Anti-ADDL monoclonal antibody and use thereof

Abstract

The present invention relates to antibodies that differentially recognize multi-dimensional conformations of Aβ-derived diffusible ligands, also known as ADDLs. The antibodies of the invention can distinguish between Alzheimer's Disease and control human brain extracts and are useful in methods of detecting ADDLs and diagnosing Alzheimer's Disease. The present antibodies also block binding of ADDLs to neurons, assembly of ADDLS, and tau phosphorylation and are there useful in methods for the preventing and treating diseases associated with soluble oligomers of amyloid β 1-42.

Description

BACKGROUND OF THE INVENTION

Alzheimer's Disease is a progressive and degenerative dementia (Terry, et al. (1991) Ann. Neurol. 30:572-580; Coyle (1987) In: Encyclopedia of Neuroscience, Adelman (ed.), Birkhäuser, Boston-Basel-Stuttgart, pp 29-31,). In its early stages, Alzheimer's Disease manifests primarily as a profound inability to form new memories (Selkoe (2002) Science 298:789-791), reportedly due to neurotoxins derived from amyloid beta (Aβ). Aβ is an amphipathic peptide whose abundance is increased by mutations and risk factors linked to Alzheimer's Disease. Fibrils formed from Aβ constitute the core of amyloid plaques, which are hallmarks of an Alzheimer's Disease brain. Analogous fibrils generated in vitro are lethal to cultured brain neurons. These findings indicate that memory loss is a consequence of neuron death caused by fibrillar Aβ.

Despite strong experimental support for fibrillar Aβ and memory loss, a poor correlation exists between dementia and amyloid plaque burden (Katzman (1988) Ann. Neurol. 23:138-144). Moreover, transgenic hAPP mice (Dodart, et al. (2002) Nat. Neurosci. 5:452-457; Kotilinek, et al. (2002) J. Neurosci. 22:6331-6335), which develop age-dependent amyloid plaques and, most importantly, age-dependent memory dysfunction, show that within 24 hours of vaccination with monoclonal antibodies against Aβ memory loss can be reversed with no change in plaque levels. Such findings are not consistent with a mechanism for memory loss dependent on neuron death caused by amyloid fibrils.

Additional neurologically active molecules formed by Aβ self-assembly have been suggested. These molecules include soluble Aβ oligomers, also referred to as Aβ-derived diffusible ligands or ADDLs. Oligomers are metastable and form at low concentrations of Aβ1-42 (Lambert, et al. (1998) Proc. Natl. Acad. Sci. USA 95:6448-6453). Aβ oligomers rapidly inhibit long-term potentiation (LTP), a classic experimental paradigm for memory and synaptic plasticity. As such, memory loss stems from synapse failure, prior to neuron death and synapse failure by Aβ oligomers, not fibrils (Hardy & Selkoe (2002) Science 297:353-356). Soluble oligomers have been found in brain tissue and are strikingly elevated in Alzheimer's Disease (Kayed, et al. (2003) Science 300:486-489; Gong, et al. (2003) Proc. Natl. Acad. Sci. USA 100:10417-10422) and in hAPP transgenic mice Alzheimer's Disease models (Kotilinek, et al. (2002) J. Neurosci. 22:6331-6335; Chang, et al. (2003) J. Mol. Neurosci. 20:305-313).

A variety of Alzheimer's Disease treatment options have been suggested. Vaccine clinical trials have revealed that persons mounting a vigorous immune response to the vaccine exhibit cognitive benefit (Hock, et al. (2003) Neuron 38:547-554); however, frequency of CNS inflammation caused early termination of part of the trial (Birmingham & Frantz (2002) Nat. Med. 8:199-200). As an alternative to a vaccine, therapeutic antibodies that target ADDLs without binding monomers or fibrils have been suggested (Klein (2002) Neurochem. Int. 41:345-352). ADDLs are highly antigenic, generating oligomer-selective polyclonal antibodies in rabbits at concentration of ˜50 μg/mL (Lambert, et al. (2001) J. Neurochem. 79:595-605). Results from transgenic mice models also suggest that antibodies can be successful in reversing memory decline (Dodart, et al. (2002) Nat. Neurosci. 5:452-457; U.S. patent application Ser. No. 11/194,989). Accordingly, there is a need in the art for ADDL-selective therapeutic antibodies for the prevention and treatment of Alzheimer's Disease. The present invention meets this need.

SUMMARY OF THE INVENTION

The present invention is an isolated antibody, or fragment thereof, capable of differentially recognizing a multi-dimensional conformation of one or more Aβ-derived diffusible ligands. In particular, the antibody of the instant invention has a complementary determining region (CDR) of Arg-Xaa₁-Leu-Xaa₂-Xaa₃-Xaa₄-Xaa₅-Xaa₆-Asp-Ala-Met-Asp-Tyr (SEQ ID NO:9), wherein Xaa₁ is Gln or Ala; Xaa₂ is Ser or Gly; Xaa₃ is Pro, Ala, Lys, Arg, or Thr; Xaa₄ is Lys or Arg; Xaa₅ is Gly, Ser, or Lys; Xaa₆ is Val, Thr, Ile or Arg. In particular embodiments, the antibody of the present invention is in admixture with a pharmaceutically acceptable carrier. In other embodiments, the antibody of the present invention is in a kit. Still other embodiments embrace an antibody having heavy and light chain variable region sequences as set forth in SEQ ID NO:108 and SEQ ID NO:112. An antibody having heavy and light chain sequences as set forth in SEQ ID NO:138 and SEQ ID NO:140 is also provided.

Methods for preventing binding of Aβ-derived diffusible ligands to a neuron and inhibiting assembly of Aβ-derived diffusible ligands employing an antibody or antibody fragment which binds a multi-dimensional conformation of one or more Aβ-derived diffusible ligands are also provided.

The present invention further embraces a method for prophylactically or therapeutically treating a disease associated with Aβ-derived diffusible ligands using an antibody of the instant invention. Administration of an antibody of the invention can prevent binding of Aβ-derived diffusible ligands to a neuron thereby preventing or treating the disease associated with Aβ-derived diffusible ligands.

The present invention is also a method for identifying a therapeutic agent that prevents the binding of Aβ-derived diffusible ligands to a neuron. This method of the invention involves contacting a neuron with Aβ-derived diffusible ligands in the presence of an agent and using an antibody of the present invention to determine binding of the Aβ-derived diffusible ligands to the neuron in the presence of the agent.

The present invention also embraces a method for detecting Aβ-derived diffusible ligands in a sample and a method for diagnosing a disease associated with Aβ-derived diffusible ligands. Such methods involve contacting a sample with an antibody of the instant invention so that the Aβ-derived diffusible ligands can be detected and a disease associated with Aβ-derived diffusible ligands can be diagnosed.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 shows the nucleic acid sequences for the heavy (FIG. 1A) and light (FIG. 1B) chain variable regions for murine anti-ADDL antibody 20C2. Lower case letters indicate the antibody leader sequences and uppercase letters indicate antibody variable region sequences. The nucleotides coding for the complementary determining regions (CDRs) are underlined.

FIG. 2 shows comparisons of heavy (FIG. 2A) and light (FIG. 2B) chain variable region amino acid sequences of murine antibody 20C2 and humanized antibodies, Hu20C2 (CDR grafted) and Hu20C2A3 (veneered). Sequences are presented as comparisons between the 20C2 mouse sequence, the most homologous human genomic sequence and the humanized sequences. Sequence differences in the frame regions between murine 20C2 and humanized Hu20C2A3 are in bold. Sequence differences in the underlined CDR regions between humanized Hu20C2A3 and murine 20C2 are in bold and indicated with an *. CDRs are underlined.

FIG. 3 shows nucleic acid sequences for the heavy (FIGS. 3A and 3B) and light (FIG. 3C) chain variable regions (HCVRs and LCVRs, respectively) for humanized anti-ADDL antibody Hu20C2 (CDR grafted). Two humanized versions of the Hu20C2 heavy chain were generated (HCVRA and HCVRB) that differ by one amino acid at position 24. In Hu20C2 HCVRA the human amino acid was used and in Hu20C2 HCVRB the mouse amino acid was used. Variable region sequences were cloned into full heavy and light chain antibody expression vectors.

FIG. 4 shows the annotated amino acid sequences and nucleotide sequences of Hu20C2 humanized antibody in Fab phage-display vector pFab4. Amino acid sequence for heavy chain version A (FIG. 4A), heavy chain version B (FIG. 4B), and the light chain (FIG. 4C) of Hu20C2 humanized antibody in Fab phage-display vector pFab4 are in italic and underlined regions are as indicated. Nucleotide sequence of heavy chain version A fused with the light chain of Hu20C2 in pFab4 vector is shown in FIG. 4D-4E with sequences encoding the Hu20C2 antibody sequences shown in lowercase.

FIG. 5 depicts the design and primers employed in preparing two light chain CDR3 libraries, namely LC3-1 and LC3-2 (FIG. 5A), and three heavy chain CDR3 libraries, namely 20C2B-39HC₃-1, 20C2B-39HC₃-2, and 20C2B-39HC₃-3 (FIG. 5B), for respectively generating affinity matured Hu20C2 light and heavy chain CDR3s. Restriction endonuclease recognition sites used for cloning are indicated in italic. Uppercase indicates nucleic acids encoding antibody variable region sequences. Nucleic acids encoding CDRs are underlined. Biotin-labeled primers are indicated.

FIG. 6 shows a comparison of the amino acid sequence of human antibody constant regions and the sequence of IgG2m4. The asterisk indicates a glycosylation site at Asn297. Regions of FcRn binding are indicated. Sequences in which IgG2m4 is different from IgG2 are underlined.

FIG. 7 shows the amino acid (FIGS. 7A and 7C) and nucleotide (FIGS. 7B and 7D) sequences for the full IgG2m4 humanized heavy chain (FIGS. 7A and 7B) and humanized Kappa light chain (FIGS. 7C and 7D) for anti-ADDL antibody Hu20C2A3. Underlining indicates variable region sequences. The remaining sequences are constant region sequences.

FIG. 8 shows interactions between Aβ40 monomer or ADDLs with Hu20C2A3 produced by two different systems, CHO (FIG. 8A) or Pichia (FIG. 8B), as determined by ELISA.

FIG. 9 shows Hu20C2A3 inhibition of bADDL binding to primary hippocampal neurons.

FIG. 10 shows fluorescent thermal melt analysis of Hu20C2A3.

FIG. 11 shows plasma Aβx-40 levels (pM) of APP-YAC mice following intravenous injection of Hu20C2A3, irrelevant control or vehicle. Hu20C2A3 was prepared by stable transfection of CHO cells or Pichia. Aβx-40 was determined using a 4G8/G2-10 ELISA. ***, p<0.001 by Tukey-Kramer HSD post-hoc testing. Error bars=SEM; N=6/group.

DETAILED DESCRIPTION OF THE INVENTION

Monoclonal antibodies, which differentially recognize multi-dimensional conformations of Aβ-derived diffusible ligands (i.e., ADDLs), have now been generated. Antibodies of this invention are derived from the murine monoclonal antibody 20C2. Murine 20C2 is known in the art for exhibiting the following characteristics. Murine 20C2 is an IgG1 antibody which binds to both synthetic and endogenous ADDLs bound to cultured hippocampal cells. Furthermore, this antibody can block both endogenous and synthetic ADDL binding to cultured cells, abate the binding of biotinylated ADDLs (bADDLs) to neurons, and prevent tau phosphorylation. The core linear epitope for 20C2 is Glu-Phe-Arg-His-Asp-Ser (SEQ ID NO:1), corresponding to amino acid residues 3-8 of Aβ1-42, with a conformational epitope that is dependent upon elements from within residues 17-42 of Aβ, but only when assembled.

The instant antibodies are humanized and, in some embodiments affinity-matured derivatives of murine 20C2. Like the murine 20C2 antibody, the antibodies disclosed herein exhibit a high degree of selectivity for multi-dimensional conformations of ADDLs, with minimal detection of monomer Aβ peptides. Advantageously, the instant antibodies identify endogenous oligomers in Alzheimer's Disease brain slices and inhibit binding of bADDLs to neurons. Moreover, the instant antibodies provide a significant and robust increase in plasma Aβx-40 levels, an increase in which is known to be associated with an ultimate lowering of brain Aβ. Accordingly, the antibodies of this invention find use in the prevention of ADDL binding to neurons and assembly of ADDLs and the treatment of ADDL-related diseases including Alzheimer's Disease.

Accordingly, the present invention is an isolated antibody that differentially recognizes one or more multi-dimensional conformations of ADDLs. An antibody of the instant invention is said to be isolated when it is present in the substantial absence of other biological macromolecules of the same type. Thus, an “isolated antibody” refers to an antibody which is substantially free of other antibodies; however, the molecule may include some additional agents or moieties which do not deleteriously affect the basic characteristics of the antibody (e.g., binding specificity, neutralizing activity, etc.).

An antibody which is capable of specifically binding one or more multi-dimensional conformations of ADDLs, binds particular ADDLs derived from the oligomerization of Aβ1-42, but like murine 20C2 does not cross-react with other Aβ peptides, namely Aβ1-12, Aβ1-28, Aβ1-40, and Aβ12-28 as determined by western blot analyses; and preferentially bind ADDLs in solution. Specific binding between two entities generally refers to an affinity of at least 10⁶, 10⁷, 10⁸, 10⁹, or 10¹⁰ M⁻¹. Affinities greater than 10⁸ M⁻¹ are desired to achieve specific binding.

In particular embodiments, an antibody that is capable of specifically binding a multi-dimensional conformation of one or more ADDLs is also raised against (i.e., an animal is immunized with) multi-dimensional conformations of ADDLs. In other embodiments, an antibody that is capable of specifically binding a multi-dimensional conformation of one or more ADDLs is raised against a low n-mer-forming peptide such as Aβ1-42[Nle35-Dpro37].

The term “epitope” refers to a site on an antigen to which B and/or T cells respond or a site on a molecule against which an antibody will be produced and/or to which an antibody will bind. For example, an epitope can be recognized by an antibody defining the epitope.

A linear epitope is an epitope wherein an amino acid primary sequence comprises the epitope recognized. A linear epitope typically includes at least 3, and more usually, at least 5, for example, about 6 to about 10 amino acids in a unique sequence.

A conformational epitope, in contrast to a linear epitope, is an epitope wherein the primary sequence of the amino acids comprising the epitope is not the sole defining component of the epitope recognized (e.g., an epitope wherein the primary sequence of amino acids is not necessarily recognized by the antibody defining the epitope). Typically a conformational epitope encompasses an increased number of amino acids relative to a linear epitope. With regard to recognition of conformational epitopes, the antibody recognizes a three-dimensional structure of the peptide or protein. For example, when a protein molecule folds to form a three-dimensional structure, certain amino acids and/or the polypeptide backbone forming the conformational epitope become juxtaposed enabling the antibody to recognize the epitope. Methods of determining conformation of epitopes include but are not limited to, for example, x-ray crystallography, two-dimensional nuclear magnetic resonance spectroscopy and site-directed spin labeling and electron paramagnetic resonance spectroscopy. See, for example, Epitope Mapping Protocols in Methods in Molecular Biology (1996) Vol. 66, Morris (Ed.).

Aβ-derived diffusible ligands or ADDLs refer to soluble oligomers of amyloid β1-42 which are desirably composed of aggregates of less than eight or nine amyloid β1-42 peptides and are found associated with Alzheimer's Disease. This is in contrast to high molecular weight aggregation intermediates, which form stings of micelles leading to fibril formation.

As exemplified herein, the instant antibody binds or recognizes at least one multi-dimensional conformation of an ADDL. In particular embodiments, the instant antibody binds at least two, at least three, or at least four multi-dimensional conformations of an ADDL. Multi-dimensional conformations of ADDLs are intended to encompass dimers, trimers, tetramers pentamers, hexamers, heptamers, octamers, nonamers, decamers, etc. as defined by analysis via SDS-PAGE. Because trimer, tetramer, etc. designations can vary with the assay method employed (see, e.g., Bitan, et al. (2005) Amyloid 12:88-95) the definition of trimer, tetramer, and the like, as used herein, is according to SDS-PAGE analysis. As such, the antibody of the instant invention has oligomer-specific characteristics. In particular embodiments, a multi-dimensional conformation of an ADDL is associated with a specific polypeptide structure which results in a conformational epitope that is recognized by an antibody of the present invention. In other embodiments, an antibody of the invention specifically binds a multi-dimensional conformation ADDL having a size range of approximately a trimer or tetramer, which have molecular weights in excess of >50 kDa.

In certain embodiments, in addition to binding to a multi-dimensional conformation, the instant antibody binds to a selected linear epitope of amyloid β1-42. A linear epitope of an ADDLs is intended as a four, five, six or more amino acid residue peptide located in the N-terminal 10, 11, 12, 15 or 20 amino acid residues of amyloid β1-42. In particular embodiments, an antibody of the invention specifically binds to a linear epitope within residues 1-10, 1-8, 3-10, or 3-8 of amyloid β1-42. An exemplary linear epitope of amyloid β1-42 which is bound by a humanized antibody of the invention is amino acid residues Glu-Phe-Arg-His-Asp-Ser (SEQ ID NO:1).

While antibodies of the instant invention may have similar linear epitopes, such linear epitopes are not wholly indicative of the binding characteristics of the instant antibodies (i.e., ability to block ADDL binding to neurons, prevent tau phosphorylation and inhibit ADDL assembly) because, as is well-known to the skilled artisan, the linear epitope may only correspond to a portion of the antigen's epitope (see, e.g., Breitling and Dübel (1999) In: Recombinant Antibodies, John Wiley & Sons, Inc., NY, pg. 115). For example, murine 20C2 is known to bind assemblies of charge-inverted, truncated Aβ7-42 peptide, which lack the linear epitope for 20C2 (i.e., amino acid residues 3-8) and contain a very different sequence corresponding to residues 7-16 of Aβ. Therefore, 20C2, as well as humanized derivatives thereof, bind to conformational epitopes that depend upon elements from within residues 17-42 of Aβ, but only when in a multidimensional conformation. The antibody of the instant invention can be distinguished from those of the art as being capable of differentially recognizing multi-dimensional ADDLs and accordingly differentially blocking ADDL binding to neurons, differentially preventing tau phosphorylation and differentially inhibiting ADDL assembly.

An antibody, as used in accordance with the instant invention includes, but is not be limited to, monoclonal antibodies, and chimeric, human (e.g. isolated from B cells), humanized, neutralizing, bispecific or single chain antibodies thereof. In one embodiment, an antibody of the instant invention is monoclonal. For the production of antibodies, various hosts including goats, rabbits, chickens, rats, mice, humans, and others, can be immunized by injection with synthetic or natural ADDLs. Methods for producing antibodies are well-known in the art. See, e.g., Kohler and Milstein ((1975) Nature 256:495-497) and Harlow and Lane (Antibodies: A Laboratory Manual (Cold Spring Harbor Laboratory, New York (1988)).

Depending on the host species, various adjuvants can be used to increase the immunological response. Adjuvants used in accordance with the instant invention desirably augment the intrinsic response to ADDLs without causing conformational changes in the immunogen that affect the qualitative form of the response. Particularly suitable adjuvants include 3 De-O-acylated monophosphoryl lipid A (MPL™; RIBI ImmunoChem Research Inc., Hamilton, Mont.; see GB 2220211) and oil-in-water emulsions, such as squalene or peanut oil, optionally in combination with immune stimulants, such as monophosphoryl lipid A (see Stoute, et al. (1997) N. Engl. J. Med. 336:86-91), muramyl peptides (e.g., N-acetylmuramyl-L-threonyl-D-isoglutamine (thr-MDP), N-acetyl-normuramyl-L-alanyl-D-isoglutamine (nor-MDP), N-acetylmuramyl-L-alanyl-D-isoglutaminyl-L-alanine-2-(1′-2′dipalmitoyl-sn-glycero-3-hydroxyphosphoryloxy)-ethylamine (E-PE), N-acetylglucsaminyl-N-acetylmuramyl-L-Al-D-isoglu-L-Ala-dipalmitoxy propylamide (DTP-DPP)), or other bacterial cell wall components. Specific examples of oil-in-water emulsions include MF59 (WO 90/14837), containing 5% Squalene, 0.5% TWEENT™ 80, and 0.5% SPAN 85 (optionally containing various amounts of MTP-PE) formulated into submicron particles using a microfluidizer such as Model 110Y microfluidizer (Microfluidics, Newton, Mass.); SAF containing 10% Squalene, 0.4% TWEEN™ 80, 5% PLURONIC®-blocked polymer L121, and thr-MDP, either microfluidized into a submicron emulsion or vortexed to generate a larger particle size emulsion; and RIBI™ adjuvant system (RAS) (Ribi ImmunoChem, Hamilton, Mont.) containing 2% squalene, 0.2% TWEEN™ 80, and one or more bacterial cell wall components such as monophosphoryllipid A, trehalose dimycolate (TDM), and cell wall skeleton (CWS).

Another class of adjuvants is saponin adjuvants, including ISCOMs (immunostimulating complexes) and ISCOMATRIX® (CSL Ltd., Parkville, Australia). Other suitable adjuvants include Complete Freund's Adjuvant (CFA), Incomplete Freund's Adjuvant (IFA), mineral gels such as aluminum hydroxide, and surface-active substances such as lysolecithin, PLURONIC® polyols, polyanions, peptides, CpG (WO 98/40100), keyhole limpet hemocyanin, dinitrophenol, and cytokines such as interleukins (IL-1, IL-2, and IL-12), macrophage colony stimulating factor (M-CSF), and tumor necrosis factor (TNF). Among adjuvants used in humans, BCG (bacilli Calmette-Guerin) and Corynebacterium parvum are particularly suitable.

An antibody to a multi-dimensional conformation ADDL is generated by immunizing an animal with ADDLs. Generally, ADDLs can be generated synthetically or by recombinant fragment expression and purification. Synthetic ADDLs can be prepared as disclosed herein or in accordance with the methods disclosed in U.S. Pat. No. 6,218,506 or in co-pending applications U.S. Ser. Nos. 60/621,776, 60/652,538, 60/695,528 and 60/695,526. Further, ADDLs can be fused with another protein such as keyhole limpet hemocyanin to generate an antibody against the chimeric molecule. The ADDLs can be conformationally constrained to form an epitope useful as described herein and furthermore can be associated with a surface for example, physically attached or chemically bonded to a surface in such a manner so as to allow for the production of a conformation which is recognized by the antibodies of the present invention.

Monoclonal antibodies to multi-dimensional conformations of ADDLs can be prepared using any technique which provides for the production of antibody molecules by continuous cell lines in culture. These include, but are not limited to, the hybridoma technique, the human B-cell hybridoma technique, and the EBV-hybridoma technique (Kohler, et al. (1975) Nature 256:495-497; Kozbor, et al. (1985) J. Immunol. Methods 81:31-42; Cote, et al. (1983) Proc. Natl. Acad. Sci. 80:2026-2030; Cole, et al. (1984) Mol. Cell Biol. 62:109-120).

In particular embodiments, the instant antibodies are humanized. Humanized or chimeric antibodies can be produced by splicing of mouse antibody genes to human antibody genes to obtain a molecule with appropriate antigen specificity and biological activity (see Morrison, et al. (1984) Proc. Natl. Acad. Sci. 81, 6851-6855; Neuberger, et al. (1984) Nature 312:604-608; Takeda, et al. (1985) Nature 314:452-454; Queen, et al. (1989) Proc. Natl. Acad. Sci. USA 86:10029-10033; WO 90/07861). For example, a mouse antibody is expressed as the Fv or Fab fragment in a phage selection vector. The gene for the light chain (and in a parallel experiment, the gene for the heavy chain) is exchanged for a library of human antibody genes. Phage antibodies, which still bind the antigen, are then identified. This method, commonly known as chain shuffling, provided humanized antibodies that should bind the same epitope as the mouse antibody from which it descends (Jespers, et al. (1994) Biotechnology NY 12:899-903). As an alternative, chain shuffling can be performed at the protein level (see, Figini, et al. (1994) J. Mol. Biol. 239:68-78).

Human antibodies can also be obtained using phage-display methods. See, e.g., WO 91/17271 and WO 92/01047. In these methods, libraries of phage are produced in which members display different antibodies on their outer surfaces. Antibodies are usually displayed as Fv or Fab fragments. Phage displaying antibodies with a desired specificity are selected by affinity enrichment to ADDLs. Human antibodies against ADDLs can also be produced from non-human transgenic mammals having transgenes encoding at least a segment of the human immunoglobulin locus and an inactivated endogenous immunoglobulin locus. See, e.g., WO 93/12227 and WO 91/10741. Human antibodies can be selected by competitive binding experiments, or otherwise, to have the same epitope specificity as a particular mouse antibody. Such antibodies generally retain the useful functional properties of the mouse antibodies. Human polyclonal antibodies can also be provided in the form of serum from humans immunized with an immunogenic agent. Optionally, such polyclonal antibodies can be concentrated by affinity purification using ADDLs as an affinity reagent.

As exemplified herein, humanized antibodies can also be produced by veneering or resurfacing of murine antibodies. Veneering involves replacing only the surface fixed region amino acids in the mouse heavy and light variable regions with those of a homologous human antibody sequence. Replacing mouse surface amino acids with human residues in the same position from a homologous human sequence has been shown to reduce the immunogenicity of the mouse antibody while preserving its ligand binding. The replacement of exterior residues generally has little, or no, effect on the interior domains, or on the interdomain contacts. (See, e.g., U.S. Pat. No. 6,797,492).

Human or humanized antibodies can be designed to have IgG, IgD, IgA, IgM or IgE constant regions, and any isotype, including IgG1, IgG2, IgG3 and IgG4. In particular embodiments, an antibody of the invention is IgG or IgM, or a combination thereof. Other embodiments of the present invention embrace a constant region formed by selective incorporation of human IgG4 sequences into a standard human IgG2 constant region. An exemplary mutant IgG2 Fc is IgG2m4, set forth herein as SEQ ID NO:140. Antibodies can be expressed as tetramers containing two light and two heavy chains, as separate heavy chains and light chains or as single chain antibodies in which heavy and light chain variable domains are linked through a spacer. Techniques for the production of single chain antibodies are well-known in the art.

Exemplary humanized antibodies derivatives of murine 20C2 monoclonal antibody are provided herein by CDR grafting and veneering. Amino acid sequences for IgG2M4 heavy chain variable regions, as well as kappa light chain variable regions for humanized 20C2 (i.e., Hu20C2A3) generated by veneering are presented in FIGS. 7A and 7C and set forth herein as SEQ ID NO:141 and SEQ ID NO:143.

Diabodies are also contemplated. A diabody refers to an engineered antibody construct prepared by isolating the binding domains (both heavy and light chain) of a binding antibody, and supplying a linking moiety which joins or operably links the heavy and light chains on the same polypeptide chain thereby preserving the binding function (see, Holliger et al. (1993) Proc. Natl. Acad. Sci. USA 90:6444; Poljak (1994) Structure 2:1121-1123). This forms, in essence, a radically abbreviated antibody, having only the variable domain necessary for binding the antigen. By using a linker that is too short to allow pairing between the two domains on the same chain, the domains are forced to pair with the complementary domains of another chain and create two antigen-binding sites. These dimeric antibody fragments, or diabodies, are bivalent and bispecific. The skilled artisan will appreciate that any method to generate diabodies can be used. Suitable methods are described by Holliger, et al. (1993) supra, Poljak (1994) supra, Zhu, et al. (1996) Biotechnology 14:192-196, and U.S. Pat. No. 6,492,123, incorporated herein by reference.

Fragments of an isolated antibody of the invention are also expressly encompassed by the instant invention. Fragments are intended to include Fab fragments, F(ab′)₂ fragments, F(ab′) fragments, bispecific scFv fragments, Fd fragments and fragments produced by a Fab expression library, as well as peptide aptamers. For example, F(ab′)₂ fragments are produced by pepsin digestion of the antibody molecule of the invention, whereas Fab fragments are generated by reducing the disulfide bridges of the F(ab′)₂ fragments. Alternatively, Fab expression libraries can be constructed to allow rapid and easy identification of monoclonal Fab fragments with the desired specificity (see Huse, et al. (1989) Science 254:1275-1281). In particular embodiments, antibody fragments of the present invention are fragments of neutralizing antibodies which retain the variable region binding site thereof. Exemplary are F(ab′)₂ fragments, F(ab′) fragments, and Fab fragments. See generally Immunology: Basic Processes (1985) 2^nd edition, J. Bellanti (Ed.) pp. 95-97.

Peptide aptamers which differentially recognize multi-dimensional conformations of ADDLs can be rationally designed or screened for in a library of aptamers (e.g., provided by Aptanomics SA, Lyon, France). In general, peptide aptamers are synthetic recognition molecules whose design is based on the structure of antibodies. Peptide aptamers consist of a variable peptide loop attached at both ends to a protein scaffold. This double structural constraint greatly increases the binding affinity of the peptide aptamer to levels comparable to that of an antibody (nanomolar range).

Exemplary nucleic acid sequences encoding heavy and light chain variable regions for use in producing antibody and antibody fragments of the instant invention are respectively disclosed herein in FIGS. 7B and 7D (i.e., SEQ ID NOs:142 and 144). As will be appreciated by the skilled artisan, the heavy chain variable regions disclosed herein can be used in combination with any one of the light chain variable regions disclosed herein to generate antibodies with modified affinities, dissociate constants, epitopes and the like.

Antibodies or antibody fragments of the present invention can have additional moieties attached thereto. For example, a microsphere or microparticle can be attached to the antibody or antibody fragment, as described in U.S. Pat. No. 4,493,825, the disclosure of which is incorporated herein by reference.

Moreover, particular embodiment embrace antibody or antibody fragments which are mutated and selected for increased antigen affinity, neutralizing activity (i.e., the ability to block binding of ADDLs to neuronal cells or the ability to block ADDL assembly), or a modified dissociation constant. Mutator strains of E. coli (Low, et al. (1996) J. Mol. Biol. 260:359-368), chain shuffling (Figini, et al. (1994) supra), and PCR mutagenesis are established methods for mutating nucleic acid molecules encoding antibodies. By way of illustration, increased affinity can be selected for by contacting a large number of phage antibodies with a low amount of biotinylated antigen so that the antibodies compete for binding. In this case, the number of antigen molecules should exceed the number of phage antibodies, but the concentration of antigen should be somewhat below the dissociation constant. Thus, predominantly mutated phage antibodies with increased affinity bind to the biotinylated antigen, while the larger part of the weaker affinity phage antibodies remains unbound. Streptavidin can then assist in the enrichment of the higher affinity, mutated phage antibodies from the mixture (Schier, et al. (1996) J. Mol. Biol. 255:28-43). Exemplary affinity-maturated light chain CDR3 amino acid sequences are disclosed herein (see Tables 6 and 7), with particular embodiments embracing a light chain CDR3 amino acid sequence of Xaa₁-Gln-Xaa₂-Thr-Arg-Val-Pro-Leu-Thr (SEQ ID NO:2), wherein Xaa₁ is Phe or Leu, and Xaa₂ is Ala or Thr. Affinity-maturated heavy chain CDR3 amino acid sequences are also provided herein. An exemplary heavy chain CDR3 amino acid sequence is set forth herein as Arg-Gln-Leu-Gly-Thr-Arg-Gly-Thr-Asp-Ala-Met-Asp-Tyr (SEQ ID NO:3). The present invention also embraces derivatives of this CDR3, e.g., Arg-Ala-Leu-Ser-Pro-Arg-Ser-Ile-Asp-Ala-Met-Asp-Tyr (SEQ ID NO:4), Arg-Gln-Leu-Gly-Ala-Arg-Lys-Thr-Asp-Ala-Met-Asp-Tyr (SEQ ID NO:5), Arg-Gln-Leu-Gly-Pro-Arg-Lys-Arg-Asp-Ala-Met-Asp-Tyr (SEQ ID NO:6), Arg-Gln-Leu-Gly-Lys-Leu-Lys-Thr-Asp-Ala-Met-Asp-Tyr (SEQ ID NO:7), or Arg-Gln-Leu-Gly-Arg-Arg-Ser-Val-Asp-Ala-Met-Asp-Tyr (SEQ ID NO:8), wherein differences with the Hu20C2A3 heavy chain CDR3 are underlined. In this regard, the present invention specifically embraces an anti-ADDL antibody having a CDR3 amino acid sequence of Arg-Xaa₁-Leu-Xaa₂-Xaa₃-Xaa₄-Xaa₅-Xaa₆-Asp-Ala-Met-Asp-Tyr (SEQ ID NO:9), wherein Xaa₁ is Gln or Ala; Xaa₂ is Ser or Gly; Xaa₃ is Pro, Ala, Lys, Arg, or Thr; Xaa₄ is Lys or Arg; Xaa₅ is Gly, Ser, or Lys; Xaa₆ is Val, Thr, Ile or Arg

Other antibody derivatives encompassed within the scope of the present invention include any humanized antibody identical to Hu20C2A3's variable regions except with a one amino acid residue difference in the frame region of the light chain (e.g., Leu-Pro-Val-Thr-Pro-Gly-Glu-Pro-Ala-Ser, SEQ ID NO:10).

For some therapeutic applications it may be desirable to reduce the dissociation of the antibody from the antigen. To achieve this, phage antibodies are bound to biotinylated antigen and an excess of unbiotinylated antigen is added. After a period of time, predominantly the phage antibodies with the lower dissociation constant can be harvested with streptavidin (Hawkins, et al. (1992) J. Mol. Biol. 226:889-96).

Various immunoassays including those disclosed herein can be used for screening to identify antibodies, or fragments thereof, having the desired specificity for multi-dimensional conformations of ADDLs. Numerous protocols for competitive binding (e.g, ELISA), latex agglutination assays, immunoradiometric assays, kinetics (e.g., BIACORE™ analysis) using either polyclonal or monoclonal antibodies, or fragments thereof, are well-known in the art. Such immunoassays typically involve the measurement of complex formation between a specific antibody and its cognate antigen. A two-site, monoclonal-based immunoassay utilizing monoclonal antibodies reactive to two non-interfering epitopes is suitable, but a competitive binding assay can also be employed. Such assays can also be used in the detection of multi-dimensional conformations of ADDLs in a sample.

An antibody or antibody fragment can also be subjected to other biological activity assays, e.g., displacement of ADDL binding to neurons or cultured hippocampal cells or blockade of ADDL assembly, in order to evaluate neutralizing or pharmacological activity and potential efficacy as a prophylactic or therapeutic agent. Such assays are described herein and are well-known in the art.

Antibodies and fragments of antibodies can be produced and maintained as hydridomas or alternatively recombinantly produced in any well-established expression system including, but not limited to, E. coli, yeast (e.g., Saccharomyces spp. and Pichia spp.), baculovirus, mammalian cells (e.g., myeloma, CHO, COS), plants, or transgenic animals (Breitling and Dübel (1999) In: Recombinant Antibodies, John Wiley & Sons, Inc., NY, pp. 119-132). Antibodies and fragments of antibodies can be isolated using any appropriate methods including, but not limited to, affinity chromatography, immunoglobulins-binding molecules (e.g., proteins A, L, G or H), tags operatively linked to the antibody or antibody fragment (e.g., His-tag, FLAG®-tag, Strep tag, c-myc tag) and the like. See, Breitling and Dübel (1999) supra.

Antibodies and antibody fragments of the instant invention have a variety of uses including, diagnosis of diseases associated with accumulation of ADDLs, blocking or inhibiting binding of ADDLs to neuronal cells, blocking ADDL assembly, prophylactically or therapeutically treating a disease associated with ADDLs, identifying therapeutic agents that prevent binding of ADDLs to neurons, and preventing the phosphorylation of tau protein at Ser202/Thr205.

Antibody and antibody fragments of the instant invention are also useful in a method for blocking or inhibiting binding of ADDLs to neuronal cells. This method of the invention is carried out by contacting a neuron, in vitro or in vivo, with an antibody or antibody fragment of the present invention so that binding of ADDLs to the neuron is blocked. In particular embodiments, an antibody or antibody fragment of the instant invention achieves at least a 15%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 95%, or 97% decrease in the binding of ADDLs as compared to binding of ADDLs in the absence of the antibody or antibody fragment. The degree to which an antibody can block the binding of ADDLs to a neuron can be determined in accordance with the methods disclosed herein, i.e., immunocytochemistry or cell-based alkaline phosphatase assay or any other suitable assay. Antibodies particularly useful for decreasing binding of ADDLs to neuronal cells include anti-ADDL antibodies having a CDR3 amino acid sequence set forth in SEQ ID NO:9, as well as derivatives and fragments thereof.

Antibody and antibody fragments of the instant invention are further useful in a method for blocking or inhibiting assembly of ADDLs. This method involves contacting a sample containing amyloid β1-42 peptides with an antibody or antibody fragment of the instant invention so that ADDL assembly is inhibited. The degree to which an antibody can block the assembly of ADDLs can be determined in accordance with the methods disclosed herein, i.e., FRET or fluorescence polarization or any other suitable assay. Antibodies particularly useful for blocking the assembly of ADDLs include anti-ADDL antibodies having a CDR3 amino acid sequence set forth in SEQ ID NO:9, as well as derivatives and fragments thereof.

Antibodies disclosed herein are also useful in methods for preventing the phosphorylation of tau protein at Ser202/Thr205. This method involves contacting a sample containing tau protein with an antibody or antibody fragment of the instant invention so that binding of ADDLs to neurons is blocked thereby preventing phosphorylation of tau protein. The degree to which an antibody can prevent the phosphorylation of tau protein at Ser202/Thr205 can be determined in accordance with the methods disclosed herein or any other suitable assay.

Blocking or decreasing binding of ADDLs to neurons, inhibiting assembly of ADDLs, and preventing the phosphorylation of tau protein at Ser202/Thr205 all find application in methods of prophylactically or therapeutically treating a disease associated with the accumulation of ADDLs. Accordingly, the present invention also embraces the use of an antibody or antibody fragment of the instant invention to prevent or treat a disease associated with the accumulation of ADDLs (e.g. Alzheimer's or similar memory-related disorders). Evidence in the art indicates that elevated levels of Aβ, but not necessarily aggregated plaque, are causative for Alzheimer's Disease-associated dementia and subsequent tau abnormalities. Aβ-derived diffusible ligands are directly implicated in neurotoxicity associated with Alzheimer's Disease. The art indicates that ADDLs are elevated in transgenic mice and Alzheimer's Disease patients and modulate functional activity associated with mnemonic processes in animal models. Thus, removing this form of Aβ could provide relief from the neurotoxicity associated with Alzheimer's Disease. As such, treatment with the instant antibody, which reduces central nervous system ADDL load, could prove efficacious for the treatment of Alzheimer's Disease. Patients amenable to treatment include individuals at risk of disease but not exhibiting symptoms, as well as patients presently exhibiting symptoms. In the case of Alzheimer's Disease, virtually anyone is at risk of suffering from Alzheimer's Disease if he or she lives long enough. Therefore, the antibody or antibody fragments of the present invention can be administered prophylactically to the general population without the need for any assessment of the risk of the subject patient. The present methods are especially useful for individuals who have a known genetic risk of Alzheimer's Disease. Such individuals include those having relatives who have been diagnosed with the disease, and those whose risk is determined by analysis of genetic or biochemical markers. Genetic markers of risk for Alzheimer's Disease include mutations in the APP gene, particularly mutations at position 717 and positions 670 and 671 referred to as the Hardy and Swedish mutations respectively. Other markers of risk are mutations in the presenilin genes, PS1 and PS2, and ApoE4, family history of Alzheimer's Disease, hypercholesterolemia or atherosclerosis. Individuals presently suffering from Alzheimer's Disease can be recognized from characteristic dementia, as well as the presence of risk factors described above. In addition, a number of diagnostic tests are available for identifying individuals who have Alzheimer's Disease. These include measurement of CSF tau and Aβ1-42 levels. Individuals suffering from Alzheimer's Disease can also be diagnosed by ADRDA criteria or the method disclosed herein.

In asymptomatic patients, treatment can begin at any age (e.g., 10, 20, 30 years of age). Usually, however, it is not necessary to begin treatment until a patient reaches 40, 50, 60 or 70 years of age. Treatment typically entails multiple dosages over a period of time. Treatment can be monitored by assaying for the presence of ADDLs over time.

In therapeutic applications, a pharmaceutical composition or medicament containing an antibody or antibody fragment of the invention is administered to a patient suspected of, or already suffering from such a disease associated with the accumulation of ADDLs in an amount sufficient to cure, or at least partially arrest, the symptoms of the disease (biochemical, histologic and/or behavioral), including its complications and intermediate pathological phenotypes in development of the disease. In prophylactic applications, a pharmaceutical composition or medicament containing an antibody or antibody fragment of the invention is administered to a patient susceptible to, or otherwise at risk of, a disease associated with the accumulation of ADDLs in an amount sufficient to achieve passive immunity in the patient thereby eliminating or reducing the risk, lessening the severity, or delaying the outset of the disease, including biochemical, histologic and/or behavioral symptoms of the disease, its complications and intermediate pathological phenotypes presenting during development of the disease. In some methods, administration of agent reduces or eliminates myocognitive impairment in patients that have not yet developed characteristic Alzheimer's pathology. In particular embodiments, an effective amount of an antibody or antibody fragment of the invention is an amount which achieves at least a 15%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 95%, or 97% decrease in the binding of ADDLs to neurons in the patient as compared to binding of ADDLs in the absence of treatment. As such, impairment of long-term potentiation/memory formation is decreased.

Effective doses of the compositions of the present invention, for the treatment of the above described conditions vary depending upon many different factors, including means of administration, physiological state of the patient, whether the patient is human or an animal, other medications administered, and whether treatment is prophylactic or therapeutic. Usually, the patient is a human but nonhuman mammals such as dogs or transgenic mammals can also be treated.

Treatment dosages are generally titrated to optimize safety and efficacy. For passive immunization with an antibody or antibody fragment, dosage ranges from about 0.0001 to 100 mg/kg, and more usually 0.01 to 5 mg/kg, of the host body weight are suitable. For example, dosages can be 1 mg/kg body weight or 10 mg/kg body weight or within the range of 1-10 mg/kg. In some methods, two or more antibodies of the invention with different binding specificities are administered simultaneously, in which case the dosage of each antibody administered falls within the ranges indicated. Antibodies are usually administered on multiple occasions, wherein intervals between single dosages can be weekly, monthly or yearly. An exemplary treatment regime entails subcutaneous dosing, once biweekly or monthly. Advantageously, subcutaneous administration has been found to reduce the flu-like symptoms associated with intravenous infusions (Lundin, et al. (2002) Blood 100:768-773). Intervals can also be irregular as indicated by measuring blood levels of antibody to ADDLs in the patient. In some methods, dosage is adjusted to achieve a plasma antibody concentration of 1-1000 μg/mL and in some methods 25-300 μg/mL. Alternatively, the antibody or antibody fragment can be administered as a sustained-release formulation, in which case less frequent administration is required. Dosage and frequency vary depending on the half-life of the antibody in the patient. In general, human and humanized antibodies have longer half-lives than chimeric antibodies and nonhuman antibodies. As indicated above, dosage and frequency of administration can vary depending on whether the treatment is prophylactic or therapeutic. In prophylactic applications, a relatively low dosage is administered at relatively infrequent intervals over a long period of time. Some patients continue to receive treatment for the rest of their lives. In therapeutic applications, a relatively high dosage at relatively short intervals is sometimes required until progression of the disease is reduced or terminated, and preferably until the patient shows partial or complete amelioration of symptoms of disease. Thereafter, the patient can be administered a prophylactic regime.

Antibody and antibody fragments of the instant invention can be administered as a component of a pharmaceutical composition or medicament. Pharmaceutical compositions or medicaments generally contain the active therapeutic agent and a variety of other pharmaceutically acceptable components. See Remington: The Science and Practice of Pharmacy, Alfonso R. Gennaro, editor, 20th ed. Lippincott Williams & Wilkins: Philadelphia, Pa., 2000. The preferred form depends on the intended mode of administration and therapeutic application. Pharmaceutical compositions can contain, depending on the formulation desired, pharmaceutically-acceptable, non-toxic carriers or diluents, which are defined as vehicles commonly used to formulate pharmaceutical compositions for animal or human administration. Diluents are selected so as not to affect the biological activity of the combination. Examples of such diluents are distilled water, physiological phosphate-buffered saline, Ringer's solutions, dextrose solution, and Hank's solution.

Pharmaceutical compositions can also contain large, slowly metabolized macromolecules such as proteins, polysaccharides such as chitosan, polylactic acids, polyglycolic acids and copolymers (such as latex-functionalized SEPHAROSE™, agarose, cellulose, and the like), polymeric amino acids, amino acid copolymers, and lipid aggregates (such as oil droplets or liposomes).

Administration of a pharmaceutical composition or medicament of the invention can be carried out via a variety of routes including, but not limited to, oral, topical, pulmonary, rectal, subcutaneous, intradermal, intranasal, intracranial, intramuscular, intraocular, or intra-articular injection, and the like. The most typical route of administration is intravenous followed by subcutaneous, although other routes can be equally effective. Intramuscular injection can also be performed in the arm or leg muscles. In some methods, agents are injected directly into a particular tissue where deposits have accumulated, for example, intracranial injection. In some embodiments, an antibody or antibody fragment is injected directly into the cranium. In other embodiments, antibody or antibody fragment is administered as a sustained-release composition or device, such as a MEDIPAD™ device.

For parenteral administration, antibody or antibody fragments of the invention can be administered as injectable dosages of a solution or suspension of the substance in a physiologically acceptable diluent with a pharmaceutical carrier that can be a sterile liquid such as water, oils, saline, glycerol, or ethanol. Additionally, auxiliary substances, such as wetting or emulsifying agents, surfactants, pH buffering substances and the like can be present in compositions. Other components of pharmaceutical compositions are those of petroleum, animal, vegetable, or synthetic origin, for example, peanut oil, soybean oil, and mineral oil. In general, glycols such as propylene glycol or polyethylene glycol are suitable liquid carriers, particularly for injectable solutions. Antibodies can be administered in the form of a depot injection or implant preparation which can be formulated in such a manner as to permit a sustained-release of the active ingredient.

An exemplary composition contains the instant antibody or antibody fragment formulated as a sterile, clear liquid at a concentration of at least 10 mg/ml in isotonic buffered saline (10 mM histidine, 150 mM sodium chloride, 0.01% (w/v) POLYSORBATE 80, pH 6.0). An exemplary antibody formulation is filled as a single dose, 0.6 ml glass vials filled with 3.3 ml of solution per vial. Each vial is stopped with a TEFLON-coated stopper and sealed with an aluminum cap.

Typically, compositions are prepared as injectables, either as liquid solutions or suspensions; solid forms suitable for solution in, or suspension in, liquid vehicles prior to injection can also be prepared. The preparation also can be emulsified or encapsulated in liposomes or micro particles such as polylactide, polyglycolide, or copolymer for enhanced delivery.

For suppositories, binders and carriers include, for example, polyalkylene glycols or triglycerides; such suppositories can be formed from mixtures containing the active ingredient in the range of 0.5% to 10%, or more desirably 1%-2%.

Oral formulations include excipients, such as pharmaceutical grades of mannitol, lactose, starch, magnesium stearate, sodium saccharine, cellulose, and magnesium carbonate. These compositions take the form of solutions, suspensions, tablets, pills, capsules, sustained-release formulations or powders and contain 10%-95% of active ingredient, or more suitably 25%-70%.

Topical application can result in transdermal or intradermal delivery. Topical administration can be facilitated by co-administration of the agent with cholera toxin or detoxified derivatives or subunits thereof or other similar bacterial toxins (see Glenn, et al. (1998) Nature 391:851). Co-administration can be achieved by using the components as a mixture or as linked molecules obtained by chemical crosslinking or expression as a fusion protein.

Alternatively, transdermal delivery can be achieved using a skin path or using transferosomes (Paul, et al. (1995) Eur. J. Immunol. 25:3521-24; Cevc, et al. (1998) Biochem. Biophys. Acta 1368:201-15).

An antibody or antibody fragment of the invention can optionally be administered in combination with other agents that are at least partly effective in treatment of amyloidogenic disease. For example, the instant antibody can be administered with existing palliative treatments for Alzheimer's Disease, such as acetylcholinesterase inhibitors such as ARICEPT™, EXELON™, and REMINYL™ and, the NMDA antagonist, NAMENDA™. In addition to these approved treatments, the instant antibody can be used to provide synergistic/additive benefit for any of several approaches currently in development for the treatment of Alzheimer's Disease, which include without limitation, inhibitors of Aβ production and aggregation.

Antibody and antibody fragments of the instant invention also find application in the identification of therapeutic agents that prevent the binding of ADDLs to neurons (e.g., a hippocampal cell) thereby preventing downstream events attributed to ADDLs. Such an assay is carried out by contacting a neuron with ADDLs in the presence of an agent and using an antibody of antibody fragment of the invention to determine binding of the ADDLs to the neuron in the presence of the agent. As will be appreciated by the skilled artisan, an agent that blocks binding of ADDLs to a neuron will decrease the amount of ADDLs bound to the neuron as compared to a neuron which has not been contacted with the agent; an amount which is detectable in an immunoassay employing an antibody or antibody fragment of the instant invention. Suitable immunoassays for detecting neuronal-bound ADDLs are disclosed herein.

Agents which can be screened using the method provided herein encompass numerous chemical classes, although typically they are organic molecules, preferably small organic compounds having a molecular weight of more than 100 and less than about 2,500 daltons. Agents encompass functional groups necessary for structural interaction with proteins, particularly hydrogen bonding, and typically include at least an amine, carbonyl, hydroxyl or carboxyl group, preferably at least two of the functional chemical groups. The agents often contain cyclical carbon or heterocyclic structures and/or aromatic or polyaromatic structures substituted with one or more of the above functional groups. Agents can also be found among biomolecules including peptides, antibodies, saccharides, fatty acids, steroids, purines, pyrimidines, derivatives, structural analogs or combinations thereof. Agents are obtained from a wide variety of sources including libraries of natural or synthetic compounds.

A variety of other reagents such as salts and neutral proteins can be included in the screening assays. Also, reagents that otherwise improve the efficiency of the assay, such as protease inhibitors, nuclease inhibitors, anti-microbial agents, and the like can be used. The mixture of components can be added in any order that provides for the requisite binding.

Agents identified by the screening assay of the present invention will be beneficial for the treatment of amyloidogenic diseases and/or tauopathies. In addition, it is contemplated that the experimental systems used to exemplify these concepts represent research tools for the evaluation, identification and screening of novel drug targets associated with amyloid beta induction of tau phosphorylation.

The present invention also provides methods for detecting ADDLs and diagnosing a disease associated with accumulation of ADDLs using an antibody or antibody fragment of the instant invention. A disease associated with accumulation of ADDLs is intended to include any disease wherein the accumulation of ADDLs results in physiological impairment of long-term potentiation/memory formation. Diseases of this type include, but are not limited to, Alzheimer's Disease and similar memory-related disorders.

In accordance with these methods, a sample from a patient is contacted with an antibody or antibody fragment of the invention and binding of the antibody or antibody fragment to the sample is indicative of the presence of ADDLs in the sample. As used in the context of the present invention, a sample is intended to mean any bodily fluid or tissue which is amenable to analysis using immunoassays. Suitable samples which can be analyzed in accordance with the methods of the invention include, but are not limited to, biopsy samples and fluid samples of the brain from a patient (e.g., a mammal such as a human). For in vitro purposes (e.g., in assays monitoring oligomer formation), a sample can be a neuronal cell line or tissue sample. For diagnostic purposes, it is contemplated that the sample can be from an individual suspected of having a disease associated with accumulation of ADDLs or from an individual at risk of having a disease associated with accumulation of ADDLs, e.g., an individual with a family history which predisposes the individual to a disease associated with accumulation of ADDLs.

Detection of binding of the antibody or antibody fragment to ADDLs in the sample can be carried out using any standard immunoassay (e.g., as disclosed herein), or alternatively when the antibody fragment is, e.g., a peptide aptamer, binding can be directly detected by, for example, a detectable marker protein (e.g., β-galactosidase, GFP or luciferase) fused to the aptamer. Subsequently, the presence or absence of the ADDL-antibody complex is correlated with the presence or absence, respectively, of ADDLs in the sample and therefore the presence or absence, respectively, of a disease associated with accumulation of ADDLs. It is contemplated that one or more antibodies or antibody fragments of the present invention can be used in conjunction with current non-invasive immuno-based imaging techniques to greatly enhance detection and early diagnosis of a disease associated with accumulation of ADDLs.

To facilitate diagnosis the present invention also pertains to a kit for containing an antibody or antibody fragment of the instant invention. The kit includes a container holding one or more antibody or antibody fragments which recognizes multi-dimensional conformation of ADDLs and instructions for using the antibody for the purpose of binding to ADDLs to form an antibody-antigen complex and detecting the formation of the antibody-antigen complex such that the presence or absence of the antibody-antigen complex correlates with presence or absence of ADDLs in the sample. Examples of containers include multiwell plates which allow simultaneous detection of ADDLs in multiple samples.

The invention is described in greater detail by the following non-limiting examples.

EXAMPLE 1

General Materials and Methods

ADDL Preparation. ADDLs in F12 medium (Biosource, Camarillo, Cailf.) were prepared from Aβ1-42 in accordance with established methods (Lambert, et al. (2001) supra). Briefly, Aβ1-42 peptide (American Peptide Co., Sunnyvale, Cailf. or California Peptide Research, Inc., Napa, Cailf.) was weighed and placed in a glass vial capable of holding a sufficient quantity of HFIP (1,1,1,3,3,3-hexafluoro-2-propanol) to achieve a peptide concentration of 10 mg/mL. HFIP was added to the dry peptide, the vial was capped and gently swirl to mix, and the peptide/HFIP solution was stored at room temperature for at least one hour. Aliquots (50 or 100 μL, 0.5 or 1.0 mg, respectively) of peptide solution was dispensed into a series of 1.5 mL conical centrifuge tubes. The tubes were placed in a SPEEDVAC® overnight to remove the HFIP. Tubes containing the dried peptide film were capped and stored at −70° C. in a sealed container with dessicant.

Prior to use, the Aβ1-42 peptide film was removed from −70° C. storage and allowed to warm to room temperature. Fresh DMSO (44 μL/mg of peptide film; 5 mM) was added and the peptide/DMSO mixture was incubated on a vortex mixer at the lowest possible speed for ten minutes. F12 media (2 mL/mg peptide) was dispensed into each tube of DMSO/peptide and the tube was capped and mixed by inversion. The 100 μM preparation was stored at 2-8° C. for eighteen to twenty four hours. The samples were centrifuged at 14,000×g for ten minutes at 2-8° C. The supernatant was transferred to a fresh tube and stored at 2-8° C. until used.

Biotinylated ADDL preparations (bADDLs) were prepared in the same manner as described above for ADDL preparations using 100% N-terminal biotinylated amyloid beta peptide (American Peptide Company, Sunnyvale, Cailf.).

Monomer Preparation. HFIP dry down preparations of amyloid beta (1-40) peptide (Aβ1-40) were prepared as outlined for Aβ(1-42) peptide. The peptide film was dissolved in 2 mL of 25 mM borate buffer (pH 8.5) per mg of peptide, divided into aliquots, and frozen at −70° C. until used.

Primary Neurons. Primary hippocampal cultures were prepared from frozen, dissociated neonatal rat hippocampal cells (Cambrex, Corp., East Rutherford, N.J.) that were thawed and plated in 96-well COSTAR® plates at a concentration of 20,000 cells per well. The cells were maintained in NEUROBASAL™ media without L-glutamine (GIBCO-BRL™, Gaithersburg, Md.) and supplemented with B27 (GIBCO-BRL™, Gaithersburg, Md.) for a period of two weeks and then used for binding studies.

Immunocytochemistry. Immunocytochemistry was performed according to established methods (Lambert, et al. (2001) supra), except the secondary antibodies were conjugated to ALEXAFLUOR® 588 (Molecular Probes, Eugene, Oreg.). Antibodies and ADDLs were preincubated for 1 hour at room temperature, at a molar ratio of 1:4 antibody:ADDL before application to the 21-day hippocampal cell culture. For endogenous ADDLs, human brain protein (prepared as in Lambert, et al. (2001) supra) was incubated with cells for 1 hour before the cells were washed, fixed, and visualized as above.

Lightly fixed frozen sections (4% paraformaldehyde at 4° C. for 30 hours and cryoprotected in 40 μm sucrose) from Alzheimer's Disease and control hippocampus were incubated with antibody (1:1000 in phosphate-buffered saline (PBS)) overnight at 4° C. After removal of antibody, sections were washed 3 times with PBS and incubated with secondary antibody at room temperature. Binding was then visualized with DAB (SIGMA™, St. Louis, Mo.). Sections were then counterstained with hematoxylin, mounted, and imaged on a NIKON® ECLIPSE® E600 light microscope with a SPOT™ INSIGHT™ digital video camera (v. 3.2).

ELISA. Polyclonal anti-ADDLs IgG (M90/1; Bethyl Laboratories, Inc., Montgomery, Tex.) was plated at 0.25 mg/well on IMMULONT™ 3 REMOVAWELL™ strips (Dynatech Labs, Chantilly, Va.) for 2 hours at room temperature and the wells blocked with 2% BSA in TBS. Samples diluted with 1% BSA in F12 were added to the wells, allowed to bind for 2 hours at 4° C., and washed 3× with BSA/TBS at room temperature. Monoclonal antibodies diluted in BSA/TBS were incubated for 90 minutes at room temperature and detected with a VECTASTAIN® ABC kit to mouse IgG. The HRP label was visualized with BIO-RAD® peroxidase substrate and read at 405 nm on a Dynex MRX-TC microplate reader.

EXAMPLE 2

Isolation of Mouse Antibody Variable Region Sequences

The cDNAs coding for the variable domains of the 20C2 mouse antibody were cloned and sequenced following a polymerase chain reaction (PCR) using specially designed primers that hybridize to the 5′-ends of the mouse constant regions and to the murine leader sequences upstream of the V regions. This ensured that the mouse variable region sequences obtained were complete and accurate. In short, mRNA was extracted from mouse hybridoma cell lines using the QIAGEN® OLIGOTEX® Direct mRNA Mini Kit and subsequently converted to cDNA using a first-strand cDNA synthesis kit. The cDNA was then used as template in PCR reactions to obtain the antibody variable region sequences.

To obtain the light chain variable region sequence, eleven independent PCR reactions were set up using each of the eleven light chain 5′ PCR primers (MKV-1 to MKV-11) and the 3′ PCR primer MKC-1 (Table 1).

	TABLE 1
	SEQ
5′		ID
Primer	Sequence	NO:
MKV-1	GAT CTC TAG ATG AAC ATT GCC TGT TAG GCT	11
	GTT GGT GCT G
MKV-2	GAT CTC TAG ATG GAG WCA GAC ACA CTC CTG	12
	YTA TGG GTG
MKV-3	GAT CTC TAG ATG AGT GTG CTC ACT CAG GTC	13
	CTG GSG TTG
MKV-4	GAT CTC TAG ATG AGG RCC CCT GCT CAG WTT	14
	YTT GGM WTC TTG
MKV-5	GAT CTC TAG ATG GAT TTW CAG GTG CAG ATT	15
	WTC AGC TTC
MKV-6	GAT CTC TAG ATG AGG TKC YYT GYT SAY CTY	16
	CTC TGR GG
MKV-7	GAT CTC TAG ATG GGC WTC AAA GAT GGA GTC	17
	ACA KWY YCW GG
MKV-8	GAT CTC TAG ATG TGG GGA YCT KTT TYC MMT	18
	TTT TCA ATG
MKV-9	GAT CTC TAG ATG GTR TCC WCA SCT CAG TTC	19
	CTT G
MKV-10	GAT CTC TAG ATG TAT ATA TGT TTG TTG TCT	20
	ATT TCT
MKV-11	GAT CTC TAG ATG GAA GCC CCA GCT CAG CTT	21
	CTC TTC C
	SEQ
3′		ID
Primer	Sequence	NO:
MKC-1	GAT CGA GCT CAC TGG ATG GTG GGA AGA TGG	22
Underlined and italic sequences denote XbaI and SacI restriction sites, respectively. W = A or T, M = A or C, K = G or T, Y = C or T, and R = A or G.

To obtain the heavy chain variable region sequences twelve independent PCR reactions were set up using each of the twelve heavy chain 5′ PCR primers (MHV-1 to MHV-12) and the appropriate isotype specific 3′ primer (MHCG-1, MHCG-2A, MHCG-2B, MHCG-3) (Table 2).

	TABLE 2
	SEQ
5′		ID
Primer	Sequence	NO:
MHV-1	GAT CTC TAG ATG AAA TGC AGC TGG GGC ATS	23
	TTC TTC
MHV-2	GAT CTC TAG ATG GGA TGG AGC TRT ATC ATS	24
	YTC TT
MHV-3	GAT CTC TAG ATG AAG WTG TGG TTA AAC TGG	25
	GTT TTT
MHV-4	GAT CTC TAG ATG RAC TTT GGG YTC AGC TTG	26
	RTT T
MHV-5	GAT CTC TAG ATG GGA CTC CAG GCT TCA ATT	27
	TAG TTT TCC TT
MHV-6	GAT CTC TAG ATG GCT TGT CYT TRG SGC TRC	28
	TCT TCT GC
MHV-7	GAT CTC TAG ATG GRA TGG AGC KGG RGT CTT	29
	TMT CTT
MHV-8	GAT CTC TAG ATG AGA GTG CTG ATT CTT TTG	30
	TG
MHV-9	GAT CTC TAG ATG GMT TGG GTG TGG AMC TTG	31
	CTT ATT CCT G
MHV-10	GAT CTC TAG ATG GGC AGA CTT ACC ATT CTC	32
	ATT CCT G
MHV-11	GAT CTC TAG ATG GAT TTT GGG CTG ATT TTT	33
	TTT ATT G
MHV-12	GAT CTC TAG ATG ATG GTG TTA AGT CTT CTG	34
	TAC CTG
	SEQ
3′		ID
Primer	Sequence	NO:
MHCG-1	GCATC GAG CTC CAG TGG ATA GAC AGA TGG GGG	35
MHCG-2A	GCATC GAG CTC CAG TGG ATA GAC CGA TGG GGG	36
MHCG-2B	GCATC GAG CTC CAG TGG ATG AGC TGA TGG GGG	37
MHCG-3	GCATC GAG CTC CAA GGG ATA GAC AGA TGG GGC	38
Underlined and italic sequences denote XbaI and SacI restriction sites, respectively. W = A or T, M = A or C, K = G or T, Y = C or T, and R = A or G.

Each of the light chain PCR reactions contained 46 μL INVITROGEN™ PLATINUM® PCR Super Mix, 1.0 μL of one of the 100 μM 5′ primers (MKV-1 to MKV-11), 1.0 μL of the 100 μM 3′ primer (MKC-1), and 2.0 μL of hybridoma cDNA. Similar PCR reactions were employed to clone the mouse heavy chain variable region sequences. Reactions were placed in a DNA thermal cycler and, after an initial denaturation step at 97° C. for 2.0 minutes, subjected to 30 cycles of: 95° C. for 30 seconds, 55° C. for 45 seconds, and 72° C. for 90 seconds. Following the last cycle, a final extension step at 72° C. for 10 minutes was employed. To determine which PCR reactions yielded product, 5 μL aliquots from each reaction were separated on 1.5% (w/v) agarose/1X TAE buffer gels, containing 0.5 μg/mL ethidium bromide. PCR products from reactions that produced fragments of the expected size (420 to 500 bp) were then gel purified, digested with XbaI and SacI and ligated into the XbaI and SacI sites in the multicloning region of plasmid pNEB193 (New England Biolabs, Beverly, Mass.). Alternatively, PCR products were ligated directly into plasmid pCR® 2.1 using the INVITROGEN™ TA CLONING® kit. Ligation products were then transformed into XL-1 cells and aliquots of the transformed E. coli were plated onto LB agar plates containing 50 μg/mL ampicillin and overlaid with 40 μL of X-Gal stock (50 mg/mL) and 40 μL IPTG (100 mM) solution for blue/white selection. Plates were incubated overnight at 37° C. and potential clones were identified as white colonies. DNA from at least 24 independent clones for each PCR product were sequenced on both strands using universal forward and reverse primers for pNEB193 and pCR® 2.1. The resulting sequences were then assembled into a contig to generate a consensus sequence for each antibody light and heavy chain variable region. Using this approach the sequences were determined for the light and heavy antibody variable regions of hybridoma 20C2 (FIGS. 1A-1B). The six complementarity-determining regions (CDRs), which form the structure complementary to the antigen, are underlined in FIGS. 1A-1B.

EXAMPLE 3

Humanization of Mouse Anti-ADDL Antibody Variable Region Sequences

Mouse antibody heavy and light variable domain nucleic acids obtained from mouse hybridoma cell line 20C2 were humanized using a CDR grafting approach. It will be appreciated by those skilled in the art that humanization of mouse antibody sequences can maximize the therapeutic potential of an antibody by improving its serum half-life and Fc effector functions thereby reducing the anti-globulin response.

Humanization by CDR grafting was carried out by selecting the human light and heavy chain variable regions from the NCBI protein database with the highest homology to the mouse variable domains. The mouse variable region sequences were compared to all human variable region sequences in the database using the protein-protein Basic Local Alignment Search Tool (BLAST). Subsequently, mouse CDRs were joined to the human framework regions and the preliminary amino acid sequence was analyzed. All differences between the mouse and human sequences in the framework regions were evaluated particularly if they were part of the canonical sequences for loop structure or were residues located at the VL/VH interface (O'Brien and Jones (2001) In: Antibody Engineering, Kontermann and Dubel (Eds.), Springer Laboratory Manuals). Framework regions were also scanned for unusual or rare amino acids in comparison to the consensus sequences for the human subgroup and for potential glycosylation sites. Wherein amino acid sequence differences existed between the mouse and human framework region sequences that were not found to be involved in canonical sequences, or located at the VL/VH interface, the human residue was selected at that position. Wherein a difference in a key residue existed, two versions of the variable region sequence were generated for evaluation. The CDR grafting strategy made the minimum number of changes to the human framework region so that good antigen binding was achieved while maintaining human framework regions that closely matched the sequence from a natural human antibody. The heavy chain and light chain variable region amino acid sequences of the resulting humanized antibody generated by CDR grafting of murine 20C2 are shown in FIGS. 2A and 2B, respectively. This antibody is designated herein as Hu20C2.

Humanized sequences for 20C2 were also designed using a veneering strategy (See, e.g., U.S. Pat. No. 6,797,492). Humanization was carried out by selecting the human light and heavy chain variable regions from the NCBI protein database with the highest homology to the mouse variable domains, as well as to the closest human antibody germline family or families (see, Kabat, et al. (1991) Sequences of proteins of immunological interest, 5^th ed., U.S. Dept. Health and Human Services, NIH, Washington D.C.). The mouse variable region sequences were compared to all human variable region sequences in the database using protein-protein BLAST. The murine variable sequences and their closest human homologues were modeled to the closest crystallized human antibody as determined by computer modeling as practiced in the art. From the model of the murine VH and VL sequences, a surface area map was constructed, which dictated the solvent accessibility of the amino acids in the mouse heavy and light variable regions. To confirm the modeling, these exposed residues were compared position-by-position with known surface accessible residues (see, e.g., Padlan (1994) Mol. Immunol. 31(3):169-217). A score was assigned for each residue in the sequence designating it as exposed, mostly exposed, partly buried, mostly buried and buried according to established methods (see, U.S. Pat. No. 6,797,492, incorporated herein by reference in its entirety). Mouse framework residues that scored as exposed or mostly exposed and differed from the homologous human sequence were changed to the human residue at that position. The designed veneered sequences retained the mouse CDRs, residues neighboring the CDRs, residues known be involved in canonical sequences, residues located at the VL/VH interface, and residues at the N-terminal sequences of the mouse heavy and light chain. The N-terminal sequences are known to be contiguous with the CDR surface and are potentially involved in ligand binding. Once the veneered sequences were finalized they were remodeled to look for are any potential obvious structural issues. A total of 12 and 9 amino acid residues were changed in the heavy chain and light chain frames, respectively. The heavy chain and light chain variable region amino acid sequences of the resulting humanized antibody generated by veneering of murine 20C2 are shown in FIGS. 2A and 2B, respectively. This antibody is designated herein as Hu20C2A3.

In comparison to 20C2, it is noted that the light chain substitutions resulting in Hu20C2A3, but not the heavy chain substitutions, are in common with Hu20C2 (FIG. 2). In particular, heavy chain variable region CDR3 is unique to Hu20C2A3.

Once the humanized amino acid sequences were selected the sequences were reverse-translated to obtain the corresponding DNA sequence. The DNA sequences were codon-optimized using art-established methods (Lathe (1985) J. Mol. Biol. 183(1):1-12) and designed with flanking restriction enzyme sites for cloning into human antibody expression vectors. The nucleotide sequences encoding the light chain variable region and two versions of the heavy chain variable region for Hu20C2 are presented in FIGS. 3A-3C. The two heavy chain variable region versions differ by a single amino acid substitution at position 24; heavy chain variable region for version A of Hu20C2 is Phe at position 24 and heavy chain variable region of version B of Hu20C2 is Leu at position 24.

EXAMPLE 4

Affinity Maturation

Affinity maturation was carried out on the Hu20C2 antibody. Nucleic acid molecules encoding humanized Hu20C2 versions A and B variable heavy chain only, light chain only or heavy chain version A and light chain together were cloned in the Fab phage-display vector pFab4. Nucleic acid sequence analysis confirmed sequence and orientation in pFab4. The annotated Hu20C2 Fab sequences in pFab4 are presented in FIGS. 4A-4C and set forth herein as SEQ ID NO:116 for heavy chain version A, SEQ ID NO:117 for heavy chain version B, and SEQ ID NO:118 for the light chain. The nucleotide sequence for heavy chain version A and light chain together in the pFab4 vector is presented in FIGS. 4D-4E. These constructs were used in the Hu20C2 maturation program using art-established phage-displayed Fab library methods.

Light Chain Maturation. Two libraries were designed to mutate the nine wild-type amino acids of CDR3 of the light (kappa) chain of Hu20C2 (i.e., Phe-Gln-Gly-Ser-Leu-Val-Pro-Leu-Thr; SEQ ID NO:39). These libraries were designated LC3-1 and LC3-2 representing light chain CDR3 sequences of Xaa-Xaa-Xaa-Xaa-Xaa-Val-Pro-Leu-Thr (SEQ ID NO:40) and Phe-Gln-Gly-Ser-Xaa-Xaa-Xaa-Xaa-Xaa (SEQ ID NO:41), respectively. Biotinylated reverse primers, 20C2LC3-1 (SEQ ID NO:123) and 20C2LC3-2 (SEQ ID NO:126), were used in combination with forward primer 20C2LC3F (SEQ ID NO:120) to generate the LC3-1 and LC3-2 libraries (see FIG. 5A). Primers were purified by polyacrylamide gel electrophoresis, whereas the vector DNA was purified by gel electrophoresis and electroelution. The two light chain libraries were designed to be randomly mutated. The final diversities of the three 10G5H6 LC₃ libraries were 4.76×10⁸ and 7.45×10⁸, respectively (Table 3). Sequence analysis of approximately 100 clones from the libraries showed 100% diversity of mutant clones at the designed amino acid positions.

	TABLE 3
	Light Chain Library
Characteristic	LC3-1	LC3-2
Vector	pFab3d20C2HS	pFab3d20C2HS
Number of	4.76 × 108	7.45 × 108
Transformants
Library Diversity	4.76 × 108 × 0.89 = 4.24	7.45 × 108 × 0.90 = 6.71
	108	108
Primary Library	2 mL	2 mL
Volume
Primary Library	2.13 × 1011	*9.3 × 1010
Titer
*Higher titers are achieved by concentration or phage rescue.

Soluble panning of the two light chain libraries against high molecular weight bADDL was completed. Briefly, four rounds of panning were carried out using biotinylated high molecular weight ADDL (bADDL). The first three rounds were carried out using approximately 1.5 μM antigen concentration (input=1×10¹⁰ to 1×10¹¹). Upon completion of the third round, the outputs of the two libraries were combined and divided into three groups for analysis with 10 nM, 100 nM and approximately 1.5 μM antigen to increase panning stringency. As such, a total of 58 output plates were tested in phage ELISA assays, i.e., two plates per library in the first round (a total of four plates), six plates per library in the second round (a total of 12 plates), eight plates for LC3-1 and 10 plates for LC3-2 libraries in the third round (a total of 18 plates) and eight plates for each antigen concentration in the fourth round (a total of 24 plates).

Panning resulted in 1000 hits, 436 of which were sequenced (Table 4).

TABLE 4
Round	Antigen	Input	Output	% Recovery	ELISA Screen*	Sequenced
1a	1.6	μM	2.13 × 1010	7.3 × 104	3.42 × 10−6	0%	0
						(0/176)
2a	2.0	μM	1.55 × 1011	1.88 × 105	1.21 × 10−6	1.5%	8
						(8/528)
3a	1.1	μM	1.80 × 1010	7.8 × 104	4.3 × 10−6	5.8%	41
						(41/704)
1b	1.6	μM	9.30 × 109	5.7 × 104	6.13 × 10−6	2.3%	4
						(7/176)
2b	2.0	μM	1.23 × 1011	1.07 × 105	8.7 × 10−7	4.5%	24
						(24/528)
3b	1.1	μM	1.37 × 1010	3.32 × 105	2.42 × 10−5	15%	134
						(134/880)
4c	1.1	μM	3.0 × 1011	1.37 × 105	4.6 × 10−7	39%	—
						(274/704)
4c	100	nM	3.0 × 1011	3.88 × 105	1.29 × 10−6	41%	—
						(290/704)
4c	10	nM	3.0 × 1011	1.6 × 105	5.3 × 10−7	32%	225
						(225/704)
					Total	1000/5104	436
a20C2 LC3-1 versus high molecular weight 10% bADDL.
b20C2 LC3-2 versus high molecular weight 10% bADDL.
c20C2 LC3-1 + 20C2 LC3-2 versus high molecular weight 10% bADDL.
*Hits per total number of colonies.

Sequence and frequency of highly enriched clones are presented in Table 5.

	TABLE 5
	SEQ
Clone		ID	Round	Round	Round
Designation	LC CDR3	NO:	2	3	4	Total
Hu20C2LC	FQGSLVPLT	39	6	15	14	35
SJ-p1-31	ADTTHVPLT	42		1	2	3
SJ-p1-14	AHSTFVPLT	43	1	1	2	4
4P2-12-E3	AQASFVPLT	44			2	2
SJ-p1-38	AQATKVPLT	45		1	1	2
4P3-59	AQSSKVPLT	46			2	2
SJ-p2-14	AQSTLVPLT	47		1	2	3
4P3-11	FAASSVPLT	48			2	2
4P3-1	FESTYVPLT	49			2	2
SJ-p2-10	FESSRVPLT	50		1	1	2
SJ-p2-11	FNATWVPLT	51		2		2
SJ-p2-60	FQASRVPLT	52		1	5	6
SJ-p1-18	FQATRVPLT	53		1	5	6
SJ-p3-51	FQGSFIGLS	54	1		1	2
SJ-p3-16	FQGSFIPGT	55		2	3	5
SJ-p8-8F	FQGSFLPPS	56		1	1	2
SJ-p3-26	FQGSFLPQL	57	1	2		3
SJ-p3-15	FQGSLFPPV	58	1	2		3
SJ-p2-70	FQGSLFSPS	59	1	5		6
SJ-p3-24	FQGSRIPIS	60		1	1	2
SJ-p3-33	FQGSRLPVS	61		2	3	5
SJ-p3-14	FQGSRVPLV	62		2	1	3
SJ-p2-1F	FQSSFVPLT	63		6	8	14
4P1-22	FQSSRVPLT	64			15	15
SJ-p2-44	GQTTLVPLT	65		1	3	4
SJ-p1-56	HESTLVPLT	66		2	1	3
4P1-40	HQSSKVPLT	67			4	4
SJ-p2-20	IQTSLVPLT	68		2		2
SJ-p1-41	IQAALVPLT	69		1	1	2
SJ-p2-13	LQSSFVPLT	70	1	4		5
4P1-26	LETSRVPLT	71			3	3
SJ-p1-33	LASSHVPLT	72		2	1	3
SJ-p2-27	LNSTTVPLT	73		2	4	6
SJ-p2-62	LQSKSVPLT	74		2		2
4P2-26-E5	LQSVRVPLT	75			3	3
4P1-32	LQSSLVPLT	76			5	5
SJ-p2-37	LQTGRVPLT	77		2	2	4
SJ-p2-64	LQTSFVPLT	78		3		3
4P1-20	LQTSNVPLT	79			5	5
SJ-p2-39	LQTTRVPLT	80		2	6	8
SJ-p2-52	LSSTFVPLT	81		3	1	4
SJ-p2-6L	LSSTHVPLT	82		2	1	3
4P1-77	LTSSAVPLT	83			2	2
SJ-p1-59	LVSSLVPLT	84		2		2
SJ-p2-23	METANVPLT	85		2		2
SJ-p1-9M	MQSSFVPLT	86		1	3	4
SJ-p2-28	MQSSLVPLT	87		1	2	3
SJ-p1-21	MQTSKVPLT	88		1	1	2
4P1-17	SQARMVPLT	89			3	3
SJ-p2-66	SQASRVPLT	90		1	2	3
SJ-p1-49	TQSTQVPLT	91		2	1	3
SJ-p2-24	VCATFVPLT	92		1	1	2
4P1-41	VQSSAVPLT	93			2	2
SJ-p2-51	VQTSLVPLT	94		12	31	43
4P1-64	VQTSVVPLT	95			3	3
SJ-p2-55	VQTTAVPLT	96		2		2
SJ-p1-25	LQTARVPLT	97		1	3	4

Fab fragments from the 10 top clones based on enrichment frequency were prepared and a total of 15 clones were converted into IgG1 humanized A version and two clones, 20C2-6 and 20C2-8, were converted to IgG1 humanized B version. K_D values for these clones were measured by BIACORE™ using biotin-Aβ1-20 (Table 6) and bADDL (Table 7) as antigens. Dramatic improvements in affinity were observed as compared to parental humanized 20C2A and 20C2B, as well as mouse 20C2 antibodies. In particular, low nanomolar to sub-picomolar K_Ds were achieved with a light chain CDR3 of the sequence Xaa₁-Gln-Xaa₂-Thr-Arg-Val-Pro-Leu-Thr (SEQ ID NO:2), wherein Xaa₁ is Phe or Leu, and Xaa₁ is Ala or Thr. Moreover, a comparison between K_D values obtained with BIACORE™ using biotin-Aβ1-20 and bADDL further demonstrates that anti-ADDL antibodies such as Hu20C2 preferentially bind multi-dimensional conformations of ADDLs over monomeric Aβ peptides.

	TABLE 6
	SEQ
	ID	KD (Biotin-Aβ1-20)
Name	Clone	LC-CDR3	NO:	Fab	IgG1 #1	IgG1 #2
20C2-1A	SJ-p2-60	FQASRVPLT	52	91 nM	1.2 nM	—
20C2-2A	SJ-p1-18	FQATRVPLT	53	28 nM	686 pM	2 nM
20C2-3A	SJ-p3-16	FQGSFIPGT	55	—	1.7 nM	—
20C2-5A	SJ-p2-1F	FQSSFVPLT	63	41 nM	912 pM	1.5 nM
20C2-6A	4P1-22	FQSSRVPLT	64	18 nM	544 pM	714 pM
20C2-6B	4P1-22	FQSSRVPLT	64	—	53 pM	—
20C2-7A	SJ-p2-27	LNSTTVPLT	73	128 nM	—	—
20C2-8A	SJ-p2-39	LQTTRVPLT	80	14 nM	140 pM	376 pM
20C2-8B	SJ-p2-39	LQTTRVPLT	80	—	46 pM	64 pM
20C2-9A	SJ-p2-51	VQTSLVPLT	94	36 nM	241 pM	420 pM
20C2-10A	SJ-p3-33	FQGSRLPVS	61	—	84 nM	—
20C2-11A	SJ-p3-6	FQGSLLPLS	98	—	—	—
20C2-12A	4P1-32	LQSSLVPLT	76	617 nM	1.5 nM	—
20C2-13A	4p1-20	LQTSNVPLT	79	94 nM	3 nM	—
20C2-18A	SJ-p1-9M	MQSSFVPLT	86	126 nM	1.8 nM	—
20C2-20A	SJ-p3-15	FQGSLFPPV	58		21 nM
20C2-22A	SJ-p2-66	SQASRVPLT	90		2.3 nM
20C2-23A	4P1-40	HQSSKVPLT	67		649 pM	1.5 nM
20C2-24A	SJ-p2-44	GQTTLVPLT	65		1.9 nM
20C2A		FQGSLVPLT	39		27 nM
20C2B		FQGSLVPLT	39		5.4 nM
Mouse-20C2		FQGSLVPLT	39	83 nM	3.4 nM

	TABLE 7
	SEQ
	ID	KD (bADDL)
Name	Clone	LC-CDR3	NO:	Fab	IgG1 #1	IgG1 #2
20C2-1A	SJ-p2-60	FQASRVPLT	52	85 nM	75 pM	—
20C2-2A	SJ-p1-18	FQATRVPLT	53	28 nM	15 pM	0.3 pM
20C2-3A	SJ-p3-16	FQGSFIPGT	55	—	3.7 nM	—
20C2-5A	SJ-p2-1F	FQSSFVPLT	63	41 nM	317 pM	68 pM
20C2-6A	4P1-22	FQSSRVPLT	64	42 nM	4.3 pM	24 pM
20C2-6B	4P1-22	FQSSRVPLT	64	—	53 pM	—
20C2-7A	SJ-p2-27	LNSTTVPLT	73	435 nM	—	—
20C2-8A	SJ-p2-39	LQTTRVPLT	80	13 nM	3 pM	0.7 pM
20C2-8B	SJ-p2-39	LQTTRVPLT	80	—	13 pM	0.8 pM
20C2-9A	SJ-p2-51	VQTSLVPLT	94	40 nM	—	2 pM
20C2-10A	SJ-p3-33	FQGSRLPVS	61	—	7.7 nM
20C2-11A	SJ-p3-6	FQGSLLPLS	98	—	—	—
20C2-12A	4P1-32	LQSSLVPLT	76	238 nM	15 pM	—
20C2-13A	4p1-20	LQTSNVPLT	79	567 nM	764 pM
20C2-18A	SJ-p1-9M	MQSSFVPLT	86	85 nM	149 pM
20C2-20A	SJ-p3-15	FQGSLFPPV	58		6.9 nM
20C2-22A	SJ-p2-66	SQASRVPLT	90		198 pM
20C2-23A	4P1-40	HQSSKVPLT	67		85 pM	66 pM
20C2-24A	SJ-p2-44	GQTTLVPLT	65		114 pM
20C2A		FQGSLVPLT	39
20C2B		FQGSLVPLT	39
Mouse-20C2		FQGSLVPLT	39	62 nM	4.1 nM

Heavy Chain Maturation. The heavy chain of Hu20C2 was also subjected to optimization by generation of 3 libraries covering the heavy chain-CDR3 (RQLGLRSIDAMDY; SEQ ID NO:99). These libraries were designated 20C2B-39HC₃-1, 20C2B-39HC₃-2, and 20C2B-39HC₃-3 representing heavy chain CDR3 sequences of XXXXXRSIDAMDY (SEQ ID NO:100) and RQLGLRSIXXXXX (SEQ ID NO:101) and RQLGXXXXXAMDY (SEQ ID NO:102), respectively. Biotinylated reverse primers, 20C2HC3-1 (SEQ ID NO:130), 20C2HC3-2 (SEQ ID NO:133), and 20C2HC3-3 (SEQ ID NO:136) were used in combination with forward primer 20C2HC3F (SEQ ID NO:127) to generate the 3 libraries (see FIG. 5B). The libraries contained >10⁸ functional diversity and covered all combinations of amino acids at every position randomized in each set (see Table 8).

TABLE 8
Monoclonal
Antibody	Diversity	Sequence	SEQ ID NO:
Hu20C2		RQLGLRSIDAMDY	99
20C2B-39HC3-1	5.78 × 108	XXXXXRSIDAMDY	100
20C2B-39HC3-2	6.16 × 108	RQLGLRSIXXXXX	101
20C2B-39HC3-3	3.99 × 108	RQLGXXXXXAMDY	102

A total of 18 output plates from 4 rounds of panning were tested in a phage ELISA assay. A total of 1235 hits were found, of which 704 were sequenced. Based on BIACORE™ K_D values of Fab fragments against biotinylated-Aβ1-20 and biotinylated ADDL (bADDL) antigens, as well as single-point BIACORE™3000 analysis (Table 9), a total of 6 Fab clones were converted into IgG1 and IgG2m4 using either CDR grafting or veneering humanization techniques. One of these 6 Fabs, designated 4a-A3, was isolated from library 20C2B-39HC₃-3 and carried 3 amino acid substitutions (RQLGTRGTDAMDY; SEQ ID NO:3) in the middle section of the heavy chain. As is evident from FIG. 2B, this heavy chain CDR3 sequence is that of Hu20C2A3.

	TABLE 9
	bADDL	Aβ 1-20	BIACORE ™		SEQ
	Binding	Binding	3000		ID
Clone	KD (M)	KD (M)	Off-rate	Sequence	NO:
4a-A3	8.23E−11	8.49E−10	6.65E−05	RQLGTRGTDAMDY	3
4b-A7	3.76E−10	1.72E−09	1.23E−04	RQLGKLALDAMDY	142
4b-H11	1.05E−09	1.33E−09	1.32E−04	RQLGRRSVDAMDY	8
20C28B	1.10E−09	1.10E−09	8.76E−05	RQLGLRSIDAMDY	143
4a-F5	1.39E−09	1.33E−09	1.17E−04	RQLGKLKTDAMDY	7
4a-B2	1.92E−09	1.29E−09	1.13E−04	RQLGARKTDAMDY	5
4b-D8	2.23E−09	1.69E−09	1.45E−04	RALSPRSIDAMDY	4
4a-A4	2.67E−09	1.58E−09	1.20E−04	RALSPRSIDAMDY	4
4b-A1	2.87E−09	2.85E−09	1.23E−04	RQLGPRKRDAMDY	6
4a-A7	3.24E−09	2.21E−09	1.55E−04	RQLGQRQTDAMDY	144
4a-B9	3.44E−09	3.54E−09	1.94E−04	RAIQPRSIDAMDY	145
4a-B3	4.17E−09	3.64E−09	1.59E−04	RQLGLRSIDAHTR	146
4a-G10	4.52E−09	2.72E−09	1.78E−04	RQLGQPSVDAMDY	147
4a-E11	4.93E−09	3.48E−09	1.65E−04	RQLGFQSTDAMDY	148
4a-C9	8.43E−09	2.46E−09	1.75E−04	RQLGQAGHDAMDY	149
4a-D5	1.17E−09	3.91E−09	1.74E−04	RQLGDNVADAMDY	150
4a-E10	1.85E−08	3.60E−09	1.39E−04	RQLGFQSTDAMDY	148
4b-D4	1.86E−08	4.87E−09	1.89E−04	RQLGMATPDAMDY	151
4b-B10	6.28E−08	7.43E−09	1.81E−04	RQLGAHWLDAMDY	152
4b-A12	1.54E−07	1.01E−08	1.69E−04	RQLGPEPQDAMDY	153

EXAMPLE 5

Generation of IgG2m4 Antibodies

IgG2m4 antibody derivatives were prepared to decrease Fc receptor engagement, C1q binding, unwanted cytotoxicity or immunocomplex formation while maintaining both the long half-life and pharmacokinetic properties of a typical human antibody. The basic antibody format of IgG2m4 is that of IgG2, which has been shown to possess a superior half-life in experimental models (Zuckier, et al. (1994) Cancer Suppl. 73:794-799). The structure of IgG2 was modified to eliminate C1q binding, through selective incorporation of IgG4 sequences, while maintaining the typical low level of FcγR binding (Canfield and Morrison (1991) J. Exp. Med. 173:1483-1491). This was achieved by using cross-over points wherein sequences of IgG2 and IgG4 were identical, thereby producing an antibody containing natural Fc sequences rather than any artificial mutational sequences. The advantages of using the instant IgG2m4 antibody which exhibits minimal effector-related activity is comparable to the deglycosylated antibody disclosed by Wilcock et al. ((2006) J. Neurosci. 26:5340-6).

The IgG2m4 form of the human antibody constant region was formed by selective incorporation of human IgG4 sequences into a standard human IgG2 constant region, as shown in FIG. 6. Conceptually, IgG2m4 resulted from a pair of chain-swaps within the CH2 domain as shown in FIG. 6. Four single mutations were made corresponding to sequences from IgG4. The Fc residues mutated in IgG2 included His268Gln, Val309Leu, Ala330Ser, and Pro331Ser, which minimized the potential for neoepitopes. The specific IgG4 amino acid residues placed into the IgG2 constant region are shown in Table 10, along with other alternatives from the basic structure.

TABLE 10
Residue			Residue	Alternative
(Kabat	Residue	Residue	in	residue in
numbering)	in IgG2	in IgG4	IgG2m4	IgG2m4	Comment
189	Pro or	Pro	Thr	Pro	Key polymorphism of IgG2;
	Thr*				Pro residue present in
					IGHG*01 allotype and Thr
					residue present in
					IGHG2*02 allotypea,b.
268	His	Gln	Gln	—	Change in the B/C loop
					known to be involved in
					FcγRII bindingc.
309	Val	Leu or	Leu	Val	FcRn binding domain
		Val
330	Ala	Ser	Ser	—	Key residue for C1q
					bindingd; also
					potentially involved in
					binding FcγRII and
					FCγRIIIe.
331	Pro	Ser	Ser	—	Key residue for C1q
					bindingd,f and FcγRI
					bindingg; also
					potentially involved in
					binding FcγRII and
					FcγRIIIe.
397	Met or	Val	Met	Val	Val residue present in
	Val*				IGHG*01 allotype and Met
					residue present in
					IGHG2*02 allotypea.
*Positions marked with an asterisk are subject to allelic variations.
aHougs, et al. (2001) Immunogenetics 52(3-4): 242-8.
bWO 97/11971.
cMedgyesi, et al. (2004) Eur. J. Immunol. 34: 1127-1135.
dTao, et al. (1991) J. Exp. Med. 173: 1025-1028.
eArmour, et al. (1999) Eur. J. Immunol. 29: 2613.
fXu, et al. (1994) J. Biol. Chem. 269: 3469-3474.
gCanfield and Morrison (1991) J. Exp. Med. 173: 1483.

Human IgG1/kappa and IgG2m4/kappa versions of humanized Hu20C2 and Hu20C2A3 antibodies were constructed. The complete amino acid sequence of the light and heavy chain Hu20C2A3 IgG2m4 antibody is shown in FIGS. 7A and 7C.

EXAMPLE 6

Binding Affinity and Specificity of Humanized Anti-ADDL Antibodies

Affinity maturation was carried out to improve affinity and improve preferential binding to ADDL. To evaluate ADDL binding affinity of the humanized antibodies, BIACORE™ and titration ELISAs were conducted as disclosed herein. Briefly, Streptavidin-coated, 96-well microtiter plates (Sigma, St. Louis, Mo.) were coated with 10% biotinylated ADDL antigen (1 μM). A series of 2-fold dilutions of purified antibody, starting at 500 ng/mL was added to the ADDL captured plates and the plates were incubated for 2 hours at 25° C. After washing five times with PBS solution using a plate washer (Bio-Tek, Winooski, Va.), polyclonal goat anti-human kappa light chain antibody (Biomeda, Foster City, Cailf.) was added at a 1/2000 dilution in 3% non-fat milk blocker and incubated at room temperature for 1 hour. A rabbit anti-goat IgG (H+L) HRP-conjugated (Bethyl Laboratories, Inc., Montgomery, Tex.) detection antibody was then added at a 1/2000 dilution in blocking solution and incubated for 1 hour at room temperature. After washing with PBS, HRP substrate, 3,3′,5′5-tetramethylbenzidine (ready-to-use TMB; Sigma, St. Louis, Mo.) was added and the reaction was stopped after 10 minutes with 0.5 N H₂SO₄. Absorbance at wavelength of 450 nm was read in a plate reader (model VICTOR V; Perkin Elmer, Boston, Mass.) and data were processed using EXCEL® work sheet. Assay variations between plates were estimated within 20%.

The K_D of Fab clone A3, as measured by BIACORE™, was 849 pM against biotinylated-Aβ1-20. The K_D of the same Fab clone was 82 pM against bADDLs, indicating that the A3 Fab demonstrated preferential binding to ADDLs. When the clone was humanized by veneering and converted to a full IgG1 molecule or full IgG2m4 molecule (i.e., Hu20C2A3), the K_D values against Aβ1-20 and ADDL were below the reliable detection limit of the BIACORE™ instrument, indicating a significant improvement of Hu20C2A3's binding equilibrium constant against Aβ1-20 and ADDL as compared to Hu20C2.

Hu20C2A3, which is the veneered version of clone A3 with an IgG2m4 isotype, was expressed in both CHO and Pichia. The two sources of Hu20C2A3 were evaluated for their ability to interact with Aβ monomer and ADDLs by ELISA. As shown in FIG. 8, Hu20C2A3 produced in either CHO (FIG. 8A) or Pichia (FIG. 8B) showed preferential ADDL binding versus Aβ40 monomer binding (6-fold). Binding constants (IgGk₅₀ values) as determined from these curves yielded values of 64 pM and 376 pM for ADDLs and Aβ monomer, respectively for Hu20C2A3 produced in Pichia and 58 pM and 361 pM for ADDLs and Aβ monomer, respectively for Hu20C2A3 produced in CHO cells.

EXAMPLE 7

Inhibition of ADDL Binding to Neurons Using Humanized Anti-ADDL Antibodies

The humanized anti-ADDL antibodies were further evaluated for their ability to block ADDL binding to primary hippocampal neurons using the methods disclosed herein. Hu20C2A3 antibody, or PBS as a control, was mixed at various molar ratios with bADDLs and incubated for one hour at 37° C. on a slow rotator. After the preincubation, the antibody/bADDL preparations were added to primary neuron cultures and incubated for an additional hour at 37° C. At the end of the incubation period, the bADDLs/antibody mixture was removed and the plates washed six times with media. The cells were then fixed in 4% paraformaldehyde for ten minutes at room temperature, the solution removed, fresh fixative added, and the cells fixed for an additional ten minutes. The cells were permeabilized with 4% paraformaldehyde containing 0.1% TRITON™ X-100 (2 times, each for ten minutes at room temperature), washed six times in PBS and then treated with 10% BSA in PBS for one hour at 37° C. Alkaline phosphatase-conjugated streptavidin (1:1,500 in 1% BSA; Molecular Probes, Eugene, OR) was then added to the cells for one hour at room temperature. The cells were rinsed six times with PBS, the alkaline phosphatase substrate (CDP-STAR® with SAPPHIRE-II™; Applied Biosystems, Foster City, Cailf.) added to the cells and incubated for thirty minutes prior to determining the luminescence on a LJL Luminometer (Analyst AD; LJL Biosystems, Sunnyvale, Cailf.). In this analysis, Hu20C2A3 was found to effectively inhibit bADDL binding to neurons at a sub-stoichiometric antibody to peptide ratio (EC₅₀=0.16; FIG. 9).

The inhibition of bADDL binding indicated that Hu20C2A3 interacts with ADDLs in a biologically relevant manner. To demonstrate that Hu20C2A3 interacts with ADDLs relevant to the human Alzheimer's Disease condition, the ability of biotinylated Hu20C2A3 to immuno-label Aβ containing plaques in human Alzheimer's Disease brain tissue was evaluated. Immunohistochemical localization showed avid labeling of Aβ in both dense core and diffuse plaques in human Alzheimer's Disease brain tissue. The specificity of this binding was demonstrated by a loss of immunoreactivity following pre-incubation with increasing ADDL:antibody amounts. Similar to Hu20C2, Hu20C2A3, efficiently labels both dense core and diffuse Aβ deposits.

EXAMPLE 8

Thermal Stability of Hu20C2A3

An evaluation of the protein stability of Hu20C2A3 was assessed using SEC-HPLC, fluorescent thermal melt analysis, and particle size analysis. Fluorescent thermal melt analysis indicated that Fc and Fab unfolding transitions occurred at approximately 70° C. and 80° C., respectively, consistent with acceptable inherent protein stability (FIG. 10).

EXAMPLE 9

In Vivo Pharmacodynamic and Efficacy Analysis

The prior art indicates that systemic injection of monoclonal anti-Aβ antibodies can increase plasma levels of Aβ acutely, whereas measurable lowering of brain Aβ requires chronic administration. It has been suggested that passive immunization in species with measurable Aβ results in an elevation of plasma Aβ due to a change in the equilibrium of Aβ between brain and peripheral compartments. This “peripheral sink” ultimately leads to lowering of brain Aβ. However, cognitive improvement has been observed in animals following acute antibody administration prior to notable changes in brain Aβ, indicating that changes in brain Aβ may occur in some form prior to a time point where these changes can be measured using known techniques. Alternately, elevations of plasma Aβ could be explained by a stabilization of peripheral Aβ following administration of antibody. Regardless of the interpretation, it has been established that early plasma elevations are a prerequisite for subsequent lowering of brain Aβ in animal models. Thus, the effect of Hu20C2A3 antibodies on plasma Aβ elevations were utilized as an indicator of target engagement. Following infusion of Hu20C2A3 at doses of 30, 100 and 300 μg/mouse IV, significant and robust increases in plasma Aβx-40 were observed relative to the non-relevant antibody (8B4) control group 4 hours post-injection (FIG. 11). The observed increases in plasma Aβx-40 for CHO-derived material were 491% (30 μg, p>0.001), 826% (100 μg, p<0.001), and 755% (300 μg, p<0.001) of the 8B4 levels. Similarly, the increases in plasma Aβx-40 for Pichia-derived material were 395% (30 μg, p>0.001), 729% (100 μg, p<0.001), and 838% (300 μg, p<0.001) of the 8B4 levels.

Claims

1. An isolated antibody, or fragment thereof, capable of differentially recognizing a multi-dimensional conformation of one or more Aβ-derived diffusible ligands, wherein said antibody comprises a complementary determining region (CDR) of Arg-Xaa1-Leu-Xaa2-Xaa3-Xaa4-Xaa5-Xaa6-Asp-Ala-Met-Asp-Tyr (SEQ ID NO:9), wherein Xaa1 is Gln or Ala; Xaa2 is Ser or Gly; Xaa3 is Pro, Ala, Lys, Arg, or Thr; Xaa4 is Lys or Arg; Xaa5 is Gly, Ser, or Lys; Xaa6 is Val, Thr, Ile or Arg.

2. The isolated antibody of claim 1, wherein the heavy and light chain variable regions of said antibody respectively comprise the amino acid sequence set forth in SEQ ID NO:108 and SEQ ID NO:112.

3. The isolated antibody of claim 1, wherein the heavy and light chain of said antibody respectively comprise the amino acid sequence set forth in SEQ ID NO:138 and SEQ ID NO:140.

4. A pharmaceutical composition comprising an isolated antibody of claim 1, or fragment thereof, capable of differentially recognizing a multi-dimensional conformation of one or more Aβ-derived diffusible ligands in admixture with a pharmaceutically acceptable carrier.

5. A method for preventing binding of Aβ-derived diffusible ligands to a neuron comprising contacting the neuron with the antibody of claim 1 so that binding of Aβ-derived diffusible ligands to the neuron is prevented.

6. A method for inhibiting assembly of Aβ-derived diffusible ligands comprising contacting a sample containing amyloid β1-42 peptides with the antibody of claim 1 thereby inhibiting assembly of Aβ-derived diffusible ligands.

7. A method for blocking the phosphorylation of tau protein at Ser202/Thr205 comprising contacting a sample containing a tau protein with the antibody of claim 1 thereby blocking the phosphorylation of tau protein at Ser202/Thr205.

8. A method for prophylactically or therapeutically treating a disease associated with Aβ-derived diffusible ligands comprising administering an effective amount of the pharmaceutical composition of claim 4.

9. A method for identifying a therapeutic agent that prevents the binding of Aβ-derived diffusible ligands to a neuron comprising contacting a neuron with Aβ-derived diffusible ligands in the presence of an agent and using an antibody of claim 1 to determine binding of Aβ-derived diffusible ligands to the neuron in the presence of the agent.

10. A method for detecting Aβ-derived diffusible ligands in a sample comprising contacting a sample with the antibody of claim 1 so that Aβ-derived diffusible ligands are detected.

11. A method for diagnosing a disease associated with Aβ-derived diffusible ligands comprising contacting a sample with the antibody of claim 1 so that a disease associated with Aβ-derived diffusible ligands is diagnosed.

12. A kit for detecting Aβ-derived diffusible ligands comprising the isolated antibody, or fragment thereof, of claim 1.