AU610819B2 - Cross-protective human monoclonal antibody compositions - Google Patents

Description

required u L Y t? R t; PHILLIPS ORMONDE AND FITZPATRICK Patent and Trade Mark Attorneys 367 Collins Street Melbourne, Australia P17/2/83 7

AUSTRALIA

Patents Act COMPLETE SPECIFICATION

(ORIGINAL)

Class Int. Class Application Number: Lodged: A /7 610 8 1 9 Complete Specification Lodged: Accepted: Published: Priority o 0 Ok'O Related Art: 0 (J 0 r Name(s) of Applicant(s); Address(es) of Applicant(s): Actual Inventor(s): APPLICANT'S REF.: GS-6025A GENETIC SYSTEMS CORPORATION 3005 First Avenue, Seattle, Washington 98121, United States of America.

Howard V. Raff So' 0 Address for Service is: PHILLIPS, ORMONDE AND FITZPATRICK Patent and Trade Mark Attorneys 367 Collins Street Melbourne, Australia, 3000 Complete Specification for the invention entitled: CROSS-PROTECTIVE HUMAN MONOCLONAL ANTIBODY COMPOSITIONS The following statement is a full description of this invention, including the best method of performing it known to applicant(s): I'1931;84 P18//78 68619/87

DYF:JC

PHILLIPS ORMONDE FITZPATRICK Patent and Trade Mark Attorneys 367 Collins Street Melbourne, Australia I LI c C-L= Y~III~UI-YII I -1A- 0 0 000 0 o 0 0a o u 0n 00. 0 0g 00 C-I-P PATENT APPLICATION 9197-21-1 CROSS-PRO'TECTIVE HUMAN MONOCLONAL ANTIBODY COMPOSITIONS FIELD OF THE INVENTION The present invention relates to the application of immunological techniques to provide novel materials useful in treating and diagnosing bacterial infections and, more particularly, to the production and application of human monoclonal antibodies that are capable of protecting against infections caused by different genera of bacteria.

BACKGROUND OF THE INVENTION 15 Gram-positive and gram-negative bacteria may cause life-threatening disease in infected patients.

These bacterial infections often cause significant morbidity and mortality. There is an increased incidence of such infections in prematurely born infants, elderly patients, and patients who have serious underlying medical conditions such as burns, surgical trauma, slow-healing wounds, or malignancies. These infections are typically of nosocomial origin hospital-acquired), and occur particularly in patients who have sustained prolonged hospitalization-associated with surgical intervention, intravascular insult, or long-term therapy with immunosuppressive agents or antibiotics. In addition, newborns who have an immature immune system are apparently acutely susceptible to neonatal sepsis and meningitis caused by particular gram-negative and gram-positive bacteria.

Included among the most frequently encountered organisms in gram-negative and gram-positive disease are Escherichia coli coli), Klebsiella pneumoniae pneumoniae), Serratia marcescens tS.

marcescens), Enterobacter aerogenes and cloacae (E.

aerogenes/cloacae), Pseudomonas aeruginosa (P.

1A 10 AoT- al

T

i i i ff0 0 0 0 00 on no -v' o as a t 0. 0 0 aeruginosa), Neisseria meningitidis meningitidis), Group B Streptococcus and Staphylococcus aureus (S.

aureus) (Sonnenwirth, A. "The Enteric Bacilli and Similar Gram-Negative Bacteria," pp. 753-790, in Microbiology, 2nd Edition, Davis, B. Dulbecco, R., Eisen, H. Ginserberg, H. Wood, W. and McCarty, Eds., Harper and Row, (1973); McCabe, W. "Gram-Negative Bacteremia," Adv. Intern. Med., 19:135-138 (1974); Kreger, et al., "Gram-Negative Bacteremia III. Reassessment of Etiology, Epidemiology, and Ecology in 612 Patients," Am. J. Med. 68:332-343 (1980); Robbins, J. et al., "Escherichia coli K1 Capsular Polysaccharide Associated With Neonatal Meningitis," New Engl. J. Med., 290:1216-1220 (1974); and Hughs, J. et al., "Nosocomial Infection Surveillance, 1980-1982," Morb. Mort. Weekly Report, 32:1SS-16SS (1983)). Of these infections, usually several, but not all, serotypes of certain gram-negative bacteria, E. coli, K. pneumoniae, E_ aerogenes/cloacae, P. aeruginosa, and S. marcescens, cause bacteremia among the adult population. In contrast to adults, the immunologically immature neonate is particularly susceptible to septicemia and meningitis caused by the encapsulated strains of E.

coli, N. meningitidis Group B, Hemophilus influenzae type B, and the five type strains of Group B Streptococcus. Although other bacteria may also cause these infections, the bacteria cited above are the predominant isolates from the aforementioned blood infections.

Antibiotics have long been the primary 'therapeutic tool for the control and eradication of grampositive and gram-negative infections. However, the continued incidence and severity of the infections, the continual emergence of antibiotic resistant bacterial strains, and the inherent toxicity of some antibiotics, point to the limitations of antibiotic therapy. These j1 r i _i I ;i*_ii i _li__ili l o eo cs o 0000 o0 o oo o 0.0o oo o ro o o 0 0 0..

Sooo 00 0 00 0 00 0 0000 observations have prompted the search for alternative prophylactic and therapeutic approaches.

It is widely believed that antibodies reactive with structures accessible (externally exposed) on live bacteria may facilitate bacterial destruction by any of several mechanisms. Included among these mechanisms are: direct lysi: of the bacteria in the presence of serum complement, bacteriostasis, by the blockading of nutrient scavenger receptors, (3) opsonization and subsequent phagocytosis of the bacteria in the presence or absence of serum complement, or prevention of attachment of the bacteria to host tissues (Mims, "Recovery from Infection," in The Pathogenesis of Infectious Disease, pp. 198-222, Mims, 15 C. Ed., Academic Press (1982)). For bacteria that possess surface carbohydrate molecules, such as lipopolysaccharide (LPS) and/or capsules, antibody appears to be most effective via opsonization mechanisms (Kaijser, et al., "The Protective Effect Against E. coli of 0 and K Antibodies of Different Immunoglobulin Classes," Scand. J. Immunol., 1:276 (1972)). Therefore, antibodies directed to these accessible carbohydrate structures may provide an effective means for bacterial elimination.

25 In general, mammals that are exposed to disease-producing bacteria produce antibodies that are specific for LPS or capsule. These antigens are chemically diverse structures composed of frequently repeating oligosaccharide molecules and whose presence 3 determines the serotype of bacterial strains. Since they are often the immunodominant bacterial antigens, serotype specific antibodies (anti-LPS or capsule) have been the most studied of potentially therapeutic antibodies. However, because of the limited cross-reactivity of these antibodies, and the apparent highly diverse nature of carbohydrate antigens on pathogenic gram-positive and gram-negative bacteria, it -4- 0000 0 0 0 00 .0 03 0 0 40& O O 0 0 0 On O 0 0 JO 0 Co me! U 0 O U U 0 would be extremely difficult and costly to produce a therapeutic formulation containing only serotype specific antibodies (see, Kaijser, B. and Ahlstedt, "Protective Capacity of Antibodies Against Escherichia coli O and K Antigens," Infect.

Immun., 17:286-292 (1977); and Morrison, D. C. and Ryan, J. "Bacterial Endotoxins and Host Immune Response," Adv. Immunol., 28:293-450 (1979)). Regardless, various reports have stimulated visions that immunotherapeutic approaches could be found to treat gram-negative bacterial disease.

Fractionated human plasma, enriched for immune globulins containing specific and protective antibodies against the infecting organisms, have been some- 15 what effective against P. aeruginosa infections.

(Collins, M. S. and Robey, R. "Protective Activity of an Intravenous Immune Globulin (Human) Enriched in Antibody Against Lipopolysaccharide Antigens of Pseudomonas aeruginosa," Amer. J. Med., 3:168-174 20 (1984)). However, commercial products are not yet readily available due to certain inherent limitations which have prevented their widespread use in the treatment of life-threatening bacterial disease.

One such limitation associated with immune 25 globulin compositions is that they are assembled from large pools of plasma samples that have been preselected for the presence of a limited number of particular antibodies. Typically, these pools consist of samples from a thousand donors who may have low titers to some pathogenic bacteria. Thus, at best, there is only a modest increase in the resultant titer of desired antibodies.

Another such limitation is that the preselection process itself requires very expensive, continuous screening of the donor population to assure product consistency. Despite considerable effort, product lots can still vary between batches and geographic regions.

0 0(0 00 0 oo oa 0 CO000 o 0 00 o 0 0 0 0000 0 0 Yet another such limitation inherent in immune globulin compositions is that their use results in coincident administration of large quantities of extraneous proteinaceous substances viruses) having the potential to cause adverse biologic eftects. The combination of low titers of desired antibodies and high content of extraneous substances often limits, to suboptimal levels, the amount of specific and thus beneficial immune globulin(s) administrable to the patient.

In 1975, Kohler and Milstein reported that certain mouse cell lines could be fused with mouse spleen cells to create hybridomas which would secrete pure "monoclonal" antibodies (Kohler, G. and Milstein, "Continuous Cultures of Fused Cells Socreting Antibody of Predefined Specificity," Nature, 256:495-497 (1975)). With the advent of this technology, the potential existed to produce murine antibodies to any particular determinant or determinants on antigens.

20 Usi -this-te-hi.-al-og-y-,-mouse-monoG--ona antibodies have been \erived from mice immunized with polysaccharide from Neisseria meningitidis Group B. These murine IgM monoclonal 'antibodies were observed to bind and opsonize several Kl ositive E. coli strains regardless of their LPS ser ypes (Cross, supra, Soderstrom, supra, and Cross, A. et al., "The Importance of the Kl Capsule in Invasive Infections Cause by Escherichia coli," J. Inf. Dis., 149,:184-193 (1984)).

Moreover, the monoclonal antibodi were protective in mice against lethal challenges with E. coli K1 and Group B meningococcal organisms (Cross, supra, and Sunderstrom, supra). In another example mouse monoclonal antibodies specific to type III Grup B Streptococcus were reported to be protectiv in a mouse experimental infection model (Egan, M. et l.> "Pr nt e r t-i-on--f-Mie-f-r-Gm-E-x-pe-r-i-me-n-a-1 0 Using this technology, mouse monoclonal antibodies have been derived from mice immunized with poly-saccharide from Neisseria meningitidis Group B. These murine IgM monoclonal antibodies were observed to bind -and opsonize several Kl-positive E. coli strains regardless of their LPS serotypes (Cross, "Evaluation of Immunotherapeutic Approaches for the Potential Treatment of Infections caused by Kl-positive Escherichia coli," J. Infect Dis. 147 68-76 (1983) Soderstrom, "serological and Functional Properties of monoclonal Antibodies to Escherichia coli Type 1 Pilus and Capsular Antigens," Prog. Allergy 33 259-274 (1983), and Cross, et al., "The Importance of the K1 Capsule in Invasive Infections Caused by Escherichia coli," J. Inf.

Dis., 149:184-193 (1984). Moreover, the monoclonal antibodies were protective in mice against lethal challenges with E. coli K1 and Group 3 meningococcal organisms (Cross, J. Infect. Dis. 147:68-76 (1983), soderstrom supra). In another example, mouse monoclonal antibodies specific to type III Group B Streptococcus were reported to be protective in a mouse experimental infection model (Egan, et al., "Protection of Mice from Experimental Infection with 0 o 0 o 0 0 0 O 00 0000 000C 0:00 o o o 0 0 0 00 0 oooooo o cO 0 00 0 0 0 C 0 0~ 0 0 0 0 0 CO i Type III Group B Streptococcus Using Monoclonal Antibodie3," J. Exp. Med., 1:1006-1011 (1983)).

A mouse monoclonal antibody, while useful in treating mice, has major disadvantages for use in hu- Smans. The human immune system is capable of recognizing any mouse monoclonal antibody as a foreign protein.

This can result in accelerated clearance of the antibody and thus abrogation of its pharmacological effect (Levy, R. and Miller, R. "Tumor Therapy with Monoclonal Antibodies," Fed. Proc., 42:2650-2656 (1983)).

More seriously, this could conceivably lead to shock and even death from allergic reactions analogous to "serum sickness." Clinical experience has shown that S,.o'o anti-mouse immunoglobulin responses have limited the 15 utility of these antibodies in approximately one-half of the patients receiving mouse monoclonal antibodies for treatment of various tumors (Sears, H. et al., 0°00° "Phase I Clinical Trial of Monoclonal Antibody in So Treatment of Gastrointestinal Tumor," Lancet, 1:762-764 (1982); and Miller, R. et al., "Monoclonal Antibody Therapeutic Trials in Seven Patients with T-Cell Lymphoma," Blood, 62:988-995 (1983)).

o a Accordingly, there is a need for human mono- 0 a 4 clonal antibodies which are protective against gramnegative and gram-positive bacterial disease. However, the diverse antigenicity of gram-positive and gram-negative disease-causing bacteria strongly suggests that producing serotype specific human monoclonal o"'a antibodies to each of the many important bacterial 3b pathoges would be impractical.

The diverse antigenicity of gram-negative bacteria is attributed to the variable regions of the lipopolysaccharide (LPS), a molecule associated with the outer membrane of gram-negative organisms. The LPS molecule is generally considered to be composed of three structural regions. The region closest to the outer membrane is the so-called lipid A portion of LPS.

F y t\ ',1 -7oo 0300 o o 0 0 0 0 0 00 .0 0 o 0 00 Q Do 0 00 S00 0 00 0 000 0 0 0 :j7 This structurally conserved region possesses the endotoxic activity associated with gram-negative disease. The second structural region, termed core, is linked to a lipid A often via a K t-d 2-4et4-3-deoxy-D-mannooctonate residue (KDO) and, similar to the lipid A region, is not usually accessible to antibody when the third outermost region of LPS is present. Although this region is partially conserved within some gram-negative bacterial species, many deviations in complete core have been found among members of the family Enterobacteriaceae. The outermost region of an LPS molecule is composed of repeating oligosaccharide units and is known as the O-specific side chain. The sugars in these oligo- 15 saccharide units comprise molecular entities that exhibit serotype specific structural antigenic diversity.

Thus, the sugars themselves, their sequence, and their linkages determine O-side chain antigenicity via their tertiary structure. Antibodies to these O-groups have generally been found to be serotype specific.

Serotypes are typically defined by their reactivity with monospecific antisera, which possess binding activity for only one particular antigenic determinant.

See generally, Mayer et al., Meths. Microbiology, 18:157-201 (1985).

Antisera to the core and lipid A regions of LPS have been produced in efforts to demonstrate protection against gram-negative infection.

Sakulramrung and Domingue, J. Inf. Dis., 151:995-1104 (1985); McCabe, et al., J. Infect. Dis., 1365:516 (1977); and Mullan, et al., Infect. Immun., 10:1195-1201 (1974). More recently, mouse and human monoclonal antibodies reactive with the conserved regions have been produced. Although these antibodies have sometimes shown partial in vivo efficacy in, tailored model systems (Teng, et al., Proc. Natl. Acad.

Sci. USA 82:1790 (1985); and Bogard and Kung, Patent Application No. W085/01659), other laboratories have not been able to demonstrate similar effects. See Elkins and Metcalf, Infect. Immun. 48:597 (1985); and Gigliotti and Shenap, J. Inf. Dis. 151:1005-1011 (1985). Moreover, these antibodies generally do not react with (bind to) intact, viable gram-negative bacteria or to purified LPS molecules. These findings suggest that it is doubtful the core or lipid A portions of LPS w l--be--aess1 on bacteria in their natural and infectious state would be accessible to antibody. It is also well accepted that anti-core or anti-lipid A antibodies will not react with grampositive bacteria because the latter do not possess LPS. In view of these findings, it is unlikely that 15 monoclonal antibodies to the conserved core or lipid A regions of LPS will be efficacious in the treatment of human gram-negative or, for that matter, gram-positive bacterial disease.

Thus, there still exists a significant need 20 for human monoclonal antibodies that are broadly (in- 20 tergenus) cross-protective against gram-positive and gram-negative bacterial diseases, as well as for methods for practical production and use of such antibodies. The present invention fulfills these needs.

O o 08 0 c: uo o eu o o 0 34 0 4 0 o.)i~l 0 .9.e SUMMARY OF THE INVENTION Novel cell lines are provided which produce human monoclonal antibodies capable of specifically cross-reacting with a plurality of bacterial species by binding an accessible epitope comprising a non-core carbohydrate moiety present on at least two different bacterial species. Additionally, methods are provided for prophylactically treating a human patient susceptible to bacterial infection and therapeutically treating a patient suffering from such an infection by administering an effective amount of a composition comprising a plurality of human monoclonal antibodies, -9- 00099 00 0 o 0 j* 0 oi r 0 o 0o o 00 0 G 0d O o o o 0 0 0 Sa a 00 0 R s i 6~ fl O 1 wherein at least one of these antibodies is capable of reacting with a non-core carbohydrate antigenic determinant shared by two or more bacterial species.

The composition preferably includes a physiologically acceptable carrier, and may also contain any one or more of the following: additional human monoclonal antibodies capable of reacting with other bacterial genera; a gammaglobulin fraction from human blood plasma; a gammaglobulin fraction from human blood plasma, where the plasma is obtained from a human exhibiting elevated levels of immunoglobulins reactive with one or more bacterial genera; and one or more antimicrobial agents.

15 DESCRIPTION OF THE SPECIFIC EMBODIMENTS In accordance with the present invention, novel cells capable of producing human monoclonal antibodies and compositions comprising such an-ibodies are provided, such compositions being capable of selectively reacting with a plurality of bacterial genera responsible for nosocomial, neonatal, or other infections, where individual antibodies typically react with non-core carbohydrate epitopes present on multiple bacterial genera. The subject cells have identifiable 25 chromosomes in which the germ-line DNA from them or a precursor cell has rearranged to encode an antibody or binding fragment thereof having a binding site for an antigenic determinant (epitope) shared by carbohydrate molecules found on at least some serotypes of two or more bacterial genera. These human monoclonal antibodies can be used in a wide variety of ways, including for diagnosis, prophylaxis and therapy of bacterial disease.

Typically, the cells of the present invention will be cells capable of stable production of a human antibody in culture, particularly immortalized human lymphocytes that produce protective human monoclonal Llj_ 4,no 0 4 0 4 S P 0' o 04 Pl antibodies to non-core carbohydrate determinants on accessible molecules shared by at least two bacterial species. By "accessible" is meant that the non-core carbohydrate determinants are physically available in the environment of use for direct interaction with the monoclonal antibodies. The monoclonal antibodies so provided are useful in the treatment or prophylaxis of serious disease caused by a wide range of bacterial infections. Furthermore, those non-core carbohydrate molecules that are released into the surrounding environment are also free to interact directly with the antibody molecules and be cleared via the reticuloendothelial system.

The compositions containing the monoclonal antibodies of the present invention will typically be useful in the therapeutic and prophylactic treatment of nosocomial, neonatal, and other infections. As nosocomial infections are typically caused by infections from the following bacteria: Escherichia coli, Pseudomonas aeruginosa, Klebsiella pneumoniae, Enterobacter aerogenes/cloacae, Serratia marcescens, and Streptococcus agalactiae Group B, antibody compositions protective against two, three, four, or more of such bacteria would be preferred. Similarly, for neonatal use, such as in neonatal sepsis and meningitis, the antibodies are desirably specific for two or more of the following bacterial organisms: Escherichia coli Kl, Neisseria meningitidis Group B, Streptococcus agalactiae Group B, and Hemophilus influenzae type B.

Other common infectious bacteria include: Staphylococcus aureus, Staphylococcus epidermidis, Streptococcus pneumoniae, Proteus mirabilis, Proteus vulgaris, Bacteroides fragilis, Pseudomonas cepacia, Mycobacterium tuberculosis, Providencia morganii, Salmonella typhi, Pneumocystis carinii, Acinetobacter herellea, Pasturella multocida, Klebsiella oxytoca. For additional relevant pathogenic bacteria known to those or more bacterial species of different genera and a plurality of serotypes in at least one species, wherein said species /2 -11t P 0 ,1 skilled in the art, see, Hughs, et al., "Nosocomial Infection Surveillance, 1980-1982," Morb. Mort.

Weekly Report, 32:1SS-16SS (1983), and, generally, Microbiology, 3rd Edition, Davis, Dulbecco, R., Eisen, Ginserberg, Wood, and McCarty, Eds., Harper and Row (1980), both of which are incorporated here 1 .n by reference. The monoclonal antibodies will react with individual members or all of the members of a 'particular bacterial species, where the members may be distinguished by their surface epitopes, particularly LPS or capsule sites, e.g. serotypes.

The unexpected discovery of monoclonal antibody cross-reactivity across various bacterial species, including the clinically important species listed 15, above, provides novel means for therapeutic and prophylactic treatments. By utilizing pre-selected crossreactive antibodies in combination, a mixture of a few antibodies can be produced for treatment against a number of different species of infectious bacteria.

By way of example, and not of limitation, a mixture of two monoclonal antibodies, one cross-reactive with at least two bacterial species of clinical significance and the second cross-reactive with at least two or three different species, will be useful in treat.;ent against four, five, six or more different species. Adding a third or fourth monoclonal Antibody, each one og which is cross-reactive with at least two clinicall.' important species--even if one or more of the specios is the same as that recognized by the first and/or second antibody, will increase usefulness in treatment against five to ten or more species. Of course, it may be necessary to also add one or more monoclonal antibodies, each specific for just a single pre-selected bacterial species, for example, when monoclonal antibodies cross-reactive with that species are unavailable.

111 -12- SSimilarly, new methods of treating bacterial infections are also provided based Tn the discovery.

Again, by way of example and not limitation, one novel S. method entails treating a patient suspected of having or being susceptible to a bacterial infection caused by a selected bacterial species. The treatment includes administering a composition comprising a monoclonal antibody reactive with the bacterial species suspected Sof causing the infection, wherein the monoclonal antibody was initially characterized as reactive with a different bacterial species.

Another example is a method of treating bacterial infections by administering compositions comprising a plurality of monoclonal antibodies reactive S 15 with a substantial proportion greater than Spreferably 60% to 80% or more, most preferably about o 90%) of pze-selected, clinically important bacterial o species, wherein the number of antibodies is at least about two less than the number of bacterial species.

Typically, if represents the number of bacterial species, the composition will comprise about n 2 antibodies, more typically about n 4 to n 8 or less antibodies for treatment against up to about 15 tc -0 G' j bacterial species. In situations where treatment against a broad spectrum 25 to 50'or more) of bacterial species is desired, the composition will typically comprise n 10 to n 20 antibodies or less.

Preparation of monoclonal antibodies can be So°7 accomplished by immortalizing the expression of nucleic acid sequences that encode for antibodies or binding fragments thereof specific for a non-core carbohydrate epitope present on multiple bacterial species. Typically, the monoclonal antibodies are produced by celldriven Epstein-Barr Virus (EBV) transformation of lymphocytes obtained from human donors who are, or h-ive been exposed to the respective gram-negative bacteria.

The antibody-secreting cell lines so produced are I- i, The following statement is a full description of this invention, including the best method of performing it known to applicant(s): PI9/3/84 1 riLB i -II III. -13characterized as continuously growing lymphoblastoid cells that possess a diploid karyotype, are Epstein-Barr nuclear antigen (EBNA) positive, and secrete monoclonal antibody of either IgG, IgM, IgA, or IgD isotype. The cell-driven transformation process itself is an invention assigned to Genetic Systems Corporation and is described in detail in United States Patent No. 4,464,465 which is incorporated herein by reference. The monoclonal antibodies may be used intact, or as fragments, such as Fv, Fab, F(ab') 2 but usually intact.

Alternatively, cell lines producing the antibodies could be produced by cell fusion between suitabhy drug-marked human myeloma, mouse myeloma, or .on 15 human lymphoblastoid cells with human B-lymphocytes to o yield hybrid cell lines.

0; The cell lines of the present invention may o find use other than for the direct production of the o human monoclonal antibodies. The cell lines may be '0o 20 fused with other cells (such as suitably drug-marked a, so human myeloma, mouse myeloma, or human lymphoblastoid cells), to produce hybridomas, and thus provide for the transfer of the genes encoding the human monoclonal ooo antibodies. Alternatively, the cell lines may be used o 25 as a source of the DNA encoding the immunoglobulins, which may be isolated and transferred to cells by techniques other than fusion. In addition, the genes encoding the monoclonal antibodies may be isolated and used in accordance with recombinant DNA techniques for co0o 30 the production of the specific immunoglobulin in a variety of hosts. Particularly, by preparing cDNA libraries from messenger RNA, a single cDNA clone, coding for the 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 producvion.

b

T

7 ~uLUUL.

-14- 00 *0 00 "t r "oa 0 0 C 0 000 oa 0 0 o oooo 0 0 o 0O 0 0 000 0 r0 0 07 000 oe o 0 00 0 (0 0000 00 0 The lymphoblastoid or hybrid cell lines may be cloned and screened in accordance with conventional techniques, with the antibodies that are capable of binding to the epitopes of different bacterial genera detected in the cell supernatants.

The monoclonal antibodies of this invention find particular utility as components of pharmaceutical compositions containing a therapeutic or prophylactic amount of at least one of the monoclonal antibodies of 10 this invention in conjunction with a pharmaceutically effective carrier. A pharmaceutical carrier can be any compatible, non-toxic substance suitable for delivery of the monoclonal antibodies to the patient. Sterile water, alcohol, fats, waxes, and inert solids may be included in the carrier. Pharmaceutically accepted adjuvants (buffering agents, dispersing agents) may also be incorporated into the pharmaceutical composition. Such compositions can contain a single monoclonal antibody cross-reactive with non-core carbo- 20 hydrate epitopes shared by two or more bacterial species that cause, for example, nosocomial and neonatal sepsis or meningitis) infections. Alternatively, a pharmaceutical composition can contain two or more monoclonal antibodies to form a "cocktail." For example, a cocktail containing human monoclonal antibodies each protective against two or more gram-negative bacterial genera responsible for human infections, would have activity against the great majority of the common clinical isolates. If desired, one or more of 30 the monoclonal antibodies could be selected to be 30 cross-reactive with gram-positive bacteria as well, making even broader product applications feasible.

Of interest are prophylactic and/or therapeutic monocional antibody compositions capable of reacting with non-core carbohydrate determinants shared by three or more, usually at least five, and more usually at least ten, and up to fifteen or more rri.

0 0 oooo o o0 0 0 *o0,0 0 0 a oo S0oo 0 oo 0 C o 0 0 O Oo 0 00 'Uo bacterial serotypes, which includes at least two, usually at least three, more usually at least five, and usually fewer than ten bacterial genera.

Of particular interest are monoclonal antibody compositions which react with at least about three, preferably at least about five and up to and including all of the following common nosocomial infection-causing bacteria: Escherichia coli, Pseudomonas aeruginosa, Klebsiella pneumoniae, Enterobacter aerogenes/cloacae, Serratia marcescens, and Streptococcus agalactiae Group B. For treatment of neonatal infections, desirably the compositions will react with at least two, usually at least three, and

I

more usually at least four and up to and including all 15 of the following infection-causing bacterial genera: Escherichia colitKl, Neisseria meningitidis Group B, Streptococcus agalactiae Group B, Hemophilus influenzae type B, Staphylococcus aureus, and Staphylococcus epidermidis.

20 Each of the compositions will include at least two, usually at least three to five, and more usually six to ten human monoclonal antibodies, where at least one antibody reacts with non-core carbohydrate epitopes of the LPS molecules) shared by two or more bacterial genera and providing protection. Typically, the antibody will not bind to all serotypes of each bacterium, but may bind to two, three or more serotypes. Desirably, there will be at least one monoclonal antibody which binds to an accessible 30 non-core carbohydrate moiety of at least two genera of gram-negative bacteria and at least one monoclonal antibody that binds to an accessible carbohydrate moiety of a gram-negative bacterium and a gram-positive bacterium.

The mole ratio of the various monoclonal antibody components will usually not differ one from the other by more than a factor of 10, more usually by not Ar-- 0000 So 00 o o 0 0 o ii o S I 00 Q co, c o 0n o oo 0. 4 ft Qd I fl -16more than a factor of 5, and will usually be in a mole ratio of about 1:1-2 to each of the other antibody components.

The human monoclonal antibodies may also find use individually, particularly where the pathogen has been identitied or is limited to a narrow range of pathogens within the binding spectrum of the particular antibody.

The human monoclonal antibodies of the present invention may also be used in combination with other monoclonal antibodies (e.g.,jcm y acigne4 application entitled "Monoclonal Antibodies Cross- Reactive and Protective Against aeruginosa Serotypes," t-e pa r m-..1 15 S0, la8 which is incorporated herein by reference) as well as existing blood:plasma products, such as commercially available gamma globulin and immune globulin products used in prophylactic or therapeutic treatment of bacterial disease in humans. Preferably, 20 for immune globulins the plasma will be obtained from human donors exhibiting elevated levels of immunoglobulins reactive with various infectious bacterial genera.

See generally, the compendium "Intravenous Imune Globulin and the Compromised Host," Amer. J. Med., 76(3a), 25 March 30, 1984, pgs 1-231, which is incorporated herein by reference.

The monoclonal antibodies of the present invention can be used as separately administered compositions given in conjunction with antibiotics or antimicrobial agents. Typically, the antimicrobial agents may include a penicillin or cephalosporin carbenicillin, penicillin G, or the like) in conjunction with an aminoglycoside gentamicin, tobramycin, etc.), but numerous additional agents cephalosporins, sulfa drugs, etc.) well-known to those skilled in-,the art may also be utilized.

-17o o o Or) 0 0 0i n S C- 00(0 o co 0 00 o o o 0 S0 0 00 0000 0 0 o 0 The human monoclonal antibodies and pharmaceutical compositions thereof of this invention are particularly useful for oral or parenteral administration. Preferably, the pharmaceutical compositions may be administered parenterally, subcutaneously, intramuscularly or intravenously. Thus, this invention provides compositions for parenteral administration which comprise a solution of the human monoclonal antibody or a cocktail thereof dissolved in an acceptable carrier, preferably an aqueous carrier. A variety of aqueous carriers can be used, water, buffered water, 0.4% saline, 0.3% glycine and the like. These solutions are sterile and generally free of particulate matter. These compositions may be sterilized by conventional, well known sterilization techniques. The compositions may contain pharmaceutically acceptable auxiliary substances as required to approximate physiological conditions such as pH adjusting and buffering agents, toxicity adjusting agents and the like, for 20 example sodium acetate, sodium chloride, potassium chloride, calcium chloride, sodium lactate, etc. The concentration of antibody in these formulations can vary widely, from less than about usually at or at least about 1% to as much as 15 or 20% by 25 weight and will be selected primarily based on fluid volumes, viscosities, etc., in accordance with the particular mode of administration selected.

Thus, a typical pharmaceutical composition for intramuscular injection could be made up to contain 30 1 ml sterile buffered water, and 50 mg of monoclonal antibody. A typical 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 apparent to those skilled in the art and are described in more detail in, for example, Remington's Pharmaceutical Science, 15th Ed., ~Y c :i ur~m~-;Lii, -18- 0 00 o 0 o000 0 0 0 0 00 0" P. o 00 O O 0 00 0 00 a c -u 0 0 ORQ 0 S00 0 0 00009 C.0 0 Mack Publishing Company, Easton, Pennsylvania (1980), which is incorporated herein by reference.

The monoclonal antibodies of this invention can be lyophilized for storage and reconstituted in a suitable carrier prior to use. This technique has been shown to be effective with conventional immune globulins and art-known lyophilization and reconstitution techniques can be employed. It will be appreciated by those skilled in the art that lyophilization and reconstitution can lead to varying degrees of antibody activity loss with conventional immune globulins, IgM antibodies tend to have greater activity loss than IgG antibodies) and that use levels may have to be adjusted to compensate.

The compositions containing the present human monoclonal antibodies or a cocktail thereof can be administered for the prophylactic and/or therapeutic treatment of bacterial infections. In therapeutic application, compositions are administered to a patient already infected, in an amount sufficient to cure or at least partially arrest the infection and its complications. An amount adequate to accomplish this is defined as a "therapeutically effective dose." Amounts effective for this use will depend upon the severity of 25 the infection and the general state of the patient's own immune system, but generally range from about I 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 mate- 30 rials of this invention may generally be employed in serious disease states, that is life-threatening'or potentially life-threatening situations, especially bacteremia and endotoxemia. In such cases, in view of the absence of extraneous substances and the absence of "foreign substance" rejections which are achieved by the present human monoclonal antibodies of this invention it is possible and may be felt desirable by -Y -19o o o (0 0 0 O 000 o 00 0: e 0 00b 00t 00 O 0 0 V 0 the treating physician to administer substantial excesses of these antibodies.

In prophylactic applications, compositions containing the present antibody or a cocktail thereof are administerrl to a patient not already infected by the corresponding bacteria 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 ate of health and general level of immunity, but generally range 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 15 pattern being selected by the treating physician. In any event, the pharmaceutical formulations should provide a quantity of the antibody(ies) of this invention sufficient to effectively treat the patient.

Monoclonal antibodies of the present invention can further find a wide variety of utilities in vitro. By way of example, the monoclonal antibodies can be utilized for bacterial typing, for isolating specific bacterial strains or fragments thereof, for vaccine preparation, or the like.

25 For diagnostic purposes, the mbnoclonal antibodies may either be labeled or unlabeled. Typically, diagnostic assays entail detecting the formation of a complex through the binding of the monoclonal antibody to the LPS of the 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 human immunoglobulin. Alternatively, the monoclonal antibodies can be directly labeled.

A wide variety of labels may be employed, such as radionuclides, fluorescers, enzymes, enzyme substrates, ~11.1~ j 00 0 jO 0 00 00 0 .o r 0 03 0) 0j 0 P~ 0 0 0000 0 0 0 00 0 enzyme cofactors, enzyme inhibitors, ligands (particularly haptens), etc. Numerous types of immunoassays are available, and by way of example, some of the assays are 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.

Commonly, the monoclonal antibodies of the present invention are utilized in enzyme immunoassays, 10 where the subject antibodies, or second antibodies from a different species, are conjugated to an enzyme. When a sample, such as human blood or lysate thereof, containing one or more bacteria of a certain genus or serotype, is combined with the subject antibodies, binding occurs between the antibodies and those molecules exhibiting the selected epitopes. Such cells may then be separated from the unbound reagents, and a second antibody (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 in the detection of bacterial in- 25 fection or for the presence of a selected antigen.

Thus, the subject monoclonal antibody composition of the present invention may be provided, usually in a lyophilized form in a container, either alone or in conjunction with additional antibodies specific for other gram-negative bacteria. The antibodies, which may be conjugated to a label or unconjugated, are included in the kits with buffers, such as Tris, phosphate, carbonate, etc., stabilizers, biocides, inert proteins, bovine serum albumin, or the like. Generally, these materials will be presenb in less than about 5% wt. based on the amount of active antibody, and usually present in total amount of at O 44 O 4 o 0 4 415) Do 5 a 0 .43 -21least 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 may be present in from about 1 to 99% wt. of the total composition.

Where a second antibody capable of binding to the monoclonal antibody is employed, this will 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.

The following experimental data and information are offered by way of example and not limitation.

EXAMPLE I This EXAMPLE demonstrates methods for the production of a human monoclonal antibody that possesses intergenus cross-reactivity against members of the genera Escherichia coli coli), Serratia marcescens 20 marcescens), Klebsiella pneumoniae pneumoniae), and Enterobacter aerogenes aerogenes). Further, this EXAMPLE demonstrates the in vivo protective activity of said antibody, against a lethal challenge of homologous (cross-reacting) E. coli, and E.

aerogenes serotypes.

A. Obtaining Suitable Human Cells Suitable human B cells (lymphocytes) were obtained from the peripheral blood of an individual 30 known to have harbored the disease cystic fibrosis.

Mononuclear cells were separated from the peripheral blood by standard centrifugation techniques on Ficoll- Paque (Boyum, "Isolation of Mononuclear Cells and Granulocytes From Human Blood," Scand. J. Clin. Lab.

Invest., 21. Suppl. 97:77-89, (1968)) and washed twice in calcium-magnesium free phosphate buffered saline (PBS) prior to suspension in 1 ml of 90% fetal'bovine 5)0 4 5555 55 555 5,u 5)4 i. comprising a p±uraiity or numan monociouni- auIL.nuu o,

~P-W~

7 -22serum (FBS) and 10% dimethyl sulfoxide and freezing at -196 0 C in liquid nitrogen.

When the mononuclear cells were to be transformed, an ampule containing 5 x 107 cells was rapidly 5 thawed at 37 0 C. The cell suspension was added to 10 ml of Iscove's medium containing 15% FBS and centrifuged at room temperature for 10 min at 250 x g. The mononuclear cells were depleted of T cells (lymphocytes) using a modified E-rosetting procedure. Briefly, the cells were resuspended to a concentration of 1 x 107 cells/ml in PBS containing 20% FBS at 4 0 C. One ml of this suspension was placed in a 17 x 100 mm polystyrene round bottom tube to which was added 1 x 109 2-aminoethyl-isothiouronium bromide (AET)-treated sheep red blood cells from a 10% solution in Iscove's medi- Sum (Madsen, M. and Johnson, "A Methodological 0 Study of E-rosette Formation Using AET Treated Sheep Red Blood Cells," J. Immun. Methods, 27:61-74, (1979)).

o The suspension was vigorously mixed for 5-10 min at 4°C S 20 and the E-rosetted cells were removed by centrifugation through Ficoll-Paque for 8 min at 2500 x g at 4 0

C.

E-rosette negative peripheral blood mononuclear cells (E PMBC) banding at the interface were washed once in K. Iscove's medium and resuspended in same containing 25 w/v FBS L-glutamine (2 mM/1), sodium pyruvate (1 mM/l), penicillin 100 IU/ml), streptomycin (100 pg/ ml), hypoxanthine (1 x 10 4 and thymidine (1.6 x -7 This medium is hereafter referred to as HAT medium.

B. Cell-Driven Transformation of Peripheral BLood Mononuclear Cells Cell-driven transformation of the E PBMC was accomplished by cocultivating the E PBMC with a transforming cell line. The transforming cell line was an EBNA positive human lymphoblastoid cell line derived by ethylmethane sulphonate mutagenesis of the GM 1500 I -23lymphoblastoid cell line. Selection in the presence of pg/ml 6-thioguanine rendered the cells hypoxanthineguanine phosphoribosyl transferase deficient and thus HAT sensitive. The cell line is denominated the 1A2 cell line and was deposited at the American Type Culture Collection on March 29, 1982 under A.T.C.C. No. CRL 8119. 1A2 cells in logarithmic growth phase were suspended in HAT medium and combined with the E PBMC at a ratio of 30 1A2 cell per E PBMC. The cell mixture was plated into 10 round-bottom 96-well microtiter plates at a concentration of 62,000 cells/well in a volum- of 200 pl/well, and the culture incubated at 37 0 C in a humidified atmosphere containing 6% CO 2 Cultures were fed five days post transforma- 15 tion by replacement of half the supernatant with fresh r a HAT medium. The wells were observed every other day on an inverted microscope for signs of cell proliferation.

STen days after plating the cell mixture and after the So. 1A2 cells had died due to HAT selection, feeding of the 20 the wells was accomplished with a new media formulation o identical to HAT media except that it lacked the aminopterin component. Fifteen days post plating, it was observed that 100% of the wells contained prolifo f erating cells and that in most of the wells, the cells a 25 were of sufficient density for removal and testing of supernatants for anti-E. coli or anti-S. marcescens a antibody.

C. Detection of Specific Antibody Secreting Cells 3 .30 Supernatants were screened for the presence of anti-E. coli or anti-S. marcescens antibodies 'using an enzyme-linked immunosorbent assay (ELISA) technique (Engvall, "Quantitative Enzyme Immunoassay (ELISA) in Microbiology," Med. Biol., 55:193-200, (1977)). The antigen plates consisted of a series of flat-bottom 96-well Immunlon 2 microtiter plates, the wells of which contained a mixture of either viable E. coli or -24o 00 0 o 0 o o o 00 .0 0 C. U£ 0 0 0 S. marcescens serotypes affixed to the well surfaces with poly-L-lysine (PLL). Briefly, 50 pl of PLL (1 pg/ml) in PBS was added to each well for 30 min at room temperature The plates were washed three S times with PBS and either PBS or 50 pl of a mixed bacteria suspension at O.D.

6 60 0.2 was added to each well. The plates were incubated at 37 0 C for 60 min and washed 3 times with saline/0.02% Tween 20 (saline/T) to removed unattached bacteria. Various antigen plates used in the screen included: 1) a mixture of E. coli serotypes 01 No. 23499) and 04 No.

12791); 2) a mixture of S. marcescens serotypes 07, 015, 016 and 018 (all reference typing strains were obtained from the Communicable Disease Center (CDC) Atlanta, GA); and 3) a microtiter plate with no bacteria.

For the ELISA procedure, assay wells were first blocked with 200 pl of a mixture containing w/v dry non-fat milk, 0.0001% Pea+nIA, and 0.01% w/v Thimerosal in 500 ml PBS to prevent non-specific protein binding. After incubation for 1 hour at RT, the plates were washed three times with 200 pl/well/wash of saline/T. To each well was added 50 pl of a mixture containing 0.1% Tween 20 and 0.2% bovine serum albumin in PBS (PTB). Supernatants from wells of the culture plate were replica plated into corresponding wells of the antigen and control plates (50 pl/well) and the plates were incubated at RT for 30 min. The supernatants were then removed, plates were washed five 30 times with saline/T, and 50 pl of biotinylated goat anti-human immunoglobulin (Ig)(TAGO #9303040 diluted 1:250 in PTB) was added to each well. After a 30 min incubation at RT the biotinylated reagent was removed, the wells washed five times with saline/T and 50 Up of a preformed avidin:biotinylated horseradish peroxidase complex (Vectastain ABC Kit, Vector Laboratories) was S added to each well. After 30 min. at RT the Vectastain Fa 0r En t ABC reagent was removed, the wells were washed five times with saline/T, and 100 pl of substrate (0.8 mg/ml ortho-phenylenediamine dihydrochloride in 100 mM citrate buffer, pH 5.0 plus 0.03% H 2 0 2 in deionized mixed in equal volumes just before plating) added to each well. After 30 min incubation in the dark, 50 ul of 3N H 2 S0 4 was added to each well to terminate the reaction. Culture supernatants which contained antibody reactive with the bacteria coated plates were detected by measuring absorbance at 490 nm on a Dynatech MR 580 microELISA reader.

Culture supernatants from six transtormations were analyzed by the above method resulting in the identification of one well (7D7) which possessed activity on the E. coli and S. marcescens serotype plates, but not on the control plates, It was determined in Ssubsequent ELISA's with individual E. coli serotypes, that this well contained antibody reactive with at least, the E. coli serotypes: 08, No. 23504), and 075 No. 1.798), but not 04, 06:K2, 08:K8, 09:K9 or 022:K13 Nos. 12791, 19138, 23501, 23505 and 23517, respectively). Further, this well contained antibody reactive with the S. marcescens serotypes 012, 013 and 015, but not any other of the twenty known S. marcescens LPS serotypes.

D. Cloning of Specific Antibody Producing Cells The cells in well 7D7 were subjected to several rounds of cloning (four) until all clonal supernatants assayed by the above ELISA procedure gave a positive reaction on E. coli serotypes 08 and 075 and on S. marcescens serotypes 012, 013 and 015. There was never an example when any clonal supernatant demonstrated segregation in its reactivity pattern, suggesting that the culture supernatant from well 7D7 possessed true intergenus cross-reactivity to the E. coli and S. marcescens serotypes set forth and did not -26contain more than one cell line (each demonstrating individual serotype reactivity). Cells were cloned by limiting dilution in round-bottom 96-well plates in the absence of feeder cells. Media consisted of Iscove's S 5 medium containing 15% v/v FBS, L-glutamine (2 mM/1), sodium pyruvate (1 mM/1), penicillin (100 IU/ml), and streptomycin (100 pg/ml). Cultures were fed every three days by reF .acement of half the supernatant with fresh media. In general, wells were of sufficient lymphoblastoid cell density between 2 and 3 weeks postplating for analysis of anti-E. coli and S. marcescens serotype specificity.

Thus, in this experiment one cloned transformed human cell line was achieved which is continuous (immortal) and secretes a human monoclonal antibody to a determinant on the surface of the E. coli and S.

oo marcescens serotypes set forth.

0 o .o Prior to filing of this patent application, the continuous transformed human cell line identified 20 as 7D7 was deposited with the American Type Culture .o Collection, Rockville, MD as A.T.C.C. No. CRL 9009.

E. Further Characterization of Intergenus Cross-Reactivity Antibody from the cloned 7D7 cell line was assayed for further intergenus cross-reactivity by a o modification ot the standard immunoblotting technique.

Specifically, cross-reactivity to the bacteria K.

pneumoniae, E. aerogenes, and E. cloacae was inves- 0 30 tigated by spotting bacteria onto a gridded nitrocellulose paper disc, reacting the disc containing the bacteria with said antibody, and developing the antibody reactions with an alkaline phosphatase/nitroblue tetrazolium enzyme system. Briefly, 1.0 pl of bacteria

(O.E.

660 0.4) was spotted per grid section of a.

nitro-cellulose paper disc (Schleicher and Schuell, 37 mm nitrocellulose disc, gridded, 0.45 pm). Each disc c -27can conveniently hold 60 different bacterial samples.

The spotted discs were air-dried, fixed in 25% v/v isopropanol for 30 min, and blocked for 10 min in the non-fat dry milk reagent as described for the ELISA method. The blocked discs were \ashed three times for min each in PBS/Tween 20 and were transferred to the lids of 35 x 10 mm tisrue culture dishes. The antibody containing supernatant (1.0 ml) was added to the lid and was incubated at RT for 60 min. Following three 1 min washes in PBS/Tween 20, 1-2 ml of 1:1000 diluted (PBS) alkaline phosphatase conjugated goat-anti-human immunoglobulin (TAGO, Burlingame, CA) was added for min at RT. The discs were washed as above, and were submerged in 1-2 ml of fresh substrate prepared as follows: 16.5 mg of bromo-chlorindolylphosphate and o o o mg nitroblue tetrazolium were dissolved in 50 ml of alkaline phosphatase buffer (0.1 M Tris-HC1, pH with 0.1 M NaCl and 5 mM MgCl 2 the solution was kept in the dark, and filtered immediately before using.

After appropriate color development (10-15 min), the reaction was quenched by rinsing the disc in several changes of distilled water. The developed discs can be stored after drying.

The discs contained; 50 K. pneumoniae cap- S 25 sule-typed reference strains obtained from Dr. George Kenney, University of Washington, Department of Pathobiology, Seattle, WA and the American Type Culture Cola" lection, 4 E. aerogenes clinical blood isolates, and 6 E. cloacae clinical blood isolates (blood isolates obtained from Harborview Hospital, Seattle, WA). The Enterobacter isolates were typed (speciated) using the API 20E System of 23 standardized biochemical tests (API Analytab Products, Plainview, NY). This method identities the genus and species of gram-negative bacteria, but not the serotype. Therefore, the Enterobacterj unlike the E. coli, S. marcescens, and K.

-28pneumoniae, are not identified as to serotype, but only as to genus and specie.

From these experiments, the 7D7 antibody was observed to possess further intergenus crossreactivity. This antibody reacted with the following serotypes: K. pneumoniae E. coli marcescens E. aerogenes K14,57,60 08,75 012,13,15 Clinical Isolates Thus, the human monoclonal antibody 7D7 was observed to possess intergenus cross-reactivity to bacteria belonging to the species E. coli, S. marcescens, K.

pneumoniae, E. aerogenes but not E. cloacae.

F. Characterization of Monoclonal Antibodies no., 15 The finding that the monoclonal antibody .o 0 ,cross-reacted with several different bacterial genera suggested the antibody was directed against a shared 0o protein or carbohydrate. These two molecular species have been shown to account for intragenus cross-reactions (Mutharia, L. and Hancock, "Characterization of Two Surface-Localized Antigenic Sites on Porin Protein F of Pseudomonas aeruginosa," Canadian J. of Microbiol., 31:381-386, (1985) and Orskov, F. and S Orskov, "Serotyping of Escherichia Coli," in Meth- S 25 ods in Microbiology, Vol. 14, Bergan, ed. Academic Press, Orlando, FL, 43-112 (1984)).

Biochemical characterization of the molecular species recognized by the 7D7 antibody was accomplished by immunoblot analysis. Briefly, washed bacteria from onuu- 30 a 20 ml overnight broth culture (for E. coli, S. marce- Sscens, and E. aerogenes serotypes) or from overnight grown plates pneumoniae) were extracted in 1.0 ml of a solution containing 64 ml of 50 mM Tris pH 7.6, ml of glycerol, 0.3 gm of deoxycholate (sodium salt), 0.14 ml beta-mercaptoethanol, and 6 ml deionizedwater (Schechter, I. and Block, "Solubilization and Purification of trans-Formesyl Pyrophosphate-Squalene Lr r.

t) -29- Synthetase," J. Biol. Chem., 246:7690-7696, (1971)).

After 18 hour incuba:ion at 4 0 C, the suspension was centrifuged at 10,000 x g for 10 min. The supernatant was removed and the protein was quantitated using the c Bio-Rad Protein Assay (Bio-Rad Laboratories, Richmond, CA). Between 100 and 1000 ng of protein (varies for each bacterial extract) from each bacteria were each subjected to sodium dodecyl sulphate (SDS) polyacrylamide gel electrophoresis (Kusecek, et al., "Lipopolysaccharide, Capsule, and fimbriae as Virulence Factors Among 01, 07, 016, 018 or 075 and K1, K5 or K100 Escherichia coli," Infection and Immunity, 43:368-379 (1984)). Separated molecular species were transferred t from the gel to a nitrocellulose membrane (NCM) as described elsewhere (Towbin, et al., "Electrophoretic Transfer of Proteins From Polyacrylamide Gels to Nitroocellulose Sheets: Procedure and Some Applications," o Proc. Natl. Acad. Sci., 76:4350-4354, (1979)) and the NCM blot blocked for 1 hour in PBS-Tween (Batteiger, B, 20 et al., "The Use of Tween 20 as a Blocking Agent in the Sc, Immunological Detection of Proteins transferred to Nitrocellulose Membranes," J. Immunol. Meth. 55:297- 307, (1982)). The blot was incubated for 1 hour at RT eS 6 in 10 ml of spent culture supernatant from the 7D7 o o 25 line. Following four 5 min rinses in PBS-Tween, the blot was incubated in goat anti-human Ig conjugated to 0a o alkaline phosphatase and developed as described herein o for the bacteria-nitrocellulose disc assay. Positive reactions were noted in all tracks that contained deo 30 oxycholate extracts of the bacteria described herein.

co 0 2 In these tracks the 7D7 antibody appeared to recognize a series of regularly spaced molecular entities giving rise to a ladder-like pattern on the immunoblot. This profile was entirely consistent with that seen in polyacrylamide gel electrophoretic analysis of LPS inthe presence of SDS, where it has been demonstrated that the heterogeneous size profile exhibited by the bands I _r is due to a population of LPS molecules differing by weight increments equivalent to the number of O-antigenic oligosaccharide side chain units present per molecule (Pavla, E.T. and Makela, "Lipopolysaccharide Heterogeneity in Salmonella typhimurium Analyzed by Sodium Dodecyl Sulfate/Polyacrylamide Gel Electrophoresis," Eur. J. Biochem., 107:137-143, (1980) and Goldman, R.C. and Leive, "Heterogeneity of Antigenic-Side-Chain Length in Lipopolysaccharide from Escherichia coli 0111 and Salmonella typhimurium LT2, Eur. J. Biochem., 107:143-154, (1980)). These data indicate that the monoclonal antibody 7D7 is directed against an antigenic determinant shared by the LPS molecules found on some serotypes of E. coli, S. marcescens, K. pneumoniae, and E. aerogenes.

e To further define the molecular nature of the antigen, the deoxycholate extracts were treated, prior u 0 a o to their electrophoresis, with the proteolytic enzyme o Proteinase K (Eberling, et al., "Proteinase K from °o c '-420 -i-<iachium album i-mbe,-- e U. l.

20 alu S 4(1974)) To prepare the sample, 10 pg of sample protein and he mixture was heated to 65 0 C for 6U min.

The samples were electrophoresed and immunoblotted as described herei. The immunoblot patterns observed C, 0 or 25 after Proteinase treatment were identical to those v patterns observed wi out treatment and thus suggest ao that the antigen reacti e with the 7D7 antibody is not

S

t protein in nature.

To specifically a ress whether 7D7 reacted 00 30 with a carbohydr:-.e epitope, t electrotransferred oa" deoxycholate sample was subjecte to mild periodate oxidation prior to reacting the nit cellulose paper with antibody. This reaction has been hown to destroy carbohydrate determinants, and thus their subsequent reactivity with antibody, without alterin proteiin or lipid epitopes (Woodward, et al., "Det ction of i lona ndrat i 0 Tritirachium album Limber," Eur. J. Biochem. 47:91-97 (1974)). To prepare the sample, 10ug of proteinasek was added to 50mg of sample protein and the mixture was heated to 65 C for 60 min. The samples were electrophoresed and immunoblotted as described herein. The immunoblot patterns observed after Proteinase K treatment were identical to those patterns observed without treatment and thus suggest that the antigen reactive with the 7D7 antibody is not protein in nature.

To specifically address whether 7D7 reacted with a carbohydrate epitope, the electrotransferred deoxycholate sample was subjected to mild periodate oxidation prior to reacting the nitrocellulose paper with antibody. This reaction has been shown to destroy carbohydrate determinants, and thus their subsequent reactivity with antibody, without altering protein or lipid epitopes o o (Woodward, et al., "Detection of Monoclonal Antibodies Specific for Carbohydrate o Gl o 0 0 0 0 0 o0 0 o 000 0 00 3 0 So 006 00 0 o i _IY~ __W_1 -31- *004 00 o o 0 a0 0 0 0 0 0 0 0000 D eo o 0 00 oo 0 0 S000 o 0 00 0 Epitopes Using Periodate Oxidation," J. of Immunol.

Methods, 78:143-153, (1985)). Briefly, after the electro-blotted nitrocellulose paper containing the electrophoresed sample was blocked with PBS-Tween, as described herein, the paper was rinsed with 50 mM acetic acid sodium acetate pH 4.5 buffer. The nitrocellulose paper was incubated in 50 mM periodic acid dissolved in the acetic acid buffer for 60 min in the dark at RT. The treated paper was rinsed three times in PBS-Tween and reacted with the antibody as described herein. Electro-blotted deoxycholate extracts treated in this manner were no longer reactive with the 7D7 monoclonal antibody. These data strongly indicate that the epitope recognized by this antibody is a carbohydrate present on the LPS molecule of the bacteria described here.

The isotype of the 7D7 monoclonal antibody was determined in an ELISA procedure similar to the specificity tests described above except that biotinyl- 20 ated goat anti-human IgG (gamma-chain specific, TAGO) or biotinylated goat anti-human IgM (mu-chain specific, TAGO) was used as the second step reagent instead of the more broadly reactive biotinylated goat anti-human Ig. Both reagents were used at a 1:500 dilution and 25 the antigen plate contained pools of PLL'immobilized E.

coli 08 and 075 strains. Positive reaction of the 7D7 monoclonal antibody with the E. coli strains was observed only with the anti-IgM reagent, demonstrating an IgM isotype for the monoclonal antibody. It will be appreciated by those skilled in the art that if the above process were repeated several times and the isotypes of intergenus cross-reactive moncclonal antibodies so obtained were determined, one would find additional IgM and IgG isotypes (Frosch, et al;, "NZB Mouse System For Production of Monoclonal Antibodies to Weak Bacterial Antigens: Isolation of an IgG Antibody to the Polysaccharide Capsules of I; 1 II 1-

-I_

-32rroup (3 Escherichia coli K1 and G~epMeningocci," Proc. Natl.

Acad. Sci., 82:1194-1198, (1985)).

G. In Vitro Activity ::In vitro functional activity of the 7D7 monoclonal antibody was examined in an in vitro opsonophagocytosis assay which compared the bacteriocidal activity of the antibody in the presence and absence of both human neutrophils and human complement.

Bacteria were prepared by either inoculating ml of tryptic soy broth (TSB) with 50 pl of an overnight broth culture (for E. coli, S. marcescens, and E.

aerogenes) or for K. pneumoniae, streaking a petri dish containing Worfel-Ferguson Agar. For broth cultures, 15 the tubes were incubated at 37 0 C on a shaker for 3 0 "o0 hours at which time 1.5 ml of the culture was o 00 Scentrifuged for 1 min at 10,000 x g, the spent culture 000 00 .o o media discarded, and the pellet was suspended in 3.5 ml 0 00 of Hank's balanced salt solution containing 0.1% S 0"o gelatin and 5 mM HEPES (HBSS/Gel). For the agar plate 20 o o' grown bacteria, the colonies were scraped off the plate o- into sterile HBSS/Gel. The bacterial concentrations for bacteria grown under both conditions were adjusted °a to 3 x 10 bacteria/ml by measuring the O.D.660 and S 660 o 25 making the appropriate dilutions (approximately 1:50,000). Human neutrophils were isolated according to van Furth and Van Zwet ("In Vitro Determination of ,Phagocytosis and Introcellular Killing by Polymorphonuclear and Mononuclear Phagocytes," in Handbook of Experimental Immunology, Vol. 2, D.M. Weir, ed., 2nd S edition, Blackwell Scientific Publications, Oxford, 36.1-36.24 (1973)) with several modifications. Buffy coat from 10 ml of heparinized blood was underlayed with Ficoll-Pacque and centrifuged. The red blood cell (RBC) pellet was washed once with RPMI 1640 medium and resuspended in an equal volume of 37 0 C PBS. Three ml tV of this suspension was added to 6 ml of 2% dextran (in *o -33ooo o 0C 0 o oa o 0 0 oo a o a o0 0 0 C 00 0 37 0 C PBS) and the contents gently but thoroughly mixed end over end. Atter a 20 min incubation at 37 0 C to allow the RBC's to sediment, the supernatant (containing neutrophils) was removed, washed twice in 4°C PBS, S once in HBSS/Gel, and suspended in same to 5x10 7 neutrophils/ml. For the complement source used with E.

coli and S. marcescens, human serum was twice adsorbed with live bacteria pools (Bjornson, A.B. and Michael, "Factors in Human Serum Promoting Phagocytosis of Pseudomonas aeruginosa I. Interaction of Opsonins with the Bacterium," J. of Inf. Dis., 130Suppl:Sl19-S126 (1974)) corresponding to the organisms used in the assay. This serum was further adsorbed with boiled Zymosan (Bjornson, supra) to remove the serum component properidin, a molecule necessary for the activation of the alternate complement pathway. For opsonophagocytic assays using K. pneumoniae and E. aerogenes, the complement source was unadsorbed normal human serum used at 1% final concentration.

Plates used to quantify the number of surviving/destroyed bacteria were prepared beginning with warming of 24 well plates at 37 0 C for 3-5 hours. A 0.4% solution of agarose in TSB was prepared by autoclaving the mixture for 5 min and allowing it to 25 cool to 50°C in a water bath. Approximately 15 min before the end of the final incubation period in the opsonophagocytosis assay, a 24 well plate was removed from the 37 0 C incubator, placed on a 42°C hot plate and 0.4 ml of TSB/agarose was added to each well. The plate was immediately returned to the 37 0 C incubator such that the agarose never cooled below 37°C.

For the assay, 25 pl of 7D7 culture supernatant and 25 ul of an appropriate bacterial strain were added in duplicate to 96 well round bottom microtiter plates and incubated at RT for 30 min. This was iollowed by the addition of 15 ul of human complement, pl of human neutrophils (5 x 10 6 and 70 ol -34o- 0 0 0 o .a 0 oooo o' o o 00 S 0 .0 90 So0 O O Q u0 o HBSS/Gel. The entire surface of the plate was wiped with a sterile cotton swab, an adhesive plastic plate sealer was applied to securely cover the entire plate and interwell areas, and the plate was rotated at 37 0

C

for 1 hour. After incubation, the plate was centrifuged for 5 min at 100 x g, the plate sealer was gently removed, and the plate surface was dried with a sterile cotton swab dipped in 70% ethanol. Fifty microliters was removed from each microtiter well and was added to individual wells of the 24 well quantitation plates which already contained the 0.4m\/well of melted (38 0 -40 0 C) 0.4% TSB/agarose. These plates were placed on a flatbed shaker for 1 min at 150 RPM and the agarose was allowed to harden for 15 min at RT. Final- 15 ly, a 0.4 ml TSB/agarose overlay was added to each well, followed by a hardening period of 15 min at 4 0

C

before the plates were incubated overnight at 37 0

C.

After 18 hours the colonies were enumerated and the data was reported as colony forming units (CFU) for each condition.

The bacterial serotypes used herein, except K. pneumoniae, were only inactivated in the presence of monoclonal antibody 7D7, an active complement source, and human neutrophils (Table When this experiment 25 was repeated with several non-7D7 reactive bacterial serotypes, no bacteria des.:ruction was observed (data not shown), thus demonstrating the functional specificity of monoclonal antibody 7D7 and its capacity to opsonize bacteria and promote their phagocytosis.

Since the combined actions of opsonins (specific antibodies) and polymorphonuclear leukocytes (neutrophils) appeared to be the primary mechanism for immunity to these bacterial strains, these data suggested that antibody 7D7, would, after appropriate administration, provide protection against lethal challenges with. the bacteria serotypes described herein.

-r

TABLE

Bacteria Neutrophils Antibody Destruction of Complement Input Bacteria o 0 0000D no co o cc C, Cr CCc u 00n 00 0 E. coli 08 and 075 E. coli 08 and 075 E. coli 08 and 075 E. coli 08 and 075 S. marcescens 012 and 015 S. marcescens 012 and 015 S. marcescens 012 and 015 15 S. marcescens 012 and 015 E. aerogenes Isolates (2) E. aerogenes Isolates (2) E. aerogenes Isolates (2) E. aerogenes Isolates (2) 7D7 6 Fllb 7D7 7D7 6F11 6F11 zD a heat-inactivated (56°C for 30 min) human complement.

(6F11) culture supernatant containing an IgM human monoclonal antibody to Pseudomonas aeruginosa Fisher type 2.

H. In Vivo Activity To test the above hypothesis, animal protection studies were performed with the 7D7 antibody and at least one organism from the E. coli and E. aerogenes species described herein. 7D7 and negative control (6F11, human IgM monoclonal antibody specific to Pseudomonas aeruginosa Fisher immunotype 2) antibodies were first concentrated from spent culture supernatants by precipitation with solid ammonium sulphate r- rr=;POIEl~~ -36final concentration) (Good, A.J. et al., "Purification of Immunoglobulins and Their Fragments," in Selected Methods in Cellular Immunology, Mishell, B.B. and Shiigi, eds., W.H. Freeman and Company, San Francisco, CA (1980) 279-286). Precipitated material Swas reconstituted in sterile water and extensively dialyzed against PBS.

The IgM antibody in the ammonium sulphate salt precipitate was purified by affinity chromatography on a murine monoclonal anti-human IgM antibody affinity column. To prepare the column, one gram of dehydrated cyanogen bromide activated Sepharose 4B (Pharmacia) was mixed with 15 ml ice cold 1 mM HC1 in distilled water. The hydrated gel was washed in 30 ml 15 coupling buffer (0.1 M carbonate (NaHCO 3 in 0.5 M oo NaC1, pH drained to form a moist cake and was combined with the ammonium salt precipitate dissolved o in 1-3 ml of coupling buffer. The gel suspension was mixed end-over-end for 2 hr at RT and subsequently ceno) 20 trifuged at 200 x g for 5 min. To block still avail- .o 0 able reactive sites, the supernatant was removed, 10 ml of 1 M ethanolamine was added to the gel, and mixing was continued as above. The suspension was centrifuged o 0 at 200 x g for 5 min and the supernatant was discarded.

2o2o 5 The gel was prepared for use with 1 wash'in 0.1 M acetate/saline buffer (6.8 g sodium acetate trihydrate and o 14.6 g NaCl were dissolved in 500 ml distilled water containing 2.9 ml glacial acetic acid, pH two washes in coupling buffer, and two washes in PBS. The 30 gel was poured into a Pharmacia C10/10 column and 0 a stored at 4 0 C until use.

To purify the immunoglobulin, 0.5 ml of salt fractionated material was diluted to 2.0 ml in PBS and was added to the affinity column. Following sample loading, the column was washed with PBS, pH 8.0 until the absorbancy monitor indicated no further protein in the flow-through. The bound antibody was eluted with 2 r -37- M MgCl2in PBS, the protein concentration was determined for each fraction at O.D.

280 and the peak fractions pooled. The antibody containing fraction was desalted on a G-25 Sephadex column and, if necessary, was concentrated by microconcentration centrifugation (Centricon 30, Amicon Corp., Danver, MA) to 1-2 mg/ml.

The final preparation was tested for purity by SDSpolyacrylamide gel electrophoresis followed by silver nitrate staining of proteins (Morrissey, "Silver Stain for Proteins in Polyacrylamide Gels: A Modified Procedure with Enhanced Uniform Sensitivity," Anal.

Biochem., (1981) 117:307-310), and for ar"'body activity by ELISA as stated herein.

For each bacteria challenge, female, outbred 15 Swiss-Webster mice weighing between 20 and 22 gm were divided into three groups of ten mice each. A representative experiment was performed as follows: 0 0 0 0 0 o o o 0 0 0 o 00 so 0 2 Group

A

B

C

Bacteria E. coli 08 E. coli 08 S. marcescens 014 Antibody 7D7 6F11 7D7 Each group receiving antibody was injected intravenously with 200 pl of sterile PBS containing 25 ug of purified antibody. Two hours later, all animals were challenged intraperitoneally with 300 pl of live bacteria containing 3 LD 5 0 of their respective bacterial strain. The bacterial suspension had been prepared from a broth culture in logarithmic phase growth, from which the bacteria was centrifuged, washed twice in PBS, and resuspended to the appropriate concentration in PBS. Animals were observed for a period of five days. Twenty-four to forty-eight hours post-challenge all animals in Group B (irrelevant antibody) and Group C (irrelevant organism) were dead. In contrast, those

-I

-38animals that had received the 7D7 (Group A) antibody were all alive and symptom free.

This animal protection model was used to demonstrate the therapeutic effect of the 7D7 antibody against bacterial challenges with organ' "ms belonging to the three genera stated herein. A summary of the data is presented in Table 2.

TABLE 2 Survival/ Challenge Bacteria Antibody Challenge Survival E. coli 08 7

D

7 10/10 100 E. coli 08 6F11 0/10 0 S. -mareescens 0 1 4 7D7 0/10 0 a 6F11 antibody is specific to Pseudomonas aeruginosa Fisher immunotype 2 and serves as negative control antibody.

S. Marcescens 014 is not reactive with the 7D7 antibody and serves as a nonspecific control organism.

TABLE 2a Survivals Antibody 0 1 2 3 4 6F11 12 2 2 1 1 7D7 12 10 6 4 4 c LD 50 's and protection studies were performed with mice rendered neutropenic by in3ection with cyclo- 2 phosphamide as follows: Day 0 150 mg/kg, Day 2 50 mg/kg. On Day 4 mice received antibody and bacteria as described herein.

These data demonstrate that the human monoclonal antibody 7D7 is able to protect mice from lethal challenges with bacteria belonging to three different gram-negative bacteria species. secause 7D7 is reactive with a carbohydrate epitope present on LPE, but LPS molecules on K. pnoumoniae are less accessible (Orskov, I. and Orskov, "Serotyping of Klebsiella," in Methods in Microbiology, Vol. 14, Bergan, Academic Press, Orlando, PL (1984) 143-146) protection i -I p4Yuiii-*ii -39by the 7D7 antibody against K. pneumoniae infections was not evident (data not shown), Nonetheless, the intergenus cross-reactive human monoclonal antibody 7D7 was able to afford protection with 25 pg of purified 5 antibody against infection by organisms belonging to the gram-negative bacteria species E. coli and E.

aerogenes.

EXAMPLE II Example II demonstrates methods for the production and selection of a human monoclonal antibody that possesses intergenus cross-reactivity against members of the species Serratia marcescens, Klebsiella pneumoniae and Enterobacter aerogenes. Further, this example demonstrates the in vitro opsonic activity of said antibody against homologous S. marcescens, K.

c pneumoniae, and E. aerogenes serotypes. The process of Example I (essentially as described in parts A through G) was repeated to produce a human monoclonal antibody that was cross-protective against infections caused by the bacteria described herein, except that it was necessary to make specific modifications to characterize and assay the antibody described in this Example. The following are changes in assay procedures and the results obtained with the monoclonal antibody described herein.

1. Culture supernatants from six transformations were analyzed by the above method resulting in o 30 the identification of one well (4F10) which possessed activity on the S. marcescens serotype plate, but not on the E. coli or control plate. It was determined in subsequent ELISA's with individual S. marcescens serotypes, that this well contained antibody reactive w 4 th the S. marcescens serotypes 015 and 018, but notany other of the twenty known S. marcescens LPS serotypes.

/j

-N

Prior to the filing of this patent application, the continuous transformed human cell line identified as 4F10 was deposited with the American Type Culture Collection, Rockville, MD as A.T.C.C. No. CRL 9007.

2. Antibody from the cloned 4F10 cell line was assayed for further intergenus cross-reactivity by a modification of the standard immunoblotting technique. Specifically, cross-reactivity to the bacteria K. pneumoniae, E. aerogenes, and E. cloacae was investigated by spotting bacteria onto a gridded nitrocellulose paper disc, reacting the disc containing the bacteria with said antibody, and developing the antibody reactions with an alkaline phosphatase/nitroblue o0,9n tetrazolium enzyme system.

o From these experiments, the 4F10 antibody was observed to possess further intergenus cross-reactivo o ity. This antibody reacted with the following seroo 20 types: o, K. pneumoniae S. marcescens E. aerogenes o K3,12,29,31,68,72 015,18 Clinical Isolates Thus, the human monoclonal antibody 4F10 possessed inoo tergenus cross-reactivity to bacteria belonging to the S 2i species S. marcescens, K. pneumoniae, E. aerogenes.

3. Using the immunoblot technique, positive 'o cS reactions were noted in all tracks that contained deoxycholate extracts of the bacteria described herein.

In these tracks, the 4F10 antibody appeared to recogoa 0 nize either a broad band of components or a series of regularly spaced molecular entities giving rise to a ladder-like pattern. This profile was entirely consistent with that seen in polyacrylamide gel electrophoretic analyses of carbohydrate moieties that demonstrate either extensive molecular weight heterogeneity due to an often repeating specific sugar sequence or to LPS i -41o 0 0 U 0 0 on a o o 0o 0 0oo 00 0 0 0 Q 0 molecules differing by weight increments equivalent to the number of O-antigenic oligosaccharide side chain units per molecule (Vimr, et al., "Use of Procaryotic-Derived Probes to Identify Poly (Sialic Acid) in Neonatal Neuronal Membranes," Proc. Natl. Acad.

Sci., (1983) 81:1971-1975; and Holden, K.G. et al., "Gel Electrophoresis of Mucous Glycoproteins, I. Effect of Gel Porosity," Biochemistry (1971) 10:3105-3109).

These data indicate that the monoclonal antibody 4F10 is directed against an antigenic determinant shared by carbohydrate molecules found on some serotypes of S.

marcescens, K. pneumoniae, and E. aerogenes.

The immunoblot patterns observed after Proteinase K treatment were identical to those patterns observed without treatment and thus suggest that the antigen reactive with the 4F10 antibody is not protein in nature.

To specifically address whether 4F10 reacted with carbohydrate epitope, the electrotransferred de- 20 oxycholate sample was subjected to mild periodate oxidation prior to reacting the nitrocellulose paper with antibody. Electro-blotted deoxycholate extracts treated in this manner were no longer reactive with the 4F10 monoclonal antibody. These data strongly indicate that 25 the epitope recognized by this antibody is a carbohydrate moiety present on molecules possessed by the bacteria described herein.

4. The isotype of the 4F10 monoclonal antibody was determined in an ELISA pro'eiure similar to the specificity tests described in Example I except that the antigen plate contained a pool of PLL immobilized S. marcescens 015 and 018 serotypes. Positive reaction of the 4F10 monoclonal antibody with the S.

marcescens serotypes was observed only with the anti- IgM reagent, demonstrating an IgM isotype for the monoclonal antibody. It will be appreciated by those 18Ll~ i -42skilled in the art that if the process of this example were repeated several times and the isotypes of intergenus cross-reactive monoclonal antibodies so obtained were determined, one would find additional isotypes, IgM and IgG isotypes.

In vitro functional activity of the 4F10 monoclonal antibody was examined in an opsonophagocytic assay which compared the bacteriocidal activity of the antibody in the presence and absence of both human neutrophils and human complement.

The bacterial serotypes used herein were only inactivated in the presence of monoclonal antibody 4F10, an active complement source, and human neutrophils (Table When this experiment was repeated with several non-4F10 reactive bacterial serotypes, no i bacterial destruction was observed (data not shown) thus demonstrating the functional specificity of monoclonal antibody 4F10 and its capacity to opsonize bac- 20 teria and promote their phagocytosis.

0o 0 0000OD 0 00 o 00 00 u n n 0o 00 0 V 1 0 (0 0 r i -I~-III-IP~UY~F -43- TABLE 3 Bacteria Neutrophils Antibody Destruction of Complement Input Bacteria S. marcescens 018 and 015 S. marcescens 018 and 015 S. marcescens 018 and 015 S. marcescens 018 and 015 K. pneumoniae K3 and K12 K. pneumoniae K3 and K12 K. pneumoniae K3 and K12 K. pneumoniae K3 and K12 E. aerogenes Isolates (2) 20 E. aerogenes Isolates (2) E. aerogenes Isolates (2) E. aerogenes Isolates (2) 6Fllb 7D7 Lito 6F11 -AD-7-- -77 4 m- 0 0 oooa 0 0 o 0 o 0 00 0 o 0 *0 0 0 o 0 U 0 0u 9

U

'-7 6F11 4 4W? heat-inactivated (56°C for 30 min) human complement.

(6F11) culture supernatant containing an IgM human monoclonal antibody to Pseudomonas aeruginosa Fisher type 2.

EXAMPLE III Example III demonstrates methods for the production of a human monoclonal antibody that is reactive with both the Escherichia coli capsular type Kl and Neisseria meningitidis meningitidis) Group B polysaccharide and further demonstrates the protective activity of said antibody in vivo against a lethal challenge of homologous E. coli and N. meningitidis ii kV wvv V'

L

o oe o on n 0 00 0 00 o B 0Oc o ,0 0 00 0oo 0 C 0 0 0 0 -44bacterial species. The process of Example I (essentially described in parts A through G) was repeated to produce a human monoclonal antibody that was cross-protective against infections caused by the bacteria described herein, except that it was necessary to make specific modifications to characterize and assay the antibody described in this Example. The following are changes in assay procedures and the results obtained with the monoclonal antibody described herein.

1. Culture supernatants from five transformations were analyzed by the above method resulting in the identification of four wells (5D4, 2C10, 9B10, and 8A8) which contained anti- E. coli specificity on the E. coli serotype plate, but not on the S. marcescens or control plates. It was determined in subsequent ELISA's, conducted as set forth, on individual E. coli serotypes, that these wells contained antibody reactive with, at least, the E. coli serotypes: 01 (ATCC 20 23499), 07:K1 (ATCC 12792), 016:K1 (ATCC 23511), and 050 (CDC 1113-83), but not 04 (ATCC 12792), 06:K2 (ATCC 19138), 08:K8 (ATCC 23501), 09:K9 (ATCC 23505), or 022:K13 (ATCC 23517). Due to its better performance during the cloning procedure and increased antibody 25 production, the 9B10 monoclonal antibody'was selected for further analysis.

Prior to filing ot this patent application the continuous transformed human cell line identified herein as 9B10 was deposited in the American Type Culture Collection, Rockville, MD as A.T.C.C. No. CRL 9006.

2. The finding that the monoclonal antibodies from each of the clones reacted with the identical group of E. coli 0-antigen serotypes, indicated that these antibodies were directed against a bacterial surface structure common to these serotypes. Several I _I 0o 0 0 a0 o 00 O 0 4 000 0n 0 ro approaches were used to define the surface structure common to these E. coli serotypes. As set forth, two (07:K1 and 016:Kl) of the four E. coli serotypes identified by the 9B10 antibody possessed the K1 capsular serotype, while the other two (01 and 050) had not been typed for their K-antigen serotype. Thus, the possibility was pursued that the 9B10 antibody contained reactivity to the K1 antigen and that the E. coli strains possessing the 0-antigen serotypes 01 and 050 also possessed the K1 capsular serotype.

Others have taken advantage of the thermolability of the K1 capsule to establish its presence.

Heating of K1 positive E. coli serotypes in a boiling water bath at 100 0 C for 60 minutes removes the subsequent ability of these strains to react with anti-Kl sera and enhances their ability to react with anti-0 antigen sera (Orskov, F. and Orskov, "Serotyping of Escherichia coli," in Methods in Microbiology, Vol. 14, T. Bergan, ed., Academic Press, London (1984) pp. 44- 20 105). The reciprocal effects ot boiling are most likely due to the removal of the capsule and the increased accessibility of antibody for lipopolysaccharide (LPS) molecules. The E. coli K1 positive serotypes (07 and 016) and the non-Kl typed serotypes (01 and 050) were heated as set forth and reacted with the' 9B10 antibody and LPS serotype specific heterologous sera (Difco Bacto-E. coli Typing Reagents) in the ELISA procedure.

Heat treated organisms lost all reactivity to the 9B10 antibody and had increased their reactivity with their homologous LPS serotype specific sera, while nontreated (control) organisms remained strongly reactive with 9B10 culture supernatants and poorly reactive with their respective LPS serotype specific antisera.

The polysaccharide (carbohydrate) from Neisseria meningitidis Group B bacteria (a homopolymer of sialic acid, alpha 2, 8-linked poly-N-acetyl neuraminic acid) has been proven to be chemically and 1 I r i -46antigenically homogeneous with the E. coli K1 polysaccharide (Grados, 0. and Ewing, "Antigenic Relationship between Escherichia coli and Neisseria meningitidis," J. Immunol. (1973) 110:262-268). These data suggest that if the 9B10 monoclonal antibody contains specificity to E. coli K1 capsule then the antibody should also contain specificity to N. meningitidis Group B polysaccharide and further, that monoclonal antibodies containing specificity to the Group B polysaccharide of N. meningitidis should also demonstrate reactivity to E. coli strains possessing the K1 capsule. Two experimental protocols were used that tested for the ability of the 9B10 antibody to react with N. meningitidis and the ability of an an- 15 tibody against N. meningitidis Group B polysaccharide to react with the four E. coli 9B10 reactive serotypes.

n h Highly purified Group B polysaccharide (Connaught Laboratories, Toronto, Canada) and viable N.

ooo meningitidis Group B bacteria were reacted with the S0 9B10 antibody in an ELISA as set forth. The 9B10 mono- 0 20 0 clonal antibody strongly reacted against both antigen °preparations. To prove the converse specificity, a commercially available N. meningitidis Group B Meningioo tis Test Kit ("Directagen" Direct Antigen Detection 25 System, Hynson, Westcott, and Dunning, B'altimore, MD), Oo 25 a°o that utilizes latex spheres coated with a murine monoclonal antibody to the Group B polysaccharide, was Suused. In agglutination assays using the 9B10 positive E. coli serotypes, all four serotypes demonstrated strong react'vity with the antibody coated spheres. E.

coli serotypes known to be Kl-antigen negative were also negative in this test system. Collectively, these data indicate that the 9B10 monoclonal antibody is reactive with the E. coli Kl capsule and the type-speci- S fic carbohydrate on N. meningitidis Group B. Further, since ~ay of these assays were performed with intact, S viable bacteria, it can be inferred that monoclonal i __r ."LU /111) ana /u ii o0 0 o o oo 0 0 0 0 o 0 ,oo .0 4 00 4 0 0 I o 3 1 -47antibody 9B10 is specific for some portion of an externally exposed region of the poly-sialic acid molecule.

3. The isotype of the 9B10 monoclonal antibody was determined in an ELISA procedure similar to the specificity tests described in Example I except that the antigen plate contained a pool of PLL immobilized E. coli K1 positive serotypes. Positive reaction of the 9B10 monoclonal antibody with the K1 positive E.

coli serotypes was observed only with the anti-IgM reagent, demonstrating an IgM isotype for the monoclonal antibody. It will be appreciated by those skilled in the art that if the process of this Example were repeated several times and the isotypes of Kl-specific monoclonal antibodies so obtained were determined, one would find additional isotypes, IgM and IgG isotypes.

4. In vitro functional activity of the 9B10 20 monoclonal antibody was examined in an opsonophagocytic assay which compared the bacteriocidal activity of the antibody in the presence and absence of both human neutrophils and human complement.

K1 positive E. coli serotypes were only inactivated in the presence of monoclonal antibody 9B10, an active complement source, and human neutrophils (Table When this experiment was repeated with several K1 negative E. coli serotypes, no bacterial destruction assay was observed (data not shown) thus demonstrating the Kl specificity of monoclonal antibody 9B10 and its capacity to opsonize bacteria and promote their phagocytosis. Since the combined action of opsonins (specific antibodies) and polymorphonuclear leukocytes (neutrophils) appeared to be the primary mechanism for immunity to K1 positive E. coli serotypes, these data suggested that antibody 9B10, would, after appropriate administration, provide protection against a lethal i I ,rp a -48challenge with any E. coli K1 encapsulated serotypes, regardless of its O-antigen serotype.

TABLE 4 co a 0 0 a 4 a o 0 0 0 00 0 0 0 0 00 0 0 0 00 0 C 0 0 0 0 oa 0 0 00 0 0 0 00 0 o' o a o Bacteria Neutrophils Antibody E. coli 01:K1 9B10 and 018:K1 E. coli 01:K1 6 Fllb and 018:K1 E. coli 01:K1 9B10 and 018:K1 E. coli 01:K1 9B10 and 018:K1 Destruction of Complement Input Bacteria N. meningitidis Group B N. meningitidis Group B N. meningitidis Group B N. meningitidis 20 Group B 9B10 6F11 9B10 9B10 a heat-inactivated (56°C) for 30 min) human complement.

(6F11) culture supernatant containing an IgM human monoclonal antibody to Pseudomonas aeruginosa Fisher type 2.

5. To test the above hypothesis, animal 25 2 protection studies were performed with the 9B10 antibody and several K1 positive and K1 negative E. coli serotypes, as well as a N. meningitidis Group B serotype (strain H313, obtained from Dr. Carl Frasch, Laboratory of Bacterial Polysaccharides, Office of Biologics, Food and Drug Administration, Bethesda, MD).

For each bacteria challenge, female, outbred Swiss-Webster mice weighing between 20 and 22 gm were divided into three groups of ten mice each. A representative experiment was performed as follows: 1,

~_P

-49- Group

A

B

C

Bacteria E. coli K1 E. coli K1 E. coli K2 Antibody 9B10 6F11 9B10 Each group receiving antibody was injected intravenously with 200 Pl of sterile PBS containing 25 Pg of purified antibody. Two hours later, all animals were challenged intraperitoneally with 300 il of live bacteria containing 3 LD 5 0 of their respective bacterial strain. The bacterial suspension had been prepared from a broth culture in logarithmic phase growth, from which the bacteria was centrifuged, washed twice in PBS, and resuspended to the appropriate concentration in PBS. Animals were observed for a period of five days. Twenty-four to forty-eight hours post-challenge all animals in Group B (irrelevant antibody) and Group C (irrelevant organism) were dead. In contrast, those animals that had received the 9B10 (Group A) antibody were all alive and symptom free.

This animal protection model was used to demonstrate the therapeutic effect of the 9B10 antibody against bacterial challenges with organisms belonging to the two species stated herein. A summary of these data is presented in Table 0400 0 04D 000 0 'r 0 a'a -3: 0 u L i- the flow-through. The bound antibody was elutea witn z Challenge Bacteria E. coli K1 E. coli K1 E. coli K2 b N. meningitidis Group B N. meningitidis 'Group B E. coli K2 TABLE Antibody 9B10 6 Fll a 9B10 9B10 6F11 9B10 Survival/ Challenge 10/10 0/10 0/10 5/5 0/5 0/5 Survival 100% 0% 0% 100% 0% 0% 00 4 00 0 0J 0 00 00 0 0 ii 2 0 S.

0. 00 o ooo 0 i 00 0

CO

6F11 antibody is specific to Pseudomonas aeruginosa Fisher b immunotype 2 and serves as negative control antibody.

E. coli K2 is not reactive with the 9B10 antibody and serves as nonspecific control organisms.

These data demonstrate that the human monoclonal antibody 9B10 is able to protect mice from lethal challenges with bacteria belonging to two different gram-negative bacterial species. The intergenus cross-protective antibody was able to passively protect 20 against infection by organisms belonging to the gramnegative bacterial species E. coli and N. meningitidis Group B.

EXAMPLE IV 25 Example IV demonstrates methods for the production of a human monoclonal antibody that possesses intergenus cross-reactivity against members of the species Escherichia coli coli), Enterobacter cloacae cloacae) and Group B Streptococcus. Further this Example demonstrates an antibody cross-reactive with species belonging to the two main bacteria divisions; gram-negative coli and E. cloacae) and gram-positive (Group B Streptococcus). Even further, this Example demonstrates the in vivo protective activity of said antibody against a lethal challenge of homologous E. coli, and Group B Streptococcus serotypes. The process of Example I (essentially described in parts A

I!LII~I

o i O -o C '3O -51- through G) was repeated to produce a human monoclonal antibody that was cross-protective against infections caused by the bacteria described herein, except that it was necessary to make specific modifications to charac- S terize and assay the antibody described in this Example. The following are changes in assay procedures and the results obtained with the monoclonal antibody described herein.

1. Supernatants were screened for the presence of anti-Group B Streptococcus antibodies using an enzyme-linked immunosorbent assay (ELISA) technique as described in Example I. The antigen plates consisted of a series of flat-bottom 96-well Immunolon 2 microtiter plates, the wells of which contained mixtures of Group B Streptococci capsule types affixed to the well surfaces with poly-L-lysine (PLL). Various antigen plates used in the screen included: a mixture of Group B Streptococcus types Ia No. 12400), Ib No. 12401), Ic No. 27591); a mixture of types II No. 12973) and III (clinical isolate obtained from Dr. C. Wilson, Children's Orthopedic Hospital, Dept. Infectious Disease, Seattle, WA); and a microtiter plate with no bacteria.

Culture supernatants from two transformations were analyzed by the above method resulting in the identification of one well (4B9) which possessed activity on both Group B Streptococcus typing plates, but not the control plates. It was determined in subsequent ELISA's with individual Group B Streptococcus types, that this well contained antibody reactive with all five reference typing strains.

Thus, in this experiment one cloned transformed human cell line was achieved which is continuous (immortal) and secretes a human monoclonal antibody to a determinant on the surface of the Group B Streptococcus types set forth.

3 0 -52- Prior to filing of this patent application, the continuous transformed human cell line identified as 4B9 was deposited with the American Type Culture Collection, Rockville, MD as A.T.C.C. No. CRL 9008.

2. Antibody from the cloned 4B9 cell line was assayed for cross-reactivity to gram-negative and gram-positive bacteria by a modification of the standard immunoblotting technique. Specifically, crossreactivity to the bacteria E. coli, K. pneumoniae, S.

marcescens, E. aerogenes E. cloacae, Hemophilus influenzae, and Staphy 1 .us aureus was investigated by spotting bacteria o, a gridded nitrocellulose paper disc, reacting the disc containing the bacteria with -i'd antibody, and developing the antibody reacttuns with in alkaline phosphatase/nitroblue tetrazolium enzyme system (as described in Example I).

From these experiments, the 4B9 antibody was observed to possess cross-reactivity with particular gram-negative bacterial species. This antibody reacted with the E, coli LPS serotypes 04, 07, 018, and 025, and the E. cloacae clinical isolates. Thus the human monoclonal antibody 4B9 possesses intergenus crossroactivity between the gram-negative and gram-positive bacteria belonging to the species coli, E. cloacae, and Group B Streptococcus.

3. The finding that the monoclonal antibody cross-reacted with several different bacterial genera belonging to both gram-positive and gram-negative bacterial divisions, suggested the antibody was directed against a shared protein or carbohydrate. The biochemical characterization of the molecular species recognized by the 489 antibody was accomplished by immunoblot analysis. For analysis of the gram-qegative genera, washed bacteria were extracted in deoxycholate as described in EXAMPLE I. For the gram-positive r r; -53bacteria, 1.0 L of bacteria cultured for 6 hours in modified Todd-Hewitt Broth (Difco, Todd-Hewitt Broth containing 2.8 gm/L anhydrous sodium phosphate, pH 7.8) at 37 0 C were harvested by centrifugation and washed three times in PBS. The bacteria were resuspended in ml of protoplast medium (40% sucrose w/v in 0.03 M potassium phosphate buffer, pH 6.8 containing 10 mM MgCl 2 and the suspension was warmed to 37 0 C for min. Approximately 3000 units of the mutanolysin (SIGMA) were added and the mixture was shaken at 37°C for 90 min or until the OD 660 of the suspension had been reduced by The digested material was 04t centrifuged at 2000 x g for 15 min 9s"RT and the i

A

supernatant was dialyzed against PBS for 48 hr (Young, M.K. and Mattingly, "Biosynthesis of Cell Wall Pe-tidoglycan and Polysaccharide Antigens by Protoplasts of Type III Group B Streptococcus," J. Bact., a (1983) 154:211-220). The dialysate was concentrated o ten-fold by positive pressure dialysis through a filter (Amicon Corp., Danvers, MA).

Carbohydrates binding to wheat germ agglu- S' tinin were purified by affinity chromatography on a wheat germ lectin Sepharose 6MB column (SIGMA). The bound digest, described herein, was eluted from the column with 10 ml of 0.1 M N-acetylglucosamine and the eluate was dialyzed against distilled water at 4 0

C.

The affinity purified eluate was dried by lyophiliz4o tion and the dry weight of the resulting material was obtained (Gray, et al., "Interaction of Group B Streptococcal Type-Specific Polysaccharides with Wheat Germ Agglutinin and Other Lectins," J. or Immunol.

Meth., (1984) 72:269-277). Positive reactions were noted in all tracks that contained deoxycholate extracts of the bacteria described herein. In those tracks containing extracts from gram-negative bacteria, the 4B9 antibody appeared to recognize a series of regularly spaced molecular entities giving rise to a .r 4f A

T-

_s~ -54ladder-like pattern on the immunoblot. This profile was entirely consistent with that seen in polyacrylamide gel electrophoretic analysis of LPS in the presence of SDS, where it has been demonstrated that the heterogenous size profile exhibited by the bands is due to a population of LPS molecules differing by weight increments equivalent to the number of 0-antigenic oligosaccharide side chain units present per molecule (Pavla, E.T. and Makela, supra and Goldman, R.D.

and Leive, supra). In those tracks containing extracts from the Group B Streptococcus types, the 4B9 antibody appeared to recognize components present on a broad band. This profile was consistent with that seen in polyacrylamide gel electrophoresis analyses of carbohydrate moieties that demonstrate extensive molecular S weight heterogeneity with a frequently repeating specific sugar sequence (Vmir, E.R. et al., supra and S. Holden, supra). These data indicate that the monoclonal antibody 4B9 is directed against an antigenic determinant shared by molecules found on some serotypes of E. coli, E. cloacae, and Group B Streptococcus.

To further define the molecular nature of the antigen, the deoxycholate extracts were treated with proteolytic enzyme Proteinase K prior to'their electrophoresis (Eberling, supra). The immunoblot patterns observed after Proteinase K treatment were identical to those patterns observed without treatment and thus suggest that the antigen reactive with the 4B9 antibody is not protein in nature.

To specifically address whether 4B9 reacted with a carbohydrate epitope, the electrotransferred deoxycholate and wheat germ agglutination affinity purified samples were subjected to mild periodate oxidation prior to reacting the nitrocellulose paper with antibody (see EXAMPLE Electro-blotted deoxycholate extracts treated in this manner were no longer reactive 0 0 0f 3 00~i 0 0 0oi with the 4B9 monoclonal antibody. These data strongly indicate that that epitope recognized by this antibody is a carbohydrate moiety present in molecules possessed by both the gram-negative and gram-positive bacteria described herein.

4. The isotype of the 4B9 monoclonal antibody was determined in an ELISA procedure similar to the specificity tests described above except that the antigen plate contained a pool of PLL immobilized Group B Streptococcus types II and III. Positive reaction of the 4B9 monoclonal antibody with the Group B Streptococcus strains was observed only with the anti-IgM reagent, demonstrating an IgM isotype for the monoclonal antibody.

5. In vitro functional activity of the 4B9 monoclonal antibody was examined in an opsonophagocytic assay which compared the bacteriocidal activity of the antibody in the presence and absence of both human neutrophils and human complement.

The bacterial strains used herein were only inactivated in the presence of monoclonal antibody 4B9, an active complement source, and human neutrophils (Ta- 25 ble When this experiment was repeated with several non-4B9 reactive bacterial serotypes, no bacteria destruction was observed (data not shown) thus demonstrating the functional specificity of monoclonal antibody 4B9 and its capacity to opsonize bacteria and promote their phagocytosis. Since the combined actions of opsonins (specific antibodies) and polymorphonuclear leukocytes (neutrophils) appeared to be the primary mechanism for immunity to these bacterial strains, these data suggest that antibody 4B9, would, after appropriate administration, provide protection against lethal challenges with the bacteria strains described herein.

M

-56- TABLE 6 Destruction of Complement Input Bacteria Bacteria E. coli 018 and 025 E. coli 018 and 025 E. coli 018 and 025 E. coli 018 and 025 Neutrophils Antibody 4B9 6 Fllb nuo 3 a a "3 '2

D

II r t O 0 O 1 7 O ;1 U D

O

J

O

3 E. cloacae Isolates E. cloacae Isolates E. cloacae Isolates E. cl6acae Isolates Group B Strep.

Types Ia and III 20 Group B Strep.

Types Ia and III Group B Strep.

Types Ia and III Group B Strcp, Types la and III 6F11 4B9 4B9 4B9 6F11 heat-inactivated (56°C for 30 min) human b6F11) culture supernatant containing an IgM antibody to Pseudomonas aeruginosa Fisher type complement.

human monoclonal 2.

a 4 A 6. To test the above hypothesis, animal protection studies were performed with the 4B9 antibody and at least one organism from each genus described herein.

From each gram-negative bacteria challenge, female, outbred Swiss-Webster mice weighing between and 22 gm were divided into three groups of ten mice each. A representative experiment was performed as follows: U~___lq Group

A

B

C

-57- Bacteria E. coli 018 E. coli 018 S. marcescens 014 Antibody 4B9 6F11 4B9 o 00 0oo 0 0 0 00 0 0o o o0 0o t 00 o auov Each group receiving antibody was injected intravenously with 200 pl of sterile PBS containing 25 pg of purified antibody. Two hours later, all animals were challenged intraperitoneally with 300 ul of live bacteria containing 3 LD 50 of their respective bacterial strain.

The bacterial suspension had been prepared from a broth culture in logarithmic phase growth, from which the bacteria was centrituged, washed twice in PBS, and resuspended to the appropriate concentration in PBS.

Animals were observed for a period of five days. Twenty-four to forty-eight hours post-challenge all animals Group B (irrelevant antibody) and Group C (irrelevant organism) were dead. In contrast, those animals that had received the 4B9 (Group A) antibody, were all alive 20 and symptom free.

For the Group B Streptococcu, protection studies, a neonatal rat model was used. Outbred Sprague- Dawley rat pups (housed with their mothers), less than 48 hours old received antibody and bacteria 25 essentially as described for the mouse protection studies. Primary differences were as tollows: 1) both the antibodies and bacterial challenges were injected intraperitoneally, and 2) the inoculum size was reduced to 20 pl.

These animal protection models were used to demonstrate the therapeutic effect of the 4B9 antibody against bacterial challenges with organisms belonging to the three species stated herein. A summary of these data is presented in Table 7.

'7 0 0 u 0

_II_

-58- TABLE 7 Survival/ Challenge Bacteria Antibody Challenge Survival E. coli 025 4B9 10/10 100 E. coli 025 6F11 a 0/10 0 b 5 S. marcescens 014 4B9 0/10 0 Group B 4B9 10/10 100 Streptococcus Ia and III Group B 6F11 0/10 0 Streptococcus Ia'and III S. marcescens 014 9B10 0/10 0 2 10 a 6F11 antibody is specific to Pseudomonas aeruginosa Fisher b immunotype 2 and serves as negative control antibody.

S. marcescens 014 is not reactive with the 7D7 antibody and serve as a non-specific control organisms.

15 Thi.e data demonstrate that the human monoclonal antibody 4B9 is able to protect mice and rats from lethal challenges with bacteria genera belonging oo 0 to both gram-negative and gram-positive bacterial divi- O.0 sions. The intergenus cross-reactive human monoclonal Soo 20 antibody 4B9 was able to afford protection v'ith 25 pg of purified antibody against infection by organisms be- "o0 longing to the gram-negative bacteria genera E. coli V. and the gram-positive bacteria belonging to Group B Streptococcus.

0 EXAMPLE V Son EXAMPLE V demonstrates methods for the production and selection of a human monoclonal antibody 0: 0 that possesses intergenus cross-reactivity against memhers of the genera Serratia marcescens, Klebsiella pneumoniae, and Enterobacter aerogenes. Further,. this (,ow example demonstrates the in vitro opsonic activity of said antibody against homologous S. marcescens, K.

pneumoniae, and E. aerogenes serotypes. The process of Example I (essentially as described in parts A through G) was repeated, except that it was necessary to make specific modifications to characterize and assay the -59- 0 o o o o o 0no od 0 0 00 0 0 o 0 o0 0 antibody described in this Example. The following are changes in assay procedures and the results obtained with the monoclonal antibody described herein.

1. Culture supernatants from four transformations were analyzed by the above method resulting in the identification of one well (7E1G) which possessed binding activity on at least one of four K. pneumoniae serotype plates, containing the capsule serotypes; 1, 2, 3, 4, 6, 8, 9, 19, 20, 21, 24, 27, 31, 43, 44, and but not on the control plate (no bacteria). Antibody from the 7E10 cell line was assayed for further intergenus cross-reactivity by a modification of the standard immunoblotting technique. Specifically, cross-reactivity to the bacteria K. pneumoniae, E.

aerogenes, S. marcescens, E. coli, and P. aeruginosa was investigated by spotting bacteria onto a gridded nitrocellulose paper disc, reacting the disc containing the bacteria with said antibody, and developing the 20 antibody reactions with an alkaline phosphatase/nitroblue tetrazolium enzyme system.

From these experiments, the 7E10 antibody was observed to possess further intergenus cross-reactivity. This antibody reacted with the fol- 2 lowing species and serotypes: K. pneumoniae S. marcescens E. aerogenes K2,8,11,12,13,21, 04,12 Clinical Isolates 26,29,30,33,42,68,69 Thus, the human monoclonal antibody 7E10 possessed intergenus cross-reactivity to bacteria belonging to the genera K. pneumoniae, S. marcescens, and E. aerogenes.

2. The isotype of the 7E10 monoclonal antibody was determined in an ELISA procedure similar to 3 the specificity tests described in Example I except that the antigen plate contained a pool of PLL (poly-L-lysine) immobilized K. pneumoniae K8 and Kll i serotypes. Positive reaction of the 7E10 monoclonal antibody with the K. pneumoniae serotypes was observed only with the anti-IgM reagent, demonstrating an IgM isotype for the monoclonal antibody. It will be appreciated by those skilled in art that if the process of this Example were repeated '.veral times and the isotypes of intergenus cross-reactive monoclonal antibodies so obtained were determined, one would find additional isotypes, IgM and IgG isotypes.

3. In vitro functional activity of the 7E10 monoclonal antibody was examined in an opsonophagocytic assay which compared the bacteriocidal activity of the antibody in the presence and absence of both human neutrophils and human complement. The bacterial serotypes used herein were only inactivated in the presence of monoclonal antibody 7E10, an active complement source, and human neutrophils (Table When this experiment was repeated with serotypes unreactive with antibody 20 7E10, no bacterial destruction was observed (data not shown). These experiments demonstrated the functional specificity of monoclonal antibody 7E10, as well as its capacity to opsonize bacteria and promote their phagocytosis.

0 0 n O o o0 0 00 0 Ooa0 0 0 o O 0 0 0 00 0 O O0 0 uo JE C yd 00 0 -61- TABLE 8 Bacteria S. marcescens 012 S. marcescens 012 S. marcescens 012 S. marcescens.

012 K. pneumoniae K8 and Kll K. pneumoniae K8 and Kll K. _pneumnoniae K8 and K11 K. pneumoniae KS and K11 >eutrophils Antibody 7E10 Fll b 7E10 7E10 Destruction of Input Complement Bacteria -a 0 0 0 86% 7E10 6F11 7E10 7E10 7E10 6F11 7E10 7E1U 0 0 94% 0 (CO 00 0 0 00 00 0 S .0 00 0 C. 0 0,) 00 0.

0 000 0 00' 0 0 0 E. aerogenes Isolate E. aerogenes Isolate E. aerogenes 20 Isolate E. aerogenes Isolate a and b see Table 3 footnotes

C.

0 I -62o 0 .0

III"

*0i EXAMPLE VI EXAMPLE VI demonstrates methods for the production and selection of a human monoclonal antibody that possesses intergenus cross-reactivity against mem- S bers of the genera Serratia marcescens and Klebsiella pneumoniae. Further, this example demonstrates the in vitro opsonic activity ot said antibody against homologous S. marcescens and K. pneumoniae serotypes. The process of Example I (essentially as described in parts A through G) was repeated, except that it was necessary to make specific modifications to characterize and assay the antibody described in this Example. The following are changes in assay procedures and the results obtained with the monoclonal antibody described herein.

1. Culture supernatants from four transformations were analyzed by the above method resulting in the identification of one well (8C9) which possessed binding activity on at least one of four K. pneumoniae 20 serotype plates, containing the capsule serotypes: 1, 2, 3, 4, 6, 8, 9, 19, 20 21, 24 27, 31, 43, 44, and but not on the control plate (no bacteria). Antibody from the 8C9 cell line was assayed for further intergenus cross-reactivity by a modification of the standard immunoblotting technique. Specifically, cross-reactivity to the bacteria K. pneumoniae, E. aerogenes, S.

marcescens, E. coli, and P. aeruginosa was investigated by spotting bacteria onto a gridded nitrocellulose paper disc, reacting the disc containing the bacteria with said antibody, and developing the antibody reactions with an alkaline phosphatase/nitroblue tetrazolium enzyme system.

From these experiments, the 8C9 antibody was observed to possess further intergenus cross-reactivity. This antibody reacted with the following species and serotypes: K. pneumoniae S. marcescens K5,6,7,14,27,36,55,64 03

I

-63- Thus, the human monoclonal antibody 8C9 possessed intergenus cross-reactivity to bacteria belonging to the genera K. pneumoniae and S. marcescens.

2. The isotype of the 8C9 monoclonal antibody was determined in an ELISA procedure similar to the specificity tests described in Example I except that the antigen plate contained a pool of PLL immobilized K. pneumoniae K14 and K27 serotypes. Positive reaction of the 8C9 monoclonal antibody with the K. pneumoniae serotypes was observed only with the anti-IgM reagent, demonstrating an IgM isotype for the monoclonal antibody. It will be appreciated by those skilled in the art that if the process of this example were repeated several times and the isotypes of intergenus cross-reactive monoclonal antibodies so obtained were determined, one would find additional isotypes, e.g., IgM and IgG isotypes.

3. In vitro functional activity of the 8C9 monoclonal antibody was examined in an opsonophagocytic assay which compared the bacteriocidal activity of the antibody in the presence and absence of both human neutrophils and human complement. The bacterial serotypes used herein were only inactivated in the presence of monoclonal antibody 8C9, an active complement source, and human neutrophils (Table When this experiment was repeated with serotypes unreactive with antibody 8C no bacterial destruction was observed (data not shown). These experiments demonstrated the functional specificity of monoclonal antibody OC9, as well as its Scapacity to opsonize bacteria and promote their phagocytosis.

SIIII l l l 1" ll -64- TABLE 9 Destruction of Input Bacteria Neutrophils Antibody Complement Bacteria i5 S. marcescens 8C9 0 03 b S. marcescens 6F1 0 03 S. marcescens 8C9 0 03 S. marcescens' 8C9 03 K. pneumoniae 8C9 0 K14 and 27 K. pneumoniae 6Fll 0 K14 and 27 K. pneumoniae 8C9 0 SK14 and 27 uo 0 o on. EXAMPLE VII no EXAMPLE VII demonstrates methods for the prooo o20 L l duction and selection of a human monoclonal antibody nthat possesses intergenus cross-reactivity against members of the genera Serratia marcescens, Klebsiella pneumoniae, Enterobacter aerogenes, and Enterobacter cloacae. Further, this example demonstrates the in vitro opsonic activity of said antibody against homologous S. marcescens, K. pneumoniae, E. aerogenes, and E.

cloacae serotypes. The process of Example I (essen- 0o o tially as described in parts A through G) was repeated, except that it was necessary to make specific modificationc to characterize and assay the antibody described in this Example. The following are changes in assay Iprocedures and the results obtained with the monoclonal antibody described herein.

1. Culture supernatants from four transformations were analyzed by the above method resulting in ~_Yil-_PYYI oo 00 O 0 O 0~ C 00 0) 0 C o 0 the identification of one well (1E4) which possessed binding activity on at least one of four K_ pneumoniae serotype plates, containing the capsule serotypes: 1, 2, 3, 4, 6, 8, 9, 19, 20, 21, 24 27, 31, 43, 44, and 55, but not on the control plate (no bacteria). Antibody from the 1E4 cell line was assayed for further intergenus cross-reactivity by a modification of the standard immunoblotting technique. Specifically, cross-reactivity to the bacteria K. pneumoniae, E.

aerogenes, S. marcescens, E. coli, E. cloacae, and P, aeruginosa was investigated by spotting bacteria onto a gridded nitrocellulose paper disc, reacting the disc containing the bacteria with said antibody, and developing the antibody -~actions with an alkaline phosphatase/ nitroblue tetrazolium enzyme system.

From these experiments, the 1E4 antibody was observed to possess further intergenus cross-reactivity. This antibody reacted with the following species and serotypes: 20 K. pneumoniae S. marcescens E. aerogenes E. cloacae Kl,3,8,9,13, 015 Clinical Clinical 15,29,31,33,36, Isolates Isolates 68,69 Thus, the human monoclonal antibody 1E4 possessed intergenus cross-reactivity to bacteria belonging to the species K. pneumoniae, S. marcescens, E. cloacae and E_ aerogenes.

2. The isotype of the 1E4 monoclonal antibody was determined in an ELISA procedure similar to the specificity tests described in Example I except that the antigen plate contained a pool of PLL immobilized K. pneumoniae K3 and K8 serotypes. Positive reaction of the 1E4 monoclonal antibody with the K. pneumoniae serotypes was observed only with the anti-IgM reagent, demonstrating an IgM isotype for the monoclonal antibody. It will be appreciated by those I _i ,i -66skilled in the art that if the process of this example were repeated several times and the isotypes of intergenus cross-reactive monoclonal antibodies so obtained were determined, one would find additional isotypes, IgM and IgG isotypes.

3. In vitro functional activity of the 1E4 monoclonal antibody was examined in an opsonophagocytic assay which compared the bacteriocidal activity of the antibody in the presence and absence of both human neutrophils and human complement. The bacterial serotypes used herein were only inactivated in the presence of monoclonal antibody 1E4, an active complement source, and human neutrophils (Table 10). When this experiment was repeated with serotypes unreactive with antibody 1E4, no bacterial destruction was observed (data not shown). These experiments demonstrated the functional specificity of monoclonal antibody 1E4, as well as its capacity to opsonize bacteria and promote their phago- 20 cytosis.

or"" o o o04 C 00 o o C *011 00 o K o 0 o CC 0

CC'

C ii

CO:)

-67- TABLE %Destruction of Input Complement Bacteria Bacteria Neutrophils Antibody S. marcescens 015 S. marcescens 015 S. marcescens 015 S. marcescens 015 K. pneumoniae K3 and 29 K. pneumoniae K3 and 29 K. pneumoniae K3 and 29 K. pneumoniae K3 and 29 E. aerogenes Isolate (2) E. aerogenes Isolate (2) E. aerogEnes 20 Isolate (2) E. aerones Isolate (2) E. cloacae Isolate E. cloacae 25 Isolate E. cloacae Isolate E. cloacae Isolate 1E4 GUll b 1E4 1E4 6Fl1 1E4 000 U 0 n0 0 00 0000 0 0 0 0 00 0 0 00 C0 0 6F11 1E4 1E4 1E4 6Fl1 1E4 1E4 a and b see Table 3 footnotes 0000 0 00 00 0 ita o o uo o 0 o 0 a0 00 0 00 0 0 0 O u 0 -68- EXAMPLE VIII EXAMPLE VIII demonstrates methods for the production and selection ot a human monoclonal antibody that possesses intergenus cross-reactivity against members of the genera Serratia marcescens, Klebsiella pneumoniae, Enterobacter -erogenes, and Pseudomonas aeruginosa. Further, this example demonstrates the in vitro opsonic activity ot said e.rtibody against homologous S. marcescens, K. pneumoniae, E. aerogenes, and P.

aeruginosa serotypes. The process of Example I (essentially as described in parts A through G) was repeated, except that it was necessary to make specific modifications to characterize and assay the antibody described in this Example. The following are changes in assay procedures and the results obtained with the monoclonal antibody described herein.

1. Culture supernatants from four transformations were analyzed by the above method resulting in the identification of one well (9D1) which possessed binding activity on at least one of four K. pneumoniae serotype plates, containing the capsule serotypes; 1, 2, 3, 4, 6, 8, 9, 19, 20, 24, 27, 31, 43, 44, and but not on the control plate (no bacteria). Antibody 25 from the 9D1 cell line was assayed for further intergenus cross-reactivity by a modification of the standard immunoblotting technique. Specifically, cross-reactivity to the bacteria K. pneumoniae, E. aerogenes, S.

marc.scens, E. coli, and P. aeruginosa was investigated by spotting bacteria onto a gridded nitrocellulose paper disc, reacting the disc containing the bacteria with said antibody, and developing the antibody reactions with an alkaline phosphatase/nitroblue tetrazolium enzyme system.

From these experiments, the 9D1 antibody was observed to possess further intergenus -69cross-reactivity. This antibody reacted with the following species and serotypes: K. pneumoniae S. marcescens E. aerogenes P. aeruginosa E9,13,15,29, 03,9,15,18 Clinical F6 33 Isolates Thus, the human monoclonal antibody 9D1 possessed intergenus cross-reactivity to bacteria belonging to the genera K. pneumoniae, S. marcescens, E. aerogenes, and P. aeruginosa.

2. The isotype of the 9D1 monoclonal antibody was determined in an ELISA procedure similar to the specificity tests described in Example I except ,-that the antigen plate contained a pool of PLL immobilized K. pneumoniae K13 serotype. Positive reaction of the 9D1 monoclonal antibody with the K. pneumoniae serotypes was observed only with the anti-IgM reagent, i demonstrating an IgM isotype for the monoclonal antibody. It will be appreciated by those skilled in the as: 20 art that if the process of this example were repeated several times and the isotypes of intergenus cross- S o reactive monoclonal antibodies so obtained were detero a mined, one would find additional isotypes, IgM and IgG isotypes.

S3. In vitro functional activity of the 9D1 oo monoclonal antibody was examined in an opsonophagocytic assay which compared the bacteriocidal activity of the i antibody in the presence and absence of both human neutrophils and human complement. The bacterial serotypes used herein were only inactivated in the presence of monoclonal antibody 9D1, an active complement source, s and human neutrophils (Table 11). When this experiment was repeated with serotypes unreactive with antibody 9D1, no bacterial destruction was observed (data iot shown). These experiments demonstrated the functional specificity of monoclonal antibody 9D1, as well as its capacity to opsonize bacteria and promote their phagocytosis.

TABLE 11 Destruction of Input Bacteria Neutrophils Antibody Complement Bacteria S. marcescens +9D1 0a 03 S. marcescens Fllb 0 03 marcescens -9D1 0 03 S. marcescens 9D1 87% 03 K. pneumoniae 9D1 0 K13 K. pneumoniae +6F11 0 K13 K. pneumoniae -9D1 0 K13 K. pneumoniae 9D1 K13 E. aerogenes +9D1 -0 Isolates (2) aerogenes +6F11 0 E. aerogenes -9D1 0 K Isolates (2) E. aerogenes +9D1 Isolates (2) P. aeruginosa 9D1 -0 F6 P._aeruginosa 6Fll 0 P. aerugiriosa -9D1 0 F6 P. aeruginosa 9D1 -A 30 F6 aand b see Table 3 footnotes -71- EXAMPLE IX EXAMPLE IX demonstrates methods for pioduction of a human monoclonal antibody that possesses intergenus cross-reactivity against members of the spe- Scies Pseudomonas aeruginosa aeruginosa), Escherichia coli coli), and Serratia marcescens marcescens). Further, this Example demonstrates the in vivo protective activity of said anitbo jagainst a lethal challenge of homologous P. aeruginosa, E. coli, and S.

marcescens serotypes. The process of Example I (essentially as described in parts A through G) was repeated to produce a human monoclonal antibody that was crossprotective against infections caused by the bacteria to which it binds. Specific modifications to the process of Example I, to characterize and assay the antibody, Got* are described in this Example. The changes in assay procedures and the results obtained with the monoclonal o antibody were as follows.

o 0 O o ao 0 G 0o'" 1. Supernatants were screened for the presence 04 20 of anti-P. aeruginosa antibodies using an ELISA techoo nique as described in Example I. The antigen plate consisted of a flat-bottom 96-well Immunolon 2 micro- S titer plate (Dynatech, Alexandria, VA), the wells of 't which contained a mixture of poly-L-lysine (PLL) immo- -o oo 25 bilized bacteria belonging to the seven P. aeruqinosa Fisher reference strains (Fisher, et al., J. of Bacteriology (1969) 98:835-836, A.T.C.C. Nos. 27312- 27318).

Culture supernatants from one transformation o4 were analyzed by the above method and resulted in:the identification of one well that possessed activity on the P. aeruginosa plate, but not the PLL control plate.

It was determined in subsequent ELISA's,with the seven- 3 teen individual P. aeruginosa serotypes/s honing to the International Antigenic Typing Scheme (IATS, A.T.C.C.

Nos. 33348-33364), that one master well 9C3 contained /-Ti .1Su i ~il-iiii;- I v 00 0 o 09 0 00 0 0 0 0 0 0 o o0 U O 00 9 0 0 -72antibcoies which bound to IATS serotype type 1 (Liu, Int. J. Syst. Bacteriol. (1983) 33:256-264, which is incorporated herein by reference).

Thus, from this experiment, one cloned transformed human cell line was achieved which is continuous (immortal) and secretes a single human monoclonal antibody which binds to a determinant on the surface of the P. aeruginosa IATS type 1.

Prior to filing this patent application, the continuous transformed human cell line identified as 9C3 was deposited with the American Type Culture Collection, Rockville, MD, at A.T.C.C. No. CRL 9239.

2. Antibody from the cloned 9C3 cell line was also assayed for cross-reactivity to gram-negative and gram-positive bacteria by a modification of the standard immunoblotting technique. Specifically, crossreactivity to the bacteria E. coli, K. pneumoniae, S.

marcescens, E. aerogenes, E. cloacae, Haemophilus influenzae, and Staphylococcus aureus was investigated by 20 spotting bacteria onto a gridded nitrocellulose paper disc, reacting the disc containing the bacteria with 9C3 antibody, and visualizing the antibody reactions with an alkaline phosphatase/nitroblue tetrazolium enzyme system (as described in Example I).

From these experiments, antibody 9C3 was observed to bind to E. coli serotype 06 and the S. marcescens serotypes 012 and 014. Thus, the human monoclonal antibody 9C3 possesses intergenus cross-reactiv- 3 ity among the gram-negative bacteria belonging to specific serotypes of the species E. coli, S. marcescens, and P. aeruginosa.

3. The finding that the monoclonal antibody cross-reacted with several different bacterial gepera, suggested the antibody may bind to a shared antigenic determinant. The biochemical characterization of the I~ -73molecular species recognized by the 9C3 antibody was accomplished by immunoblot analysis as described in Example I. Reactions were noted in deoxycholate extracts of reactive serotypes from P. aeruginosa and E.

coli, but not S. marcescens. Although it is ncc clear why antibody 9C3 was unreactive w .th the S. marcescens preparation, it is possible the antibody recognizes a conformational epitope that was destroyed by the preparation treatments. For the E. coli and P. aeruginosa bacterial extracts, the 9C3 antibody appeared to recognize a series of regularly spaced molecular entities giving rise to a ladder-like pattern on the immunoblot.

This profile was entirely consistent with that seen in polyacrylamide gel electrophoretic analysis of LPS in the presence of SDS, where it has been demonstrated "o that the heterogeneous size profile exhibited by the bands is due to a population of LPS molecules differing Sby weight increments equivalent to the number of Oe °0 antigenic oligosaccharide side chain units present per o2o molecule (Pavla, and Makela, supra, and o oe 20 0o Goldman, and Leive, supra).

To further define the molecular nature of the antigen, the deoxycholate extracts were treated with o proteolytic enzyme, Proteinase K, prior to their electrophoresis (Eberling, supra). The immunoblot pat- Son 25 0 25o terns observed after Proteinase K treatment were identical to those patterns observed without treatment and, 00 0 o thus, suggested that the antigen reactive with the 9C3 antibody was not protein in nature.

o 0 4. The isotype of the 9C3 monoclonal antibody L0 0 was determined in an ELISA procedure similar to the specificity tests described above, except that the antigen plate contained PLL-immobilized P. aeruginosa Fisher serotype 4 bacteria. Positive reaction of the 9C3 monoclonal antibody with the P. aeruginosa strain a

II_

ls13PII--1 -74was observed only with the anti-IgM reagent, demonstrating an IgM isotype for the monoclonal antibody.

In vitro functional activity of the 9C3 monoclonal antibody was examined in an opsonophagocytic assay which compared the bacteriocidal activity of the antibody in the presence and absence of both human neutrophils and human complement.

The'bacterial strains used herein were only 1 killed in the presence on monoclonal antibody 9C3, an active complement source, and human neutrophils (Table 12). When this experiment was repeated with several non-9C3 reactive bacterial serotypes, no bacterial destruction was observed, thus demonstrating the functional specificity of monoclonal antibody 9C3 and its capacity to opsonize bacteria and promote their phagocytosis. Since the combined action of opsonins (specific antibodies) and polymorphonuclear leukocytes (neutrophils) appeared to be the primary mechanism for im- 2 munity to these bacterial strains, these data indicate that antibody 9C3 would, after appropriate administration, provide protection against lethal challenges with the bacteria strains described herein.

o o o u o oo 00 0 0 o Coun S 0 0 0 0 000 b 0 0 0 o& 0 f L i~ I~ TABLE 12 Destruction of Bacteria Neutrophils Antibody Complement Input Bacteria (a) E. coli 06 9C3 (a 0 (b) E. coli 06 6F11 b 0

E.

E. coli 06 9C3 0 E. coli 06 9C3 98% S. marcescens S. 014 9C3 0 S. marcescens 014 6F11 0 S. marcescens 014 9C3 0 S. marcescens 014 9C3 94% P. aeruginosa Fisher 4 9C3 0 o o P. aeruginosa o Fisher 4 6F11 0 P. aeruginosa S o Fisher 4 9C3 0 2o P. aeruginosa 20 Fisher 4 9C3 79% .1 0 0 heat-inactivated (56C for 30 min) human complement.

or (b)6F11 culture supernatant containing an IgM human monoclonal o antibody to Pseudomonas aeruginosa Fisher type 2.

o '1 u25 6. To test the protective characteristics of the o a 9C3 antibody, animal protection studies were performed So with at least one organism from each genus described herein.

The burned mouse model was used for protec- 0 o 30 0,o" tion experiments with P. aeruginosa Fisher 4 and S.

marcescens 014. For each bacterial challenge, female, outbred Swiss-Webster mice weighing 22-25 gm were divided into three groups of 7 or 8 mice each. An exemplary experiment was performed as follows: -76- Group Bacteria Antibody A P. aeruginosa F4 9C3 B P. aeruginosa F4 6F11 C P. aeruginosa F2 9C3 The day before the experiment, each mouse was shaved and treated with a depilatory agent to remove all hair on the back (burn site). On the experiment day, each animal received 0.1 ml in each thigh of an 1 anesthetizing saline solution containing 0.7 ml 0.85% NaC1, 0.2 ml xylazine (20 mg/ml) and 0.1 ml ketamine (100 mg/ml), such that the dosage per mouse was mg/kg zylazine and 180 mg/kg ketamine. The anesthetized mice received a 10% of total body surface area, full-thickness, third degree gas flame burn. Immedio"°;Oo ately after wound infliction, the mice were injected sub-eschar with 0.5 ml of 4 0 C antibody containing spent S° culture fluid pre-mixed with 5-10 LD 50 's of bacteria.

L. l0 The bacterial suspension had been prepared from a broth .02 culture in logarithmic phase growth, from which the o" u20 Sbacteria were centrifuged, washed twice in PBS, and resuspended to the appropriate concentration in PBS.

Animals were observed for a period of ten days. Three oo to five days post-challenge, all animals in Group B o (irrelevant antibody) and Group C (irrelevant organism) I o .o were dead. In contrast, those animals that had received the 9C3 (Group A) antibody were all alive and symp- ,tom-free (see Table 13).

For the E. coli 06 protection studies, healthy Swiss-Webster mice (20-22 gm) were divided into °Oo o three groups of ten mice each. Each group receiving antibody was injected intravenously with 200 pl of sterile PBS containing 25 pg of purified antibody. Two hours later, all animals were challenged intraperitone- 3 ally with 300 pl of live bacteria containing 3 LD0' s of their respective bacterial strain (for results, see Table 13).

-77- TABLE 13 Survival/ Expt. Challenge Bacteria Antibody Challenge Survival 1 P. aeruginosa F4 9C3 6/7 86% P. aeruginosa F4 6 F11(a) 0/7 0 P. aeruginos?. F 2 9C3 0/7 0 2 E. coli 06 903 6/10 E. coli 06 6F11 0/10 0 3 S. marcescens 014 9C3 8/8 100% S. marcescens 014 6F11 0/8 0 (a) 6

F

11 antibody is specific to Pseudomonas aeruginosa Fisher immunotype 2 and surves as negative control antibody.

(b) P. aeruginosa F2 is not reactive with the 9C3 antibody and serves as a non-specific control organism.

These data demonstrate that the human monoclonal antibody 9C3 is able to protect mice from lethal challenges with bacteria belonging to three gram-negative genera. The intergenus cross-reactive human monoclonal antibody 9C3 was ablu to afford protection with antibody containing culture supernatant or purified antibody against infection by organismsbelonging to t.ie gram-negative genera E. coli, S. marcescens, and P.

aeruginosa.

From the foregoing, it will be appreciated that the cell lines of the present invention provide compositions of human monoclonal antibodies and fragments thereof cross-reative for and cross-protective against various bacterial species, both gram-negative and gram-positive. This allows prophylactic and therapeutic compositions to be more easily developed that can be effective against nosocomial and neonatal infections due to most, if not all, bacterial genera. By combining the antibodies, it is possible to obtain 1 -78broad protection against a large portion, usually less than all, of the clinically significantkaILt bod-E s. In addition, the cell lines provide antibodies which find uses in immunoassays and other well-known procedures.

Although the present invention has been described in some detail by way of illustration and example for purposes of clarity of understanding, it will be obvious that certain changes and modifications may be practiced Within the scope of the appended claims.

PB

0 o oI pc 0i 0 0, n a .9

Claims (13)

1. A composition useful in the propylactic and/or therapeutic treatment of bacterial infection, said composition comprising a prophylatic or therapeutic effective amount of a plurality of human monoclonal antibodies and being reac- tive with at least two bacterial species of at least two genera, at least one of said antibodies or binding fragment thereof being capable of specifically reacting with a non-core carbohydrate epitope shared by serotypes of two different species.

2. A composition according to claim 1, wherein at least one of said antibodies or binding fragments thereof reacts with serotypes of species of at least three bacterial species and genera.

3. A composition according to claim 1 or claim 2 wherein at least one of said bacterial species is gram-pos- itive and a second is gram-negative. o, 4. A composition according to claim 1 or claim 3, wherein each of at least two of said antibodies is protective 0 against infection caused by bacteria belonging to at least o o two different genera. 0 5. A composition useful in the prophylactic and/or 0 oo therapeutic treatment of bacterial infection, said composition comprising a prophylatic or therapeutic amount of at least two human monoclonal antibodies or binding fragment thereof, 00 wherein each of said antibodies specifically reacts with I a non-core carbohydrate epitope shared by serotypes of two or more bacterial species of different genera and a plurality of serotypes in at least one species, wherein said species comprise Escherichia coli, Neisseria meningitidis, Serratia marcescens, Klibsiella pneumoniae, Enterobacter aerogenes, Enterobacter cloacae, Pseudomonas aeruginosa, and Streptoco- ccus agalactiae Group B.

6. A composition according to any one of claims 1 to 5, having at least one human monoclonal antibody which reacts with at least one of the following bacterial species combinations: E. coli, S. marcescens and E. aerogenes; E. coli and N. meinigitidis; E. coli, E. cloacae 39 and Streptococcus agalactiae Group B; K. pneumoniae, -79- i S. marcescens and E. aerogenes; K. pneumoniae and S. marcescens; K. pneumoniae, S. marcescens, E. aerogenes and E. cloacae; K. pneumoniae, S. marcescens, E. aero- genes, and P. aeruginosa; or E. coli, S. marcescens and P. aeruginosa.

7. A composition useful in the prophylactic and/or therapeutic treatment of bacterial infection substantially as hereinbefore described with reference to any one of the examples.

8. A pharmaceutical composition comprising a comp- osition according to any of claims 1 to 7, and a physiolog- ically acceptable carrier.

9. A method of inhibiting symptoms of bacterial disease in a human host which comprises: administering to said host a therapeutically or prophylactically effective dose of a composition according to any of claims 1 to 7, a physiologically acceptable carrier, and an antimicrobial agent and/or a gamma glogulin fraction from human blood plasma.

10. An immortal cell line which secretes a human mono- clonal antibody or binding fragment thereof specifically cross-reactive with an epitope comprising an accessible o non-core carbohydrate moiety present on at least two serotypes of two different species of different bacterial genera.

11. A cell line according to claim 10, wherein said S.oo bacterial species are at least two of the species Escherichia oo o coli, Serratia marcescens, Neisseria meningitidis, Klebsiella pneumoniae, Enterobacter cloacae, Enterobacter aerogenes, 0o o Pseudomonas aeruginosa, and Streptococcus agalactiae Group B. 12 A accord-i-ng- -to- cl-aim -de-s-ig-n--ed-- 0o S ATCC No. CRL 9006, CRL 9007, CRL 9008, CRL 9009, and CRL 9239.

13. A cell line substantially as hereinbefore described.

14. A human monocI-tGal antibody or binding fragment thereof capable of binding to aa bohydrate epitope reactive with a monoclonal antibody produced by aell line according to any one of claims 10 to 13. A kit for use in detecting the presence of a east 3 9. +-we--member-----ef-species -e-f-d-i-ffe-re-n-t -baGer-i-g-ane-r-a-s-ad- r KW (O. I U-~c~ 12. A cell line according to claim 10, designated ATCC No. CRL 9006, CRL 9007, CRL 9008, CRL 9009, and CRL 9239 as herein before described. 13. A cell line substantially as hereinbefore described. 14. A human monoclonal antibody or binding fragment thereof capable of binding to a carbohydrate epitope reactive with a monoclonal antibody produced by a cell line according to any one of claims 10 to 13. 0 0 0 o~a o 0 o 0 os O 0 o0 0 00 0 0 0 So noo (4 0 LR O 0 00 0Q0 o i 0 0 0 00 00 0 0 00 0 (4 0 0 0 00 0 0 r 0 00 S00 0 f 0 0 0 0 (0 VA V, $Ir~p rl! I Irrr~rrrlsrr*lrr~lll~13~ r~rrCI coQmprising a mqnoc.onai antiDody co s o .a.n.Ing at least one monoclonal antibody, wherein sai L-acibody reacts with a non-core carbohydrate determin-a t shared by at least two members of different bacqte-rial species, and a label pro- viding for a detecta b-e-signal covalently bonded to said anti- body or bond a second antibody reactive with said mono- A composition useful in the prophylactic and/or e rosau c aLHPrpnt.il treatment of neonatal sepsis or meningitis, said composition comprising a prophylactic or therapeutic amount of at least two human monoclonal antibodies or binding frag- ment thereof, wherein each of said antibodies specifically reacts with a non-core carbohydrate epitope shared by sero- types of two or more bacterial species of different genera, wherein said species comprise Escherichia coli Kl, Neisseria meningitidis Group B, Hemophilus influenzae Type B, and Streptococcus agalsotiae Group B. Lo A composition useful in the prophylactic and/or therapeutic treatment of a pre-selected number of bacterial species, said composition comprising a prophylactic or thera- peutic effective amount of two or more human monoclonal anti- 0 Q bodies or binding fragments thereof, 0 the antibodies being capable of reacting with a total of at least four of the bactezial species, and a compatible carrier. D .o 1. A method for producing a pharmaceutical composition o o useful in the prophylactic or therapeutic treatment of bacterial infection, said method comprising: 0. 1 o4 combining a plurality of human monoclonal anti- bodies wherein said composition reacts with at least two bacterial species of at least two genera, wherein at least coe one of said monoclonal antibodies or binding fragments thereof is capable of specifically reacting with a non-core carbo- hydrate epitope shared by serotypes of two different species. A method for producing a pharmaceutical composition useful in the prophylactic and/or therapeutic treatment of neonatal sepsis or meningitis, wherein said method comprises: combining a prophylactic or therapeutic amount 3 9 at least two human monoclonal antibodies or binding frag- S-1W- "T -81- 3 n.' ments thereof, wherein each of said antibodies sPecifically reacts with a non-core carbohydrate epitope shared by sero- types of two or more bacterial species of different genera, wherein said species comprise E. coli Kl, Neisseria menin- gitidis, Hemophilus influenzae, and Streptococcus algalactiae Group B. A method for producing a pharmaceutical compos- ition substantially as hereinbefore described with reference to any one of the examples. 2O. A method of producing a human monoclonal antibody specifically cross-reactive with a non-core carbohydrate epitope present on at least two serotypes of at least two different species of different bacterial genera, comprising: cultivating one of the cell lines of claims 11 or 12 and recovering said antibodies. A method for producing human monoclonal antibodies capable of specifically binding to a non-core carbohydrate o("O epitope shared by serotypes of two different bacterial a species, said method comprising: isolating the B cells from an individual; -o immortalizing said B cells to form a plurality S of antibody producing clones; screening said clones for the production of anti- bodies specific for a non-core carbohydrate epitope of at least two different bacterial species; and oa cultivating said selected clones and harvesting o" the monoclonal antibody produced therefrom. S2. A method according to claim 2\ wherein said B cells are immortalized by EBV transformation. A method for producing human monoclonal antibodies substantially as hereinbefore described with reference to any one of the examples. BADTED---EFebruary,-1-98-7- PHILLIPS ORMONDE AND FITZPATRICK Attorneys for: ENETIQC SYSTEMS CORPORATION 7 -39 SKW -82- 83

24. A composition useful in the prophylactic and/or therapeutic treatment of bacterial infection due to group B streptococci, said composition comprising a protective human monoclonal antibody or binding fragment thereof which binds with the Group 93 carbohydrate of said group B streptococci and also similarly binds to another gram-negative bacterial species. A corr -ition according to claim 24, wherein the human monoclonal antibody also binds with E. coli or Serratia o 1 t o Marcescens. o 0oo 0

26. A composition according to claim 25 wherein the Bn E. coli includes serotypes 04 and 018 as hereinbefore 00o described. o 00 0 0. .o 27. A composition useful in the prophylactic and /or Soo 0 i5 therapeutic treatment of bacterial infection due to E. coil, said antibody or binding fragment thereof which binds to a capsule carbohydrate moiety present on E. coli K1 and also o=oo similarly bind to another gram-negative bacterial species. eo o o 28. A composition according to claim 27, wherein the human gJ_ monoclonal antibody also binds with N. meningitidis group B. o DATED: 25 February 1991 o0 o u 0o o PHILLIPS ORMONDE FITZPATRICK Attorneys for: GENETIC SYSTEMS CORPORATION f) J. f a~ Lr 1 L- I