Leukotoxin Gene of Pasteurella Haemolytica

This invention relates to the cloning and expression of the leukotoxin gene of Pasteurella haemolytica in a suitable host microorganism.

Bovine pneumonic pasteurellosis associated with Pasteurella haemolytica Al is a major cause of sickness and death in feedlot cattle. Although vaccination with this organism might be expected to produce immunity to the disease, experimental trials and field studies using conventional formalinized bacterins have failed to show a protective effect, in fact, vaccinated animals are frequently more suspceptible to the disease than their non-vaccinated counterparts.

Immunization with live P. haemolytica has been shown to protect cattle against experimental challenge exposure to the bacterium. P. haemolytica produces a cytotoxin with specificity for ruminant leukocytes. This may contribute to the pathogenesis of pneumonic pasteurellosis by impairing primary lung defense and subsequent immune response or by induction of inflammation as a consequence of leukocyte lysis.

The protective capability of cytotoxic supernate from P. haemolytica has been used as a vaccine. This preparation contains numerous soluble antigens from the bacterium which may be important in protection. These soluble antigens include a ruminant-leukocyte-specific cytotoxin, serotype-specific soluble surface antigens, neuraminidase and protease. An example of such vaccine and its development is disclosed in applicant's copending U.S. patent application Ser. No. 821,197 filed Jan. 22, 1986. Developments of vaccines from the crude cytotoxic supernate requires the purification and characterization of these antigens which can become a difficult and costly process.

Recent advances in molecular biology have provided a new approach in the characterization of bacterial determinants involved in pathogenicity. The particular genes which code for bacterial antigens can be isolated by molecular cloning using various recombinant DNA techniques. For example, it is known the genes which code for the heat-labile enterotoxins of E. coli of porcine and human origin have been isolated. The cholera toxin genes from Vibrio cholera have also been isolated using the cloned E. coli elt genes as a hybridization probe. Genes encoding the enzymatic moiety of the exotoxin A from Pseudomonas aeruginosa and the phospholipase C determinant of P. aeruginosa have also been successfully cloned and expressed in E. coli. These cloned genes greatly facilitate the analyses of the pathogenic and immunogenic characteristics of the toxin protein products as well as enabling the characterization of the genetic organization and regulation of expression of these bacterial toxins. Hence, an alternative method involving the isolating of genes coding for the soluble antigens of P. haemolytica would be an important advance over existing vaccine developed from culture supernatant.

According to an aspect of the invention a nucleotide sequence coding for leukotoxin which is a protein produced by the metabolism of Pasteurella haemolytica is provided.

According to another aspect of the invention a recombinant plasmid may be developed which includes the nucleotide sequence coding for the leukotoxin.

According to another aspect of the invention a phage may be developed having the nucleotide sequence.

According to another aspect of the invention the nucleotide sequence is a DNA fragment of approximately 8.7 kbp.

According to another aspect of the invention a micro-organism may be transformed with the recombinant plasmid and, in particular, may be that having the identifying characteristics of ATCC deposition #86/025.

According to another aspect of the invention, the nucleotide sequence of the leukotoxin may be altered to produce modified derivatives of the leukotoxin.

According to another aspect of the invention a process for producing a foreign proteinaceous leukotoxin having a molecular weight of approximately 100,000 daltons comprises transforming a micro-organism with the nucleotide sequence. The transformed micro-organism is then cultured under suitable conditions to yield the leukotoxin. Accordingly, the leukotoxin may be isolated and a vaccine may be developed against Pasteurella haemolytica by mixing isolated leukotoxin and/or its derivatives with suitable carriers.

According to another aspect of the invention polyclonal and monoclonal antibodies may be raised to the leukotoxin and/or derivatives.

According to another aspect of the invention the entire nucleotide sequence or a portion thereof may be used as a DNA probe.

Preferred embodiments of the invention are exemplified in the drawings wherein:

FIG. 1 is an Agarose gel electrophoresis profile of the Pasteurella haemolytica plasmid clone bank;

FIGS. 2A, 2B, and 2C are restriction maps of the recombinant plasmids which contain the Pasteurella haemolytica leukotoxin gene;

FIG. 3 is a Southern blot hybridization analysis of insert DNA from the plasmid pPH5 against total Pasteurella haemolytica genomic DNA;

FIG. 4 is a Western blot analysis of the cellular proteins from the E. coli clones carrying pPH5, pPH10 and the vector pBR322; and

FIGS. 5A, 5B, 5C, and 5D are restriction maps of additional recombinant plasmids which contain the P. haemolytica leukotoxin gene.

Based on the work disclosed in applicant's copending U.S. patent application Ser. No. 821,197, non-viable vaccines protecting calves against experimental challenge by intrabronchular inoculation with live P. haemolytica can be developed from the culture supernatant. Vaccinated calves had significantly lower pneumonic scores based on clinical signs and postmortem lesions compared to unvaccinated calves or bacterin vaccinated calves. As noted, however, such vaccines include other antigens aside from the leukotoxin present in the culture supernate. In an attempt to locate a DNA sequence which might code for the leukotoxin, as expressed in P. haemolytica, a clone bank of P. haemolytica A1 genomic DNA was constructed in Escherichia coli using the plasmid vector pBR322. From this clone bank, a collection of recombinant plasmids coding for the soluble antigens of P. haemolytica A1 were isolated. The E. coli clones were screened for the presence of P. haemolytica antigens by the colony ELISA technique (enzyme-linked immunosorbant assay) using a rabbit antiserum related to the soluble antigens of P. haemolytica A1. From this collection, plasmids coding for the leukotoxin were identified by screening protein preparations from the E. coli clones for leukotoxin activity using the neutral red assay. The presence of the leukotoxin in the E. coli cells was confirmed by serum neutralization using both rabbit and calf sera. The cloned leukotoxin was identified as a protein having a molecular weight of approximately 100,000 daltons. Such identification was through SDS-PAGE and Western blot analysis of the cytoplasmic proteins from the E. coli clones. DNA sequence analysis revealed the complete sequence of the leukotoxin gene on the cloned fragment in the plasmid.

Looking specifically at the methodology and techniques for isolating and identifying the leukotoxin gene, the following demonstrates a preferred embodiment thereof. In the following procedures and examples, the materials and methods were:

P. haemolytica A1 ATCC 43270 (biotype A, serotype 1) was originally obtained from E. L. Biberstein, University of California, Davis. Stock organisms were maintained as lyophilized cultures after freeze-drying in distilled water containing (wt/vol) 5% dextran (molecular weight, 70,000), 7% sucrose, and 1% monosodium glutamate.

Escherichia coli HB101 used as the recipient, and plasmid pBR322 was used as the vector for the construction of the clone bank. The E. coli HB101 is characterized by deposit at ATCC 33694. Their genotypes and characteristics were described by Maniatis et al (1982, Molecular cloning: a laboratory manual; Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.) and Bolivar et al (1977, Construction and characterization of new cloning vehicles. II. A multipurpose cloning system; Gene 2:95-113), respectively.

P. haemolytica was cultured in brain heart infusion broth (Difco Laboratories, Detroit, Mich.) while E. coli cultures were grown in LT medium as described by R. Y. C. Lo and L. A. Cameron, A Sample Immunological Detection Method for the Direct Screening of Genes from Clone Banks (Can. J. of Biochem. and Cell Biol., Vol 64, No 1, 73-76, 1986). Where appropriate the antibiotics ampicillin and tetracycline were supplemented at 100 and 12.5 mg/liter, respectively, after the medium was autoclaved. Restriction endonuclease, T4 DNA ligase, E. coli DNA polymerase I, and bacterial alkaline phosphate were purchased from Bethesda Research Laboratories, Inc., Gaithersburg, Md.; GIBCO, Burlington, Ontario, Canada; or Boehringer Mannheim, Dorval, Quebec, Canada and used as described by the suppliers. Low-melting-point agarose was from Bethesda Research Laboratories. [alpha-.sup.32 P]dATP (3,000 Ci/mmol) was purchased from New England Nuclear Corp., Lachine, Quebec.

A clone bank of P. haemolytica A1 was constructed by the method of Nasmyth and Reed (1980, Isolation of genes by complementation in yeast: molecular cloning of a cell-cycle gene. Proc. Natl. Acad. Sci. USA 77:2119-2123). Genomic DNA was extracted from P. haemolytica A1 by sodium dodecyl sulfate (SDS) lysis of the bacterial cells and phenol extraction (Marmer, J. 1961. A procedure for the isolation of deoxyribonucleic acid from microorganisms. J. Mol. Biol. 3:208-218). The DNA was then partially digested by the restriction endonuclease Sau3A, and fragments of 5 to 10 Kilobase pairs were recovered by a sucrose gradient centrifugation by the method of Maniatis et al (1978. The isolation of structural genes from libraries of eukaryotic DNA, Cell 15: 687-701). These fragments were inserted into the unique BamHI site of the plasmid vector pBR322 which had been dephosphorylated by bacterial alkaline phosphatase. The ligated DNA mixture was transformed into E. coli competent cells, and recombinants were recovered by selecting for ampicillin-resistant (Amp.sup.r) colonies. The recombinants were pooled and amplified in broth cultures, and plasmid DNA was prepared by cesium chloride-ethidium bromide centrifugation (Thompson, R., S. G. Hughes, and P. Broda. 1974. Plasmid identification using specific endonucleases. Mol. Gen. Genet. 133:141-149) and stored at -20.degree. C.

New Zealand White rabbits (Riemens Fur Ranches Ltd., St. Agatha, Ontario, Canada) were immunized with lyophilized P. haemolytica culture supernatant prepared essentially as described previously (Shewen, P. E., and B. N. Wilkie. 1982. Cytotoxin of Pasteurella haemolytica acting on bovine leukocytes. Infec. Immun. 35:91-94) except that 7% autologous rabbit serum was substituted for fetal bovine serum in the growth medium. Each rabbit received a total of seven inoculations of 10 mg of crude superantant in 1 ml of saline, divided equally intradermally and intravenously, at intervals of 10 to 12 days. The first two intradermal inoculations were emulsified in complete Freund adjuvant (Difco). Serum was collected 1 week after the seventh immunization. This serum had a titer to P. haemolytica surface antigens of 1/32, as measured by direct bacterial agglutination (Shewen, P. E., and B. N. Wilkie. 1982. Antibody titer to Pasteurella haemolytica A1 in Ontario beef cattle. Can. J. Comp. Med. 46:354-356), and a neutralizing titer to P. haemolytica leukotoxin of 1/640, as measured in the micro-plate assay described below.

The antibody preparation was used to screen the clone bank for production of P. haemolytica antigens by the colony enzyme-linked immunosorbent assay (ELISA) blot method (Lo and Cameron, supra). Briefly, the E. coli recombinant clones were grown on nitrocellulose paper, and the cells were lysed with chloroform vapor. The nitrocellulose paper was immersed in a blocking solution of TBS buffer (20 mM Tris hydrochloride [pH 7.5], 500 mM NaCl) containing 3% gelatin, 150 mM NaCl, 5 mM MgCl.sub.2, 1 ug of DNase I per ml, and 40 ug of lysozyme per ml for 1 h and then inoculated with the rabbit antiserum (1/200 dilution) in TBS buffer overnight. After being washed twice in TBS buffer, the nitrocellulose paper was reacted with goat anti-rabbit immunoglobulin G (IgG) conjugated with horseradish peroxidase (Bio-Rad Laboratories, Mississauga, Ontario, Canada) at a 1/2,000 dilution in TBS buffer for 1 h. The blots were washed twice in TBS buffer and developed in horseradish peroxidase color development reagents (Bio-Rad).

The restriction fragments of the recombinant plasmids were analyzed by agarose gel electrophoresis as previously described by Lo and Cameron, (supra). For Southern blot analysis (Southern, E. M. 1975. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 98:503-517), P. haemolytica genomic DNA was digested with an appropriate restriction endonuclease, electrophoresed on an agarose gel, and blotted onto nitrocellulose paper by electrophoretic transfer. The DNA fragments to be used as probes were recovered from the recombinant plasmids after enzyme digestion and purification on low-melting-point agarose gels by the method of Wieslander (Weislander, L. 1979. A simple method to recover intact high molecular weight RNA and DNA after electrophoretic separation in low gelling temperature agarose gels. Anal. Biochem. 98:305-309) and labeled with [alpha-.sup.32 P]dATP by nick-translation by the procedure of Rigby et al (Rigby, P. W. J., M. Dieckmann, C. Rhodes, and P. Berg. 1977. Labelling deoxyribonucleic acid to high specific activity in vitro by nick translation with DNA polymerase I.J. Mol. Biol. 113:237-251).

The E. coli cells carrying the recombinant plasmids were subjected to osmotic shock treatment by the method of Neu and Heppel (1965. The release of enzymes from Escherichia coli by osmotic shock and during the formation of spheroplasts. J. Biol. Chem. 240:3685-3692). The resulting cell suspension was stirred for 10 min on ice and then centrifuged, and the supernatant was recovered. The supernatant was adjusted to 0.01M Tris hydrochloride (pH 8), concentrated by centircon units (Amicon Corp., Oakville, Ontario, Canada), stored at 4.degree. C., and designated as the periplasmic protein preparation. The cell pellet was suspended in 10 ml of 0.01M Tris hydrochloride (pH 8), sonicated at 100 W for 1 min, and centrifuged at 100,000.times.g for 1 h at 4.degree. C. The sonicated supernatant recovered was designated as the cellular protein preparation. The enzymes cyclic phosphodiesterase and beta-galactosidase were assayed as markers for periplasmic and cellular proteins, respectively, as described previously (Neu, H. C., and L. A. Heppel. 1965. The release of enzymes from Escherichia coli by osmotic shock and during the formation of spheroplasts. J. Biol. Chem. 240:3685-3692). The protein concentrations of the preparations were determined by the method of Lowry (1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275).

The periplasmic and cellular proteins from the E. coli recombinant clones were analyzed by SDS-polyacrylamide gel electrophoresis by the method of Hancock and Cary (1979. Outer membrane of Pseudomonas aeruginosa: heat- and 2-mercaptoethanol-modifiable proteins. J. Bacteriol. 140:902-910). The separating gel consisted of 9% (wt/vol) acrylamide (acrylamide/bisacrylamide ratio, 40:0.8) in 0.4M Tris hydrochloride (pH 8.8), 0.09M NaCl-1% SDS, while the stacking gel consisted of 3% (wt/vol) acrylamide (acrylamide/bisacrylamide ratio, 20:0.8) in 0.13M Tris hydrochloride (pH 7), 1% SDS. For each lane, approximately 15 .mu.g of protein was mixed with the solubilization reduction mixture (Hancock, R. E. W., and A. M Carey. 1979. Outer membrane of Pseudomonas aeruginosa: heat- and 2-mercaptoethanol-modifiable proteins. J. Bacteriol. 140:902-910) and electrophoresed at 150 V with 0.025M Tris, 0.2M glycine (pH 8.4), 1% SDS as the running buffer. Protein bands were visualized by staining the 0.05% Coomassie blue R 250 (wt/vol) in 10% acetic acid 14% methanol.

For Western blot analysis (Burnette, W. N. 1981. "Western blotting": electrophoretic transfer of proteins from sodium dodecyl sulfate-polyacrylamide gels to unmodified nitrocellulose and radiographic detection with antibody and radioiodinated protein A. Anal. Biochem. 112:195-203), the proteins were transferred to nitrocellulose paper after SDS-polyacrylamide gel electrophoresis in a Trans-Blot blotting cell apparatus (Bio-Rad) with 0.025M Tris, 0.2M glycine (pH 8.4), 20% methanol as the blotting buffer. The P. haemolytica proteins were then detected by ELISA with the rabbit antiserum preparation described above.

The leukotoxic activities of periplasmic and cellular protein preparations were measured in a microplate cytotoxicity assay C. N. Greer and P. E. Shewen "Automated Colourimetric Assay for Detection of Pasteurella haemolytica Leukotoxin", Vet. Micro. 12, 31-42, 1986 with BL-3 cells, a bovine leukemia-derived B-lymphocyte cell line (originally obtained from G. Theilen, University of California, Davis), as targets. Cell viability, measured as the uptake of the vital dye neutral red, was determined by reading the optical density of each well at 540 nm with an automated spectrophotometer. After overnight dialysis against RPMI 1640 medium, quadruplicate 200-.mu.l samples (at protein concentrations of 6 mg/ml) were incubated with 2.times.10.sup.5 cell in each of four wells of a microtiter plate for 1 h at 37.degree. C. Lyophilized P. haemolytica culture supernatant (Shewen, P. E., and B. N. Wilkie. 1982. Cytotoxin of Pasteurella haemolytica acting on bovine leukocytes. Infec. Immun. 35:91-94), reconstituted at 3 mg/ml in RPMI 1640 medium, was used as the positive control for toxic activity. Percent toxicity was calculated as the percent loss of viability by comparing the mean optical density in test wells with that in control wells containing cells incubated with RPMI 1640 medium only. The heat liability of toxic activity was determined by preheating an aliquot of each sample at 56.degree. C. for 30 min. before testing. Host species specificity in toxic activity was confirmed by retesting toxic samples with canine, porcine, and human peripheral blood lymphocytes purified by density gradient centrifugation on Ficoll-Hypaque (Shimizu, M., I. C. Pan, and W. R. Hess. 1976. T and B lymphocytes in porcine blood. Am. J. Vet. Res. 36:309-317) as targets. In addition, the rabbit antiserum and a bovine serum with antitoxic activity, obtained from an infected calf, were tested in serial twofold dilutions for the ability to neutralize toxicity in one of the clone-derived samples.

In accordance with the above procedures the following methodology was used to achieve the results noted with reference to the following figures.

About 4.times.10.sup.3 E. coli colonies were recovered after transformation with the P. haemolytica DNA fragments and the pBR322 DNA ligation mixture, of which less than 1% were also tetracycline resistant. The transformants were pooled from the agar plates and amplified in broth cultures in LT medium containing ampicillin, and plasmid DNA was prepared by cesium chloride-ethidium bromide centrifugation for storage (-20.degree. C.) as the clone bank. Plasmid DNA of the clone bank was analyzed by agarose gel electrophoresis, which showed that it contained plasmids larger than the vector pBR322. In FIG. 1 lane a is plasmid vector pBR322; lane b is plasmid DNA from the P. haemolytica clone bank. Shown are covalently closed circular (ccc), and (oc) open circular forms of pBR322.

Plasmids DNA from the clone bank was transformed into E. coli competent cells, and the transformants were screened for P. haemolytica antigens by the colony ELISA blot method to detect the production of P. haemolytica antigens, Lo and Cameron, (supra). Twenty-seven positive recombinants were identified. Periplasmic and cellular proteins were prepared from the positive clones to assay for the P. haemolytica leukotoxin.

Plasmid DNA from the positive recombinant clones was analyzed by restriction endonuclease mapping, and the results indicated that some of the recombinant plasmids had the same insert DNA. Four recombinant plasmids, 10, 11, 16, and 18, were found to have the same restriction map in that a 6.3-kilobase-pair (kbp) insert was cloned in the vector pBR322 as shown in FIG. 2. The heavy lines represent pBR322 sequences and light lines represent insert sequences.

(a) Plasmid pBR322 is represented linearly at the coordinates of 3 kdbp

(b) Recombinant plasmids pPH 5 and pPH 6

(c) Recombinant plasmids pPH 10, pPH 11, pPH 16, and pPH 18

The nomenclature used in FIG. 2 has the following meaning:

P is equivalent to restriction site PstI

H is equivalent to restriction site HincII

C is equivalent to restriction site ClaI

A is equivalent to restriction site AvaI

PI is equivalent to restriction site PvuI

B is equivalent to restriction site BamHI

PII is equivalent to restriction site PvuII

B/S is equivalent to restriction site BamHI-Sau3A junctions

In this restriction mapping there are no restriction sites for the endonucleases on the insert DNA of EcoRI, BamHI, HindIII, SalI, NdeI, KpnI, SmaI and XbaI. More interestingly, the same 6.3-kbp insert DNA was also cloned in plasmids 5 and 6 in the opposite orientation (as shown in FIG. 3).

To demonstrate that the insert DNA was of P. haemolytica origin, the PstI-AvaI fragment from plasmid 5 was purified, nick-translated with [alpha-.sup.32 P]dATP, and used as a probe in Southern blot analysis against P. haemolytica genomic DNA. The results (shown in FIG. 3) indicate that the insert DNA hybridized to unique fragments of the P. haemolytica DNA digest.

Periplasmic and cellular proteins from the E. coli clones carrying the recombinant plasmids 1, 5, 8, 9, 10, 11 and 13 as well as plasmid pBR322 were assayed for leukotoxin activity. None of the periplasmic protein preparations shows significant cytotoxic activity. Cellular proteins recovered after sonication from three recombinant clones were found to be toxic for BL-3 cells. These were clones carrying plasmids 5, 10, and 11, which showed 95.5, 53.1, and 55.5% toxicity, respectively. For plasmids 10 and 11 all activity was heat labile (56.degree. C., 1 h), while 26.1% of the toxicity in the plasmid 5 preparation was heat stable. In comparison, the P. haemolytica culture supernatant (Shewen, P. E., and B. N. Wilkie. 1982. Cytotoxin of Pasteurella haemolytica acting on bovine leukocytes. Infec. Immun. 35:91-94) used as a toxicity control was 93.4% and 19.6% toxic after heating.

Cellular proteins from the clone bearing plasmid 5 showed no toxicity for canine, porcine or human peripheral blood lymphocites when tested. Likewise, the P. haemolytica control was not toxic for these non-ruminant cells.

Both bovine and rabbit antitoxic sera neutralized the toxic activity of cellular proteins from the clone bearing plasmid 5 at a 1/256 dilution, the highest dilution tested. No neutralization occurred at any dilution with the normal rabbit serum pool or the fetal calf serum pool, which were used as controls.

The protein preparations from the recombinant clones carrying plasmid 5 and plasmid 10 were analyzed by SDS-polyacrylamide gel electrophoresis followed by electrophoretic transfer to nitrocellulose paper and ELISA. Results from Coomassie blue staining of the gels indicated no new protein bands in the periplasmic protein preparations. However, an additional protein band was detected after Western blot analysis of the cellular protein preparation from the recombinant clone carrying plasmid 5 (as shown in FIG. 4). This additional protein band migrated to a position which corresponds to one of the soluble antigens in the P. haemolytica culture supernatant. The molecular weight of the extra protein was estimated to be about 100,000.

In the Western blot analysis (as shown in FIG. 4), several other bands were detected in all of the protein preparations. Since these bands were present also in the control sample, E. coli carrying pBR322, they are probably E. coli proteins which react with the antibody preparation and do not affect the present interpretation.

With reference to FIG. 5, the plasmid pPH5 is one of the recombinant plasmids isolated in accordance with the above procedures. Upon further characterization of the insert DNA of pPH5 by DNA sequencing analysis it became apparent that two open reading frames (ORF.sub.1 and ORF.sup.2) were identified. It was found, however, that the ORF.sup.1 was incomplete in that it was lacking a termination signal. By subsequent probing of the P. haemolytica A1 clone bank a second plasmid pPH5A was recovered. Such probe was accomplished on the basis of using the 1.7 kbp BamHI-PstI fragment from the pPH5 DNA sequence which was purified and used as a probe to screen the P. haemolytica A1 bank. This plasmid contains insert sequences further to the left of the BamHI site on the pPH5. With reference to FIG. 5, however, the pPH5A was lacking sequences to the right of the PvuI site on the pPH5 sequence. The sequence from the PvuI site to EcoRV on the pPH5 was determined to be part of the coding region of ORF.sub.2. From this information a new plasmid was constructed using sequences from pPH5 and pPH5A to produce the plasmid pLKT5. This plasmid contains all the insert sequences covering the regions of ORF.sub.2 and ORF.sub.1. The legend for FIG. 5 is as follows:

P is equivalent to the restriction site PstI

P.sub.1 is equivalent to the restriction site PvuI

H is equivalent to the restriction site HincIII

Ev is equivalent to the restriction site EcoRV

B is equivalent to the restriction site BamHI

P.sub.2 is equivalent to the restriction site PvuII

BII is equivalent to the restriction site BglII

B/S is equivalent to the restriction site BamHI-Sau3A junction

The ORF.sub.2 is a coding sequence of 498 nucleotides which codes for 166 amino acids. The deduced molecular weight then of the resultant protein is 19,820 daltons. ORF.sub.1 is a coding sequence of 2856 nucleotides which in turn codes for protein having 952 amino acid groups with a deduced molecular weight of 101,883 daltons. This is the leukotoxin which is produced on expression of the plasmid pLKT5 in the suitable host micro-organism such as E. coli.

The identifying characteristics of pBR322 and E. coli HB101 are already reported as noted above. The leukotoxin gene as found in the recombinant plasmid pLKT4 and as transferred in host E. coli HB101 is characterized as deposited at ATCC, deposition No. 68025.

When the protein preparations from a number of the E. coli recombinant clones were assayed for leukotoxic activity, three preparations were found to be toxic for the BL-3 cells. These were the cellular proteins from the clones carrying plasmids 5, 10, and 11. Interestingly, they all carried the same P. haemolytica insert DNA in either of the two orientations (FIG. 2). After the initial toxicity assay, three more recombinant plasmids, 6, 16 and 18, were also identified as carrying the same insert DNA. These results suggest that the leukotoxinc gene is carried on the 6.3-kbp insert DNA. Furthermore, the 6.3-kbp DNA must also carry the necessary regulatory regions for the expression of the leukotoxin.

The leukotoxic activity of these protein preparations was heat labile, which is characteristic of the P. haemolytica leukotoxin. Furthermore, similar to the P. haemolytica leukotoxin, the preparation from the recombinant clones carrying pPH5 was found to be non-toxic for canine, porcine, and human peripheral blood lymphocytes (Shewen, P. E., and B. M. Wilkie. 1982. Cytotoxin of Pasteurella haemolytica acting on bovine leukocytes. Infect. Immun. 35:91-94). In addition, both rabbit and bovine antitoxic sera neutralized the toxicity in the pPH5 preparation. Therefore, the P. haemolytica leukotoxin gene is carried in this 6.3-kbp DNA fragment. Results from Southern blot analysis, in which the insert DNA hybridized to unique DNA fragments of the P. haemolytica genome, indicate that there is only one copy of the leukotoxin gene in P. haemolytica.

The leukotoxic activity of the protein preparation from the E. coli clones carrying pPH5 was found to be twice that of the protein preparations from the clones carrying pPH10 and pPH11. Furthermore, an additional protein band was found in the pPH5 protein preparation after Western blot analysis (FIG. 4). This additional band was absent in the pPH10 and the control pBR322 protein preparations. We suggest that there is a higher level of leukotoxin produced when the 6.3-kbp insert DNA is cloned in the orientation seen in pPH5. This increased level of leukotoxin production may be a result of expression from both the leukotoxin promoter and the tetracycline promoter on pBR322 in pPH5, while in the other orientation, i.e., pPH10, the leukotoxin can be expressed only from its own promoter.

The migration of the P. haemolytica leukotoxin produced in the E. coli recombinant clone carrying pPH5 during SDS-polyacrylamide gel electrophoresis corresponded to one of the soluble antigens present in the P. haemolytica culture supernatant (FIG. 4), confirming that the molecular weight of the leukotoxin is about 100,000. Other researchers, by using physical techniques for purification, have reported molecular weights for the P. haemolytica leukotoxin of 150,000 (Himmel, M. E., M.D. Yates, L. H. Lauerman, and P. G. Squire, 1982. Purification and partial characterization of a macrophage cytotoxin from Pasteurella haemolytica. Am. J. Vet. Res. 43:764-767) and 300,000 (Baluyut, C. S., R. R. Simonson, W. J. Bemrick, and S. K. Maheswaran. 1981. Interaction of Pasteurella haemolytica with bovine neutrophils: identification and partial characterization of cytotoxin. Am. J. Vet. Res. 42:1920-1926). These may be overestimated due to the aggregation of binding of serum materials with the leukotoxin. A more accurate determination of the molecular weight of the leukotoxin is possible from the DNA sequence anlaysis of the cloned gene.

During the initial screening of leukotoxic activity from the recombinant clones, toxic activities were observed only with the cellular protein preparations from E. coli containing pPH5, pPH10 or pPH11. No leukotoxic activity was found in the respective periplasmic protein preparations. The failure of the P. haemolytica leukotoxin to be excreted by the E. coli cells is not unexpected since it has been observed on a number of occasions that foreign proteins produced in E. coli from cloned genes were not exported (Coleman, K., G. Dougan, and J. P. Arbuthnott. 1983. Cloning and expression in Escherichia coli K-12, of the chromosomal hemolysin (phospholipase C) determinant of Pseudomonas aeruginosa, J. Bacteriol. 153:909-915); (Gray, G. L., D. H. Smith, J. S. Baldridge, R. N. Haskins, M. L. Vasil, E. Y. Chen, and H. L. Heyneker. 1984. Cloning, nucleotide sequence and expression in Escherichia coli of the exotoxin A structural gene of Pseudomonas aeruginosa. Proc. Natl. Acad. Sci. USA 81:2645-2649).

By way of DNA sequencing analysis in accordance with standard procedures, the nucleotide sequence of ORF.sub.1 and ORF.sub.2 is as follows:

The sequence is numbered from -469 at the rightmost EcoRV site of FIG. 5 on the pLKT5 plasmid and counting leftward. Nucleotide No. 1 is the first nucleotide of the coding region of ORF.sub.2 at approximately nucleotide 510 there is the break between ORF.sub.2 and ORF.sub.1. ##STR1##

Beneath the codons of the DNA sequence, the corresponding amino acid has been identified. It is appreciated that the letters indicating the nucleic acids of the DNA sequence and the expressed amino acids are in accordance with International lettering which is well understood by those skilled in the art.

Having isolated the gene coding for the leukotoxin of P. haemolytica it is appreciated that many steps may be taken in improving animal health care. Based on this information and in the expression of the gene, a suitable vaccine is prepared without the impurities of a vaccine prepared from the culture supernatant. The DNA sequence may be used in its entirety or fragment or fragments thereof as a DNA probe. The expressed leukotoxin protein may be used as the pure antigen in developing poly- and monoclonal antibodies. It is appreciated that the DNA sequence coding for the leukotoxin may be transformed into a suitable host micro-organism by either a suitable plasmid such as pBR322 or a suitable phage lamda.

The suitable host micro-organism such as the E. coli may be cultured under suitable conditions to produce the leukotoxin. The leukotoxin may be isolated from the cells after culture is complete. A sonicated host cell protein preparation can be fractionated by preparative polyacrylamide gel electrophoresis using a denaturing system at pH 9.7 where the leukotoxin has been shown to migrate at a low rate. Using in vivo expression of the leuktoxin in the E. coli mini cell system, the position of the leukotoxin on a non-denaturing system by PAGE and Western blot analysis is achieved. Using the cloned protein, rabbits can be immunized by subcutaneous and intravenous inoculation and evaluated for development of antibodies to the cloned antigen by micro immunodiffusion and dot blot analysis. The antibodies are tested for reactivity with P. haemolytica surface antigens by micro-agglutination and for the ability to neutralize the cytotoxin.

In establishing that the cloned leukotoxin antigen proves to be immunogenic in rabbits, tests may then be conducted for immunogenicity and protective capabilities in challenge experiments involving calves. Such analysis establishes the effectiveness of vaccines prepared from the genetically produced leukotoxin against pneumonia in calves.

While preferred embodiments of this invention have been described and illustrated herein, the person skilled in the art will appreciate that changes and modifications may be made therein without departing from the spirit and scope of this invention as defined in the appended claims.