fragments respectively, were used in the following cloning steps.
Cloning procedure 9-16 kb and 15-22 kb DNA fragments with 5'GATC overhangs were ligated with BclI restricted pUN121 (refs. 27 and 19) by means of T4 DNA
ligase. Insertion of DNA into the unique BcII site of this vector leads to inactivation of the cI gene, encoding the lambda cI
repressor, which subsequently is unable to repress transcription from the plasmid-encoded J1 PL promoter into the tetracycline resistance gene. The resulting plasmids were tranformed to competent E. coli MT
102 cells as described in (ref. 27). Positive selection for clones with plasmid inserts is achieved by adding tetracycline to the medium (10 ~eg/ml). Using standard transformation techniques (ref. 27), 3332 tetracycline resistant recombinant E. coli clones were obtained, -100X of them containing inserts, thus constituting a P. multocida strain 45/78 gene library in E. coli. Colonies of the E. coli clones were grown on LB-plates containing 10 ~g/ml of tetracycline. A scrape of thesQ colonies was stored at -80°C in a 20X glycerol solution.

Identification of P. mu.Itocida toxin producing E. coli clones Screening procedure The gene library was screened using the colony blot method to trans-fer colonies to nitroceillulose (ref. 14).
PMT-producing clones were then detected by incubation of the nitrocellulose filters as follows: A) 15 min. in 50 mM Tris, pH 9.6, 150 mM NaCl, 0.0-'iX Tween~20 (washing buffer) and 2 ~g/ml DNaseI, B) 2 x 10 min. in washing buffer without DNaseI, C) 30 min. in washing buffer and 3X gelatine and heated, D) 2 x 10 min. in washing buffer with 1X Triton X-100, E) 60 min. in a 10-fold dilution in washing buffer of a. previously described hybridoma supernatant P3F51, F) 3 x 5 min. in washing buffer, G) 60 min. in washing buffer with horsera-dish peroxidase linked-anti-mouse Ig whole antibodies from Amersham (NA.931) diluted 1:1000, H) 3 x 5 min. in washing buffer, I) 1 min.
in 10 mM Na2HP04, 10 mM citric acid, pH 5.0 (C/P buffer), and J) approximately 5 min. i.n a staining solution mixed immediately prior to use consisting of 80 mg dioctylsodiumsulfosuccinate (DONS), 24 mg 3,3',5,5'-tetrameahylbenzidine, 10 ml ethanol, 30 ml C/P buffer and 20 ~1 H202. The enzyme reaction was terminated by incubation in 100 mg DONS in 12.5 ml ethanol and 37.5 ml H20.
The following clones were found to be positive by this procedure: SPE
301, 308, 311, 31.2 and 315.
Western blot The positive clones obtained in the screening procedure were further analyzed using the Western blot technique (ref. 32).
* Trade-mark X

In tire Western blot procedure, one ml overnight cultures were pelleted (6000 x g for 5 min.) and resuspended in 0.5 M Tris-HCL, pH
6.8, 3X SDS, 15X glycerol, 5X mercaptoethanol and bromophenol blue.
The samples were boiled for 5 minutes prior to loading on a gel.
Proteins were separated by sodium dodecylsulfate polyacrylamide gel electrophoresis (SDS-PAGE) (11), the separating gel consisting of 7X
(w/v) acrylamide (acrylamide/bisacrylamide ratio of 40:1) in 0.4 M
Tris, pH 6.8, O.1X SDS, 0.05X glycerol.
Subsequent transfer to nitrocellulose filters was performed in a semi-dry electroblotter as described by Kyhse-Andersen (25). Further handling of the filters was as described above in the screening procedure.
The results appear from Fig. 3 which shows a Western blot of the 5 positive clones.
The transformant strains SPE 308 (lanes 2 and 3) and SPE 312 (lanes 5 and 6) harbouring the plasmids pSPE 308 and pSPE 312, respectively, were shown to be PMT-producing, whereas SPE 301, SPE 315 and SPE 311 were shown not to be. Purified PMT (lane 8) was used as control. The recombinant toxin produced by SPE 308 and SPE 312 shows the same properties as the native protein with respect to recognition in the competitive ELIS,A as described in Example 1; the size and the toxic activity in the EBL test (ref. 29) are unchanged.

Restriction mapping and Iocalfzation of the PMT gene The plasmids pSPE 308 and pSPE 312 were analyzed for recognition sequences for restriction enzymes (restriction sites), testing for restriction sites for most of the known enzymes with a 6 base pair (bp) recognition sequence. The results are shown in Fig. 4 (pPSPE
308) and Fig. 5 (pSPE 312). It appears from the figures that the two plasmids carry an approximately 6 kb overlap (from position 11000 to 17000 in pSPE308). The pmt gene is therefore localized within this area.

By constructing end-point deletions in the overlapping areas in pSPE
308 and pSPE 312 a number of plasmids, the pSPE 308 derivatives pSPE
336, pSPE 341, p~SPE 344. and pSPE 350 as well as the pSPE 312 deriva-tives pSPE 338, pSPE 34.3, pSPE 345, pSPE 347/349 and pSPE 525 were constructed. The extenta of the resulting plasmids are shown in Figs.
6 and 7.
These plasmids were transformed to E. coli strain MT 102 and analyzed for the production of PMT by Western blotting as described above.
In the Western blot, a single protein with an apparent molecular weight of 125000 daltons was found. The blot is shown in Fits. 8. The plasmid pSPE 349 coding for this protein is deleted of a region to the right of the EcoRl site at position 8200 in pSPE 312 (cf. Fig. 5 and 7). The gene product of plasmid pSPE 349 therefore localizes the position and orientation of the pmt gene (shown as shaded area in Fig. 5). The coding region begins about 3.3 kb upstream of the EcoRl site at position 8200 in pSPE 312 i.e. around the CIaI site at posi-tion 4900. Since the total coding region is estimated to be about 3.9 kb the structural gene ends about position 8800 on the map shown in Fig. 5.

Sequencing of the pmt gene The nucleotide sequence carrying the pmt gene as localized in Example 4 was determined using the method described by Sanger et al. (ref.
33). The sequence of 4381 consecutive by were determined. The DNA
sequence of the region is shown in Fig. 10 (a)-(j) in which the deduced amino acid sequence is indicated above the DNA sequence starting at position 219. The methionine codon of position 219-221 is preferred as a starting codon to the methionine codon of position 213-218 due to its perfect spacing to the putative ribosome binding site at position 201-210. The region containing the pmt gene was subjected to more detailed restriction mapping by a computer search x ~o for all restriction sites for restriction enzymes with 6 by recogni-tion sequences.
The results are shown in Fig. 9 which shows a high degree of confor-mity with the pr~eviousl:y constructed restriction map.

Expression of P. multocida toxin in E. coli The amount of PM'T produced in different recombinant E. coli clones was determined b:y estimating the incorporation of S-35 labelled methionine using the following procedure. Cells were grown in AB
minimal medium (:ref. 21) supplemented with 0.2X glucose, 1 ~g/ml thiamine and 50 psg/ml l~eucine at 37°C. At an optical density (OD450) of 0.6, 1 ~1 of ~3 1 mCi,/ml S-35 methionine solution (purchased from Amersham, SJ. 10:L5) was added to a 1 ml sample of the culture at 37°C. 3 minutes :later, :50 ~1 of 10 mg/ml unlabelled methionine was added as a chase and after further 3 minutes, the sample was put on ice. Samples were pelleted (6000xg for 5 minutes) and resuspended in loading buffer (1).SM Tr:is-HC1 pH 6.8, 3X SDS, 15X glycerol, 5X
mercaptoethanol, bromoplnenol blue). Subsequently, proteins were separated in SDS-PAGE a;s describad in Example 5. Furthermore, the gei was dried and subjected to autoradiography overnight on a KODAK XAR-5~
X-ray film. Relative amounts of PMT were estimated by scanning of the X-ray film, using the ~ and ~' subunits of the E. coli RNA-polymerase as a reference.
Recombinant clones tested in this way were MT102 strains harbouring the plasmids pSPI: 312, pSPE 525 (shown in Fig. 11) and pSPE 481 (shown in Fig. 1;?), respectively. Plasmid pSPE 481 consists of the 7 kb Pstl fragment of pSPE 525 ligated to a 1.3 kb PstI fragment of plasmid pPL 195, containing part of the ampicillin resistance gene and the lambda Pl:. promoter and operator region. pPL 195 was con-structed by ins erting a pUC8 EcoRI-HindIII polylinker (ref. 33) into a EcoRI and HindaII restricted PLc28 vector (ref. 34). The resulting pSPE 481 plasmid carries the lambda PL promoter transcribing into the f pmt gene.
g * Trade-mark Table 4 Estimated amount of PMT' in the recombinant E. coli clones molecules/cell <500 3000-4000 12000-15000 Data used in the above calculations PMT
Methionine content (X) 2.76 2.56 2.80 (23) (24) Size (kD) 150 155 146.5 (23) (24) Molecules/cell 4500 4500 (at 2.5-3 doublings/hour) (25) (25) Furthermore, the content of PMT in SPE481 was estimated using the following procedure. A 100 ml culture with an optical density (OD450) of 5 was pelleted (6000 x g for 5 minutes) and resuspended in 10 ml of 50 mM Tris-HC1 pH 7Ø The supernatant, which did not contain any PMT, was discarded and the harvested cells were disrupted by sonica-tion. PMT was then further purified on an anion-exchange column as described in ref. 6. The purified PMT was then subjected to a quantitative ELISA as described in Example 2, obtaining an estimated value of 2-5 ~g PMT per ml of culture fluid.
The Escherischia coli K-12 strain MT102 harbouring the pSPE 481 plasmid has been deposited 21 March, 1988 according to the Budapest Treaty in Deutsche Sammlung von Mikroorganismen and Zellkulturen GmbH
under the strain designation Escherichia coli K-12 SPE 481. The accession number is DSM 4488.
'X

~2 ~34~~4~

Production and characterization of toxin derivatives The following de:rivativ~es of toxin-encoding plasmids were constructed with the purpose of producing truncated, i.e. detoxified, toxins which are potentially uaeful for immunogenic purposes. The construc-tions were prepared on the basis of the restriction mappings disclosed in Examples 6 and 7. The hypothetical toxin-derived proteins produced from ~plasmids pSPE A through pSPE Q, proteins A
through R, are shown in Fig. 13. All derivatives were optimally expressed from the respective plasmids in strain SG 21059 kindly provided by Susan Gottesmann. The known genotype of this strain is ~ga1 Ion146:: ATnlO ~Iac.
1) pSPE A. The p7Lasmid was constructed by restricting pSPE 481 with the restriction esnzyme :ituI prior to ligation. This deletion causes a change of reading; frame, However, as described below, a PMT der-ivative could be detected in Western blotting as well as in the EBL
toxicity test. This cou~Ld be due to a small amount of erroneous frame-shifting in the translation procedure. See Fig. 13.
2) pSPE B. The p7.asmid was constructed by restricting pSPE 481 with the restriction enzyme XbaI prior to ligation. The plasmid codes for a hypothetical PMT derivative of about 108 kD, lacking amino acids 169 through 468 of PMT. See Fig. 13.
3) pSPE C. The pl.asmid was constructed by restricting pSPE 481 with the restriction enzyme NdeI prior to ligation. This deletion causes a change of reading frame,. However, as described below, a PMT der-ivative could be detectE:d in small amounts. This could be due to erroneous frame-:ohiftinf; in the translation procedure. See Fig. 13.
4) pSPE D. The pl.asmid was constructed by restricting pSPE 481 with the restriction enzymes NdeI and SnaBI and subsequently blunt-ending the resulting ends, using T4 polymerise (purchased from New England Biolabs) as described by the manufacturer, prior to ligation. This deletion causes a change: of reading frame. However, as described .... .w below, a PMT derivative: could be detected in small amounts. This could be due to erroneous frame-shifting in the translation pro-cedure. See Fig. 13.
5) pSPE E. The plasmid was constructed by restricting pSPE 481 with the restriction enzyme HindIII prior to ligation. The plasmid codes for a hypothetical PMT derivative of about 53 kD, since this deletion causes a change of reaf.ing frame. See Fig. 13.
6) pSPE F. The plasmid was constructed by restricting pSPE 312 with the restriction enzyme HindIII prior to ligation. Like pSPE E, this plasmid codes for a hypothetical PMT derivative of about 53 kD. See Fig. 13.
7) pSPE G. The plasmid was constructed by restricting pSPE 481 with the restriction enzymes SnaBI and XhoI and subsequently blunt-ending the resulting ends as above, prior to ligation. The plasmid codes for a hypothetical PMT derivative of about 135 kD, lacking amino acids 507 through 568 of PMT. See Fig. 13.
8) pSPE H. The plasmid was constructed by restricting pSPE 481 with the restriction enzymes. SnaBI and SpeI and subsequently blunt-ending the resulting ends as above, prior to ligation. The plasmid codes for a hypothetical fMT derivative of about 88 kD, lacking amino acids 569 through 1058 of PMT. Se:e Fig. 13.
9) pSPE I. The p~lasmid was constructed by restricting pSPE 481 with the restriction enzyme Nsil prior to ligation. The plasmid codes for a hypothetical fMT derivative of about 79 kD, lacking amino acids 634 through 1204 of PMT. Se:e Fig. 13.
10) pSPE J. The plasmid was constructed by restricting pSPE 312 with the restriction enzyme NsiI prior to ligation. The plasmid codes for a hypothetical fMT der3.vative of about 70 kD. See Fig. 13.
il) pSPE K. The plasmid was constructed by restricting pSPE 481 with the restriction enzyme SpeI and blunt-ending the resulting ends as x above prior to ligation. The plasmid codes for a hypothetical PMT
derivative of about 117 kD. See Fig. 13.
12) pSPE L. The ;plasmid was constructed by restricting pSPE 312 with the restriction .enzyme EcoRI prior to ligation. The plasmid codes for a hypothetical PI~iT derivative of about 124 kD. See Fig. 13.
13) pSPE 0. The plasmid was constructed by partially restricting non-methylated pSPE ~+81 with the restriction enzyme BclI prior to ligation. The plasmid codes for a hypothetical PMT derivative of 133 kD, lacking amino acids 30 through 150 of PMT. See Fig. 13.
14) pSPE P. The plasmid was constructed by restricting pSPE 481 with the restriction Enzyme ;ipeI, subsequent treatment with the exonuclea-se Ba131, restricaion wiCth EcoRI and finally treatment with Klenow fragment of DNA polymerise I in the presence of all four deoxyribo-nucleotides prior to li~;ation. The plasmid codes for a hypothetical PMT derivative of about 136 kD, lacking amino acids 1043 through 1130 of PMT. See Fig. 13.
15) pSPE Q. The p~lasmid was constructed from a derivative of pSPE
481, pSPE 680. pSPE 680 was constructed by restricting pSPE 481 with the res~riction enzymes namHl and Glai, and treatment with Klenow fragment of DNA polymerise I in the presence of all four deoxynucleo-tides prior to ligation. Subsequently, pSPE Q was constructed by restricting pSPE 680 with EcoRV prior to ligation. The plasmid codes for a hypothetical PMT dlerivative of 127 kD, lacking amino acids 175 through 246 of PM.T.
16) pSPE R. The plasmid was constructed by restricting pSPE 481 with the restriction enzymes SpeI and EcoRI and blunt-ending the resulting ends, as described above, prior to ligation. The resulting plasmid codes for a hypothetical PMT derivative of about 117 kD. See Fig.
13.
The reactivity of selected derivatives with a panel of anti-PMT MAbs has been studied by sandwich ELISA's based on detection with non-competitive combination pairs of MAbs. By this method an x ~s antigen-code (A~;-cods), i.e.. a term indicating the epitope-difference of the derivative compared to PMT, was determined for each der-ivative.
The minimal cytopathic dose (MCD) on 120 ~1 of 1.5 x 105 embryonic bovine lung cells has been estimated for some affinity-purified PMT
derivatives.
The results of t:he two studies are indicated in the following table (Table 5):

Derivative L 0 P B* C* G*

Ag-code PMT-a: PMT-~9PMT-~ PMT-~ PMT-y PMT

MCD (ng) 250 30000 5 n.d. n.d. n.d.

* not affinity-purified. (sonicates) n.d.: not done The antigen-code "PMT" indicates that the derivative reacts in-differently compared to PMT by the sandwich-ELISA, i.e. all epitopes characterized by the panel of MAbs, which are present on PMT can be detected on the derivative. The Ag-code "PMT-x", where x is a, ~ or 7 indicates that the MAb(s) reacting with epitope x on PMT are not reacting with the derivative. The results of these tests shows that epitope a is missing on. derivative L, ~ is missing on 0 and P and 7 is missing on B and C.
The MCD of PMT is approx. 0.01-0.03 ng and it is therefore obvious that the abovementioned derivatives L and 0 are practically non-cytopathic compared to PMT, whereas P shows slightly remaining cytopathic activity.
A Western blot was made using a mouse anti-PMT antiserum produced as described in Example 1, and using horseradish perioxidase rabbit f anti-mouse Ig whole antibody (purchased from Amersham) as the secondary antibody. Strains harbouring the plasmids pSPE A, B, C, D, E, G, H, I and L were shown to react with the anti-PMT antiserum (cf.
Fig. 14).
Trials carried out on mice for elucidation of the immunogenic effect of the O-derivative Purpose:
To study the immunogenic: effect of PMT after deletion in the N-terminal end (0-derivati.ve).
Method:
For the trials, P~ALB/c. mice were used. Sexually mature female mice were immunized subcutaneously 2 times at 14 days interval with 0.3 ml 0-derivative (2.5 - S ~g,/ml) in 20X Alhydrogel (cf. Example 12) in PBS + O.1X zero-u~ouse-serum.
Simultaneously with the 1st vaccination the female mice were mated.
Thus, the 2nd vaccination took place about 1 week before expected birth.
1) About 10 days old t:he baby mice were divided into 2 groups. Half of them were dripped intranasally with PMT (totally 60 ng PMT), the other half was injected intraperitoneally (i.p.) with PMT in different concentrations. The number of dead animals was recorded.
2) All the adult female mice were bled after killing of the surviving baby mice. The blood samples were analysed by ELISA as described ir: Exampl.e 13 for the presence of antibodies against PMT. Immediately after blood sampling the mice were injected i.p. with different: concentrations of PMT. The number of dead animals was recorded.

Schematic trial plan Trial week No.
- 3 - 31 female BALB/c mice were mated and subsequently immunized with 0-derivative 1 - 31 female BAL1B/c mice were immunized for the second time with 0-derivative 0 - 31 fem;3le mice delivered 122 baby mice 1 - the baby mice were divided into group I (61) which was treated intranasally with PMT. Group II (61) which was treated i.p. with PMT
2 - survivCng baby mice were killed female BALB/c~mice were bled female BALB/c mice were injected i.p. with PMT
3 - the number of surviving female BALB/c mice was recorded and the animals killed Trial i:inished ~s 134'040 Results are shown in tables 6 and 7.
LD75 in non-protected baby mice was about 20 ng PMT injected i.p.
X50 in non-protected adult mice was about 70 ng PMT injected i.p.
Conclusion:
It can be concluded that mice born from mothers vaccinated with 0-derivative vaccine in the doses described can survive i.p.
injection of min. 25 x LD50 of PMT. The protection is obtained via antibodies transferred via colostrum from mother to offspring.
Furthermore it c,an be concluded that 0-derivative vaccinated animals develop antibodies against PMT, even if some variation is seen. The mice can survive i.p. injection of min. SO x LD50 of PMT.
Thus, mice vaccinated with 0-derivative can transfer a considerable protection against PMT to the offspring via colostrum. The mice themselves devel~~p antilbodies against PMT and are protected against even high concentrations of PMT.
-x 1 34~ 04 0 o ~ w o° "~~ '' 0~0 N o n -r- - e~

o ~r x N O

en O

C ~ - b O

i O

v r-1C N r~

O

U O
oD

V

- - 'Oa O O~ O

v H H ~ 0 O

w ~ n y ~ ~

O tai ~ W

~i O -ri let r~

N . ~
yJ

O O~ t~~

~ O

~ *

O O v r~

a ~ ~ ~~ v ~' a~

.. O

Z U O U

~. ~ O .i U1 lit v N

~ ro ro v ~d s %d ~

o~ t~ ~a w ~ ~ w o o o y ~~o No a..~ 0 0 0 .. .. .. o C ~C n1 e-1 ~ N
O .1 rl r, .. .. .. v v z b z .a rl N v1 rl N 1"1 x ExAMP3.E 10 Differentiation ~of PMT+ and PHT- strains by PMT-ELISA
615 field isolates and 7 reference strains of P. multocida were exam-fined. The field isolates were obtained from nasal swabs (603 iso-lates) and lungs (12 :isolates) of pigs from 156 Danish herds and were identified by the following criteria: acid produced from glucose, saccharose, mann:itol, s~orbitol and not from maltose, arabinose, dul-citol and inositol; and production of indole, ornithine decarboxyl-ase, catalase, o:~cidase and not of urease.
Extracts far toxin analyses were prepared by harvesting blood agar (9 cm Petri dish) overnight (37°C) cultures into 2 ml of sterile water by means oi: a spar_ula. The suspensions were left for extraction at 37°C for approximate:Ly 18 hours. One part of the extract was examined directly by PM'.f-ELISA as described in Example 2. All absor-bances (A) were E:xpresse d as percentages of the absorbance obtained by a positive control (E~). This control was a 1:1 dilution of an extract, freshly preparESd for each test of the toxigenic type D
reference strain P. mult:ocida ssp. multocida 45/78.
Another part was centrifuged (30 min. at 1500 x g), the supernatant sterile filtered and subsequently examined in the EBL-cell test as described earlier- refs. 22 and 29).
The 615 field isolates were characterized as toxigenic (250) or non-toxigenic (365) by the F:BL-cell test and were of capsular type A (119 toxigenic and 92 non-toxigenic isolates) or D (131 toxigenic and 273 non-toxigenic isolates).
Full agreement beaween the EBL-cell test and the PMT-ELISA was ob-tained for the 61.5 field isolates and the 7 reference strains (Table 8).

sl 131040 EBL-cell-test PMT-ELISA
a) b) 250 field isolates of P. multocida ssp. multo~cida+ +

365 field isolates of P. multocida ssp. multo~cida- -Type strain (CCUG 17977) P. multocida ssp. septica- -Type strain (NCTC. 10204) P. multocida ssp. galicida- -Type strain (NCTC 10322) P. multocida ssp. multocida,type A - -Reference strain (ATCC
12945) P. multocida ssp. multocida,type A - -Reference strain (NCTC ' 12177) P. multocida ssp. multocida,type A + +

Reference strain (ATCC
7707) P. multocida ssp. multocida,type D - -Reference strain (NCTC
12178) P. multocida ssp. multocida,type D + +

a) All EBL-positive (+) bacterial extracts had EBL-titres above 103 (median 104, range 103-106), in the EBL-cell test, EBL-negative (-) extracts were non-cytopathic.
b) All 1:1 diluted ELISA-positive (+) bacterial extracts had relative absorbances above 39X (mean t SD: 94X ~ 13X) in the PMT-ELISA, where-as all ELISA-negative (-) extracts had relative absorbances below 9X
(2.iX ~ 1.9X).
The cytopathic and non-cytopathic extracts of the 615 field isolates were separated in two clearly distinguishable groups by the PMT-ELISA
(Fig. 15). Since the mean t SD of the absorbances (A) obtained from the 1:1 diluted extracts of the 250 toxigenic isolates was 1.72 t s2 1341040 0.48,-visual readings 3.nstead of photometric measurements of the ELI-SA-results would'. be satisfactory for the differentiation of extracts of P. multocida. The mean t SD of the PMT-concentration in the ex-tracts of the toxigenic. isolates of P. multocida was estimated being 2.8 t 1.9 ~g/ml, and since the detection limit of the PMT-ELISA is approx. 50 pg (1 ng/ml) PMT (cf. Example 2), dilutions of the ex-tracts (Fig. 16) and extracts with low PMT-concentrations may appro-priately be tested by PMT-ELISA. The main advantages of the PMT-ELISA
compared to existing tests are the independence of cell culture or laboratory animal facilities, the ability of a single laboratory worker to handle several hundreds of samples per day and the possi-bility of obtaining quantitative objective results from bacterial extracts in 4 hours.

Neutralization o.f PMT with monoclonal anti-PMT-antibodies Samples (30 ~1) ~~f either PMT in PBS or PMT in a crude cell-free ex-tract of P. multocida 4.5/78 (ref. 6) containing PMT in amounts up to 12 ng and 1 pg o:E purified MAb (P3F51) were incubated for 15 min. at 20°C before application to embryonic bovine lung (EBL) cells (120 ~1, 1.5 x 105 cells/ml) as described for the n_rig,'_nal EBL-cell test (ref.
29). The minimal cytopathic dose (MCD) of PMT was estimated when no MAb was present :in the ;sample. The neutralization titer was recorded as the number of MCD which could be neutralized by 1 ~g of MAb.
The results appear from Table 9 below.
~.

s3 1341040 Table 9 Hybridoma Represe:nta- Neutralization in group No. tive MAb EBL-cell test (x MCD)a 2 P3F64. 70 3 P3F3T <2 9 P3F50~ 400 ________________________________________________________________ a) Neutralization of the cytopathic effect on PMT was estimated as the number of minimal cytopathic doses (MCD) neutralized by 1 ~g of MAb. The MCD of PMT is about 30 pg.
As indicated in Table 9, addition of 1 ~g of MAb to PMT 15 min. be-fore addition to the EBL-ells resulted in a 30 to 400 times increase of the MCD for 9 out of the 10 representative MAbs, whereas MAb P3F37 had a very low neutralizing effect on PMT. The neutralization of the cytopathic effect of PM:T was also achieved when a crude cell-free extract of P. multocida 45/78 was used instead of pure PMT.
Samples (200 ~1) containing PMT in variable amounts up to 2.56 ~g and purified MAb (P3F51) in: amounts between 0.15 and 15 ~g were incubated for 15 min. at 20°C and. injected intraperitoneally (i.p.) in female BALB/c mice (6 weeks old, 15 to 20 g). Mice dying within a week from the time of PMT injection were recorded and the lethal dose of PMT
and the neutralizing effect of the MAb was estimated. When 1.5 or 15 ~g of P3F51 were added the lethal dose of PMT increased about 4 and 32 times, respectively, whereas 0.15 ~g of the MAb had no neutraliz-ing effect.
~x 54 ' 349 04 0 To study the in vivo neutralization ability of anti-PMT monoclonal antibodies a 200 ~1 solution containing 15 ~cg of purified monoclonal antibody (P3F51) was injjected (i.p.) in female BALB/c mice (6 weeks old, 15-20 g) 2 clays before i.p. administration of a 200 ~1 solution containing PMT in varying amounts up to 2.56 fag either in a pure form or as a crude cell-free extract of P. multocida 45/78 (ref. 6). The neutralizing effect was estimated as described above.
The lethal dose of PMT increased about 32 times when mice were pas-sively immunized with 15~ ~g of P3F51 2 days before challenge with PMT
or a crude cell-free extract of P. multocida 45/78.

Vaccination with purified PMT or derivative O
mg of PMT purified as described in Example 3 in 45 ml of PBS was dialyzed against 0.35% formaldehyde in PBS, pH 7.3-7.9, for 36 hours 15 at 4°C after which 1 g/1 lysine-HC1 was added to the dialysis li-quid, and after 18 hours the dialysis was continued with repeated changes of PBS. The thus produced detoxified PMT was analyzed for lack of (or sufficiently reduced) toxic activity in the mouse letha-lity test and the cytopathic test on EBL-cells descri.hed ahove as well as a dermonecrotic test in guinea pies as described by Foged et al. (1).
10 mg of biologically inactive (detoxified) PMT in 40 ml PBS was then coupled to 10 ml aluminium hydroxide gel purchased from Superfos, Denmark, under the trade name Alhydrogel as recommended by the manufacturer and diluted in 20X aluminium hydroxide in PBS to a final concentration of about 5 ~g/ml or 125 pg/ml detoxified PMT.
Gestating gilts were immunized subcutaneously 4-6 weeks and 2-3 weeks before farrowing 'with a dosis of 3 ml of the detoxified PMT vaccine composition prepared above. After farrowing piglets were inoculated intranasally with Bordetella bronchiseptica and P. multocida and the protective effect of the immunization of the sows was estimated by measuring the average daily weight gain before slaughtering of the ss 1341040 pigs (at about 90 kg live weight) and determination of osteopatholo-gical conditions in the snout of the pigs at slaughtering. Pigs from immunized sows were compared to pigs from non-immunized sows and the protective effect of the immunization is shown in Table 10.

No. of animals Mean daily No. of animals (litters) weight gain with severe tur-after weaning binate atrophy (X) Pigs from non-immunized iil (8) 781 g 49 (80.3X) sows Pigs from immunized 1'l4 (20) 848 g 20 (11.5X) sows In a study still in progress gilts were immunized with 50 ~cg/dose of affinity-purified derivative 0 from sonicates of an E.coli clone containing pSPE 0 as described is Example 9. t:o modifications of v were performed e~tcept for coupling to Alhydroger': Preliminary results of the vaccination study indicates that:
a) the serum- and colostrum titres against native PMT are similar for gilts vaccinated with derivate 0 and formaldehyde treated PMT, b) the specific: antibodies are transferred to piglets through colostrum edually well in both vaccine groups.
c) the clinica7L symptoms of atrophic rhinitis are prevented equally well in the progeny from gilts vaccinated with 0 (0-piglets) and formaldehydes treated PMT (P-piglets), and that this prevention seems to be close to 100X, when compared to piglets born by ,.
unvaccinated gifts (control piglets).
* Trade-mark s6 1341040 d) the toxigenic P. multocida used for the experimental infection can be reis~~lated in significantly higher rates from control piglets thaw from ~0- or P-piglets at 5 weeks of age.

Detection of an ti-PMT-antibodies By proceeding substantially as described in Example 2, but by incu-bating the coating monoclonal anti-PMT-antibody with a premixed pre-paration of serum and a constant amount of PMT, it is possible to detect anti-PMT-~intibod:ies in serum of for instance pigs infected with P. muXtocida or of animals vaccinated with a vaccine of the in-vention. The mix~:ure wh:LCh was prepared for concentrated or diluted serum samples was incubated for 30 min. at 37°C before incubation for min. in the microtiter plate. The presence of anti-PMT-antibodies in the serum sample was detected by a decrease in absorbance measured 15 substantially as described in Example 1 (the section entitled "ELISA
for estimating epitope specificity"). The results are shown in Fig.
17 which shows the 50X b locking titers of serum from an anti-PMT-antibody negativE~ pig (~:2), a pig infected with a toxin-producing P.
multocida strain (about 14) and a gilt vaccinated with the vaccine described in Exaiuple 12 (about 250).

Detection of PMT by colony blot and i~oblotting The presence of fMT in samples may be detected by a colony blot method (ref. 14) as described in Example 5 (the section entitled "Screening procedure").
Similarly, the presence of PMT in samples may be detected by separat-ing proteins in the samples electrophoretically by SDS-PAGE (as de-scribed in Examp7~e 1) and transferring them electrophoretically to a nitrocellulose mE:mbrane where PMT, if present, can be visualized by immunoblotting as descr:Lbed in Example 1 (in the section entitled "Immunoblotting"). The electrophoretic location of the stained protein band also gives. the apparent molecular weights of PMT
(approximately 143 kd).

Genetic distinction between PMT+ and PMT- isolates of P. multocida as determined by' colony hybridization P. multocida isolates (17 toxin-positive and 18 toxin-negative strains as determined by ELISA and EBL tests as described above) were inoculated on Tryptic Soy Broth Agar plates (purchased from DIFCO).
After incubation overnight at 37°C, a replica was made on a nitrocel-lulose membrane filter (Schleicher & Schull BA 85). This replica was placed (face up) on top of 4 consecutive Whatman 3M~filters which were soaked in 10% SDS, denaturation buffer (0.5M NaOH, 1.5M NaCl), neutralization buffer (0.5M Tris-HC1 pH 8.0, 1.5M NaCl) and 2xSSPE
(360 mM NaCl, 20 mM NaH2P04, 2 mM EDTA pH 7.4), respectively.
Incubation was carried out for 5 minutes on each filter at room temperature. Subsequently, the nitrocellulose filter was dried and DNA was fixed to the filter by baking at 80°C for 2 hours. Prehy-bridization and hybridizations were done in 6 x SSC (0.15M NaCl, 0.015M sodium citrate, pH 7), 0.5X SDS and 5 x Denhardt solution for 2 hours and overnight, respectively, at 65°C. The probe was a radio-actively labelled XbaI fragment from position 1623 to 4376 in the se-quence shown in Fig. 10 (a)-(j) prepared by the nick translation method (ref. 22). After hybridization the filter was washed at 25°C
in 2 x SSC, 0.5X SDS for 2 x 15 min. and in 0.2 x SSC, 0.5X SDS for 2 x 1 hour at 65°C and left for sutoradiography overnight.
The results appear from Fig. 18 which shows that colonies at posi-tions 5, 6, 8, 9, 10, 12, 13, 14, 15, 17, 19, 22, 34, 36, 37, 39, 45 and 50 were PMT+ and colonies at positions 7, 23, 24, 26, 28, 30, 32, 42, 44, 47, 48, 53, 56, 58, 63, 66, 73 and 75 were PMT-. These results are in accordance with the ELISA and EBL determinations.
Hence, non-toxigenic strains of P. multocida owe their lack of toxin production to a lack of the PMT encoding pmt gene.
* Trade-mark Purification of rPMT and comparison of rPHT with PMT
In the toxin purification procedure, cells harvested from a 1 1 overnight stationary culture of SPE312 were resuspended in 10 ml of H20 and sonicated several times for 0.5 min, at 0°C using a Branson sonifier 250 (Br,anson, Conn., U.S.A.). The sonicate was diluted to SO
ml in 0.1 M Tris-HC1, p:H 7.8 containing 0.5 M NaCl before appliation to the affinity ~~olumn which was prepared by immobilizing the anti-PMT Mab P3F51 as descrilbed in Example 3. After repeated washings of the affinity column, rPIKT was eluted with 0.1 M glycin-HC1, pH 2.8 as earlier described for the affinity purification of PMT from extracts of toxigeni.c P. tQUltocida. All fractions were immediately neutralized with 1 M K2HP04.
A diluted bacterial sonicate of SPE312 containing approximately 82 ~cg of rPMT as determined by the quantitative ELISA described in Example 2 was applied to a 1-ml affinity column to which was coupled approximately 5 mg of anti-PMT MAb P3F51. No rPMT could be detected in the effluent from they cohimn. Upon elution approximately 75 ~g of rPMT was obtainecL in the: two main fractions of 1.4 ml each. This corresponds to a recover~y ef 91% ef the applied rPMT.
PMT assays Quantification of rPMT was done as described for PMT (Example 2) using the capture anti-fMT MAb P3F51 and the biotinylated detector MAb P3F37 in the PMT-ELLSA, a sandwich ELISA based on the same technique as explained below for the study of epitopes on rPMT and PMT. Quantification by the PMT-ELISA was compared to results obtained in a modified Coomassie brilliant blue dye-binding microassay previously used for the determinations of protein concentrations and dye-binding ability of fMT compared to bovine serum albumine (BSA).
Comparison of epitopes on rPMT and PMT was done by sandwich ELISA's based on 10 anti-PMT MAbs (Example 1) purified from hybridoma supernatants on protein A-agarose columns. These MAbs have been shown to react with different epitopes on PMT. The sandwich-ELISA's were done as described in Example 2. Dual determinations were performed for both antigens in all 100 combinations of the 10 catching MAbs and the same biotinylated detecting MAbs. Combination pairs of MAbs resulting in abs~orbances below 0.3 were considered competitive. For the non-competitive conibination pairs, the results were described as the mean of dual determinations of absorbance obtained for rPMT
relative to the mean of dual determinations of absorbance for PMT.
The dermonecrotic and lethal effects of rPMT and PMT were determined by injecting 200 ~cl of dilutions of the previously ELISA-quantified samples intradermally into guinea pigs or intraperitoneally into BALB/c mice, respectively. Samples resulting in a dermal lesion of 10 mm or more at 48 h a:Eter intradermal injection were scored as dermonecrotic and samples resulting in death in less than 5 days after intraperitoneal injection were scored as lethal. All results were based on at least duplicate determinations.
Affinity-purified rPMT :and PMT had very similar patterns of reactions in the structuraJ~ ELISA test based on 100 combination pairs of 10 different anti-PZiT MAbs and 100 ng/ml of affinity purified antigen.
For PMT 25 pairs resulte d in an absorbance value (A4g2) below 0.3 which was considE:red to indicate competitiveness. The same 25 pairs showed competitive reactions when the antigen was 100 ng/ml rPMT. The remaining 75 non-competitive combination pairs resulted in A4g2 values above 0.3 both when PMT and rPMT was used. The overall mean ~
SD for the 75 ca7.culate<i values of the relative absorbances of rPMT
compared to PMT was 1121: ~ 8X. Only minor differences from the overall mean were: observed for the mean values for the 10 catching MAbs and the 10 t~iotinyRated detector MAbs.
PMT and rPMT reacted very similarly when tested for cytopathic effect on EBL-cells, for dermonecrotic activity in guinea pigs, and for lethality in mice:, and their ability to bind Coomassie brilliant blue were equal and aF>proximately 2.5 times weaker than the dye-binding ability of BSA (7.'able 1~~) .
X

Cytopathic, dermonecrat:ic, lethal and dye-binding effects of PMT and rPMT
Sample cuinimal minimal minimal dye-binding c:ytopathic dermonecrotic lethal (X)a dose (pg) dose (ng) dose (ng) rPMT 20-40 35 30 35-45 a The concentration oi: BSA relative to the concentration of sample resulting in equal colour formation in the Coomassie brilliant blue dye-binding mic:roassay.

Examination of E.coli arid P. multocida sonicates for cytopathic activity.
Sonicates of E.coli SPE312 and P, multocida 45/78, prepared as described in Example 1.6 were tested fo~_- cytopathic effect in the embryonic bovine lung (E:BL) cell test (ref. 29). A row of 5-fold dilutions was prepared for each soni.cate and 30 pl of each sample was applied to 1.8 x 104 EBL-cells in 120 pl of culture medium and the mixture incubated for 3 days at 37°C before fixation and staining.
Samples which resulted i:n monolayers of EBL-cells morphologically discernible from the epiLthelial-like swirling patterns of negative control culture, were scored as cytopathic. The cytopathic effects for affinity purified rPMT and PMT in the EBL-cell test were determined in the: same way. The minimal cytopathic dose (MCD) for the samples was calculated as minimal amount of rPMT or PMT, determined by the quantitative PMT-~ELISA causing a cytopathic effect.
Neutralization of.' the cytopathic effect of E.coli SPE 313 sonicate by anti-PMT MAbs wa:o compared to neutralization of pure PMT: Samples (30 pl) containing approximately 1 ~cg of MAb and varying amounts of X

~ 34~ 04 0 ~1 sonicate or PMT were incubated for 15 min. at 20°C before application to EBL-cells. The resulta were recorded as the number of MCDs neutralized by each MAb, and as the ratio between the number of neutralized MCDs of the sonicate and pure PMT for each MAb.
Sonicates of SPE308 and SPE312 were shown to cause morphological changes of embryonic bovine lung (EBL) cells, identical to those caused by toxigen.ic strains of P. multocida (Fig. 19 (data for SPE308 not shown)). As observedl for pure PMT, the cytopathic effect of the sonicate of E.coli SPE312 could be neutralized by incubation with anti-PMT MAbs. Between 5~ and 125 times MCD of the sonicate could be neutralized by various anti-PMT MAbs, whereas between 3 and 125 times MCD of the pure fMT were: neutralized. The overall mean ~ SD for the 10 calculated values of the relative number of neutralized MCDs of E.coli SPE312 sor~icate compared to PMT was 95X t 32X. A PMT-unrelated MAb used as a control dial not neutralize the effects of the two cytopathic preparations.

Analysis of the mature of the DNA flanking the pmt gene In an attempt to investigate the nature of the DNA flanking the pmt gene in P. multoc:ida 45/78, chromosome walking was performed as described in ref. 37. By using a colony hybridization procedure plasmids carrying P. multocida DNA were isolated from the P.
multocida gene library described in Example 4.
Probes The plasmid pLOL03 was constructed by subcloning a 0.8 kb Accl-HindIII DNA fragment of pSPE344 (Fig. 20) in the vector pGEM-blue (Promega, Wi, US~~). The plasmid pLOR02 was likewise constructed by subcloning the 2.4 kb E~.oRI-BglII fragment of pSPE312 (Fig. 20) in the vector pGEM-blue. The E.coli K12 strain DHSalpha (BRL, Md, USA) was used as host strain for pLOL03 and pLOR02. pLOL03 and pLOR02 in linearized forms were used for generating RNA probes of the P.
multocida DNA carried b:y these plasmids. The RNA probes were radioac-X

w_ ~34~0r 0 tivel_y labelled using the Riboprobe System I~procedure (Promega, Wi, USA), and used in the colony hybridizations and Southern blots described below.
Colony hybridization The P. multocida gene library was spread in appropriate dilution on several LB-plates containing 10 ~g/ml tetracycline, and incubated overnight at 37°C. Replicas of the plates were made on nitrocellulose membrane filters, and the cells were lysed and the DNA fixed to the filters as described in example 15.
Prehybridization and hybridization was performed at 65°C in 50%
formamide, 6 x SSC (0.15 NaCl, 0.015 M tri-sodium citrate, pH 7.0), O.1X SDS, 5 x Denhardt's solution and 200 ~g/ml denatured Salmon sperm DNA for at least 2 hours and overnight, respectively. After hybridization, the filters were washed twice at room-temperature in 1 x SSC, O.1X SDS, and twice at 65°C in 0.1 x SSC, O.1X SDS. After washing, the filters were left overnight for autoradiography.
This procedure resulted in the isolation of a number of clones carrying P. multocida DNA flanking the inserts in pSPE308 or pSPE312.
These clones were further analyzed using the Southern blot technique (ref. 17). The Southern blots showed that the following plasmids were recognized by the RNA probe coded for by pLOL03: pL0A01, pL0A02 and pL0A03. Similarly the plasmids pLDB0l, pLOB02 and pLOB03 were recognized by the RNA probe coded for by pLOR02.
pL0A03 (approx. 14.2 kb) and pLOB03 (approx. 12.7 kb) carried the largest inserts. Their restriction maps and a Southern blot analysis show that pL0A03 and pSPE308 contain overlapping DNA of approximately 4.0 kb and that pLOB03 and pSPE312 contain overlapping DNA of approximately 1.7 kb as shown in Fig. 20.
A Southern blot was made using DNA extracted as described in Example 4 (a KI gradient (0.875 g/m1) was used instead of the CsCl2 gradient) from the toxigenic P. a~ultocida 45/78 and from a non-toxigenic P.
multocida strain MH81P8~, type D (ref. 36) and the plasmids pL0A03, * Trade-mark ~3 ~ 3 4 1 4 4 0 pL0A0-2, pSPE308, pSPE312 and pLOB03 digested by restriction enzymes as indicated in Fig. 21. The probe was the 2.4 kb BglII-EcoRI
fragment of pLOB03 radioactively labelled by nicktranslation (Rigby et al., 19'l7, (ref. 19)). The result shows that:
1) The probe recognizes a DNA sequence on each of the plasmids pL0A03 and pLOB03. Thus, there is a homologous sequence on each side of the ,pmt gene. The distance between these homologous sequences is approximately 25 kb.
2) The probe recognizes distinct fragments of the chromosomal DNA of both P. multocida strains used in this Southern blot.
The above findings could indicate that the DNA flanking the pmt gene and thus the pmt gene itself has originally been carried by a bac-teriophage, a transposon, a plasmid or another genetic element which is integrated into the bacterial chromosome.
Dot blot DNA from 24 bacteriophages isolated from P. multocida strains and all shown to be different in their lysis patterns towards a range of P.
multocida strains were bound to a nylon filter by dot blotting. The plasmid pL0A03 was radioactively labelled by nicktranslation, and used as a probe against the filter. Hybridization and washing conditions were as described above. The results are shown in Fig.
22. The probe hybridized to 22 out of 24 bacteriophages and, as expected, to the four positive controls. By using pSPE308 and pLOB03 as probes, similar results were obtained. pSPE312 gave only a slight hybridzation to some of the bacteriophage genomes. The 4.5 kb pmt gene containing Clal-PvuII fragment of pSPE312 (Fig. 5) did not show any homology to any of the bacteriophage genomes (autoradiographs are not shown).
These results snow that: there are sequences homologous to P.
multocida bacteriophage DNA on both sides of the pmt gene. This further substantiates t:he notion that the pmt gene is carried by a prophage.

~ 341 04 0 BIBLIOGRAPHY
1. K.B. Pedersen and K. Barfod, 1981. The aetiological significance of Bordetella bronchisErptica amd P. multocida in atrophic rhinitis of swine. Nord. Vet.-Med. 33: pp. 513-522.
2. J.M. Rutter a:nd X. Rojas, 1982. Atrophic rhinitis in piglets:
Differences in t:he patrnogenicity of Pasteurella multocida in combined infections with BordetE~lla bronchiseptica. Vet. Rec. 110: pp. 531-535.
3. F. Elling and K.B. federsen, 1985. The pathogenesis of persistent turbinate atrophy induced by toxigenic Pasteurella multocida in pigs.
Vet. Pathol. 22: pp. 46.9-474.
4. K.B. Pederser:, J.P. Nielsen, N.T. Foged, F. Elling, N.C. Nielsen and P. Willeberg;, 1988. Atrophic rhinitis in pigs: Proposal for a revised definition. Vet. Rec. 22: pp. 490-491.
5. K.B. Pedersen and F. Elling, 1984. The pathogenesis of atrophic rhinitis in pigs induced by toxigenic Pasteurella multocida. J. Comp.
Pathol. 94: pp. 203-214..
6. N.T. Foged, K.B. Pedersen and F. Elling, 1987, Characterization and biological effect of the P. multocida toxin. FEMS Microbiol.
Lett. 43: pp. 4'~-51.
7. E.M. Kamp, P.J. v.d. Heijden and B.J. Tetenburg, 1987, Purifica-tion of a heat labile ctermonecrotic toxin from culture fluid of Pasteurella mult:ocida. Vet. Microbiol. 13: pp. 235-248.
8. T. Nakai, A. Sawata" M. Tsuji, Y. Samejima and K. Kume, 1984.
Purification of dermanecrotic toxin from a sonic extract of Pas-teurella multoc~~:da SP-~~2 serotype D. Infect. Immun. 46: pp. 429-434.
9. C.L. Trammel, J.O. Cisar, M.J. Pabst and P. Goforth, 1979.
Stimulation of tone resorption by a factor from Actinomyces viscosus.
J. Perdont. Res.. 14: pp. 263-264.
X
,r 10. P.A. Price, 1987. Structure and function of vitamin K-dependent bone proteins. In: C. C'hristiansen, J.S. Johansen, B.J. Riis (eds.) Osteoporosis, NOrhaven ,A/S, Viborg, Denmark, pp. 656-663.

11. DELETED

12. J. Mestecky, 1987. 'The common mucosal immune system and current strategies for induction of immune response in external secretions.
J. Clin. Immunol. 7 (4): pp. 265-276.
10 13. H.W.D. Matthes, W.H. Zenke, T. Grundstrom, A. Staub, M. Wintze-rith and P. Chambon, 1984. Simultaneous rapid chemical synthesis of over 100 oligonucleotides on a microscale. The EMBO Journal 3: pp.
801-805.
14. 14. T. Mania~tis, E.:F. Frisch and J. Sambrook, 1982. Molecular 15 cloning. Cold Spicing Harbor Laboratory, Cold Spring Harbor, Nea York.
15. 15. G. Kohle:r and C. Milstein, 1975. Continuous cultures of fused cells secreting ~intibod;y of prede=fined specificity. Nature, 256: pp.
495-497.
16. G. Klein, J. Luka and J. Zeuthen, 1980. Transformation induced by 20 Epstein-Barr virus and 'the role of the nuclear antigen. Cold Spring Harbor Spmp. Qua~zt. Biol. 44: pp. 253-261.
17. J.W. Goding, 1983. Monoclonal antibodies: Principles and practice. Academic Press, London, 267 pages.
18. E. Southern, 1975, ',Detection of specific sequences among DN BA
25 fragments separated by ,gel electrophoresis. J. Mol. Biol. 98, pp.
503.
19. Rigby, P.W.J., Diec'I~ann, M., Rhodes, C., and Berg, P. 1977.
X

Labelling deoxyribonucleic acid to high specific activity in vitro by nick translation with DNA polymerase. I.J.Mol.Biol. 113: pp. 237-251.
20. Gebeyechu, G., Rao, P.Y., SooChan, P., Simms, D.A., and Klevan, L. 1987. Novel biotinylated nucleoticle - analogs for labelling and colorimetric detection o f DNA. Nucleic Acids Res. 15: p.p. 4513-4534.
21. R.K. Saiki, .9.J. Sclaarf, F. Faloona, K.B. Mullis, G.T. Horn, H.A.
Ehrlich and N.A. Arnhiem, 1985. Enzymatic amplification of ~-globin genomic sequence:a and restriction site analysis for diagnosis of sickle-cell anem:Ca. Sri~ence 230: pp. 1350-1354.
22. J.P. Nielsen, M. Bisgaard, and K.B. Pedersen, 1986, Production of toxin in strains previously classified as P. multocida. Acta Path.
Microbiol. Immunol. Scared. Sec. B, 94: pp. 203-204.
23. S. Fazekas, Groth, ;i. and D. Scheidegger, Production of mono-clonal antibodie:~: Strategy and tactics, J. Immunol. Meth. 35, 1980, pp. 1-21.
24. U.K. Laemmli., 1970. Cleavage of structural proteins during the assembly of the tread of bacteriophage T4. Nature 227: pp. 680-685.
25. J. Kyhse-AndE:rsen, X984, Eiectroblotting of multiple gels: A sim-ple apparatus wig=hout buffer tank for rapid transfer of proteins from polyacrylamide to vitro<:ellulose. J. Biochem. Biophys. Methods 10:
pp. 203-209.
26. O.J. Bjerrum, K.P. l:.arsen and M. Wilken, 1983. Some recent deve-lopments of the E:lectro:Cmmunochemical analysis of membrane proteins.
Application of Z~Jittergent, Triton X-114 and western blotting tech-nique, pp. 79-12~a. In H. Tschesche (ed.), Modern methods in protein chemistry. WalthE:r de Gruyter Berlin, New York.
27. L.J. Anderson, J.C. Hierholzer, Y.O. Stone, C. Tsou and B.F.
Fernie, 1986. IdE:ntification of epitopes on respiratory syncytial virus proteins b~~ competitive binding immunoassay. J. Clin. Micro-biol. 23: pp. 47'.i-480.
.n .. 6~ X34 40 28. J.L. Guesdon, T. Ternynck and T. Avrameas, 1979. The use of avidin-biotin interaction in immunoenzymatic techniques. J. Xis-tochem. Cytochem. 27: pp. 1131-1139.
29. H. Towbin, T. Staehlin and J. Gordon, 1979, Electrophoretic transfer of proteins from polyacrylamide gel to nitrocellulose sheets: Procedure: and applications. Proc. Natl. Acad. Sci. USA 76:
pp. 4350-4354.
30. M.J. Casabad~~n and 5~. Cohen, 1980. Analysis of gene control signals by DNA fusion and cloning in Escherichia coli. J. Mol. Biol.
138: pp. 179-207.
31. W.B. Wood, 1566. Host specificity of DNA produced by Escherichia coli: Bacterial mutations affecting the restriction and modification of DNA. J. Mol. ~~iol. IE.: pp. 118-133.
32. B. Nilsson, M. Uhlen, S. Josephson, S. Gatenbeck and L. Philip-son, 1983, An improved positive selection plasmid vector constructed by oligonucleotide mediated mutagenesis, Nucleic Acids Res. 11(22):
pp. 8019-8030.
33. F. Sanger, S. Nickli:n and A.R. Coulson, 1977. DNA sequencing With chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74: pp.
5463-5467.
34. D.J. Clark and 0. Maalee, 1967. DNA replication and the division cycle in Escheric:his co_Li. J. Mol. Biol. 23:99-112.
35. K.K. Stanley, 1983, Solubilization and immune-detection of ~-galactosidase hybrid proteins carrying foreign antigenic determi-pants. Nucleic A~:ids Rea. 11(12): pp. 4077-4092.
36. J.M. Rutter, 1983. 'ilirulence of Pasteurella multocida in atrophic rhinitis of gnotobiotic pigs infected with Bordetella bronchiseptica.
Res. Vet. Sci. 3~+: pp. :287-295.
X

~ 34~ 044 37. K-. Kaiser and N. Murry, 1985: The use of phage lambda replacement vectors in the construction of representative genomic DNA libraries.
In: DNA cloning, Vol. I, A practical approach, D.M. Glover (ed.) IRL
Press, Oxford.
X

Claims (35)

1. A vaccine for immunizing an animal, including a human being, against diseases caused by Pasteurella multocida, the vaccine comprising an immunogenically effective amount of a recombinant, immunogenic detoxified Pasteurella multocida toxin or toxin analogue having an amino acid sequence that is encoded by a DNA sequence substantially as in plasmid pSPE L, pSPE O
or pSPE P, or an immunogenic subsequence of said toxin or toxin analogue encoded by a subsequence or analogue of said DNA sequence.

2. A vaccine according to claim 1, wherein the toxin or toxin analogue is in substantially pure form.

3. A vaccine according to claim 1 or 2, in which said immunogenic subsequence of the toxin or toxin analogue is capable of binding to a monoclonal antibody which monoclonal antibody specifically binds Pasteurella multocida toxin.

4. A vaccine according to claim 3, characterised in that the monoclonal antibody is Moab P3F51.

5. A vaccine according to any one of claims 1, 2, 3 or 4, wherein the toxin or toxin analogue is fused to another polypeptide.

6. A vaccine according to claim 5, wherein the polypeptide to which the toxin is fused is selected from bovine serum albumin, ovalbumin, keyhole limpet hemocyanin, lysozyme and another immunogenic peptide.

7. A vaccine according to claim 1, which further comprises an adjuvant.

8. A vaccine according to claim 7, wherein the adjuvant is selected from Freund's complete or incomplete adjuvant, aluminium hydroxide, Bordetella pertussis, a saponin, a muramyl dipeptide, an ISCOM, peanut oil and silicone oil.

9. A vaccine according to claim 1, wherein the carrier is a macromolecular carrier.

10. A vaccine according to claim 9, wherein the macromolecular carrier is selected from polysaccharides and polypeptides.

11. A vaccine according to claim 9 or 10, wherein the toxin or toxin analogue is multivalently coupled to the macromolecular carrier.

12. A vaccine according to claim 1, wherein the toxin or toxin analogue is self-polymerized.

13. A vaccine according to claim 1, which is in the form of a tablet, granule or capsule intended for oral administration and provided with an enteric coating.

14. A vaccine according to claim 13, wherein the enteric coating is selected from shellac, cellulose acetate esters, hydroxypropylmethyl cellulose esters, polyvinyl acetate esters, and polymers of methacrylic acid and methacrylic acid esters.

15. An isolated and purified DNA fragment comprising a nucleotide fragment as shown in Fig. 10(a) - (j), encoding a Pasteurella multocida toxin or toxin analogue, or a subsequence or analogue of said DNA fragment encoding an immunogenic subsequence of the toxin or toxin analogue.

16. A DNA fragment according to claim 15, said DNA fragment comprising a nucleotide sequence substantially as shown in plasmid p SPE L, pSPE O or pSPE P, or a subsequence or analogue thereof encoding an immunogenic subsequence of the toxin or toxin analogue.

17. A DNA fragment according to claim 15, which further comprises a nucleotide sequence encoding another polypeptide fused to the nucleotide sequence encoding the toxin or toxin analogue.

18. An expression vector which is capable of replicating in a host microorganisms and which carries DNA fragment according to claim 15.

19. An expression vector according to claim 18 which is any one of the plasmids pSPE L, O or P as depicted in figure 13 under L, O or P.

20. A microorganism capable of expressing a DNA fragment according to claim 15 and containing a vector according to claim 18 or 19.

21. A microorganism according to claim 20 which is a bacterium.

22. A microorganism according to claim 21 which is a gramnegative bacterium.

23. A method of producing an immunogenic detoxified Pasteurella multocida toxin or toxin analogue, the method comprising:
(a) isolating a DNA fragment according to claim 16 or 17;
(b) inserting said sequence, resulting in the expression of the detoxified toxin or toxin analogue or a subsequence coding for an immunogenic subsequence of the toxin or toxin analogue, in an expression vector;
(c) transforming a host microorganism with the vector produced in step (b);
(d) cultivating the microorganism produced in step (c) under suitable conditions for expressing the toxin or toxin analogue;
(e) harvesting the toxin or toxin analogue from the culture.

24. A method of producing a vaccine for immunizing an animal, including a human being, against diseases caused by microorganisms producing an osteolytic toxin, the method comprising formulating the toxin or toxin analogue produced by the method of claim 23 with an immunologically acceptable earner or vehicle.

25. A non-pathogenic microorganism which carries and is capable of expressing an inserted DNA fragment according to claim 16 or 17, for use as a live vaccine for the immunization of an animal against diseases caused by microorganisms producing an osteolytic toxin.

26. A microorganism according to claim 25, wherein the nucleotide sequence coding for the toxin or toxin analogue is expressed on the outer surface of the host cell.

27. A microorganism according to claim 26, wherein the nucleotide sequence coding for the toxin or toxin analogue is fused to another nucleotide sequence encoding a surface protein or a subsequence thereof which causes the toxin or toxin analogue to be expressed on the outer surface of the host cell, optionally as a fused polypeptide.

28. A microorganism according to claim 27, wherein the surface protein is selected from fimbrial proteins or adhesins or the LamB protein from E. coli.

29. A recombinant immunogenic detoxified Pasteurella multocida toxin or toxin analogue or an immunogenic subsequence thereof, having an amino acid sequence that is encoded by a DNA sequence substantially as shown in plasmid pSPE L, pSPE O or pSPE P, or a subsequence or analogue of that DNA sequence encoding said immunogenic subsequence of the toxin or toxin analogue.

30. Use of a recombinant detoxified immunogenic Pasteurella multocida toxin or toxin analogue, according to claim 29, for preparing a vaccine for the immunization of an animal, including a human being, against diseases caused by microorganisms producing an osteolytic toxin.

31. A diagnostic agent for the detection of Pasteurella multocida toxic-producing microorganisms or bacteriophages containing a pmt gene, which comprises a labelled DNA sequence homologous with an isolated and purified DNA fragment comprising a nucleotide fragment as shown in Fig. 10(a)-(j), encoding a Pasteurella multocida toxin or toxin analogue, or a subsequence or analogue thereof.

32. A diagnostic agent according to claim 31, wherein the label is selected from radioactive isotopes, enzymes and complexing agents.

33. A method for detection of the presence of a Pasteurella multocida toxin producing microorganisms in a sample selected from bacterial suspensions, bacterial extracts, culture supernatants, animal body fluids and intermediate or final vaccine products, the method comprising amplifying a DNA sequence coding for Pasteurella multocida toxin or toxin analogue contained in said microorganism by us of oligonucleotide sequences substantially complementary to at least part of a DNA fragment comprising a nucleotide sequence coding for a Pasteurella multocida toxin or toxin analogue in the polymerase chain reaction (PCR) followed by detection with a diagnostic agent according to claim 31 or 32.

34. A method according to claim 33, wherein the detecting nucleotide sequence is provided with a label.

35. A method according to claim 34, wherein the label is selected from radioactive isotopes, enzymes and complexing agents.