WO2009124068A1 - Iterative screening and subtractive process for marker discovery - Google Patents

Abstract

Methods and systems designed to produce a preparation of polyantibody antigens are provided.

Description

Iterative Screening and Subtractive Process for Marker Discovery

The application claims benefit of U.S. Provisional Patent Application No. 61/072,672 filed March 31, 2008, which is incorporated by reference herein in its entirety.

Field of Disclosure

Methods and systems designed to make and use polymarkers and polymarker antibodies are provided.

Background

The availability of commercial antibodies against a common disease-causing bacterial agent or disease suggests that straightforward antibody-based assays development, of adequate diagnostic sensitivity and specificity, is possible using off-the-shelf materials. One source, (Meridian Life Science-Biodesign, 2008) lists 44 monoclonal and 22 polyclonal antibody products against the common bacterium Escherichia coli in its current catalog. Unfortunately, commercial antibodies, potentially useful for diagnosis and treatment of diseases, fail to deliver adequate diagnostic sensitivity and specificity upon sampling a large number of member organisms from the Escherchia coli target organism population in spite of a growing body of knowledge related to the antibody development know-how, protein and DNA sequenced information, and expanding receptor-ligand binding pair options currently available.

There are a number of conventional approaches to making antibodies.

The panmarker approach:

A large number of products have been developed using what can generally be referred to as a "panmarker" strategy. Such strategies generally utilize an unpurified source of virtually all target molecules potential marker such as heat killed, fixed, and/or inactivated bacterial culture containing for immunization and subsequent attempts at the generation of target-specific antibodies. Pathogenic agents often express highly prevalent repeating surface structures such as viral coat proteins, outer bacterial membranes or cell wall components that are primary stimulators of the immune system. Such immunodominant molecules are simple and, when immunized, can result in a monospecific antibody preparation that reacts primarily with the immunodominant molecules. Other potentially useful markers, such as unique surface proteins, do not produce strong antibody responses when co-injected with immunodominant molecules.

An additional challenge associated with panmarker approaches is that bacteria, viruses and other pathogens may be related to each other in unknown ways. For example, a capsular (K) type of E. coli, type A, is similar to the most common capsular type of Haemophilus influenzae, Type B (Schneerson et al, The Journal of Immunology, 1972, 108: 1551-1562.). Horizontal gene transfer may allow sharing of the same immunodominant structure in related bacteria. For example, the Shigella flexneri type 6 O-antigen was recently found for the first time in a Klebsiella pneumoniae strain (Ansaruzzaman M, et.al. Eur J Biochem. 1996 May 1 ;237(3):786-91). Immunization with a panmarker may have the unintended consequence of stimulating responses to markers which are found in non-target organisms.

The panmarker strategy for generating antibodies relies on immunodominant structures, as with serotyping, to avoid the prohibitive costs, possible lack of specificity, and potential for host animal harm. Alternative marker selection methods are required to maintaining the overall ability to specifically detect marker more uniquely associated with specific disease-causing targets and target organisms.

The monoinarker approach:

Monomarker directed antibody development toward a particular "representative" target characteristic molecular structure, or monomarker, based on specific information, such as DNA and/or protein sequence information. In this case, a purified or synthetic version of the monomarker is used to immunize animals in a more directed approach. In spite of the enormous amount of information and the ease with which a synthetic marker such as a peptide can be made, many monomarker development projects have failed employing this technique.

One significant failure mode for the monomarker approach is that the biological diversity of disease-causing target organisms is substantial. Staphylococcus aureus, for example, has a variety of well-known and clinically important identifying marker: the potential to express methicillin-resistance, capsule, coagulase activity, protein A, proteases, adhesions and other virulence factors, which may or may not be expressed depending on the strain and the sample environment. Organism diversity, on a genetic and environmental level, is not fully described by sequence information, thus complicating the in-silico selection of a single or monomarker that is omnipresent in a target organism population.

An improvement over the monomarker approach would include the marker discovery methods that do not require prior genomic or proteomic knowledge of the target and adequately represent the target organism population. Additionally, marker would ideally be available for antibody binding at all virtually all times in a target population.

There are many additional challenges associated with the generation of antibody against marker. An efficient antibody development program would quickly reduce or eliminate the use of weakly immunogenic marker prior to the initiation of an immunization project, would not require prediction of immunogenic structures, and would use native marker structures rather than synthetic versions.

Brief Description of the Drawings

Figure 1 depicts IFM staining of ATCC^® 49139 with prototype ATCC^® 19606 (Type 1) chicken IgY antibody and alexa-555 conjugated goat anti-chicken secondary antibody. Panel

A shows a phase contrast image of bacteria immobilized onto the PLL slide surface and panel B shows the matched image of the same cells under fluorescent excitation. The region shown here represents approximately 20% of the whole analyzed image. Staining efficiency and intensity values calculated for this sample were 93% stained with average intensity of 2.4- fold over the slide background.

Summary

This disclosure is directed towards the methods of preparing polymarker antibodies for the rapid discovery of target organism polymarker, biotyping of target organisms with in a collection of target organisms using polymarker antibodies, utilizing polymarker antibodies to substantially detect the target, and subtracting cross reactive antibodies from the polymarker antibodies for diagnostic and therapeutic applications.

Detailed Description

Methods of preparing a polyantigen antibody preparation that bind the cell surface of a cell type are described. The method includes obtaining a cell membrane portion of lysed cells. The cell membrane portion includes a substantial quantity of markers, e.g. cell surface antigens of the cell type. The cell surface antigens are immobilized on a solid surface. The immobilized cell membrane portion is contacted with a plurality of antibodies to immobilize a plurality of cell surface antibodies that bind one or more markers. The cell surface antibodies are collected from said solid surface to produce a polyantigen antibody preparation.

hi other embodiments, methods of identifying markers are described. Those of skill in the art will understand that the requirements and suitability of a particular class of target molecules or marker can be adapted to antibody binding assay conditions (i.e., lysed cells, whole cells, protein extract identification, etc.).

In order that the disclosure may be more completely understood, several definitions are set forth below. Such definitions are meant to encompass grammatical equivalents.

Definitions

As used herein, the term "molecule class" refers to a conventional class of molecule types, such as proteins, polysaccharides, or lipids.

As used herein, the term "marker" refers to a molecule present in the target organism. Examples of marker include a surface protein, capsular polysaccharide, and lipopolysaccharide. The term marker can refer to a cell surface antigens.

As used herein, the term "polymarker" refers to a plurality of markers.

As used herein, the term "biotype" refers to a common physiological characteristic of a group of target organisms. The aggregate biotypes are representative of the target biological diversity. Biotypes can correlate to a cell type, such as cells from a particular tissue, a class, family, species, strain, phenotype, and/or serotype of microorganism.

As used herein, the term "antibody" refers to a protein consisting of one or more polypeptides substantially encoded by all or part of the recognized immunoglobulin genes. The recognized immunoglobulin genes, for example in humans, include the kappa (K), lambda (λ), and heavy chain genetic loci, which together comprise the myriad variable region genes, and the constant region genes mu (υ), delta (δ), gamma (γ), sigma (σ), and alpha (α) which encode the IgM, IgD, IgG (IgGl, IgG2, IgG3, and IgG4), IgE, and IgA (IgAl and IgA2) isotypes respectively. Antibody herein is meant to include full length antibodies and antibody fragments, as described in more detail herein, and may refer to a natural antibody from any organism, an engineered antibody, or an antibody generated recombinantly for experimental, therapeutic, or other purposes.

As used herein, the term "marker-antibody" refer to an antibody capable of binding to a marker.

As used herein, the term "polymarker-antibody" refer to a collection of marker- antibodies, that together are capable of binding to member organism or class of organisms from the target organism population.

As used herein, the term "confidence" as used herein refers to the lower limit of the 95% confidence interval (LLCI₉₅), meaning that there is a 95% chance that at least x% of all samples tested will have the correct result. The value for LLCI₉₅ with a given number of tests is calculated by the R.G. Newcombe method of efficient score (corrected for continuity) as described for the web-based calculator at http://faculty.vassar.edu/lowry/clinl .html. Testing a collection of 20 strains would provide an LLCI₉₅ predictive confidence value of 80%, providing substantial confidence of the assay result on an unknown target organism. A collection of 10 strains would provide 65% LLCI₉₅ predictive confidence while a collection of 100 strains would provide 95% LLCI₉₅ predictive confidence. A positive result on a strain provides a degree of confidence that an unknown sample containing a target organism will also provide a positive result.

Preparing Polyantigen Antibody Preparations

A. Markers

Antibody preparations are prepared that identify markers, such as markers on the surface of a class of cells. In the case of diagnostics the marker may correlate to a disease state or identifiable medical condition. In a therapeutic context the binding of a functional antibody to the marker may be useful for blocking, inhibiting, modifying the role of the marker in the disease state, resulting in a therapeutic benefit. Examples of a marker include: a native biological structure (such as a protein), a mutated biological structure, a modified biological structure, a cell, an organism, a class of organisms, a family of organisms, a genus, a species, a strain of microorganism, or a functional grouping which distinguishes one organism from a similar one (such as an antibiotic resistant bacterium). Markers may also be a collection of cell surface antigens, whether bound or not bound to the membrane.

In certain embodiments, target cells presenting the markers include animal, plant, yeast, and microbial cells. The cells can be selected from a specific type of tissue. Alternatively, the cells can be diseased or disordered cells (e.g. tumor cells). Examples of non-cellular targets that may present markers includes but is not limited to; viral cells, soluble proteins, secreted compounds, such as secreted toxins, or deposited matrices.

In certain embodiments, the target is a target organism such as a microorganism. Non-limiting examples of target organisms include, but are not limited to, a single organism species (e.g. Staphylococcus aureus, fungi, parasites, mycobacteria, or any whole organism), a single organism genus (e.g. Staphylococcus), a broad category of organisms (e.g. gram- positive bacteria), or a narrow category of organisms (methicillin resistant Staphylococcus aureus). Representative target organism strains are obtained from ATCC^® or from clinical sample repositories. The representative strain set usually contains "type strains" and those used as controls for conventional testing (i.e., control strains for clinical microbiological CLSI testing).

The number of organisms included in the collection of cells making up the cell type may depend on the target organism properties including but not limited to factors associated with the sample environment in which the organisms are found, geographic factors, and the organism natural history. The degree of representation of the strain sets is additionally dependent on the biological diversity expressed at various levels. Biological diversity between individual targets with regards to their structure (structural diversity), differences between targets presented by individual organisms (member diversity), differences between types of target organism (type diversity), differences in target marker presentation based on sample environment (sample diversity), and differences in target marker presentation based on external stimuli (phenotypic diversity). A single bacterial species may have many types or strains which differ significantly from one another in many aspects: marker may have slight differences between strains due to mutations, a sample containing a clonal population of bacteria may exhibit different levels of marker expression among individuals, different strains of the species may express different marker, a sample of the bacteria in blood may present different marker or levels of a given marker than the same bacteria in urine, and the bacteria may present different marker before and after antibiotic treatment of the patient. The collection of target organisms should include a sufficient number of individual examples of the target to represent each type of diversity that might be present in a sample containing the target.

The number of selected strains may also influenced by the diagnostic assay sensitivity objectives and the required assay statistical confidence level and a direct correlation between the number of strains in the collection and its' ability to represent the entire target population.

By way of illustration, but not limitation, a target organism strain collection designed for a Klebsiella clinical pneumonia assay consisted of 3 Klebsiella pneumoniae strains from ATCC (including a type strain), 12 Klebsiella pneumoniae clinical isolates from pneumonia patients collected from several sites in the U.S. over a prior year period, 1 Klebsiella oxytoca strain from ATCC and 4 Klebsiella oxytoca clinical isolates from pneumonia patients collected from several sites in the U.S. over the last most recent prior year period. The collection represent a pneumonia-associated species, of the Klebsiella genus, temporally recent isolates, and a broad sampling of geographic locations within the U.S.

A representative member organism may be selected as a first portion from the target collection. In some cases, the first portion is a "type strain" as defined by the scientific community, however it may be chosen by any of a number of criteria.

For illustration purposes, the type strain of Klebsiella pneumoniae ATCC^® 13883 is chosen as the first portion of the target organism collection.

B. Markers

Markers may encompass multiple types of biological molecules. A marker may be selected from extracellular matrix (biofilm), aqueous soluble, aqueous insoluble, or detergent solubilized, associated molecules. Additionally, the marker may be further refined by location or cellular compartment. Membrane associated, intracellular, cell signal associated, cell surface molecules, cell membrane, or any combination of the above molecules may be selected. Further refinement of marker would be at the molecule class level and could also be proteins, peptidoglycan, and lipopolysaccharides. A single molecule can be a marker.

In other aspects, member organism marker represents the expression profile or "phenotype" for the selected molecules. Many molecular classes from a first portion of a target organism may be required to detect the target organism polymarker.

For illustration purposes, surface immobilized live cell detection of the Klebsiella genus in an immunofluorescence microscopy assay format is required. The marker selected for this purpose will be directed toward molecule class that are surface exposed and available for antibody binding on live cells. Several molecule classes are possible candidate groups of markers, including outer membrane lipopolysaccharides, capsular polysaccharides, extracellular matrix polysaccharides, secreted proteins, and surface anchored proteins. Since Klebsiella are not motile and have an outer membrane covering their cell wall, flagellar structures and cell wall peptidoglycans are excluded from the candidate marker options. The first molecule class examined will be surface-anchored proteins, since these marker are cell- associated.

C. Obtaining Markers

Any one of many types of strategies may be used to obtain cell surface markers, including cell surface antigens in various molecular classes. Cells can be first lysed to release the cell membrane portion, including all or a portion of molecular classes, from the whole cell. Examples include, but are not limited to: mechanical disruption methods, liquid homogenization, sonication, freeze/thaw methods, mortar and pestle methods, hypertonic or hypotonic suspension, detergent treatment, and enzymatic treatment. Mechanical methods refer to the use of blades to grind and disperse large amounts of tissue. The Waring blender and the Polytron are commonly used for this purpose.

In general, liquid-based homogenization lyses cells by forcing them through a narrow space, thereby shearing the cell membranes. Sonication physically disrupts cells through the use of pulsed, high frequency sound waves to agitate and lyse cells, bacteria, spores and finely diced tissue. Freeze/thaw methods lyse cells by freezing and then thawing them at room temperature or 37°C in a process that causes cells to swell and ultimately break as ice crystals form during the freezing process and then contract during thawing. Cells (especially plant cells) can be manually ground, for example by using a mortar and pestle. Alternatively, cells can be treated with glass beads to facilitate the crushing of cell walls. Lysis can be promoted by suspending cells in a hypotonic (low ionic strength) buffer causing the cell to take up fluid from the surroundings to maintain its osmotic balance. Alternatively, a hypertonic (high ionic strength) buffer causes the cell to secrete fluid which may also damage the cell membrane or wall sufficiently to lyse the cells. Detergents are capable of dissolving the cell membranes of many cell types, as well as having the property of dissociating surface structures without fully lysing the cells, depending on the type and concentration of detergent used. Cells may also be treated with a chemical agent or enzyme to remove specific types of material. For example, lysozyme can be used to digest the polysaccharide component of yeast and bacterial cell walls.

In certain embodiments, molecular classes may be separated from each other. Molecule classes are separated by a number of methods. Examples include separation of organelles by differential centrifugation, ammonium sulfate precipitation of proteins, extraction of lipids from solubilized membranes using organic solvents, differential extraction of lipopolysaccharides using phenol, precipitation of extracellular proteins using acetone, size exclusion chromatography, ion exchange chromatography, and affinity chromatography. Separation of the chosen molecular class from other classes is desirable since each type of molecule class may have limitations which are not well understood. However, a given or known degree of purification homogeneity of the molecule class is not required.

One or more markers may be first tagged to enable facile post lysis separation. The tag can be a molecule that binds to a solid surface. The tagging can be performed using a number of methods available from commercial vendors and described in product manuals such as the Pierce Biotechnology Product Catalog and references cited therein, all expressly incorporated herein by reference. Examples of tagging molecules include but are not limited to, primary amine-reactive activated tagging compounds, nucleotide-reactive tagging compounds, carboxyl-reactive tagging compounds, sulfhydryl-reactive tagging compounds, and photoreactive tagging compounds. The tag can be a small molecule or unique structure such as biotin, glutathione FLAG, or multi(His), that can be selectively bound or captured by an immobilized affinity binding reagent such as avadin, streptavidin, glutathione-S- transferase, lectins or metal chelating columns. Such a strategy can be used to selectively label the desired marker present in a cellular compartment (such as the cell surface) for inclusion or removal from all other molecules. For example surface proteins may be selectively biotin tagged in order to separate them from other molecules (such as intracellular proteins) and from other molecule classes, such as outer membrane lipopolysaccharides.

For illustration, and not limitation, surface proteins from approximately 10¹⁰K pneumoniae ATCC^® 13883 bacteria may be tagged and separated from other molecules using a non-permeable activated biotin, sulfo-NHS-SS-biotin reagent(Pierce). This strategy is designed to covalently tag only surface proteins having exposed primary amine groups with the biotin structure using a spacer arm that contains a cleavable disulfide bond that may be cleaved to remove the tag after purification. The surface protein class of molecules are a group of markers for the said strain. The cells are washed to remove media components and resuspended in 1 mL of phosphate buffered saline (PBS) at pH 7.4. Next, 100 μL of the 6 mg/mL sulfo-NHS-SS-biotin is added to the cells and the tagging reaction is allowed to proceed at 4⁰C. After 30 minutes exposure to the activated biotin, the reaction is quenched using a large excess of primary amine (100 mM Tris, pH 8.0) and the bacteria are washed to remove unbound biotin. The bacteria are then treated with an extraction reagent (such as Pierce B-PER^® or Triton X-100 detergent) and/or enzymes (such as lysozyme) so that the tagged proteins are liberated from the surface and released into the solution. Inhibitors of degrading enzymes such as proteases may also be added to maintain the target molecule class in a more complete state.

D. Marker immobilization on a solid surface

The group of markers is then immobilized on a solid surface. All or a portion of the markers can be immobilized on a solid surface. In various embodiments, a substantial quantity of markers may be immobilized, hi this context, a substantial quantity of markers may be at least 20% of markers, 30% of markers, 40% of markers, 50% of markers, 60% of markers, 70% of markers, 80% of markers or 90% of markers.

Immobilization can be accomplished by any method known in the art. Non-limiting examples include, tag affinity capture, such as biotin capture using immobilized streptavidin; covalent coupling, such as coupling proteins through primary amines onto cyanogen bromide-activated beads; or adsorption, such as binding of nucleic acids to glass beads or adsorption of proteins to a polystyrene plate.

Markers can be prepared for immobilization by chemical activation, buffer change, or pH adjustment. The solid surface may also be chemically activated or otherwise made capable of accepting the marker for binding. Alternatively, the marker may be chemically activated or other made capable of binding to the solid surface. It will be recognized by those skilled in the art that there are many potential methods for immobilizing the markers on a solid surface, hi the case of the activated markers or solid surface, the markers and solid surface are combined and incubated for a period of time to allow the binding to take place. Preferably, the interaction is covalent, however the interaction may include but not be limited to ionic, hydrogen bonding, and or Van der Waals forces in combination or individually immobilizing the group of one or more markers to the solid surface. Receptor-ligand interactions, including but not limited to, biotin streptavidin, in which proteins tagged with activated biotin are captured streptavidin coupled to agarose beads in the column. In this case, a substantial portion of untagged molecular classes are removed by washing the beads with a buffer solution. Depending on the chemistry and mechanisms, durations of up to 24 hours and temperatures up to 4O⁰C may be required. After the reaction is complete, the mixture may require quenching with an excess amount of reactant capable of binding the remaining activated group on the substrate. The column is washed to remove unbound material and the immobilized group of one or more markers bound to the solid surface are now ready for use as, for example, an antibody affinity matrix.

In certain embodiments, the solid surface immobilization can be part of an analytical method. The solid surface is preferably a particulate material that is subsequently packed into a column matrix and may be the stationary phase of a chromatography column.

In the illustration, the biotin-tagged K. pneumoniae ATCC^® 13883 surface proteins are exposed to beads that have a covalently-attached biotin receptor, streptavidin, which is capable of capturing the tagged proteins. Beads of this type are available through vendors (such as Pierce # 20351). Unbound materials from the lysate are removed with extensive washing with a neutral pH buffer containing detergent that maintains the solubility of the attached surface proteins and facilitates extraction of surrounding membrane components from the immobilized proteins. After washing, the beads are suitable for use as an affinity matrix.

A substantial quantity of markers can be used.

E. Contacting the immobilized markers with antibodies

The immobilized markers are then contacted with a plurality of antibodies. The antibody source material is prepared so that it is compatible with antibody binding (neutral pH, isotonic) and then brought into contact with the affinity matrix to allow specific antibodies to bind to the immobilized marker structures. Antibodies that do not bind any structures from the immobilized marker are removed by washing.

The unbound material from the antibody source can be removed by any method known in the art. For example, the unbound material can be removed by washing with a neutral buffer such as PBS pH 7.4. In addition, the bound antibodies may be stringently washed using a variety of solutions that are selected for their ability to remove non- specifically bound materials as well as weakly avid antibodies. For example, a low tonicity buffer, such as 10 mM Tris pH 7.4; a high tonicity buffer, such as IM NaCl in 10 mM Tris pH 7.4; a reduced pH buffer, such as 10 mM sodium acetate at pH 5.5; an elevated pH buffer such as 10 mM sodium borate at pH 9.0, or detergent containing buffer such as PBS containing 0.1% Tween-20 may be used as stringent wash buffers. Selection of a stringent wash buffer is empirical and should be understood as an optional step by those skilled in the art.

The antibody source may be of any type known in the art. In general, antibodies are immunological proteins that bind one or more structures. In most mammals, including humans and mice, antibodies are constructed from paired heavy and light polypeptide chains. The light and heavy chain variable regions show significant sequence diversity between antibodies, and are responsible for binding the structure. Each chain is made up of individual immunoglobulin (Ig) domains, and thus the generic term immunoglobulin is used for such proteins.

Ig structural units typically comprise a tetramer. Each tetramer is typically composed of two identical pairs of polypeptide chains, each pair having one "light" chain (typically having a molecular weight of about 25 kDa) and one "heavy" chain (typically having a molecular weight of about 50-70 kDa). Each of the light and heavy chains are made up of two distinct regions, referred to as the variable and constant regions. For the IgG class, the heavy chain is composed of four immunoglobulin domains linked from N- to C-terminus in the order V_H-CHI -CH2-CH3, referring to the heavy chain variable domain, heavy chain constant domain 1, heavy chain constant domain 2, and heavy chain constant domain 3 respectively (also referred to as V_H-C7l-Cγ2-C73, referring to the heavy chain variable domain, constant gamma 1 domain, constant gamma 2 domain, and constant gamma 3 domain respectively). The IgG light chain is composed of two immunoglobulin domains linked from N- to C- terminus in the order V_L-C_L, referring to the light chain variable domain and the light chain constant domain respectively. The constant regions show less sequence diversity, and are responsible for binding a number of natural proteins to elicit important biochemical events.

The variable region of an antibody contains the antigen binding determinants of the molecule, and thus determines the binding characteristics of an antibody for its target antigen. The variable region is so named because it is the most distinct in sequence from other antibodies within the same class. In the variable region, three loops are gathered for each of the V domains of the heavy chain and light chain to form an antigen-binding site. Each of the loops is referred to as a complementarity-determining region (hereinafter referred to as a "CDR"), in which the variation in the amino acid sequence is most significant. There are 6 CDRs total, three each per heavy and light chain, designated V_H CDRl, V_R CDR2, V_H CDR3, V_L CDRl, V_L CDR2, and V_L CDR3. The variable region outside of the CDRs is referred to as the framework (FR) region. Although not as diverse as the CDRs, sequence variability does occur in the FR region between different antibodies. Overall, this characteristic architecture of antibodies provides a stable scaffold (the FR region) upon which substantial antigen binding diversity (the CDRs) can be explored by the immune system to obtain specificity for abroad array of antigens. A number of high-resolution structures are available for a variety of variable region fragments from different organisms, some unbound and some in complex with antigen. Sequence and structural features of antibody variable regions are disclosed, for example, in Morea et al., 1997, Biophys Chem 68:9-16; Morea et al., 2000, Methods 20:267-279, and the conserved features of antibodies are disclosed, for example, in Maynard et al., 2000, Annu Rev Biomed Eng 2:339-376, each incorporated herein it its entirety by reference.

Antibodies described herein may be substantially encoded by genes from any organism, e.g., mammals, including but not limited to humans, rodents including but not limited to mice and rats, lagomorpha including but not limited to rabbits and hares, camelidae including but not limited to camels, llamas, and dromedaries, and non-human primates, including but not limited to Prosimians, Platyrrhini (New World monkeys), Cercopithecoidea (Old World monkeys), and Hominoidea including the Gibbons and Lesser and Great Apes. Antibodies may also derive from non-mammalian sources including, but not limited to avians, such as chickens, turkeys, and ducks; reptiles, such as snakes or lizards; and amphibians, such as frogs or salamanders.

The antibodies can be human antibodies. The antibodies of the present disclosure may be substantially encoded by immunoglobulin genes belonging to any of the antibody classes. In one embodiment, the antibodies of the present disclosure comprise sequences belonging to the IgG class of antibodies, including human subclasses IgGl, IgG2, IgG3, and IgG4. In an alternate embodiment, the antibodies of the present disclosure comprise sequences belonging to the IgA (including human subclasses IgAl and IgA2), IgD, IgE, IgG, or IgM classes of antibodies. The antibodies of the present disclosure may comprise more than one protein chain. That is, the present disclosure may find use in an antibody that is a monomer or an oligomer, including a homo- or hetero-oligomer. Antibodies can be full-length antibodies, which have an antibody structure that constitutes the natural biological form of an antibody, including variable and constant regions. For example, in most mammals, including humans and mice, the full length antibody of the IgG class is a tetramer and consists of two identical pairs of two immunoglobulin chains, each pair having one light and one heavy chain, each light chain comprising immunoglobulin domains V_L and C_L, and each heavy chain comprising immunoglobulin domains V_H, CHl (Cγl), CH2 (Cγ2), and CH3 (Cγ3). In some mammals, for example in camels and llamas, IgG antibodies may consist of only two heavy chains, each heavy chain comprising a variable domain attached to the Fc region.

Alternatively, the antibodies can be a variety of structures, including, but not limited to antibody fragments. Antibody fragments include but are not limited to bispecific antibodies, minibodies, domain antibodies, synthetic antibodies, antibody mimetics, chimeric antibodies, antibody fusions (sometimes referred to as "antibody conjugates"), and fragments of each, respectively. Specific antibody fragments include, but are not limited to, (i) the Fab fragment consisting of VL, VH, CL and CHl domains, (ii) the Fd fragment consisting of the VH and CHl domains, (iii) the Fv fragment consisting of the VL and VH domains of a single antibody; (iv) the dAb fragment, which consists of a single variable region, (v) isolated CDR regions, (vi) F(ab')2 fragments, a bivalent fragment comprising two linked Fab fragments (vii) single chain Fv molecules (scFv), wherein a VH domain and a VL domain are linked by a peptide linker which allows the two domains to associate to form an antigen binding site (viii) bispecific single chain Fv dimers and (ix) "diabodies" or "triabodies", multivalent or multispecifϊc fragments constructed by gene fusion. The antibody fragments may be modified. For example, the molecules may be stabilized by the incorporation of disulfide bridges linking the VH and VL domains. Examples of antibody formats and architectures are described in Holliger & Hudson, 2006, Nature Biotechnology 23(9): 1126-1136, and Carter 2006, Nature Reviews Immunology 6:343-357 and references cited therein, all expressly incorporated by reference.

Antibodies may include multispecific antibodies, notably bispecific antibodies, also sometimes referred to as "diabodies". These are antibodies that bind to two (or more) different antigens. Diabodies can be manufactured in a variety of ways known in the art, e.g., prepared chemically or from hybrid hybridomas. In one embodiment, the antibody is a minibody. Minibodies are minimized antibody-like proteins comprising a scFv joined to a CH3 domain. In some cases, the scFv can be joined to the Fc region, and may include some or all of the hinge region. For a description of multispecific antibodies see Holliger & Hudson, 2006, Nature Biotechnology 23(9):1126-1136 and references cited therein, all expressly incorporated herein by reference.

Antibodies may be obtained from any readily available source, whether naturally derived or synthetic. In various embodiments, the antibodies may be in, or alternatively obtained from, blood, blood serum or egg yolk. Any source that contains a plurality of antibodies can be used.

F. Collecting polymarker antibodies

Once the affinity matrix is sufficiently washed, the bound antibodies are removed using an elution buffer. Typical elution buffers are designed to temporarily interrupt the bonds that exist between an antibody and its conjugate structure and then reconstitute the binding ability by returning the eluted antibody to a native environmental state. Since these bonds can vary considerably in number and strength of binding, the necessary conditions for elution of any given antibody require empiric determination. Most antibodies are eluted by a low pH buffer such as 100 mM glycine pH 3.0 or a high pH buffer such as 100 mM ethanolamine at pH 10.0. More stringent conditions such as lower or higher pH buffers, use of dissociating agents or temporary denaturation by heating or ionic detergent treatment may be necessary for some strongly bound antibodies. Following elution, the antibodies are neutralized or diluted in a buffer solution that is capable of returning them to their original configuration. It should be understood by those skilled in the art that elution may instead affect the immobilized marker structure to which the antibody is bound, causing a conformational change that release the antibody. The antibody can be identified by any method known in the art. The resulting preparation of eluted polymarker antibodies contains a substantial quantity of antibodies directed at a marker or group of markers identifying the target organism polymarker. In this context, a substantial quantity of antibodies refers to at least 20% of antibodies, 30% of antibodies, 40% of antibodies, 50% of antibodies, 60% of antibodies, 70% of antibodies, 80% of antibodies or 90% of antibodies.

The eluted polymarker antibodies may also contain antibodies directed at structures from the immobilization matrix, hi some embodiments, these may be removed by exposing the neutralized eluted antibodies to a second matrix that is identical to the described affinity matrix, without the immobilized markers. This is termed here a "dummy matrix". For the purposes of illustration, a preparation of chicken egg yolk immunoglobulin (IgY) purified from normal eggs is used as the source of antibodies. The antibodies are passed over the affinity matrix containing the biotin-tagged surface proteins bound to the covalently attached streptavidin - agarose beads. Unbound materials are washed out using PBS, and specific antibodies are released by exposing to a 100 mM glycine buffer at pH 3.0. The released antibodies are neutralized with buffer (100 mM Tris at pH 8.0) to regenerate their binding ability. The eluted antibodies are then passed over the dummy matrix (i.e., streptavidin agarose beads) to remove antibodies directed against the matrix further purifying the polymarker antibodies.

G. Testing the polymarker antibodies

The polymarker antibodies can be tested and detected by any of a number of methods known in the art. There are many types of assays that may be used to detect antibody binding. Examples include but are not limited to immunofluorescence microscopy, immunohistochemical staining, plate immunofluorescence, flow cytometry, ELISA, radioimmunoassay, western blot, immunochromatography, lateral flow, slide agglutination, latex bead agglutination, magnetic particle capture, immunodiffusion, immuno-MALDI- TOFMS, and immunomicroarray.

Polymarker antibodies suitability depends on the desired application. One important parameter used to determine suitability is the minimum amount of the signal required to obtain measurable antibody binding. The measured antibody binding signal is typically compared to an assay "negative control" to define the sensitivity against non-specific background binding. Negative control samples typically use the same binding assay procedure measuring the readout obtained in the absence of the target organism. Additional specificity controls may be used to measure the contribution of secondary detection reagents to the signal obtained on the target sample as well as on the negative control sample, hi this way, the true contribution of the polymarker antibodies to the measured signal can be demonstrated. Detection occurs when the antibodies have sufficient affinity or avidity to generate reliably measurable signal in the desired application.

Furthermore, those skilled in the art will recognize that there are many methods by which detection can be defined including but not limited to, statistical signal processing techniques such as the student's T- test, image analysis techniques including pixel co- localization, as well as the more commonly used "signal-to-noise" ratio based detection criteria.

For the purposes of illustration, a bulk sample analysis fluid-phase assay such as ELISA requires that the polymarker antibodies bind to an immobilized target organism. The target organism may be intact or lysed. The antibody binding is then detected using enzymatic substrate conversion over a defined time period. If the polymarker antibodies' binding is sufficient, a detectable amount of conversion over a negative control sample which does not contain any target organism will occur. A threshold level of conversion above a multiple of the negative control readout (such as 2-fold signal-to-noise ratio) is commonly used to determine the positive or negative test result and thus the sample sensitivity threshold. Assay controls may also be performed in which the secondary reagent enzyme conjugate is tested without the marker primary antibody against the target and negative control samples.

In contrast to the above illustration and for further description, an immunofluorescence microscopy assay of discrete cells requires not only a sufficient antibody binding signal above the background for each cell stained, but may also require that a certain percentage of the cells are stained at detectable levels. In this illustration, detection means a minimum percentage of cells are stained at a minimum individual cell staining intensity compared to the surrounding background. Staining efficiency is the measure of the percentage of cells stained at detectable levels. Controls may also be useful in this assay type.

Live bacteria can be immobilized on a solid surface by any method known in the art, including antibody adhesion or non-specific binding methods utilizing adhesion including but not limited to forces of electrostatic, hydrophobic, and van der Waals origin.

Cell preparation viability can also be tested in conjuction with the polymarker antibodies binding. Live bacteria can be identified by a number of different methods known in the art, including microscopy based growth methods and fluorescence based detection methods as described in the Molecular Probes, "The Handbook — A Guide to Fluorescent Probes and Labeling Technologies" and references cited therein, all expressly incorporated herein by reference.

In cases of therapeutic applications, it may be desirable for a preparation of polymarker antibodies to competitively block antibody binding. An example of blocking assay would be inhibiting the bacterium binding to an epithelial surface using a preparation of polymarker antibodies. Unbinding of the polymarker antibodies using a competitive ligand may also be desired. In certain processes, antibodies that can quickly capture and remove a target structure from a process flow are desirable. An example of such a process might be to remove a protease from a fermentation process, used to produce a recombinant protein, by cycling the media through an affinity matrix containing a preparation of polymarker antibodies that binds and captures the protease simultaneously blocking its function.

For the purposes of illustration, the preparation of polymarker antibodies obtained from K. pneumoniae ATCC 13882 surface protein markers may be examined by immunofluorescent microscopy (IFM). The IFM assay has the following pre-determined parameters for determining a true positive result: the cells must be stained with at least 85% efficiency and have an average staining intensity of at least 2-fold over the surrounding slide background.

H. Determination of target collection biotypes

In another embodiment of the disclosure, the prepared polymarker antibodies bind to a second portion of target organisms with sufficient affinity and/or avidity resulting in detection, indicating that the second portion of target organisms share significant amounts for one or more of the markers with the polymarker from the first portion of target organisms. The first portion and second portion of target organisms are said to be of the same biotype.

Members from second portion of the target organisms that are not detected are considered to be of a different biotype since they do not share sufficient markers from the first portion of target organisms. It should be understood by those of sufficient knowledge in the art that this does not mean that one or more the markers are not present, but that they can be masked by other structures, not be present in sufficient amounts for detection, or be modified in such as way as not to be bound by the preparation of polymarker antibodies. The lack of antibody binding may also be considered to be a result of different target organism phenotype. In all such cases above the member organisms are said to be of different biotype.

The polymarker antibodies' target organism relatedness can be determined by testing with other portions of the target organism collection. For example, the polymarker antibodies may stain the type strain ATCC^® 13882, and 4 other strains from the collection and the 5 strains are said to be of similar biotype. The biotype classification can be used for epidemiology studies or other applications where classification is desirable. This type of study may also be used to examine dynamic phenotype changes of the target organism where markers may be dynamically expressed as a result of stimulation, such as treatment of a resistant bacterium with an antibiotic, or as the result of temporal progression.

In other aspects of this disclosure, detection of target cells expressing mutant protein may be determined by the relatedness of suspected mutated target cells to normal cells. Relatively rare and unique marker intermixed with related molecules from non-target cells may be used to trace a particular feature, such as the expression of a mutant protein by a tumor cell or an exogenous protein in recombinant protein fermentation.

I. Iterative marker discovery hi other aspects of the disclosure, the second marker or group of markers are obtained from a second portion of the target organism collection. A second preparation of polymarker antibodies are obtained, as previously described herein.

For an illustration of this embodiment, surface proteins from the K. pneumoniae ATCC^® 9997 are selected as the second marker. Purification of this marker is performed as described above for the surface protein of K. pneumoniae ATCC^® 13883, antibodies are introduced to the affinity matrix, unbound materials are removed by washing, and marker antibody against the K. pneumoniae ATCC^® 9997 marker are eluted by pH change. The marker antibody preparation is used in the immunofluorescence assay to analyze all members of the target collection and to determine the expressed bio types related to the second portion of the target organism collection K. pneumoniae ATCC 9997.

In other aspects, the collection of antibodies is not possible using the immobilized group of markers from the first portion of target organisms. In this case, the first portion of target organisms may be considered "untypeable" using markers of this molecular class, hi some cases this may be due to a variety of factors including but not limited to, masking of the molecule class, lack of sufficient amounts of markers, or variable phenotypical expression. hi other aspects of the disclosure, the second group of markers are obtained from a second molecule class from the first portion of the target organism collection. A second preparation of polymarker antibodies are obtained, as previously described.

For illustration, the process for purifying capsular polysaccharides from the K. pneumoniae ATCC^® 13883 and obtaining the corresponding preparation of polymarker antibodies is described. Purification of this molecule class is performed by washing the cells to remove media, heat treating the cells for 1 hour at 8O⁰C, and removing the cells by centrifuging, retaining the released crude capsular polysaccharides. The capsular polysaccharides may be further purified by treatment with nucleases and proteases, and then precipitated with 75% ethanol at O⁰C. The precipitated capsular polysaccharides are then centrifuged and the pellet resuspended in an immobilization buffer. The purified capsular polysaccharides may be immobilized by covalent binding to an epoxy-activated sepharose bead. The epoxy-activated sepharose beads are treated with 1,4-bis (2,3-epoxy-propoxy-) butane to produce an activated epoxy group which is capable of linkage to carbohydrates through hydroxyl groups. The capsular polysaccharides are diluted with IM NaOH to raise the pH to a level higher than pH 12. The beads are washed with 100 mM NaOH, combined with the capsular polysaccharides, and allowed to react for 24 hours at 35°C. The mixture is quenched and neutralized by the addition of IM Tris at pH 7.4. The beads are subsequently washed to restore the neutral pH and prepare for affinity binding of antibodies. As detailed herein, antibodies are introduced to the affinity matrix, unbound materials are removed by washing, and polymarker antibodies against AT. pneumoniae ATCC^® 13883 capsular polysaccharides are eluted by pH change. The collected polymarker antibodies are tested against all members of the target collection, using the immunofluorescence microscope assay, to determine the capsular polysaccharide biotypes of the identified K. pneumoniae ATCC^® 13883 polymarker. hi other aspects of the disclosure, the second preparation of polymarker antibodies are tested against a portion of the other member organisms from the target organism collection. The extent that the second preparation detects the target organism population is assessed. hi other aspects of the disclosure, additional preparations of polymarker antibodies are iteratively produced, as described herein. The iterative process is repeated, identifying target organism polymarker and biotypes thereof, until detecting with at least one preparation of polymarker antibodies preferably all target organisms from the target organism collection.

For illustration, K. pneumoniae ATCC^® 700603, did not produce a true positive result with the first or second polymarker antibody preparations directed against the surface protein polymarkers from K. pneumoniae ATCC 13883 and K. pneumoniae ATCC 13882 respectively. Surface protein markers from this strain were prepared using the procedure described herein and the prepared polymarker antibodies were tested against the target collection. The collected polymarker antibodies, in this case, produced a true positive result on the K. pneumoniae ATCC^® 700603 as well as 2 additional member organisms, indicating that these three strains are of a different biotype than K. pneumoniae ATCC^® 13883 and K. pneumoniae ATCC^® 13882. Subsequent iteration of this process on the remaining target organisms identifies two additional biotypes that, in total, yield five biotypes for the entire 20 member target organism collection.

In other aspects, the group of markers from a plurality of target organisms are combined and immobilized.

In other aspects of the disclosure, the preparation of polymarker antibodies can be refined. For example, the preparation can be iteratively bound a second time to the immobilized group of markers.

A second refinement involves a preparation of polymarker antibodies against a marker or group of markers using a different antibody source, either comprising a different sample from the same antibody source species or a sample from a second antibody source species. In this way, the antibody host species' capability to recognize the markers can be assessed and the preparations can be further improved. For illustration, a second preparation of polymarker antibodies against the K. pneumoniae ATCC^® 13883 surface protein markers may be purified using a different lot of IgY. The second and first preparations may be combined in order to control the lot-to-lot variation of preparations. Alternatively, a different antibody source, such as goat serum, may be used to obtain the polymarker antibodies. This strategy may be useful in selecting an antibody type that has desirable properties for detection in the antibody binding assay, such as better avidity for the target or less background in the assay. Polymarker antibodies prepared from different lots of antibodies or from different species hosts may be combined.

In other aspects, the collected polymarker antibodies are combined.

In another embodiment, the processes described herein may be used to generate polymarker antibodies or a combination of polymarker antibodies that detect a substantial quantity of the target organism collection.

Animal immunization with polymarker

In other aspects, a target organism polymarker may be used for animal immunization.

In other aspects, combined polymarker from a plurality of portions of the target organisms may be used for animal immunization.

In other aspects, the polymarker may be used to construct a monoclonal antibody library containing a number of B-lymphocyte hybridoma clones that each recognize a single epitope of a marker molecule. The library can be used to produce a large quantity of antibodies against the polymarker.

In other aspects, a preparation of polymarker antibodies may be covalently or otherwise immobilized onto an affinity matrix, enabling additional immunoaffmity purification of target organism polymarker. The purification step may limit the number of molecules in the preparation and may increase the specificity of antibody preparations described herein.

In other aspects, the purified polymarker may be used for animal immunization.

In other aspects, a combined purified polymarker may be used for animal immunization.

In other aspects, the specificity of the preparation of polymarker antibodies is improved using subtraction techniques described herein prior to purification of the polymarker.

For illustration, the polymarker from K. pneumoniae ATCC^® 9997 may be purified for the purpose of immunizing animals, hi this case, the tagging, lysis, and binding to the streptavidin agarose beads would proceed as described above. After washing, the beads may be injected directly into host animals or the purified polymarker may be removed from the captured biotin tag by digesting the beads with dithiothreitol to cleave the disulfide bond in the linker region of the tagging molecule. The purified preparation may then be injected into a host animal, such as a chicken, in several doses over several months. Polyclonal antibodies collected from serum or egg yolks of the host animal will, after approximately 6 to 8 weeks, contain elicited antibodies against the purified polymarker preparation. The polyclonal antibodies may be used directly from the whole serum, the IgG or IgY fraction and may be enriched by a purification protocol such as protein L affinity columns, or may be purified using polymarker affinity column techniques described above.

For illustration, a combination of preparations of polymarker antibodies, representing all Klebsiella biotypes, may be covalently attached to an affinity matrix such as epoxy- activated sepahrose CL-6B by reaction through primary amines at physiologic pH. Once the antibodies are bound to the affinity matrix, unbound materials are washed out and the markers can be introduced to the matrix. The source of marker may be a combination of lysates from all or a single Klebsiella biotype, enriched for the surface protein marker by removal of insoluble debris, Triton X-100 detergent extraction, and size-exclusion chromatography on a sephadex G-50 column to remove small protein fragments and other debris. After binding, the unbound materials may be removed from the column using normal or stringent wash conditions, and then the bound proteins may be eluted by pH change. The eluted markers may be neutralized and exchanged into a physiologic buffer such as PBS before injection. Antibodies elicited by the purified polymarker provide increased specificity over those produced by injection of the polymarker as previously described.

Subtraction

In another embodiment, a collection of non-target organisms can be tested with a preparation of polymarker antibodies in order to determine target organism specificity. A preparation may undesirably detect a non-target organism. This type of reactivity is termed "cross-reactivity". The cross reactivity may be due to target and non target marker molecules having significant molecular homology and/or presence of at least one shared marker in target organism and non-target organisms. Specificity is a measure of the number of non- target organisms having detectable cross-reactivity with the preparation of polymarker antibodies.

In other aspects, a combination of a preparations of polymarker antibodies may be used for specificity testing. hi this embodiment, the collection should generally include a sufficient sampling of non-target organisms. The number of non-target organisms determines the predictive confidence of specificity and may be adjusted to achieve a desired statistical confidence level. Additionally, the number of organisms included in the collection depends on the specificity objectives and the desired statistical confidence level for the test. Additionally, the types of organisms which may be found in the sample, the corresponding predictive values, the sample type, geographic and/or epidemiological factors associated with the non-target and target organism, natural history, organism encapsulation, biofilm formation, nutrient depleted stasis, motility, antibiotic exposure, and other factors that could result in changes to organism phenotype are also taken into account when assembling the collection.

An example of a non-target organism collection designed for a Klebsiella clinical pneumonia assay included 5 Escherichia coli strains and isolates, 5 Enterobacter spp. strains and isolates, a Citrobacter freudii strain, 4 Pseudomonas aeruginosa strains and isolates, 3 Haemophilus influenzae strains and isolates, a Moraxella spp. strain, 2 Staphylococcus aureus strains, 2 Streptococcus pneumoniae strains, a coagulase-negative Staphylococcus, and a viridans Streptococcus. This collection of strains and isolates is designed to examine close relatives to Klebsiella, as well as other predominant bacterial types found in clinical pneumonia samples. The 25 non-target organism collection further provides 83% LLCI₉₅ value, providing fair confidence that any positive or negative test result will provide the same result on any target organism from the general population of bacteria, provided that the sample is prepared in a similar way, is of the same clinical sample type, is geographically and temporally similar, and is not otherwise biased relative to the non-target collection. A similar collection of 10 organisms would provide 65% LLCI₉₅ while a similar collection of 100 organisms would provide 95% confidence.

In another embodiment, the specificity of a preparation of polymarker antibodies is refined. Cross-reactivity can be subtracted by exposing the preparation to an immobilized marker, panmarker or polymarker from a non-target organism or group of non-target organisms. Cross-reactive antibodies bind to the non-target and are removed from the preparation of polymarker antibodies.

In another embodiment, the immobilized non-target marker may be a whole cell of the non-target organism. The live, fixed, or otherwise whole cell and cross-reactive antibody subsequently may be removed from the marker antibody solution preferably by centrifugation or filtration or other methods know by those skilled in the art.

In another embodiment, the remaining polymarker antibodies can have increased specificity for the target organism.

In another embodiment, the subtractive removal of cross-reactive antibodies maybe performed iteratively against several non-target members. hi another embodiment, a marker antibody preparation may have significant cross- reactivity with non target organism and may not be used in the preparation of polymarker antibodies.

For illustration, six polymarker antibodies for Klebsiella are individually tested against a panel of 24 gram negative non-target organisms, including several closely related members of the Enterobacteήaceae family: E. coli, Enterobacter cloacae, Enterobacter aerogenes, Citrobacter freudii, and Serratia marcescens; as well as several less related organisms: Pseudomonas aeruginosa, Acinetobacter baumannii, Stenotrophomonas maltophilia, and Haemophilus influenzae. This non-target collection represents a broad sampling of the most common organisms reported for the pneumonia applications requiring diagnostic Klebsiella detection. The five preparations of polymarker antibodies, using surface protein derived markers, exhibited some low-level reactivity against one or more members of the non-target collection. Subsequent adsorption against the non-target panmarker (whole cells) results in improved assay specificity with no change to the assay sensitivity. However, the preparation of polymarker antibodies directed against the K. pneumoniae ATCC 13883 capsular polysaccharides results in strong reactivity against a Enterobacter aerogenes strain. Adsorption against the Enterobacter aerogenes results in a complete loss of activity, indicating that a marker or group of markers from the target and non-target have substantially overlapping biotype. Thus, the diagnostic utility of the polymarker antibodies prepared from the capsular polysaccharide marker is reduced.

In another embodiment, the polymarker antibodies' cross-reactivity is removed performed prior to sensitivity testing against member organisms.

Further marker characterization

In another embodiment, the monoclonal library or marker antibody from immunized animals can be screened to identify molecules within the marker preparation. A minimum number of molecules from the marker may be identified to reliably detect the target organism.

In other aspects, marker antibody may be used to purify marker for subsequent characterization and determination a minimum number of molecules from the marker required to reliably detect the target organism.

In other aspects, monoclonal antibody may be used to purify the marker for the purpose of further molecular characterization and identification using methods known to those skilled in the art of unknown molecule detection.

In another embodiment, the purified marker from a said marker antibody immunoaffmity column may be further purified in order to further characterize the individual marker in the preparation. This purification may be accomplished by any number and combination of methods known to those skilled in the art of said purification processes. These methods include but are not limited to chromatographic size separation, 1-D or 2-D gel electrophoresis, ion exchange chromatography, reverse phase chromatography, or affinity chromatography. Characterization may be accomplished by any number and combination of methods known in the art, including spectrophotometry, such as ultraviolet spectroscopy, infrared spectroscopy, or atomic adsorption spectroscopy; mass determination, such as by MALDI-TOFMS or SDS-PAGE; protein sequencing, such as by Edman degradation; fragmentation, such as by proteolytic digest; charge characterization, such as by isoelectric focusing; and magnetic characterization, such as by nuclear magnetic resonance spectroscopy. Iteration of immunoaffmity purification using marker antibody followed by a traditional purification step can increase the efficiency of purification of a desired single marker.

Characterization or identification of individual molecules in the marker is not a requirement of the marker discovery process. Such characterization can be beneficial for the diagnosis and treatment of disease. Discovery of a surface protein molecule from antibiotic resistant Klebsiella may link the protein to resistance mechanisms in other strains and other types of bacteria and thus, provide a path to better antibiotic combination therapy. Furthermore, if the molecule is known to be related to a particular resistance phenotype, such as ESBL expression, the detection of this molecule could provide a rapid means to diagnosing a particular type of resistance without the requirement for culture and growth of bacteria in antibiotics to demonstrate resistance. An example of a previous discovered single molecule marker or monomarker is the mecA-associated modified penicillin binding protein PbP 2a in methicillin resistant Staphylococcus aureus samples. The detection of this protein has enabled the development of rapid MRSA screening assays that do not require culture.

Detection of organism phenotvpes

In another embodiment, the target organism may be a specific shared phenotype that supersedes classification at a species, genus, family, or super family classification, including but not limited to resistance mechanisms, enzymes, toxin production, and mutated proteins.

In other aspects, the non-target organisms could include members of a species that do not express the target phenotype. Marker discovery may be focused on the detecting the expression of a particular virulence phenotype within the species as in the case of enterohemmorhagic and enterotoxigenic E. coli (EHEC and ETEC). EHEC and ETEC have acquired toxic virulence factors from Shigella. The methods described herein can be used to generate polymarker antibodies for the EHEC and ETEC phenotype. Additionally, the methods described herein can be used to remove cross-reactive antibodies common to both the non-target E. coli and the target ETEC or EHEC. The refined marker antibody or polymarker antibody preparation would therefore be specific for the ETEC or EHEC phenotype. The preparations would demonstrate significant improvements over existing serotype methods, such as the antibody-based assay for E. coli O157:H7, since this particular serotype is only one of many serotypes that have acquired toxic virulence factors from Shigella, such as the 055, Ol 11, 026 and various less common but more varied serotypes.

In other aspects of the disclosure, detection of antibiotic resistance phenotypes, such as extended spectrum beta lactamase (ΕSBL) production, can be desirable. The ΕSBL mechanism of resistance is associated with a changes to a number of surface-expressed proteins and the cell wall. The resistance mechanism is shared by organisms within the Enterobacteriaceae family. The target is the organisms harboring the resistance mechanism and the non target organism are those organisms that do not have the resistance mechanism. The methods described herein are used to develop marker antibody for the ΕSBL resistance mechanisms. A plurality of marker antibody or polymarker antibody may be required to substantially detect the target organisms containing the resistance mechanism.

In other aspects of the disclosure, the resistance mechanism is induced to further assist the marker discovery process. Induction can be performed using many different methods known by those skilled in the art. In such cases, the non-target can be the same strain grown without induction.

EXAMPLES

The following examples describe in detail methods and systems for identifying polymarker, determining biotypes, and making polymarker antibody. It will be apparent to those skilled in the art that many modifications, both to materials and methods, may be practiced without departing from the scope of the disclosure.

Example I: Immuno Fluorescent Microscopy Assay

Live bacteria were immobilized on a microscope slide coated with poly-L-lysine and then brought into contact with a dilute solution of antibodies in a staining medium consisting of tryptic soy broth (TSB) + 5 mg/mL bovine serum albumin (BSA). Unbound antibodies are washed out of the sample using staining medium and bound antibodies are detected using a fluorescent secondary antibody conjugate specific for the antibody. For example, goat anti- chicken IgY conjugated with Alexa 555 was used when detecting chicken antibodies. Stained cells are visualized by IFM with phase-contrast and fluorescent images and electronically-captured. Image analysis software was used to assess fluorescent intensity and percentage of cell staining or stain efficiency. The detection criteria was determined by assay objectives. For example, the detection staining threshold could be a staining efficiency of at least 75% with intensity at least 2-fold higher than intensity of slide background. An average of 400 cells were analyzed in a single region of interest (ROI) for each sample.

Example II: Polymarker Identification, Biotyping, Combining Polymarker and Detecting Substantial Quantity of Target Organism Collection

A group of surface protein markers from a population of Acinetobacter spp. bacteria were purified and polymarker antibody was obtained. The polymarker antibodies were used to determine the number of biotypes present in the target organism collection. Seven biotypes were identified. The preparations of polymarker antibodies were collected and combined yielding. The combined preparation detected 95% of strains and isolates tested in an IFM assay. In certain embodiments, substantial quantity if polymarker antibodies can refer to a polymarker-antibodies that bind to at least 50% of a target, such as a class of organisms.

Acinetobacter spp. are a diverse group of loosely-related organisms that express few unique characteristics other than some respiratory functions that group them with other non- fermenting gram-negative bacteria. The genus is divided into several named species as well as at least 17 genomospecies (GSP) groups of which 3 are of clinical significance: GSP 2 {A. baumanniϊ), GSP 3, and GSP 13. Definitive identification of these organisms at the genus level is generally performed using growth-based phenotyping assays and do not readily differentiate the clinically relevant groups from others. Further identification at the species (or genomospecies) level employs nucleic acid-based assays. An accurate antibody-based assay that can be performed on live bacteria directly from clinical samples, potentially containing multiple types of bacteria, would represent a significant improvement over existing technologies.

A target organism collection of 20 Acinetobacter spp. strains and isolates was created using 7 strains, well know to those skilled in the art, from the American Type Culture Collection (ATCC^®) along with 13 recent pneumonia-related clinical Acinetobacter spp. isolates obtained from Jones Medical Institute (JMI). All strains and isolates were verified to be Acinetobacter spp. using BioMerieux API-20NE testing.

Each strain or isolate was streaked onto tryptic soy/sheep blood agar (TSA/B) plates and grown overnight at 35⁰C. A single colony was inoculated into tryptic soy broth (TSB) to produce a mid-log density culture. Bacteria from this initial culture were frozen in TSB + 15% glycerol aliquots for storage at -8O⁰C. For antibody IFM studies, a streak culture was inoculated from frozen stocks on TSA/B plates and grown overnight at 35⁰C. Several colonies were used to inoculate a liquid TSB culture which was grown for 2-3 hours to obtain a mid-log culture containing approximately 2 x 10⁸ CFU/mL.

Surface proteins from the type strain (ATCC^® 19606) were purified using a surface protein biotin tagging process. In brief, the surface proteins from approximately 10¹⁰ bacteria are labeled and separated from other antigen classes using a ImM non-permeable activated biotin, sulfo-NHS-SS-biotin (Pierce), to covalently tag only surface proteins having exposed primary amine groups. After a 30 minute exposure to the activated biotin, the reaction is quenched using a large excess of primary amine (100 mM Tris, pH 8.0) and the bacteria are washed to remove unbound biotin. The bacteria are then treated with non-ionic detergent (such as Triton X-100) and/or enzymes (such as lysozyme) and/or solvents (such as ethanol) so that the tagged proteins are liberated from their surrounding membrane or cell wall milieu and released into the solution. The biotin labeled proteins are polymarker for the ATCC® 19606 type target organism. The polymarker are then exposed to beads that have a covalently-attached biotin receptor, streptavidin, capturing the tagged proteins. Unbound materials from the lysate are removed with extensive washing with a neutral pH buffer containing detergent that maintains the solubility of the attached surface proteins.

The polymarker was immobilized on a solid surface by binding the biotin-tagged proteins to streptavidin-linked Sepharose CL-6B beads. The solid surface was transferred to a column and prior to use, the column was washed with binding, elution and storage buffers to remove any loosely-bound materials that could be released in any of the conditions used for purification of antibodies.

IgY antibodies were obtained from egg yolk. A crude egg yolk immunoglobulin (IgY) preparation was made from 24 commercially available chicken eggs. Yolks were removed from the whites, and the lipids and lipoproteins were removed by precipitation in a weak acid solution (10 mM sodium acetate at pH 5.2) at 4⁰C overnight. The remaining soluble proteins were further purified using tangential flow filtration (TFF) with a 100 kDa filter to concentrate and remove low molecular weight proteins and contaminants. The concentrated solution was neutralized by buffer exchange with phosphate-buffered saline (PBS) at pH 7.4 using the TFF system. This crude IgY preparation was then added to the affinity column to allow specific IgY antibodies (cell surface antibodies) to bind to the immobilized surface protein marker. After extensive washing with PBS, the antibodies were collected by eluting them from the affinity column by lowering the pH in the matrix using glycine buffer at pH 3.0. The eluted antibodies were collected from the column and neutralized using 100 mM Tris pH 8.0. This treatment effectively removed all bound antibodies from the affinity matrix, leaving the polymarker intact.

The polymarker antibodies were tested against each of the 20 strains and isolates in the target organism collection using the immunofluorescence microscopy (IFM) assay described above. Detection, in this example, required a 75% minimum percentage of cells staining with a minimum individual cell staining intensity compared to the surrounding background of at least 1.2-fold. An example of images obtained by IFM staining of A. baumannii ATCC^® 49139 using polymarker antibodies is shown in Figure 1. Strains and isolates detected with the polymarker antibodies were defined as belonging to biotype 1.

Other biotypes were identified by repeating the process described in the section above. Table 1 shows biotype determined by staining using preparations of polymarker IgY antibodies in IFM assay of live Acinetobacter spp. Strains 1, 2, 6, 8, 9, 17 and 19 were selected as representative strains for the 7 biotypes.

Table 1

Staining of some cells was observed with ATCC^® 19003, a strain that was not detected using its own polyantigen antibodies. Thus, the ATCC^® 19003 strain was recognized as its own biotype (Type 3) since none of the other antibodies detected the strain, nor did the its polymarker antibody detect any other stains or isolates. Non detection of ATCC^® 19003 may have been due to encapsulation which obscured the surface protein on these cells. Little overlapping staining of other biotypes was observed with each of the corresponding marker antibody, suggesting only non detectable levels of expressed surface protein sharing across the various biotypes.

The combined polymarker antibodies detected a substantial quantity (19/20) Acinetobacter spp. target organism collection as shown in table 1.

For comparison, polyclonal custom chicken antibodies were generated against a peptide monomarker. The peptide sequence was based on a predicted surface exposed outer membrane protein (omp) sequence. Peptide-specific polyclonal antibodies were obtained that recognized the protein, of the expected size, in a western blot assay of A. baumannii lysates. The polyclonal custom antibody stained 2 of 20 A. baumannii strains in the IFM assay. This may have been due to insufficient amounts of the specific peptide sequence available on the surface, hiding of the antigen or epitope within the protein or other structures, or differences in structure or folding of the native protein antigen verses the peptide on the non detected strains.

Table 2 contains the results of chicken anti-Acinetobacter omp peptide polyclonal antibody preparation against A. baumannii target collection. The highlighted strains were not detected by the omp peptide polyclonal antibody using the previously described IMF detection criteria.

Table 2

The chicken anti-Acinetobacter omp peptide polyclonal antibody preparation failed to detect a substantial quantity of the target organism collection.

EXAMPLE 3 Immunization with Polymarker

A combined polymarker preparation was used to immunize chickens in order to generate additionally antibodies directed against Klebsiella pneumoniae polymarker.

Polymarker identification and biotyping on a collection of 18 K. pneumoniae strains and isolates was performed as described above. Polymarker antibodies, from the 5 identified biotypes, were detected all 18 members of the target organism collection.

Surface protein polymarker from each of the 5 biotypes were prepared and immobilized on a solid bead surfaces as described previously. Beads from each of the 5 preparations were combined to produce a combined polymarker preparation. The combined preparation was washed extensively and then resuspended at a 50% slurry in sterile PBS for inj ection/immunization.

The following immunization protocol was followed: • Day 0 - inject a chicken with 0.3 mL of slurry in 0.3 mL complete Freund's adjuvant • Day 12 - boost with 0.3 mL of slurry in 0.3 mL incomplete Freund's adjuvant

• Day 21 - boost with 0.3 mL of slurry in 0.3 mL incomplete Freund's adjuvant

Eggs were collected from the chicken beginning on day 36 and continuing through day 71. Each week, eggs were received from the vendor, yolks were separated and pooled for the antibody preparation, and then frozen at -8O⁰C. The IgY preparations were tested against the target collection. Detection for the K. pneumoniae example assay was having at least 75% staining efficiency of 2-fold intensity over the slide background. The data from the week 5 preparation are presented in Table 3.

Table 3. IFM staining assay results using antibodies from a chicken host immunized with combined K. pneumoniae polymarker. Polymarker from strains in bold represented the 5 K. pneumoniae biotypes.

12 of the 18 strains (67%) were detected above threshold with the 5 week preparation. A commercial K. pneumoniae monoclonal antibody (United States Biological # K1891-05) was tested against the target collection. Binding of primary antibody was detected using a goat anti-mouse IgG (H+L) fluorescent secondary antibody, Alexa 546 conjugate (Nitrogen). This monoclonal antibody detected 2 of the 18 strains and isolates from the target organism collection.

Table 4. IFM sensitivity results for anti-K. pneumoniae monoclonal antibody, detected with goat anti-mouse IgG (H+L), Alexa 546 conjugate.

These results indicate that the individual epitope detected by this antibody is variably expressed or present among the cells of a single culture and many strains and isolates do not express the epitope at detectable levels.

The combined polymarker antibody preparation detected 12 of 18 K. pneumoniae strains/isolates while a commercial monoclonal antibody detected only 2 strains demonstrating the utility of polymarker antibodies for the detection of substantial quantities of target organisms in a target organism collection. Example 4 Subtraction of Cross Reactive Antibodies

Staphylococci share the bulk of their genetic complement across the entire genus increasing the difficulty of identifying S. aureus specific polymarker and the development of S. aureus species specific polymarker antibodies. Specificity improvements can be achieved by removal of cross reactive antibodies within the preparation of polymarker antibodies. The following example demonstrates the specificity improvements upon polymarker antibodies.

A target organism collection of 25 Staphylococcus aureus strains and isolates was created using 9 strains from the American Type Culture Collection (ATCC^®) along with 16 recent clinical S. aureus isolates obtained from JMI, CDC, and the University of Texas (UTX). Strains were verified to be S. aureus by gram-stain, morphology, and coagulase testing. Bacteria were cultured and maintained as described in Example 1.

A non-target organism collection of 7 coagulase-negative staphylococci (CNS) strains were obtained from ATCC^®. All strains were verified to be CNS by gram-stain, morphology, and coagulase testing.

A group of surface protein markers were prepared from the type strain ATCC^® 12600 as described above. Polymarker for ATCC® 12600 was identified and corresponding polymarker antibodies were prepared by the methods described previously.

The preparation of polymarker antibodies were tested against the 25 target strain collection using the IFM assay. In this example, detection occurred when at least 85% cell staining efficiency of at least 2-fold intensity over background was obtained. Test results are presented in Table 5.

Cultures for S. haemolyticus, S. epidermidis, and S. saprophyti s were grown to mid- log density, washed with PBS and fixed using 2% formaldehyde in PBS for 15 minutes. Residual fixing buffer was removed by washing in PBS, after which, the cells were added to the preparation of polymaker antibodies. Following 1 hour of incubation at 2O⁰C, the bacteria were removed by centrifugation and 0.2 μm filtration. The adsorption process was repeated two more times with the fixed S. haemolyticus cells. The resulting adsorbed preparation of polymarker antibodies was tested again against the target and non-target collections by IFM assay. The cross-reactivity detection threshold was defined as having no more than 10% staining efficiency of detectable intensity. The results are also presented in table 6.

Table 5. Detection of the S. aureus target collection using the 12600 polymarker antibodies.

Table 6. Cross reactivity of the 12600 polymarker antibodies against CNS before and after adsorption 3 times against S. haemolyticus and 1 time each against S. epidβrmidis and S. saprophyticus.

All S. aureus strains and isolates as well as all CNS strains were detected by the preparation of polymarker antibodies. The lack of specificity for S. aureus demonstrated the wide expression across the genus of at least one marker within the identified polymarker.

The preparation of polymarker antibodies detection of S. aureus was reduced to approximately 50% after adsorption, indicating that the S. aureus ATCC^® 12600 polymaker has several surface protein marker or group of markers that are common to approximately half of all S. aureus. This result also suggests that other S. aureus biotypes exist expressing different S. aureus specific markers.

The adsorbed polymarker antibodies cross-reactivity with S. epidermidis and S. saprophytics was reduced below thresholds, but S. haemolyticus activity was virtually unchanged after 1 round of adsorption (95% stained at 9.0-fold). Iterative adsorption of the polymarker antibodies with 2 additional rounds against S. haemolyticus was successful in reducing the activity to slightly above the detection level, as shown in table 5. Adsorption to these three species resulted in moderate to complete reduction of cross-reactivity against other members of the non-target organism panel. Cross-reactivity against S. lugdunensis and S. hominis was not eliminated, indicating the uniqueness of these species with respect to the adsorbed cells.

The use of a single surface protein preparation of polymarker antibodies for identification of S. aureus initially showed 100% functional sensitivity by IFM assay. This reagent also demonstrated low specificity, staining all CNS strains tested. A process of removing cross-reactive antibodies by adsorbing against whole fixed CNS cells improved the specificity of the reagent with respect to CNS. However, this process simultaneously reduced the detection of approximately 50% of S. aureus strains and isolates, while leaving binding ability essentially intact on the remaining 50%. This is indicative that the remaining surface proteins that were specific to S. aureus ATCC^® 12600 were shared by some, but not all, strains and isolates. This is a similar result to biotyping processes for Acinetobacter spp. and K. pneumoniae in which a minority of strains and isolates were stained with the prototype antibody. However, in the case of S. aureus, the biotyping process was initially hindered by a large amount of highly related surface protein marker shared among all Staphylococci. Biotypes for the polymarker on ATCC^® 12600, were identified after removing the antibodies responsible for this cross-reactivity.

Claims

We claim

1. A method of preparing a first polyantigen antibody preparation that bind the cell surface of a cell type comprising:

a) immobilizing a first plurality of markers from a collection of target organisms of said cell type on a solid surface to form an immobilized plurality of markers;

b) contacting said immobilized plurality of markers with a plurality of antibodies to immobilize at least a portion of said plurality of immobilized antibodies on said solid surface; and

c) collecting said portion of said plurality of immobilized antibodies from said solid surface to produce said first polyantigen antibody preparation.

2. The method of claim 1 , comprising contacting the polyantigen antibody preparation to a second plurality of markers from the collection of target organisms to prepare a second polyantigen antibody preparation with a biotype common to said collection of target organisms.

3. The method of one of claims 1-2, comprising lysing said collection of target organisms to obtain said first plurality of markers.

4. The method of one of claims 1-3, wherein the first plurality of markers comprises a substantial quantity of cell surface antigens of said cell type.

5. The method of claim 4, wherein the immobilizing step comprises immobilizing said substantial quantity of cell surface antigens on said solid surface.

6. A method of detecting a cell in said cell type comprising:

a) preparing a first polyantigen antibody preparation according to the method of one of claims 1-5;

b) providing a polymarker antibody prepared to a composition; and

c) detecting the polymarker antibody to determine the presence of the cell.

7. The method of claim 1, further comprising: a) determining the affinity and/or avidity of the polymarker antibodies for a second plurality of target organisms; and

b) obtaining antibodies that bind to said plurality of target microorganisms to prepare a second polyantigen preparation with a biotype common to said second plurality of target microoganisms.

8. The method of one of claims 1-5, further comprising repeating the steps of claim 1.

9. The method of claim 1, wherein a host animal is immunized with said target organism polymarker to generate polymarker antibodies.

10. The method of claim 9, wherein said polymarker is purified using polymarker antibodies prior to immunization.

11. The method of one of claims 1 -5, further comprising:

a) contacting said preparation of polymarker antibodies with a plurality of cells of said cell type;

b) collecting a first group of antibodies in said preparation of antibodies that do not bind said cell type; and

c) removing a second group of antibodies in said preparation of antibodies that bind at least one of said plurality of cells of said cell type.

12. The method of claim 1-5, further comprising purifying a class of markers from said first plurality of markers before said immobilizing step.

13. The method of claim 12, wherein said class of markers is selected from the group consisting of cell surface polysaccharides, cell surface polysaccharide fragments, cell surface proteins, and cell surface protein fragments.

14. The method of claim 13, wherein said antigen class is cell surface polysaccharides or cell surface polysaccharide fragments.

15. The method of claim 14, wherein said cell surface polysaccharides or cell surface polysaccharide fragments are immobilized on said solid surface by chemical conjugation.

16. The method of one of claims 1 -5, wherein said plurality of antibodies is obtained from a source selected from the group consisting of blood, blood serum and egg yolk.

17. The method of one of claims 1-5, wherein said solid surface is the stationary phase of a chromatography column.

18. The method of claim 17, wherein said plurality of markers are tagged with a tag capable of binding said stationary phase.

19. The method one of claims 1-5, wherein said cell type is a pathogenic microorganism.

20. A method of one of claims 1-5, further comprising labeling said antibodies.

21. The method of claim 20, wherein said label is selected from the group consisting of a fluorescent label and a chemiluminescent label.

22. The method of one of claims 1 -5, further comprising:

e) contacting said preparation of polyantigen antibodies with a plurality of cells of said cell type;

f) removing a first group of antibodies in said preparation of antibodies that do not bind said cell type; and

g) collecting a second group of antibodies in said preparation of antibodies that bind at least one of said plurality of cells of said cell type.

23. A method of refining a preparation of polyantigen antibodies that bind the cell surface of a cell type comprising:

a) preparing a first polyantigen antibody preparation according to one of claims 1-5;

b) preparing a second polyantigen antibody preparation according to one of claims 1-

5;

c) combining the first polyantigen antibody preparation and the second polyantigen antibody preparation.

24. A method of selecting a polyantigen antibody preparation that bind the cell surface of a first cell type and do not bind the cell surface of a second cell type, comprising:

a) selecting a first polyantigen antibody preparation that bind the first cell type according to the method of claim 1 ;

b) contacting said first polyantigen antibody preparation with a surface immobilized cell membrane antigens of said second cell type; and

c) collecting the subset of antibodies do not bind said surface immobilized antigens of said second cell type.

25. A method for measuring the growth of a plurality of individual microorganisms of a cell type in a sample comprising: contacting said sample with a biosensor comprising: a concentration module; and at least one detection surface; and concentrating said individual microorganisms onto said detection surface, wherein each said individual microorganism binds to said detection surface in spatially discrete sites; allowing said individual microorganisms to grow for a first period of time; contacting said individual microorganisms with a preparation of polyantigen antibodies specific to the cell surface of a cell type; and detecting said one or more of said polyantigen antibodies to detect the growth of individual microorganisms as an indication of the presence of said microorganisms of said cell type.