GB2192185A - Monoclonal antibodies to Pseudomonas aeruginosaflagella - Google Patents

Description

1 GB 2 192 185 A 1

SPECIFICATION

Monoclonal antibodiesto Pseudomonas aeruginosa flagella Fieldof the invention 5

The present invention relatestothe application of immunological techniquesto provide novel materials useful in diagnosing andtreating bacterial infections and, more particularly, to the production and application of monoclonal antibodies that are capable of recognizing Pseudomonas aeruginosa flagella.

Backgroundof theinvention 10

Gram-negative diseaseand its most serious complications, e.g., bacteremia and endotoxemia, arethe cause of significant morbidityand mortality in human patients.This is particularly true of thegram-negative organism Pseudomonas aeruginosa, which has been increasingly associated with bacterial infections,especially nosocomial infections, overthe lastfiftyyears.

Duringthe pastfewdecades, antibiotics have been thetherapyof choice for controlling gram-negative 15 disease. The continued high morbidity and high mortality associated with gram-negative bacterial disease, however, is indicative of the limitations of antibiotic therapy, particularly with respectto P. aeruginosa. (See, forexample, Andriole, V.G., "Pseudomonas Bacteremia: Can Antibiotic Therapy Improve Survival?",J. Lab. Clin. Med. [1978194:196- 199). This has prompted the search for alternative methods of prevention and treat- ment. 20 One method that has been considered is augmentation of the host's immune system by active or passive immunization. For instance, it has been observed that active immunization of humans or experimental ani mals with whole cell bacterial vaccines or purified bacterial endotoxins from A aeruginosa leads tothe development of specific opsonic antibodies directed primarily against determinants on the repeating oli gosaccharicle units of the lipopolysaccharide (LPS) molecules located on the outer cell membrane of A 25 aeruginosa (see Pollack, M., Immunoglobulins: Characteristics and Uses ofIntravenous Preparations, AIv ing, B.M., and Finlayson, J.S., eds., pp. 73-79, U.S. Department of Health and Human Services, 1979). Such antibodies, whether actively engendered or passively transferred, have been shown to be protective against the lethal effects of A aeruginosa infection in a variety of animal models (Pollack, supra) and in some preliminary investigations with humans (see Young, L.S. and Pollack, M., Pseudomonas aeruginosa, Sabath, 30 L.,ed., pp. 119-132, Hans Huber, 1980).

The above reports suggest that immunotherapeutic approaches could be utilized to prevent and treat bacterial disease due to A aeruginosa, such as by administering pooled human immune globulins thatcon tain antibodies againstthe infecting strain(s). Human immune globulins are defined herein as that portion of fractionated human plasma that is enriched for antibodies, among which are represented specific antibodies 35 to strains of P. aeruginosa. Dueto certain inherent limitations in using human immune globulin components, this approach to treatment of disease due to A aeruginosa remains under investigation (see, forexample, Collins, M.S. and Roby, R.E.,Am. J. Med., 76(3A): 168-174, [1984]), and as yet there are no commercial prod ucts available utilizing these components.

One such limitation associated with immune globulin compositions is thatthey consist of pools of samples 40 from a thousand or more donors, such samples having been preselected for the presence of particular anti Pseudomonas antibodies. This pooling leads to an averaging of individual antibodytiters which, at best, results in modest increases in the resultanttiter of the desired antibodies.

Another limitation is thathe preselection process itself requires expensive, continuous screening of the donor pool to assure product consistency. Despite these efforts, the immune globulin products can still have 45 considerable variability from batch to batch and among products from different geographic regions.

Yet another such limitation inherent in immune globulin compositions is thattheir use results in the coinci dent administration of large quantities of extraneous proteinaceous substances (which may includeviruses, such as those recently shown to be associated with Acquired Immune Deficiency Syndrome, orAIDS), hav ing the potential to cause adverse biologic effects. The combination of low titers of desired antibodies and 50 high content of extraneous substances may often limit, to suboptimal levels, the amount of specific and thus beneficial immune globulin(s) administrable to the patient.

In 1975 Kohler and Milstein reported their seminal discovery that certain mouse cell lines could be fused with mouse spleen cells to create hybridomas each of which which would secrete antibodies of a single specificity, i.e., monoclonal antibodies (Kohler, G., and Milstein, 6., Nature, 256:495-497 [1975]). With the 55 advent of this technology it became possible, in some cases, to produce large quantities of exquisitely specific murine antibodies to a particular determinant or determinants on antigens. Subsequently, using later-developed technologies, it became possible to produce human monoclonal antibodies (see, e.g., U.S.

Patent No. 4,464,465, which is incorporated herein by reference).

It is recognized that in some situations mouse monoclonal antibodies or compositions of such antibodies 60 may present problems for use in humans. For example, it has been reported that mouse monoclonal anti bodies used in trial studies forthe treatment of certain human disease can elicit an immune responsethat renders them noneffective (Levy, R.L., and Miller, R.A., Ann. Rev. Med., 34:107-116 [19831). However, with recent advances in recombinant DNAtechnology, such as the production of chimeric mouse/human mono- clonal antibodies, these problems maybe abated. Also, methods forthe production of human monoclonal 65 2 GB 2 192 185 A 2 antibodies are now available (see, Human HybridomasandMonoclonalAntibodies, Engleman, E.G., etal., eds., Plenum Publishing Corp. [19851, which is incorporated herein by reference).

Using hybridoma and/or cell transformation technology, a number of groups have reported the production of monoclonal antibodies protective against A aeruginosa infections. Monoclonal antibodies have been produced that are reactive with various epitopes of A aeruginosa, including single and multi-serotype 5 specific surface epitopes, such as those found in LPS molecules of the bacteria (see, forexample, commonly assigned pending U.S. patent application serial numbers 734,624 and 807, 394, which are both incorporated herein by reference). Also, protective monoclonal antibodies specific for A aeruginosa exotoxin A have been produced (see, forexample, commonly assigned U.S. patent application serial number 742,170, which is incorporated herein by reference). 10 While utilizing monoclonal antibodies specific for the LPS region of A aeruginosa, or the bacteria's exo toxins, may provide sufficient protection in some situations, generally it is preferable to have broader protec tioncapability. For example, in prophylactic treatments for potential infections in humans, it would be prefer able to administer an antibody or antibodies protective against a plurality of A aeruginosa strains. Similarly, in therapeutic applications where the serotype(s) of the infecting strain(s) is not known, itwould be prefer- 15 able to administer an antibody or combination of antibodies effective against most, if not all, of the clinically important A aeruginosa serotypes, ideally by providing antibodies reactive across traditional serotyping schemes.

One aspect of A aeruginosa physiology that has been shown to contribute to the organism's virulence is motility, a capability resulting primarily from the presence of a flagellum (see, Montie, T., et al. [19821, Infect 20 andImmun., 38,1296-1298). A aeruginosa is characterized by having a single flagellum atone end of its rod-shaped structure. Burned mouse model studies have shown that a greater percentage of mice survived when non-motile A aeruginosa strains were inoculated into experimental burns than if motile strains were utilized. (McManus, A., et al. [19801, Burns, 6:235-239 and Montie, T., et al. [19821, Infect, and Immun., 38:1296-1298). Other studies on the pathogenesis of P. aeruginosa have alleged that animals immunized with 25 flagella antigen preparations were protected when burned and infected with motile strains of the bacteria (see, Holder, I.,etal. [1982],Infectandlmmun, 35:276-280).

Importantly, A aeruginosa flagella have been studied by serological methods and have been reported to fall into two major antigenic groups designated H1 and H2 by B. Lanyi (1970, Acta Microbiol. Acad. Sci. Hung., 17:35-48) and type a and type b byAnsorg, R. (1978, ZbI. Bakt. Hyg., 1. Abt. Orig. A, 242:228-238). Serological 30 typing of flagella by both laboratories showed that H1 flagella (Lanyi, B. , supra) orflagella type b (Ansorg, R., supra) was serologically uniform, i.e., no subgroups have been identified. This serologically uniform flag ellar type will be referred to as type b. The other major antigen, H2 (Lanyi, B., supra) or type a flagella (Ansorg, R., supra) contained five subgroups. This antigen will be referred to as flagella type a, and the fivesubgroups as ao, al, a2, a3, and a4. The five subgroups of type a are expressed in varying combinations on different strains 35 of typea bearing P. aeruginosa with the exception of the antigen ao. The ao antigen was found on all type a flagella, although the degree to which it was expressed varied among strains.

A serotyping scheme based on the heat stable major somatic antigens of A aeruginosa is referred to as the Habs scheme, which has recently been incorporated into the International Antigenic Typing System scheme.

(See, Liu, Int. J Syst. Bacteriol., 33:256 [19831.) The flagella types of A aeruginosa Habs reference strains 40 have been characterized by immunofluorescence with polyclonal sera by R. Ansorg (1978, Zbl. Bakt. Hyg., 1.

Abt. Orig. A, 242:228-238) or by slide coagglutination (Ansorg, R., et al. , 1984,J. Clin. Microbiol., 20:84-88).

Habs strains 2,3,4,5,7, 10, 11, and 12 are flagella type b bearing strains, and Habs strains 1, 6,13, and 9 bear type a flagella. Thus, a large number of strains of A aeruginosa could possibly be recognized by a small number of monoclonal antibodies specific for flagellar proteins. 45 Accordingly, there exists a significant need for monoclonal antibodies capable of reacting with epitopes on flagellar proteins and, in some cases, also providing protection against multiple serotypes of A aeruginosa.

Further, some of these antibodies should be suitable for use as prophylactic and therapeutic treatments of P.

aeruginosa infections, as well as the diagnosis of such infections. The present invention fulfills these needs.

50 Summary ofthe invention

Novel cell lines are provided which can produce monoclonal antibodies capable of binding to flagella present on most strains of A aeruginosa bacteria. The monoclonal antibodies specifically react with epitopes on flagella proteins of A aeruqinosa and can distinguish between type a and type b flagella of the bacteria.

Additionally, a method is provided fortreating a human susceptible to infection or already infected with A 55 aeruginosa by administering a prophylactic ortherapeutic amount of a composition comprising at least one monoclonal antibody or binding fragment thereof capable of reacting with the flagella of A aeruginosa strains, the composition preferably also including a physiologically acceptable carrier. The composition may also contain any one ormore of the following: additional monoclonal antibodies capable of reacting with A aeruginosa exotoxin A; monoclonal antibodies capable of reacting with serotype determinants on the LPS of 60 A aeruginosa; a gamma globulin fraction from human blood plasma; a gamma globulin fraction from human blood plasma, where the plasma may be obtained from a human exhibiting elevated levels of im munoglobulins reactive with A aeruginosa; and one or more antimicrobial agents. Further, clinical uses of the monoclonal antibodies are provided, including the production of diagnostic kits.

65 3 GB 2 192 185 A 3 Description of the specific embodiments

In accordance with the present invention, novel cells capable of producing monoclonal antibodies and compositions comprising such antibodies are provided, such compositions being capable of selectively re cognizing the flagella presenton a pluralityof A aeruginosa strains, where individual antibodies typically recognize one type of A aeruginosa flagella. The subject cells have identifiable chromosomes in which the 5 germ-line DNA from them or a precursor cell has rearranged to encode an antibody having a binding site for an epitope on a flagellar protein common among certain A aeruginosa strains. Fortype a flagellar proteins, pan-reactive monoclonal antibodies can be produced; and fortype b flagellar proteins, antibodies that are pan-reactive or react with at least about 70% of the flagellar bearing strains are included. These monoclonal antibodies can be used in a wide variety of ways, including diagnosis andtherapy. 10 The monoclonal antibodies so provided are particularly useful in the treatment or prophylaxis of serious disease caused by A aeruginosa. The surface proteins on the flagella of A aeruginosa would be availablefor direct contact bythe antibody molecules, thus likely inhibiting the motility of the organism and/orfacilitating other effects beneficial to the infected hosts.

The preparation of monoclonal antibodies can be accomplished by immortalizing a cell line capable of 15 expressing nucleic acid sequences that code for antibodies specific for an epitope on the f lagel lar proteins of multiple strains of A aeruginosa. The immortalized cell line maybe a mammalian cell line that has been transformed through oncogenesis, bytransfection, mutation, orthe like. Such cells include myeloma lines, lymphoma lines, or other cells lines capable of supporting the expression and secretion of the im- munoglobulin, or binding fragment thereof, in vitro. The imunoglobu [in or fragment maybe a natu rally- 20 occurring immunoglobulin of a mammal otherthan the preferred mouse or human sources, produced by transformation of a lymphocyte, particularly a splenocyte, by means of a virus or byfusion of the lymphocyte with a neoplastic cell, e.g., a myeloma, to produce a hybrid cell line. Typically, the splenocyte will be obtained from an animal immunized against flagellar antigens or fragments thereof containing an epitopic site. Im munization protocols are we] I known and can vary considerably yet remain effective. (See, Golding, Mono- 25 clonalAntibodies: Principles andPractice, Academic Press, N.Y. [19831, which is incorporated herein by reference).

The hybrid cell lines maybe cloned and screened in accordance with conventional techniques, and anti bodies in the cel I supernatants detected that are capable of binding to A aeruginosa flagellar determinants.

The appropriate hybrid cell lines may then be grown in large-scale culture or injected into the peritoneal 30 cavity of an appropriate host for production of ascites fluid.

In one embodiment of the present invention, the cells are transformed human lymphocytes that produce human monoclonal antibodies, preferably protective in vivo, to accessible epitopes specific for at least one flagellar protein. The lymphocytes can be obtained from human donors who are or have been exposed tothe appropriate flagella-bearing strains of P, aeruginosa. A preferred cell- driven transformation process is described in detail in U.S. Patent No. 4,464,465, which is incorporated herein by reference.

Byvirtue of having the antibodies of the present invention, which are known to be specific fortheflagellar proteins, in some cases the supernatants of subsequent experiments may be screened in a competition assaywith the subject monoclonal antibodies as a means to identify additional examples of anti-flagellar monoclonal antibodies. Thus, hybrid cell lines can be readily produced from a variety of sources based on the 40 availability of present antibodies specific forthe particular f lagel lar antigens.

Alternatively, where hybrid cell lines are available that produce antibodies specificforthe subject epitopic sites, these hybrid cell lines may be fused with other neoplastic B-cells, where such other B-cells may serve as recipients for genomic DNA coding forthe receptors. While rodent, particularly murine, neoplastic B-cells are most commonly utilized, other mammalian species may be employed, such as lagomorpha, bovine, ovine, 45 equine, porcine, avian or the like.

The monoclonal antibodies may be of any of the classes orsubclasses of immunoglobulins, such as IgM, IgD, IgA, IgE, or subclasses of IgG known to each species of animal. Generally, the monoclonal antibodies may be used intact, or as binding fragments, such as Fv, Fab, F(ab')2, but usually intact.

The cell lines of the present invention mayfind use otherthan forthe direct production of the monoclonal 50 antibodies. The cell lines may be fused with other cells (such as suitably drug-marked human myeloma, mouse myeloma, or human lymphoblastoid cells), to produce hybridomas, and thus provide forthetransfer of the genes encoding the monoclonal antibodies. Alternatively, the cell lines may be used as a source of the chromosomes encoding the immunoglobulins, which may be isolated and transferred to cells bytechniques otherthan fusion. In addition, the genes encoding the monoclonal antibodies may be isolated and used in 55 accordance with recombinant DNAtechniques forthe production of the specific immunoglobulin in a variety of hosts. Particularly, by preparing cDNA libraries from messenger RNA, a single cDNA clone, coding forthe immunoglobulin and free of introns, may be isolated and placed into suitable prokaryotic or eukaryotic expression vectors and subsequently transformed into a host for ultimate bulk production. (See,generally, U,S. Nos 4,172,124; 4,350,683; 4,363,799; 4,381,292; and 4,423,147. See also, Kennett et al.,Monoclonal 60 Antibodies, Plenum, NewYork [19801, and references cited therein, all of which are incorporated herein by reference.) More specifically, in accordance with hybrid DNAtechnology, the immunoglobulins orfragments of the present invention may be produced in bacteria oryeast. (See, Boss, et al., Nucl. Acid. Res., 12:3791 and Wood et al., Nature 314:446, both of which are incorporated herein by reference.) For example, the messenger RNA 65 4 GB 2 192 185 A 4 transcribed from the genes coding forthe light and heavy chains of the monoclonal antibodies produced by a cell line of the present invention maybe isolated by differential cDNA hybridization employing cDNAfrom BALB/c lymphocytes other than the subject clone. The mRNAthat does not hybridize will be rich forthe messages coding forthe desired immunoglobulin chains. As necessary, this process can be repeated to further enhance the the desired m RNA levels. The subtracted mRNA composition may then be reverse- 5 transcribed to provide for a cDNA mixture enriched for the desired sequences. The RNA maybe hydrolyzed with an appropriate RNase and the ssDNA made double-stranded with DNA polymerase I and random primers, e.g,, random lyfragmented calf thymus DNA. The resulting dsDNA maythen be cloned by insertion into an appropriate vector, e.g., virus vectors, such as lambda vectors or plasmid vectors (such as pBR322, pACYC1 84, etc.). By developing probes based on known sequences forthe constant regions of the light and 10 heavy chains, those cDNA clones having the gene coding forthe desired light and heavy chains can be identified by hybridization. Thereafter, the genes may be excised from the plasmids, manipulated to remove superfluous DNA upstream forthe initiation codon or constant region DNA, and the introduced in an appropriate vectorfor transformation of a hostand ultimate expression of the gene.

Conveniently, mammalian hosts (e.g., mouse cells) may be employed to process chain (e.g., join the heavy 15 and light chains) to produce an intact immunoglobulin, and furthermore, secrete the immunoglobulin free of the leader sequence, if desired. Alternatively, one may use unicellular microorganisms for producing thetwo chains, where further manipulation maybe required to remove the DNA sequences coding for the secretory leader and processing signals, while providing for an initiation codon at the 5'terminus of the sequence coding for the heavy chain. In this manner, the immunoglobulins can be prepared and processed so as to be 20 assembled and glycosylated in cel Is other than mammalian cells. If desired, each of the chains maybe truncated so as to retain at least the variable region, which regions may then be manipulated to provide for other immunoglobulins orfragments specific forthe flagellate epitopes.

The monoclonal antibodies of the present invention are particularly usefu I because of their specificityfor antigens across almost all P. aeruginosa variants presently known. Also, some of the monoclonal antibodies 25 are protective in vivo, permitting incorporation into pharmaceutical products, such as antibody com binations for bacteria[ infections.

Monoclonal antibodies of the present invention can also find a wide variety of utilities in vitro. Byway of example, the monoclonal antibodies can be utilized for microorganism typing, for isolating specific P. aerugi nosa strains, for selectively removing A aeruginosa cells in a heterogeneous mixture of cells, or the like. 30 For diagnostic purposes, the monoclonal antibodies may either be labeled or unlabeled. Typically, dia gnostic assays entail detecting the formation of a complex through the binding of the monoclonal antibody to the flagellum of the A aeruginosa organism. When unlabeled, the antibodies find use in agglutination assays. In addition, unlabeled antibodies can be used in combination with other labeled antibodies (second antibodies) that are reactive with the monoclonal antibody, such as antibodies specific for immunoglobulin. 35 Alternatively, the monoclonal antibodies can be directly labeled. A wide variety of labels maybe employed, such as radionuclides, fluorescers, enzymes, enzyme substrates, enzyme cofactors, enzyme inhibitors, ligands (particularly haptens), etc. Numerous types of immunoassays are available, and byway of example, some include those described in U.S. Patent Nos, 3,817,827; 3,850,752; 3, 901,654; 3,935,074; 3,984,533; 3,996,345; 4,034,074; and 4,098,876, all of which are incorporated herein by reference. 40 Commonly, the monoclonal antibodies of the present invention are utilized in enzyme immunoassays, where the subject antibodies, or second antibodies from a different species, are conjugated to an enzyme.

When a sample containing A aeruginosa of a certain serotype, such as human blood or lysate thereof, is combined with the subject antibodies, binding occurs between the antibodies and those molecules exhibit ing the desired epitope. Such cells may then be separated from the unbound reagents, and a second antibody 45 (labeled with an enzyme) added. Thereafter, the presence of the antibody- enzyme conjugate specifically bound to the cells is determined. Other conventional techniques well known to those skilled in the art may also be utilized.

Kits can also be supplied for use with the subject antibodies for detecting A aeruginosa in solutions orthe % presence of P. aeruginosa flagellar antigens. Thus, the subject monoclonal antibody composition of the 50 present invention maybe provided, usually in a lyophilized form, either alone or in conjunction with additi onal antibodies specific for other gram-negative bacteria. The antibodies, which maybe conjugated to a label or unconjugated, are included in the kits with buffers, such as Tris, phosphate, carbonate, etc., stabilizers, biOGides, inert proteins, e.g., bovine serum albumin, or the like. Generally, these materials will be present in less than about 5% wt. based on the amount of active antibody, and usually present in total amount of at least 55 about 0.001 %wt. based again on the antibody concentration. Frequently, it will be desirable to include an inert extender or excipient to dilute the active ingredients, where the excipient maybe present in from about 1 to 99% wt. of the total composition. Where a second antibody capable of binding to the monoclonal anti body is employed, this wil I usually be present in a separate vial. The second antibody is typically conjugated to a label and formulated in an analogous manner with the antibody formulations described above. 60 The monoclonal antibodies, particularly human monoclonal antibodies, of this invention can also be incor porated as components of pharmaceutical compositions containing a therapeutic or prophylactic amount of at least one of the monoclonal antibodies of this invention with a pharmaceutically effective carrier. A pharm aceutical carrier should be any compatible, nontoxic substance suitable to deliverthe monoclonal antibodies to the patient. Sterile water, alcohol, fats, waxes, and inert solids maybe used as the carrier. Pharmaceutic- 65 GB 2 192 185 A 5 ally accepted adjuvants (buffering agents, dispersing agents) may also be incorporated into the pharmaceuti cal composition. Such compositions can contain a single monoclonal antibody so as to be specific for strains of one flagellar type of P. aeruginosa. Alternatively, a pharmaceutical composition can contain two or more monoclonal antibodies to forma "cocktail". For example, a cock-tail containing monoclonal antibodies against both types off lagella or against groups of the various A aeruginosa strains (e.g., different serotypes) 5 would be a universal product with activity against the great majority of the clinical isolates of that particular bacterium.

The mole ratio of the various monoclonal antibody components wil I usually not differ by more than a factor of 10, more usually by not more than a factor of 5, and wil I usually be in a mole ratio of about 1:1-2 to each of the other antibody components. 10 The monoclonal antibodies of the present invention may also be used in combination with existing blood plasma products, such as commercially available gamma globulin and immune globulin products used in prophylactic ortherapeutic treatment of A aeruginosa disease in humans. Preferably, for immune globulins the plasma will be obtained from human donors exhibiting elevated levels of immunog lobu lins reactive with A aeruginosa. (Seegenerally, the compendium "Intravenous Immune Globulin and the Compromised 15 Host," Amer. J. Med., 76(3a), March 30,1984, pp. 1-231, which is incorporated herein by reference.) The subject monoclonal antibodies can be used as separately administered compositions given in con junction with antibiotics or antimicrobial agents. Typically, the antimicrobial agents may include an anti pseudomonal penicillin (e.g., carbenicillin) in conjunction with an aminoglycoside (e.g., gentamicin,tobra mycin, etc.), but numerous additional agents (e.g., cephalosporins) well- known to those skilled in the art may 20 also be utilized.

The monoclonal antibodies and pharmaceutical compositions thereof of this invention are particularly useful for oral or parenteral administration. Preferably, the pharmaceutical compositions may be administe red parenterally, i.e., subcutaneously, intramuscularly or intravenously. Thus, this invention provides com positionsfor parenteral administration which comprise a solution of the monoclonal antibody or a cocktail 25 thereof dissolved in an acceptable carrier, preferably an aqueous carrier. Avariety of aqueous carriers can be used, e.g., water, buffered water, 0.4% saline, 0.3% glycine andthe like. These solutions are sterile and generallyfree of particulate matter. These compositions may be sterilized byconventional, well known steril ization techniques. The compositions maycontain pharmaceutically acceptable auxiliary substances as re quired to approximate physiological conditions such as pH adjusting and buffering agents, toxicity adjusting 30 agents and the like,for example sodiumacetate, sodium chloride, potassium chloride, calcium chloride, sodium lactate, etc. The concentration of antibody in these formulations can varywidely, i.e.,from lessthan aboutO.5%, usually at orat least about 1%to as much as 15 or20% byweight and will be selected primarily based on fluid volumes, viscosities, etc., preferably for the particular mode of administration selected.

Thus, a typical pharmaceutical composition for intramuscular injection could be made up to contain 1 ml 35 sterile buffered water, and 50 mg of monoclonal antibody. Atypical composition for intravenous infusion could be made up to contain 250 ml of sterile Ringer's solution, and 150 mg of monoclonal antibody. Actual methods for preparing parenterally administrable compositions will be known or apparentto those skilled in the art and are described in more detail in, for example, Remington's Pharmaceutical Science, 15th Ed., Mack Publishing Company, Easton, Pennsylvania 0 980), which is incorporated herein by reference. 40 The monoclonal antibodies of this invention can be Iyophilized forstorage and reconstituted in a suitable carrier priorto use. This technique has been shown to be effective with conventional immune globulins and art-known lyphilization and reconstitution techniques can be employed. It will be appreciated bythose skilled in the art that Iyophilization and reconstitution can lead to varying degrees of antibody activity loss (e.g.,with conventional immune globulins, IgM antibodies tend to have greater activity loss than IgG antibodies) and 45 that use levels may have to be adjusted to compensate.

The compositions containin, g the present monoclonal antibodies or a cocktail thereof can be administered for the prophylactic and/or therapeutic treatment of A aeruginosa infections. In therapeutic application, com positions are administered to a patient already infected with one or more A aeruginosa serotypes, in an amount suff icientto cure or at least partially arrestthe infection and its complications. An amount adequate 50 to accomplish this is defined as a "therapeutically effective dose." Amounts effective forthis use will depend upon the severity of the infection and the general state of the patient's own immune system, but generally range from about 1 to about 200 mg of antibody per kilogram of body weight with dosages of from 5 to 25 mg per kilogram being more commonly used. It must be kept in mind that the materials of this invention may generally be employed in serious disease states,that is, life-threatening or potentially life-threatening situ- 55 ations, especially bacteremia and endotoxemia, due to A aeruginosa.

In prophylactic applications, compositions containing the present antibody or a cocktail thereof are admin istered to a patient not already infected by A aeruginosa to enhance the patient's resistance to such potential infection. Such an amount is defined to be a "prophylactically effective dose." In this use, the precise amounts again depend upon the patient's state of health and general level of immunity, but generally range 60 from 0.1 to 25 mg per kilogram, especially 0.5 to 2.5 mg per kilogram.

Single or multiple administrations of the compositions can be carried out with dose levels and panern being selected bythe treating physician. In any event, the pharmaceutical formulations should provide a quantity of the antibody(ies) of this invention suff icient to effectively treat or prophylaxthe patient.

65 6 GB 2 192 185 A 6 Experimental Example 1

Example 1 demonstratesthe methodology used to preparea murine monoclonal antibodythat binds specificallyto A aeruginosa flagella. 5 Three-month-old BALB/c mice were immunized intraperitoneally eighttimes with viable A aeruginosa Fisher immunotype land Fisher immunotype 2 (A.T.C.C. 427312 and #27313) bacteria everyone to two weeks for a total of nine weeks. The initial doses of bacteria were 8 X 106 and 1 X 107 organisms per mouse for A aeruginosa Fisher immunotype 1 and Fisher immunotype 2, respectively, andthe dosage was increased 30-to 60-fold during thecourseof immunizations. 10 Threedays afterthe last injection, the spleen from one mousewas removed asepticallyand a singlecell suspension was prepared bygentle rotation of the organ between thefrosted ends of two sterile glassslides.

Spleen mononuclear cells were combined in a 4:1 ratiowith log phase mouse myeloma cells (NSI-l,ob tainedfrom Dr. C. Milstein, Molecular Research Council, Cambridge, England) andfusedto create hybrid omas according to the procedure described by Tam et al. (1982, Infect, Immun., 36:1042-1053). Thefinal 15 hybrid cell suspension was diluted to a concentration of 1.5 X 106 cells per mi in RPMI-hybrid-HAT(RPMI 1640 [Gibco, Grand Island, NY] containing 15% heat-inactivated fetal calf serum, 1 mM sodium pyruvate, 100 I.Lg/ml of penicillin and streptomycin, 1.0 X 10 -4 M hypoxanthine,4.0 X 10-7 M aminopterin, and 1.6 X 10-5 M thymidine), which included 2.0 X 106 per mi freshly prepared BALB/cthymocytes as feedercells.

The mixturewas plated (20ORI perwell) into 96-well plates (3596, Costar, Cambridge, MA). Cultureswere 20 fed by removal and replacementof 50% of thevolume of each well withfresh RPM 1-hybrid-HAT every two to three days. Culture supernatants were assayedforthe presence of anti-P. aeruginosa antibodies byenzyme linked immunosorbant assay (ELISA) when the cell growth reached approximately 40% confluency in the wells, usuallywithin 7-10 days.

The culture supernatants of the hybrid cells were assayed simultaneously on outer membrane pre- 25 parationsfrom each of thetwo immunizing bacteria. Outer membrane preparations were isolated by a modi fication of the method of Tam etal. (1 982,lnfect Immun., 36:1042-1053), which is incorporated herein by reference. Bacteria (P. aeruginosa Fisher immunotype 1 and Fisher immunotype 2) were inoculated into trypticase soy broth (TSB) and grown 16-18 hours at 340C with aeration in a gyratory shaker bath. The bacteria were harvested by centrifugation and washed twice with phosphate buffered saline (PBS, 0.14 M NaCl, 3 mM 30 KCI, 8 mM Na2HP04-7 H20,1.5 mM KH2PO4, PH 7.2) containing 150 trypsin inhibitory units (T.I.U.) aprotinin per mi (Sigma, St. Louis, MO).

The pelletfrom the final centrifugation was resuspended in 0.17 M triethanolamine, 20 mM disodiurn ethylenediamine tetraacetic acid (EDTA), and then homogenized on ice for 10 minutes. The debris was pelle ted from the homogenate at 14,900 X g and discarded, and the supernatantwas centrifuged again as above. 35 The pelletwas again discarded, andthe membranes were pelleted from the supernatant by centrifugation at 141,000 X g for one hour. The supernatantwas discarded and the membrane pellets were stored overnight at 40C in 10 mi PBS containing 75 T.W. aprotinin per mi. The next day the pellets were resuspended byvortex ing and then were aliquoted and stored at -70'C. The protein content of each was determined bythe method of Lowry et al. (1951,J. Biol, Chem., 193:265-275). 40 The antigen plates forthe ELISAwere prepared as follows. The outer membrane preparations were diluted to 5 Kg per mi protein in PBS and 50KI of the solutions were plated into each well of 96-well plates (Linbro 076-031-05, Flow Laboratories, Inc., McLean, VA), sealed and incubated overnight at 37'C. Unbound antigen was flicked out of the plates and 1 001il of 5% (w/v) bovine serum albumin (BSA) in PBS was added to each well and the plates incubated for one hour at 37'C. 45 After flicking out the u nadsorbed BSA, culture supernatants (50111) from each well of the fusion plates were replica-plated into the corresponding wells of antigen plates and incubated 30 minutes at37'C. The unbound antibody was flicked out of the wells and the plates washed three times with 1 001fl of 1% (w/v) BSA-PBS per well. Next.501LI perwell of appropriately diluted biotinylated goat anti- mouse IgG (Tago, Inc., Burlingame, 14 CA) was added to each well- and incubated for 30 minutes at 370 C. The plates were washed three times as 50 described above and then 50KI of a preformed avidin:biotinylated horseradish peroxidase complex (Vectas tain ABC Kit, Vector Laboratories, Inc., Burlingame, CA) prepared according to manufacturer's specifications was added to the wells. After 30 minutes at room temperature, the Vectastain reagentwas flicked out of the wells, the wells were washed as above, and then 1 001il per well of substrate, o-phenylenediamine (0.8 mg/ml in 0. 1 M citrate buffer, pH 5.0, mixed with an equal volume of 0.03% [v/v] H202) was added. Substrate was 55 incubated 30 minutes at room temperature in the dark, the reactions were then terminated by the addition of 500/well of 3N H2SO4 Hybridoma cells secreting monoclonal antibodiesthat bound to eitherof thetwo antigen preparations were located by measuring the absorbance at490nm of the colorimetric reactions in each well on a Dynatech Model 580 MicroELISA reader (Alexandria, VA). The cells in one well, designated Pa3 IVC2, produced anti- 60 bodythat bound to the Fisher immunotype 2 antigen plate only. This well was studied further as described below. The monoclonal antibody and clonal cell line from this well are both identified by the Pa3 IVC2 design ation in the following text. Pa3 IVC2 cells from the master wel I were mini-cloned and cloned by limiting dilution techniques as described by Tam et al. (1982, Infect. Immun, 36:1042-1053).

65 7 GB 2 192 185 A 7 Ascites fluid containing high titred monoclonal antibodywas prepared in CB6 F, mice (BALB/c [female] x C57BL/6 [male] FI) according to procedures described byTam etal. (1982, 1nfect Immun36:1042-1053).

Two- to three-month-old male C136 F, micewere injected i ntra peritonea I ly with 0.5 ml Pristane (2,6, 10, 14-Tetra m ethyl pentadecan e, Aldrich Chemical Co., Milwaukee,WI) 10-21 days priorto intraperitoneal injec tion with log phase Pa3 IVC2 cells RPMI. Each mousewas injected with 0.5- 1 X 107 cells in 0.5 mI.After 5 approximately two weeks, the fluid that accumulated was removedfrom the mice everytwoto three days.

The concentration of antibody inthe ascitesfluid was determined byagarose gel electrophoresis (Paragon, Beckman Instruments, Inc., Brea, CA) and all ascites that contained 5 mg/mI or greater antibody was pooled, aliquoted, and frozen at -70'C.

10 Characterization of the moleculartargetboundby the monoclonalantibody Culture supernatant from the cloned Pa3 IVC2 cell linewasassayed by ELISA as described above on outer membrane preparations from all seven A aeruginosa Fisher immunotype strains (A.T.C.C. 27312-27318),P.

aureofaciens (A.T.C.C. 13985), and Klebsiellapneumoniae (A.T.C.C. 8047), all prepared as described above.

Antibody Pa3 IVC2 bound to the outer membrane preparations of A aeruginosa Fisher immunotypes 2,6, 15 and 7, and not to other Fisher immunotypes, A aurefaciens, or K. pneumoniae.

The specific antigen identified by the antibody Pa3 IVC2 was identified by radioimmune precipitation.

Briefly, this analysis entails incubating radio-labeled antigens with Pa3 IVC2 antibody and a particulate source of protein A which results in the formation of insoluble antibody: antigen complexes. These com plexes are washed to remove any nonspecifically bound antigen and then the complexes are dissociated and 20 separated in a polyacrylamide gel. The predominant radioactive species found in the gel are thereby identi fied as the corresponding antigen(s) to which the antibody Pa3 IVC2 binds.

Aliquots (25 Lg) of soluble A aeruginosa Fisher immunotypes 2,3,4, and 5 outer membrane preparations were radiolabeled in solid phase with 1251 using lodo-gen (Pierce Chemical Co., Rockford, IL) (Fraker and Speck, 1978, Biochem, Biophys, Res. Commun., 80:849-857; Markwell and Fox, 1978, Biochemistry, 17:4807- 25 4817). This procedure resulted in the iodination of exposed tyrosine residues of most if not all proteins contained in the outer membrane preparations.

To diminish nonspecific binding ofthe outer membrane antigens to antibody Pa3 IVC2, the radio-labeled preparations (5 X 106 counts per minute per assay) werefirst incubated forone hourat4'Cwith BALB/c normal mouse serum (1:40 final dilution). Pa3 IVC2 culture supernatant (0. 5 ml) containing the Pa3 IVC2 30 antibodywasthen added to each outer membrane sample. After incubation of antigen and antibodyforone hourat40C,the protein Asource, IgGSORB (0.095 ml persample) (The Enzyme Center, Inc., Boston, MA)was added and incubated foran additional 20 minutes at4'C (Kessler, S.W., 1975,J. Immunol., 115:1617-1622).

IgGSORB was prepared according to manufacturer's specifications, and just priorto use nonspecific reac tions were prevented by blocking potentially reactive sites with culture media bywashing the IgGSORBtwice 35 with RPMI-hybrid (RPMI-hybrid-HAT media excluding HAT).

The antigen-antibody-IgGSORB complexes were pelleted at 1500 x g forten minutes at4'C, washed twice with phosphate-RIPA buffer (10 mM phosphate, pH 7.2,0.15 M NaCl, 1.0% [v/vl Triton X-1 00, 1.0% [w/vl sodium deoxycholate, 0.1 % [w/vl sodium dodecyl sulfate [SDSJ, and 1.0% [v/vl aprotinin); twicewith high saltbuffer(O.IM Tris-HCI, pH 8.0,0.5 M LiCl, 1 % [v/vl beta- mercaptoethanol); and once with lysis buffer (0.02 40 M Tris-HCI, pH 7.5,0.05 M NaCl, 0.05% [v/vl Nonidet P-40) (Rohrschneider et al., 1979, Proc. Nati. Acad. Scl., U.S.A., 76:4479-4483. Antigen bound to the complexwas released by incubation with sample buffer (0.125 M Tris-HCI, pH 6.8 2% [w/v] SDS, 2% [v/v] beta-mercaptoethanol, and 20% [v/v] glycerol) at 95o C forten minutes and collected in the supernatant after centrifugation at 1500 x g for 10 minutes.

The supernatant samples were then applied to 14% polyacrylamide gels containing SDS prepared accord- 45 ing to the method of B. Lugtenberg et al. (1978, FEBS Lett., 58:254-258) as modified by Hancock and Carey (1 979,J. Bacteriol., 140:902-910, which is incorporated herein by reference), and the antigens were separated in the gel by electrophoresis overnight at 50V constant voltage. Following fixation of the gel in 40% (v/v) methanol, 10% (v/v) acetic acid, and 5% (v/v) glycerol overnight, itwas dried onto Whatman 3 MM papervia a Biorad gel dryer (Richmond, CA). The dried gel was covered with plastic wrap and exposed to KodakX-AR 50 film for 18 hours at room temperature.

Results of this experiment illustrated that Pa3 IVC2 bound to only one antigen in the outer membrane preparation of A aeruginosa Fisher immunotype 2 only and notto any antigen present in the otherouter membrane preparations. The molecular weight (MW) of the antigen in the gel was about 53,000 daltons as determined by comparing its mobility to that of 14C_Iabeled protein standards (phosphorylase B, 92,500 MW; 55 BSA, 69,000 MW; ovalburnin, 46,000 MW; carbonic anhydrase, 30,000 MW; cytochrome C, 12,000 MW) (New England Nuclear, Boston, MA) that were separated in the same gel. The molecularweight of this antigen correlated with the molecularweight of flagellin, the protein comprising the flagella of A aeruginosa, as reported by Montie et al. (1982, Infect. Immun., 35:281-288), which is incorporated herein by reference.

In addition, Pa3 IVC2 was examined by ELISA using A aeruginosa Habs strains 1-12 (A.T.C.C. 33348-33359) 60 thatwere fixed with ethanol to 96-well microtiter plates. The antigen plates were prepared asfollow, s.

Overnight broth cultures of each organism were pelleted, washed twice with PBS and then resuspended in PBS to an A660 of 0.2 O.D. units. The diluted bacteria were plated into wells (50W perwell) and then centrifu ged at 1500 x g for 15 minutes at room temperature. The PBS was aspirated and then ethanol (95%) was added to the wells for 15 minutes at room temperature. Afterthe ethanol was flicked out of the wells, the 65 8 G13 2 192 004 A 8 plates were air-dried and then covered and stored at4'C until use.

Results ofthe ELISAtests, performed asdescribed above, showedthatPa3 IVC2 bound to ethanol-fixed Habs strains 2,3,4,5,7, 10, 11, and 12. This pattern of specificity indicated that Pa3 IVC2 bound to type b flagella of A aeruginosa (Ansorg, R., 1978,Zbl. Bakt. Hyg., I. Abt. Orig. A, 242:228-238; Ansorg, R., et al., 1984, J. Clin. MicrobidL, 20:84-88, both of which are incorporated herein by reference). Based upon this assign- 5 ment of specificityto monoclonal Pa3 IVC2,the A aeruginosa reference strains, Fisher immunotype 2, Fisher immunotype 6, and Fisher immunotype 7 beartype bflagella. From the preceding experimental data it has been concluded that Pa3 IVC2 binds specificallyto P. aeruginosa flagellin type b.

In vivo protective activity of Pa3 IVC2 10 Animal experiments were conducted to determine if monoclonal antibody PA3 IVC2 would protect a mouse challenged with multiple LD50's of live A aeruginosa bacteria. The model chosen was the burned mouse model (Collins, M.S., and Roby, R.E., 1983, J. Trauma, 23:530-534, which is incorporated herein by reference). Groups ofmice were given a serious burn according to the authors'protocol and then im mediately challenged with 5-10 LD5,,'s of Fisher immunotype 7. Monoclonal antibody was administered intra- 15 peritoneally as high titred ascites (0.2 ml intraperitoneally) priorto burn and challenge. No increase in number of survivors was observed in Pa3 IVC2treated animals compared to those that did not receive anti body.

ExampI62 20 Example 2 demonstratesthe methodology for the preparation of a murine hybridoma cell line producing a murine monoclonal antibodyto A aeruginosaflagellin type bthat is protective in vivo.

Adultfemale BALB/c micewerefirst injected intraperitoneal ly with viable A aeruginosa Fisher im munotype 6 (ATCC No. 27317) (8 x 106 organisms) followed two weeks laterwith an injection of viableP.

aeruginosa Fisher immunotype 5 (ATCC No. 27316) (4 X 106 organisms). During the subsequent two-week 25 period, viable A aeruginosa Fisher immunotype 5 and Fisher immunotype 6 were administered together in two weekly injections. The dosage of each organism was increased such that the final dosage was ten-fold greater than the initial dosage. Afinal injection of A aeruginosa Fisher immunotype 6 outer membrane preparations (50 VLg protein) prepared according to the method of R.E.W. Hancock and H. Nikaido (1 978,J.

BacterioL, 136:381-390) was given four days afterthe last viable bacteria injection. Three days after the last 30 immunization, the spleen was removed from one mouse and the spleen cells prepared by hybridization as described in Example 1.

Culture supernatants of the hybridoma cellswere assayedforthe presence of anti-P. aeruginosa anti bodies of ELISA on day 10 post-fusion according to the procedures stated in Example 1, exceptthatthe antigen forthe ELISA plateswasviable bacteria immobilized in thewells of the 96-well microtiter plates.The 35 plateswere prepared asfollows.

Fifty microliters poly-L-lysine (PLL) (1 jig/ml in PBS) (Sigma #P-1 524, St. Louis, MO) were added to each well of 96-well plates (Linbro) and incubated for30 minutes at room temperature. Unadsorbed PLLwas flicked out andthe wellswerewashed threetimeswith PBS. Bacterial cultures grown overnight in TSBwere washed oncewith PBS andthen resuspended in PBSto O.D-660nm = 0.2. Fifty microliters of the bacterial 40 suspensionswere added to each well of the plate and allowed to bind at37'Cforone hour. Unbound bacteria were removed byflicking the plates and then washing the wells three times with saline-Tween (0.9% [w/vJ NaCl, 0.05% [v/vl Tween-20).

Nonspecific binding of the antibodies was blocked by the addition of 200pl/well of blocking buffer (PBS containing 5% [w/vl non-fqt dry mil k, 0.01 % [v/vl Antifoam A [Sigma, St. Louis, MO], and 0.01 % [w/vJ thimer45 osal) to the wells and incubation for one hou r at room temperature. Excess blocking buffer was expelled and the wells were washed th ree times with saline-Tween as previously described.

Culture supernatants (50111) were replicaplated into the corresponding wells of the assay plates and incuba ted at room temperature for 30 minutes. The culture supernatants were removed by flicking the plates and washing the wells five times with saline-Tween. 50 An enzyme-conju gated second step antibody (horseradish peroxidase-conj u gated goat anti-mouse IgG + IgM) (Tago, Inc., Burlingame, CA) was diluted in PBS containing 0.1 % (v/v) Tween-20 and 0.2% (w/v) BSA according to the previously determined titrations, and then 50pl of the reagent was added to each well and incubated 3Q minutes at room temperature. The excess reagent was expelled; the wells washed five times in saline-Tween; and 1 00pi/well of o-phenylenediami ne substrate was added and incubated for 30 minutes as 55 described in Example 1. Reactions were terminated as stated in Example land then read at A490 nm on a Bio-Tek EL-31 0 Automated EIA Plate Reader.

By the above-described methods, the culture supernatants from the fusion were assayed for the presence of antibodies that bound to A aeruginosa Fisher immunotypes 1, 2,3, or 4, but not to control plates prepared by the same PLL and blocking procedure, but without bacteria. Supernatants containing antibody that bound 60 to any of these four Fisherimmunotypes were assayed a second time using each of the seven Fisher im munotype bacteria separately. Antibody present in the supernatant from one well, PaF4 IVE8, bound only to A aeruginosa Fisher immunotypes 2,6, and 7. Cells from well PaF4 IVE8 were cloned by limiting dilution methods as described in Example 1. The monoclonal antibody and the clonal cell line from this well are both identified bythe PaF4 IVE8 designation in the following text. High titred monoclonal antibody-containing 65 9 GB 2 192 185 A 9 ascites fluid was generated as described in Example 1, except that BALB/c mice were u sed instead of CB6F1.

Specificity of PAF41VE8 One assay performed to identify the antigen bound by monoclonal antibody PaF4 IVES was indirect im munofluorescence on bacterial organisms. Each of the seven reference Fisher immunotypes of A aerugi- 5 nosa, plus a nonflagellated strain of A aeruginosa (PA103,A.T.C.C. 29260, Leifson, 1951,J. BacterioL,62:377 389), and Escherichia co/i(G.S.C. A25) were grown overnight at37'C in TSB. The bacteria were pelleted by centrifugation and then washed twice in PBS. Each strain was resuspended in PBS to an O.D-660nm = 2.2.

The bacterial suspensions were then further diluted 1: 150 and 20[d samples were placed in individual wells of Carlson slides (Carlson Scientific Inc., Peotone, IL) and dried onto the slide at40'C. Culture supernatants 10 (25d) of PaF4 IVES were incubated on the dried bacterial samples on the slides in a humidified chamberat room temperature for 30 minutes. Unbound antibody was washed off the slides by dipping theslides in distilledwater.

Afterthe slides dried, fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse IgG + IgM (25 [LI per well of a 1:40 dilution PBS) (Tago, Burlingame, CA) was incubated on the slides forthirty minutes at room 15 temperature in a humidified chamber in the dark. The slides were again washed in distilled water, dried, and then covered with a coverslip mounted with glycerol in PBS (9: 1). Slides were viewed with afluorescence microscope.

Fluorescent staining was observed only on A aeruginosa Fisher immunotypes 2,6, and 7, and was obse rved to be a sinusoidal pattern (line) emanating from one end only of the organisms. This is consistentwith 20 the morphology and location of the single polar flagel I um of these bacteria.

The reaction of PaF4 IVES with flagella was confirmed by immunoblot analysis. Outer membrane antigens from A aeruginosa Fisher immunotype 6 (see Example 1) were separated by electrophoresis in a 14% poly acrylamide gel containing SIDS as described in Example 1, except that electrophoresis was run for five hours at 80 mAmps constant amperage. Prestained molecular weight markers (lysozyme, 14,300 MW; beta- 25 lactoglobulin, 18,400 MW; al pha-chymotrypsinogen, 25,700 MW; oval burnin, 43,000 MW; bovine serum albumin, 68,000 MW; phosphorylase B, 97,400 MW; and myosin, 200,000 MW) (BRL, Gaithersburg, MD) were included in the same polyacrylamide gel.

Antigens were transferred from the polyacrylamide gel to a nitrocellulose membrane, (NCM), (0.45 um, Schleicher& Schuell, Inc., Keen, NH) in a Trisglycine-methanol buffer (Towbin etaL [19791, Proc. Nati. Acad. 30 Sci., U.S.A., 76:4350-4354) containing 0.05% (w/v) SDS overnight at 4C at a constant amperage of 200 mA.

Aftertransfer, the NCM was incubated in 0.05% (v/v) Tween-20 in PBS (PBSTween) (Batteiger, B., et al., 1982, J. Immuno/. Meth., 55:297-307) for one hour at room temperature. Forthis step and all subsequent steps,the tray containing the NCM was placed on a rocking platform to ensure distribution of solution overthe entire NCM. 35 After one hour, the PBS-Tween solution was poured off and PaF4 IVES ascites (diluted 1: 1000 in PBS Tween) was added and incubated with the NCM for one hour at room temperature. The NCM wasthen washed fivetimes, five minutes each, with PBS-Tween, to remove unbound antibody. Alkaline phosphatase conjugated goat anti-mouse IgG +IgM (Tago, Inc.) was diluted according to manufacturer's specifications and incubated with the NCM for one hour at room temperature. The NCM was washed five times as described 40 above, and the substrate containing bromochloroindolyl phosphate and nitrobluetetrazolium (Sigma, St.

Louis, MO), prepared as described by Leary et al. (1983, Proc. Natl. Acad. Sci., U.S.A., 80:4045-4049), was added and incubated 10-20 minutes at room temperature. The reaction was terminated by washing substrate away with distilled water.

The results of this experiment showed that PaF4 IVES bound specificallyto a single antigen with a molec- 45 ular weight of 53,000 daltons in the outer membrane preparation. The results of the indirect immunofl uoresc ence assay and immunoblotting demonstratethat PaF4 IVES binds to the flagel " la of A aeruginosa.

The flagella type that PaF4 IVES recognized was determined by ELISA. Habs strains 1-12 (A.T.C.C. #33348 33359) were each bound to wells of 96-well micro-titer plates (Linbro) with PLL, and the ELISA performed as described earlier in this Example. The source of PaF4 IVE8antibody was culture supernatant. Positive reac- 50 tions were noted in wells containing Habs strains 2,3,4,5,7, 10, 11, and 12, thus indicating that PaF41VE8 binds to type b flagella. The in vivo protection data are presented below in Example 4.

Example3

Example 3 demonstratesthe methodology for the preparation of a murine hybridoma cell line producing a 55 monoclonal antibody, reactive with anti-P. aeruginosa flagella type a, that is protective in vivo.

The lymphoid cell source forthe fusion was a spleen from an immunized BALB/c mouse that had been injected fourtimes intraperitoneally over a six-week period with purified type a flagella (10-20 [tg protein) from Habs strains 6 and 8 (A.T.C.C. #33353 and 33355). Flagella were purified according to the method T.C.

Montie etaL (1982, Infect. Immun., 35:281-288, which is incorporated herein by reference) with the exception 60 thatthefinal centrifugation of flagella was at 100,000 x g for one hour ratherthan 40,000 X g forthree hours.

A second modification adopted for some procedures was to shear the flagella from the bacteria 30 seconds in a blender rather than three minutes. (Allison et al., 1985, Infect. Immun. , 49:770-774).

Protein concentrations oi each preparation were determined with the BioRad Protein Assay (Bio-Rad, Richmond, CA) and the presence of contaminating lipopolysaccharide (LPS) was assessed by measuring the 65 GB 2 192 185 A 10 KDO content (Karkhanis,Y.D., etal., 1978,Anat Biochem., 85:595-601). The molecular weights of the flagella proteins were determined by comparing their migration in an SDS polyacrylamide gel with the migrationof standareprotein markers (BRL) (see Example 2). The molecular weight of Habs6flagellin was 51,700daltons andthatof Habs 8 flagellin was 47,200 daltons. These values agree with those obtained byJ.S.Allison etal.

(1985,lnfect. Immun., 49:770-774, which is incorporated herein byreference). 5 Fusion of splenocytes from flagellin-immunized mice andNS-1 myeloma cells was performed three days afterthe last immunization, as described in Examples 1 and 2. When hybridoma cells grewto approximately 40% confluency (day 7), culture supernatants were replica-plated into corresponding wells of three different antigen plates, PLL-bound A aeruginosa Fisher immunotype 1 (see Example 2 for preparation), and formalin fixed Habs 6 and Habs 9. 10 The bacteria forthe formalin-fixed antigen plates were grown, washed, and diluted as described for PLL bound antigen plates. Diluted bacteria (0.2 O.D. units atA660) were added to individual wells (50ld perwell) of Linbro 96-well micro-titer plates and the plates were then centrifuged at 1200 X g for 20 minutes at room temperature. The supernatants were flicked out of the wells and 751.d of 0.2% (v/v) formalin in PBS was added to each well and incubated for 15 minutes at room temperature. After the formalin was flicked out of the 15 wells, the plates were air-dried and stored at 4C until used. Formalin did not alter antigen icity of the flagella as shown bythe ability of anti-flagella antisera to agglutinate formalin- treated organisms (lanyi, B., 1970, Acta Microbiol. Acad. Sci., Hung., 17:35-48). A aeruginosa Fisher immunotype 1 strain was included asa control because this strain was nonflagellated, as shown by mordant dye staining (Manualof Clin. Micro biol., 1985, Lennette, ed. Amer. Soc. Microbiol., Wash., D.C., p. 1099). Hybrid cells in the well designated FA6 20 IIG5 produced an antibody that bound to Habs 6 and Habs 9 (both flagella type a bearing strains) but not Fisher immunotype 1.

Cells from well FA61IG5 were subcultured and cloned as described in previous examples. The monoclonal antibody and the clonal cell line from this well are both identified by the FA61IG5desig nation in the following text. Ascites was produced in BALB/c mice as described in Example 1. 25 Specificity of FA611G5 The specificity of the antibody FA6 IIG5 was determined by indirect imunofluorescence and im munoblotting. Indirect immunofluorescence was performed essentially as described in Example 2, with the following modifications. 30 Bacteria cultures grown overnight on trypticase soy agarat30'Cwere removed from the plateswith cotton swabs and resuspended in PBS to an A660 of 0.2 O.D. units. Formalin (0. 37% [v/v] in PBS final concentration) was addedto the suspension with vortexing. Afteran incubation at room temperaturefor 15 minutes,the bacteria were diluted 1:12 in PBS and 20KI of this suspension was placed in individual wells of Carlson slides.

Afterdrying,the slides were prepared forviewing as described in Example 2. The source of antibodywas 35 culture supernatantfrom the FA6 IIG5 cell line.

Fluorescent staining by the FA6 IIG5 antibody was observed only with A aeruginosa strains bearing type a flagella and none of those bearing type b. Thefluorescence pattern observed was a sinusoidal line pattern indicating thatFA6 IIG5 bound to theflagella. The fluorescent signal was enhanced bytreating the bacteria with formalinr butthe treatment was not required to visualize flagellar staining with the antibody. 40 Immunoblotting was performed as described in Example 2. The sources of flagella type a antigens were the purified flagellar preparations (see this Example). Antigens were separated in 10% polyacrylamide gels containing SDS (Laemmli, U.K., 1970, Nature [London], 227:680-685) and transferred to an NCM. Pre parations of FA6 IIG5, either culture supernatant or ascites diluted 1: 1000, were reacted with the NCM, and the reaction detected with an appropriate enzyme-conjugated reagent and enzyme substrate as described in 45 Example 2. The immunoblot illustated that FA6 IIG5 bound specificallyto the 51,700 MWflagellin of Habs6 andthe47,20OMWflagellinofHabs8.

Confirmation that FA6 IIG5 reacted with only flagella type a and not type b was obtained by ELISA, in which Habs strains 1-12 were bound individually with PLL to the wells of Lin bro 96-wel I microtiter plates. The antibody bound to only Habs strains 1, 6,8, and 9, which are the only flagella type a bearing strains of the 50 twelve (see, Ansorg, R., et al., 1984,J. On. Microbiol., 20:84-88, which is incorporated herein by reference). In vivo protection studies are presented in the following Example 4. 47 Example 4

Example 4 demonstrates protection of mice passively immunized with antibodies PaF4 IVE8 and FA6 I IG5 55 against challenge with A aeruginosa in the bu rned mouse model.

The anti-flagella monoclonal antibodies were tested in the burned mouse model according to the method of M.S. Collins and R.E. Roby (1983,J. Trauma, 23:530-534, which is incorporated herein by reference). For the protection studies, all antibodies were purified by protein A- Sepharose chromatography (Ey, P.L., et al., 1978,1mmunochemistry, 15:429-436, which is incorporated herein by reference) and dialyzed into P13S buf- 60 fer. The flagella type a bearing strain used in the animal trials was A aeruginosa PA220 (from Dr. James Pennington, Boston, MA) and the flagella type b strain was the reference A aeruginosa Fisher immunotype 2 (A.T.C.C. 0 27313).

Forty micrograms of purified monoclonal antibody was given per mouse intravenously one to two hours priorto burn and challenge. Immediately afterthe burn, the animals received 0.5 ml cold PBS subeschar 65 11 GB 2 192 185 A 11 containing the challenge bacteria. The challenge dose was approximately 10 LD50'sforeach organi sm. The results of the animal trials are presented in Tables 1 and 11.

Table 1. Protection Study of an Anti-Flagella Type a Monoclonal Antibody in the Burned Mouse Model' 5 Percent SurvivalbyDay2 Treatment 1 2 3 4 5 6 7 8 9 Anti-Flagellaa, 10 FA6 11G5 100 100 100 100 100 90 90 80 80 Anti-Flagella b, PaF4WE8 100 20 10 10 10 10 10 10 10 15 Non-specific anti LPS monocional antibody 100 40 30 20 10 10 10 10 10 PBS, no bacteria 80 80 80 80 80 80 80 80 80 20 1 Micewere challenged subescharwith approximately 10 I-Dr,()'s of P. aeruginosa PA220.

2 The percent is based upon survival of mice in ten animal groups with the exception of the PBS onlycontrol group which consisted of five mice. Days are post-burn and challenge. 25 Table 11. Protection Study of Anti-Flagella Type b Monoclonal Antibody in the Burned Mouse Model' Percent Survivalby Day2 30 Treatment 1 2 3 4 5 6 7 8 9 Anti-Flagelia b, PaF4WE8 100 100 90 90 90 90 90 90 90 35 Anti-Flagella a, FA6 HG5 100 20 10 10 10 10 10 10 10 Non-specific anti 1-PSMonoclonal 40 antibody 100 20 20 20 20 20 20 20 20 PBS, no bacteria 100 100 100 100 100 100 100 100 100 45 1 Micewere challenged subescharwith approximately 10 LD50's of P. aeruginosa Fisher immunotype 2.

2 Percentsurvival was based on numberof mice surviving pergroup of ten with the exception of the PBS onlycontrol group,which consisted of five mice. Days are post-burn and challenge.

Very significant survival was observed in mice that were treated with the anti-a antibody oranti-b antibody 50 and then challenged with the corresponding antigen. Conversely, 80-90% of the untreated butchallenged mice oranimals treated with a nonmatching antiflagellar monoclonal antibody or nonspecific anti-LPS anti body died. The inability of the anti-flagella type a antibodyto protect micefrom a lethal challenge offlagella type b bearing P. aeruginosa Fisher immunotype 2, and the inability of the anti-flagelia type b antibodyto protect mice from a lethal challenge of type a bearing P. aeruginosa PA220, corroborated in vivo the specifi- 55 city of the antibodies observed in vitro. Survival of mice burned, but not infected, indicated that the burn itself wasnotlethai.

Example 5

Example 5 demonstrates the extensive cross-reactivity of PaF4 IVE8 and FA6 11G5 with P. aeruginosq clini- 60 cat isolates, indicating the clinical utility of these antibodies in immunotherapy of P. aeruginosa infections.

Clinical isolates were obtained,,from hospitals and clinics. The isolates were from a variety of isolation sites, including blood, wounds, respiratory tract, urine, and ears. A total of 157 isolates were examined.

PaF4 IVE8 bou nd specifically to 34 clinical isolates (22%), while the flagel la type a antibody, FA6 11G5, bound to 102 clinical isolates (65%), fora total of 136 of 157 isolates (87%). Of the 21 strains that were not recognized 65 12 G13 2 192 185 A 12 by either antibody, 19 were non-flagellated as shown by mordant dye staining. Therefore, both antibodies in combination bound to 136 of 138(98%) of flagellated clinical isolates, confirming prior reports (see, R.Ansorg, 1978, Zbl. Bakt. Hyg., I Abt. Orig. A, 242:228-238, which is incorporated herein by reference).

Example 6 5

Example 6 demonstrates methods forthe production of human monoclonal antibodies that bind to A aeruginosa type bfiagelia.

A peripheral blood sample from an individual immunized with a high molecularweight polysaccharide preparation (Pier et al., 1984, Infect. Immun., 45:309) served as a source of B cells. Mononuclear cellswere separated from the blood by standard centrifugation techniques on FicollPaque (Boyum (1968) Scand. J. 10 Clin. Lab. Invest., 21:77) and washed twice in calcium/magnesium-free phosphate buffered saline (PBS).

The mononuclear cells were depleted of T-cells using a modified Erosetting procedure. Briefly, the cells were first resuspended to a concentration of 1 x 107 cells/ml in PBS containing 20% fetal calf serum (FCS) at 4'C. One ml of this suspension was then placed in a 17 X 100 mm polystyrene round-bottom tube to which wasaddedl x 109 2-amino-isothiouronium bromide (AET)-treated sheep red blood cells from al 0% (v/v) 15 solution in Iscove's modified Dulbecco's medium (Iscove's medium) (Madsen and Johnson (1979)J, Immun.

Methods, 27:61). The suspension was very gently mixed for 5-10 minutes at 4C and the E-rosetted cel Is then removed by centrifugation on Ficoll-Paque for 8 minutes at 2500 x gat 40C. E-rosette negative peripheral blood mononuclearcells (E-PBMQ banding atthe interface were collected and washed once in Iscove's medium and resuspended in same containing 15% (v/v) FCS, L-glutamine (2 mmol/1), penicillin (100 IU/ml), 20 streptomycin (100 jig/ml), hypoxanthine (1 X 10-4 M), aminopterin (4 x 1 O'M) and thymidine 0.6 X 10-5M).

This medium is hereafter referred to as HAT-medium.

Cell-driven transformation of the E-PBMC was accomplished by cocultivating these cells with a transform ing cell line. The transforming cell line was an Epstein-Barr antigen (EBNA) positive human lymphoblastoid cell line derived by ethyl methane-sulphonate (EMS) mutagenesis of the GM 1500 lymphoblastoid cell line 25 followed by selection in the presence of 30 jLgIml 6-thioguanine to render the cells hypoxanthine-guanine phosphoribosyl transferase (HGPRT) deficient and thus HATsensitive. This cell line is denominated the 1 A2 cell line and was deposited atthe American Type Culture Collection (A.T.C. C.) on March 29,1982, under A.T.C.C. No. CRL 8119. lA2 cells in logarithmic growth phase were suspended in HAT- medium and then combined with the E-PBMC's at a ratio of fifteen lA2 cells per PBMC. The cell mixture was plated intothirty 30 round-bottom 96-well microtiter plates (Costar3799) at a concentration of 32,000 cells/well in a volume of 200 jil perwell, and incubated at370C in a humidified atmosphere containing 6% C02. Cultures were fed on days 5 and 8 post-plating by replacement of half the supernatantwith fresh HATmedium. Sixteen days after plat ing, 100% of the wells contained proliferating cells and in most of the wells the cells were of sufficient density for removal and testing of supernatants for anti-P. aeruginosa antibodies. 35 Supernatants were screened forthe presence of anti-P. aeruginosa antibodies using the ELISAtechnique as described in Example 2 with the following modifications. A pool of the seven Fisher immunotype refer ence strains (A.T.C.C. Nos. 27312-27318) (A660 = 0.2 O.D. units) were bound to flat-bottom 96-well microtiter plates (Immulon 11, Dynatech) pre-treated with poly-L-lysine, incubated, and washed as described in Example 2. After blocking non-specific binding sites and washing the plates, 50 RI of PBS containing 0.1 % (v/v) Tween- 40 and 0.2% (w/v) BSAwas added per well. Culture supernatants (50 pl) were then replica-plated intothe corresponding wells of assay plates and into control plates that were treated with PLL and blocked, butdid not contain bacteria. After incubation and washing, enzyme-conju gated second step antibodies (50 ml per well), horseradish peroxidase-conjugated goat anti-human IgG and goat anti-human IgM, diluted app ropriately in PBS containing 0.1 % (v/v) Tween-20 and 0.2% (w/v) BSA, were added to the wells and the assay 45 completed as described in Example 2.

Supernatants containing antibodythat bound to the pool of Fisher immunotypes, but notto the control plate, were assayed a second time using each of the seven Fisher immunotype bacteria separately. Antibody present in the supernatantfrom one well, 20H1 1, bound only to A aeruginosa Fisher immunotypes 2,6, and 7. The cells were subcultured repeatedly at decreasing low cell densities until all wells with growth were 50 secreting antibody. The cell line and the monoclonal antibody (IgM isotype) are both identified by the 20H1 1 designation in the following text.

A second transformation was performed in which the source of B cel Is was from the peripheral blood of a cystic fibrosis patient known to have had a chronic A aeruginosa infection. E-PBNC's were prepared as described above and co-cultivated with the transforming cell line, lA2, at a ratio of 72 1A2 cells per E-PBMC. 55 The cell mixture was plated into fifteen round-bottom 96-well microtiter plates at a concentration of 7.4 x 104 cells perwell and cultured as above.

Supernatants were assayed by ELISAforthe presence of anti-P. aeruginosa antibodies sixteen days after the transformation was plated. The assay was performed as described for the previous transformation, ex cept that the pool of A aeruginosa strains used forthe initial screening was composed of Fisher immunotype 60 reference strains, F2, F4, F6, and F7 (A.T.C.C. Nos. 27313,27.315,27316, and 27317), and three clinical isolates from the Genetic Systems Corporation Organism Bank (GSCOB) that had different LPS immunotypps and Flagella types. The clinical isolate PSA 1277 (GSCOB) bears type a flagella and Fisher immunotype 1 LPS; the second isolate PSA G98 (GSCOB) bears type a flagella and Fisher immunotype 3 LPS; and the third, PSA F625 (GSCOB)l bears type b flagella and Fisher immunotype 5 LPS. This mixture of reference strains and clinical 65 13 GB 2 192 185 A 13 isolateswill be referred to astheP. aeruginosa flagellated pool. Supernatants containing antibodythat boundto plates containingtheP. aeruginosa flagellated pool, butnottothe PILL-coated control plates,were assayed byELISAa secondtimeon the individual strains of the pool. Onewell, 3Cl, boundto reference strains F2, F6and F7 and to the clinical isolate F625.

Cloning of the3Cl cell linewas accomplished byfirst subculturing the cells intwo rounds of lowdensity 5 subculture, first at 20 cells perwell of 96-well platesfollowed byculturing at2 cells perwell. Formal cloning of the specific antibody-producing cellswas performed by plating the cellsata densityof about 1 cell/well in 72-well Terasaki plates (Nunc 01-36538) in avolume of 10 I.Ll/well of HAT- medium lacking intheaminopterin component (HT-medium). The plateswere placed in an incubatorfor2-3 hoursto allowthe cellsto settleto the bottom of the wel Is and were then microscopically scored by two individuals for wells that contained a 10 single cell. The wells were fed daily with HT-medium and when outgrowth was sufficient, the cells were transferred to a 96-well round-bottom plate. All wells with outgrowth were assayed by ELISA on A aerugi- la nosa strains bearing type b flagella, and all were found to be producing the appropriate antibody. The cell line and the monoclonal antibody (IgM isotype) are both identified by the 3Cl designation in the following text.

The antigen identified by20Hl 1 and 3Cl was flagella as shown by indirect immunofluorescence and im- 15 mu noblotting. The techniques were performed basically as described in Examples 2 and 3. Forthe indirect immunofluorescence assayP. aeruginosa strains bearing.type b flagella, reference Fisher immunotypes F2, F6, and F7 (A.T.C.C. Nos. 27313,27317, and 27318) and a strain bearing type a flagella, reference Fisher immunotype 4 (A.T.C.C. No. 27315), were prepared as described in Example 3. The flagella type of the refer- ence strains was determined by typing with the murine monoclonal antibodies, PaF4 IVE8 and FA6 IIG5. The 20 slides were prepared for viewing as described in Example 2. The sources of both antibodies were culture supernatants, and the FITC-conjugated reagent was FITC-conjugated goat anti-human Ig (polyvalent) (Tago, Burlingame, CA) diluted 1: 100 in PBS containing 0.5% (w/v) bovine gamma globulins (Miles Scientific, Cat.

No. 82-041-2, Naperville, IL) and 0.1% (w/v) sodium azide as a preservative.

Fluorescent staining by the 20H 11 and3Cl antibodies was observed only with P. aeruginosa strains bear- 25 ing type b flagella and notwith the flagella type a bearing strain, reference Fisher immunotype 4. Thefluores cence pattern observed was a sinusoidal line pattern emanating from one end of the bacteria indicating that the antibodies bound to the flagella of the bacteria.

Immunoblotting was performed as described in Example 2. Purified type b flagella from the A aeruginosa reference strains Fisher immunotype 2 (A.T.C.C. No. 27313), and purified flagella type a from reference 30 strains Habs 6 and Habs 8 (A.T.C.C. Nos. 33353 and 33355) were prepared as described in Example 3. Anti gens were separated in a 10% polyacrylamide gel (see Example 3) and transferred to an NCM. Culture super natants containing 20H1 I or3Cl antibodies, culture supernatant containing a non-specific human antibody, and culture media were incubated with the NCM and the reaction detected with an alkaline phosphatase conjugated goat anti-human Ig (polyvalent) (Tago, Burlingame, CA) diluted in PBS containing 0.05% Wv) 35 Tween-20. Enzyme substrate was prepared as described in Example 2. The immunoblot illustrated that both antibodies bound to the 53,000 MWflagellin protein of Fisher immunotype 2, and not to the 51,700 MW flagellin protein of Habs 6 nor to the 47,200 MW flagellin protein of Habs 8. No reaction was observed with eitherthe non-specific human antibody or the culture media.

Additional confirmation that antibodies 201-11 land 3Cl bound onlyto type b flagella and notto type a was 40 obtained by ELISA, in which Habs strains 1-12 were bound individually with PLL to the wel Is of Im m ulon 96-well microtiter plates. The antibodies bound only to Habs strains 2,3, 4,5,7, 10, 11, and 12, which aretype b bearing strains (Ansorg eta[. (1984)J. Clin. Microbio/., 20:84).

Example 7 45

Example 7 demonstrates methodsforthe production of a human monoclonal antibodythat bindstop.

aeruginosaty.pe a flagella.

A peripheral blood samplefrom an individual immunized with a high molecularweight polysaccharicle preparation (Pier etal. (1981) Infect Immun., 34:461) served as a source of B cells. The mononuclear cells were separated from the blood and then depleted of T-cells as described in Example 6. The cellswerethen 50 frozen in FCS containing 10% climethyl sulfoxide in a liquid nitrogen vaportank. At a laterdata the cellswere thawed quicklyat370C, washed once in Iscove's medium and resuspended in HAT-medium. Cell-driven transformation was accomplished by co-cultivating the E-PBMC's with lA2 cells ata ratio of 30 1A2 cells per E-PBMC. The cell mixturewas plated into 30 96-well tissue culture plates at a concentration of 62,000 cells per well. Cultures were fed the seventh day after plating by replacement of half the volume with HAT-medium. 55 Cell proliferation was observed in 100% of the wells on the fourteenth day post-plating, and supernatants were removed from the wells and assayed at this time.

Supernatants were assayed by ELISA for the presence of anti-P. aeruginosa antibodies by using the flag ellated A aeruginosa pool and I'LL-treated plates as a control, as described in Example 6. Supernatants containing antibodies that bound to the flagellated pool, but notto the PLL control plates, were assayed again 60 on the individual bacteria strains of the flagellated pool. One well, 21 B8, contained antibodythai bound to PSA 1277, PSA G98, and reference Fisher immunotype 4, which arethe three strains of the flagellated pool that beartype a flagella.

Cloning of the 21 B8 cell line was accomplished in Example 6 forthe 3Cl cell line with the following modi fications atthe formal cloning step. Afterthe wells of the Terasaki plates were scored for the presence of only 65 14 GB 2 192 185 A 14 a singlecell,each cell was transferred from the Terasaki platetoan individualwell of a96-well round-bottom culture plateinavolumeof 100 jil HAT-medium lacking the aminopterin component (HT-medium). Non transforming, HAT-sensitive lymphoblastoid cells were included in allwellsatadensityof 500cells/wellas feedercells. Five days post-plating, 10ORI of HAT-medium was added to the wells to selectively kill thefeeder cells. Wells were again fed on days7and9 post-plating by replacementof half the supernatant with HAT- 5 medium. The cells were then fed with HT-medium until the cells were of sufficient density to detectthe presenceof antibody by ELISA. All wells with outgrowth produced antibodythat bound to flagella type a bearing A aeruginosa strains. The cell line and the monoclonal antibody (IgG, isotype) are both identified by the 21 B8 designation in the following text.

The antigen identified by 21 B8 wasflagella as shown by indirect immunofluorescence and im- 10 munoblotting (see Example 6 for descriptions of the techniques). Fluorescent staining bythe 21 B8 antibody was observed only with P. aeruginosa reference strain Fisher immunotype4(A.T.C.C. No. 27315), which bears a type a flagella, and notwith A aeruginosa reference strain immunotype 2 (A.T.C.C. No. 27313)that bears type a flagella. Thefluorescence pattern observed was a sinusoidal line pattern emanating from one end of the bacteria indicating thatthe antibody bound to the flagella of the bacteria. 15 Immunoblotting was performed as described in Example 2. Purified type a flagella from the P. aeruginosa reference strain Habs 6 (A.T.C.C. No. 33353) and purified flagella type b from the P. aeruginosa reference strain Fisher immunotype 2 (A.T.C.C. No. 27313) were prepared as described in Example 3. Antigenswere separated in a 10% polyacrylamide gel (see Example 3) and transferred to an NCM. Culture supernatants containing either 21 B8 or a non-specific human antibody and culture media were reacted with the NCM and 20 the reaction detected with an alkaline phosphatase-conju gated goat anit- human Ig (polyvalent) and enzyme substrate as described in Examples 2 and 6. The immunoblot illustrated that the 21 B8 antibody bound onlyto the 51,700 MWflagellin protein of Habs 6 and notto the 53,000 MWflagellin protein of Fisher immunotype 2.

No reaction was observed with eitherthe non-specific human antibody orthe culture media.

25 Example 8

Example 8 demonstrates protection of mice passively immunized with human anti-flagella antibodies, 20H1 1, 3C1, and 21 B8, against challenge with A aeruginosa in the burned mouse model.

The human anti-flagella monoclonal antibodies weretested in the burned mouse model (see Example4).

21 B8 and 20H1 1 antibodies were prepared by precipitation of culture supernatants generated from the re- 30 spective cell lines with ammonium sulfate (50%final concentration) (Good et aL, SelectedMethods in Cel lularlmmunology, Mishell, B.B., and Shiigi, S.M., eds., W.J. Freeman & Co. , San Francisco, CA, 1980,279 286). The precipitate was solubilized in PBS, dialyzed against PBS overnight at 4%C, and then sterilefiltered priorto administration in animals. The source of antibody3C1 and the nonspecific anti-LPS antibody used as a negative control in that study was culture supernatant. As a positive control for each study, the appropriate 35 purified murine monclonal antibody, PaF4, IVE8 or FA6 IIG5, was included.

The flagella type a bearing strain used in the animal trials was the clinical isolate PSA A522 (GSCOB),which expresses Fisher immunotype 1 LPS, and theflagella type b strain was clinical isolate PSA A447 (GSCOB), which expresses Fisher immunotype 6 LPS. The human antibodies (0.45 ml) were premixed with the bacteria (greaterthan 5 ILD100's in 0.05 ml) and inoculated subeschar immediately after the burn was administered. 40 The results of the animals'trials are presented in Tables III, IV, and V.

Table Ill. Protection Study of the Human Anti-Flagella Type a, Monoclonal Antibody 21 B8 in the Burned Mouse Model.

45 Percent Survival by Day' Treatment 1 2 3 4 5 6 7 8 9 Murineanti-flagellaa, 50 FA6 IIG5' 100 100 100 100 88 88 88 88 88 Human anti-flagella a, 21 B8 100 100 100 100 100 100 100 100 100 55 Human anti-flagella b, 20H1 1 100 0 0 0 0. 0 0 0 0 PBS 100 25 12 12 12 12 12 12 12 60 1 Mice were challenged subsescharwith greaterthan 5 1_13100's of A aeruginosa PSAA522.

2 Percent survival was basej on the number of mice surviving per group of eight animals. Days are post burn and challenge.

'Purified antibody (10 pg in 0.45 ml PBS) was pre-mixed with the bacteria and administered subeschar 65 GB 2 192 185 A 15 after burn and challenge.

Table IV. Protection Study of the Human Anti-Flagella Type b Monoclonal Antibody 20H1 1 in the Burned Mouse Model' 5 Percent SurvivalbyDay2 Treatment 1 2 3 4 5 6 7 8 9 Murine anti-flagella b, PaF41VE83 100 100 100 100 100 100 100 100 100 10 Human anti-flagella b, 1 20H11 100 100 100 100 100 100 100 100 100 Human anti-flagella a, 15 21 B8 100 0 0 0 0 0 0 0 0 Media 80 0 0 0 0 0 0 0 0 'Mice were challenged subescharwith greaterthan 5 LD100's of A aeruginosa clinical isolate, PSAA447. 20 2 Percent survival was based on the number of mice surviving per group of five animals. Days are post-burn and challenge.

3 Purified antibody (10 Kg 0.45 ml PBS) was pre-mixed with the bacteria and administered subescharafter burn and challenge.

25 Table V. Protection Study of the Human Anti-Flagella Type b Monoclonal Antibody 3C1 inthe Burned Mouse Model' Percent Survivalby Day' 30 Treatment 1 2 3 4 5 6 7 8 9 Murine anti-flagella b, OaF41VE8 3 100 100 100 100 100 100 100 100 100 35 Human anti-flagella b, 3C1 100 100 80 80 80 80 80 80 80 Non-specificanti-LPS monoclonal antibody 100 0 0 0 0 0 0 0 0 40 1 Micewere challenged subescharwith greaterthan 5 1_13100's of PSAA447.

2 Percentsurvival was based on survival of mice in five animal groups. Days are post-burn and challenge.

3 Purified antibody (40 Lg) was administered intravenously two hours priorto burn and challenge.

45 Very significant survival was observed in mice that were treated with the anti-flagella type a antibody, or either of thetwo anti-flagella type b antibodies, and then challenged with the corresponding antigen. Con versely, 88%- 100% of the untreated butchallenged mice, orthose treated with a nonmatching anti-flagellar monoclonal antibody or nonspecific anti-LPS antibody, died. As was observed with the murine monoclonal antibodies (see Example 4), the human anti-flagella antibodies specifically protect against lethal challenge 50 onlywith the organisms bearing the corresponding flagella type, i.e., the human anti-flagella type a antibody provided protection against lethal challenge with the flagella type a bearing organism and notthe type b, and the anti-flagella type b antibodies protected mice that were challenged with flagella type b bearing strains, but notthe type a bearing organism.

55 Example 9

Example 9 demonstratesthe cross-reactivity of the human anti-flagella antibodies 20H1 1, 3C1, and 21 B8 with A aeruginosa clinical isolates.

A aeruginosa clinical isolates (115) obtained from hospitals and clinics and isolated primarilyfrom burn wounds and blood were identified as bearing flagel [a type a or type b by typing with the murine monoclonal 60antibodies, FA6 IIG5 or Pa F4 IVE8 (see Examples 2,3, and 5). Fifty-five of the clinical isolates were identified as bearing type a flagella by reacting with the murine monoclonal antibody, FA6 IIG5, and 59 were identified as type b bearing bytheir reaction with the murine monoclonal antibody, PaF4 IVE8.

Cross-reactivity of the anti-flagella type a human monoclonal antibody, 21 B8, was extensive in thatthe antibody recognized 54 of the 56 flagella type a bearing clinical isolates (96%). Cross-reactivity of 20H1 1 with 65 16 GB 2 192 185 A 16 the isolates bearing type b flagella was also extensive in that 20H1 1 recognized all 59 isolates (100%).In contrast, the other anti-flagella type b monoclonal antibody, 3C1, bound to only 43 of the 59 isolates (73%).

These results demonstrate that 20H1 1 binds to a pan-reactive epitope (i. e., an epitope present on at least about 95% of flagellated A aeruginosa strains), whereas 3C1 bindsto an epitope not present on all type b flagel lin molecules. Even though the flagella type b antigen is serological ly uniform when analyzed with 5 polyclonal antisera (Lanvi, B.,supra, and Ansorg, R.,supra), the cross- reactivity patterns of 20H1 1 and3C1 surprisingly show that the type b flagella has at least two separate epitopes that can be identified by monoclonal antibodies.

The extensive cross-reactivity of antibodies 21 B8 and 20H1 1 with A aeruginosa clinical isolates indicates the particular clinical utility of these antibodies in immunotherapy of A aeruginosa infections. 10 Example 10

Example 10 demonstrates methods forthe production of another exemplary human monoclonal antibody thatbindstoP. aeruginosa type b flagella, as wel I as that antibody's protective activity against challenge with A aeruginosa in the burned mouse model. 15 Atransformed cell line was prepared and cloned essentially as described in Example 7, except thatthe transformedcell mixture was plated on 20 96-well tissue culture plates at a concentration of about 2250 E-PBMC per well. Supernatants were initially assayed by ELISA on the flagellated pool as described in Ex ample 6 except that Fisher immunotype 2 and 4 reference strains were absentfrom the pool. Positivewells were subsequently assayed on each of the strains in the flagellated pool including Fisher immunotype 2 and 20 4. The ultimately isolated cell line and the monoclonal antibody(IgG isotype)areboth identified bythel2D7 designation in the following text.

The12137human monoclonal antibody showed extensive cross-reactivity with anti-flagella type a isolates, recognizing 54 of the 56 flagella type-a-bearing clinical isolates (96%) tested. The protective activity of the 12137 monoclonal antibody is shown in Table VI. The protection studies were performed essentially as descri- 25 bed in Example 4, exceptthatthe challenge dose was 1 LDj()O and the challenge strain was 1624, a clinical isolate expressing Fisher immunotype 2 LPS and type a flagella. Table VI. Protection Study of the Human Anti-Flagella Type a Monoclonal Antibody 12D7 in the Burned Mouse Model' 30 Percent SurvivalbyDay2 Treatment3 1 2 3 4 5 6 7 8 9 35 Human anti-flagella a, 12D7 100100 100 100 100 100 100 100 100 Human anti-Fisher2 LPS, 2H9 90 90 90 90 90 90 90 90 90 40 Murine anti-flagella a, 11G5 100 100 100 100 100 100 100 100 100 Human anti-flagellab, 45 15F4 100 37.5 25 25 25 25 25 2525 Mice were challenged subescharwith 1 1_13100 (950 CFU) of A aeruginosa 1624.

2 Percent survival was based on the number of mice surviving per group often animals except for the 15F4 group where N =8. Days are post-burn and challenge. 50 Purified antibody (50 jig in 0.5 ml PBS) was administered Lp. two hours priorto burn and challenge.

Example 11

Example 11 demonstrates the production of a human monoclonal antibody reactive with flagella b type P.

aeruginosa and the protection of mice passively immunized with this antibody against challenge withP. 55 aeruginosa in the burned mouse model.

Atransformed cell line was prepared and cloned essentially as described in Example 10, except thatthe 1A2 to B cell transformation ratio was about 60:1 and the transformed cell mixture was plated on 15 plates at a concentration of about 1930 E-PBMC per well. Also, the cells were assayed on a pool of A aeruginosa strains comprising G98 (Fisher 3 immunotype, flagella type a) and 1739 (a clinical isolate of Fisher 5 imrqunotype, 60 flagella type b), and conflrred on a different pool of A aeruginosa strains: 1277, G98,1739 and Firherim munotypes F2, F4, F6 and F7. The ultimately isolated cell line and the secreted monoclonal antibody (IgG, isotype) are both referred to by the 2138 designation in the following text.

An immunofluorescence assay performed with 2138 was positive on a flagella type b, Fisher 2 immunotype strain, but negative in a flagella type a, Fisher immunotype4 reference strain. In clinical isolate testing 59/59 65 17 GB 2 192 185 A 17 (100%) of flagellated type b isolates tested positive.

The protective activity of 2138 is shwon in Table Vil. The protective studies were performed as described in Example 10, except that clinical isolate F164 (Fisher immunotype4, flagella type b) was used for the challenge.

5 Table VII. Protection Study of the Human Anti-Flagella Type b Antibody Monoclonal Antibody 2B8 in the Burned Mouse Model' Percent Survivalby Day2 10 Treatment 1 2 3 4 5 6 7 8 9 Human anti-flagella b, 2138 3 100 89 89 89 89 89 89 89 89 15 Murine anti-flagella b, PaF41VE8 4 100 100 100 100 100 100 90 90 90 Murineanti-Fisher2 LPS, VH3 4 90 20 20 20 20 20 20 20 20 20 Mice were challenged subescharwith 1 LD,00(105CFU)ofP.aeruginosaFl64.

2 Percent survival based on the number of mice surviving per group often animals except for the 2138 group where N=9. Days are post-burn and challenge.

3 Purified antibody (50 Lg in 0.5 ml PBS) was administered i.p. two hours prior to burn and challenge. 25 4 Purified antibody (5 pg in 0.5 ml PBS) was administered i.p. two hours priorto burn and challenge.

Example 12

Example 12 demonstrates the production of anotherhuman monoclonal antibody reactive with flagella b type A aeruginosa and the protection of mice passively immunized with this antibody against challenge with 30 P. aeruginosa in the burned mouse model.

Atransformed cel I line was prepared and cloned essentially as described in Example 7, with the following exceptions. A different individual was immunized (Pier et al. (1984) 45:309) and served as a B cell source, and the plating level of E-PBMC was about 2000 cells perwell. The screening was performed on pooled Fisher 1-7 immunotypes. The ultimately isolated cell line and the secreted monoclonal antibody (IgG, isotype) are both 35 referred to bythe 14C1 designation in the following text.

In clinical testing, 59/59 (100%) of flagellated type b isolates tested positive with 14C1. The protective studies were performed as described for Example 11 above and are shown in Table Vill.

Table Vill. Protection Study of the Human Anti-Flagella Type b Monoclonal Antibody 14C1 in the Burned 40 Mouse Model' Percent SurvivalbyDay2 Treatment3 1 2 3 4 5 6 7 8 9 45 Human anti-flagella b, 14C1 100 100 100 100 100 90 90 90 90 Murineanti-flagellab, 50 PaF41VE8 100 100 100 100 100 100 100 100 100 Human anti-Fisher 2 LPS, VH3 100 40 20 20 20 20 20 20 20 55 Mice were challenged subeschar with g reaterthan 1 LD100's (90 CFU) of A aeruginosa F1 64.

2 Percent survival was based on the number of mice surviving per group often animals exceptforthe PaR IVE8 group where N=9. Days are post-burn and challenge.

3 Purified antibody (50 I.Lg in 0.5 ml PBS) was administered i.p. two hours priorto burn and challenge.

60 From the foregoing, it will be appreciated that the cel I lines of the present invention provide mearisfor producing monoclonal antibodies and fragments thereof reactive with A aeruginosa flagella and cross protective against various A aeruginosa strains. Surprisingly, a number of monoclonal antibodies were isolated thatwere reactive with different epitopes on each type of flagella. The antibodies of the present invention allow prophylactic and therapeutic compositions to be more economically and easily produced for 65 18 GB 2 192 185 A 18 use against infections due to mostP. aeruginosa strains. In addition, the cell lines provide antibodies which find uses in immunoassays and otherwell-known procedures.

Although the present invention has been described in somedetail byway of illustration and examplefor purposes of clarity of understanding, itwill be obvious that certain changes and modifications may be practi ced within the scope of the appended claims. 5

Claims (48)

1. A composition comprising a monoclonal antibody or binding fragment thereof capable of specifically reacting with pseudomonasaeruginosa bacteria flagella. 10

2. A composition according to claim 1, wherein the monoclonal antibody or binding fragmentthereof inhibits motility of the bacteria.

3. A composition according to claim 1, wherein the monoclonal antibody or binding fragment thereof is protective in vivo.

4. A composition comprising a monoclonal antibody capable of specifically reacting with a flagellar pro- 15 tein epitope ofpseudomonas aeruginosa.

5. A composition according to claim 4, wherein the monoclonal antibody is capable of blocking the bind ing to the epitope of monoclonal antibodies produced by cell lines designated A.T.C.C. Accession Numbers HB9129, HB9130, CRI_9300, CRI_9301, CRI_9422, CRL9423, and CRL9424.

6. A composition comprising one or more monoclonal antibodies, wherein a first of said monoclonal 20 antibodies are capable of reacting with an epitope of type a ortype b flagella of pseudomonasaeruginosa.

7. A composition according to claim 6, wherein a second of said monoclonal antibodies is capable of reacting with a type of flagella not reactive with said first monoclonal antibody.

8. A composition according to either of claims 6 and 7, wherein said first antibody is a human monoclonal antibody. 25

9. A composition according to claim 6, wherein at least one of said monoclonal antibodies is protective in vivo.

10. A composition according to any of claims 1 to 9, further comprising a gamma globulin fraction from human blood plasma and/or an antibiotic agent.

11. A pharmaceutical composition comprising a composition according to any of claims 1 to 10 and a 30 physiologically acceptable carrier.

12. A pharmaceutical composition useful fortreating or preventing apseudomonasaeruginosa infection, said composition comprising a monoclonal antibody reactive with a flagella protein of A aeruginosa and protective in vivo, and antimicrobial agent, a gamma globulin fraction from human blood plasma and a physiologically acceptable carrier. 35

13. A pharmaceutical composition according to claim 12, wherein the antibody is a human monoclonal antibody and the gamma globulin fraction from human blood plasma is obtained from humans exhibiting elevated levels of immunoglobulins reactive withpseudomonas aeruginosa bacteria and/or products thereof.

14. A pharmaceutical composition comprising at least two monoclonal antibodies, each specifically reac- 40 ting with a different type ofpseudomonas aeruginosa flagellar protein and capable of treating or preventing pseudomonas aeruginosa infections.

15. A composition according to claim 14, wherein at least one of the monoclonal antibodies is a human monoclonal antibody.

16. A composition according to eitherof claims 14 and 15 further comprising at least one human mono- 45 clonal antibody capable of reacting with at least one serotypic determinant on a lipopolysaccharicle molecule ofpseudomonasaeruginosa and/or a monoclonal antibody reactive with exotoxin A.

17. Use of a composition according to anyone of claims 1 to 16 in a process for making a pharmaceutical forthe treatment of bacteremia and/or septicemia.

18. A cell line which produces a monoclonal antibody capable of specifically reacting with type a ortype b 50 pseudomonas aeruginosa flagella.

19. A cell line according to claim 18, wherein the monoclonal antibody is protective in vivo against pseudomonas aeruginosa.

20. A cell line according to either of claims 18 and 19, which is a hybrid cell line.

21. A cell line according to anyone of claims 18 to 21, which produces human monoclonal antibodies. 55

22. A cell line according to anyone of claims 18to 21, which is one of A. T.C.C. Accession Numbers HB9129,HB9130,CRL9300,CRL9301,CRL9422,CRL9423orCRL9424.

23. A method of producing monoclonal antibodies specific forflagella proteins of pseudomonas aerugi nosa and capable of treating or preventing pseudomonas aeruginosa infections, said method comprising:

cultivating at least one of the cell lines of claim 22 and recovering said antibodies. 60

24. Use of a monoclonal antibody produced according to claim 23 in a process of preparing a composi tion for the treatment of bacteremia and/or septicemia.

25. A monoclonal antibody orfragmentthereof reactive with an epitope which is capable of binding to a monoclonal antibody produced according to claim 23.

26. A composition comprising a human monoclonal antibody specifically reactive with an eptiope pre- 65 19 GB 2 192 185 A 19 sent on pseudomonas aeruginosa flagella.

27. A composition according to claim 26, wherein the epitope is present on type a ortype b flagella, but not both.

28. A composition according to claim 26, wherein the epitope is exhibited by at least about 70% of type b flagella. 5

29. A composition according to claim 26, wherein the epitope is exhibited solely by type b flagella.

30. A composition according to claim 29, wherein the antibody is panreactive.

31. A composition for the treatment of bacteria infections which contains a prophylactic ortherapeutic amount of a monoclonal antibody capable of binding to the flagellum of pseudomonas aeruginosa in com bination with one or more of: a prophylactic or therapeutic amount of a monoclonal antibody capable of 10 reacting with pseudomonas aeruginosa exotoxin A; a monoclonal antibody capable of reacting with at least one serotype determinant on a lipoplysaccharide fraction of pseudomonas aeruginosa; a gamma globulin fraction from human blood plasma; a gamma globulin fraction from a human blood plasma exhibiting eleva ted levels of imm unog lobul ins reactive with pseudomonas aeruginosa and/or products thereof; or an anti mocrobial agent. 15

32. A monoclonal antibody composition having the same binding specificity as anyone of the mono clonal antibodies designated FA6 IIG5, Pa3 IVC2, PaF4 lVE8,20H1 1, or2l 138.

33. A monoclonal antibody according to claim 32, conjugated to a label capable of providing a detectable signal.

34. A monoclonal antibody according to claim 33, wherein the label is a fluorescer or an enzyme. 20

35. A method for determining the presence ofpseudomonasaeruginosa in a sample, which comprises combining said sample with a monoclonal antibody reactive with pseudomonas aeruginosa flagella and detecting complex formation.

36. A kitfor use in detecting the presence ofpseudomonasaeruginosa bacteria, said kit comprising a monoclonal antibody composition containing at least one monoclonal antibody, wherein said antibody 25 reacts with a type-specificflagella protein of said bacteria, and labels providing for a detectable signal coval ently bonded to said antibody or bonded to second antibodies reactive with each said monoclonal antibody.

37. A composition according to claim 1 substantially as hereinbefore described.

38. A composition according to claim 6 substantially as hereinbefore described.

39. A composition according to claim 12 substantially as hereinbefore described. 30

40. A composition according to claim 14 substantially as hereinbefore described.

41. A cell line according to claim 18 substantially has hereinbefore described.

42. A method according to claim 23 substantially as hereinbefore described.

43. An antibody orfragment according to claim 25 substantially as hereinbefore described.

44. A composition according to claim 26 substantially as hereinbefore described. 35

45. A composition according to claim 31 substantially as hereinbefore described.

46. A composition according to claim 32 substantially as hereinbefore described.

47. A method according to claim 35 substantially as hereinbefore described.

48. A kit according to claim 36 substantially as hereinbefore described.

Printed for Her Majesty's Stationery Office by Croydon Printing Company (UK) Ltd, 11187, D8991685.

Published by The Patent Office, 25 Southampton Buildings, London WC2A1AY, from which copies maybe obtained.