WO2011107989A1

WO2011107989A1 - Improved therapeutic antibodies against flagellated pseudomonas aeruginosa

Info

Publication number: WO2011107989A1
Application number: PCT/IL2011/000208
Authority: WO
Inventors: Lewis F. Neville
Original assignee: Lostam Biopharmaceuticals Ltd
Priority date: 2010-03-01
Filing date: 2011-03-01
Publication date: 2011-09-09
Also published as: EP2542577A1; US20130096282A1; US8653242B2

Abstract

Improved antibodies are provided selected from human, dual-specific, chimeric or humanized antibodies, wherein said human chimeric and humanized antibodies specifically bind to flagellin type A or type B of P. aeruginosa, and said dual-specific antibodies specifically binds to flagella type A and type B of Pseudomonas aeruginosa, and said antibodies are protective against infection caused by P. aeruginosa. These antibodies as well as pharmaceutical composition comprising them are useful for the treatment of indications caused by P. aeruginosa infection.

Description

IMPROVED THERAPEUTIC ANTIBODIES AGAINST FLAGELLATED

PSEUDOMONAS AERUGINOSA

TECHNICAL FIELD

The present invention relates to antibodies that specifically binds to flagella type A or type B of P. aeruginosa, or specifically binds to flagellin type A and type B of P. aeruginosa, but excluding mouse monoclonal antibodies, to pharmaceutical compositions comprising them and to methods for prophylactic or therapeutic treatment of infections caused by these bacteria.

BACKGROUND ART

Pseudomonas aeruginosa is an opportunistic pathogen that causes a variety of acute infections especially in cystic fibrosis patients, mechanically ventilated patients, burn victims and immunocompromised individuals. P. aeruginosa infections are recognized by the medical community as particularly worrisome and difficult to treat, especially multidrug resistant (MDR) strains in the hospital setting. A patient's prognosis for recovery from an infection caused by P. aeruginosa is enhanced when the diagnosis is made and appropriate treatment initiated as early in the course of the infection as possible, before the number of bacteria in the host becomes overwhelming and much more difficult to bring under control.

Patients suffering from major burns are especially vulnerable as they have extensive disruption of the skin barrier, with a concurrent suppression of the immune system. These conditions expose the burn area to bacterial wound colonization that can lead to burn wound infection, sepsis, multi-organ failure and subsequent death. Bacterial infection is the leading cause of death in major burns (responsible to 50-80% of overall thermal injury mortality), and P. aeruginosa is the most prevalent pathogen isolated from burn wound infections mainly due to its high persistence in the environment and its high intrinsic antibiotic resistance. Moreover, excessive antibiotic pressure in burn units has resulted in the emergence of multidrug-resistant strains of P. aeruginosa (Edward- Jones et al., 2003, Singh et al, 2003), the emergence of which underscores the clinical need to develop new classes of antibacterial therapeutics that can target centrally important proteins implicated in P. aeruginosa virulence. P. aeruginosa is the scourge of hospital-associated pneumonias (HAP) of which ventilator-associated pneumonia (VAP) is routinely observed in Intensive Care Units. P. aeruginosa also causes infections in immune suppressed individuals (eg. cancer patients, patients awaiting transplantation, AIDS patients, premature babies). Also, P. aeruginosa causes infections at surgical sites, is responsible for urinary tract infections and of course is a relentless pulmonary infection in Cystic Fibrosis patients. Additionally, patients suffering from chronic obstructive pulmonary dysfunction (COPD) and bronchiectasis are highly susceptible to P. aeruginosa infections. Healthy individuals may also be susceptible to P. aeruginosa; for example, contact lens users are prone to P. aeruginosa infections and Swimmer's ear (also called otitis externa) is often caused by P. aeruginosa.

Immunotherapeutic strategies focusing on immunization (active or passive) and treatment targeting P. aeruginosa virulence-associated factors, such as elastase, protease, and exotoxin A, have been described in the literature and have showed limited success in the prevention and treatment of P. aeruginosa infections in animal models (Drake et al., 1987; Landsperger et al., 1994; Matsumoto et al., 1999; Steiner et al., 1998; Eaves-Pyles et al, 1999). Moreover, anti-lipopolysaccharide antibodies were found to be ineffective against a broad spectrum of P. aeruginosa isolates because of the presence of various lipopolysaccharide serotypes.

Flagellin protein, the principal component of bacterial flagella, has long been recognized as an important virulence factor in P. aeruginosa infections. Two major antigenic groups of flagella have been identified; type A, also known as HI, comprising five sub-types designated ao_-4, and type B, also known as H2, having no sub-groups (Lanyi et al, 1970; Ansorg, 1978). The 'propeller-like' rotation as a result of flagellin polymerization is crucial for bacterial locomotion and survival and it is a contributing factor to bacterial invasion. Indeed, sub-type monoclonal antibodies (mAbs) raised against endogenous P. aeruginosa flagellins have been shown to afford protection in different animal models of P. aeruginosa infection (Drake et al., 1987; Rosok et al., 1990; Oishi et al., 1993; Landsperger et al., 1994; Matsumoto et al., 1999). US 4,834,976 discloses monoclonal antibodies or antigen binding fragment thereof capable of specifically reacting with P. aeruginosa bacteria flagella type A or type B, but not both, which are said to be protective in vivo against said bacteria.

The flagellum, which mediates rapid bacterial movement, has an important role in bacterial dissemination and ultimately the progression of local bacterial colonization into an overwhelming invasive disease. Studies using various gram-negative flagellated bacteria, as well as flagellin deletion strains have demonstrated that flagellin is a 'double- edged sword'. Besides its role in bacterial motility and cellular invasion, isolated flagellin protein or intact P. aeruginosa harboring the flagellum appendage, can trigger inflammatory cells to produce an array of different pro-inflammatory molecules, including those involved in innate immunity. Flagellin's mode of action is mediated through high affinity binding towards the Toll-5 receptor (TLR5) whose activation results in the nuclear, translocation of NF- κΒ and Elk-1 with enhanced transcription of immune response genes. Recent molecular studies revealed that the proinflammatory induction by flagellin is located at both the N' and C'-terminal regions of the molecule, rendering flagella, a highly feasible, "appendage-like" target for immunotherapy. The noxious effects of flagellin clearly highlights flagellin's role in experimental systemic tissue injury and circulatory shock.

Recently, it has been shown that a large percentage of P. aeruginosa strains from CF patients are non-motile, yet do possess an intact flagellum (and associated flagellin proteins) which has been essentially rendered disabled in terms of motility. Nevertheless, these non-motile strains are still damaging since their disabled flagellin retains its biological activity at lung TLR5 causing profound local inflammation. Thus, antibodies targeting P. aeruginosa flagellin should be protective against both motile and non-motile (flagellin positive) strains. Because most clinical P. aeruginosa isolates are flagellated, and given that there are only two major antigenic types, antiflagellum immunotherapy has been proposed as a possible treatment for P. aeruginosa infections.

SUMMARY OF INVENTION

The present invention relates, in one aspect, to an antibody that specifically binds to flagella type A or type B of P. aeruginosa, or specifically binds to flagellin type A and type B of P. aeruginosa, but excluding a mouse monoclonal antibody. This antibody may be a fully human antibody or a chimeric, humanized or dual-specific antibody or an antigen-binding fragment of the antibody.

The antibody may be covalently linked, optionally via a cleavable linker, to an antibiotic agent and/or it may be covalently linked to a nonproteinaceous polymer, such as polyethyleneglycol. The present invention further relates to a nucleic acid molecule comprising a nucleotide sequence which encodes the antibody of the present invention, to vectors comprising said nucleic acid molecule operably linked to a promoter capable of driving the expression of said nucleic acid molecule, and to a host cell, such as mouse myeloma NSO and Chinese Hamster Ovary (CHO) cells, or plant cells, such as tobacco, carrot and rice cells, comprising the vector(s).

In yet another aspect, the present invention relates to a pharmaceutical composition comprising at least one antibody or antigen binding fragment thereof, according to the present invention, or said antibody or an antigen binding fragment of the antibody covalently linked to an antibiotic agent or a nonproteinaceous polymer, and a pharmaceutically acceptable carrier.

The present invention further relates to methods for treatment of a subject suffering from or being susceptible to infection caused by P. aeruginosa, comprising administering to said patient a therapeutically effective amount of an antibody according to the present invention or antigen binding fragment thereof, or a derivative of said antibody or fragment thereof or an antigen binding fragment of the antibody covalently linked to an antibiotic agent or a nonproteinaceous polymer.

In still another aspect, the present invention relates to a method for diagnosing and monitoring a host response to severe P. aeruginosa infection, comprising (a) obtaining a patient sample; (b) contacting said patient sample with an antibody of the invention; and (c) measuring binding of said antibody to said patient sample, wherein detection of binding of said antibody to said patient sample indicates the presence or level of P. aeruginosa in the patient sample. In order to facilitate convenient detection of the antibody, it may be labeled with a detectable label such as, but not limited to, antigenic peptide tags fluorophores, enzymes, luminescent compounds, radioisotopes and micro- or nano- particles.

In yet another aspect, the present invention relates to a kit for detecting P. aeruginosa in a biological sample comprising a container containing an antibody according to the present invention; a container containing a set of reagents required for the detection and quantification of said antibody; and instructions for use. BRIEF DESCRIPTION OF DRAWINGS

Fig. 1 shows a schematic drawing of a dual-specific antibody using "knobs-into- holes" CH3 mutations. CH2 and CH3, second and third constant regions of human IgGl . Two chains are heterodimerized by "knobs-into-holes" engineered CH3 domains. Two naturally occurring hinge region disulfide bonds are indicated by horizontal lines.

Fig. 2 shows a schematic drawing of a dual-specific antibody consisting of a chimeric, humanized or human antibody (empty block arrow), for example LST-003, linked via a linker peptide (thin black arrow) to two single chain variable fragments (scFvs; filled block arrow), derived for example from LST-002, that are connected via a spacer peptide (dashed arrow).

Figs. 3A-B show motility of clinical P. aeruginosa (PA) strains PA 27853, known to harbor type A flagellin, in the presence of anti-flagella type A mAb (LST-001) antibody (A) or the multi-drug-resistant Ka02 PA strain and laboratory strain PaOl (B). Quantitative data based on measurements done on the diameter of the halo formed by motile bacteria grown on agar is depicted.

Figs. 4A-B show a chromatogram (A) with a large absorbance peak (280 nm) representing a purified protein fraction collected from the supernatant of a CHO cell culture containing chimeric LST-003 mAb and a corresponding Coomassie-Blue stained polyacrylamide gel (B) showing the presence of intact chimeric LST-003 heavy and light chain polypeptides in the collected fractions 4-7 as indicated on the X-axis of the chromatogram with the majority of staining in fractions 4 and 5.

Figs. 5A-B show two separate immunoblot profiles in which different laboratory or multi-drug resistant (MDR) strains of P. Aeruginosa lysates were screened with either mouse LST-001 mAb (A) or chimeric LST-003 mAb (B) that target P. aeruginosa, flagellin type A. Five bacterial strains (P. aeruginosa. 141, 49, 1, 409 and PAK) were shown to be flagellin type A, since immunoreactivity was observed with mouse LST-001 (A) and chimeric LST-003 (B) mAbs. In contrast, 2 bacterial strains (PA21 and PA01), previously known to harbor flagellin type B since they reacted solely with LST-002 mAb, failed to react with either LST-001 or LST-003 mAbs.

Fig. 6 shows specific, whole P. Aeruginosa bacterial binding, as measured by

ELISA, of mouse monoclonal anti-flagella type A and B antibody (LST-001 and LST-002, respectively), chimeric anti-flagella type A antibody (LST-003), and human monoclonal IgM anti-flagella type B antibody (LST-007). Black columns, whole bacterial binding to flagella type B; White columns, whole bacteria binding to flagella type A. OD, Optical Density.

Figs. 7A-B show the purification profile of recombinant human LST-008 transiently expressed in CHO. A chromatogram of the purification profile is depicted (A) and the peak fraction containing LST-008 was stained with Coomassie following SDS- PAGE (B; lane 2; lane 1 = marker proteins).

Fig. 8 shows the binding profile of LST-008 on ELISA plates coated with recombinant PA flagellin type B.

Fig. 9 shows binding of human mAbs LST-007 and LST-008 targeting P. aeruginosa flagellin type B. KaOl, Ka02, Ka03 and a04 - clinical P. Aeruginosa isolates of which Ka02 and Ka04 are multi-drug or partial drug resistant; 25619 , NCTC 10662, 27853 - commercial PA strains.

Figs. lOA-C show details of the chromatographic purification of the monoclonal dual-specific anti-flagella type A and B antibody LST-009 (A) and the corresponding Coomassie-gel stained analysis of the purified protein without (B) or with (C) reducing agent (dithiothreitol, DTT).

Fig. 11 shows binding of the monoclonal dual-specific anti-flagella type A and B. antibody LST-009 to recombinantly expressed flagellin type A, by an ELISA assay; OD, Optical Density. The binding profile of LST-009 towards PA flagellin type A is directly compared with the chimeric mAb LST-003.

Fig. 12 shows binding of the monoclonal dual-specific anti-flagella type A and B antibody LST-009 to recombinantly expressed flagellin type B, by an ELISA assay; OD, Optical Density. The binding profile of LST-009 towards PA flagellin type B is directly compared with the chimeric mAb LST-004.

Fig. 13 shows binding of the monoclonal dual-specific anti-flagella type A and B antibody LST-009 to flagellin type A in formaldehyde-fixed PAK bacteria, by an ELISA assay; OD, Optical Density.

Fig. 14 shows binding of the monoclonal dual-specific anti-flagella type A and B antibody LST-009 to flagellin type B in formaldehyde-fixed PaOl bacteria, by an ELISA assay; OD, Optical Density.

Fig. 15A-B depict a comparisons of binding of the dual-specific LST-010 mAb as compared to LST-003 (A) or LST-004 (B) towards ELISA plates coated with purified PA flagellin type A (A) or purified PA flagellin type B (B). OD, Optical Density; abscissa, concentration (nM) of mAb added to each well.

Fig. 16 shows binding properties of dual-specific mAb LST-010 (1 nM) towards PA bacteria containing flagellin type A (PA27853) or type B (Ka02). OD, Optical Density.

Fig. 17 shows binding properties of dual-specific mAb LST-010, LST-003 or LST-

004 towards PA bacteria containing flagellin type A (PA27853) and type B (Ka02). LST- 010 binding to wells containing an equal mixture of immobilized PA bacteria harboring flagellin types A and B constituted the total binding of the control LST-003 and LST-004 mAbs. OD, Optical Density.

Fig. 18 demonstrates the superiority of LST-002 vs. imipenem in preventing mortality in a mouse model of pneumonia driven by Ka02, a MDR PA bacterial strain. Data represents Kaplan-Meier survival curves of infected mice treated with saline, LST- 002 or imipenem over a 3 day post-infection period. Survival at 3 days in mice treated with LST-002 was highly significant (p < 0.0001) as compared to saline or imipenem-treated animals using Fisher's exact test. Identical survival numbers and statistical analysis were obtained at day 9 post-infection.

Fig. 19 depicts the effect of LST-002 on Ka02-mediated thigh muscle infection. Data represents the Ratio-Infection Index (RII; ordinate axis) of the surface-area of thigh muscle infection in ipsilateral muscle to contralateral muscle in saline and LST-002 treated mice as a function of days (abscissa) following 2 separate injections of cyclophosphamide (day 0 and day 3), with Ka02 being injected at day 4. LST-002 was administered i.v. at 20 mg/kg, 60 min after administration of Ka02. Further dosing of LST-002 (10 mg/kg) was given i.p. at time points + 5 hr, + 24 hr and + 48 hr after infection.

DETAILED DESCRIPTION OF THE INVENTION

An antibody molecule of the IgG type comprises two light chains and two heavy chains linked via disulfide bonds. Both the light chain and the heavy chain contains a domain of relatively variable amino acid sequences, known as the variable region, which in turn contains hypervariable regions, also known as complementarity-determining regions (CDR), that are interspersed among relatively conserved framework regions, as defined by Kabat (1970), Brochet et al (2008) or Chothia (1987; 1992). Together, the CDR and framework region determine the three-dimensional structure of the IgG binding site and thus, the antigen specificity of the antibody. The complete IgG molecule also contains a domain of relatively conserved amino acid sequences, called the constant region consisting of three constant domains (C_H1-3).

The IgG molecule is often referred to in terms of its functional fragments. Cleavage of an IgG with the protease papain produces two identical antigen-binding fragments (Fab) and an "Fc" fragment conferring the biological activity of the antibody, such as binding to the first component of the complement cascade or binding to Fc-receptor bearing cells, such as phagocytes, mast cells, neutrophils and natural killer cells. The Fc fragment comprises the heavy constant regions CH2 and CH3, and the Fab fragment comprises the heavy (CHI) and light (CL) constant regions and the variable regions of the heavy (V_H) and light (V_L) chains. The terms "Fab", "Fab-fragment" and "Fab-region" are used interchangeably herein.

Two mouse, and two human monoclonal antibodies have been disclosed in US 4,834,976 that bind to flagellin proteins of P. aeruginosa. Two of the antibodies, the mouse FA6 IIG5 (ATCC HB9130) termed herein mouse LST-001 and the human 21B8 (ATCC CRL 9301), are specific to P. aeruginosa flagellin type A, and the two other antibodies, the mouse PaF4 IVE8 (ATCC HB9129), termed herein mouse LST-002, and the human 20H11 (ATCC CRL9300), termed herein LST-007, are specific to P. aeruginosa flagellin type B.

None of the above mouse antibodies is suitable alone as an active agent for treatment of P. aeruginosa infections, due to its murine origin. Monoclonal antibodies of murine origin are highly antigenic causing a human anti-mouse antibody (HAMA) response, and therefore often are rapidly removed from circulation and may be the cause of systemic inflammatory effects. Both mouse LST-001 and LST-002 antibodies, or derivatives thereof, were never developed into antibacterial drugs.

The above human antibodies are expressed in Epstein-Barr virus (EBV) transformed B-cells, which are an extremely poor source for commercial quantities of antibody. This is because the resultant lymphoblastoid cell line (LCL) that secretes the human antibodies grow very slowly and as aggregates requiring high concentrations of fetal bovine serum (15%) for support. In order to provide for the production of acceptable levels of antibody amenable for purification in serum-free media, the transformed B-cells may be fused with myeloma cells or heteromyeloma cells. Alternatively, the respective V_H and V_L genes of the antibodies must first be sequenced, linked to their appropriate human constant domains and sub-cloned into a suitable mammalian expression vector for recombinant expression in an appropriate eukaryotic cell line. Thereafter, further antibody versions can be created by for example expression of VH and VL (single chain fragments) derived from a single antibody, as shown hereinafter in Example 9. Alternatively, a single antibody molecule may comprise such fragments derived from two separate antibodies to create a dual-specific or heterodimeric molecule.

Described herein are novel improved antibodies specific for flagellar proteins of P. aeruginosa, which are protective in vivo against said bacteria. One preferred antibody is a fully human IgG antibody based on the variable regions of the human IgM monoclonal antibody LST-007. Other antibodies utilize the variable fragments of mouse LST-001 and LST-002 mAbs, or functional variants thereof, while reducing, i.e. abrogating or decreasing in a statistically or biologically significant manner, antigenicity of the antibodies when presented to the human immune system. In particular, the antibodies are specific for flagellin, type A and/or type B, as evidenced by their specific binding to purified flagellin or to flagella, either isolated or attached to a bacterial cell membrane. Several methods for producing antibodies with reduced antigenicity exist involving the reduction of antigenicity of non-human antibodies by replacing immunogenic regions with non-immunogenic regions, as disclosed in for example US 6881557 (Foote), US 5,869,619 (Studnicka), US 5,225,539 (Winter et al.), Yamashita et al. (2007) and Almagro et al. (2008). For example, antibodies with reduced immunogenicity may be chimeric antibodies consisting of the Fab fragment of a non-human antibody, which contains the CDRs, fused to the constant region of a human antibody. A "humanized antibody" is a chimeric antibody in which a larger part of the protein is derived from human sequences. Commonly, humanized antibodies consists of 5-10% sequences derived from non-human antibodies and 90-95% sequences derived from human antibodies. Thus, the term "humanized antibody" as used herein refers to an antibody comprising the CDRs of a murine monoclonal antibody transplanted onto a human conserved framework region. In order to ensure that the binding specificity is maintained, certain "human" amino acids may be replaced with corresponding amino acids from the equivalent murine sequences.

The antibodies of the present invention may be specific for P. aeruginosa flagella type A or type B, or they may be dual-specific, i.e. they specifically bind to epitopes present on P. aeruginosa type A flagella and to epitopes present on P. aeruginosa type B flagella. In preferred embodiments, the antibody is a human antibody comprising a human IgG constant region fused to human variable regions of an anti- . aeruginosa monoclonal antibody that specifically binds to flagella type B.

Many designs have been put forward for creating dual-specific antibodies. One example is based on the so called "knobs into holes" concept (US Patent No. 7,642,228) in which, in the case of antibodies, a first human IgG heavy chain molecule is associated with a second human IgG heavy chain molecule via an interface, said interface having introduced therein at least one protuberance or cavity such that: (a) the interface of the first human IgG heavy chain molecule comprises a protuberance that is positionable in a cavity in the interface of the second human IgG heavy chain molecule and/or (b) the interface of the first human IgG heavy chain molecule comprises a cavity that accommodates a protuberance of the second human IgG heavy chain molecule. In particular, a knob is created by replacing a small amino side chain at the interface between CH3 domains of the human IgGl Fc region (hinge, constant region 2 and 3 of immunoglobulin heavy chain (CH2 and CH3), GenBank accession no. AF 150959) with a larger one, whereas a hole is constructed by replacing a large side chain with a smaller one. Variable regions having specificity to a first antigen are fused, for example, to the first human IgG heavy chain molecule and other variable regions, having specificity to a second antigen, are fused to the second human IgG heavy chain molecule. Of course, the variable regions having specificity to the first antigen may be fused to the second human IgG heavy chain molecule and then, reciprocally, the variable regions having specificity to the second antigen may be fused to the first human IgG heavy chain molecule. In any case, a heteromultimer is formed having binding-specificity for both the first and the second antigens (Fig. 1). In the case of the present invention, the first antibody binding domain may target P. aeruginosa flagella type A, while the second antibody binding domain may target P. aeruginosa flagella type B.

Alternatively, the dual-specific antibody of the invention may comprise a chimeric antibody specific for P. aeruginosa flagella type A fused via a linker peptide to single chain variable fragments (scFv) that specifically bind to flagella type B of P. aeruginosa or a "mirror image" antibody that comprise a chimeric antibody specific for P. aeruginosa flagella type B fused via a linker peptide to scFvs that specifically bind to flagella type A of P. aeruginosa (Fig. 2). The chimeric antibody component of the dual-specific antibody may be replaced with a humanized or human antibody of the same specificity, and the scFv may be humanized to minimize its immunogenicity or it may be derived from a human monoclonal antibody. The scFv component of the dual-specific antibody comprises one V_H and one V_L fragment linked by a spacer peptide such as a glycine-serine linker.

A different kind of dual-specific antibody, often referred to as a "bispecific" antibody, comprises two scFvs having different binding specificities that are interconnected via a linker peptide. In particular, one scFv that specifically binds flagella type A of P. aeruginosa is linked via a linker peptide to a scFv that specifically binds flagella type B of P. aeruginosa. In principle, a bispecific antibody maybe combined with a chimeric or humanized antibody directed at either flagella of type A or B. Several other formats of dual-specific antibodies, have been disclosed (Kufer et al., 2004) and are all contemplated by the present invention, such as quadroma, an intact antibody wherein each light/heavy chain pair has a different binding specificity, F(ab)₂, essentially a Fab fragment of a quadroma, heterodimeric Fab, diabodies and DVD-Ig, in which two variable fragment pairs, each pair specific to a different antigen, are present on each one of the heavy and light chains (see for example US 20090304693).

Thus, in certain embodiments the dual-specific antibody is selected from the group consisting of:

(i) a dual-specific antibody comprising (a) a scFv that specifically binds to flagella type A, fused to the Fc-region of a first human IgG .heavy chain molecule; and (b) a scFv that specifically binds to flagella type B, fused to the Fc-region of a second human IgG heavy chain molecule, wherein a small amino acid residue of the Fc-region of the first human IgG heavy chain molecule is replaced with a large amino acid residue, and a large amino acid residue of the Fc-region of the second human IgG heavy chain molecule is replaced with a small amino acid residue, and/or a large amino acid residue of the Fc-region of the first human IgG heavy chain molecule is replaced with a small amino acid residue, and a small amino acid residue of the Fc-region of the second human IgG heavy chain molecule is replaced with a large amino acid residue, whereby the large amino acid residue forms a protuberance that is positionable in the cavity formed by the small amino acid residue and thereby the first and the second human IgG heavy chain molecules form a heterodimer that specifically binds both flagella type A and flagella type B; (ii) a dual-specific antibody comprising an antibody composed of a human IgG constant region fused to variable regions of an anti-P.

aeruginosa monoclonal antibody that specifically binds to flagella type A, wherein said antibody is connected at its terminal Fc-region to two single chain variable fragments (scFvs) that specifically bind to flagella type B of

P. aeruginosa.

(iii) a dual-specific antibody comprising an antibody composed of a human IgG constant region fused to variable regions of an anti-P.

aeruginosa monoclonal antibody that specifically binds to flagella type B of P. aeruginosa, wherein said antibody is connected at its terminal Fc-region to two scFvs that specifically bind to flagella type A of P. aeruginosa; and

(iv) a dual-specific antibody comprising two scFvs that specifically bind to flagella type A of P. aeruginosa fused via a linker peptide to two other scFvs that specifically bind to flagella type B of P. aeruginosa.

In certain embodiments the antibody is a dual-specific antibody of (i) or (ii).

In one embodiment the antibody is a chimeric antibody comprising a human IgG or IgM constant region fused to mouse variable regions of an anti-P. aeruginosa monoclonal antibody that specifically binds to flagella type A or type B of P. aeruginosa, and in another embodiment the antibody is a humanized antibody comprising a human IgG or IgM constant region fused to humanized mouse variable regions of an anti-P. aeruginosa monoclonal antibody that specifically binds to flagella type A or type B of P. aeruginosa. Obviously, the human IgG heavy and light constant domains may be derived from any one of IgGl, IgG2, IgG3, and IgG4 subclasses of human IgG antibodies, and it may comprise one, two or three intact or truncated constant domains (CH1-3), which may optionally be mutated to alter effector function or provide for heteromultimer formation, or modified post-translationally (e.g. glycosylation) to improve the half-life of the antibody. In certain embodiments the IgG constant region is a human IgGl constant region.

Alternatively, the human constant domains may consist of the human IgM type composed of CL for the light chain and Ο_μ1 , Ο_μ2, Ο_μ3, Ο_μ4 for the heavy chain with an adjoining J chain to permit antibody polymerization and formation of a pentameric species. In certain embodiments, the IgM constant region of the antibody of the present invention is a human ¾μ and a human Ig/c constant region, for example as shown in Example 6 herein below.

As shown below in Examples 1 and 6, and in accordance with the present invention, the genes encoding for the variable regions of the mouse monoclonal antibodies LST-001 and LST-002 and the human monoclonal antibody LST-007 have been cloned and their nucleotide sequences have been determined. The CDR regions present within the variable regions of LST-001 and LST-002 may be identified, for example, by using an algorithm disclosed by Wu and Kabat (1970), hereinafter referred to as "Kabat", or an algorithm known as the IMGT/V -QUEST disclosed by Brochet et al (2008) or according to Chothia (1987; 1992). However, it should be understood that any combination of CDRs and framework regions that confers to the antibody specificity towards P. aeruginosa flagellin type A or B or capability to compete with LST-001, LST-002 or LST-007 for binding to P. aeruginosa flagellin type A or B is encompassed by the present invention as long as the constant regions are not mouse constant regions.

In certain embodiments the variable regions or scFv that specifically bind to flagella type B comprises (a) a human VH fragment comprising CDRs present in SEQ ID NO: 1 ; and (b) a human VL fragment comprising CDRs present in SEQ ID NO: 2.

In certain embodiments, CDR1, CDR2 and CDR3 of the heavy chain (VH) of LST- 007 as defined by IMGT/V-QUEST comprises the amino acid sequence set forth in SEQ ID NO: 3, SEQ ID NO: 4 and SEQ ID NO: 5, respectively, while CDR1, CDR2 and CDR3 of the heavy chain (VH) of LST-007 as defined by Kabat comprises the amino acid sequence set forth in SEQ ID NO: 6, SEQ ID NO: 7 and SEQ ID NO: 8, respectively. The CDR1, CDR2 and CDR3 of the V_L fragment of LST-007 as defined by IMGT/V -QUEST comprises the amino acid sequence set forth in SEQ ID NO: 9, the amino acid sequence AAS and the amino acid sequence set forth in SEQ ID NO: 10, respectively, while CDR1 , CDR2 and CDR3 of the VL fragment as defined by Kabat comprises the amino acid sequence set forth in SEQ ID NO: 11, SEQ ID NO: 12 and SEQ ID NO: 10, respectively (an identical CDR3 was predicted by the two algorithms).

In preferred embodiments, the human VH fragment comprises the amino acid sequence of SEQ ID NO: 13; and the VL fragment comprising the amino acid sequence of SEQ ID NO: 14, and in particular the human antibody comprises the VH fragment fused to a human IgGl heavy constant domain as set forth in SEQ ID NO: 15 and the VL fragment fused to a human IgGl light constant domain as set forth in SEQ ID NO: 16.

In certain embodiments, the antibody comprises the variable region or scFv that specifically binds to flagella type A comprising (a) a mouse V_H fragment comprising CDRs present in SEQ ID NO: 17; and (b) a mouse V_L fragment comprising CDRs present in SEQ ID NO: 18.

In certain embodiments, CDRl, CDR2 and CDR3 of the heavy chain (V_H) of LST- 001 as defined by IMGT/V-QUEST comprises the amino acid sequence set forth in SEQ ID NO: 19, SEQ ID NO: 20 and SEQ ID NO: 21, respectively, while CDRl , CDR2 and CDR3 of the heavy chain (VH) of LST-001 as defined by Kabat comprises the amino acid sequence set forth in SEQ ID NO: 22, SEQ ID NO: 23 and SEQ ID NO: 24, respectively. The CDRl, CDR2 and CDR3 of the V_L fragment of LST-001 as defined by IMGT/V- QUEST comprises the amino acid sequence set forth in SEQ ID NO: 25, the amino acid sequence WAS and the amino acid sequence set forth in SEQ ID NO: 26, respectively, while CDRl, CDR2 and CDR3 of the VL fragment as defined by Kabat comprises the amino acid sequence set forth in SEQ ID NO: 27, SEQ ID NO: 28 and SEQ ID NO: 26, respectively (an identical CDR3 was predicted by the two algorithms).

In certain embodiments, the antibody comprises the VH fragment comprising the amino acid sequence of SEQ ID NO: 17 or SEQ ID NO: 29 and the V_L fragment comprising the amino acid sequence of SEQ ID NO: 18 or SEQ ID NO: 30.

The variable region of the VH and VL of LST-002 that is specific to P. aeruginosa flagellin type B, comprise amino acid sequences of SEQ ID NO: 31 and SEQ ID NO: 32 or SEQ ID NO: 33, respectively. Two alternative sequences are designated for the VL of LST- 002 because, as shown herein in Example 2, two variant sequences were determined for this fragment; the amino acid sequence of SEQ ID NO: 32 was present in the majority (five out of seven) of the clones sequenced, while the amino acid sequence of SEQ ID NO: 33 was present in the minority (two out of seven) of the clones sequenced.

In certain embodiments, CDRl, CDR2 and CDR3 of the heavy chain (V_H) of LST- 002 as defined by IMGT/V -QUEST comprises the amino acid sequence set forth in SEQ ID NO: 34, SEQ ID NO: 35 and SEQ ID NO: 36 respectively, while CDRl , CDR2 and CDR3 of the heavy chain (VH) of LST-002 as defined by Kabat comprises the amino acid sequence set forth in SEQ ID NO: 37, SEQ ID NO: 38 and SEQ ID NO: 39 respectively. The CDRl, CDR2, and CDR3 of the majority V_L fragment as defined by IMGT V-QUEST comprises the amino acid sequence set forth in SEQ ID NO: 40, the amino acid sequence WTS and the amino acid sequence set forth in SEQ ID NO: 41 , respectively, while CDR1 , CDR2 and CDR3 of the majority V_L fragment as defined by Kabat comprises the amino acid sequence set forth in SEQ ID NO: 42, SEQ ID NO: 43 and SEQ ID NO: 41 , respectively (an identical CDR3 was predicted by the two algorithms). The CDR1, CDR2, and CDR3 of the minority V_L fragment as defined by IMGT/V-QUEST and Kabat comprises the amino acid sequence set forth in SEQ ID NO: 44, the amino acid sequence GAS and the amino acid sequence set forth in SEQ ID NO: 45, respectively, while CDR1, CDR2 and CDR3 of the minority VL fragment as defined by Kabat comprises the amino acid sequence set forth in SEQ ID NO: 46, SEQ ID NO: 47 and SEQ ID NO: 48, respectively.

In certain embodiments, the antibody comprises the VH fragment comprising the amino acid sequence of SEQ ID NO: 49; and the VL fragment comprising the amino acid sequence of SEQ ID NO: 50, or the VL fragment comprising the amino acid sequence of SEQ ID NO: 51.

Consequently, the antibody of the invention may comprise variable regions within framework regions that place the CDR regions in a correct three dimensional position enabling high affinity binding of the specific antigen. In particular, the variable regions as defined by IMGT/V-QUEST or Kabat comprise the CDRs of the LST-001, LST-002 or LST-007 monoclonal antibody within framework regions that place the CDR regions in a correct three dimensional position enabling high affinity binding of P. aeruginosa flagellin type A or type B. As taught by Winter et al., Studnicka and Queen (US 5,693,761), there exist many alternative sets of framework regions that in conjunction with a specific set of CDR regions confer specific binding to an antigen, and anyone of the possible framework regions that will allow specific binding of the antibody of the invention to P. aeruginosa flagellin type A or type B are considered by the invention. Alternatively, the antibody may comprise a variable region comprising the specific CDRs and the specific framework region of the LST-001, LST-002 or LST-007 monoclonal antibodies.

The framework regions may have at least 80%, 85%, 90%, or 95% identity to the framework region comprising the amino acid sequences of the LST-001 , LST-002 or LST- 007 monoclonal antibodies, and each one of the CDR regions may have at least 80%, 85%, 90%, or 95% identity to the corresponding CDR regions of the LST-001, LST-002 or LST- 007 monoclonal antibodies, respectively, as long as the variable region comprising these CDRs and framework regions confers specific binding to P. aeruginosa flagellin type A or type B. In the case of the human antibody LST-007, its framework regions or even CDR's may be modified to remove putative T-cell epitopes, so called de-irnmunization, while maintaining binding specificity towards flagellin type B.

The antibody of the invention may have variable regions including the CDRs and the interspersing framework regions that are identical to that of LST-001 or LST-002. The resultant chimeric mAbs derived from the LST-001 and LST-002 niAbs are referred to herein as LST-003 and LST-004 respectively.

As described below in Example 2, sequences of humanized antibodies were predicted by using a certain algorithm that compares the murine antibody variable regions with a database of human germline Ig variable regions. In this way, the specificity of the murine antibody to its antigen is maintained in the predicted humanized antibody. It should be understood that any humanized sequences arrived at by using alternative algorithms, for example those described by Yamashita (2007) and Almagro (2008), that still maintain the original antigen binding specificity may be used in the present invention.

The resultant humanized mAbs derived from the LST-001 and LST-002 mAbs have been coined herein LST-005 and LST-006 respectively. In view of the above, in one embodiment, the antibody that specifically binds to flagellin type A, is a humanized or a dual-specific antibody, comprising the VH fragment consisting of SEQ ID NO: 52 and the VL fragment consisting of SEQ ID NO: 53. In another embodiment, the antibody that specifically binds to flagellin type B is a humanized or a dual-specific antibody, comprising a VH fragment consisting of SEQ ID NO: 54 and the VL fragment consisting of SEQ ID NO: 55 or SEQ ID NO: 56. Each one of the humanized V_H and V_L fragments may have at least 80%, 85%, 90%, or 95% identity to the corresponding VH and VL fragments of the LST-001 or LST-002 monoclonal antibodies, respectively, as long as the variable regions comprising these sequences confer specific binding to P. aeruginosa flagellin type A or type B or capability to compete with LST-001 or LST-002 for binding to P. aeruginosa flagellin type A or type B.

In certain embodiments the antibody of the invention is a dual-specific antibody comprising a scFv that specifically binds to flagella type A, fused to the Fc-region of a first human IgG heavy chain molecule; and (b) a scFv that specifically binds to flagella type B, fused to the Fc-region of a second human IgG heavy chain molecule, wherein a small amino acid residue of the CH3 domain of the first human IgG heavy chain molecule is replaced with a large amino acid residue, and a large amino acid residue of the CH3 domain of the second human IgG heavy chain molecule is replaced with a small amino acid residue, and/or a large amino acid residue of the Fc-region of the first human IgG heavy chain molecule is replaced with a small amino acid residue, and a small amino acid residue of the Fc-region of the second human IgG heavy chain molecule is replaced with a large amino acid residue, whereby the large amino acid residue forms a protuberance that is positionable in the cavity formed by the small amino acid residue and thereby the first and the second human IgG heavy chain molecules form a heterodimer that specifically binds both flagella type A and flagella type B,

As mentioned above, the variable fragments comprising the scFv of the dual- specific antibody are linked by a peptide termed "spacer peptide", while the peptide used to link the scFv with the constant region of the dual-specific antibody or to link two scFvs in a "bispecific" antibody is termed herein " linker peptide".

Thus, in certain embodiments, each one of the scFvs of the dual-specific antibody is composed of a V_H fragment fused via a spacer peptide to a V_L fragment and the length of the spacer peptide is selected from the range of 5 to 20 amino acid residues or 7 to 15 amino acid residues.

In one embodiment, the spacer peptide of the dual-specific antibody consists of three consecutive peptides of the sequence of SEQ ID NO: 57 (GGGGS) or one peptide having the amino acid sequence of SEQ ID NO: 58 (GGGSAAA).

In certain embodiments, the scFvs are connected via a linker peptide to the C- terminus of the Fc-region of an IgG heavy chain molecule and the length of the linker peptide is selected from the range of 5 to 15 amino acid residues or 9 amino acid residues. In one embodiment, the linker peptide is of the amino acid sequence of SEQ ID NO: 59 (PGSAGGSGG) or it consists of two or three consecutive peptides of SEQ ID NO: 57.

The "knob" and the "hole" in the dual-specific "knobs-into-holes" type antibody are created by the presence of a "large" amino acid (that has replaced a "small" amino acid of the wild-type sequence) and a "small" amino acid (that has replaced a "large" amino acid of the wild-type sequence), respectively. Gly, Ala, Ser, Pro, Val, Thr, Cys are considered in the art to be small amino acid residues, while Lys, Gin, Glu, Met, Phe, Tyr, Tryp are considered to be large amino acid residues (see for example Livingstone & Barton (1993) Protein sequence alignments: a strategy for the hierarchical analysis of residue conservation. CABIOS 9, 745-756). In certain embodiments, the small amino acid residue is threonine and the large amino acid residue is tyrosine.

In certain embodiments, the threonine at position 366 (T366) of the CH3 domain of the first human IgG heavy chain molecule is replaced with a tyrosine, and tyrosine at position 407 (Y407) of the CH3 domain of the second human IgG heavy chain molecule is replaced with a threonine; and/or Y407 of the CH3 domain of the first human IgG heavy chain molecule is replaced with a threonine and T366 of the CH3 domain of the second human IgG heavy chain molecule is replaced with a tyrosine.

In certain embodiments, the scFv of the dual-specific antibody that specifically binds to flagella type A comprises a VH fragment comprising the amino acid sequence of SEQ ID NO: 29 or SEQ ID NO: 17 and the V_L fragment comprising the amino acid sequence of SEQ ID NO: 18 or SEQ ID NO: 30, and the scFv that specifically binds to flagella type B comprises a VH fragment comprising the amino acid sequence of SEQ ID NO: 31 or SEQ ID NO: 49 and the V_L fragment comprising the amino acid sequence of SEQ ID NO: 50 or SEQ ID NO: 32.

In particular, the dual-specific antibody comprises a first amino acid chain comprising the scFv that specifically binds to flagella type A linked via a linker peptide to a human IgGl Fc-fragment, said first amino acid chain having the amino acid sequence of SEQ ID NO: 60; and a second amino acid chain comprising the scFv that specifically binds to flagella type B linked via a linker peptide to a human IgGl Fc-fragment, said second amino acid chain having the amino acid sequence of SEQ ID NO: 61.

In certain embodiments the antibody of the invention is a dual-specific antibody comprising a chimeric antibody composed of a human IgG constant region fused to mouse variable regions of an anti-P. aeruginosa monoclonal antibody that specifically binds to flagella type A, wherein said chimeric antibody is connected at its Fc-region to scFvs that specifically bind to flagella type B of P. aeruginosa.

Each scFv is composed of a VH fragment fused via a spacer peptide to a VL fragment and the length of the spacer peptide is selected from the range of 5 to 20 amino acid residues or 7 to 15 amino acid residues. In particular, the spacer peptide is of the amino acid sequence of SEQ ID NO: 58 .

Also, each of the scFvs is connected via a linker peptide to the C-terminus of the Fc-region of an IgG heavy chain molecule and the length of the linker peptide is selected from the range of 5 to 15 amino acid residues or 9 amino acid residues. In particular, the linker peptide consists of two consecutive peptides of the sequence of SEQ ID NO: 57.

In preferred embodiments, the chimeric antibody, that specifically binds to flagella type A of the dual-specific antibody, comprises the VH fragment comprising SEQ ID NO: 29 and the V_L fragment comprising SEQ ID NO: 30, fused to a human IgGl Fc-fragment, and each one of said two scFvs comprises the VH fragment comprising SEQ ID NOs: 31 and the VL fragment comprising SEQ ID NO: 32.

In particular, the dual-specific antibody comprises an amino acid chain comprising the heavy chain of said chimeric antibody fused to the scFvs, said amino acid chain comprising the amino acid sequence of SEQ ID NO: 62, and the light chain of LST-001 linked to a human IgGl Fc-fragment.

The invention is also directed to an antibody of the invention or an antigen binding fragment thereof covalently linked, optionally via a cleavable linker, to an antibiotic agent. Non-limiting examples of antibiotic agents are Amikacin, Ampicillin/Sulbactam, Amoxicillin/Calvulanic acid, Aztreonam, Cefepime, Cefotaxime, Ceftazidime, Chloramphenical, Ciprofloxacin, Colistin, Doripenem, Gentamicin, Imipenem, Levofloxacin, Meropenem, Minocycline, Piperacillin, Piperacillin/Tazobactam, Ticracillin, Tigecycline, Tobramycin, Trimethoprim-Sulfamethoxazole.

The invention is further directed to an antibody of the invention or an antigen binding fragment thereof attached to a tag useful for purification and detection, or a chemical moiety that alters the physical properties of the antibody such as stability. Thus, the antibody of the invention may be fused to a tag that enables efficient purification such as, but not limited to, a Histidine-tag or an antigenic peptide tag, and it may be covalently linked to a nonproteinaceous polymer, such as polyethyleneglycol (PEG) to increase the stability of the antibody and change the rate at which the antibody is eliminated from a subject after administration thereto. The PEG is a substituted or unsubstituted polymer having a molecular weight of from about 1000 to about 5000 Da or more. Other non- limiting examples of such polymers are poly (propyleneglycol), or poly (oxyalkylene). An antibody attached to such a tag or chemical moiety is referred to herein as a "derivative" of the unmodified antibody.

Antigen binding fragments of the human, chimeric or humanized antibody of the present invention, or derivatives thereof, wherein said fragment binds flagellin type A or type B of P. aeruginosa bacteria are also contemplated by the invention. The antigen- binding fragments can be prepared from full-length antibody isolates, for example, by digestion with proteases, or they may be produced using standard recombinant DNA methodology.

In another aspect, the present invention relates to a nucleic acid molecule comprising nucleotide sequences which encode an antibody according to the present invention or an antigen binding fragment thereof.

In certain embodiments, the nucleic acid molecule encodes a human antibody that specifically binds to flagella type B, comprising the nucleic acid sequence of SEQ ID NO: 63 encoding the heavy chain of said human antibody, and the nucleic acid sequence of SEQ ID NO: 64 encoding the light chain of said human antibody.

In certain embodiments the nucleic acid molecule encodes a dual-specific antibody that specifically binds to flagella type A and B, comprising the nucleic acid sequence of SEQ ID NO: 65 encoding for a scFv that specifically binds to flagella type A fused to a human IgGl fragment, and the nucleic acid sequence of SEQ ID NO: 66, encoding for a scFv that specifically binds to flagella type B fused to a human IgGl fragment.

In certain embodiments the nucleic acid molecule encodes a dual-specific antibody that specifically binds to flagella type A and B, comprising the nucleic acid sequence of SEQ ID NO: 67 encoding the heavy chain of a chimeric antibody that specifically bind to flagella type A fused to a scFv that specifically binds to flagella type B, and the nucleic acid sequence of SEQ ID NO: 68 encoding the light chain of an antibody that specifically binds to flagella type A.

In certain embodiments the nucleic acid molecule encodes a chimeric antibody selected from:

(i) a chimeric antibody that specifically binds to flagella type A, comprising the nucleic acid sequence of SEQ ID NO: 69 encoding the heavy chain of said chimeric antibody, and the nucleic acid sequence of SEQ ID NO: 70 encoding the light chain of said chimeric antibody; or

(ii) a chimeric antibody that specifically binds to flagella type B, comprising the nucleic acid sequence of SEQ ID NO: 71 encoding the heavy chain of said chimeric antibody, and the nucleic acid sequence of SEQ ID NO: 72 encoding the light chain of said chimeric antibody. In still another aspect, the present invention relates to a vector comprising the nucleic acid molecule defined hereinbefore, and optionally a nucleotide sequence encoding a heterologous polypeptide such as an antigenic peptide tag or enzyme, operably linked to at least one expression control sequence such as a promoter capable of driving the expression of said nucleic acid molecule.

The present invention further relates to a host cell such as mouse myeloma NSO and Chinese Hamster Ovary cells (CHO), or plant cells, such as tobacco, carrot and rice cells, which comprises at least one vector as defined hereinbefore and which produces an antibody or antigen binding fragment thereof according to the invention.

In certain embodiments, the human antibody is an IgM antibody that specifically binds to flagella type B, and comprises the VH fragment comprising the amino acid sequence of SEQ ID NO: 13; and the VL fragment comprising the amino acid sequence of SEQ ID NO: 14.

In one aspect, the present invention relates to a pharmaceutical composition comprising at least one antibody or antigen binding fragment thereof, or a derivative of said at least one antibody or antigen binding fragment thereof according to the invention, preferably a human antibody or a dual-specific antibody, and a pharmaceutically acceptable carrier. The term "carrier" refers to a diluent, adjuvant, excipient, or vehicle with which the therapeutic, i.e. the antibody, is administered. The carriers in the pharmaceutical composition may comprise a binder, such as microcrystalline cellulose, polyvinylpyrrolidone (polyvidone or povidone), gum tragacanth, gelatin, starch, lactose or lactose monohydrate; a disintegrating agent, such as alginic acid, maize starch and the like; a lubricant or surfactant, such as magnesium stearate, or sodium lauryl sulphate; and a glidant, such as colloidal silicon dioxide. The term "derivative" as used herein refers to antibodies modified by addition of molecules such as antibiotic agents, artificial polymers such as PEG, or post-translational modifications such as the attachment of acetate, phosphate, lipids or carbohydrates.

Any patient who is immunosuppressed and/or admitted to the Intensive Care Unit (ICU) could receive an antibody targeting P. Aeruginosa flagella prophylactically even in the absence of an underlying P. Aeruginosa infection, especially patients who are intubated or catheterized. P. Aeruginosa can also cause infections in immunocompetent patients not associated with the ICU setting. Such infections include, but are not limited to, those of the eye (keratitis), and ear infections ("swimmer's ear" or otitis externa). The flagella of P. Aeruginosa aids it in invading the host and disseminating systemically. It has been found in accordance with the present invention that the antibodies disclosed herein are capable of impeding the motility P. Aeruginosa by specifically binding to it (see Example 2 hereinafter). This implies that the antibodies of the present invention may be active in creams or gels intended for treatment of topical P. aeruginosa infections associated with wounds, eyes, ears, etc.

It has further been found in accordance with the present invention that LST-002 antibodies are efficacious in treating multi-drug resistant P. Aeruginosa infections in a pneumonia model in mice (Example 10).

It has further been found in accordance with the present invention that LST-002 antibodies are efficacious in the treatment of multidrug-resistant P. aeruginosa model of thigh muscle infection. In Example 10 hereinafter, it is shown that administration of LST- 002 antibodies to neutropenic mice totally prevented infection lesions in thigh muscle infected with Ka02, while in saline treated mice the lesions increased 1.8 and 2.7 fold as compared with saline injected muscle, 5 and 7 days after infection, respectively.

Since it is also shown hereinafter in the examples that the chimeric and dual- specific antibodies of the present invention comprising the variable regions of LST-002 indeed binds specifically and with high affinity to flagella type B, and since it is known (US 4,834,976) that the human IgM LST-007 does so too, it can be expected that the human, chimeric, humanized and dual-specific antibodies of the present invention comprising the variable regions of LST-002 or LST-007, or based thereon, would be at least as effective in treating all P. Aeruginosa infections harboring the type B flagellin protein

Moreover, it is shown hereinafter in the examples that the chimeric and dual- specific antibodies of the present invention comprising the variable regions of LST-001 indeed binds specifically and with high affinity to flagella type A; thus, drawing from the positive experience with LST-002 antibodies, it can be expected that also the chimeric, humanized and dual-specific antibodies of the present invention comprising the variable regions of LST-001, or based thereon, would be at least as effective as LST-002 in treating P. Aeruginosa infections harboring the type A flagellin protein.

Interestingly, the dual-specific antibody LST-010 was shown hereinafter in Example 7 to be capable of binding mixed P. Aeruginosa populations (flagellin types A and B) at its presumed KD (InM or 0.2 / g/ml), thereby representing a therapeutic target concentration. As mentioned above, LST-002 was capable of treating and preventing P.

Aeruginosa infections in mice; the therapeutic dose given in Example 10 was on average

35 mg LST-002 antibody per kg body weight. An expected approximate equivalent dose for administration to a human can be calculated using known formulas to be 2.8 mg/kg or 170 mg for a 60 kg adult and 280 mg for a 100 kg adult. Thus, the dose for systemic administration in a human should be in the range of 10 mg to 3000 mg. The frequency of administration is, for example, once a week.

The present invention thus further relates to methods for treatment of an infection caused by P. aeruginosa, comprising administering to a subject in need of such treatment a therapeutically effective amount of at least one, i.e. one or more, antibodies according to the present invention or antigen binding fragment thereof. For example, a mixture of two mono-specific antibodies, an anti- P. aeruginosa type A flagella antibody and an anti- P. aeruginosa type B flagella antibody, or one dual-specific antibody, may be administered to treat a mixed infection of P. aeruginosa type A and P. aeruginosa type B.

The present invention also relates to prophylactic methods for reduction of a risk of

P. aeruginosa infection in a subject susceptible to P. aeruginosa that may be exposed to P. aeruginosa, comprising administering to said patient a therapeutically effective amount of at least one antibody.

Similarly, the present invention further relates to the antibody of the present invention for use in treating an infection caused by P. aeruginosa, comprising administering to a subject in need of such treatment a therapeutically effective amount of at least one antibody according to the present invention or antigen binding fragment thereof; and to the antibody of the present invention for use in reducing a risk of P. aeruginosa infection in a subject susceptible to P. aeruginosa that may be exposed to P. aeruginosa, comprising administering to said patient a therapeutically effective amount of at least one antibody.

The term "prophylactic method" is used herein interchangeable with the term "preventive method" and refer to the institution of measures to protect a person from a disease to which he or she has been, or may be, exposed. The preventive or prophylactic action is the prevention of, the delay of, or the interference with, the establishment of a P. aeruginosa infection, by administering the antibodies to a susceptible subject that has not an established P. aeruginosa infection. The treatment, i.e. the therapeutic action, is the treatment obtained by administering the antibodies to a subject having an established P. aeruginosa infection and thus diminishing the number of bacteria or eliminating the bacteria, slowing or stopping the proliferation of the bacteria, slowing or preventing the dissemination of the bacteria, and attenuating or eliminating the symptoms of the infection, in said subject.

Examples of subjects amenable for treatment according to the present invention are subjects suffering from a disease selected from the group consisting of chronic colonization in the lungs, bacteremia, sepsis, surgical wound infection, urinary tract infection, ventilator-associated pneumonia, non-ventilator associated pneumonia, obstructive pulmonary dysfunction (COPD), bronchiectasis, keratitis, and ear infection (otitis externa), and examples of subjects susceptible to infection are immune suppressed subjects selected from the group consisting of cancer patients, patients awaiting transplantation, AIDS patients, burn patients and very low or extremely low birth weight infants (VLBWI and ELBWI, respectively.) VLBWI is defined as a birth weight less than 1500 g, and ELBW is defined as a birth weight less than 1000 g (2 lb, 3 oz). Most extremely low birth weight infants are also the youngest of premature newborns, usually born at 27 weeks' gestational age or younger.

The term "bacteremia" as used herein refers to the presence of bacteria in the blood, and the term "sepsis" refers to a medical condition characterized by a whole-body inflammatory state and the presence of a known or suspected infection in the blood, urine, lungs, skin, or other tissues.

The antibody or antigen binding fragment thereof, or the derivative of said antibody or antigen binding fragment thereof, may be administered by intravenous administration as a bolus or by continuous infusion over a period of time, by intramuscular, intraperitoneal, intracerebrospinal, subcutaneous, intra-articular, intrasynovial, intrathecal, oral, topical, or inhalation routes, and it may be administered in combination with the administration of an antibiotic drug, an antibody that targets P. aeruginosa excluding flagellin, such as, but not limited to, anti-LPS or anti-PcrV antibodies (PcrV is one component of the P. aeruginosa type III secretion system), or an anti-inflammatory antibody, such as an anti-TNF-a, an anti-IL8 or an anti-ILl antibody.

The antibodies disclosed herein would be most valuable when used in methods of treatment of multidrug-resistant P. aeruginosa that is resistant to at least two drugs, non- limiting examples of which are: strains resistant to an aminoglycoside, such as gentamicin, amikacin, tobramycin; strains resistant to a quinolone such as ciprofloxacin, levofloxacin, moxifloxacin, clinafloxacin, trovafloxacin and sitafloxacin; strains resistant to a cephalosporin, such as ceftazidime, cefepime, cefpirome, cefuroxime, ceftriaxone, cefotaxime and ceftobiprole; strains resistant to a penicillin such as ampicillin - optionally in combination with sulbactam, ureidopenicillin, piperacillin - optionally in combination with tazobactam, and ticarcillin; strains resistant to a carbapenem, such as biapenem, meropenem, imipenem, ertapenem, doripenem, panipenem and faropenem; strains resistant to a polymyxin, such as polymyxin B and colistin; or a monobactam, such as aztreonam, and pandrug-resistant P. aeruginosa that is resistant to all classes of available antimicrobial agents.

In still another aspect, the present invention relates to a method for diagnosing and monitoring a host response to P. aeruginosa infection, comprising (a) obtaining a patient sample; (b) contacting said patient sample with an antibody of the invention; and (c) measuring binding of said antibody to said patient sample, wherein detection of binding of said antibody to said patient sample indicates the presence or level of multi-drug or pan- drug resistant P. aeruginosa in the patient sample. The term "presence" as used herein in the context of the method for diagnosis and monitoring a host response to P. aeruginosa infection indicates that the method is a qualitative method giving information only regarding the presence or absence of said bacteria, while the term "level" indicates that the method is a qualitative method producing information regarding the amount or number of cells present in the patient sample. Thus, monitoring is achieved by assessing the level of bacteria over time subsequent to the initiation of administration of the antibodies.

In order to facilitate convenient detection of the antibody, it may be labeled with a detectable label, such as an antigenic peptide tag, fluorophores, enzymes, radioisotopes and micro- and nano- particles. Non-limiting examples of useful fluorophores are fluorescein, rhodamine and alexa; useful enzymes are for example Horse radish peroxidase and alkaline phosphatase; useful radioisotopes are for example H³, P³² and S³⁵; and particles often used to immobilize antibodies are magnetic or polystyrene micro- and nano- particles.

In yet another aspect, the present invention is directed to a kit for detecting P. aeruginosa in a biological sample comprising a container containing an antibody according to the present invention; a container containing a set of reagents required for the detection and quantification of said antibody; and instructions for use.

The invention will now be illustrated by the following non-limitative examples. EXAMPLES

Example 1. Sequencing the variable fragments of the mouse LST-001 and LST-002 mAbs.

The dideoxy termination methodology of polynucleotide sequencing of Sanger et al. (1977) was used to determine the sequences encoding the variable fragments of the heavy and light chains of the mouse LST-001 and LST-002 mAbs, and of the human LST- 007. The sequences of the polypeptides are as follows:

(Ϊ) mAbs LST-001

(la) The full V_H sequence (SEQ ID NO: 29)

(lb) The mature V_H sequence (SEQ ID NO: 17)

(lc) The CDR sequences of the V_H sequence as defined by IMGT/V-QUEST:

CDR1 (LST-001 V_H) (SEQ ID NO: 19)

CDR2 (LST-001 V_H) (SEQ ID NO: 20)

CDR3 (LST-001 V_H) (SEQ ID NO: 21)

(Id) The CDR sequences of the VH sequence as defined by Rabat:

CDR1 (LST-001 V_H) (SEQ ID NO: 22)

CDR2 (LST-001 VH) (SEQ ID NO: 23)

CDR3 (LST-001 VH) (SEQ ID NO: 24)

(le) The full V_L sequence (SEQ ID NO: 30)

(If) The mature V_L sequence. (SEQ ID NO: 18)

(Ig) The CDR sequences of the Vi sequence as defined by IMGT V-QUEST:

CDR1 (LST-001 V_L) (SEQ ID NO: 25)

CDR2 (LST-001 V_L) (the amino acid sequence WAS)

CDR3 (LST-001 V_L) (SEQ ID NO: 26)

(lh) The CDR sequences of the Vi sequence as defined by Rabat:

CDR1 (LST-001 V_L) (SEQ ID NO: 27)

CDR2 (LST-001 V_L) (SEQ ID NO: 28)

CDR3 (LST-001 V_L) (SEQ ID NO: 26)

(2) mAbs LST-002

For the V_L of LST-002, two separate yet similar sequences were discovered; one sequence is a consensus sequence from five separate recombinant clones and is termed a "majority sequence"; the other sequence is a consensus sequence from two separate recombinant clones and is termed a "minority sequence". (2a) The full V_H sequence (SEQ ID NO: 49)

(2b) The mature V_H sequence. (SEQ ID NO: 31)

(2c) The CDR sequences of the V_H sequence as defined by IMGT/V-QUEST: CDR1 (LST-002 V_H) (SEQ ID NO: 34)

CDR2 (LST-002 V_H) (SEQ ID NO: 35)

CDR3 (LST-002 V_H) (SEQ ID NO: 36)

(2d) The CDR sequences of the VH sequence as defined by Rabat:

CDR1 (LST-002 VH) (SEQ ID NO: 37)

CDR2 (LST-002 VH) (SEQ ID NO: 38)

CDR3 (LST-002 VH) (SEQ ID NO: 39)

(2e) The full V_L "majority" sequence (SEQ ID NO: 50)

(2f) The mature VL "majority" sequence (SEQ ID NO: 32)

(2g) The CDR sequences of the Vi "majority" sequence as defined by IMGT V- QUEST

CDR1 (LST-002 V_Lmaj) (SEQ ID NO: 40)

CDR2 (LST-002 V_Lmaj) (the amino acid sequence WTS)

CDR3 (LST-002 V_Lmaj) (SEQ ID NO: 41)

(2h) The CDR sequences of the VL "majority" sequence as defined by Rabat CDR1 (LST-002 V_Lmaj) (SEQ ID NO: 42)

CDR2 (LST-002 V_Lmaj) (SEQ ID NO: 43)

CDR3 (LST-002 V_Lmaj) (SEQ ID NO: 41)

(2i) The full V_L "minority " sequence (SEQ ID NO: 51)

(2j) The mature V_L "minority " sequence (SEQ ID NO: 33)

(2k) The CDR sequences of the VL "minority" sequence as defined by IMGT/V- QUEST

CDR1 (LST-002 V_Lmin) (SEQ ID NO: 44)

CDR2 (LST-002 V_Lmi_n) (the amino acid sequence GAS)

CDR3 (LST-002 V_Lmin) (SEQ ID NO: 45)

(21) The CDR sequences of the VL "minority " sequence as defined by Rabat CDR1 (LST-002 V_Lmin) (SEQ ID NO: 46)

CDR2 (LST-002 V_Lmin) (SEQ ID NO: 47) CDR3 (LST-002 V_Lmin) (SEQ ID NO: 48)

Example 2. LST-001 inhibits motility of P. Aeruginosa bearing flagellin type A but not P. Aeruginosa bearing flagellin type B.

A commercially available, P. aeruginosa strain derived from a blood culture (ATCC 27853) was grown overnight at 37°C in LB media. This strain was shown by us to be a type A flagellin strain and is sensitive to all antibiotics. After overnight culture, bacteria were diluted in fresh LB media to generate an OD₆oo of ~ 0.2.

To separate wells of 6-well sterile culture plate, a liquefied solution of 3 ml of soft motility agar (0.3%) was added containing LST-001 mAb (0.3-20 μg/ml) or a negative control LST-002 mAb (20 μg/ml). Thereafter, the mAb-impregnated agar was allowed to solidify for 2 hrs at room temperature after which time, 10 μΐ of bacterial suspension at OD₆oo ⁼ 0-2 was stabbed into the centre of the wells, at a depth of approximately 2/3 into the agar. Plates were transferred to an incubator at 30°C for 18-22 hr to allow bacterial growth and swarming (i.e. motility), the latter indicated by the presence of a halo formation encircling the localized bacterial growth. The diameter of the halo formation was measured for each well enabling determination of the inhibitory effect of the mAb of bacterial motility (Fig. 3A). As can be clearly seen, at 20 μg/ml, LST-001 is capable of inhibiting close to 100% of the bacterial swarming.

Using an alternative in-vitro motility technology, the effects of LST-002 on motility of a MDR P. aeruginosa (PA) strain (Ka02) was performed. In 10 cm plastic culture dishes, a 6 ml underlayer of 0.6% agar in LB was poured and allowed to solidify. During this period, 6 ml of LB was aliquoted into separate tubes to which LST-002 was added at final concentrations 80, 20, 5, 1.25, 0.3 and 0 μg/ml. Thereafter, 10 μΐ of bacterial culture (containing ~ 20-100 cfu) was added followed by 6 mis of 0.6% agar in LB. This ~ 12 ml solution was poured onto the solidified agar under-layer. Following solidification, a final upper layer comprising of 6 mis 0.6% agar in LB was added. Plates were transferred to an incubator at 37°C for 18-22 hr to allow bacterial growth and swarming (i.e. motility) and the diameter of the bacterial colonies measured. As can be seen from Figure 3B, using this "sandwich" motility assay, LST-002 caused a significant reduction in the size of bacterial colonies even at very low concentrations for both a laboratory strain (PaOl) and a MDR strain (Ka02). In preliminary pharmacokinetic studies, we have shown that following a single i.v. injection of LST-002 at 10 mg/kg, early plasma concentrations of this mouse mAb was ~ 200 μg/ml, significantly higher than those concentrations required to inhibit P. Aeruginosa motility in vitro. One may therefore assume that inhibition of motility would be very significant in vivo and a potential major mechanism to attenuate bacterial virulence.

Table 1. Ka02 is an example of a multi-drug resistant P. Aeruginosa strain (S=antibiotic sensitive; R=antibiotic resistance) with MIC values determined at two separate testing sites using Vitek technology to

generate the antibiograms.

Interestingly, prior reports on inhibition of migration of P. aeruginosa with anti- flagellar mAbs are done with ex-vivo neutralization of P. aeruginosa prior to spotting of "neutralized" P. aeruginosa on soft agar. In our hands, it seems that the mAbs retain bioactivity within a soft, yet solidified agar.

This implies that the antibodies of the present invention are active in creams or gels intended for treatment of topical P. aeruginosa infections associated with wounds, eyes, ears, etc.

Example 3. Production of chimeric antibodies LST-003 and LST-004 derived from the VH and VL domains of their corresponding mouse monoclonal antibodies LST-001 and LST-002.

Chimeric mAb LST-003, contains the LST-001 mouse VH and VL domains (SEQ ID NO: 29 and SEQ ID NO: 30, respectively) fused to its respective human IgGl constant domains. The light chain of this chimeric antibody, consisting of the VL and CL domains, is encoded by the nucleic acid molecule having the nucleic acid sequence of SEQ ID NO: 70, and the heavy chain of this chimeric antibody, consisting of the VH and ⁽¾ domains, is encoded by the nucleic acid molecule having the nucleic acid sequence of SEQ ID NO: 69.

In the same way, chimeric mAb LST-004, contains the LST-002 mouse VH and VL domains (SEQ ID NO: 49 and SEQ ID NO: 50) fused to a human IgGl constant backbone. The light chain of this chimeric antibody, consisting of the VL and CL domains, is encoded by the nucleic acid molecule having the nucleic acid sequence of SEQ ID NO: 72, and the heavy chain of this chimeric antibody, consisting of the VH and CH domains, is encoded by the nucleic acid molecule having the nucleic acid sequence of SEQ ID NO: 71.

Similarly, an additional chimeric mAb can be generated containing the LST-002 mouse V_H domain (SEQ ID NO: 49) and the V_L minority domain (SEQ ID NO: 51) fused to a human IgGl constant backbone.

To produce LST-003, the LST-001 VL and VH domains were fused directly to their respective human IgGl constant coding sequences and the resultant VL-CL and VH- CH coding sequences were ligated into one of the 2 available multiple cloning sites (MCS)_ present in a bicistronic mammalian expression vector (pVitro-neo-mcs; In- Vivogen). To that end, Bglll and Nhel restriction motifs sites were included at the 5' and 3 '-termini respectively of the VL-CL PCR fragment, which was restricted with Bglll/Nhel, and thereafter ligated into the identically excised expression plasmid. Similarly, the VH- CH PCR fragment included a 5' BamHl linker and a 3' Avrll linker. Following excision, the purified PCR fragment was ligated into the expression plasmid previously digested with BamHl and Avrll.. In all instances, the V_L-C_L and V_H-C_H DNA sequences containing the desired restriction sites for the purpose of cloning, were chemically synthesized using overlapping synthetic primers (GENEART, Regensburg, Germany). This in-vitro "gene- construction" strategy allows optimal codon usage for expression in mammalian cells. Following PCR on the synthetic gene using external primers and sub-cloning into a suitable shuttle vector for acceptance of PCR products (eg.TOPO TA cloning kits, In- Vitrogen), plasmid DNA can be extracted from the host bacterium using conventional plasmid kits. Thereafter, the recombinant plasmids are digested with the appropriate restriction enzymes (Bglll/Nhel for the V_L-C_L fragment and BamHl /Avrll for the V_H-C_H fragment) and fragments gel purified for subsequent ligations. Using the current bicistronic plasmid, sequential cloning of the heavy and light chains were performed. To that end, the bicistronic expression plasmid was excised with Bglll/Nhel followed by ligation of the prepared V_L-CL fragment. Following identification and recovery of appropriate recombinant plasmid by PCR colony screening, this intermediary recombinant plasmid was excised with BamHl/Avrll for acceptance of the V_H-C_H fragment. Following identification and recovery of this latter recombinant plasmid, large plasmid preparations were made, followed by extensive bidirectional sequencing covering the inserted antibody fragments across the multiple cloning sites prior to expression studies.

For the chimeric mAb LST-004, the cloning strategies for the V_L and V_H were identical as described for LST-003. Furthermore, the expression plasmid was the same with the exception that the dihydrofolate reductase (DHFR) gene had been cloned into the available EcoRI site (5' GAATTC 3') located 1378 bp down stream from the Avrll site of MCS-1. The purpose was to allow plasmid amplification for eventual generation of stable CHO cell lines containing all the recombinant mAbs.

The protocol used was as follows: CHO-S cells were cultured in PROCH04 CDM medium (Lonza). On the day of transfection, the cells were centrifuged at 600 rpm for 5 minutes and resuspended in fresh PROCH05 CDM medium (Lonza) at a cell density of 2 x 10⁶ cells/ml. Transfection of 500 ml of culture was performed in a spinner flask using 2.5 μg of DNA and 10 μg Polyethylenimine (PEI) per 1 ml of culture, diluted in 150 mM NaCl. The concentration of plasmid preparation was 1.99 mg/ml (Abs₂6o/Abs₂₈o = 1-7). After 5 hr, the transfected culture was diluted with 500 ml of fresh PR05 medium and incubated at 37°C in 6% C0₂ with agitation at 60 rpm. Six days later, cell-free supernatant containing the chimeric mAb was purified on protein-A chromatography (Fig. 4A).

SDS-PAGE analysis following Coomassie gel staining showed that peak fractions collected (2-7) according to the enhanced absorbances contained two major polypeptides of approximately 25 and 55 kD, respectively (primarily in fractions #'s 4 and 5), indicating the presence of intact light and heavy chains thereby representing a bona-fide, recombinantly-expressed chimeric LST-003 antibody (Fig. 4B).

The binding specificity of the chimeric LST-003 mAb was assessed in a western blot experiment as compared to the parental mouse LST-001 mAb. Two identical, yet separate nitrocellulose membranes containing a panel of blotted P. aeruginosa lysates of type A or B flagellin following electrophoresis were hybridized with either mouse LST- 001 (Fig. 5A) or chimeric LST-003 (Fig. 5B) mAbs at identical final concentrations (10 μg ml) with the ensuing detecting secondary peroxidase-conjugated antibodies being anti- mouse Fc or anti-human Fc respectively. Laboratory strains PAK (type A flagellin) or PA01 (type B flagellin) were included as appropriate controls whereas other strains represented multi-drug resistant (MDR) P. aeruginosa strains. Fig. 5A shows the positive reactivity of mouse LST-001 to strains PA141, PA49, PA1, PA409 and PAK, indicating all 5 strains harbor flagellin type A. The lack of reactivity towards PA21 and PA01 indicates these strains harbor flagellin type B. A very similar profile, yet with stronger binding, was observed with chimeric LST-003 (Fig.5B) confirming the authenticity of this recombinantly-produced mAb in binding flagellin type A strains (PA141, PA49, PA1, PA409 and PAK) but not flagellin type B (PA21 and PA01). Figure 6 also depicts binding of LST-001, LST-002 and LST-003 to immobilized, whole P. aeruginosa as measured in an ELISA and confirms the western blot findings. See below in Example 7 for the ELISA protocol.

Example 4. Humanized VH and V_L sequences based on LST-001 and LST-002 mAb V_H and V_L sequences.

Humanized mAbs termed "LST-005" and "LST-006" that bind P. Aeruginosa flagellin type A and B respectively, are designed by predicting humanized V_H and V_L sequences based on LST-001 and LST-002 mAb VH and V_L sequences using CDR grafting methods. The mouse variable region sequences were compared to databases of human germline Ig variable region genes (e.g. http://www.ncbi.nlm.nih.gov/igblast/) and the closest human variable heavy and kappa light chain genes that are also frequently expressed in the human repertoire, were selected as the templates for humanization. The CDRs (as defined by Kabat, except for CDRH1 which also includes the preceding five amino acids) of each mouse variable chain were then transferred to the selected human germline frameworks in replacement of the equivalent human sequences. Selected mouse framework residues, that are believed to be important for CDR conformation, were also used to replace their human equivalents, resulting in the CDR grafted humanized antibody sequences. The amino acid sequences are as follows:

(1) LST-005

(la) Predicted humanized LST-005 V_H Sequence: (SEQ ID NO: 52)

(lb) Predicted humanized LST-005 V_L Sequence: (SEQ ID NO: 53)

(2) LST-006

(2a) Predicted humanized LST-006 VH Sequence:(SEQ ID NO: 54)

(2b) Predicted humanized LST-006 V_L Sequence ("Majority"): (SEQ ID NO: 55)

(2c) Predicted humanized LST-006 V_L Sequence ("Minority"): (SEQ ID NO: 56). Example 5. Production of humanized antibodies corresponding to monoclonal antibodies LST-005 (versus P. aeruginosa type A flagellin) and LST-006 (versus P. aeruginosa type B flagellin)

An artificial gene encoding for humanized LST-005 mAb is made by fusing nucleotide sequences encoding for the LST-005 humanized VH and VL domains (SEQ ID NOs: 52 and 53) to a nucleotide sequence encoding for human IgGl constant backbone. The resulting molecule is cloned downstream to an appropriate promoter and intrinsic signal peptide in an expression vector, and a CHO cell is transfected with said expression vector and produces the humanized antibody.

Humanized LST-006 mAb, containing the LST-006 humanized VH and VL (SEQ ID NOs: 54 and 55) domains fused to a human IgGl constant backbone, is produced in a similar fashion. Additionally, a further humanized mAb sub-type containing the LST-006 humanized VH and VL (SEQ ID NOs: 54 and 56) domains disclosed above in Example 2 fused to a human IgGl constant backbone, is produced in a similar fashion. Example 6. Sequencing the variable and constant fragments of the human IgM antibody LST-007 mAb and linking the variable fragments to a human IgGl backbone.

6.1. LST-007 sequences

(6a) The full V_H sequence (SEQ ID NO: 13)

(6b) The mature V_H sequence. (SEQ ID NO: 1)

(6c) The CDR sequences of the V_H sequence as defined by IMGT/V-QUEST: CDR1 (LST-007 V_H) (SEQ ID NO: 3)

CDR2 (LST-007 V_H) (SEQ ID NO: 4)

CDR3 (LST-007 V_H) (SEQ ID NO: 5)

(6d) The CDR sequences of the VH sequence as defined by Rabat:

CDR1 (LST-007 V_H) (SEQ ID NO: 6)

CDR2 (LST-007 VH) (SEQ ID NO: 7)

CDR3 (LST-007 VH) (SEQ ID NO: 8) (6e) The full V_L sequence (SEQ ID NO: 14)

(6f) The mature V_L sequence. (SEQ ID NO: 2)

(6g) The CDR sequences of the V_L sequence as defined by IMGT/V-QUEST: CDR1 (LST-001 V_L) (SEQ ID NO: 9)

CDR2 (LST-001 V_L) (the amino acid sequence AAS)

CDR3 (LST-001 V_L) (SEQ ID NO: 10)

(6h) The CDR sequences of the Vi sequence as defined by Rabat:

CDR1 (LST-001 V_L) (SEQ ID NO: 11)

CDR2 (LST-001 V_L) (SEQ ID NO: 12)

CDR3 (LST-001 V_L) (SEQ ID NO: 10)

6.2 Binding of human IgM antibody LST-007 to intact, immobilized P. aeruginosa bacterium. P. aeruginosa laboratory strains P AK (flagellin type A) and PaO 1 (flagellin type B) were grown overnight, diluted to an OD of 0.3 and washed with PBS. Fifty μΐ bacteria were then plated onto ELISA plates pre-coated with 50 μΐ poly-L-lysine (1 μg/ml in PBS) which were blocked with PBS-10% fetal calf serum (FCS). Following overnight incubation with bacteria at 4°C, plates were washed with 0.9% saline-0.05% Tween-20 and incubated with 1 :2 dilutions of LST-001, LST-002, LST-007 supernatant or purified LST-003 (~ 2 μg/ml) for 2 hrs at room temperature. Following washing, secondary goat anti-mouse (LST-001, LST-002) or goat anti-human detecting antibodies (LST-003, LST-007) were added and coloration generated following the addition of TMB substrate and plates read for absorbance (405 nm) following addition of 10% H₂S0₄.

As can be seen in Fig. 6 and Fig. 9, LST-007 specifically binds to all seven P. Aeruginosa bacteria strains tested (one in Fig. 6 and 6 in Fig. 9) harboring flagellin type B, but not flagellin type A (Fig 9, strain 27853). This confirms that following its secretion, LST-007 is present in supernatants as an intact mAb that is in a conformationally correct state.

6.3 Cloning and expression of the human antibody LST-007. Cloning of the VH and VL domains of LST-007 was performed by PCR. Briefly, total RNA was extracted from cells previously confirmed to secrete the human antibody. Thereafter, cDNA was created from the RNA by reverse transcription with an oligo(dT) primer. PCR was performed using a mixture of IgM designed, degenerate primers; the forward primer annealing to the signal/leader peptide; and the reverse primer annealing to the beginning of the first constant domain.

LST-007 is currently produced in Epstein-Barr virus (EBV) transformed human B- lymphocytes, termed lymphoblastoid cell line (LCL). LCL's suffer from a number of drawbacks, such as they grow extremely slowly and as aggregates, fail to reach cell confluency, are low level antibody producing cells, require high concentrations of fetal bovine serum (15%) to support growth, might contain residual virus (eg. EBV) and problematic to adapt to serum-free media to expedite purification. Therefore, it is highly desirable to produce true immortalized cell lines, either by fusing the transformed B-cells with myeloma cells, for example derivatives of P3X63 such as P3X63Ag8.653 (ATCC; cat # CRL-1580) or SP2/0 (ATCC; cat # CRL-1646), or with heteromyeloma cells such as F3B6 (ATCC; cat # HB-8785).

Another attractive possibility is to express the antibody in a eukaryotic cell that is easily handled and which produce large amounts of antibody. Thus, the nucleic acid sequences encoding for the heavy and light chains are cloned into a vector optionally in conjunction with nucleic acid sequences encoding for the J-chain. Thus, the human antibody may be expressed in a cell, such as CHO cells, either as IgG or IgM type, depending on the type of heavy chain and on the presence or absence of the J-chain.

The IgM type confers certain advantages such as avidity towards the flagella target where it has been purported to adopt a profound conformational change to create a "staple" position in which the IgM appears as 10-legged spiders attached to the surface of the flagella (Roax, 1999). Additionally, IgM molecules fix complement, promote opsonophagocytosis of bacteria, are resistant to proteolytic digestion by P. Aeruginosa elastase and might be administered at lower doses to produce desired therapeutic effects.

6.4. Cloning of LST-008 comprising V_H and V_L of human monoclonal antibody LST-007 linked to hlgGl. To generate a fully human recombinant LST-008 equipped with a human IgGl domain, DNA encoding the V_H domain of LST-007 (SEQ ID NO: 13) together with its respective heavy constant human IgGl domain were synthesized and subcloned into a mammalian expression plasmid so that the entire coding sequence was ligated into the available Avrll/BamHI sites (see) as detailed above in Example 3 regarding the chimeric mAb LST-004. The heavy chain of this human monoclonal antibody, consisting of the V_H and C_H domains, is encoded by the nucleic acid molecule having the nucleic acid sequence of SEQ ID NO: 63.

The entire amino acid sequence of the heavy chain sequence is set forth in SEQ ID NO: 15.

In a similar mode, V_L domain of LST-007 (SEQ ID NO: 14) together with its respective light constant domain of human IgGl were synthesized and and subcloned into a mammalian expression plasmid so that the entire coding sequence was ligated into the available Bglll/Ndel sites (see Figure 11). The entire amino acid sequence of the light chain sequence is set forth in (SEQ ID NO: 16).

The light chain of this human monoclonal antibody, consisting of the V_L and C_L domains, is encoded by the nucleic acid molecule having the nucleic acid sequence of SEQ ID NO: 64.

The bicistronic plasmid harboring the V_H and V_L of LST-007 (SEQ ID NO: 13 and SEQ ID NO: 14, respectively) and fused to its respective human IgGl constant domains, was transiently transfected into CHO cells. The protocol used was as follows: CHO-S cells were cultured in PROCH04 CDM medium (Lonza). On the day of transfection, the cells were centrifuged at 600 rpm for 5 minutes and resuspended in fresh PROCH05 CDM medium (Lonza) at a cell density of 2 x 10⁶ cells/ml. Transfection of 500 ml of culture was performed in a spinner flask using 2.5 g of DNA and 10 μg Polyethylenimine (PEI) per 1 ml of culture, diluted in 150 mM NaCl. The concentration of plasmid preparation was 1.99 mg/ml (Abs₂6o Abs28o = 1 -7). After 5 hr, the transfected culture was diluted with 500 ml of fresh PR05 medium and incubated at 37°C in 6% C0₂ with agitation at 60 rpm. Six days later, cell-free supernatant containing the fully human mAb was purified on protein-A chromatography (Fig. 7) with the purified peak on Coomassie gel staining depicting the V_H and VL fragments at approximately 55 and 25 kD respectively (Fig. 7, inset, lane 2) according to the protein markers in Fig. 7, inset, lane 1.

To confirm authenticity of recombinantly-produced LST-008, its reactivity to P.Aeruginosa flagellin type B by ELISA was determined (Fig. 8). Maxisorp ELISA plates were coated with recombinant flagellin type B at 5 μg/ml (250 ng/well), blocked with PBS-10% FCS and incubated with LST-008 over a concentration range of 0.43 ng/ml-2 μg/ml. Thereafter, wells were washed and replenished with a goat-anti-human Fc-HRP conjugated secondary antibody at 1 :10,000 dilution. Reaction was catalyzed by the addition of TMB substrate and plates read for absorbance (405 nm) following addition of 10% H2SO4. A sigmoidal concentration-response curve was obtained with EC50 ~ 10 ng/ml. Additionally, in a bacterial binding assay employing a panel of P.Aeruginosa strains fixed by formaldehyde to ELISA plates (Fig. 9), LST-008 (0.5 μg/ml final concentration) bound all strains known to harbor flagellin type B (KaOl, Ka02, Ka03, Ka04, 25619, NCTC 10662) but not flagellin type A (27853). LST-008's binding profile was identical to that observed with its parental hybridoma LST-007, from which its VH and VL regions were derived (Fig. 9).

Example 7. Production of dual-specific antibodies specifically binding flagellin, type A and type B of P. aeruginosa.

7.1 Strategy and cloning of dual-specific antibodies.

A most attractive design is the so called "knobs-into-holes" concept in which a knob is created by replacing a small amino side chain at the interface between CH3 domains of the human IgGl Fc region (hinge, constant region 2 and 3 of immunoglobulin heavy chain (CH2 and CH3), GenBank accession no. AF 150959) with a larger one, whereas a hole is constructed by replacing a large side chain with a smaller one. In particular, a knob variant is obtained by replacement of a small amino acid (threonine at position 366) with a large amino acid (tyrosine) in the CH3 domain of the human IgGl Fc region, creating the mutant T366Y. A hole variant is constructed by replacement of a large residue (tyrosine at position 407) with a small one (threonine) in the CH3 domain (Ridgway et al., 1996), creating the mutant Y407T. Anti- P. aeruginosa flagellin type A and anti- P. aeruginosa flagellin type B scFvs are fused to the knob and hole variants, respectively (Fig. 1). The two chains are then produced in CHO cells and heterodimer antibodies with dual specificity are produced.

The dual-specific antibody LST-009 was produced according to the above strategy, as follows: the LST-001 anti-flagella type A antibody VH and VL regions were fused as a scFv to the constant region of IgGl (Hinge-CH2-CH3). Optimized codon usage for expression in mammalian cells was designed in which the VH of LST-001 was the most upstream sequence translated preceded by its own signal peptide. At VH'S C'-terminus, a coding sequence conforming to an amino acid spacer peptide, GGGGSGGGGSGGGGS (SEQ ID NO: 57 repeated three times), was placed followed immediately by the mature coding sequence of VL of LST-001. At the end of VL, a linker peptide DNA sequence conforming to amino acids PGSAGGSGG (SEQ ID NO: 59) was placed followed by a human IgGl sequence constituting the hinge, CH2 and CH3 domains. At amino acid # 366, the wild-type small threonine (T) residue was replaced by a large tyrosine (Y) residue. This substitution created a "knob" variant within the CH3 domain. The entire amino acid sequence is set forth in SEQ ID NO: 60.

The nucleic acid molecule encoding for the sequence of SEQ ID NO: 60, which has the nucleic acid sequence as set forth in SEQ ID NO: 65, was subcloned into pVitro-neo- MCS (equipped with the DHFR gene) via BamHl-Avrll restriction sites at the MCS-1 as detailed above in Example 3 regarding the chimeric mAb LST-004.

Similarly, LST-002 VL and VH were cloned as an scFv with human IgGl Fc (Hinge-CH2-CH3). The format is V_L majority- linker- V_H-linker-CH2-CH3. Optimized codon usage for expression in mammalian cells was designed in which the VL majority of LST-002 was the most upstream sequence translated preceded by its own signal peptide. At VL'S C'-terminus, a coding sequence conforming to an amino acid spacer linker, GGGSAAA (SEQ ID NO: 58), was placed followed immediately by the mature coding sequence of VH of LST-002. At the end of VL, a linker peptide DNA sequence conforming to amino acids PGSAGGSGG (SEQ ID NO: 59) was placed followed by a human IgGl sequence constituting the hinge, CH2 and CH3 domains. At amino acid # 407, the wild- type and large tyrosine (Y) residue was replaced by a small threonine (T) residue. This substitution created a "hole" variant within the CH3 domain. The entire amino acid sequence is set forth in SEQ ID NO: 61.

The nucleic acid molecule encoding for the sequence of SEQ ID NO: 61, which has the nucleic acid sequence as set forth in SEQ ID NO: 66, was subcloned into pVitro-neo- MCS (equipped with the DHFR gene) via Bglll-Nhel restriction sites at the MCS2.

Additionally, a proline has been included in the linker peptide since this is known to disrupt alpha helices and augment folding/heterodimerization. It should also be noted that the signal peptide of the VH domain is used in the anti-flagella type A scFv, whereas the signal peptide of the VL domain is used in the anti-flagella type B scFv. This is important, because the presence of a VL signal peptide is crucial for antibody secretion.

The constructs were cloned into the multi cloning sites of a pVITRO-neo-mcs vector (equipped with the DHFR gene). This vector represents an example of a commercially available multigenic plasmid which can be selectable in E. Coli via kanamycin or selectable in mammalian cells with G418. By incorporation of the dihydrofolate reductase (DHFR) gene into the plasmid and transfection into mammalian cells (eg. CHO), addition of methotrexate will cause gene amplification of DHFR with its associated transfected DNA. This results in multiple copies of the recombinant plasmid in the transfected, stable cell and higher levels of the recombinant protein.

7.2 Transient Transfection of CHO Cells with DNA encoding for dual-specific antibody LST-009. Suspension culture adapted CHO cells were cultured in PROCH04 CDM medium (Lonza). On the day of the transfection, the cells were centrifuged at 200 g for 5 min and resuspended in fresh PROCH05 CDM or Ultra CHO medium (Lonza) at a cell density of 2 x 10⁶ cells/ml. CHO cell cultures were then transfected with 2.5 μg of LST-009 DNA and 10-15 μ polyethylenimine (PEI) per 1 ml of culture, diluted in 150 mM NaCl using a plasmid preparation with a Abs260/Abs280 >1.7) in spinner flasks agitated at 60 rpm 37°C in 8% C02. After 5 hr, the transfected culture was diluted with an equal culture volume of fresh PROCH05 or Ultra CHO medium and incubated at 37°C in 8% C02 with agitation at 60 rpm. The supernatant was collected after 9-14 days post- transfection.

7.3. Purification of LST-009 and analysis of purified mAb. Supernatant was purified using an Amersham Biosciences AKTA Chromatography system. Purification method was performed using 1 ml Protein A columns (GE Healthcare) using a standard antibody purification protocol. A typical purification chromatogram trace of heterodimeric LST-009 mAb is shown in Figure 10.

After successful elution of bound protein from the column, fractions were collected that corresponded to elution peak. Bradford analysis was performed to determine which fractions contained suitable levels of protein for dialysis. Fractions containing protein were dialysed overnight at 4°C in PBS pH7.2 at a ratio of 1 : 100. Purified fractions were collected and analysed by SDS-PAGE and Coomassie gel staining to determine purity (see Fig. 16, inset).

Yields of purified LST-009 were ~ 4 mg/1. On Coomassie gel staining in the presence of reducing agent (50 mM DTT final concentration), LST-009 was stained as ~ 50-55 kD band (Fig. 10, lane B). This conforms to the anticipated Mw derived from a scFv fragment and the presence of CH2-CH3 of the human IgGl heavy chain. The apparent single band consists of the scFv for P. aeruginosa flagellin type A (with associated heavy chains) and the scFv for P. aeruginosa flagellin type B (with associated heavy chains). In the absence of reducing agent (Fig. 10, inset lane A), a doublet was observed between ~ 90-100 kD.

7.4. Binding of dual specific LST-009 to ELISA plates coated with recombinantly-expressed P. aeruginosa flagellin types A or B. Two hundred and fifty ng (250 ng) of P. aeruginosa flagellin type A or B (50 μΐ from 5 μg/ml) were dispensed into each well of ELISA plates (Nunc, cat # 442404) and coating allowed to proceed for 2 hours at room temperature with gentle shaking. Plates were then blocked with 200 μΐ PBS- 10% fetal bovine serum (FBS) overnight at 4°C.

LST-009 diluted in PBS-10% FBS was added to wells over a concentration range of 0.43 μg/ml - 2 μg/ml and incubation allowed to proceed for 2 hrs at room temperature. As suitable control mAbs for LST-009 and in order to guage binding potency, LST-003 and LST-004 were included in the ELISA since these mAbs harbored the same VH and V_L of LST-009 with all 3 mAbs being detected with the same anti-human Fc secondary antibody. To that end, following three washes with PBS-0.05% Tween-20, 50 μΐ of a goat anti-human IgG (Fc)-HRP (Cat # A80-104P; Bethyl, TX, USA) was added at a dilution of 1 : 10,000 in PBS-10% FBS with incubation allowed to proceed for 60 min at room temperature. Following 3 separate washes with PBS-Tween-20 (0.05%) and gentle tapping of plates on paper towels to remove final traces of wash buffer, 50 μΐ TMB/E substrate solution (Cat # ES001-500 ml; Millipore) was added for 15 min at room temperature. Thereafter, signal was quenched with 50 μΐ 10% H₂S0₄ and plates read at 450 nm in a Microtiter plate reader (Anthos MultiRead 400).

On plates coated with PA flagellin type A, LST-009 bound the immobilized antigen, albeit with a right hand shift in potency as compared to its control mAb LST-003 (Fig. 11). To that end, LST-009's EC50 towards P. aeruginosa flagellin type A was approximately 600 ng/ml as compared to LST-003 being approximately 65 ng/ml (Fig. 17). In contrast, LST-009 demonstrated high potency binding towards P. aeruginosa flagellin type B by ELISA (Fig. 12). LST-009's EC₅₀ towards this immobilized antigen was approximately 250 ng/ml as compared to the control LST-004 mAb being approximately 300 ng/ml.

7.5. Binding of LST-009 to formaldehyde-treated, whole P. aeruginosa bacteria by ELISA. From a single fresh colony, P. aeruginosa bacteria were grown overnight in 5 ml LB at 37°C. Thereafter, bacterial cultures were centrifuged at 1500 g for 30 min and following removal of the clear supernatant, pellets were washed twice with 20 ml PBS with intervening centrifugation. The final pellet was resuspended in a volume of PBS (~ 10 ml) and adjusted to generate an OD₆oo nm of - 0.2 and maintained on ice prior to coating on ELISA plates which were prepared as follows: to the wells of flat-bottomed ELISA plates (Nunc, cat # 442404), 50 μΐ poly-L-lysine (PLL, Sigma Cat # P-1524) of a 1 μg/ml solution was added and plates were incubated for 30 min at room temperature. Thereafter, non-adsorbed PLL was removed.

Fifty μΐ of the bacterial suspension at 0.2 OD600 nm was added to the PLL-coated ELISA plates which were then centrifuged at 1500 rpm for 20 min to expedite bacterial adsorption. Thereafter, supernatant was removed and 75 μΐ of 0.2% formaldehyde added to the adsorbed bacteria for 15 min at room temperature to cause irreversible fixation of bacteria to the plates. Following removal of formaldehyde, plates were air-dried for 5 min at room temperature and taken for ELISA to characterize LST-009 binding as follows:

ELISA plates containing the fixed P. aeruginosa bacteria, were blocked by adding 200 μΐ PBS-10% FBS for 60 min at room temperature. After 60 min, block solution was flicked out, plates washed once with PBS and 50 μΐ LST-009 in PBS-10% FBS added over a range of concentrations and incubated for 60 min at 37°C. Following removal of the primary antibody and 3 washes with PBS-Tween (0.05%), 50 μΐ of a goat anti-human IgG (Fc)-HRP (Cat # A80-104P; Bethyl, TX, USA) was added at a dilution of 1 : 10,000 in PBS- 10% FBS with incubation allowed to proceed for 60 min at 37°C. Following 3 separate washes with PBS-Tween (0.05%) and gentle tapping of plates on paper towels to remove final traces of wash buffer, 50 μΐ TMB/E substrate solution (Cat # ES001-500 ml; Millipore) was added with incubation allowed to proceed for 15 min at room temperature. Thereafter, signal was quenched with 50 μΐ 10%H₂SO₄ and plates read at 450 ran in a Microtiter plate reader (Anthos MultiRead 400).

Over a mAb concentration of 0.125-11.25 μg/ml, LST-009 bound PAK in a concentration-dependent manner (Fig. 13) with an EC50 of ~ 4 μg/ml. In contrast, LST- 009 binding towards PaO 1 was much higher with an EC50 of approximately 80 ng/ml (Fig. 14). These data confirm LST-009's binding profile to recombinantly-expressed PA flagellin types A and B.

These findings have major implications for the design of effective bispecific mAbs targeting P. Aeruginosa flagellin types A and B. It would appear that the order of translation of the VH and VL fragments are critical in allowing efficient heterodimer formation and ultimate intact binding towards target antigen in which VL should be cloned upstream from VH for both scFv's. Example 8. Production of alternatively designed dual-specific antibodies specifically binding flagellin, type A and type B of . aeruginosa

In this example, dual-specific antibodies comprising a chimeric antibody that binds specifically to P. aeruginosa flagellin type A, i.e. LST-003 as described in Example 3, consisting of the original LST-001 mouse V_H and V_L domains (SEQ ID NO: 29 and SEQ ID NO: 30, respectively) fused to a nucleotide sequence encoding for human IgGl constant backbone, the two heavy chains of which are connected via linker peptides consisting of (G₄S)₂ (two consecutive peptides of SEQ ID NO: 57) to the N-termini of two scFvs which binds specifically to P. aeruginosa flagellin type B were designed. The scFvs are identical to each other and each consists of a V_H and a V_L domain of LST-002 (SEQ ID NO: 31 and 50, respectively) linked together via a 7 amino acid spacer peptide of the amino acid sequence GGGSAAA (SEQ ID NO: 58). Practically, the heavy chain of LST-003 was fused to the scFvs having binding specificity for P. aeruginosa flagellin type B, and this fusion polypeptide, having variable regions at each extreme of the human heavy constant backbone is of SEQ ID NO: 62. and is encoded by the nucleic acid molecule having the nucleic acid sequence as set forth in SEQ ID NO: 67

The light chain consists simply of the VL and CL domain of LST-001 and is encoded by the nucleic acid molecule having the nucleic acid sequence as set forth in SEQ ID NO: 68.

8.1 Expression and purification of LST-010. A recombinant, bicistronic mammalian expression plasmid harboring the complete chimeric LST-003 mAb sequence (i.e. its VH, VL domains and respective human constant chains) as well as the VH-VL domains of LST-004, was synthesized and optimized for expression in CHO cells (GeneArt, Germany) . The resultant dual-specific chimeric mAb, coined LST-010, is structurally represented in Figure 2.

The recombinant LST-010 mammalian plasmid was prepared at a concentration of

1 mg/ml through standard molecular biology methodologies and taken for transfection into mammalian cells. In brief, CHO-S cells were cultured in ProCH05 CDM media (Lonza, UK), centrifuged at 200 g for 5 min and resuspended in fresh ProCH05 CDM media at a density of 2 x 10⁶ cells/ml. Transfection of 700 ml CHO-S cells was performed in an Erlenmeyer flask using 2.5 μg DNA and 15 μg Polyethylenimine (PEI) per ml culture, diluted in 150 mM NaCl. After 5 hrs, the transfected culture was diluted with 700 ml fresh media and incubated at 37°C in 8% C0₂ with gentle agitation at 92 rpm. The supernatant was collected after 7 days post-transfection at which point it was centrifuged, clarified by filtration through 0.8 μτη gyrodisc filters and taken for purification.

LST-010 was purified from two separate 700 ml batches of clarified CHO-S supernatants using an Amersham Bioscience AKTA Chromatography system. Purification method was a standard antibody purification protocol using a 1 ml Protein-A column (GE Healthcare). Following elution, peak fractions from both 700 ml purifications were taken for Bradford analysis of protein concentration and those specific fractions containing the highest concentration of LST-010 were pooled and dialyzed overnight against PBS at 4°C. Concentration of the final purified LST-010 was 130 μg/ml which was then taken for binding assays using the following two strategies: a) Binding to ELISA plates coated with recombinant PA flagellin types A or B; b) Binding to ELISA plates containing, whole, immobilized PA bacteria of flagellin types A or B.

8.1 Binding of dual-specific LST-010 to ELISA plates coated with recombinantly- expressed P. aeruginosa flagellin types A or B. Two hundred and fifty ng (250 ng) of highly purified, recombinantly-expressed P. aeruginosa flagellin types A or B (50 μΐ from 5 μg/ml) were dispensed into each well of ELISA plates (Nunc, cat # 442404) and coating allowed to proceed for 2 hours at room temperature with gentle shaking. Plates were then blocked with 200 μΐ PBS-10% fetal bovine serum (FBS) overnight at 4°C. LST-010 diluted in PBS-10% FBS was added to the antigen-coated wells. Similarly, in separate wells, LST-003 and LST-004 were included as positive controls for PA flagellin type A and B ELISA plates respectively. Since the Mw of the mAbs differ, (LST-003/004 = 150 kD, LST-010 = 200 kD), an identical mAb concentration range 0.0067 - 6.9 nM (as opposed to ng/ml) was employed in the ELISA. Following binding with these different primary mAbs for 2 hrs at room temperature, plates were washed with PBS-0.05% Tween- 20. Thereafter, 50 μΐ of a goat anti-human IgG (Fc)-HRP (Cat # A80-104P; Bethyl, TX, USA) was added at a dilution of 1: 10,000 in PBS-10% FBS and incubation allowed to proceed for 60 min at room temperature. Following 3 separate washes with PBS-Tween-20 (0.05%), 50 μΐ TMB/E substrate solution (Cat # ES001-500 ml; Millipore) was added for 15 min at room temperature. Colorimetric development was then quenched with 50 μΐ 10% H₂S0₄ and plates read at 450 nm in a Micro titer plate reader (Anthos MultiRead 400).

As can be seen from Figure 15A, binding of LST-010 towards PA flagellin type A paralleled that of chimeric LST-003. Specific binding of LST-010 towards PA flagellin type B was also observed (Fig. 15B), although its signal was approximately 20-50% as compared to the positive control of LST-004.

8.2. Binding of dual-specific LST-010 to immobilized, formaldehyde-treated, whole P. aeruginosa bacteria by ELISA. From a single fresh bacterial colony, PA strains 27853 (harboring flagellin type A) or Ka02 (harboring flagellin type B) were grown overnight in 5 ml LB at 37°C. Thereafter, bacterial cultures were centrifuged at 1500 g for 30 min and following removal of the clear supernatant, pellets were washed twice with 20 ml PBS with intervening centrifugation. The final pellet was resuspended in a volume of PBS (~ 10 ml) and adjusted to generate an OD₆oo of ~ 0.2 and maintained on ice prior to coating on ELISA plates which were prepared as follows: to the wells of flat-bottomed ELISA plates (Nunc, cat # 442404), 50 μΐ poly-L-lysine (PLL, Sigma Cat # P-1524) of a 1 μg/ml solution was added and plates were incubated for 30 min at room temperature. Thereafter, non-adsorbed PLL was removed. Fifty μΐ of the appropriate bacterial suspensions from the different strains at 0.2 OD600 nm were added to the PLL-coated ELISA plates. In some wells, a "mixed-bacterial population" was added. To that end, 25 μΐ of PA 27853 and 25 μΐ of Ka02 bacterial suspensions were added to the same wells. Plates were then centrifiiged at 1500 rpm for 20 min to expedite bacterial adsorption, supernatant removed and 75 μΐ of 0.2% formaldehyde added to the adsorbed bacteria for 15 min at room temperature to cause irreversible fixation of bacteria to the plates. Following removal of formaldehyde, plates were air-dried for 5 min at room temperature and taken for ELISA to characterize LST-010 binding as well as the appropriate controls of LST-003 and LST- 004 as follows:

ELISA plates containing the fixed P. aeruginosa bacteria, were blocked by adding 200 μΐ PBS-10% FBS for 60 min at room temperature. After 60 min, block solution was flicked out, plates washed once with PBS and 50 μΐ mAbs in PBS-10% FBS added at a final concentration of 1 nM and incubated for 60 min at 37°C. Following removal of the primary antibodies and 3 washes with PBS-Tween (0.05%), 50 μΐ of a goat anti-human IgG (Fc)-HRP (Cat # A80-104P; Bethyl, TX, USA) was added at a dilution of 1 : 10,000 in PBS-10% FBS with incubation allowed to proceed for 60 min at 37°C. Following 3 separate washes with PBS-Tween (0.05%) and gentle tapping of plates on paper towels to remove final traces of wash buffer, 50 μΐ TMB/E substrate solution (Cat # ES001 -500 ml; Millipore) was added with incubation allowed to proceed for 15 min at room temperature. Thereafter, signal was quenched with 50 μΐ 10%H₂SC>4 and plates read at 450 nm in a Microtiter plate reader (Anthos MultiRead 400).

In the bacterial binding assay, we used LST-010 at a final concentration of 1 nM since this represents the approximate affinity constants for its 2 components, LST-003 and LST-004. As can be seen from Figure 16, LST-010 bound PA27853 at a similar signal observed with its control mAb LST-003. Also, binding of LST-010 towards Ka02 was positive although its signal was approximately 40%) of that observed with LST-004. As anticipated, LST-003 failed to bind Ka02 with minimal signal of LST-004 towards PA27853. Furthermore, LST-010 was capable of binding immobilized, "mixed" bacteria in wells containing PA27853 and Ka02 (Figure 17). In this experiment, individual wells were coated with half the amount of bacteria for both PA27853 and Ka02 strains as used in the study depicted in Figure 16. Thus on binding, OD's generated with 1 nM LST-003 and LST-004 were approximately 50% as compared to Figure 16. Importantly, LST-010 (1 nM) was able to bind both bacterial populations within the same well since the OD generated constituted the sum of binding of LST-003 and LST-004 (Fig. 17). These critical data would indicate that the dual specific mAb LST-010 is capable of binding mixed PA populations (flagellin types A and B) at its presumed KD, thereby representing a therapeutic target concentration.

A corresponding dual-specific antibody is produced in which the chimeric or humanized antibody module is specific for P. aeruginosa flagellin type B, and the scFvs are specific for P. aeruginosa flagellin type A. Example 9. Characterization of the mAbs binding P. aeruginosa flagellin types A and B by Surface Plasmon Resonance (SPR).

The binding affinities of 6 different mAbs (Table 2) were determined using the ProteOn™ XPR36 protein interaction array system utilizing SPR instrumentation. mAbs were bound onto a ProteOn GLM sensor chip using standard amine coupling chemistry. In general, 5,000-10,000 relative units (RU) of mAbs were immobilized. Thereafter, 5 different concentrations of analyte (PA flagellin type A or B) from 5-80 nM were injected in a volume of 100 μΐ at a flow rate of 50 ml/min. Six concentration-dependent sensorgrams were obtained for each mAb to generate RU versus run time (min) plots. Full kinetic rat constant determinations were generated using a simple 1 : 1 interaction model and the kinetic constants and subsequent KD values obtained.

Both LST-003 and LST-004 chimeric mAbs retained similar affinities towards PA flagellin type A and B respectively which were highly comparable to their parental mouse mAbs LST-001 and LST-002 (Table 2). Interestingly, the affinity of the type B binding component of the dual-specific mAb LST-009 was 0.8 nM, being essentially identical to its parental LST-002/LST-004 mAbs which harbor the identical VH and VL sequences. The newly created fully human mAb LST-008 which contained a human IgGl constant domain exhibited a high affinity of 0.94 nM towards PA flagellin type B. This finding is of interest since its parental mAb (LST-007) which is of the IgM type, are generally considered as lower affinity mAbs.

Table 2. Affinity constants (nM) of mAbs binding P. aeruginosa flagellin types A and B.

Example 10. In vivo experiments.

10.1 LST-002 is efficacious in the treatment of multidrug-resistant P. aeruginosa model of pneumonia

Experiment # 1. Sixty-one (61) male Adult C57 Black mice (25-30 g, Harlan Nossan, Milan, Italy) were housed in a controlled environment in cages with filter tops in specific pathogen-free conditions. They were briefly anesthetized with inhaled sevorane (Abbot Laboratories) in an oxygenated chamber and placed in a supine position with their heads elevated approximately 30°. Ka02 bacterial inoculums (10⁶ cfu's in 50 μΐ of lactated Ringer's solution) were instilled slowly into the left lung of each animal using a gavage needle.

Mice were randomly divided into 3 experimental treatment groups as follows:

a) Saline-Treated (n=20): saline administered i.v. 1 hr after infection followed by one further i.p. administration at 5 hr post-infection; b) LST-002-Treated (n=21): LST- 002 given i.v. 1 hr after infection at a dose of 20 mg/kg followed by one further i.p. administration at 5 hr post-infection at a dose of 15 mg/kg; c) Positive control - Imipen em-Treated (n=20): "Tienam" (Merck, Sharpe and Dohme)] (active ingredient, imipenem, a carbapenem antibiotic) given i.p. 1 hr after infection at a dose of 25 mg/kg followed by further i.p. administrations at 25 mg/kg at time points of 5 hr, 24 hr, 29 hr, 48 hr and 53 hr following infection.

Survival was monitored for all groups every 12 hrs until day 9 post-infection and data depicted in Kaplan-Meier curves (Fig. 18). As can be seen from Figure 18, LST-002 significantly improved survival over a 3 day period following infection which was superior to saline-treated and imipenem-treated animals. Animal survival at day 3 was 71.4% for LST-002 (15/21 survivors), 30% for imipenem (6/20 survivors) and 10% for saline (2/20 survivors). Survival at day 9 was unchanged from survival data observed at day 3, i.e. LST-002 was 240% more efficacious in increasing survival in mice with pneumonia than a currently accepted treatment protocol and increased survival by 700% as compared with untreated mice.

Experiment # 2: LST-002 is efficacious in the treatment of multidrug-resistant P. aeruginosa model of thigh muscle infection. The effect of LST-002 on thigh muscle abscess infection triggered by the intra-muscular administration of MDR PA strain Ka02 was examined. Female CD-I mice weighing 23-27 g (3-5 weeks in age) were rendered neutropenic by split i.p. administration of cyclophosphamide at doses of 150 mg/kg and 100 mg/kg on days 0 and +3. On day 4, 50 μΐ of freshly-grown Ka02 (2 x 10⁷ cfu's per ml) was injected at a depth of 5 mm into the right thigh muscle ("ipsilateral") whereas an equivalent volume of saline was injected into the left thigh muscle ("contralateral"). Over an ensuing 3 day period (days 5-7, i.e. 24, 48 and 72 hr following Ka02 administration), the surface areas of the infected thigh muscle lesions were measured with calipers in both ipsilateral and contralateral muscles. Ratios of surface areas of infection lesions of ipsilateral muscles divided by contralateral muscle lesions were calculated with ratios > 1 indicative of Ka02 . This calculation is called the Ratio-Infection Index (RII). In some mice, LST-002 was administered i.v. at 20 mg/kg, 60 min after administration of Ka02. Further dosing of LST-002 (10 mg/kg) was given i.p. at time points + 5 hr, + 24 hr and + 48 hr after infection.

As can be seen from Figure 19, RII in mice challenged with Ka02/saline, increased from day 5 (1.78) to day 7 (2.68). This increase was prevented in mice treated with LST- 002 since RII's on day 5 and 7 were 1.09 and 0.96 respectively.

Chimeric, humanized and dual-specific antibodies specific for P. Aeruginosa flagellin type A, B, or both, are further employed in Experiments 1 and 2 above.

10.2 P. aeruginosa infection in burn model. In another animal model, normal mice are exposed to a localized burn on their back followed by an immediate subcutaneous injection of P. aeruginosa bacteria. In this lethal model, the antibodies are given intravenously, intraperitoneally, subcutaneously or by inhalation in various experimental paradigms as described below to quantitative their biological effect in preventing lethality and/or attenuating bacterial invasion to organs.

10.3 The following experiments are performed using antibodies specific for P. aeruginosa flagellin type A or B:

· Chimeric, humanized, recombinant fully human or dual-specific antibodies given prophylactically and/or post-infection versus P Aeruginosa clinical isolates bearing type A or B flagellin.

• Chimeric, humanized, recombinant fully human or dual-specific antibodies given in combination with an anti-biotic prophylactically and/or post-infection versus P Aeruginosa clinical isolates bearing type A or B flagellin.

• Chimeric, humanized, recombinant fully human or dual-specific antibodies given prophylactically and/or post-infection versus anti-biotic resistant P Aeruginosa clinical isolates bearing type A or B flagellin.

• Combination of chimeric anti-flagellin type A and anti- flagellin type B

antibodies, humanized anti-flagellin type A and anti- flagellin type B antibodies, recombinant human anti-flagellin type A and anti- flagellin type B antibodies or dual-specific antibodies, given prophylactically and/or postinfection versus P Aeruginosa clinical isolates bearing type A and type B flagellin or versus anti-biotic resistant P Aeruginosa clinical isolates bearing type A and type B flagellin in combination with - or without - an antibiotic given prophylactically and/or post-infection.

REFERENCES

Almagro JC, Fransson J. Humanization of antibodies. Frontiers in Bioscience 13, 1619-1633, January 1, 2008

Ansorg, Flagella Specific H Antigenic Schema of Pseudomonas aeruginosa . Zbl. Bakt. Hyg., I. Abt. Orig. A, vol. 242 (1978), pp. 228-238.

Brochet X, Lefranc MP, Giudicelli V. IMGT/V-QUEST: the highly customized and integrated system for IG and TR standardized V-J and V-D-J sequence analysis. Nucleic Acids Res. 2008 Jul 1;36: W503-8.

Chothia, C. & Lesk, A.M. "Canonical Structures for the Hypervariable Regions of Immunoglobulins," J. Mol. Biol. (1987), vol. 197, pp. 901-917.

Chothia, C, Lesk, A.M., Gherardi, E., Tomlinson, I.M., Walter, G., Marks, J.D., Llewelyn, M. B. & Winter, G. "Structural Repertoire of the Human VH Segments," J. Mol. Biol. (1992), vol. 227, pp. 799-817.

Drake D, and Montie TC. Protection against Pseudomonas aeruginosa infection by passive transfer of anti-flagellar serum. Can. J. Microbiol. 1987; 33:755-63.

Edward-Jones V, Greenwood JE. On behalf of the Manchester Burns Research Group 2003. What's new in burn microbiology? James Laing Memorial Prize Essay. Burns. 2000; 29: 15-24.

Singh NP, Goyal R, Manchanda V, et al. Changing trends in bacteriology of burns in the burns unit, Delhi, India. Burns. 2003; 29: 129-132.

Eaves-Pyles T, Szabo C, Salzman AL. Bacterial invasion is not required for activation of NF-kappaB in enterocytes. Infection and Immunity 1999; 67: 800-04.

Kufer P, Ralf Lutterbuse and Patrick A. Baeuerle. A revival of bispecific antibodies. TRENDS in Biotechnology Vol.22 No.5 May 2004.

Landsperger WJ, Kelly-Wintenberg KD, Montie TC, Knight LS, Hansen MB,

Huntenburg CC, Schneidkraut MJ. Inhibition of bacterial motility with human antiflagellar monoclonal antibodies attenuates Pseudomonas aeruginosa-induced Pneumonia in the immunocompetent rat. Infection and Immunity 1994; 62: 4825-30.

Lanyi, B. Serological properties of Pseudomonas aeruginosa. II. Type-specific thermolabile (flagellar) antigens. Acta Microbiol. Acad. Sci. Hung., (1970) 17:35-48.

Matsumoto T, Tateda K, Miyazaki S, Furuya N, Ohno A, Ishii Y, Hirakata Y, and Yamaguchi K. Effect of antiflagellar human monoclonal antibody on gut-derived Pseudomonas aeruginosa sepsis in mice. Clin. Diagn. Lab. Immunol. 1999; 6: 537-41 Oishi K, Sonoda F, Iwagaki A et al. Therapeutic effects of a human antiflagella monoclonal antibody in a neutropenic murine model of Pseudomonas aeruginosa pneumonia. Antimicrobial agents and Chemotherapy 1993; 37: 164-70.

Rosok MJ, Stebbins MR, Connelly K, Lostrom ME, Siadak, AW. Generation and characterization of murine antiflagellum monoclonal antibodies that are protective against lethal challenge with Pseudomonas aeruginosa. Infect. Immun. 1990; 58:3819-28.

Sanger F, Nicklen S, and Coulson AR. DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci U S A. 1977 Dec;74(12):5463-7

Steiner TS, Lima AA, Nataro JP and Guerrant RL. Enteroaggregative Escherichia coli produce intestinal inflammation and growth impairment and cause IL-8 release from intestinal epithelial cells. J Inf Dis 1998; 177: 88-96.

Wu TT and Kabat EA. An analysis of the sequences of the variable regions of Bence Jones proteins and myeloma light chains and their implications for antibody complementarity. J Exp Med. 1970 Aug 1;132(2):211-50.

Yamashita, M., Katakura Y., Shirahata S. Recent advances in the generation of human monoclonal. Antibody. Cytotechnology (2007) 55:55-60.

Roux KH. Immunoglobulin structure and function as revealed by electron microscopy. Int Arch Allergy Immunol (1999) 120: 85-89.

Claims

1. An antibody that specifically binds to flagella type A or type B of P. aeruginosa, or specifically binds to flagellin type A and type B of P. aeruginosa, but excluding a mouse monoclonal antibody.

2. The antibody according to claim 1, which is a human antibody comprising a human IgG constant region fused to human variable regions of an anti-P. aeruginosa monoclonal antibody that specifically binds to flagella type B.

3. The antibody according to claim 1, which is a dual-specific antibody selected from:

(i) a dual-specific antibody comprising (a) a scFv that specifically binds to flagella type A, fused to the Fc-region of a first human IgG heavy chain molecule; and (b) a scFv that specifically binds to flagella type B, fused to the Fc-region of a second human IgG heavy chain molecule, wherein a small amino acid residue of the Fc-region of the first human IgG heavy chain molecule is replaced with a large amino acid residue, and a large amino acid residue of the Fc-region of the second human IgG heavy chain molecule is replaced with a small amino acid residue, and/or a large amino acid residue of the Fc-region of the first human IgG heavy chain molecule is replaced with a small amino acid residue, and a small amino acid residue of the Fc-region of the second human IgG heavy chain molecule is replaced with a large amino acid residue, whereby the large amino acid residue forms a protuberance that is positionable in the cavity formed by the small amino acid residue and thereby the first and the second human IgG heavy chain molecules form a heterodimer that specifically binds both flagella type A and flagella type B;

(ii) a dual-specific antibody comprising an antibody composed of a human IgG constant region fused to variable regions of an anti- .

P. aeruginosa. (iii) a dual-specific antibody comprising an antibody composed of a human IgG constant region fused to variable regions of an anti- .

(iv) a dual-specific antibody comprising two scFvs that specifically bind to flagella type A of P. aeruginosa fused via a linker peptide to two other scFvs that specifically bind to flagella type B off. aeruginosa.

4. The antibody according to claim 1, which is a chimeric antibody comprising a human IgG or IgM constant region fused to mouse variable regions of an anti- . aeruginosa monoclonal antibody that specifically binds to flagella type A or type B of P. aeruginosa.

5. The antibody according to claim 1, which is a humanized antibody comprising a human IgG or IgM constant region fused to humanized mouse variable regions of an anti- P. aeruginosa monoclonal antibody that specifically binds to flagella type A or type B of P. aeruginosa.

6. The antibody according to any one of claims 2 to 5, wherein said IgG constant region is a human IgGl constant region.

7. The antibody according to claim 2 or 3, wherein said variable regions or scFv that specifically bind to flagella type B comprises (a) a human VH fragment comprising CDRs present in SEQ ID NO: 1 ; and (b) a human VL fragment comprising CDRs present in SEQ ID NO: 2.

8. The antibody according to claim 7, wherein said CDRs are as defined by EVIGT/V- QUEST or Kabat.

9. The antibody according to claim 7, wherein said human VH fragment comprises the amino acid sequence of SEQ ID NO: 13; and the VL fragment comprising the amino acid sequence of SEQ ID NO: 14.

10. The antibody according to claim 9, comprising the V_H fragment fused to a human IgGl heavy constant domain as set forth in SEQ ID NO: 15 and the V_L fragment fused to a human IgGl light constant domain as set forth in SEQ ID NO: 16

11. The antibody according to any one of claims 3 to 6, wherein said variable region or scFv that specifically binds to flagella type A comprises (a) a mouse V_H fragment comprising CDRs present in SEQ ID NO: 17; and (b) a mouse V_L fragment comprising CDRs present in SEQ ID NO: 18.

12. The antibody according to claim 11, wherein said CDRs are as defined by I GT/V-QUEST or Kabat.

13. The antibody according to claim 11, comprising the V_H fragment comprising the amino acid sequence of SEQ ID NO: 17 or SEQ ID NO: 29 and the V_L fragment comprising the amino acid sequence of SEQ ID NO: 18 or SEQ ID NO: 30.

14. The antibody according to any one of claims 3 to 6, wherein said variable domains or scFv that specifically binds to flagella type B comprises (a) a mouse VH fragment comprising CDRs present in SEQ ID NO: 31; and (b) a mouse V_L fragment comprising CDRs present in SEQ ID NO: 32 or 33.

15. The antibody according to claim 14, wherein said CDRs are as defined by IMGT/V-QUEST or Kabat.

16. The antibody according to claim 14, comprising the VH fragment comprising the amino acid sequence of SEQ ID NO: 49; and the V_L fragment comprising the amino acid sequence of SEQ ID NO: 50, or the V_L fragment comprising the amino acid sequence of SEQ ID NO: 51.

17. The antibody according to claim 1, which is a dual-specific or humanized antibody comprising the V_H fragment comprising the amino acid sequence of SEQ ID NO: 52 and the VL fragment comprising the amino acid sequence of SEQ ID NO: 53 that specifically binds to flagella type A.

18. The antibody according to claim 1, which is a humanized or a dual-specific antibody comprising the V_H fragment comprising the amino acid sequence of SEQ ID NO: 54 and the VL fragment comprising the amino acid sequence of SEQ ID NO: 55 or SEQ ID NO: 56 that specifically binds to flagella type B.

19. The antibody according to any one of claims 1 1 to 16, which is a dual-specific antibody, comprising a scFv that specifically binds to flagella type A, fused to the Fc- region of a first human IgG heavy chain molecule; and (b) a scFv that specifically binds to flagella type B, fused to the Fc-region of a second human IgG heavy chain molecule, wherein a small amino acid residue of the CH3 domain of the first human IgG heavy chain molecule is replaced with a large amino acid residue, and a large amino acid residue of the CH3 domain of the second human IgG heavy chain molecule is replaced with a small amino acid residue, and/or a large amino acid residue of the Fc-region of the first human IgG heavy chain molecule is replaced with a small amino acid residue, and a small amino acid residue of the Fc-region of the second human IgG heavy chain molecule is replaced with a large amino acid residue, whereby the large amino acid residue forms a protuberance that is positionable in the cavity formed by the small amino acid residue and thereby the first and the second human IgG heavy chain molecules form a heterodimer that specifically binds both flagella type A and flagella type B.

20. The antibody according to claim 19, wherein any of the scFvs is composed of a V_H fragment fused via a spacer peptide to a V_L fragment and the length of the spacer peptide is selected from the range of 5 to 20 amino acid residues or 7 to 15 amino acid residues.

21. The antibody according to claim 20, wherein said spacer peptide consists of three consecutive peptides of the sequence of SEQ ID NO: 57 or the amino acid sequence of SEQ ID NO: 58 (GGGSAAA).

22. The antibody according to claim 19, wherein any one of the scFvs is connected via a linker peptide to the C-terminus of the Fc-region of an IgG heavy chain molecule and the length of the linker peptide is selected from the range of 5 to 15 amino acid residues or 9 amino acid residues.

23. The antibody according to claim 22, wherein said linker peptide consists of an amino acid sequence of SEQ ID NO: 59 (PGSAGGSGG).

24. The antibody according to claim 19, wherein the small amino acid residue is threonine and the large amino acid residue is tyrosine.

25. The antibody according to claim 24, wherein threonine at position 366 (T366) of the CH3 domain of the first human IgG heavy chain molecule is replaced with a tyrosine, and tyrosine at position 407 (Y407) of the CH3 domain of the second human IgG heavy chain molecule is replaced with a threonine; or Y407 of the CH3 domain of the first human IgG heavy chain molecule is replaced with a threonine and T366 of the CH3 domain of the second human IgG heavy chain molecule is replaced with a tyrosine.

26. The antibody according to claim 19, wherein the scFv that specifically binds to flagella type A comprises a VH fragment comprising the amino acid sequence of SEQ ID

NO: 29 and the VL fragment comprising the amino acid sequence of SEQ ID NO: 18, and the scFv that specifically binds to flagella type B comprises a VH fragment comprising the amino acid sequence of SEQ ID NO: 31 and the V_L fragment comprising the amino acid sequence of SEQ ID NO: 50.

27. The antibody according to claim 26, comprising a first amino acid chain comprising the scFv that specifically binds to flagella type A linked via a linker peptide to a human IgGl Fc-fragment, said first amino acid chain having the amino acid sequence of SEQ ID NO: 60; and a second amino acid chain comprising the scFv that specifically binds to flagella type B linked via a linker peptide to a human IgGl Fc-fragment, said second amino acid chain having the amino acid sequence of SEQ ID NO: 61.

28. The antibody according to any one of claims 11 to 16, which is a dual-specific antibody comprising a chimeric antibody composed of a human IgG constant region fused to mouse variable regions of an anti- . aeruginosa monoclonal antibody that specifically binds to flagella type A, wherein said chimeric antibody is connected at its Fc-region to two scFvs that specifically bind to flagella type B of P. aeruginosa.

29. The antibody according to claim 28, wherein the chimeric antibody that specifically binds to flagella type A comprises the VH fragment comprising SEQ ID NO: 29 and the V_L fragment comprising SEQ ID NO: 30, fused to a human IgGl Fc-fragment, and each one of said two scFvs comprises the VH fragment comprising SEQ ID NOs: 31 and the VL fragment comprising SEQ ID NO: 32.

30. The antibody according to claim 29, comprising an amino acid chain comprising the heavy chain of said chimeric antibody fused to the scFvs, said amino acid chain comprising the amino acid sequence of SEQ ID NO: 62.

31. A human IgG antibody that specifically binds to flagella type B, comprising the V_H fragment fused to a human IgGl heavy constant domain as set forth in SEQ ID NO: 15 and the VL fragment fused to a human IgGl light constant domain as set forth in SEQ ID NO: 16.

32. A dual specific antibody comprising a polypeptide of SEQ ID NO: 60 and a polypeptide of SEQ ID NO: 61.

33. A dual specific antibody comprising an amino acid chain comprising the heavy chain of a chimeric antibody that specifically binds to flagella type A of P. aeruginosa, fused to scFvs that specifically bind to flagella type B of P. aeruginosa, said amino acid chain comprising the amino acid sequence of SEQ ID NO: 62.

34. The antibody according to any one of claims 1-33, wherein the antibody is covalently linked, optionally via a cleavable linker, to an antibiotic agent.

35. The antibody according to claim 34, wherein said antibiotic agent is selected from the group consisting of Amikacin, Ampicillin/Sulbactam, Amoxicillin/Calvulanic acid, Aztreonam, Cefepime, Cefotaxime, Ceftazidime, Chloramphenical, Ciprofloxacin, Colistin, Doripenem, Gentamicin, Imipenem, Levofloxacin, Meropenem, Minocycline, Piperacillin, Piperacillin/Tazobactam, Ticracillin, Tigecycline, Tobramycin and Trimethoprim-Sulfamethoxazole.

36. The antibody according to any one of claims 1-35, wherein the antibody is covalently linked to a nonproteinaceous polymer, such as polyethyleneglycol.

37. An antigen binding fragment of the antibody according to claim 1, or a derivative thereof, wherein said fragment binds flagella type A or type B of P. aeruginosa.

38. A nucleic acid molecule comprising a nucleotide sequence which encodes an antibody according to any one of claims 1 to 33, or an antigen binding fragment of said antibody.

39. The nucleic acid molecule according to claim 38, encoding a human antibody that specifically binds to flagella type B, comprising the nucleic acid sequence of SEQ ID NO: 63 encoding the heavy chain of said human antibody, and the nucleic acid sequence of SEQ ID NO: 64 encoding the light chain of said human antibody.

40. The nucleic acid molecule according to claim 38, encoding a dual-specific antibody that specifically binds to flagella type A and B, comprising the nucleic acid sequence of SEQ ID NO: 65 encoding for a scFv that specifically binds to flagella type A fused to a human IgGl fragment, and the nucleic acid sequence of SEQ ID NO: 66 encoding for a scFv that specifically binds to flagella type B fused to a human IgGl fragment.

41. The nucleic acid molecule according to claim 38, encoding a dual-specific antibody that specifically binds to flagella type A and B, comprising the nucleic acid sequence of SEQ ID NO: 67 encoding the heavy chain of a chimeric antibody that specifically bind to flagella type A fused to a scFv that specifically binds to flagella type B, and the nucleic acid sequence of SEQ ID NO: 68 encoding the light chain of an antibody that specifically binds to flagella type A.

42. The nucleic acid molecule according to claim 38, encoding a chimeric antibody selected from:

(ii) a chimeric antibody that specifically binds to flagella type B, comprising the nucleic acid sequence of SEQ ID NO: 71 encoding the heavy chain of said chimeric antibody, and the nucleic acid sequence of SEQ ID NO: 72 encoding the light chain of said chimeric antibody.

43. A vector comprising the nucleic acid molecule according to any one of claims 38 to 42, and optionally a nucleotide sequence encoding a heterologous polypeptide, operably linked to a promoter capable of driving the expression of said nucleic acid molecule.

44. A host cell, such as mouse myeloma NSO and Chinese hamster ovary cells, or plant cells, such as tobacco, carrot and rice cells, which comprises at least one vector according to claim 43.

45. A host cell, such as mouse myeloma NSO and Chinese hamster ovary cells, or plant cells, such as tobacco, carrot and rice cells, which comprises at least one vector encoding for a human antibody that specifically binds to flagella type A or type B of P. aeruginosa.

46. The host cell according to claim 45, wherein said human antibody is an IgM antibody that specifically binds to flagella type B, and comprises the VH fragment comprising the amino acid sequence of SEQ ID NO: 13; and the V_L fragment comprising the amino acid sequence of SEQ ID NO: 14.

47. A pharmaceutical composition comprising at least one antibody according to any one of claims 1 to 36, or an antigen binding fragment thereof.

48. A method for treatment of an infection caused by P. aeruginosa, comprising administering to a subject in need of such treatment a therapeutically effective amount of at least one antibody according to claim 1.

49. The method according to claim 48, wherein said subject suffers from a disease selected from the group consisting of chronic colonization in the lungs, bacteremia, sepsis, surgical wound infection, urinary tract infection, ventilator-associated pneumonia, non- ventilator associated pneumonia, obstructive pulmonary dysfunction (COPD), bronchiectasis, keratitis and ear infection {otitis externa).

50. A prophylactic method for reduction of a risk of P. aeruginosa infection in a subject susceptible to P. aeruginosa that may be exposed to P. aeruginosa, comprising administering to said patient a therapeutically effective amount of at least one antibody according to claim 1.

51. The method according to claim 50, wherein said subject susceptible to P. aeruginosa infection is an immune suppressed subject selected from the group consisting of a cancer patient, a patient awaiting transplantation, an AIDS patient, a burn patient and a very low or extremely low birth weight.

52. The method according to claim 48 or 50, wherein said antibody, or a derivative of said antibody, is administered by intravenous administration as a bolus or by continuous infusion over a period of time, by intramuscular, intraperitoneal, intracerebrospinal, subcutaneous, intra-articular, intrasynovial, intrathecal, oral, topical, or inhalation routes.

53. The method according to claim 52, in combination with the administration of (a) an antibiotic drug; (b) an antibody that targets P. aeruginosa excluding flagellin; or (c) an anti-inflammatory antibody.

54. A method for diagnosing and/or monitoring a host response to P. aeruginosa infection, comprising (a) obtaining a patient sample; (b) contacting said patient sample with an antibody according to any one of claims 1-33 or an antigen binding fragment thereof; and (c) measuring binding of said antibody or antigen binding fragment thereof to said patient sample, wherein detection of binding of said antibody or antigen binding fragment thereof to said patient sample indicates the presence or level of P. aeruginosa in the patient sample.

55. The method according to claim 54, wherein said antibody or a fragment thereof is labeled with a detectable label selected from the group consisting of fluorophores, enzymes, luminescent compounds, radioisotopes or micro- or nano-particles, enzymes, luminescent compounds, radioisotopes or micro- or nano-particles.

56. A kit for detecting P. aeruginosa in a biological sample comprising a container containing an antibody according to any one of claims 1-33; a container containing a set of reagents required for the detection and quantification of said antibody; and instructions for use.