AU2012201856A1 - Anti-Abeta globulomer antibodies, antigen-binding moieties thereof, corresponding hybridomas, nucleic acids, vectors, host cells, methods of producing said antibodies, compositions comprising said antibodies, uses of said antibodies and methods of using said antibodies - Google Patents

Abstract

ANTI-As GLOBULOMER ANTIBODIES, ANTIGEN-BINDING MOIETIES THEREOF, CORRESPONDING HYBRIDOMAS, NUCLEIC ACIDS, VECTORS, HOST CELLS, METHODS OF PRODUCING SAID ANTIBODIES, COMPOSITIONS COMPRISING SAID ANTIBODIES, USES OF SAID ANTIBODIES AND METHODS OF USING SAID ANTIBODIES Abstract Anti-A globulomer antibodies, antigen-binding moieties thereof, corresponding hybridomas, nucleic acids, vectors, host cells, methods of producing said antibodies, compositions comprising said antibodies, uses of said antibodies and methods of using said antibodies. The present invention relates to anti-Ap globulomer antibodies having a binding affinity to AD(20 42) globulomer that is greater than the binding affinity of the antibody to A3 (1-42) globulomer, antigen-binding moieties thereof, hybridomas producing said antibodies, nucleic acids encoding said antibodies, vectors comprising said nucleic acids, host cells comprising said vectors, methods of producing said antibodies, compositions comprising said antibodies, therapeutic and diagnostic uses of said antibodies and corresponding methods relating to Alzheimer's disease and other amyloidoses. (6090564_1):GGG

Description

S&F Ref: 856665D1 AUSTRALIA PATENTS ACT 1990 COMPLETE SPECIFICATION FOR A STANDARD PATENT Name and Address Abbott Laboratories, of 100 Abbott Park Road, Abbott of Applicants: Park, Illinois, 60064, United States of America Abbott GmbH & Co. KG, of Max-Planck-Ring 2, 65205, Wiesbaden, Germany Actual Inventor(s): Andreas Striebinger Boris Labkovsky Ulrich Ebert Patrick Keller Stefan Barghorn Heinz Hillen Address for Service: Spruson & Ferguson St Martins Tower Level 35 31 Market Street Sydney NSW 2000 (CCN 3710000177) Invention Title: Anti-AGl globulomer antibodies, antigen-binding moieties thereof, corresponding hybridomas, nucleic acids, vectors, host cells, methods of producing said antibodies, compositions comprising said antibodies, uses of said antibodies and methods of using said antibodies, uses of said antibodies and methods of using said antibodies The following statement is a full description of this invention, including the best method of performing it known to me/us: 5845c(6161608_1) 1 Anti-An Globulomer Antibodies, Antigen-Binding Moieties Thereof, Corresponding Hy bridomas, Nucleic Acids, Vectors, Host Cells, Methods Of Producing Said Antibodies, Compositions Comprising Said Antibodies, Uses Of Said Antibodies And Methods Of Using Said Antibodies 5 The present invention relates to anti-As globulomer antibodies, antigen-binding moieties thereof, hybridomas producing said antibodies, nucleic acids encoding said antibodies, vectors comprising said nucleic acids, host cells comprising said vectors, methods of producing said antibodies, compositions comprising said antibodies, therapeutic and diagnostic uses of said 10 antibodies and corresponding methods relating to Alzheimer's disease and other amyloidoses. In 1907 the physician Alois Alzheimer first described the neuropathological features of a form of dementia subsequently named in his honour (Alzheimer 1907). Alzheimer's disease (AD) is the most frequent cause for dementia among the aged with an incidence of about 10% of the 15 population above 65 years. With increasing age, the probability of disease also rises. Globally, there are about 15 million people affected and further increases in life expectancy are ex pected to increase the number of diseased people to about threefold over the next decades. From a molecular point of view Alzheimer's disease (AD) is characterized by a deposit of ab 20 normally aggregated proteins. In the case of extra-cellular amyloid plaques these deposits con sist mostly of amyloid-p-peptide filaments, in the case of the intracellular neurofibrillary tangles (NFTs) of the tau protein. The amyloid 0 (AP) peptide arises from the p-amyloid precursor pro tein by proteolytic cleavage. This cleavage is effected by the cooperative activity of several proteases named a-, p- and y-secretase. Cleavage leads to a number of specific fragments of 25 differing length. The amyloid plaques consist mostly of peptides with a length of 40 or 42 amino acids (AP40, Ap42). The dominant cleavage product is Ap40; however, AS42 has a much stronger toxic effect. Cerebral amyloid deposits and cognitive impairments very similar to those observed in Alz 30 heimer's disease are also hallmarks of Down's syndrome (trisomy 21), which occurs at a fre quency of about 1 in 800 births. The amyloid cascade hypothesis of Hardy and Higgins postulated that increased production of AP(1-42) would lead to the formation of protofibrils and fibrils, the principal components of AP 35 plaques, these fibrils being responsible for the symptoms of Alzheimer's disease. Despite the poor correlation between severity of dementia and As plaque burden deposited this hypothesis was favoured until recently. The discovery of soluble As forms in AD brains, which correlates better with AD symptoms than plaque load does, has led to a revised amyloid-cascade hypothesis. 40 2 Active immunization with AO peptides leads to a reduction in the formation as well as to partial dissolution of existing plaques. At the same time it leads to alleviation of cognitive defects in APP transgenic mouse models. 5 For passive immunization with antibodies directed to As peptides a reduction of an As plaque burden was also found. The results of a phase Ila trial (ELAN Corporation Plc, South San Francisco, CA, USA and Dublin, UK) of active immunization with AN-1 792 (AP(1-42) peptide in fibrillary condition of 10 aggregation) suggest that immunotherapy directed to AP peptide was successful. In a sub group of 30 patients the progression of disease was significantly reduced in patients with posi tive anti-Ap antibody titer, measured by MMSE and DAD index. However, this study was stopped because of serious side effects in form of a meningoencephalitis (Bennett and Holtz man, 2005, Neurology, 64, 10-12). 15 Meningoencephalitis was characterized by neuroinflammation and infiltration of T-cells into the brain. Presumably, this was due to a T-cell immune response induced by injection of Ap(1-42) as antigen. Such an immune response is not to be expected after passive immunization. To date, there are no clinical data with reference to this available yet. However, with reference to 20 such a passive approach to immunization concerns about the side effect profile were voiced because of preclinical studies in very old APP23 mice which received an antibody directed against an N-terminal epitope of AP(1-42) once a week over 5 months. These mice showed an increase in the number and severity of microhaemorrhages compared to control animals treated with saline (Pfeifer et al., 2002, Science, 298, 1379). A comparable increase in micro 25 haemorrhages was also described in very old (> 24 months) Tg2576 and PDAPP mice (Racke et al., 2005, J Neurosci, 25, 629-636; Wilcock et al. 2004, J. Neuroinflammation, 1(1):24; De Mattos et al., 2004, Neurobiol. Aging 25(S2):577). In both mouse strains antibody injection led to a significant increase in microhaemorrhages. In contrast, an antibody directed against the central region of the Ap(1-42) peptide did not induce microhaemorrhages (de Mattos et al., 30 supra). The lack of inducing microhaemorrhages was associated with an antibody treatment which did not bind to aggregated Ap peptide in the form of CAA (Racke et al., J Neurosci, 25, 629-636). But, the exact mechanism leading to microhaemorrhages in mice transgenic for APP has not been understood. Presumably, cerebral amyloid angiopathy (CAA) induces or at least aggravates cerebral haemorrhages. CAA is present in nearly every Alzheimer's disease brain 35 and about 20% of the cases are regared as "severe CAA". Passive immunization should there fore aim at avoiding microhaemorrhages by selecting an antibody which recognizes the central or the carboxy terminal region of the Ap peptide. W02004/067561 describes stable Ap(1-42) oligomers (AP(1-42) globulomers) and antibodies 40 directed specifically against the globulomers. Digestion with unspecific proteases shows that the AP globulomer may be digested beginning with the hydrophilic N-terminus protruding from the globular core structure (Barghorn et al., 2005, J Neurochem, 95, 834-847). Such N terminal truncated As globulomers (Ap(12-42) and Ap(20-42) globulomers) represent the basic structural unit of this oligomeric AP. They are a very potent antigen for active immunization of 5 rabbits and mice leading to high antibody titers (W02004/067561). The putative pathological role of N-terminally truncated As forms in vivo has been suggested by several recent reports of their existence in AD brains (Sergeant et al., 2003, J Neurochem, 85, 1581-1591; Thai et al., 1999, J Neuropathol. Exp Neurol, 58, 210-216). During in vivo digestion certain proteases found in brain, e.g. neprilysin (NEP 24.11) or insulin degrading enzyme (IDE), may be involved 10 (Selkoe, 2001, Neuron, 32, 177-180). It was an object of the present invention to provide antibodies directed against As globulomers which improve the cognitive performance of a patient in immunotherapy while at the same time reacting only with a small portion of the entire amount of AP peptide in brain. This is expected 15 to prevent a substantial disturbance of cerebral As balance and lead to less side effects. (For instance, a therapeutically questionable reduction of brain volume has been observed in the study of active immunization with As peptides in fibrillary condition of aggregation (ELAN trial with AN1792). Moreover, in this trial severe side effects in form of a meningoencephalitis were observed. 20 The present invention solves this problem by providing globulomer-specific antibodies pos sessing high affinity for truncated forms of As globulomers. These antibodies are capable of discriminating not only other forms of AP peptides, particularly monomers and fibrils, but also untruncated forms of As globulomers. 25 Thus, the present invention relates to an antibody having a binding affinity to an AP(20-42) globulomer that is greater than the binding affinity of this antibody to an Ap(1 -42) globulomer. Further, the present invention relates to an antibody having a binding affinity to an AO(20-42) 30 globulomer that is greater than the binding affinity of this antibody to an AO(12-42) globulomer. According to a particular embodiment, the invention thus relates to antibodies having a binding affinity to the Ap(20-42) globulomer that is greater than the binding affinity of the antibody to both the Ap1(1-42) globulomer and the Ap(12-42) globulomer. 35 The term "Ap(X-Y)" here refers to the amino acid sequence from amino acid position X to amino acid position Y of the human amyloid P protein including both X and Y, in particular to the amino acid sequence from amino acid position X to amino acid position Y of the amino acid sequence DAEFRHDSGY EVHHQKLVFF AEDVGSNKGA IIGLMVGGVV IAT (corresponding 40 to amino acid positions 1 to 43) or any of its naturally occurring variants, in particular those 4 with at least one mutation selected from the group consisting of A2T, H6R, D7N, A21G ("Flem ish"), E22G ("Arctic"), E22Q ("Dutch"), E22K ("Italian"), D23N ("Iowa"), A42T and A42V wherein the numbers are relative to the start of the AO peptide, including both position X and position Y or a sequence with up to three additional amino acid substitutions none of which 5 may prevent globulomer formation, preferably with no additional amino acid substitutions in the portion from amino acid 12 or X, whichever number is higher, to amino acid 42 or Y, whichever number is lower, more preferably with no additional amino acid substitutions in the portion from amino acid 20 or X, whichever number is higher, to amino acid 42 or Y, whichever number is lower, and most preferably with no additional amino acid substitutions in the portion from 10 amino acid 20 or X, whichever number is higher, to amino acid 40 or Y, whichever number is lower, an "additional" amino acid substation herein being any deviation from the canonical se quence that is not found in nature. More specifically, the term "AP(1-42)" here refers to the amino acid sequence from amino acid 15 position 1 to amino acid position 42 of the human amyloid p protein including both 1 and 42, in particular to the amino acid sequence DAEFRHDSGY EVHHQKLVFF AEDVGSNKGA IIGLMVGGVV IA or any of its naturally occurring variants, in particular those with at least one mutation selected from the group consisting of A2T, H6R, D7N, A21G ("Flemish"), E22G ("Arc tic"), E220 ("Dutch"), E22K ("Italian"), D23N ("Iowa"), A42T and A42V wherein the numbers 20 are relative to the start of the AP peptide, including both 1 and 42 or a sequence with up to three additional amino acid substitutions none of which may prevent globulomer formation, preferably with no additional amino acid substitutions in the portion from amino acid 20 to amino acid 42. Likewise, the term "AP(1-40)" here refers to the amino acid sequence from amino acid position 1 to amino acid position 40 of the human amyloid 0 protein including both 25 1 and 40, in particular to the amino acid sequence DAEFRHDSGY EVHHQKLVFF AEDVGSNKGA IIGLMVGGVV or any of its naturally occurring variants, in particular those with at least one mutation selected from the group consisting of A2T, H6R, D7N, A21 G ("Flemish"), E22G ("Arctic"), E22Q ("Dutch"), E22K ("Italian"), and D23N ("Iowa") wherein the numbers are relative to the start of the AP peptide, including both 1 and 40 or a sequence with up to three 30 additional amino acid substitutions none of which may prevent globulomer formation, prefera bly with no additional amino acid substitutions in the portion from amino acid 20 to amino acid 40. More specifically, the term "Ap(12-42)" here refers to the amino acid sequence from amino 35 acid position 12 to amino acid position 42 of the human amyloid 5 protein including both 12 and 42, in particular to the amino acid sequence VHHQKLVFF AEDVGSNKGA IIGLMVGGVV IA or any of its naturally occurring variants, in particular those with at least one mutation se lected from the group consisting of A21 G ("Flemish"), E22G ("Arctic"), E22Q ("Dutch"), E22K ("Italian"), D23N ("Iowa"), A42T and A42V wherein the numbers are relative to the start of the 40 AP peptide, including both 12 and 42 or a sequence with up to three additional amino acid 5 substitutions none of which may prevent globulomer formation, preferably with no additional amino acid substitutions in the portion from amino acid 20 to amino acid 42. More specifically, the term "Ap(20-42)" here refers to the amino acid sequence from amino 5 acid position 20 to amino acid position 42 of the human amyloid 0 protein including both 20 and 42, in particular to the amino acid sequence F AEDVGSNKGA IIGLMVGGVV IA or any of its naturally occurring variants, in particular those with at least one mutation selected from the group consisting of A21G ("Flemish"), E22G ("Arctic"), E22Q ("Dutch"), E22K ("Italian"), D23N ("Iowa"), A42T and A42V wherein the numbers are relative to the start of the As peptide, in 10 cluding both 20 and 42 or a sequence with up to three additional amino acid substitutions none of which may prevent globulomer formation, preferably without any additional amino acid sub stitutions. The term "Ap(X-Y) globulomer" (Ap(X-Y) globular oligomer) here refers to a soluble, globular, 15 non-covalent association of AP(X-Y) peptides as defined above, possessing homogeneity and distinct physical characteristics. According to one aspect, AP(X-Y) globulomers are stable, non-fibrillar, oligomeric assemblies of AP(X-Y) peptides which are obtainable by incubation with anionic detergents. In contrast to monomer and fibrils, these globulomers are character ized by defined assembly numbers of subunits (e.g. early assembly forms, n=4-6, "oligomers 20 A", and late assembly forms, n=12-14, "oligomers B", as described in W02004/067561). The globulomers have a 3-dimensional globular type structure ("molten globule", see Barghorn et al., 2005, J Neurochem, 95, 834-847). They may be further characterized by one or more of the following features: - cleavability of N-terminal amino acids X-23 with promiscuous proteases (such as thermolysin 25 or endoproteinase GluC) yielding truncated forms of globulomers; - non-accessibility of C-terminal amino acids 24-Y with promiscuous proteases and antibodies; - truncated forms of these globulomers maintain the 3-dimensional core structure of said globulomers with a better accessibility of the core epitope Ap(20-Y) in its globulomer conforma tion. 30 According to the invention and in particular for the purpose of assessing the binding affinities of the antibodies of the present invention, the term "Ap(X-Y) globulomer" here refers in particular to a product which is obtainable by a process as described in WO 2004/067561, which is in corporated herein by reference. 35 Said process comprises unfolding a natural, recombinant or synthetic Ap(X-Y) peptide or a derivative thereof; exposing the at least partially unfolded Ap(X-Y) peptide or derivative thereof to a detergent, reducing the detergent action and continuing incubation. 40 For the purpose of unfolding the peptide, hydrogen bond-breaking agents such as, for exam- 6 pie, hexafluoroisopropanol (HFIP) may be allowed to act on the protein. Times of action of a few minutes, for example about 10 to 60 minutes, are sufficient when the temperature of action is from about 20 to 50*C and in particular about 35 to 400C. Subsequent dissolution of the residue evaporated to dryness, preferably in concentrated form, in suitable organic solvents 5 miscible with aqueous buffers, such as, for example, dimethyl sulfoxide (DMSO), results in a suspension of the at least partially unfolded peptide or derivative thereof, which can be used subsequently. If required, the stock suspension may be stored at low temperature, for example at about -20*C, for an interim period. 10 Alternatively, the peptide or the derivative thereof may be taken up in slightly acidic, preferably aqueous, solution, for example an about 10 mM aqueous HCI solution. After an incubation time of usually a few minutes, insoluble components are removed by centrifugation. A few minutes at 10000 g is expedient. These method steps are preferably carried out at room temperature, i.e. a temperature in the range from 20 to 30*C. The supernatant obtained after centrifugation 15 contains the AP(X-Y) peptide or the derivative thereof and may be stored at low temperature, for example at about -20*C, for an interim period. The following exposure to a detergent relates to the oligomerization of the peptide or the de rivative thereof to give an intermediate type of oligomers (in WO 2004/067561 referred to as 20 oligomers A). For this purpose, a detergent is allowed to act on the at least partially unfolded peptide or derivative thereof until sufficient intermediate oligomer has been produced. Preference is given to using ionic detergents, in particular anionic detergents. 25 According to a particular embodiment, a detergent of the formula (1): R-X, is used, in which the radical R is unbranched or branched alkyl having from 6 to 20 and preferably 10 to 14 car bon atoms or unbranched or branched alkenyl having from 6 to 20 and preferably 10 to 14 30 carbon atoms, the radical X is an acidic group or salt thereof, with X being preferably selected from among -COO-M', -SO3-M', and especially -OSO3-M* and M' is a hydrogen cation or an inorganic or organic cation preferably selected from alkali metal and alkaline earth metal cations and ammonium cations. 35 Advantageous are detergents of the formula (1), in which R is unbranched alkyl of which alk-1 yl radicals must be mentioned in particular. Particular preference is given to sodium dodecyl sulfate (SDS). Lauric acid and oleic acid can also be used advantageously. The sodium salt of the detergent lauroylsarcosin (also known as sarkosyl NL-30 or Gardol*) is also particularly 7 advantageous. The time of detergent action in particular depends on whether - and if yes, to what extent - the peptide or the derivative thereof subjected to oligomerization has unfolded. If, according to the 5 unfolding step, the peptide or derivative thereof has been treated beforehand with a hydrogen bond-breaking agent, i.e. in particular with hexafluoroisopropanol, times of action in the range of a few hours, advantageously from about 1 to 20 and in particular from about 2 to 10 hours, are sufficient when the temperature of action is about 20 to 50*C and in particular about 35 to 400C. If a less unfolded or an essentially not unfolded peptide or derivative thereof is the start 10 ing point, correspondingly longer times of action are expedient. If the peptide or the derivative thereof has been pretreated, for example, according to the procedure indicated above as an alternative to the HFIP treatment or said peptide or derivative thereof is directly subjected to oligomerization, times of action in the range from about 5 to 30 hours and in particular from about 10 to 20 hours are sufficient when the temperature of action is about 20 to 50*C and in 15 particular about 35 to 40*C. After incubation, insoluble components are advantageously re moved by centrifugation. A few minutes at 10000 g is expedient. The detergent concentration to be chosen depends on the detergent used. If SDS is used, a concentration in the range from 0.01 to 1% by weight, preferably from 0.05 to 0.5% by weight, 20 for example of about 0.2% by weight, proves expedient. If lauric acid or oleic acid are used, somewhat higher concentrations are expedient, for example in a range from 0.05 to 2% by weight, preferably from 0.1 to 0.5% by weight, for example of about 0.5% by weight. The detergent action should take place at a salt concentration approximately in the physiologi 25 cal range. Thus, in particular NaCl concentrations in the range from 50 to 500 mM, preferably from 100 to 200 mM and particularly at about 140 mM are expedient. The subsequent reduction of the detergent action and continuation of incubation relates to a further oligomerization to give the Ap(X-Y) globulomer of the invention (in WO 2004/067561 30 referred to as oligomers B). Since the composition obtained from the preceding step regularly contains detergent and a salt concentration in the physiological range it is then expedient to reduce detergent action and, preferably, also the salt concentration. This may be carried out by reducing the concentration of detergent and salt, for example, by diluting, expediently with wa ter or a buffer of lower salt concentration, for example Tris-HCI, pH 7.3. Dilution factors in the 35 range from about 2 to 10, advantageously in the range from about 3 to 8 and in particular of about 4, have proved suitable. The reduction in detergent action may also be achieved by add ing substances which can neutralize said detergent action. Examples of these include sub stances capable of complexing the detergents, like substances capable of stabilizing cells in the course of purification and extraction measures, for example particular EO/PO block co 40 polymers, in particular the block copolymer under the trade name Pluronic* F 68. Alkoxylated 8 and, in particular, ethoxylated alkyl phenols such as the ethoxylated t-octylphenols of the Tri ton* X series, in particular Triton* X100, 3-(3-cholamidopropyldimethylammonio)-1 propanesulfonate (CHAPS*) or alkoxylated and, in particular, ethoxylated sorbitan fatty esters such as those of the Tween* series, in particular Tween* 20, in concentration ranges around 5 or above the particular critical micelle concentration, may be equally used. Subsequently, the solution is incubated until sufficient AP(X-Y) globulomer of the invention has been produced. Times of action in the range of several hours, preferably in the range from about 10 to 30 hours and in particular in the range from about 15 to 25 hours, are sufficient 10 when the temperature of action is about 20 to 500C and in particular about 35 to 40'C. The solution may then be concentrated and possible residues may be removed by centrifugation. Here too, a few minutes at 10000 g proves expedient. The supernatant obtained after centrifu gation contains an A@(X-Y) globulomer of the invention. 15 An Ap(X-Y) globulomer of the invention can be finally recovered in a manner known per se, e. g. by ultrafiltration, dialysis, precipitation or centrifugation. It is further preferred if electrophoretic separation of the Ap(X-Y) globulomers under denaturing conditions, e. g. by SDS-PAGE, produces a double band (e. g. with an apparent molecular 20 weight of 38 / 48 kDa for AP(1-42)), and especially preferred if upon glutardialdehyde treat ment of the globulomers before separation these two bands are merged into one. It is also preferred if size exclusion chromatography of the globulomers results in a single peak (e. g. corresponding to a molecular weight of approximately 100 kDa for Apf(1-42) globulomer or of approximately 60 kDa for glutardialdehyde cross-linked AP(1-42) globulomer), respectively. 25 Starting out from AP(1-42) peptide, AP(12-42) peptide, and AP(20-42) peptide said processes are in particular suitable for obtaining Ap(1-42) globulomers, Ap(12-42) globulomers, and AP(20-42) globulomers. 30 In a particular embodiment of the invention, AD(X-Y) globulomers wherein X is selected from the group consisting of the numbers 2 .. 24 and Y is as defined above, are those which are obtainable by truncating AP(1-Y) globulomers into shorter forms wherein X is selected from the group consisting of the numbers 2 .. 24, with X preferably being 20 or 12, and Y is as defined above, which can be achieved by treatment with appropriate proteases. For instance, an 35 AP(20-42) globulomer can be obtained by subjecting an A@(1-42) globulomer to thermolysin proteolysis, and an AP(12-42) globulomer can be obtained by subjecting an AP(1-42) globu lomer to endoproteinase GluC proteolysis. When the desired degree of proteolysis is reached, the protease is inactivated in a generally known manner. The resulting globulomers may then be isolated following the procedures already described herein and, if required, processed fur 40 ther by further work-up and purification steps. A detailed description of said processes is dis- 9 closed in WO 2004/067561, which is incorporated herein by reference. For the purposes of the present invention, an Ap(1-42) globulomer is in particular the Ap(1-42) globulomer as described in example 1a herein; an Ap(20-42) globulomer is in particular the 5 Ap(20-42) globulomer as described in examples 1c herein, and an Ap(12-42) globulomer is in particular the Ap(12-42) globulomer as described in examples 1d herein. Preferably, the globulomer shows affinity to neuronal cells. Preferably, the globulomer also exhibits neuromodulating effects. 10 According to another aspect of the invention, the globulomer consists of 11 to 16, and most preferably, of 12 to 14 Ap(X-Y) peptides. According to another aspect of the invention, the term "Ap(X-Y) globulomer" here refers to a 1 5 globulomer consisting essentially of Ap(X-Y) subunits, where it is preferred if on average at least 11 of 12 subunits are of the AP(X-Y) type, more preferred if less than 10% of the globu lomers comprise any non-AP(X-Y) peptides, and most preferred if the content of non-Ap(X-Y) peptides is below the detection threshold. 20 More specifically, the term "Ap(1-42) globulomer" here refers to a globulomer consisting essen tially of Ap(1-42) units as defined above; the term "Ap(12-42) globulomer" here refers to a globulomer consisting essentially of AP(1 2-42) units as defined above; and the term "A(20 42) globulomer" here refers to a globulomer consisting essentially of Ap(20-42) units as de fined above. 25 The term "cross-linked AP(X-Y) globulomer" here refers to a molecule obtainable from an Ap(X-Y) globulomer as described above by cross-linking, preferably chemically cross-linking, more preferably aldehyde cross-linking, most preferably glutardialdehyde cross-linking of the constituent units of the globulomer. In another aspect of the invention, a cross-linked globu 30 lomer is essentially a globulomer in which the units are at least partially joined by covalent bonds, rather than being held together by non-covalent interactions only. For the purposes of the present invention, a cross-linked AO(1-42) globulomer is in particular the cross-linked Ap(1 42) oligomer as described in example 1b herein. 35 The term "Ap(X-Y) globulomer derivative" here refers in particular to a globulomer that is la belled by being covalently linked to a group that facilitates detection, preferably a fluorophore, e. g. fluorescein isothiocyanate, phycoerythrin, Aequorea victoria fluorescent protein, Dic tyosoma fluorescent protein or any combination or fluorescence-active derivative thereof; a chromophore; a chemoluminophore, e. g. luciferase, preferably Photinus pyralis luciferase, 40 Vibrio fischeri luciferase, or any combination or chemoluminescence-active derivative thereof; 10 an enzymatically active group, e. g. peroxidase, e. g. horseradish peroxidase, or any enzy matically active derivative thereof; an electron-dense group, e. g. a heavy metal containing group, e.g. a gold containing group; a hapten, e. g. a phenol derived hapten; a strongly anti genic structure, e. g. peptide sequence predicted to be antigenic, e. g. predicted to be anti 5 genic by the algorithm of Kolaskar and Tongaonkar; an aptamer for another molecule; a che lating group, e. g. hexahistidinyl; a natural or nature-derived protein structure mediating further specific protein-protein interactions, e. g. a member of the fos/jun pair; a magnetic group, e. g. a ferromagnetic group; or a radioactive group, e. g. a group comprising 'H, "C, 32 P, 35 S or 1251 or any combination thereof; or to a globulomer flagged by being covalently or by non-covalent 10 high-affinity interaction, preferably covalently linked to a group that facilitates inactivation, se questration, degradation and/or precipitation, preferably flagged with a group that promotes in vivo degradation, more preferably with ubiquitin, where is particularly preferred if this flagged oligomer is assembled in vivo; or to a globulomer modified by any combination of the above. Such labelling and flagging groups and methods for attaching them to proteins are known in 15 the art. Labelling and/or flagging may be performed before, during or after globulomerisation. In another aspect of the invention, a globulomer derivative is a molecule obtainable from a globulomer by a labelling and/or flagging reaction. Correspondingly, term "AP(X-Y) monomer derivative" here refers in particular to an A0 mono 20 mer that is labelled or flagged as described for the globulomer. Expediently, the antibody of the present invention binds to an Ap(20-42) globulomer with a K 0 in the range of x10-6 M to 1x10- 12 M. Preferably, the antibody binds to an AP(20-42) globu lomer with high affinity, for instance with a KD of 1x10- 7 M or greater affinity, e.g. with a KD Of 25 3x10-8 M or greater affinity, with a KD of 1x10-e M or greater affinity, e.g. with a KD of 3x10- 9 M or greater affinity, with a KD of 1x10- 9 M or greater affinity, e.g. with a KD of 3x10 0 M or greater affinity, with a KD of 1x1 0-10 M or greater affinity, e.g. with a KD of 3x10 1 M or greater affinity, or with a KD Of 1x10 11 M or greater affinity. 30 The term "greater affinity" here refers to a degree of interaction where the equilibrium between unbound antibody and unbound globulomer on the one hand and antibody-globulomer com plex on the other is further in favour of the antibody-globulomer complex. Likewise, the term "smaller affinity" here refers to a degree of interaction where the equilibrium between unbound antibody and unbound globulomer on the one hand and antibody-globulomer complex on the 35 other is further in favour of the unbound antibody and unbound globulomer. The term "greater affinity"is synonymous with the term "higher affinity" and term "smaller affinity"is synonymous with the term "lower affinity". According to a particular embodiment, the invention relates to an antibody which binds to the 40 A(20-42) globulomer with a KD in the range of 1x10- 6 M to 1x10~~ 2 M, to the Ap(1-42) globu- 11 lomer with a KD of 10~1 M or smaller affinity, the binding affinity to the AP(20-42) globulomer being greater than the binding affinity to the AP(1-42) globulomer. It is preferred that the binding affinity of the antibody of the present invention to the As(20-42) 5 globulomer is at least 2 times, e. g. at least 3 times or at least 5 times, preferably at least 10 times, e. g. at least 20 times, at least 30 times or at least 50 times, more preferably at least 100 times, e. g. at least 200 times, at least 300 times or at least 500 times, and even more preferably at least 1000 times, e. g. at least 2000 times, at least 3000 times or at least 5000 times, even more preferably at least 10000 times, e. g. at least 20000 times, at least 30000 or 10 at least 50000 times, and most preferably at least 100000 times greater than the binding affin ity of the antibody to the Ap(1-42) globulomer. According to a particular embodiment, the invention relates to an antibody which binds to the Ap(12-42) globulomer with a KD with a KD of 10-12 M or smaller affinity, the binding affinity to 15 the AP(20-42) globulomer being greater than the binding affinity to the Ap(12-42) globulomer. It is also preferred that the binding affinity of the antibody of the present invention to the AP(20 42) globulomer is at least 2 times, e. g. at least 3 times or at least 5 times, preferably at least 10 times, e. g. at least 20 times, at least 30 times or at least 50 times, more preferably at least 20 100 times, e. g. at least 200 times, at least 300 times or at least 500 times, and even more preferably at least 1000 times, e. g. at least 2000 times, at least 3000 times or at least 5000 times, even more preferably at least 10000 times, e. g. at least 20000 times, at least 30000 or at least 50000 times, and most preferably at least 100000 times greater than the binding affin ity of the antibody to the AP(12-42) globulomer. 25 Preferably, the antibodies of the present invention bind to at least one As globulomer, as de fined above, and have a comparatively smaller affinity for at least one non-globulomer form of Ap. 30 Antibodies of the present invention having a comparatively smaller affinity for at least one non globulomer form of As than for at least one As globulomer include antibodies having a binding affinity to the Ap(20-42) globulomer that is greater than to an AP((1-42) monomer. Further, it is preferred that, alternatively or additionally, the binding affinity of the antibody to the Ap(20-42) globulomer is greater than to an Ap(1-40) monomer. 35 In a preferred embodiment of the invention, the affinity of the antibody to the Ap(20-42) globu lomer is greater than its affinity to both the Apr(1-40) and the Ap(1-42) monomer. The term "Ap(X-Y) monomer" here refers to the isolated form of the AP(X-Y) peptide, prefera 40 bly a form of the Ap(X-Y) peptide which is not engaged in essentially non-covalent interactions 12 with other AP peptides. Practically, the Ap(X-Y) monomer is usually provided in the form of an aqueous solution. In a particularly preferred embodiment of the invention, the aqueous mono mer solution contains 0.05% to 0.2%, more preferably about 0.1% NH 4 0H. In another particu larly preferred embodiment of the invention, the aqueous monomer solution contains 0.05% to 5 0.2%, more preferably about 0.1% NaOH. When used (for instance for determining the binding affinities of the antibodies of the present invention), it may be expedient to dilute said solution in an appropriate manner. Further, it is usually expedient to use said solution within 2 hours, in particular within 1 hour, and especially within 30 minutes after its preparation. 10 More specifically, the term "AP(1-40) monomer" here refers to an AP(1-40) monomer prepara tion as described in example 2 herein, and the term "Ap(1-42) monomer' here refers to an Ap(1-42) preparation as described in example 2 herein. Expediently, the antibody of the present invention binds to one or, more preferably, both mono 15 mers with low affinity, most preferably with a KD of 1x10-8 M or smaller affinity, e. g. with a Ko of 3x10-8 M or smaller affinity, with a KD of 1x10- 7 M or smaller affinity, e. g. with a KD of 3x10 7 M or smaller affinity, or with a Ko of 1x10- M or smaller affinity, e. g. with a Ko of 3x10- 5 M or smaller affinity, or with a KD of 1x10- M or smaller affinity. 20 It is especially preferred that the binding affinity of the antibody of the present invention to the Ap(20-42) globulomer is at least 2 times, e. g. at least 3 times or at least 5 times, preferably at least 10 times, e. g. at least 20 times, at least 30 times or at least 50 times, more preferably at least 100 times, e. g. at least 200 times, at least 300 times or at least 500 times, and even more preferably at least 1000 times, e. g. at least 2000 times, at least 3000 times or at least 25 5000 times, even more preferably at least 10000 times, e. g. at least 20000 times, at least 30000 or at least 50000 times, and most preferably at least 100000 times greater than the binding affinity of the antibody to one or, more preferably, both monomers. Antibodies of the present invention having a comparatively smaller affinity for at least one non 30 globulomer form of As than for at least one AP globulomer further include antibodies having a binding affinity to the Ap(20-42) globulomer that is greater than to AP(1-42) fibrils. Further, it is preferred that, alternatively or additionally, the binding affinity of the antibody to the Ap(20-42) globulomer is greater than to Ap(1-40) fibrils. 35 The term "fibril" here refers to a molecular structure that comprises assemblies of non covalently associated, individual Ap(X-Y) peptides, which show fibrillary structure in the elec tron microscope, which bind Congo red and then exhibit birefringence under polarized light and whose X-ray diffraction pattern is a cross-P structure.

13 In another aspect of the invention, a fibril is a molecular structure obtainable by a process that comprises the self-induced polymeric aggregation of a suitable AP peptide in the absence of detergents, e. g. in 0.1 M HCI, leading to the formation of aggregates of more than 24, prefera bly more than 100 units. This process is well known in the art. Expediently, A(X-Y) fibrils are 5 used in the form of an aqueous solution. In a particularly preferred embodiment of the inven tion, the aqueous fibril solution is made by dissolving the AP peptide in 0.1% NH 4 0H, diluting it 1 : 4 with 20 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4, followed by readjusting the pH to 7.4, incu bating the solution at 37 *C for 20 h, followed by centrifugation at 10000 g for 10 min and re suspension in 20 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4. 10 The term "Ap(X-Y) fibril" here refers to a fibril consisting essentially of AP(X-Y) subunits, where it is preferred if on average at least 90% of the subunits are of the AP(X-Y) type, more pre ferred if at least 98% of the subunits are of the AP(X-Y) type, and most preferred if the content of non-AP(X-Y) peptides is below the detection threshold. 15 More specifically, the term "A(1-42) fibril" here refers to a AP(1-42) fibril preparation as de scribed in example 3 herein. Expediently, the antibody of the present invention binds to one or, more preferably, both fibrils 20 with low affinity, most preferably with a K 0 of 1x10- M or smaller affinity, e. g. with a Ko of 3x108 M or smaller affinity, with a KD of 1x10~ 7 M or smaller affinity, e. g. with a Ko of 3x10- 7 M or smaller affinity, or with a Ko of 1x1e M or smaller affinity, e. g. with a K 0 of 3x10 M or smaller affinity, or with a K 0 of 1x10- M or smaller affinity. 25 It is especially preferred that the binding affinity of the antibody of the present invention to AP(20-42) globulomer is at least 2 times, e. g. at least 3 times or at least 5 times, preferably at least 10 times, e. g. at least 20 times, at least 30 times or at least 50 times, more preferably at least 100 times, e. g. at least 200 times, at least 300 times or at least 500 times, and even more preferably at least 1000 times, e. g. at least 2000 times, at least 3000 times or at least 30 5000 times, even more preferably at least 10000 times, e. g. at least 20000 times, at least 30000 or at least 50000 times, and most preferably at least 100000 times greater than the binding affinity of the antibody to one or, more preferably, both fibrils. According to one particular embodiment, the invention relates to antibodies having a binding 35 affinity to the AP(20-42) globulomer which is greater than its binding affinity to both AP(1-40) and Ap(1-42) fibrils. According to a particularly preferred embodiment, the present invention relates to antibodies having a comparatively smaller affinity for both the monomeric and fibrillary forms of As than 14 for at least one As globulomer, in particular AP(20-42) globulomer. These antibodies hereinaf ter are referred to globulomer-specific antibodies. Antibodies of the present invention further include antibodies having a binding affinity to the 5 Ap(20-42) globulomer that is greater than to a cross-linked Ap(1-42) globulomer, in particular to a glutardialdehyde cross-linked A1(1-42) globulomer, such as that described in example 1b herein. In a particularly preferred embodiment of the invention, the binding affinity of the antibody to 10 AP(20-42) globulomer is at least 2 times, e. g. at least 3 times or at least 5 times, preferably at least 10 times, e. g. at least 20 times, at least 30 times or at least 50 times, more preferably at least 100 times, e. g. at least 200 times, at least 300 times or at least 500 times, and even more preferably at least 1000 times, e. g. at least 2000 times, at least 3000 times or at least 5000 times, even more preferably at least 10000 times, e. g. at least 20000 times, at least 15 30000 or at least 50000 times, and most preferably at least 100000 times greater than the binding affinity of the antibody to cross-linked Ap(1-42) globulomer. The antibodies of the present invention are preferably isolated, in particular monoclonal and more particularly recombinant. 20 The present invention also relates to a monoclonal antibody (5F7) that is obtainable from a hybridoma designated by American Type Culture Collection deposit number PTA-7241. The present invention also relates to a monoclonal antibody (10F11) that is obtainable from a 25 hybridoma designated by American Type Culture Collection deposit number PTA-7239. The present invention also relates to a monoclonal antibody (7C6) that is obtainable from a hybridoma designated by American Type Culture Collection deposit number PTA-7240. 30 The present invention also relates to a monoclonal antibody (4B7) that is obtainable from a hybridoma designated by American Type Culture Collection deposit number PTA-7242. The present invention also relates to a monoclonal antibody (6A2) that is obtainable from a hybridoma designated by American Type Culture Collection deposit number PTA-7409. 35 The present invention also relates to a monoclonal antibody (2F2) that is obtainable from a hybridoma designated by American Type Culture Collection deposit number PTA-7408. The present invention also relates to a monoclonal antibody (4D10) that is obtainable from a 40 hybridoma designated by American Type Culture Collection deposit number PTA-7405.

15 The present invention also relates to a monoclonal antibody (7E5) that is obtainable from a hybridoma designated by American Type Culture Collection deposit number PTA-7809. 5 The present invention also relates to a monoclonal antibody (10C1) that is obtainable from a hybridoma designated by American Type Culture Collection deposit number PTA-781 0. The present invention also relates to a monoclonal antibody (3B10) that is obtainable from a hybridoma designated by American Type Culture Collection deposit number PTA-7851. 10 These antibodies of the present invention, 5F7, 1OF1 1, 7C6, 4B7, 6A2, 2F2, 4D10, 7E5, 1OC1 and 3810, are characterized by having a binding affinity to an AP(20-42) globulomer that is greater than the binding affinity of the antibody to an A1(1-42) globulomer. 15 The present invention also relates to antibodies having a similar binding profile to that of any one of said monoclonal antibodies, 5F7, 1OF11, 7C6, 4B7, 6A2, 2F2, 4D10, 7E5, 1OC1 and 3B10. Antibodies having a similar binding profile to that of any one of said monoclonal antibod ies should be understood as not being limited to antibodies having a binding affinity to an AP(20-42) globulomer that is greater than the binding affinity of the antibody to an AP(1 -42) 20 globulomer. Antibodies having a binding profile similar to that of any one of said monoclonal antibodies, 5F7, 1OF1 1, 7C6, 4B7, 6A2, 2F2, 4D10, 7E5, 10C1 and 3B10, include antibodies which bind to the same epitope as monoclonal antibody 5F7, 1OF11, 7C6, 4B7, 6A2, 2F2. 4D10, 7E5, 10C1 25 and 3B10. All monoclonal antibodies from the group consisting of 5F7, 1OF11, 7C6, 4B7, 6A2, 2F2, 4D10, 7E5, 10C1 and 3B10 bind to an epitope contained within the 20 and 42 AP sequence range, in particular within the 20-30 As sequence range. Without being bound to theory, said epitope is 30 believed to be a structural, non-linear epitope in between subunits in the region of amino acids 20 and 42, in particular in the region of amino acids 20 and 30. The present invention also relates to antibodies which are capable of competing with at least one, preferably all, antibodies selected from the group consisting of 5F7, 10F11, 7C6, 4B7, 35 6A2, 2F2,4D10, 7E5, 10C1 and 3B10. The term "competing antibodies" herein refers to any number of antibodies targeting the same molecular or stably but non-covalently linked supermolecular entity, preferably the same mole cule, wherein at least one is capable of specifically reducing the measurable binding of an 40 other, preferably by sterically hampering the other's access to its target epitope or by inducing 16 and/or stabilizing a conformation in the target entity that reduces the target's affinity for the other antibody, more preferably by directly blocking access to the other's target epitope by binding to an epitope in sufficiently close vicinity of the former, overlapping with the former or identical to the former, most preferably overlapping or identical, in particular identical. Two epi 5 topes are herein said to be "overlapping" if they share part of their chemical structures, pref erably their amino acid sequences, and to be "identical", if their chemical structures, preferably their amino acid sequences, are identical. Thus, the present invention also relates to antibodies whose target epitopes are overlapping 10 with, preferably identical to, the target epitope of at least one of the antibodies selected from the group consisting of 5F7, 1OF1 1, 7C6, 487, 6A2, 2F2, 4D10, 7E5, 1OC1 and 3B10. Antibodies having a similar binding profile to that of any one of said monoclonal antibodies, 5F7, 10F11, 7C6, 487, 6A2, 2F2, 4D10, 7E5, 1OC1 and 3B10, thus further include antibodies 15 which comprise at least a portion of the antigen-binding moiety of any one of said monoclonal antibodies, 5F7, 1OF11, 7C6, 4B7, 6A2, 2F2, 4D10, 7E5, 1OC1 and 3B10. Preferably, said portion comprises at least one complementary determining region (CDR) of any one of said monoclonal antibodies, 5F7, 1OF11, 7C6, 4B7, 6A2, 2F2, 4D10, 7E5, 10C1 and 3810. 20 Thus, according to a further particular embodiment, the present invention relates to antibodies comprising the amino acid sequence of the heavy chain CDR3 and/or the amino acid se quence of the light chain CDR3 of monoclonal antibody 5F7, 1OF 11, 7C6, 487, 6A2, 2F2, 4D10, 7E5, 10C1 or 3B10. Specific examples of such antibodies include those which also comprise the amino acid sequence of the heavy chain CDR2 and/or the amino acid sequence 25 of the light chain CDR2 of monoclonal antibody 5F7, 1 0F1 1, 7C6, 4B7, 6A2, 2F2, 4D10, 7E5, 10C1 or 3B10, respectively. Even more specifically, such antibodies include those which also comprise the amino acid sequence of the heavy chain CDR1 and/or the amino acid sequence of the light chain CDR1 of monoclonal antibody 5F7, 1 0F1 1, 7C6, 4B7, 6A2, 2F2, 4D10, 7E5, 10C1 or 3810, respectively. 30 In one aspect, the present invention thus relates to antibodies comprising a heavy chain wherein the CDR3, CDR2 and/or CDR1 domain comprises the amino acid sequence of the heavy chain CDR3, CDR2 and/or CDR1 of monoclonal antibody 5F7, 1 OF1 1, 7C6, 4B7, 6A2, 2F2, 4D10, 7E5, 10C1 or 3B10. 35 In a further aspect, the present invention thus relates to antibodies comprising a light chain wherein the CDR3, CDR2 and/or CDR1 domain comprises the amino acid sequence of the light chain CDR3, CDR2 and/or CDR1, respectively, of monoclonal antibody 5F7, 1OF1 1, 7C6, 4B7, 6A2, 2F2, 4D10, 7E5, 10C1 or 3810. 40 17 Preferably, the antibody comprises at least one CDR comprising an amino acid sequence se lected from the group consisting of: amino acid residues 31-35 of SEQ ID NO:3, amino acid residues 50-66 of SEQ ID NO:3, amino acid residues 99-109 of SEQ ID NO:3, amino acid resi dues 24-39 of SEQ ID NO:4, amino acid residues 55-61 of SEQ ID NO:4, amino acid residues 5 94-102 of SEQ ID NO:4, amino acid residues 31-35 of SEQ ID NO:7, amino acid residues 50 66 of SEQ ID NO:7, amino acid residues 97-109 of SEQ ID NO:7, amino acid residues 24-39 of SEQ ID NO:8, amino acid residues 55-61 of SEQ ID NO:8, amino acid residues 94-102 of SEQ ID NO:8, amino acid residues 31-35 of SEQ ID NO: 11, amino acid residues 50-65 of SEQ ID NO:1 1, amino acid residues 98-107 of SEQ ID NO:1 1, amino acid residues 24-39 of 10 SEQ ID NO:12, amino acid residues 55-61 of SEQ ID NO:12, amino acid residues 94-102 of SEQ ID NO:12, amino acid residues 31-35 of SEQ ID NO:15, amino acid residues 50-66 of SEQ ID NO:15, amino acid residues 99-107 of SEQ ID NO:15, amino acid residues 24-40 of SEQ ID NO:16, amino acid residues 56-62 of SEQ ID NO:16, amino acid residues 95-103 of SEQ ID NO:16, amino acid residues 31-35 of SEQ ID NO:19, amino acid residues 50-66 of 15 SEQ ID NO:19, amino acid residues 99-109 of SEQ ID NO:19, amino acid residues 24-39 of SEQ ID NO:20, amino acid residues 55-61 of SEQ ID NO:20, amino acid residues 94-102 of SEQ ID NO:20, amino acid residues 31-35 of SEQ ID NO:23, amino acid residues 50-66 of SEQ ID NO:23, amino acid residues 99-109 of SEQ ID NO:23, amino acid residues 24-39 of SEQ ID NO:24, amino acid residues 55-61 of SEQ ID NO:24, amino acid residues 94-102 of 20 SEQ ID NO:24, amino acid residues 31-35 of SEQ ID NO:27, amino acid residues 50-65 of SEQ ID NO:27, amino acid residues 98-101 of SEQ ID NO:27, amino acid residues 24-39 of SEQ ID NO:28, amino acid residues 55-61 of SEQ ID NO:28, amino acid residues 94-102 of SEQ ID NO:28, amino acid residues 31-35 of SEQ ID NO:31, amino acid residues 50-66 of SEQ ID NO:31, amino acid residues 99-107 of SEQ ID NO:31, amino acid residues 24-40 of 25 SEQ ID NO:32, amino acid residues 56-62 of SEQ ID NO:32, amino acid residues 95-103 of SEQ ID NO:32, amino acid residues 31-35 of SEQ ID NO:35, amino acid residues 50-66 of SEQ ID NO:35, amino acid residues 99-107 of SEQ ID NO:35, amino acid residues 24-40 of SEQ ID NO:36, amino acid residues 56-62 of SEQ ID NO:36, amino acid residues 95-103 of SEQ ID NO:36, amino acid residues 31-35 of SEQ ID NO:38, amino acid residues 50-66 of 30 SEQ ID NO:38, and amino acid residues 98-109 of SEQ ID NO:38. In a preferred embodiment, the antibody comprises at least 3 CDRs selected from the group consisting of the sequences disclosed above. More preferably the 3 CDRs selected are from sets of variable domain CDRs selected from the group consisting of: 35 VH 5F7 CDR Set VH 5F7 CDR-H1 TFYIH: VH_5F7_ CDR-H1 residues 31-35 of SEQ ID NO:3 VH 5F7 CDR-H2 MIGPGSGNTYYNEMFKD: residues 50-66 of SEQ ID NO:3 VH 5F7 CDR-H3 AKSARAAWFAY: residues 99-109 of SEQ ID NO:3 18 VL 5F7 CDR Set VL 5F7 CDR-L1 RSSQSVVQSNGNTYLE: residues 24-39 of SEQ ID NO:4 KVSNRFS; VL SF7 CDR-L2 KSRS residues 55-61 of SEQ ID NO:4 VL 5F7 CDR-L3 FQGSHVPPT: residues 94-102 of SEQ ID NO:4 VH 1OF1l CDR Set VH 10F1l CDR-H1 SYVMH: residues 31-35 of SEQ ID NO:7 VH 1OF11 CDR-H2 YIYPYNDGTKYNEKFKG: residues 50-66 of SEQ ID NO:7 VHi10Fll CDR-H3 TVEGATWDGYFDV: residues 97-109 of SEQ ID NO:7 VL 1OF11 CDR Set VL 10F1l CDR-L1 KSSQSLLYSKGKTYLN: residues 24-39 of SEQ ID NO:8 VL 10F11 CDR-L2 LVSKLDS: residues 55-61 of SEQ ID NO:8 VL 10F1l CDR-L3 VQGTHFPHT: residues 94-102 of SEQ ID NO:8 VH 7C6 CDR Set VH 7C6 CDR-H1 SYAMS: residues 31-35 of SEQ ID NO:11 VH 7C6 CDR-H2 SIHNRGTIFYLDSVKG: residues 50-65 of SEQ ID NO:11 VH 7C6 CDR-H3 GRSNSYAMDY: residues 98-107 of SEQ ID NO:11 VL 7C6 CDR Set VL 7C6 CDR-L1 RSTQTLVHRNGDTYLE: residues 24-39 of SEQ ID NO:12 VL 7C6 CDR-L2 KVSNRFS: residues 55-61 of SEQ ID NO:12 VL 7C6 CDR-L3 FQGSHVPYT: residues 94-102 of SEQ ID NO:12 VH 4B7 CDR Set VH 4B7 CDR-H1 DYEMV: residues 31-35 of SEQ ID NO:15 VH 4B7 CDR-H2 YISSGSRTIHYADTVKG: residues 50-66 of SEQ ID NO:15 VH 4B7 CDR-H3 TLLRLHFDY: residues 99-107 of SEQ ID NO:15 VL 4B7 CDR Set VL 4B7 CDR-L1 RSSQSLFYRSNQKNFLA: residues 24-40 of SEQ ID NO:16 VL 487 CDR-L2 WASTRES: residues 56-62 of SEQ ID NO:16 VL 4B7 CDR-L3QQYYSYPWT: residues 95-103 of SEQ ID NO:16 VH 2F2 CDR Set VH 2F2 CDR-H1 TFYIH: residues 31-35 of SEQ ID NO:19 VH 2F2 CDR-H2 MIGPGSGNTYYNEMFKD: residues 50-66 of SEQ ID NO:19 VH 2F2 CDR-H3 AKSARAAWFAY: residues 99-109 of SEQ ID NO:19 VL 2F2 CDR Set VL 2F2 CDR-L1 RSSQSVVQSNGNTYLE: residues 24-39 of SEQ ID NO:20 VL 2F2 CDR-L2 KVSNRFS: residues 55-61 of SEQ ID NO:20 19 VL 2F2 CDR-L3 FQGSHVPPT: residues 94-102 of SEQ ID NO:20 VH 6A2 CDR Set VH 62 CD-HlTFYIH: VH 6A2 CDR-H1 residues 31-35 of SEQ ID NO:23 MIGPGSGNTYYNEMFKD: yR 6A2 CDR-H2 residues 50-66 of SEQ ID NO:23 VH 6A2 CDR-H3 AKSHRAAWFAY: residues 99-109 of SEQ ID NO:23 VL 6A2 CDR Set VL 6A2 CDR-L1 RSSQSVVQSNGNTYLE: residues 24-39 of SEQ ID NO:24 VL 62 CD-L2KVSNRFF: VL 6A2 CDR-L2 residues 55-61 of SEQ ID NO:24 VL 6A2 CDR-L3 FQGSHVPPT: residues 94-102 of SEQ ID NO:24 VH 4D10 CDR Set VH 410 CR-HlSYGVH: V-H 4010 CDR-Hl residues 31-35 of SEQ ID NO:27 V H 4D10 CDR-H2 VIWRGGRIDYNAAFMS: H residues 50-65 of SEQ ID NO:27 VH 410 CR-H3NSDV: VH 4D1 CDR-H3 residues 98-101 of SEQ ID NO:27 VL 4D10 CDR Set VL 4D10 CDR-L1 KSSQSLLDIDGKTYLN: residues 24-39 of SEQ ID NO:28 VL 410 CR-L2LVSKLDS: VL 4D10 CDR-L2 residues 55-61 of SEQ ID NO:28 VL 4D10 CDR-L3 WQGTHFPYT: residues 94-102 of SEQ ID NO:28 VH 7E5 CDR Set VH 75 CD-H1DY EM V VH 7ES CDR-Hl residues 31-35 of SEQ ID NO:31 VH 7E5 CDR-H2 YISSGSRTIHYADTVKG: residues 50-66 of SEQ ID NO:31 VH 7E5 CDR-H3 TLLRLHFDY: residues 99-107 of SEQ ID NO:31 VL 7E5 CDR Set VL 7E5 CDR-L1 .RSSQSLFYRSNQKNFLA: residues 24-40 of SEQ ID NO:32 VL 75 CD-L2 AST RE S VL 7E5 CDR-L2 residues 56-62 of SEQ ID NO:32 QQYYSYPWT: VL 7E5 CDR-L3 residues 95-103 of SEQ ID NO:32 VUII OCI CDR Set VH 11 CR SDYEMtV VH 10C1 CDR-Hl residues 31-35 of SEQ ID NO:35 VH 1OC1 CDR-H2 YINSGSGTIHYADTVKG: residues 50-66 of SEQ ID NO:35 VH 10C1 CDR-H3 TLLRLHFDY: residues 99-107 of SEQ ID NO:35 VL 1OC1 CDR Set VL 1OC1 CDR-LI KSSQSLFYSRNQKNFLA: residues 24-40 of SEQ ID NO:36 VL IC1 CR-L2WASTGES: VL 10C1 CDR-L2 residues 56-62 of SEQ ID N:36 VL 10C1 CDR-L3 QQYFSYPWT: residues 95-103 of SEQ ID NO:36 VH 3B10 CDR Set VH 3310 CDR-Hl DYVIH: residues 31-35 of SEQ ID NO:38 20 vH 3B310 CDR-H2 YINPYNDGTQYNEKFKG: residues 50-66 of SEQ ID NO:38 VH 3B10 CDR-H3 VEGGTWDGYFDV: residues 98-109 of SEQ ID NO:38 In one embodiment the antibody of the invention comprises at least two variable domain CDR sets. More preferably, the two variable domain CDR sets are selected from the group consist ing of: VH 5F7 CDR Set & VL 5F7 CDR Set; VH 1OF1 1 CDR Set & VL 1OF1 1 CDR Set; VH 5 7C6 CDR Set & VL 7C6 CDR Set; VH 4B7 CDR Set & VL 4B7 CDR Set; VH 2F2 CDR Set & VL 2F2 CDR Set; VH 6A2 CDR Set & VL 6A2 CDR Set; VH 4D10 CDR Set & VL 4D10 CDR Set; VH 7E5 CDR Set & VL 7E5 CDR Set; and VH 1OC1 CDR Set & VL 1OC1 CDR Set. In another embodiment the antibody disclosed above further comprises a human acceptor 10 framework. In a preferred embodiment the antibody is a CDR grafted antibody. Preferably the CDR grafted antibody comprises one or more of the CDRs disclosed above. 15 Preferably the CDR grafted antibody comprises a human acceptor framework. In a preferred embodiment the antibody is a humanized antibody. Preferably the humanized antibody comprises one or more of the CDRs disclosed above. More preferably the humanized antibody comprises three or more of the CDRs disclosed above. Most preferably the human 20 ized antibody comprises six CDRs disclosed above. In a particular embodiment, the CDRs are incorporated into a human antibody variable domain of a human acceptor framework. Prefera bly the human antibody variable domain is a consensus human variable domain. More pref erably the human acceptor framework comprises at least one Framework Region amino acid substitution at a key residue, wherein the key residue is selected from the group consisting of a 25 residue adjacent to a CDR; a glycosylation site residue; a rare residue; a residue capable of interacting with AP(20-42) globulomer; a residue capable of interacting with a CDR; a canoni cal residue; a contact residue between heavy chain variable region and light chain variable region; a residue within a Vernier zone; and a residue in a region that overlaps between a Chothia-defined variable heavy chain CDR1 and a Kabat-defined first heavy chain framework. 30 Preferably the human acceptor framework human acceptor framework comprises at least one Framework Region amino acid substitution, wherein the amino acid sequence of the frame work is at least 65% identical to the sequence of said human acceptor framework and com prises at least 70 amino acid residues identical to said human acceptor framework. 35 In yet a further aspect, the present invention relates to antibodies comprising both the heavy and light chain as defined above.

21 Preferably, the antibody comprises at least one variable domain having an amino acid se quence selected from the group consisting of: SEQ ID NO:3, SEQ ID NO:4, SEQ ID NO:7, SEQ ID NO:8, SEQ ID NO:11, SEQ ID NO:12, SEQ ID NO:15, SEQ ID NO:16, SEQ ID NO:19, SEQ ID NO:20, SEQ ID NO:23, SEQ ID NO:24, SEQ ID NO:27, SEQ ID NO:28, SEQ ID 5 NO:31, SEQ ID NO:32, SEQ ID NO:35, SEQ ID NO:36, and SEQ ID NO:38. More preferably, the antibody comprises two variable domains, wherein said two variable domains have amino acid sequences selected from the group consisting of: SEQ ID NO:3 & SEQ ID NO:4, SEQ ID NO:7 & SEQ ID NO:8 & SEQ ID NO:11 & SEQ ID NO:12, SEQ ID NO:15 & SEQ ID NO:16, SEQ ID NO:19,& SEQ ID NO:20, SEQ ID NO:23 & SEQ ID NO:24, SEQ ID NO:27 & SEQ ID 10 NO:28, SEQ ID NO:31 & SEQ ID NO:32, and SEQ ID NO:35 & SEQ ID NO:36. In another aspect, the antibodies of the present invention comprise a heavy chain constant region selected from the group consisting of IgG1, IgG2, IgG3, IgG4, IgM, IgA, IgD, IgE and human IgG1 Ala234 Ala235 mutant constant regions. In particular, the antibodies comprise a 15 human constant region. More preferably, the antibodies comprise an amino acid sequence selected from the group consisting of SEQ ID NOs:39-42. Antibodies comprising an IgG1 heavy chain constant region are preferred. In another embodiment the antibody is glycosylated. Preferably the glycosylation pattern is a 20 human glycosylation pattern or a glycosylation pattern produced by any one of the eukaryotic cells disclosed herein, in particular CHO cells. The present invention also relates to an antigen-binding moiety of an antibody of the present invention. Such antigen-binding moieties include, but are not limited to, Fab fragments, F(ab') 2 25 fragments and single chain Fv fragments of the antibody. Further antigen-binding moieties are Fab' fragments, Fv fragments, and disulfide linked Fv fragments. The invention also provides an isolated nucleic acid encoding any one of the antibodies dis closed herein. A further embodiment provides a vector comprising the isolated nucleic acid 30 disclosed herein. Said vector may in particular be selected from the group consisting of pcDNA; pTT (Durocher et al., Nucleic Acids Research 2002, Vol 30, No.2); pTT3 (pTT with additional multiple cloning site; pEFBOS (Mizushima, S. and Nagata, S., (1990) Nucleic Acids Research Vol 18, No. 17); pBV; pJV; and pBJ. 35 In another aspect a host cell is transformed with the vector disclosed herein. Preferably the host cell is a prokaryotic cell. More preferably the host cell is E.coli. In a related embodiment the host cell is a eukaryotic cell. Preferably the eukaryotic cell is selected from the group con sisting of a protist cell, an animal cell (e.g. a mammalian cell, an avian cell, and an insect cell), a plant cell and a fungal cell. More preferably the host cell is a mammalian cell including, but 22 not limited to, CHO and COS; or a fungal cell, eg., a yeast cell, such as Saccharomyces cere visiae; or an insect cell such as Sf9. Another aspect of the invention provides a method of producing an antibody of the invention, 5 comprising culturing any one of the host cells or a hybridoma disclosed herein in a culture me dium under conditions suitable to produce the antibody. Another embodiment provides an anti body that is obtainable by the method disclosed herein. Antibodies of the present invention can be obtained in a manner known per se. 10 B lymphocytes which, in totality, contain an antibody repertoire composed of hundreds of bil lions of different antibody specificities are a part of the mammalian immune system. A normal immune response to a particular antigen means selection of one or more antibodies of said repertoire which specifically bind to said antigen, and the success of an immune response is 15 based at least partially on the ability of said antibodies to specifically recognize (and ultimately to eliminate) the stimulating antigen and to ignore other molecules in the environment of said antibodies. The usefulness of antibodies which specifically recognize one particular target antigen has led 20 to the development of monoclonal antibody technology. Standardized hybridoma technology now allows the production of antibodies with a single specificity for an antigen of interest. More recently, recombinant antibody techniques such as in-vitro screening of antibody libraries have been developed. These techniques likewise allow antibodies having a single specificity for an antigen of interest to be produced. 25 In the method of the invention, the antigen of interest may be allowed to act on the antibody repertoire either in vivo or in vitro. According to one embodiment, the antigen is allowed to act on the repertoire by immunizing an 30 animal in vivo with said antigen. This in-vivo approach may furthermore comprise establishing from the lymphocytes of an animal a number of hybridomas and selecting a particular hybri doma which secretes an antibody specifically binding to said antigen. The animal to be immu nized may be, for example, a mouse, rat, rabbit, chicken, camelid or sheep or may be a trans genic version of any of the animals mentioned above, for example a transgenic mouse with 35 human immunoglobulin genes, which produces human antibodies after an antigenic stimulus. Other types of animals which may be immunized include mice with severe combined immuno deficiency (SCID) which have been reconstituted with human peripheral mononuclear blood cells (chimeric hu-PBMC SCID mice) or with lymphoid cells or precursors thereof, as well as mice which have been treated with a lethal total body irradiation, then protected against radia 40 tion with bone marrow cells from a mouse with severe combined immunodeficiency (SCID) and 23 subsequently transplanted with functional human lymphocytes (the "Trimera" system). Another type of an animal to be immunized is an animal (e.g. a mouse) in whose genome an endoge nous gene encoding the antigen of interest has been switched off (knocked out), for example by homologous recombination, so that, after immunization with the antigen, said animal recog 5 nizes said antigen as foreign. It is obvious to the skilled worker that the polyclonal or mono clonal antibodies produced by this method are characterized and selected by using known screening methods which include, but are not limited to, ELISA and dot blot techniques. According to another embodiment, the antigen is allowed to act on the antibody repertoire in 10 vitro by screening a recombinant antibody library with said antigen. The recombinant antibody library may be expressed, for example, on the surface of bacteriophages or on the surface of yeast cells or on the surface of bacterial cells. In a variety of embodiments, the recombinant antibody library is an scFv library or an Fab library, for example. According to another em bodiment, antibody libraries are expressed as RNA-protein fusions. 15 Another approach to producing antibodies of the invention comprises a combination of in vivo and in vitro approaches. For example, the antigen may be allowed to act on the antibody rep ertoire by immunizing an animal in vivo with said antigen and then screening in vitro with said antigen a recombinant antibody library prepared from lymphoid cells of said animal or a single 20 domain antibody library (e.g. containing heavy and/or light chains). According to another ap proach, the antigen is allowed to act on the antibody repertoire by immunizing an animal in vivo with said antigen and then subjecting a recombinant antibody library or single domain li brary produced from lymphoid cells of said animal to affinity maturation. According to another approach, the antigen is allowed to act on the antibody repertoire by immunizing an animal 25 in vivo with said antigen, then selecting individual antibody-producing cells secreting an anti body of interest and obtaining from said selected cells cDNAs for the variable region of the heavy and light chains (e.g. by means of PCR) and expressing said variable regions of the heavy and light chains in mammalian host cells in vitro (this being referred to as selected lym phocyte antibody method or SLAM), thereby being able to further select and manipulate the 30 selected antibody gene sequences. Moreover, monoclonal antibodies may be selected by ex pression cloning by expressing the antibody genes for the heavy and light chains in mammal ian cells and selecting those mammalian cells which secrete an antibody having the desired binding affinity. 35 The present invention provides defined antigens for screening and counter screening. Thus it is possible, according to the invention, to select those polyclonal and monoclonal antibodies which bind to an AP(20-42) globulomer with the binding affinities as defined above. The methods of the invention for producing antibodies can be used to produce various types of 40 antibodies. These include monoclonal, in particular recombinant antibodies, especially essen- 24 tially human antibodies, chimeric antibodies, humanized antibodies and CDR graft antibodies, and also antigen-binding moieties thereof. The present invention further relates to a hybridoma that is capable of producing (secreting) a 5 monoclonal antibody of the present invention. Hybridomas of the present invention include those designated by an American Type Culture Collection deposit number selected from the group consisting of PTA-7241, PTA-7239, PTA-7240, PTA-7242, PTA-7408, PTA-7409, PTA 7405, PTA-7809, PTA-7810 and PTA-7851. 10 It is noted that the antibodies of the present invention may also be reactive with, i.e. bind to, As forms other than the AP globulomers described herein. These antigens may or may not be oligomeric or globulomeric. Thus, the antigens to which the antibodies of the present invention bind include any As form that comprises the globulomer epitope with which the antibodies of the present invention are reactive. Such As forms include truncated and non-truncated AP(X 15 Y) forms (with X and Y being defined as above), such as Ap(20-42), Ap(20-40), Ap(12-42), Ap(12-40), Ap(1-42), and Ap(1-40) forms, provided that said forms comprise the globulomer epitope. The present invention also relates to a composition comprising an antibody of the invention or 20 an antigen-binding moiety thereof, as defined above. According to a particular embodiment, said composition is a pharmaceutical composition which comprises the antibody of the invention or the antigen-binding moiety and a pharmaceutical acceptable carrier. 25 The antibody of the invention or the antigen-binding moiety as defined above is preferably ca pable of neutralizing, both in vitro and in vivo, the activity of AP globulomer or a derivative thereof to which it binds. Said antibody or antigen-binding moiety may therefore be used for inhibiting the activity of said globulomer or derivative thereof, for example in a preparation con 30 taining said globulomer or derivative thereof or in human individuals or other mammals in which said globulomer or derivative thereof is present. According to one embodiment, the invention relates to a method of inhibiting the activity of said globulomer or derivative thereof, which method comprises allowing an antibody of the inven 35 tion or an antigen-binding moiety thereof to act on a globulomer or derivative thereof so as to inhibit the activity of said globulomer or derivative thereof. Said activity may be inhibited in vi tro, for example. For instance, the antibody of the invention or the antigen-binding moiety may be added to a preparation such as a sample derived from a subject or a cell culture which con tains or is suspected to contain said globulomer or derivative thereof, in order to inhibit the 25 activity of said globulomer or derivative thereof in said sample. Alternatively, the activity of the globulomer or derivative thereof may be inhibited in an individual in vivo. Thus the present invention further relates to the use of an antibody or an antigen-binding moi 5 ety as defined above for preparing a pharmaceutical composition for treating or preventing an amyloidosis, in particular an amyloidosis selected from the group consisting of Alzheimer's disease and the amyloidosis of Down's syndrome. One aspect of said use of the invention is therefore a method of treating or preventing an amyloidosis, in particular Alzheimer's disease or the amyloidosis of Down's syndrome, in a subject in need thereof, which comprises adminis 10 tering an antibody or an antigen-binding moiety as defined above to the subject. Using said antibody or antigen-binding moiety for treating and especially preventing the amyloidosis, in particular Alzheimer's disease or the amyloidosis of Down's syndrome, is in particular for pas sive immunization. Accordingly, in the method of treating or preventing an amyloidosis, in par ticular Alzheimer's disease or the amyloidosis of Down's syndrome, in a subject in need 15 thereof one purpose of administering the antibody or antigen-binding moiety to the subject is passively immunizing the subject against the amyloidosis, in particular Alzheimer's disease or the amyloidosis of Down's syndrome. The antibody of the invention or the antigen-binding moiety as defined above is preferably ca 20 pable of detecting, both in vitro and in vivo, an AP globulomer or derivative thereof to which it binds. Said antibody or the antigen-binding moiety may therefore be used for detecting said globulomer or derivative thereof, for example in a preparation containing said globulomer or derivative thereof or in human individuals or other mammals in which said globulomer or de rivatives thereof is present. 25 According to one embodiment, the invention relates to a method of detecting said globulomer or derivative thereof, which method comprises allowing an antibody of the invention or an anti gen-binding moiety thereof to act on a globulomer or derivative thereof so as to bind to said globulomer or derivative thereof (and thereby preferably forming a complex comprising the 30 antibody or antigen-binding moiety thereof and the globulomer or derivative thereof). Said globulomer may be detected in vitro, for example. For example, the antibody of the invention or the antigen-binding moiety may be added to a preparation, for instance a sample derived from a subject or a cell culture which contains or is suspected to contain said globulomer or deriva tive thereof, in order to detect said globulomer or derivative thereof in said preparation. Alter 35 natively, the globulomer or derivative thereof may be detected in an individual in vivo. Thus the present invention further relates to the use of an antibody or an antigen-binding moi ety as defined above for preparing a composition for diagnosing an amyloidosis, in particular Alzheimer's disease or the amyloidosis of Down's syndrom. One aspect of said use of the in 40 vention is a method of diagnosing an amyloidosis, in particular Alzheimer's disease or the 26 amyloidosis of Down's syndrom, in a subject suspect of having the amyloidosis, in particular Alzheimer's disease or the amyloidosis of Down's syndrom, which comprises administering to the subject an antibody or an antigen-binding moiety as defined above and detecting the for mation of a complex comprising the antibody or the antigen-binding moiety with the antigen, 5 the presence of the complex indicating the amyloidosis, in particular Alzheimer's disease or the amyloidosis of Down's syndrom, in the subject. A second aspect of said use of the invention is a method of diagnosing an amyloidosis, in particular Alzheimer's disease or the amyloidosis of Down's syndrom, in a subject suspect of having the amyloidosis, in particular Alzheimer's dis ease or the amyloidosis of Down's syndrom, which comprises providing a sample from the 10 subject, contacting the sample with an antibody or an antigen-binding moiety as defined above and detecting the formation of a complex comprising the antibody or the antigen-binding moi ety with the antigen, the presence of the complex indicating the amyloidosis, in particular Alz heimer's disease or the amyloidosis of Down's syndrom, in the subject. 15 Detailed description of the invention The binding affinities of the antibodies of the invention may be evaluated by using standard ized in-vitro immunoassays such as ELISA, dot blot or BlAcore analyses (Pharmacia Biosen 20 sor AB, Uppsala, Sweden and Piscataway, NJ). For further descriptions, see J6nsson, U., et al. (1993) Ann. Biol. Clin. 51:19-26; J6nsson, U., et al. (1991) Biotechniques 11:620-627; Johnsson, B., et al. (1995) J. Mol. Recognit. 8:125-131; and Johnsson, B., et al. (1991) Anal. Biochem. 198:268-277. 25 According to a particular embodiment, the affinities defined herein refer to the values obtained by performing a dot blot as described in example 8 and evaluating it by densitometry. Accord ing to a particular embodiment of the invention, determining the binding affinity by dot blot comprises the following: a certain amount of the antigen (e.g. the Ap(X-Y) globulomer, Ap(X-Y) monomer or AP(X-Y) fibrils, as defined above) or, expediently, an appropriate dilution thereof, 30 for instance in 20 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4, 0.2 mg/ml BSA to an antigen concen tration of, for example, 100 pmol/pl, 10 pmol/pl, 1 pmol/pl, 0.1 pmol/pl and 0.01 pmol/pl, is dot ted onto a nitrocellulose membrane, the membrane is then blocked with milk to prevent unspe cific binding and washed, then contacted with the antibody of interest followed by detection of the latter by means of an enzyme-conjugated secondary antibody and a colorimetric reaction; 35 at defined antibody concentrations, the amount of antibody bound allows affinity determination. Thus the relative affinity of two different antibodies to one target, or of one antibody to two dif ferent targets, is here defined as the relation of the respective amounts of target-bound anti body observed with the two antibody-target combinations under otherwise identical dot blot conditions. Unlike a similar approach based on Western blotting, the dot blot approach will 40 determine an antibody's affinity to a given target in the latter's natural conformation; unlike the 27 ELISA approach, the dot blot approach does not suffer from differences in the affinities be tween different targets and the matrix, thereby allowing for more precise comparisons between different targets. 5 The term "Kd", as used herein, is intended to refer to the dissociation constant of a particular antibody-antigen interaction as is known in the art. The antibodies of the present invention are preferably isolated antibodies. An "isolated anti body" means an antibody having the binding affinities as described above and which is essen 10 tially free of other antibodies having different binding affinities. The term "essentially free" here refers to an antibody preparation in which at least 95% of the antibodies, preferably at least 98% of the antibodies and more preferably at least 99% of the antibodies have the desired binding affinity. Moreover, an isolated antibody may be substantially free of other cellular mate rial and/or chemicals. 15 The isolated antibodies of the present invention include monoclonal antibodies. A "monoclonal antibody" as used herein is intended to refer to a preparation of antibody molecules, antibodies which share a common heavy chain and common light chain amino acid sequence, in contrast with "polyclonal" antibody preparations which contain a mixture of antibodies of different amino 20 acid sequence. Monoclonal antibodies can be generated by several novel technologies like phage, bacteria, yeast or ribosomal display, as well as by classical methods exemplified by hybridoma-derived antibodies (e.g., an antibody secreted by a hybridoma prepared by hybri doma technology, such as the standard Kohler and Milstein hybridoma methodology ((1975) Nature 256:495-497). Thus, a non-hybridoma-derived antibody with uniform sequence is still 25 referred to as a monoclonal antibody herein although it may have been obtained by non classical methodologies, and the term "monoclonal" is not restricted to hybridoma-derived anti bodies but used to refer to all antibodies derived from one nucleic acid clone. Thus, the monoclonal antibodies of the present invention include recombinant antibodies. The 30 term "recombinant" herein refers to any artificial combination of two otherwise separated seg ments of sequence, e.g., by chemical synthesis or by the manipulation of isolated segments of nucleic acids by genetic engineering techniques. In particular, the term "recombinant antibody" refers to antibodies which are produced, expressed, generated or isolated by recombinant means, such as antibodies which are expressed using a recombinant expression vector trans 35 fected into a host cell; antibodies isolated from a recombinant combinatorial antibody library; antibodies isolated from an animal (e.g. a mouse) which is transgenic due to human immu noglobulin genes (see, for example, Taylor, L.D., et al. (1992) Nucl. Acids Res. 20:6287-6295); or antibodies which are produced, expressed, generated or isolated in any other way in which particular immunoglobulin gene sequences (such as human immunoglobulin gene sequences) 40 are assembled with other DNA sequences. Recombinant antibodies include, for example, chi- 28 meric, CDR graft and humanized antibodies. The person skilled in the art will be aware that expression of a conventional hybridoma-derived monoclonal antibody in a heterologous sys tem will require the generation of a recombinant antibody even if the amino acid sequence of the resulting antibody protein is not changed or intended to be changed. 5 In a particular embodiment of the invention, the antibody is a humanized antibody. According to a multiplicity of embodiments, the antibody may comprise an amino acid se quence derived entirely from a single species, such as a human antibody or a mouse antibody. 10 According to other embodiments, the antibody may be a chimeric antibody or a CDR graft an tibody or another form of a humanized antibody. The term "antibody" is intended to refer to immunoglobulin molecules consisting of 4 polypep tide chains, two heavy (H) chains and two light (L) chains. The chains are usually linked to one 15 another via disulfide bonds. Each heavy chain is composed of a variable region of said heavy chain (abbreviated here as HCVR or VH) and a constant region of said heavy chain. The heavy chain constant region consists of three domains CH1, CH2 and CH3. Each light chain is composed of a variable region of said light chain (abbreviated here as LCVR or VL) and a con stant region of said light chain. The light chain constant region consists of a CL domain. The 20 VH and VL regions may be further divided into hypervariable regions referred to as comple mentarity-determining regions (CDRs) and interspersed with conserved regions referred to as framework regions (FR). Each VH and VL region thus consists of three CDRs and four FRs which are arranged from the N terminus to the C terminus in the following order: FRI, CDR1, FR2, CDR2, FR3, CDR3, FR4. This structure is well known to those skilled in the art. 25 The term "antigen-binding moiety" of an antibody (or simply "antibody moiety") refers to one or more fragments of an antibody of the invention, said fragment(s) still having the binding affini ties as defined above. Fragments of a complete antibody have been shown to be able to carry out the antigen-binding function of an antibody. In accordance with the term "antigen-binding 30 moiety" of an antibody, examples of binding fragments include (i) an Fab fragment, i.e. a monovalent fragment composed of the VL, VH, CL and CH1 domains; (ii) an F(ab') 2 fragment, i.e. a bivalent fragment comprising two Fab fragments linked to one another in the hinge region via a disulfide bridge; (iii) an Fd fragment composed of the VH and CH1 domains; (iv) an Fv fragment composed of the FL and VH domains of a single arm of an antibody; (v) a dAb frag 35 ment (Ward et al., (1989) Nature 341:544-546) consisting of a VH domain or of VH, CH1, CH2, DH3, or VH, CH2, CH3; and (vi) an isolated complementarity-determining region (CDR). Al though the two domains of the Fv fragment, namely VL and VH, are encoded by separate genes, they may further be linked to one another using a synthetic linker, e.g. a poly-G 4 S amino acid sequence, and recombinant methods, making it possible to prepare them as a sin 40 gle protein chain in which the VL and VH regions combine in order to form monovalent mole- 29 cules (known as single chain Fv (ScFv); see, for example, Bird et a. (1988) Science 242:423 426; and Huston et al. (1988) Proc. Nat/. Acad. Sci. USA 85:5879-5883). The term "antigen binding moiety" of an antibody is also intended to comprise such single chain antibodies. Other forms of single chain antibodies such as "diabodies" are likewise included here. Diabodies are 5 bivalent, bispecific antibodies in which VH and VL domains are expressed on a single polypep tide chain, but using a linker which is too short for the two domains being able to combine on the same chain, thereby forcing said domains to pair with complementary domains of a differ ent chain and to form two antigen-binding sites (see, for example, Holliger, P., et al. (1993) Proc. Natl. Acad. Sci. USA 90:6444-6448; Poljak, R.J., et al. (1994) Structure 2:1121-1123). 10 An immunoglobulin constant domain refers to a heavy or light chain constant domain. Human IgG heavy chain and light chain constant domain amino acid sequences are known in the art and are represented in Table 1. Table 1: Sequence of human IgG heavy chain constant domain and light chain constant do main 15 Protein Sequence ID Sequence 12345678901234567890123456789012 Ig gamma-1 SEQ ID NO:39 ASTKGPSVFFLAPSSKSTSGGTAALGCLVKDY constant region FPEPVTVSWNSGALTSGVHTFPAVLQSSGLYS LSSVVTVPSSSLGTQTYICNVNHKPSNTKVDK KVEPKSCDKTHTCPPCPAPELLGGPSVFLFPP KPKDTLMISRTPEVTCVVVDVSHEDPEVKFNW YVDGVEVHNAKTKPREEQYNSTYRVVSVLTVL HQDWLNGKEYKCKVSNKALPAPIEKTISKAKG QPREPQVYTLPPSREEMTKNQVSLTCLVKGFY PSDIAVEWESNGQPENNYKTTPPVLDSDGSFF LYSKLTVDKSRWQQGNVFSCSVMHEALHNHYT QKSLSLSPGK Ig gamma-1 SEQ ID NO:40 ASTKGPSVFPLAPSSKSTSGGTAALGCLVKDY constant region EPEPVTVSWNSGALTSGVHTFPAVLQSSGLYS mutant LSSVVTVPSSSLGTQTYICNVNHKPSNTKVDK KVEPKSCDKTHTCPPCPAPEAAGGPSVFLFPP KPKDTLMISRTPEVTCVVVDVSHEDPEVKFNW YVDGVEVHNAKTKPREEQYNSTYRVVSVLTVL HQDWLNGKEYKCKVSNKALPAPIEKTISKAKG QPREPQVYTLPPSREEMTKNQVSLTCLVKGFY PSDIAVEWESNGQPENNYKTTPPVLDSDGSFF LYSKLTVDKSRWQQGNVFSCSVMHEALHNHYT QKSLSLSPGK Ig Kappa constant SEQ ID NO:41 TVAAPSVFIFPPSDEQLKSGTASVVCLLNNFY region PREAKVQWKVDNALQSGNSQESVTEQDSKDST YSLSSTLTLSKADYEKHKVYACEVTHQGLSSP VTKSFNRGEC Ig Lambda SEQ ID NO:42 QPKAAPSVTLFPPSSEELQANKATLVCLISDF constant region YPGAVTVAWKADSSPVKAGVETTTPSKQSNNK YAASSYLSLTPEQWKSHRSYSCQVTHEGSTVE KTVAPTECS Furthermore, an antibody of the present invention or antigen-binding moiety thereof may be part of a larger immunoadhesion molecule formed by covalent or noncovalent association of said antibody or antibody moiety with one or more further proteins or peptides. Relevant to 30 such immunoadhesion molecules are the use of the streptavidin core region in order to pre pare a tetrameric scFv molecule (Kipriyanov, S.M., et al. (1995) Human Antibodies and Hybri domas 6:93-101) and the use of a cystein residue, a marker peptide and a C-terminal polyhis tidinyl, e. g. hexahistidinyl, tag in order to produce bivalent and biotinylated scFv molecules 5 (Kipriyanov, S.M., et al. (1994) Mol. Immunol. 31:1047-1058). The term "human antibody" refers to antibodies whose variable and constant regions corre spond to or are derived from immunoglobulin sequences of the human germ line, as described, for example, by Kabat et al. (see Kabat, et al. (1991) Sequences of Proteins of Immunological 10 Interest, Fifth Edition, U.S. Department of Health and Human Services, NIH Publication No. 91-3242). However, the human antibodies of the invention may contain amino acid residues not encoded by human germ line immunoglobulin sequences (for example mutations which have been introduced by random or site-specific mutagenesis in vitro or by somatic mutation in vivo), for example in the CDRs, and in particular in CDR3. Recombinant human antibodies of 15 the invention have variable regions and may also contain constant regions derived from immu noglobulin sequences of the human germ line (see Kabat, E.A., et al. (1991) Sequences of Proteins of Immunological Interest, Fifth Edition, U.S. Department of Health and Human Ser vices, NIH Publication No. 91-3242). According to particular embodiments, however, such re combinant human antibodies are subjected to in-vitro mutagenesis (or to a somatic in-vivo 20 mutagenesis, if an animal is used which is transgenic due to human Ig sequences) so that the amino acid sequences of the VH and VL regions of the recombinant antibodies are sequences which although related to or derived from VH and VL sequences of the human germ line, do not naturally exist in vivo within the human antibody germ line repertoire. According to particu lar embodiments, recombinant antibodies of this kind are the result of selective mutagenesis or 25 back mutation or of both. Preferably, mutagenesis leads to an affinity to the target which is greater, and/or an affinity to non-target structures which is smaller than that of the parent anti body. The term "chimeric antibody" refers to antibodies which contain sequences for the variable 30 region of the heavy and light chains from one species and constant region sequences from another species, such as antibodies having murine heavy and light chain variable regions linked to human constant regions. The term "CDR-grafted antibody" refers to antibodies which comprise heavy and light chain 35 variable region sequences from one species but in which the sequences of one or more of the CDR regions of VH and/or VL are replaced with CDR sequences of another species, such as antibodies having murine heavy and light chain variable regions in which one or more of the murine CDRs (e.g., CDR3) has been replaced with human CDR sequences.

31 The term "humanized antibody" refers to antibodies which contain sequences of the variable region of heavy and light chains from a nonhuman species (e.g. mouse, rat, rabbit, chicken, camelid, sheep or goat) but in which at least one part of the VH and/or VL sequence has been altered in order to be more "human-like", i.e. to be more similar to variable sequences of the 5 human germ line. One type of a humanized antibody is a CDR graft antibody in which human CDR sequences have been inserted into nonhuman VH and VL sequences to replace the cor responding nonhuman CDR sequences. The terms "Kabat numbering", "Kabat definitions" and "Kabat labeling" are used interchangea 10 bly herein. These terms, which are recognized in the art, refer to a system of numbering amino acid residues which are more variable (i.e. hypervariable) than other amino acid residues in the heavy and light chain variable regions of an antibody, or an antigen binding portion thereof (Kabat et al. (1971) Ann. NY Acad, Sci. 190:382-391 and Kabat, E.A., et al. (1991) Sequences of Proteins of Immunological Interest, Fifth Edition, U.S. Department of Health and Human 15 Services, NIH Publication No. 91-3242). For the heavy chain variable region, the hypervariable region ranges from amino acid positions 31 to 35 for CDR1, amino acid positions 50 to 65 for CDR2, and amino acid positions 95 to 102 for CDR3. For the light chain variable region, the hypervariable region ranges from amino acid positions 24 to 34 for CDR1, amino acid positions 50 to 56 for CDR2, and amino acid positions 89 to 97 for CDR3. 20 As used herein, the terms "acceptor" and "acceptor antibody" refer to the antibody or nucleic acid sequence providing or encoding at least 80%, at least 85%, at least 90%, at least 95%, at least 98% or 100% of the amino acid sequences of one or more of the framework regions. In some embodiments, the term "acceptor" refers to the antibody amino acid or nucleic acid se 25 quence providing or encoding the constant region(s). In yet another embodiment, the term "ac ceptor" refers to the antibody amino acid or nucleic acid sequence providing or encoding one or more of the framework regions and the constant region(s). In a specific embodiment, the term "acceptor' refers to a human antibody amino acid or nucleic acid sequence that provides or encodes at least 80%, preferably, at least 85%, at least 90%, at least 95%, at least 98%, or 30 100% of the amino acid sequences of one or more of the framework regions. In accordance with this embodiment, an acceptor may contain at least 1, at least 2, at least 3, least 4, at least 5, or at least 10 amino acid residues not occuring at one or more specific positions of a human antibody. An acceptor framework region and/or acceptor constant region(s) may be, e.g., de rived or obtained from a germline antibody gene, a mature antibody gene, a functional anti 35 body (e.g., antibodies well-known in the art, antibodies in development, or antibodies commer cially available). As used herein, the term "CDR" refers to the complementarity determining region within anti body variable sequences. There are three CDRs in each of the variable regions of the heavy 40 chain and of the light chain, which are designated CDR1, CDR2 and CDR3, for each of the 32 variable regions. The term "CDR set" as used herein refers to a group of three CDRs that oc cur in a single variable region capable of binding the antigen. The exact boundaries of these CDRs have been defined differently according to different systems. The system described by Kabat (Kabat et al., Sequences of Proteins of Immunological Interest (National Institutes of 5 Health, Bethesda, Md. (1987) and (1991)) not only provides an unambiguous residue number ing system applicable to any variable region of an antibody, but also provides precise residue boundaries defining the three CDRs. These CDRs may be referred to as Kabat CDRs. Chothia and coworkers (Chothia & Lesk, J. Mol. Biol. 196:901-917 (1987) and Chothia et al., Nature 342:877-883 (1989)) found that certain sub- portions within Kabat CDRs adopt nearly identical 10 peptide backbone conformations, in spite of great diversity at the level of amino acid se quence. These sub-portions were designated as L1, L2 and L3 or H1, H2 and H3 where the "L" and the "H" designates the light chain and the heavy chains regions, respectively. These regions may be referred to as Chothia CDRs, which have boundaries that overlap with Kabat CDRs. Other boundaries defining CDRs overlapping with the Kabat CDRs have been de 15 scribed by Padlan (FASEB J. 9:133-139 (1995)) and MacCallum (J Mol Biol 262(5):732-45 (1996)). Still other CDR boundary definitions may not strictly follow one of the above systems, but will nonetheless overlap with the Kabat CDRs, although they may be shortened or length ened in light of prediction or experimental findings that particular residues or groups of resi dues or even entire CDRs do not significantly impact antigen binding. The methods used 20 herein may utilize CDRs defined according to any of these systems, although preferred em bodiments use Kabat or Chothia defined CDRs. As used herein, the term "canonical" residue refers to a residue in a CDR or framework that defines a particular canonical CDR structure as defined by Chothia et al. (J. Mol. Biol. 196:901 25 907 (1987); Chothia et al., J. Mol. Biol. 227:799 (1992), both are incorporated herein by refer ence). According to Chothia et al., critical portions of the CDRs of many antibodies have nearly identical peptide backbone confirmations despite great diversity at the level of amino acid se quence. Each canonical structure specifies primarily a set of peptide backbone torsion angles for a contiguous segment of amino acid residues forming a loop. 30 As used herein, the terms "donor" and "donor antibody" refer to an antibody providing one or more CDRs. In a preferred embodiment, the donor antibody is an antibody from a species dif ferent from the antibody from which the framework regions are obtained or derived. In the con text of a humanized antibody, the term "donor antibody" refers to a non-human antibody pro 35 viding one or more CDRs. As used herein, the term "framework" or "framework sequence" refers to the remaining se quences of a variable region minus the CDRs. Because the exact definition of a CDR se quence can be determined using different systems, the meaning of a framework sequence is 40 subject to correspondingly different interpretations. The six CDRs (CDR-L1, -L2, and -L3 of 33 light chain and CDR-H1, -H2, and -H3 of heavy chain) also divide the framework regions on the light chain and the heavy chain into four sub-regions (FR1, FR2, FR3 and FR4) on each chain, in which CDRI is positioned between FR1 and FR2, CDR2 between FR2 and FR3, and CDR3 between FR3 and FR4. Without specifying the particular sub-regions as FR1, FR2, FR3 5 or FR4, a framework region, as referred by others, represents the combined FR's within the variable region of a single, naturally occurring immunoglobulin chain. As used herein, a FR represents one of the four sub- regions, and FRs represents two or more of the four sub- re gions constituting a framework region. 10 Human heavy chain and light chain acceptor sequences are known in the art. As used herein, the term "germline antibody gene" or "gene fragment" refers to an immu noglobulin sequence encoded by non- lymphoid cells that have not undergone the maturation process that leads to genetic rearrangement and mutation for expression of a particular immu 15 noglobulin. (See, e.g., Shapiro et al., Crit. Rev. Immunol. 22(3): 183-200 (2002); Marchalonis et al., Adv Exp Med Biol. 484:13-30 (2001)). One of the advantages provided by various em bodiments of the present invention stems from the finding that germline antibody genes are more likely than mature antibody genes are to conserve essential amino acid sequence struc tures characteristic of individuals in the species, hence less likely to be recognized as non-self 20 when used in that species. As used herein, the term "key" residues refers to certain residues within the variable region that have more impact on the binding specificity and/or affinity of an antibody, in particular a humanized antibody. A key residue includes, but is not limited to, one or more of the following: 25 a residue that is adjacent to a CDR, a potential glycosylation site (which can be either N- or 0 glycosylation site), a rare residue, a residue capable of interacting with the antigen, a residue capable of interacting with a CDR, a canonical residue, a contact residue between heavy chain variable region and light chain variable region, a residue within the Vernier zone, and a residue in the region that overlaps between the Chothia definition of a variable heavy chain CDR1 and 30 the Kabat definition of the first heavy chain framework. As used herein, the term "humanized antibody" specifically refers to an antibody or a variant, derivative, analog or fragment thereof which immunospecifically binds to an antigen of interest and which comprises a framework (FR) region having substantially the amino acid sequence of 35 a human antibody and a complementary determining region (CDR) having substantially the amino acid sequence of a non-human antibody. As used herein, the term "substantially" in the context of a CDR refers to a CDR having an amino acid sequence at least 80%, preferably at least 85%, at least 90%, at least 95%, at least 98% or at least 99% identical to the amino acid sequence of a non-human antibody CDR. A humanized antibody comprises substantially all of 40 at least one, and typically two, variable domains (Fab, Fab', F(ab') 2, FabC, Fv) in which all or 34 substantially all of the CDR regions correspond to those of a non-human immunoglobulin (i.e., donor antibody) and all or substantially all of the framework regions are those of a human im munoglobulin consensus sequence. Preferably, a humanized antibody also comprises at least a portion of an immunoglobulin constant region (Fc), typically that of a human immunoglobulin. 5 In some embodiments, a humanized antibody contains both the light chain as well as at least the variable domain of a heavy chain. The antibody also may include the CH1, hinge, CH2, CH3, and CH4 regions of the heavy chain. In some embodiments, a humanized antibody only contains a humanized light chain. In some embodiments, a humanized antibody only contains a humanized heavy chain. In specific embodiments, a humanized antibody only contains a 10 humanized variable domain of a light chain and/or humanized heavy chain. The humanized antibody can be selected from any class of immunoglobulins, including igM, IgG, IgD, IgA and IgE, and any subclass, including without limitation IgG 1, IgG2, IgG3 and igG4. 15 The framework and CDR regions of a humanized antibody need not correspond precisely to the parental sequences, e.g., the donor antibody CDR or the consensus framework may be mutagenized by substitution, insertion and/or deletion of at least one amino acid residue so that the CDR or framework residue at that site does not correspond exactly to either the donor 20 antibody or the consensus framework. In a preferred embodiment, such mutations, however, will not be extensive. Usually, at least 80%, preferably at least 85%, more preferably at least 90%, and most preferably at least 95% of the humanized antibody residues will correspond to those of the parental FR and CDR sequences. As used herein, the term "consensus frame work" refers to the framework region in the consensus immunoglobulin sequence. As used 25 herein, the term "consensus immunoglobulin sequence" refers to the sequence formed from the most frequently occurring amino acids (or nucleotides) in a family of related immunoglobu lin sequences (see e.g., Winnaker, From Genes to Clones (Verlagsgesellschaft, Weinheim, Germany 1987). In a family of immunoglobulins, each position in the consensus sequence is occupied by the amino acid occurring most frequently at that position in the family. Where two 30 amino acids occur equally frequently, either can be included in the consensus sequence. As used herein, "Vernier" zone refers to a subset of framework residues that may adjust CDR structure and fine-tune the fit to antigen as described by Foote and Winter (1992, J. Mol. Biol. 224:487-499, which is incorporated herein by reference). Vernier zone residues form a layer 35 underlying the CDRs and may impact on the structure of CDRs and the affinity of the antibody. The term "epitope" includes any polypeptide determinant capable of specific binding to an im munoglobulin. In certain embodiments, epitope determinants include chemically active surface groupings of molecules such as amino acids, sugar side chains, phosphoryl, or sulfonyl, and, 40 in certain embodiments, may have specific three dimensional structural characteristics, and/or 35 specific charge characteristics. An epitope is a region of an antigen that is bound by an anti body. In certain embodiments, an antibody is said to specifically bind an antigen when it pref erentially recognizes its target antigen in a complex mixture of proteins and/or macromole cules. 5 The term "polynucleotide" as referred to herein means a polymeric form of two or more nu cleotides, either ribonucleotides or deoxynucleotides or a modified form of either type of nu cleotide. The term includes single and double stranded forms of DNA but preferably is dou ble-stranded DNA. 10 The term "isolated polynucleotide" as used herein shall mean a polynucleotide (e.g., of ge nomic, cDNA, or synthetic origin, or any combination thereof) that, by virtue of its origin , the "isolated polynucleotide" is not associated with all or a portion of a polynucleotide with which the "isolated polynucleotide" is found in nature; is operably linked to a polynucleotide that it is 15 not linked to in nature; or does not occur in nature as part of a larger sequence. The term "vector", as used herein, is intended to refer to a nucleic acid molecule capable of transporting another nucleic acid to which it has been linked. One type of vector is a "plasmid", which refers to a circular double stranded DNA into which additional DNA segments may be 20 ligated. Another type of vector is a viral vector, wherein additional DNA segments may be ligated into the viral genome. Certain vectors are capable of autonomous replication in a host cell into which they are introduced (e.g., bacterial vectors having a bacterial origin of replication and episomal mammalian vectors). Other vectors (e.g., non-episomal mammalian vectors) can be integrated into the genome of a host cell upon introduction into the host cell, and thereby 25 are replicated along with the host genome. Moreover, certain vectors are capable of directing the expression of genes to which they are operatively linked. Such vectors are referred to herein as "recombinant expression vectors" (or simply, "expression vectors"). In general, ex pression vectors of utility in recombinant DNA techniques are often in the form of plasmids. In the present specification, "plasmid" and "vector" may be used interchangeably as the plasmid 30 is the most commonly used form of vector. However, the invention is intended to include such other forms of expression vectors, such as viral vectors (e.g., replication defective retroviruses, adenoviruses and adeno-associated viruses), which serve equivalent functions. The term "operably linked" refers to a juxtaposition wherein the components described are in a 35 relationship permitting them to function in their intended manner. A control sequence "operably linked" to a coding sequence is connected in such a way that expression of the coding se quence is achieved under conditions compatible with the control sequences. "Operably linked" sequences include both expression control sequences that are contiguous with the gene of interest and expression control sequences that act in trans or at a distance to control the gene 40 of interest. The term "expression control sequence" as used herein refers to polynucleotide 36 sequences which are necessary to effect the expression and processing of coding sequences to which they are ligated. Expression control sequences include appropriate transcription initia tion, termination, promoter and enhancer sequences; efficient RNA processing signals such as splicing and polyadenylation signals; sequences that stabilize cytoplasmic mRNA; sequences 5 that enhance translation efficiency (i.e., Kozak consensus sequence); sequences that enhance protein stability; and when desired, sequences that enhance protein secretion. The nature of such control sequences differs depending upon the host organism; in prokaryotes, such control sequences generally include promoter, ribosomal binding site, and transcription termination sequence; in eukaryotes, generally, such control sequences include promoters and transcrip 10 tion termination sequence. The term "control sequences" is intended to include components whose presence is essential for expression and processing, and can also include additional components whose presence is advantageous, for example, leader sequences and fusion partner sequences. 15 "Transformation", as defined herein, refers to any process by which exogenous DNA enters a host cell. Transformation may occur under natural or artificial conditions using various methods well known in the art. Transformation may rely on any known method for the insertion of for eign nucleic acid sequences into a prokaryotic or eukaryotic host cell. The method is selected based on the host cell being transformed and may include, but is not limited to, viral infection, 20 electroporation, lipofection, and particle bombardment. Such "transformed" cells include stably transformed cells in which the inserted DNA is capable of replication either as an autono mously replicating plasmid or as part of the host chromosome. They also include cells which transiently express the inserted DNA or RNA for limited periods of time. 25 The term "recombinant host cell" (or simply "host cell"), as used herein, is intended to refer to a cell into which exogenous DNA has been introduced. It should be understood that such terms are intended to refer not only to the particular subject cell, but, also to the progeny of such a cell. Because certain modifications may occur in succeeding generations due to either muta tion or environmental influences, such progeny may not, in fact, be identical to the parent cell, 30 but are still included within the scope of the term "host cell" as used herein. Preferably host cells include prokaryotic and eukaryotic cells selected from any of the kingdoms of life. Pre ferred eukaryotic cells include protist, fungal, plant and animal cells. Most preferably host cells include but are not limited to the prokaryotic cell line E.coli; mammalian cell lines CHO, HEK 293 and COS; the insect cell line Sf9; and the fungal cell Saccharomyces cerevisiae. 35 Standard techniques may be used for recombinant DNA, oligonucleotide synthesis, and tissue culture and transformation (e.g., electroporation, lipofection). Enzymatic reactions and purifica tion techniques may be performed according to manufacturer's specifications or as commonly accomplished in the art or as described herein. The foregoing techniques and procedures may 40 be generally performed according to conventional methods well known in the art and as de- 37 scribed in various general and more specific references that are cited and discussed through out the present specification. See e.g., Sambrook et al., Molecular Cloning: A Laboratory Man ual (2d ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. (1989)), which is incorporated herein by reference for any purpose. 5 "Transgenic organism", as known in the art and as used herein, refers to an organism having cells that contain a transgene, wherein the transgene introduced into the organism (or an an cestor of the organism) expresses a polypeptide not naturally expressed in the organism. A "transgene" is a DNA construct, which is stably and operably integrated into the genome of a 10 cell from which a transgenic organism develops, directing the expression of an encoded gene product in one or more cell types or tissues of the transgenic organism. Methods of producing antibodies of the invention are described below. A distinction is made here between in-vivo approaches, in-vitro approaches or a combination of both. 15 Some methods of producing antibodies of the invention are described below. A distinction is made here between in-vivo approaches, in-vitro approaches or a combination of both. In-vivo approaches 20 Depending on the type of the desired antibody, various host animals may be used for in-vivo immunization. A host expressing itself an endogenous version of the antigen of interest may be used. Alternatively, it is possible to use a host which has been made deficient in an endoge nous version of the antigen of interest. For example, mice which had been made deficient in a 25 particular endogenous protein via homologous recombination at the corresponding endoge nous gene (i.e. knockout mice) have been shown to generate a humoral response to the pro tein with which they have been immunized and therefore to be able to be used for production of high-affinity monoclonal antibodies to the protein (see, for example, Roes, J. et al. (1995) J. ImmunoL. Methods 183:231-237; Lunn, M.P. et al. (2000) J. Neurochem. 75:404-412). 30 A multiplicity of nonhuman mammals are suitable hosts for antibody production in order to pro duce nonhuman antibodies of the invention. They include mice, rats, chickens, camelids, rab bits, sheep and goats (and knockout versions thereof), although preference is given to mice for the production of hybridomas. Furthermore, a nonhuman host animal expressing a human an 35 tibody repertoire may be used for producing essentially human antibodies to a human antigen with dual specificity. Nonhuman animals of this kind include transgenic animals (e.g. mice) bearing human immunoglobulin transgenes (chimeric hu-PBMC SCID mice) and hu man/mouse irradiation chimeras which are described in more detail below.

38 According to one embodiment, the animal immunized with Ap(20-42) globulomer or derivative thereof is a nonhuman mammal, preferably a mouse, which is transgenic due to human immu noglobulin genes so that said nonhuman mammal makes human antibodies upon antigenic stimulation. Typically, immunoglobulin transgenes for heavy and light chains with human germ 5 line configuration are introduced into such animals which have been altered such that their endogenous heavy and light chain loci are inactive. If such animals are stimulated with antigen (e.g. with a human antigen), antibodies derived from the human immunoglobulin sequences (human antibodies) are produced. It is possible to make from the lymphocytes of such animals human monoclonal antibodies by means of standardized hybridoma technology. For a further 10 description of transgenic mice with human immunoglobulins and their use in the production of human antibodies, see, for example, US 5,939,598, WO 96/33735, WO 96/34096, WO 98/24893 and WO 99/53049 (Abgenix Inc.), and US 5,545,806, US 5,569,825, US 5,625,126, US 5,633,425, US 5,661,016, US 5,770,429, US 5,814,318, US 5,877,397 and WO 99/45962 (Genpharm Inc.); see also MacQuitty, J.J. and Kay, R.M. (1992) Science 257:1188; Taylor, 15 L.D. et al. (1992) Nucleic Acids Res. 20:6287-6295; Lonberg, N. et al. (1994) Nature 368:856 859; Lonberg, N. and Huszar, D. (1995) Int. Rev. Immunol. 13:65-93; Harding, F.A. and Lon berg, N. (1995) Ann. N. Y Acad. Sci. 764:536-546; Fishwild, D. M. et al. (1996) Nature Bio technology 14:845-851; Mendez, M. J. et al. (1997) Nature Genetics 15:146-156; Green, L.L. and Jakobovits, A. (1998) J. Exp. Med. 188:483-495; Green, L.L. (1999) J. Immunol. Methods 20 231:11-23; Yang, X.D. et a/. (1999) J. Leukoc. Biol. 66:401-410; Gallo, M.L. et al. (2000) Eur. J. Immunol. 30:534-540. According to another embodiment, the animal which is immunized with AP(20-42) globulomer or derivative thereof may be a mouse with severe combined immunodeficiency (SCID), which 25 has been reconstituted with human peripheral mononuclear blood cells or lymphoid cells or precursors thereof. Such mice which are referred to as chimeric hu-PBMC SCID mice produce human immunoglobulin responses upon antigenic stimulation, as has been proved. For a fur ther description of these mice and of their use for generating antibodies, see, for example, Leader, K.A. et a. (1992) Immunology 76:229-234; Bombil, F. et al. (1996) Immunobio/. 30 195:360-375; Murphy, W.J. et al. (1996) Semin. Immuno/. 8:233-241; Herz, U. et al. (1997) tnt. Arch. Allergy Immunol. 113:150-152; Albert, S.E. et al. (1997) J. Immunol. 159:1393-1403; Nguyen, H. et al. (1997) Microbiol. Immunol. 41:901-907; Arai, K. et al. (1998) J. Immunol. Methods 217:79-85; Yoshinari, K. and Arai, K. (1998) Hybridoma 17:41-45; Hutchins, W.A. et a. (1999) Hybridoma 18:121-129; Murphy, W.J. et al. (1999) Clin. Immunol. 90:22-27; Smith 35 son, S.L. et al. (1999) Mo/. Immunol. 36:113-124; Chamat, S. et al. (1999) J. Infect. Diseases 180:268-277; and Heard, C. et al. (1999) Molec. Med. 5:35-45. According to another embodiment, the animal which is immunized with Ap(20-42) globulomer or a derivative thereof is a mouse which has been treated with a lethal dose of total body irra 40 diation, then protected from radiation with bone marrow cells from mice with severe combined 39 immunodeficiency (SCID) and subsequently transplanted with functional human lymphocytes. This type of chimera, referred to as the Trimera system, is used in order to produce human monoclonal antibodies by immunizing said mice with the antigen of interest and then producing monoclonal antibodies by using standardized hybridoma technology. For a further description 5 of these mice and of their use for generating antibodies, see, for example, Eren, R. et a. (1998) Immunology 93:154-161; Reisner, Y and Dagan, S. (1998) Trends Biotechnol. 16:242 246; Ilan, E. et a/. (1999) Hepatology 29:553-562; and Bocher, W.O. et al. (1999) Immunology 96:634-641. 10 Starting from the in-vivo generated antibody-producing cells, monoclonal antibodies may be produced by means of standardized techniques such as the hybridoma technique originally described by Kohler and Milstein (1975, Nature 256:495-497) (see also Brown et al. (1981) J. Immunol 127:539-46; Brown et al. (1980) J Biol Chem 255:4980-83; Yeh et al. (1976) PNAS 76:2927-31; and Yeh et a/. (1982) /nt. J. Cancer 29:269-75). The technology of producing 15 monoclonal antibody hybridomas is sufficiently known (see generally R. H. Kenneth, in Mono clonal Antibodies: A New Dimension In Biological Analyses, Plenum Publishing Corp., New York, New York (1980); E. A. Lerner (1981) Yale J. Biol. Med., 54:387-402; M. L. Gefter et al. (1977) Somatic Cell Genet., 3:231-36). Briefly, an immortalized cell line (typically a myeloma) is fused with lymphocytes (typically splenocytes or lymph node cells or peripheral blood lym 20 phocytes) of a mammal immunized with the AP globulomer of the invention or derivative thereof, and the culture supernatants of the resulting hybridoma cells are screened in order to identify a hybridoma which produces a monoclonal antibody of the present invention. Any of the many well known protocols for fusing lymphocytes and immortalized cell lines can be ap plied for this purpose (see also G. Galfre et al. (1977) Nature 266:550-52; Gefter et a. Somatic 25 Cell Genet., cited supra; Lerner, Yale J. Biol. Med., cited supra; Kenneth, Monoclonal Antibod ies, cited supra). Moreover, the skilled worker will appreciate that there are diverse variations of such methods, which are likewise useful. Typically, the immortalized cell line (e.g. a mye loma cell line) is derived from the same mammalian species as the lymphocytes. For example, murine hybridomas may be established by fusing lymphocytes from a mouse immunized with 30 an immunogenic preparation of the invention with an immortalized mouse cell line. Preferred immortalized cell lines are mouse myeloma cell lines which are sensitive to culture medium containing hypoxanthine, aminopterine and thymidine (HAT medium). Any of a number of mye loma cell lines may be used by default as fusion partner, for example the P3-NS1/1-Ag4-1, P3 x63-Ag8.653 or Sp2/O-Ag14 myeloma lines. These myeloma cell lines are available from the 35 American Type Culture Collection (ATCC), Rockville, MD. Typically, HAT-sensitive mouse myeloma cells are fused to mouse splenocytes using polyethylene glycol (PEG). Hybridoma cells resulting from the fusion are then selected using HAT medium, thereby killing unfused and unproductively fused myeloma cells (unfused splenocytes die after several days because they are not transformed). Hybridoma cells producing monoclonal antibodies of the invention 40 are identified by screening the hybridoma culture supernatants for such antibodies, for exam- 40 ple by using a dot blot assay as described above and in example 8 in order to select those antibodies which have the binding affinities as defined above. The monoclonal antibodies 5F7, 10F11, 7C6, 487, 6A2, 2F2, 4D10, 7E5, 10C1, and 3810 all 5 have been generated using the above-described in-vivo approach and thereof are obtainable from a hybridoma as defined herein. Likewise, said hybridoma can be used as a source of nucleic acid encoding light and/or heavy chains in order to recombinantly produce antibodies of the present invention, as is described 10 below in further detail. In-vitro approaches As an alternative to producing antibodies of the invention by immunization and selection, anti 15 bodies of the invention may be identified and isolated by screening recombinant combinatorial immunoglobulin libraries with Ap(20-42) globulomer or derivative thereof to thereby isolate immunoglobulin library members which have the required binding affinity. Kits for generating and screening display libraries are commercially available (e.g. the Pharmacia Recombinant Phage Antibody System, catalog No. 27-9400-01; and the Stratagene SurfZAP* Phage Dis 20 play Kit, catalog No. 240612). In many embodiments, the display library is an scFv library or an Fab library. The phage display technique for screening recombinant antibody libraries has been adequately described. Examples of methods and compounds which can be used particu larly advantageously for generating and screening antibody display libraries can be found, for example, in McCafferty et al. WO 92/01047, US 5,969,108 and EP 589 877 (describes in par 25 ticular scFv display), Ladner et al. US 5,223,409, US 5,403,484, US 5,571,698, US 5,837,500 and EP 436 597 (describes pill fusion, for example); Dower et al. WO 91/17271, US 5,427,908, US 5,580,717 and EP 527 839 (describes in particular Fab display); Winter et al. International Publication WO 92/20791 and EP 368,684 (describes in particular the cloning of sequences for variable immunoglobulin domains); Griffiths et al. US 5,885,793 and EP 589 30 877 (describes in particular isolation of human antibodies to human antigens by using recom binant libraries); Garrard et al. WO 92/09690 (describes in particular phage expression techni ques); Knappik et al. WO 97/08320 (describes the human recombinant antibody library HuCal); Salfeld et al. WO 97/29131, (describes production of a recombinant human antibody to a hu man antigen (human tumor necrosis factor alpha) and also in-vitro affinity maturation of the 35 recombinant antibody) and Salfeld et al. U.S. Provisional Application No. 60/126,603 and the patent applications based hereupon (likewise describes production of recombinant human an tibodies to human antigen (human interteukin-12), and also in-vitro affinity maturation of the recombinant antibody).

41 Further descriptions of screenings of recombinant antibody libraries can be found in scientific publications such as Fuchs et al. (1991) BiolTechnology 9:1370-1372; Hay et al. (1992) Hum Antibod Hybridomas 3:81-85; Huse et al. (1989) Science 246:1275-1281; Griffiths et al. (1993) EMBO J 12:725-734; Hawkins et al. (1992) J Mol Biol 226:889-896; Clarkson et al. (1991) Na 5 ture 352:624-628; Gram et al. (1992) PNAS 89:3576-3580; Garrard et al. (1991) Bio/Technology 9:1373-1377; Hoogenboom et al. (1991) Nuc Acid Res 19:4133-4137; Barbas et al. (1991) PNAS 88:7978-7982; McCafferty et al, Nature (1990) 348:552-554; and Knappik et al. (2000) J. Mol. Bio/. 296:57-86. 10 As an alternative to using bacteriophage display systems, recombinant antibody libraries may be expressed on the surface of yeast cells or of bacterial cells. WO 99/36569 describes meth ods of preparing and screening libraries expressed on the surface of yeast cells. WO 98/49286 describes in more detail methods of preparing and screening libraries expressed on the sur face of bacterial cells. 15 In all in vitro approaches, a selection process for enriching recombinant antibodies with the desired properties form an integral part of the process, which is generally referred to as "pan ning" and often takes the form of affinity chromatography over columns to whose matrix the target structure has been attached. Promising candidate molecules are then subjected to indi 20 vidual determination of their absolute and/or relative affinities, preferably by means of a stan dardized dot blot assay, as described above and in example 8. Once an antibody of interest of a combinatorial library has been identified and sufficiently char acterized, the DNA sequences encoding the light and heavy chains of said antibody are iso 25 lated by means of standardized molecular-biological techniques, for example by means of PCR amplification of DNA from the display package (e.g. the phage) which has been isolated during library screening. Nucleotide sequences of genes for light and heavy antibody chains, which may be used for preparing PCR primers, are known to the skilled worker. A multiplicity of such sequences are described, for example, in Kabat, E.A., et al. (1991) Sequences of Pro 30 teins of Immunological /nterest, Fifth Edition, U.S. Department of Health and Human Services, NIH Publication No. 91-3242 and in the database of sequences of the human germ line VBASE. An antibody or antibody moiety of the invention may be produced by recombinantly expressing 35 the genes for light and heavy immunoglobulin chains in a host cell. In order to recombinantly express an antibody, a host cell is transfected with one or more recombinant expression vec tors carrying DNA fragments encoding the light and heavy immunoglobulin chains of said anti body, thereby expressing the light and heavy chains in the host cell and secreting them pref erably into the medium in which said host cells are cultured. The antibodies can be isolated 40 from this medium. Standardized recombinant DNA methods are used in order to obtain genes 42 for heavy and light antibody chains, to insert said genes into recombinant expression vectors and to introduce said vectors into host cells. Methods of this kind are described, for example, in Sambrook, Fritsch and Maniatis (eds.), Molecular Cloning; A Laboratory Manual, Second Edition, Cold Spring Harbor, N.Y., (1989), Ausubel, F.M. et al. (eds.) Current Protocols in Mo 5 lecular Biology, Greene Publishing Associates, (1989) and in US 4,816,397 by Boss et al.. Once DNA fragments encoding VH and VL segments of the antibody of interest have been obtained, said DNA fragments may be further manipulated using standardized recombinant DNA techniques, for example in order to convert the genes for variable regions to genes for full 10 length antibody chains, to genes for Fab fragments or to an scFv gene. These manipulations comprise linking a VL- or VH-encoding DNA fragment operatively to another DNA fragment encoding another protein, for example a constant antibody region or a flexible linker. The term "operatively linked" is to be understood here as meaning that the two DNA fragments are linked in such a way that the amino acid sequences encoded by said two DNA fragments re 15 main in frame. The isolated DNA encoding the VH region may be converted to a gene for a full length heavy chain by operatively linking the VH-region encoding DNA with another DNA molecule encoding heavy chain constant regions (CH1, CH2 and CH3). The sequences of human heavy chain 20 constant region genes are well known (see, for example, Kabat, E.A., et al. (1991) Sequences of Proteins of Immunological Interest, Fifth Edition, U.S. Department of Health and Human Services, NIH Publication No. 91-3242), and DNA fragments spanning said regions may be obtained by means of standardized PCR amplification. The heavy chain constant region may be a constant region from IgG1, IgG2, IgG3, igG4, igM, IgA, IgE or IgD, with preference being 25 given to a constant region from IgG, in particular IgG1 or IgG4. To obtain a gene for a heavy chain Fab fragment, the VH-encoding DNA may be operatively linked to another DNA mole cule encoding merely the heavy chain constant region CH1. The isolated DNA encoding the VL region may be converted to a gene for a full length light 30 chain (and a gene for an Fab light chain) by operatively linking the VL-encoding DNA to an other DNA molecule encoding the light chain constant region CL. The sequences of genes of the constant region of human light chain are well known (see Kabat, E.A., et al. (1991) Se quences of Proteins of Immunological Interest, Fifth Edition, U.S. Department of Health and Human Services, NIH Publication No. 91-3242), and DNA fragments spanning said regions 35 may be obtained by means of standardized PCR amplification. The light chain constant region may be a constant kappa or lambda region, a constant kappa region being preferred. In order to generate an scFv gene, the VH- and VL-encoding DNA fragments may be opera tively linked to another fragment encoding a flexible linker, for example the amino acid se 40 quence (Gly 4 -Ser) 3 so that the VH and VL sequences are expressed as a continuous single- 43 chain protein, with the VL and VH regions being linked to one another via said flexible linker (see Bird et aL. (1988) Science 242:423-426; Huston et aL. (1988) Proc. Nat. Acad. Sci. USA 85:5879-5883; McCafferty et aL., Nature (1990) 348:552-554). 5 Single domain VH and VL having the binding affinities as described above may be isolated from single domain libraries by the above-described methods. Two VH single-domain chains (with or without CH1) or two VL chains or a pair of one VH chain and one VL chain with the desired binding affinity may be useful as described herein for the antibodies of the invention. 10 In order to express the recombinant antibodies or antibody moieties of the invention, the DNAs encoding partial or full length light and heavy chains may be inserted into expression vectors so as to operatively link the genes to appropriate transcriptional and translational control se quences. In this context, the term "operatively linked" is to be understood as meaning that an antibody gene is ligated in a vector in such a way that transcriptional and translational control 1 5 sequences within the vector fulfill their intended function of regulating transcription and transla tion of said antibody gene. Expediently, the expression vector and the expression control sequences are chosen so as to be compatible with the expression host cell used. The gene for the antibody light chain and the 20 gene for the antibody heavy chain may be inserted into separate vectors or both genes are inserted into the same expression vector, this being the usual case. The antibody genes are inserted into the expression vector by means of standardized methods (for example by ligation of complementary restriction cleavage sites on the antibody gene fragment and the vector, or by ligation of blunt ends, if no restriction cleavage sites are present). The expression vector 25 may already carry sequences for antibody constant regions prior to insertion of the sequences for the light and heavy chains. For example, one approach is to convert the VH and VL se quences to full length antibody genes by inserting them into expression vectors already encod ing the heavy and, respectively, light chain constant regions, thereby operatively linking the VH segment to the CH segment(s) within the vector and also operatively linking the VL segment to 30 the CL segment within the vector. Additionally or alternatively, the recombinant expression vector may encode a signal peptide which facilitates secretion of the antibody chain from the host cell. The gene for said antibody chain may be cloned into the vector, thereby linking the signal peptide in frame to the N termi 35 nus of the gene for the antibody chain. The signal peptide may be an immunoglobulin signal peptide or a heterologous signal peptide (i.e. a signal peptide from a non-immunoglobulin pro tein). In addition to the genes for the antibody chain, the expression vectors of the invention may have regulatory sequences controlling expression of the genes for the antibody chain in a host cell. 40 44 The term "regulatory sequence" is intended to include promoters, enhancers and further ex pression control elements (e.g. polyadenylation signals) which control transcription or transla tion of the genes for the antibody chain. Regulatory sequences of this kind are described, for example, in Goeddel; Gene Expression Technology: Methods in Enzymology 185, Academic 5 Press, San Diego, CA (1990). The skilled worker will appreciate that the expression vector design which includes selection of regulatory sequences may depend on factors such as the choice of the host cell to be transformed, the desired strength of expression of the protein, etc. Preferred regulatory sequences for expression in mammalian host cells include viral elements resulting in strong and constitutive protein expression in mammalian cells, such as promoters 10 and/or enhancers derived from cytomegalovirus (CMV) (such as the CMV promoter/enhancer), simian virus 40 (SV40) (such as the SV40 promoter/enhancer), adenovirus (e.g. the adenovi rus major late promoter (AdMLP)) and polyoma. For a further description of viral regulatory elements and sequences thereof, see, for example, US 5,168,062 to Stinski, US 4,510,245 to Bell et al. and US 4,968,615 to Schaffner et al. 15 Apart from the genes for the antibody chain and the regulatory sequences, the recombinant expression vectors of the invention may have additional sequences such as those which regu late replication of the vector in host cells (e.g. origins of replication) and selectable marker genes. The selectable marker genes facilitate the selection of host cells into which the vector 20 has been introduced (see, for example, US patent Nos 4,399,216, 4,634,665 and 5,179,017, all to Axel et al.). For example, it is common for the selectable marker gene to render a host cell into which the vector has been inserted resistant to cytotoxic drugs such as G418, hygro mycin or methotrexate. Preferred selectable marker genes include the gene for dihydrofolate reductase (DHFR) (for use in dhfr host cells with methotrexate selection/amplification) and the 25 neo gene (for G418 selection). For expression of the light and heavy chains, the expression vector(s) encoding said heavy and light chains is(are) transfected into a host cell by means of standardized techniques. The various forms of the term "transfection" are intended to comprise a multiplicity of techniques 30 customarily used for introducing exogenous DNA into a prokaryotic or eukaryotic host cell, for example electroporation, calcium phosphate precipitation, DEAE-dextran transfection, and the like. Although it is theoretically possible to express the antibodies of the invention either in pro karyotic or eukaryotic host cells, preference is given to expressing the antibodies in eukaryotic cells and, in particular, in mammalian host cells, since the probability of a correctly folded and 35 immunologically active antibody being assembled and secreted is higher in such eukaryotic cells and in particular mammalian cells than in prokaryotic cells. Prokaryotic expression of an tibody genes has been reported as being ineffective for production of high yields of active anti body (Boss, M.A. and Wood, C. R. (1985) Immunology Today 6:12-13).

45 Preferred mammalian host cells for expressing recombinant antibodies of the invention include CHO cells (including dhfr- CHO cells described in Urlaub and Chasin, (1980) Proc. Natl. Acad. Sci. USA 77:4216-4220, which are used together with a DHFR-selectable marker, as de scribed, for example, in R.J. Kaufman and P.A. Sharp (1982) Mol. Biol. 159:601-621), NSO 5 myeloma cells, COS cells and SP2 cells. When introducing recombinant expression vectors encoding the antibody genes into mammalian host cells, the antibodies are produced by cultur ing the host cells until the antibody is expressed in said host cells or, preferably, the antibody is secreted into the culture medium in which the host cells grow. The antibodies may then be isolated from the culture medium by using standardized protein purification methods. 10 It is likewise possible to use host cells in order to produce moieties of intact antibodies, such as Fab fragments or scFv molecules. Variations of the above-described procedure are of course included in the invention. For example, it may be desirable to transfect a host cell with DNA encoding either the light chain or the heavy chain (but not both) of an antibody of the in 15 vention. If either light or heavy chains are present which are not required for binding of the an tigen of interest, then the DNA encoding either such a light or such a heavy chain or both may be removed partially or completely by means of recombinant DNA technology. Molecules ex pressed by such truncated DNA molecules are likewise included in the antibodies of the inven tion. In addition, it is possible to produce bifunctional antibodies in which a heavy chain and a 20 light chain are an antibody of the invention and the other heavy chain and the other light chain have specificity for an antigen different from the antigen of interest, by crosslinking an antibody of the invention to a second antibody by means of standardized chemical methods. In a preferred system for recombinant expression of an antibody of the invention or an antigen 25 binding moiety thereof, a recombinant expression vector encoding both the antibody heavy chain and the antibody light chain is introduced into dhfr CHO cells by means of calcium phosphate-mediated transfection. Within the recombinant expression vector, the genes for the heavy and light antibody chains are in each case operatively linked to regulatory CMV enhan cer/AdMLP-promoter elements in order to effect strong transcription of said genes. The re 30 combinant expression vector also carries a DHFR gene which can be used for selecting dhfr CHO cells transfected with the vector by using methotrexate selection/amplification. The se lected transformed host cells are cultured so that the heavy and light antibody chains are ex pressed, and intact antibody is isolated from the culture medium. Standardized molecular biological techniques are used in order to prepare the recombinant expression vector, to trans 35 fect the host cells, to select the transformants, to culture said host cells, and to obtain the anti body from the culture medium. Thus the invention relates to a method of synthesizing a re combinant antibody of the invention by culturing a host cell of the invention in a suitable culture medium until a recombinant antibody of the invention has been synthesized. The method may furthermore comprise isolating said recombinant antibody from said culture medium. 40 46 As an alternative to screening recombinant antibody libraries by phage display, other methods known to the skilled worker may be used for screening large combinatorial libraries to identify the antibodies of the invention. Basically, any expression system in which a close physical link age between a nucleic acid and the antibody encoded thereby is established and may be used 5 to select a suitable nucleic acid sequence by virtue of the properties of the antibody it encodes may be employed. In one type of an alternative expression system, the recombinant antibody library is expressed in the form of RNA-protein fusions, as described in WO 98/31700 to Szostak and Roberts, and 10 in Roberts, R.W. and Szostak, J.W. (1997) Proc. Natl. Acad. Sci. USA 94:12297-12302. In this system, in-vitro translation of synthetic mRNAs carrying on their 3' end puromycin, a peptidyl acceptor antibiotic, generates a covalent fusion of an mRNA and the peptide or protein en coded by it. Thus a specific mRNA of a complex mixture of mRNAs (e.g. a combinatorial li brary) may be concentrated on the basis of the properties of the encoded peptide or protein 15 (e.g. of the antibody or a moiety thereof), such as binding of said antibody or said moiety thereof to A(20-42) globulomer or a derivative thereof. Nucleic acid sequences which encode antibodies or moieties thereof and which are obtained by screening of such libraries may be expressed by recombinant means in the above-described manner (e.g. in mammalian host cells) and may, in addition, be subjected to further affinity maturation by either screening in 20 further rounds mRNA-peptide fusions, introducing mutations into the originally selected se quence(s), or using other methods of in-vitro affinity maturation of recombinant antibodies in the above-described manner. Combinations of in-vivo and in-vitro approaches 25 The antibodies of the invention may likewise be produced by using a combination of in-vivo and in-vitro approaches such as methods in which AO(20-42) globulomer or a derivative thereof is first allowed to act on an antibody repertoire in a host animal in vivo to stimulate pro duction of AP(20-42) globulomer- or derivative-binding antibodies and then further antibody 30 selection and/or antibody maturation (i.e. optimization) are accomplished with the aid of one or more in-vitro techniques. According to one embodiment, a combined method of this kind may comprise firstly immunizing a nonhuman animal (e.g. a mouse, rat, rabbit, chicken, camelid, sheep or goat or a transgenic version thereof or a chimeric mouse) with said AP(20-42) globu lomer or derivative thereof to stimulate an antibody response to the antigen and then preparing 35 and screening a phage display antibody library by using immunoglobulin sequences of lym phocytes which have been stimulated in vivo by the action of said AP(20-42) globulomer or derivative. The first step of this combined procedure may be carried out in the manner de scribed above in connection with the in-vivo approaches, while the second step of this proce dure may be carried out in the manner described above in connection with the in-vitro ap 40 proaches. Preferred methods of hyperimmunizing nonhuman animals with subsequent in-vitro 47 screening of phage display libraries prepared from said stimulated lymphocytes include those described by BioSite Inc., see, for example, WO 98/47343, WO 91/17271, US 5,427,908 and US 5,580,717. 5 According to another embodiment, a combined method comprises firstly immunizing a nonhu man animal (e.g. a mouse, rat, rabbit, chicken, camelid, sheep, goat or a knockout and/or transgenic version thereof, or a chimeric mouse) with an A(20-42) globulomer of the invention or derivative thereof to stimulate an antibody response to said AP(20-42) globulomer or deriva tive thereof and selecting the lymphocytes which produce the antibodies having the desired 10 specificity by screening hybridomas (prepared, for example, from the immunized animals). The genes for the antibodies or single domain antibodies are isolated from the selected clones (by means of standardized cloning methods such as reverse transcriptase polymerase chain reac tion) and subjected to in-vitro affinity maturation in order to improve thereby the binding proper ties of the selected antibody or the selected antibodies. The first step of this procedure may be 15 conducted in the manner described above in connection with the in-vivo approaches, while the second step of this procedure may be conducted in the manner described above in connection with the in-vitro approaches, in particular by using methods of in-vitro affinity maturation, such as those described in WO 97/29131 and WO 00/56772. 20 In a further combined method, the recombinant antibodies are generated from individual iso lated lymphocytes by using a procedure which is known to the skilled worker as selected lym phocyte antibody methods (SLAM) and which is described in US 5,627,052, WO 92/02551 and Babcock, J.S. et a. (1996) Proc. Natl. Acad. Sci. USA 93:7843-7848. In this method, a non human animal (e.g. a mouse, rat, rabbit, chicken, camelid, sheep, goat, or a transgenic version 25 thereof, or a chimeric mouse) is firstly immunized in vivo with Ap(20-42) globulomer or a de rivative thereof to stimulate an immune response to said oligomer or derivative, and then indi vidual cells secreting antibodies of interest are selected by using an antigen-specific haemo lytic plaque assay. To this end, the globulomer or derivative thereof or structurally related molecules of interest may be coupled to sheep erythrocytes, using a linker such as biotin, 30 thereby making it possible to identify individual cells secreting antibodies with suitable specific ity by using the haemolytic plaque assay. Following the identification of cells secreting antibod ies of interest, cDNAs for the variable regions of the light and heavy chains are obtained from the cells by reverse transcriptase PCR, and said variable regions may then be expressed in association with suitable immunoglobulin constant regions (e.g. human constant regions) in 35 mammalian host cells such as COS or CHO cells. The host cells transfected with the amplified immunoglobulin sequences derived from in vivo-selected lymphocytes may then be subjected to further in-vitro analysis and in-vitro selection by spreading out the transfected cells, for ex ample, in order to isolate cells expressing antibodies with the binding affinity. The amplified immunoglobulin sequences may furthermore be manipulated in vitro. 40 48 Antibodies having the required affinities defined herein can be selected by performing a dot blot essentially as described above. Briefly, the antigen is attached to a solid matrix, pref erably dotted onto a nitrocellulose membrane, in serial dilutions. The immobilized antigen is then contacted with the antibody of interest followed by detection of the latter by means of an 5 enzyme-conjugated secondary antibody and a colorimetric reaction; at defined antibody and antigen concentrations, the amount of antibody bound allows affinity determination. Thus the relative affinity of two different antibodies to one target, or of one antibody to two different tar gets, is here defined as the relation of the respective amounts of target-bound antibody ob served with the two antibody-target combinations under otherwise identical dot blot conditions. 10 Antibodies which bind to the same epitope as monoclonal antibody 5F7, 10F11, 7C6, 4B7, 6A2, 2F2, 4D10, 7E5, 1OC1, or 3B10 can be obtained in a manner known per se. In the same way as antibodies may be competing, described above, different target structures 1 5 are herein said to be "competing" for a particular antibody if at least one of these structures is capable of specifically reducing the measurable binding of another, preferably by offering an overlapping or identical epitope, more preferably an identical epitope. Competing target entities are useful for directly selecting antibodies by virtue of their relative 20 affinity to such target structures. Relative affinities may thus be determined directly by using a competition assay in which distinguishable forms of the competing entities, e. g. differently labelled competing structures, are contacted with the antibody of interest, and the relative affin ity of the antibody to each of these entities is deduced from the relative amounts of these enti ties which are bound by the antibody. 25 Such competition may be used to directly enrich for antibodies possessing a desired relative affinity to the target entity, by attaching the entity towards which greater affinity is desired to a solid matrix support and adding a suitable amount, preferably a molar excess, of the compet ing entity towards which smaller affinity is desired to the medium. Thus, the antibodies display 30 ing the desired relative affinities will tend to bind to the matrix more strongly than others and may be obtained after washing out the less desirable forms, e. g. by washing out at low salt concentrations and then harvesting the bound antibody by reversibly detaching it from its tar get by using high salt concentrations. If desired, several rounds of enrichment may be per formed. In a particular embodiment of the invention, where the genotype underlying an anti 35 body is physically linked to this antibody, e. g. in a pool of hybridomas or antigen-displaying phages or yeast cells, the corresponding phenotype may be rescued. In another embodiment of the invention, a modified dot blot is used where the immobilized an tigen competes with a solved entity for antibody binding, so that the relative affinity of the anti 40 body can be deduced from the percentage bound to the immobilized antigen.

49 Antibody moieties such as Fab and F(ab') 2 fragments may be produced from whole antibodies by using conventional techniques such as digestion with papain or pepsin. In addition, antibod ies, antibody moieties and immunoadhesion molecules may be obtained by using standardized 5 recombinant DNA techniques. The present invention also relates to pharmaceutical agents (compositions) comprising an an tibody of the invention and, optionally, a pharmaceutically suitable carrier. Pharmaceutical compositions of the invention may furthermore contain at least one additional therapeutic 10 agent, for example one or more additional therapeutic agents for the treatment of a disease for whose relief the antibodies of the invention are useful. If, for example, the antibody of the in vention binds to a globulomer of the invention, the pharmaceutical composition may further more contain one or more additional therapeutic agents useful for the treatment of disorders in which the activity of said globulomer is important. 15 Pharmaceutically suitable carriers include any solvents, dispersing media, coatings, antibacte rial and antifungal agents, isotonic and absorption-delaying agents, and the like, as long as they are physiologically compatible. Pharmaceutically acceptable carriers include, for example, water, saline, phosphate-buffered saline, dextrose, glycerol, ethanol and the like, and combina 20 tions thereof. In many cases, preference is given to using isotonic agents, for example sugars, polyalcohols such as mannitol or sorbitol, or sodium chloride in addition. Pharmaceutically suit able carriers may furthermore contain relatively small amounts of auxiliary substances such as wetting agents or emulsifiers, preservatives or buffers, which increase the half life or efficacy of the antibodies. 25 The pharmaceutical compositions may be suitable for parenteral administration, for example. Here, the antibodies are prepared preferably as injectable solutions with an antibody content of 0.1 - 250 mg/ml. The injectable solutions may be prepared in liquid or lyophilized form, the dosage form being a flint glass or vial, an ampoule or a filled syringe. The buffer may contain 30 L-histidine (1 - 50 mM, preferably 5 - 10 mM) and have a pH of 5.0 - 7.0, preferably of 6.0. Further suitable buffers include, without being limited thereto, sodium succinate, sodium cit rate, sodium phosphate or potassium phosphate buffers. Sodium chloride may be used in or der to adjust the tonicity of the solution to a concentration of 0 - 300 mM (preferably 150 mM for a liquid dosage form). Cryoprotectants, for example sucrose (e.g. 0 - 10%, preferably 0.5 35 1.0%) may also be included for a lyophilized dosage form. Other suitable cryoprotectants are trehalose and lactose. Fillers, for example mannitol (e.g. 1 - 10%, preferably 2 - 4%) may also be included for a lyophilized dosage form. Stabilizers, for example L-methionine (e.g. 51 - 50 mM, preferably 5 - 10 mM) may be used both in liquid and lyophilized dosage forms. Further suitable fillers are glycine and arginine. Surfactants, for example polysorbate 80 (e.g. 0 - 50 0.05%, preferably 0.005 - 0.01%), may also be used. Further surfactants are polysorbate 20 and BRIJ surfactants. The compositions of the invention may have a multiplicity of forms. These include liquid, semi 5 solid and solid dosage forms, such as liquid solutions (e.g. injectable and infusible solutions), dispersions or suspensions, tablets, pills, powders, liposomes and suppositories. The preferred form depends on the intended type of administration and on the therapeutic application. Typi cally, preference is given to compositions in the form of injectable or infusible solutions, for example compositions which are similar to other antibodies for passive immunization of hu 10 mans. The preferred route of administration is parenteral (e.g. intravenous, subcutaneous, intraperitoneal or intramuscular). According to a preferred embodiment, the antibody is admin istered by intravenous infusion or injection. According to another preferred embodiment, the antibody is administered by intramuscular or subcutaneous injection. 15 Therapeutic compositions must typically be sterile and stable under preparation and storage conditions. The compositions may be formulated as solutions, microemulsions, dispersions, liposomes or other ordered structures suitable for high concentrations of active substance. Sterile injectable solutions may be prepared by introducing the active compound (i.e. the anti body) in the required amount into a suitable solvent, where appropriate with one or a combina 20 tion of the abovementioned ingredients, as required, and then sterile-filtering said solution. Dispersions are usually prepared by introducing the active compound into a sterile vehicle con taining a basic dispersion medium and, where appropriate, other required ingredients. In the case of a sterile lyophilized powder for preparing sterile injectable solutions, vacuum drying and spray drying are preferred methods of preparation, which produces a powder of the active 25 ingredient and, where appropriate, of further desired ingredients from a previously sterile filtered solution. The correct flowability of a solution may be maintained by using, for example, a coating such as lecithin, by maintaining, in the case of dispersions the required particle size or by using surfactants. A prolonged absorption of injectable compositions may be achieved by additionally introducing into the composition an agent which delays absorption, for example 30 monostearate salts and gelatine. The antibodies of the invention may be administered by a multiplicity of methods known to the skilled worker, although the preferred type of administration for many therapeutic applications is subcutaneous injection, intravenous injection or infusion. The skilled worker will appreciate 35 that the route and/or type of administration depend on the result desired. According to particu lar embodiments, the active compound may be prepared with a carrier which protects the compound against rapid release, such as, for example, a formulation with sustained or con trolled release, which includes implants, transdermal plasters and microencapsulated release systems. Biologically degradable biocompatible polymers such as ethylene vinyl acetate, poly 40 anhydrides, polyglycolic acid, collagen, polyorthoesters and polylactic acid may be used. The 51 methods of preparing such formulations are well known to the skilled worker; see, for example, Sustained and Controlled Release Drug Delivery Systems, J.R. Robinson, ed., Marcel Dekker, Inc., New York, 1978. 5 According to particular embodiments, an antibody of the invention may be administered orally, for example in an inert diluent or a metabolizable edible carrier. The antibody (and further in gredients, if desired) may also be enclosed in a hard or soft gelatine capsule, compressed to tablets or added directly to food. For oral therapeutic administration, the antibodies may be mixed with excipients and used in the form of oral tablets, buccal tablets, capsules, elixirs, 10 suspensions, syrups and the like. If it is intended to administer an antibody of the invention via a route other than the parenteral one, it may be necessary to choose a coating from a material which prevents its inactivation. The present invention also relates to a method of inhibiting the activity of globulomers of the 15 invention in an individual which suffers from a disorder in which the amyloid P protein is in volved and in which in particular the activity of said globulomers of the invention is important. Said method comprises the administration of at least one antibody of the invention to the indi vidual with the aim of inhibiting the activity of the globulomer to which the antibody binds. Said individual is preferably a human being. An antibody of the invention may be administered for 20 therapeutic purposes to a human individual. In addition, an antibody of the invention may be administered to a nonhuman mammal for veterinary purposes or within the framework of an animal model for a particular disorder. Such animal models may be useful for evaluating the therapeutic efficacy of antibodies of the invention (for example for testing dosages and time courses of administration). 25 Disorders in which the globulomers of the invention play a part include in particular disorders in whose development and/or progression a globulomer of the invention is involved. These are in particular those disorders in which globulomers of the invention are evidently or presumably responsible for the pathophysiology of said disorder or are a factor which contributes to the 30 development and/or progression of said disorder. Accordingly, those disorders are included here in which inhibition of the activity of globulomers of the invention can relieve symptoms and/or progression of the disorder. Such disorders can be verified, for example, by an in creased concentration of globulomers of the invention in a biological fluid of an individual suf fering from a particular disorder (e.g. increased concentration in serum, plasma, CSF, urine, 35 etc.). This may be detected, for example, by using an antibody of the invention. The globu lomers of the invention play an important part in the pathology associated with a multiplicity of disorders in which neurodegenerative elements, cognitive deficiencies, neurotoxic elements and inflammatory elements are involved.

52 In another aspect of the invention, disorders that can be treated or prevented include those associated with amyloidoses. The term "amyloidoses" here denotes a number of disorders characterized by abnormal folding, clumping, aggregation and/or accumulation of particular proteins (amyloids, fibrous proteins and their precursors) in various tissues of the body. In Alz 5 heimer's disease and Down's syndrome, nerve tissue is affected, and in cerebral amyloid an giopathy (CAA) blood vessels are affected. The pharmaceutical compositions of the invention may include a "therapeutically effective amount" or a "prophylactically effective amount" of an antibody or antibody moiety of the inven 10 tion. A "therapeutically effective amount" refers to an amount effective, at dosages and for pe riods of time necessary, to achieve the desired therapeutic result. A therapeutically effective amount of the antibody or antibody moiety may be determined by a person skilled in the art and may vary according to factors such as the disease state, age, sex, and weight of the indi vidual, and the ability of the antibody or antibody moiety to elicit a desired response in the indi 15 vidual. A therapeutically effective amount is also one in which any toxic or detrimental effects of the antibody or antibody portion are outweighed by the therapeutically beneficial effects. A "prophylactically effective amount" refers to an amount effective, at dosages and for periods of time necessary, to achieve the desired prophylactic result. Typically, since a prophylactic dose is used in subjects prior to or at an earlier stage of disease, the prophylactically effective 20 amount will be less than the therapeutically effective amount. Moreover, the present invention includes a further method of preventing or treating Alzheimer's disease in a patient in need of such prevention or treatment. This method comprises the step of administering the vaccine noted above to the patient in an amount sufficient to effect the 25 prevention or treatment. Further, the present invention encompasses a method of identifying compounds suitable for active immunization of a patient predicted to develop an amyloidosis, e.g. Alzheimer's disease. This method comprises: 1) exposing one or more compounds of interest to one or more of the 30 antibodies described above for a time and under conditions sufficient for the one or more com pounds to bind to the antibody or antibodies; 2) identifying those compounds which bind to the antibody or antibodies, the identified compounds to be used in active immunization in a patient predicated to develop an amyloidosis, e.g. Alzheimer's disease. 35 Within the framework of diagnostic usage of the antibodies, qualitative or quantitative specific globulomer determination serves in particular to diagnose disease-relevant amyloid P forms. In this context, specificity means the possibility of being able to detect a particular globulomer or a derivative thereof, or a mixture thereof with sufficient sensitivity. The antibodies of the inven- 53 tion advantageously have detection threshold concentrations of less than 10 ng/ml of sample, preferably of less than 1 ng/ml of sample and particularly preferably of less than 100 pg/ml of sample, meaning that at least the concentration of globulomer per ml of sample, indicated in each case, advantageously also lower concentrations, can be detected by the antibodies of the 5 invention. The detection is carried out immunologically. This may be carried out in principle by using any analytical or diagnostic assay method in which antibodies are used, including agglutination and precipitation techniques, immunoassays, immunohistochemical methods and immunoblot tech 10 niques, for example Western blotting or, preferably, dot blot methods. In vivo methods, for ex ample imaging methods, are also included here. The use in immunoassays is advantageous. Competitive immunoassays, i.e. assays where antigen and labelled antigen (tracer) compete for antibody binding, and sandwich immunoas 15 says, i.e. assays where binding of specific antibodies to the antigen is detected by a second, usually labelled antibody, are both suitable. These assays may be either homogeneous, i.e. without separation into solid and liquid phases, or heterogeneous, i.e. bound labels are sepa rated from unbound ones, for example via solid phase-bound antibodies. Depending on label ling and method of measurement, the various heterogeneous and homogeneous immunoas 20 say formats can be classified into particular classes, for example RIAs (radioimmunoassays), ELISA (enzyme-linked immunosorbent assay), FIA (fluorescence immunoassay), LIA (lumi nescence immunoassay), TRFIA (time-resolved FIA), IMAC (immunoactivation), EMIT (en zyme-multiplied immune test), TIA (turbidometric immunoassay), I-PCR (immuno-PCR). 25 For the globulomer quantification of the invention, preference is given to competitive immuno assays in which a defined amount of labelled globulomer derivative serving as tracer competes with the globulomer of the sample (containing an unknown amount of unlabelled globulomers) to be quantified for binding to the antibody used. The amount of antigen, i.e. the amount of globulomer, in the sample can be determined from the amount of the displaced tracer with the 30 aid of a standard curve. Of the labels available for these purposes, enzymes have proved advantageous. Systems based on peroxidases, in particular horseradish peroxidase, alkaline phosphatase and P-D galactosidase, may be used, for example. Specific substrates whose conversion can be moni 35 tored photometrically, for example, are available for these enzymes. Suitable substrate sys tems are based on p-nitrophenyl phosphate (p-NPP), 5-bromo-4-chloro-3-indoly phos phate/nitroblue tetrazolium (BCIP/NPT), Fast-Red/naphthol-AS-TS phosphate for alkaline phosphatase; 2,2-azinobis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS), o phenylenediamine (OPT), 3,3',5,5'-tetramethylbenzidine (TMB), o-dianisidine, 5-aminosalicylic 40 acid, 3-dimethylaminobenzoic acid (DMAB) and 3-methyl-2-benzothiazotinehydrazone (MBTH) 54 for peroxidases; o-nitrophenyi-p-D-galactoside (o-NPG), p-nitrophenyl-p-D-galactoside and 4 methylumbelliphenyl-p-D-galactoside (MUG) for p-D-galactosidase. In many cases, these sub strate systems are commercially available in a ready-to-use form, for example in the form of tablets which may also contain further reagents such as appropriate buffers and the like. 5 The tracers used may be labelled globulomers. In this sense, a particular globulomer can be determined by labelling the globulomer to be determined and using it as tracer. The coupling of labels to globulomers for preparing tracers may be carried out in a manner 10 known per se. The comments above on derivatization of globulomers of the invention are re ferred to by analogy. In addition, a number of labels appropriately modified for conjugation to proteins are available, for example biotin-, avidin-, extravidin- or streptavidin-conjugated en zymes, maleimide-activated enzymes and the like. These labels may be reacted directly with the oligomer or, if required, with the appropriately derivatized globulomer to give the tracer. If, 15 for example, a streptavidin-peroxidase conjugate is used, then this firstly requires biotinylation of the globulomer. This applies correspondingly to the reverse order. Suitable methods to this end are also known to the skilled worker. If a heterogeneous immunoassay format is chosen, the antigen-antibody complex may be 20 separated by binding it to the support, for example via an anti-idiotypical antibody coupled to said support, e.g. an antibody directed against rabbit IgG. Appropriate supports, in particular microtiter plates coated with appropriate antibodies, are known and partly commercially avail able. 25 The present invention further relates to immunoassay sets having at least one antibody as described above and further components. Said sets are, usually in the form of a packaging unit, a combination of means for carrying out a globulomer determination of the invention. For the purpose of as easy handling as possible, said means are preferably provided in an essen tially ready-to-use form. An advantageous arrangement offers the immunoassay in the form of 30 a kit. A kit usually comprises multiple containers for separate arrangement of components. All components may be provided in a ready-to-use dilution, as a concentrate for diluting or as a dry substance or lyophilisate for dissolving or suspending; individual or all components may be frozen or stored at room temperature until use. Sera are preferably shock-frozen, for example at -20 0 C so that in these cases an immunoassay has to be kept preferably at temperatures 35 below freezing prior to use. Further components supplied with the immunoassay depend on the type of said immunoassay. Usually, standard protein, tracer which may or may not be required and control serum are sup plied together with the antiserum. Furthermore, microtiter plates, preferably antibody-coated, 55 buffers, for example for testing, for washing or for conversion of the substrate, and the enzyme substrate itself may also be included. General principles of immunoassays and generation and use of antibodies as auxiliaries in 5 laboratory and hospital can be found, for example, in Antibodies, A Laboratory Manual (Har low, E., and Lane, D., Ed., Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1988). Thus, the present invention also includes a method of diagnosing an amyloidosis, e.g. Alz heimer's disease, in a patient suspected of having this disease. This method comprises the 10 steps of: 1) isolating a biological sample from the patient; 2) contacting the biological sample with at least one of the antibodies described above for a time and under conditions sufficient for formation of antigen/antibody complexes; and 3) detecting presence of the antigen/antibody complexes in said sample, presence of the complexes indicating a diagnosis of an amyloido sis, e.g. Alzheimer's disease, in the patient. The antigen may be, for example, an globulomer 15 or a portion or fragment thereof which has the same functional properties as the full globu lomer (e.g., binding activity). Further, the present invention includes another method of diagnosing an amyloidosis, e.g. Alz heimer's disease in a patient suspected of having this disease. This method comprising the 20 steps of: 1) isolating a biological sample from the patient; 2) contacting the biological sample with an antigen for a time and under conditions sufficient for the formation of antibody/antigen complexes; 3) adding a conjugate to the resulting antibody/antigen complexes for a time and under conditions sufficient to allow the conjugate to bind to the bound antibody, wherein the conjugate comprises one of the antibodies described above, attached to a signal generating 25 compound capable of generating a detectable signal; and 4) detecting the presence of an anti body which may be present in the biological sample, by detecting a signal generated by the signal generating compound, the signal indicating a diagnosis of an amyloidosis, e.g. Alz heimer's disease in the patient. The antigen may be a globulomer or a portion or fragment thereof having the same functional properties as the full globulomer (e.g., binding activity). 30 The present invention includes an additional method of diagnosing an amyloidosis, e.g. Alz heimer's disease in a patient suspected of having an amyloidosis, e.g. Alzheimer's disease. This method comprises the steps of: 1) isolating a biological sample from said patient; 2) con tacting the biological sample with anti-antibody, wherein the anti-antibody is specific for one of 35 the antibodies described above, for a time and under conditions sufficient to allow for formation of anti-antibody/antibody complexes, the complexes containing antibody present in the biologi cal sample; 2) adding a conjugate to resulting anti-antibody/antibody complexes for a time and under conditions sufficient to allow the conjugate to bind to bound antibody, wherein the conju gate comprises an antigen, which binds to a signal generating compound capable of generat- 56 ing a detectable signal; and 3) detecting a signal generated by the signal generating com pound, the signal indicating a diagnosis of an amyloidosis, e.g. Alzheimer's disease in the pa tient. 5 Also, the present invention includes a kit comprising: a) at least one of the antibodies de scribed above and b) a conjugate comprising an antibody attached to a signal-generating compound, wherein the antibody of the conjugate is different from the isolated antibody. The present invention also encompasses a kit comprising: a) an anti-antibody to one of the 10 antibodies described above and b) a conjugate comprising an antigen attached to a signal generating compound. The antigen may be a globulomer or a fragment or portion thereof hav ing the same functional characteristics as the globulomer (e.g., binding activity). In one diagnostic embodiment of the present invention, an antibody of the present invention, or 15 a portion thereof, is coated on a solid phase (or is present in a liquid phase). The test or bio logical sample (e.g., whole blood, cerebrospinal fluid, serum, etc.) is then contacted with the solid phase. If antigen (e.g., globulomer) is present in the sample, such antigens bind to the antibodies on the solid phase and are then detected by either a direct or indirect method. The direct method comprises simply detecting presence of the complex itself and thus presence of 20 the antigens. In the indirect method, a conjugate is added to the bound antigen. The conjugate comprises a second antibody, which binds to the bound antigen, attached to a signal generating compound or label. Should the second antibody bind to the bound antigen, the sig nal-generating compound generates a measurable signal. Such signal then indicates presence of the antigen in the test sample. 25 Examples of solid phases used in diagnostic immunoassays are porous and non-porous mate rials, latex particles, magnetic particles, microparticles (see U.S. Patent No. 5,705,330), beads, membranes, microtiter wells and plastic tubes. The choice of solid phase material and method of labeling the antigen or antibody present in the conjugate, if desired, are determined based 30 upon desired assay format performance characteristics. As noted above, the conjugate (or indicator reagent) will comprise an antibody (or perhaps anti-antibody, depending upon the assay), attached to a signal-generating compound or label. This signal-generating compound or "label" is itself detectable or may be reacted with one or 35 more additional compounds to generate a detectable product. Examples of signal-generating compounds include chromogens, radioisotopes (e.g., 1251, 131l, 32P, 3H, 35S and 14C), chemiluminescent compounds (e.g., acridinium), particles (visible or fluorescent), nucleic ac- 57 ids, complexing agents, or catalysts such as enzymes (e.g., alkaline phosphatase, acid phos phatase, horseradish peroxidase, beta-galactosidase and ribonuclease). In the case of en zyme use (e.g., alkaline phosphatase or horseradish peroxidase), addition of a chromo-, fluro-, or lumo-genic substrate results in generation of a detectable signal. Other detection systems 5 such as time-resolved fluorescence, internal-reflection fluorescence, amplification (e.g., poly merase chain reaction) and Raman spectroscopy are also useful. Examples of biological fluids which may be tested by the above immunoassays include plasma, whole blood, dried whole blood, serum, cerebrospinal fluid or aqueous or organo 10 aqueous extracts of tissues and cells. The present invention also encompasses a method for detecting the presence of antibodies in a test sample. This method comprises the steps of: (a) contacting the test sample suspected of containing antibodies with anti-antibody specific for the antibodies in the patient sample under 15 time and conditions sufficient to allow the formation of anti-antibody/antibody complexes, wherein the anti-antibody is an antibody of the present invention which binds to an antibody in the patient sample; (b) adding a conjugate to the resulting anti-antibody/antibody complexes, the conjugate comprising an antigen (which binds to the anti-antibody) attached to a signal generating compound capable of detecting a detectable signal; and (d) detecting the presence 20 of the antibodies which may be present in the test sample by detecting the signal generated by the signal generating compound. A control or calibrator may be used which comprises anti body to the anti-antibody. Kits are also included within the scope of the present invention. More specifically, the present 25 invention includes kits for determining the presence of antigens (e.g., globulomers) in a patient suspected of having Alzheimer's disease or another condition characterized by cognitive im pairment. In particular, a kit for determining the presence of antigens in a test sample com prises a) an antibody as defined herein or moiety thereof; and b) a conjugate comprising a second antibody (having specificity for the antigen) attached to a signal generating compound 30 capable of generating a detectable signal. The kit may also contain a control or calibrator which comprises a reagent which binds to the antigen. The present invention also includes a kit for detecting antibodies in a test sample. The kit may comprise a) an anti-antibody specific (for example, one of the subject invention) for the anti 35 body of interest, and b) an antigen or portion thereof as defined above. A control or calibrator comprising a reagent which binds to the antigen may also be included. More specifically, the kit may comprise a) an anti-antibody (such as the one of the present invention) specific for the antibody and b) a conjugate comprising an antigen (e.g., globulomer) attached to a signal gen erating compound capable of generating a detectable signal. Again, the kit may also comprise 40 a control of calibrator comprising a reagent which binds to the antigen.

58 The kit may also comprise one container such as vial, bottles or strip, with each container with a pre-set solid phase, and other containers containing the respective conjugates. These kits may also contain vials or containers of other reagents needed for performing the assay, such 5 as washing, processing and indicator reagents. It should also be noted that the subject invention not only includes the full length antibodies described above but also moities or fragments thereof, for example, the Fab portion thereof. Additionally, the subject invention encompasses any antibody having the same properties of 10 the present antibodies in terms of, for example, binding specificity, structure, etc. Advantages of the invention: By immunization with AP(20-42) globulomer different monoclonal antibodies may be obtained 15 which differ in their tolerance or recognition of different AP(1-42) oligomers and A(X-42) oli gomers, as determined by comparative dot blotting as described above. This allows develop ment of an antibody directed to N-terminally truncated AP oligomers which possesses an opti mal relation between cognition enhancing effect, desired specificity over other AP forms and minimal side effect profile. Surprisingly, the AP1(1-42) and Ap~(12-42) globulomers, in spite of 20 containing the structural element, are only partly recognized by antibodies obtained by using the further truncated Ap(20-42) globulomer as antigen. By restriction of the specific oligomeric AP form to the basic structural principle (i.e. use of the N-terminally truncated AP globulomers) an antibody profile is generated in active immunization 25 which is highly specific for oligomeric AP forms. The same holds true for monoclonal antibodies for use in passive immunization. The advantage of such a specific strategy for immunization (active and passive) is that it will not induce an immune response against AP monomers, AP peptides in fibrillary states of aggregation or sAPPa. This is advantageous in several ways: 30 1) In the form of insoluble AP plaques AP peptides in fibrillary states of aggregation amount to the major part of the entire AP peptide pool in AD brains. A massive release of As by dissolution of AP plaques induced by reaction of anti-As antibodies with these plaques is to be regarded as detrimental. This massive release of AP would then cross the blood-brain barrier, enter the bloodstream and potentially increase the risk of mi 35 crohaemorrhages. In addition, in the ELAN trial this very strategy of immunization with fibrillary As peptide forms required cancellation of the trial due to 6% of cases with an onset of meningoencephalitis. 2) Immune responses directed to monomeric AP peptide forms are undesirable, as it was 40 possible to show that the latter may exert cognition-enhancing effects.

59 3) Immune responses directed to sAPPa are likewise undesirable, as this might lead to a reaction with the physiologically occurring precursor protein APP and thus to an auto immune reaction. Moreover, sAPPa was also shown to exert cognition-enhancing ef 5 fects. 4) A response directed to vascular As peptide in the form of CAA is to be avoided in order to eschew the undesirable side effect of microhaemorrhages (antibodies against the central portion of As and which in addition do not bind to AP-peptides aggregated in the 10 form of CAA induce fewer microhaemorrhages when compared to such against the N terminus, see above). 5) Antibodies which specifically react with As oligomers will have higher bioavailability with regard to the pathophysiologically relevant AP species, as they will not be bound 15 to, e.g., fibrillary AP and thus made unavailable for therapeutic effect. Deposit Information: The hybridoma which produces monoclonal antibody 5F7 was deposited with the American Type Culture Collection, 10801 University Boulevard, Manassas, Virginia 20110 on December 01, 2005 under the terms of the Budapest Treaty and received designa 20 tion PTA-7241. Further, the hybridoma which produces monoclonal antibody 1 OF1 1 was de posited with the American Type Culture Collection, 10801 University Boulevard, Manassas, Virginia 10801 on December 01, 2005 under the terms of the Budapest Treaty and received designation PTA-7239. Additionally, the hybridoma which produces monoclonal antibody 4B7 was deposited with the American Type Culture Collection, 10801 University Boulevard, Ma 25 nassas, Virginia 10801 on December 01, 2005 under the terms of the Budapest Treaty and received designation PTA-7242, and the hybridoma which produces monoclonal antibody 7C6 was deposited with the American Type Culture Collection, 10801 University Boulevard, Ma nassas, Virginia 10801 on December 01, 2005 under the terms of the Budapest Treaty and received designation PTA-7240. Additionally, the hybridoma which produces monoclonal anti 30 body 6A2 was deposited with the American Type Culture Collection, 10801 University Boule vard, Manassas, Virginia 10801 on February 28, 2006 under the terms of the Budapest Treaty and received designation PTA-7409, and the hybridoma which produces monoclonal antibody 2F2 was deposited with the American Type Culture Collection, 10801 University Boulevard, Manassas, Virginia 10801 on February 28, 2006 under the terms of the Budapest Treaty and 35 received designation PTA-7408. The hybridoma which produces monoclonal antibody 4D10 was deposited with the American Type Culture Collection, 10801 University Boulevard, Ma nassas, Virginia 10801 on February 28, 2006 under the terms of the Budapest Treaty and re ceived designation PTA-7405. The hybridoma which produces monoclonal antibody 7E5 was deposited with the American Type Culture Collection, 10801 University Boulevard, Manassas, 40 Virginia 10801 on August 16, 2006 under the terms of the Budapest Treaty and received des- 60 ignation PTA-7809. The hybridoma which produces monoclonal antibody 1OC1 was deposited with the American Type Culture Collection, 10801 University Boulevard, Manassas, Virginia 10801 on August 16, 2006 under the terms of the Budapest Treaty and received designation PTA-7810. The hybridoma which produces monoclonal antibody 3B10 was deposited with the 5 American Type Culture Collection, 10801 University Boulevard, Manassas, Virginia 10801 on September 01, 2006 under the terms of the Budapest Treaty and received designation PTA 7851. All deposits have been made on behalf of Abbott Laboratories, 100 Abbott Park Road, Abbott Park, Illinois 60064 (US). 10 In the drawings: Figure 1 shows size-exclusion chromatograms of AP(1-42) and Ap(1-40). Ap(1-42) monomer was dissolved in A) 0.1% NH 4 0H, B) 70% formic acid C) 0.1% NaOH 15 and in D) Ap(1-40) was dissolved in 0.1% NaOH. Subsequently, the samples were further diluted 1:10 in 20mM NaH 2

PO

4 , 140mM NaC, pH 7.4 These sam ples were incubated for 5 min (left column) or 1 hour (right column) after disso lution at ambient temperature, then applied to the size exclusion column; Figure 2 A) shows an SDS PAGE of standard proteins (molecular marker proteins, lane 20 1); Ap~(1-42) fibril preparation; control (lane 2); Ap~(1-42) fibril preparation + mAb 5F7, 20h, 37*C, supernatant (lane 3); Ap(1-42) fibril preparation + mAb 5F7, 20h, 370C, pellet (lane 4); Ap(1-42) fibril preparation + mAb 6E10, 20h, 370C, supernatant (lane 5); Ap(1-42) fibril preparation + mAb 6E10, 20h 37 0 C, pellet (lane 6); ); 25 B) shows the results of the quantitative analysis of mAbs bound to Ap-fibrils in percent of total antibody; Figure 3 is a bar diagram which shows the results of the object recognition test with APP/L transgenic mice after active immunization with AP(1-42) monomers in 0.1% NH 4 0H, Ap(1-42) globulomers and Ap(20-42) globulomers as compared 30 to wild-type mice (positive control) and PBS-treated APP/L mice (negative con trol), where circles indicate significant differences to PBS-treated APP/L mice and asterisks indicate highly significant differences to chance level (50%) ac cording to post-hoc t-test after P<0.05 in ANOVA for differences among groups; Figure 4 shows dot blots of the reactivity of 100 pmol/pl (row A); 10 pmol/pl (row B); 35 1 pmol/pl (row C); 0.1 pmol/pl (row D) and 0.01 pmol/pl (row E) of Ap(1-42) globulomer (column 1); of HFIP pretreated AP(1-42) monomer in Pluronic F68 (cbldr , 2); of AP(20-42) globulomer (column 3); of Ap1(12-42) globulormher (col umn 4); of HFIP pretreated Ap(1-40) monomer in DMSO (column 5); of Ap(1- 61 42) monomer, NH 4 0H (column 6); of an Ap(1-42) fibril preparation (column 7); and of sAPPa from Sigma (column 8) with various antisera obtained after an ac tive immunization of APP/L Tg mice with AP(20-42) globulomer; Figure 5 is a bar diagram which shows the concentrations of soluble and insoluble AP(1 5 42) and AP(1-40) peptide in brain extracts of actively immunized APP/PS1 Tg mice with either Ap(1-42) monomer (0.1% NH 4 0H), AP(1-42) globulomer, AP(20-42) globulomer or vehicle as control; Figure 6 is a bar diagram which shows the results of the object recognition test with APP/L transgenic mice after passive immunization with anti AP(20-42) globu 10 lomer antibodies 5F7, 1OF1 1, and 7C6 as compared to control mice for A) each antibody separately and B) for all antibodies taken together; Figure 7 A) shows a dot blot analysis of the specificity of different anti-As antibodies (-6E10, -5F7, -4B7, -10F1, -6A2, -4D10, -3B10, -2F2, -7C6, -7E5, -1OC1). The monoclonal antibodies tested here were obtained by active immunization of 15 mice with AP(20-42) globulomer followed by selection of the fused hybridoma cells (except for the commercial available 6E10, Signet No 9320). The individual As forms were applied in serial dilutions and incubated with the respective monoclonal antibodies for immune reaction: 1. Ap(1-42) monomer, 0.1% NH 4 0H 20 2. AP(1-40) monomer, 0.1% NH 4 0H 3. Ap(1-42) monomer, 0.1% NaOH 4. Ap(1-40) monomer, 0.1% NaOH 5. Ap(1-42) globulomer 6. Ap(12-42) globulomer 25 7. Ap(20-42) globulomer 8. Ap(1-42) fibril preparation 9. sAPPa (Sigma) (first dot: 1 pmol) B) Quantitative evaluation was done using a densitometric analysis of the inten sity. For each AP form, only the dot corresponding to the lowest antigen concen 30 tration was evaluated provided that it had a relative density of greater than 20% of the relative density of the last optically unambiguously identified dot of the Ap(20-42) globulomer (threshold). This threshold value was determined for every dot-blot independently. The value indicates the relation between recogni tion of AO(20-42) globulomer and the respective As form for the antibody given; 62 Figure 8 shows the binding of antibodies at different concentrations to transversal sec tions of the neocortices of Alzheimer's disease (AD) patients or old APP trans genic mice: A) Verification of amyloid deposits by Congo Red staining as plaques in brain 5 tissue and as cerebral amyloid angiopathy (CAA) in brain vessels in the APP transgenic mouse line Tg2576 and in an AD patient (RZ55); B) Strong staining of parenchymal deposits of AP(amyloid plaques) in an AD pa tient (RZ16) occurs only with 6G1 and the commercially available antibody 6E10 (left column) while antibodies 5F7, 2F2, and 6A2 (second column), 4D10, 10 1OF11, and 3810 (third column) and 7C6, 7E5, and 10C1 (right column) show no staining. All antibodies were used at a concentration of 0.7 pg/ml; C) Strong staining of parenchymal deposits of AP(amyloid plaques) in 19 month old Tg2576 occurs only with 6G1 and the commercially available antibody 6E10 (left column) while antibodies 5F7, 2F2, and 6A2 (second column), 4D10, 15 1OF11, and 3B10 (third column) and 7C6, 7E5, and 1OC1 (right column) show no staining. All antibodies were used at a concentration of 0.7 pg/ml; D)-G) Quantification of the analysis of Applaque staining in the histological im ages using image analysis. Optical density values (0% = no staining) were cal culated from the greyscale values of plaques subtracted by greyscale values of 20 background tissue: D) Staining at 0.7 pg/ml antibody in old Tg2576 mice, E) staining at 3 different concentrations of antibodies in APP/L mice, F) staining at 0.7 pg/ml antibody in an AD patient (RZ55), and G) staining at 3 different con centrations of antibodies in an AD patient (RZ16). The differences between staining of the commercially available antibodies 6E10 (asterisks) and 4G8 (cir 25 cles) and all other antibodies (three asterisks/circles: p<0.001 versus control; post-hoc Bonferroni's t-test after ANOVA with p<0.001) were statistically evalu ated (D,F). In E) and G) all antibodies except 6G1 showed always significantly less staining than the commercially available antibodies 6E10 and 4G8 (p<0.001 in post-hoc t-test after p<0.001 in ANOVA). 30 H) Strong staining of vascular deposits of As (arrows) occurs only with 6G1 and the commercially available antibody 6E10 (left column) while antibodies 5F7, 2F2, and 6A2 (second column), 4D10, 1 OF1 1, and 3B10 (third column) and 7C6, 7E5, and 10C1 (right column) show no staining. All antibodies were used at a concentration of 0.7 pg/ml. A qualitatively similar situation was found in 35 Tg2576 mice (not shown here); Figure 9 Anti-Ap-antibody titer and dot-blot selectivity profile in plasma of TG2576 mice approximately one year after active immunization. Plasma samples of Tg2576 mice approximately one year after the last immunization with A) Ap (20-42) 63 globulomer, B) AP (12-42) globulomer, C) AP (1-42) monomer and D) vehicle, were assessed for anti-As antibodies produced and still present by dot-blot. 1. AR (1-42) globulomer 2. A0 (1-42) monomer, HFIP pretreated, in 0.1% Pluronic F68 5 3. A0 (20-42) globulomer 4. AP (12-42) globulomer 5. AP (1-40) monomer, HFIP pre-treated, 5mM in DMSO 6. AP (1-42) monomer, 0.1% NH 4 0H 7. AP (1-42) fibril preparation 10 8. sAPPa (Sigma); (first dot: 1pmol); Figure 10 shows a table summarizing the levels of Ap(20-42) globulomer in brain tissue of human beings having Alzheimer's disease and a non demented control; Figure 11 illustrates nucleotide and amino acid sequences of the variable heavy and light chains of monoclonal antibodies (mAbs) as follows (complementarity determin 15 ing regions (CDRs) are underlined in each amino acid sequence): Fig. 11 SEQ ID NO: Sequence Type Chain mAb Al 1 nucleotide variable heavy (VH) 5F7 A2 2 nucleotide variable light (VL) 5F7 Al 3 amino acid variable heavy (VH) 5F7 A2 4 amino acid variable light (VL) 5F7 B1 5 nucleotide variable heavy (VH) 1OF11 B2 6 nucleotide variable light (VL) 1OF1 1 81 7 amino acid variable heavy (VH) 1OF11 B2 8 amino acid variable light (VL) 1OF1 1 C1 9 nucleotide variable heavy (VH) 7C6 C2 10 nucleotide variable light (VL) 7C6 C1 11 amino acid variable heavy (VH) 7C6 C2 12 amino acid variable light (VL) 7C6 D1 13 nucleotide variable heavy (VH) 4B7 D2 14 nucleotide variable light (VL) 4B7 D1 15 amino acid variable heavy (VH) 4B7 D2 16 amino acid variable light (VL) 4B7 64 El 17 nucleotide variable heavy (VH) 2F2 E2 18 nucleotide variable light (VL) 2F2 El 19 amino acid variable heavy (VH) 2F2 E2 20 amino acid variable light (VL) 2F2 Fl 21 nucleotide variable heavy (VH) 6A2 F2 22 nucleotide variable light (VL) 6A2 Fl 23 amino acid variable heavy (VH) 6A2 F2 24 amino acid variable light (VL) 6A2 G1 25 nucleotide variable heavy (VH) 4D10 G2 26 nucleotide variable light (VL) 4D10 G1 27 amino acid variable heavy (VH) 4D10 G2 28 amino acid variable light (VL) 4D10 Hl 29 nucleotide variable heavy (VH) 7E5 H2 30 nucleotide variable light (VL) 7E5 Hi 31 amino acid variable heavy (VH) 7E5 H2 32 amino acid variable light (VL) 7E5 11 33 nucleotide variable heavy (VH) 1oCi 12 34 nucleotide variable light (VL) loCi 11 35 amino acid variable heavy (VH) loCi 12 36 amino acid variable light (VL) loCi J1 37 nucleotide variable heavy (VH) 3B10 J1 38 amino acid variable heavy (VH) 3810 The following examples are intended to illustrate the invention, without limiting its scope. Example 1: Preparation of globulomers 5 a) A@(1-42) globulomer The AP(1-42) synthetic peptide (H- 1368, Bachem, Bubendorf, Switzerland) was suspended in 100% 1,1,1,33,3-hexafluoro-2-propano (HFIP) at 6 mg/mL and incubated for complete solubi lization under shaking at 37 0C for 1.5 h. The HFIP acts as a hydrogen-bond breaker and is 65 used to eliminate pre-existing structural inhomogeneities in the AP peptide. HFIP was removed by evaporation in a SpeedVac and Ap(1-42) resuspended at a concentration of 5 mM in di methylsulfoxide and sonicated for 20 s. The HFIP-pre-treated Ap(1-42) was diluted in phos phate-buffered saline (PBS) (20 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4) to 400 pM and 1/10 5 volume 2% sodium dodecyl sulfate (SDS) (in H 2 0) added (final concentration of 0.2% SDS). An incubation for 6 h at 37 "C resulted in the 16/20-kDa AS(1-42) globulomer (short form for globular oligomer) intermediate. The 38/48-kDa AP(1-42) globulomer was generated by a fur ther dilution with three volumes of H 2 0 and incubation for 18 h at 37 C. After centrifugation at 3000 g for 20 min the sample was concentrated by ultrafiltration (30-kDa cut-off), dialysed 10 against 5 mM NaH 2

PO

4 , 35mM NaCl, pH 7.4, centrifuged at 10000 g for 10 min and the super natant comprsing the 38/48-kDa Ap(1-42) globulomer withdrawn. As an alternative to dialysis the 38/48-kDa AP(1-42) globulomer could also be precipitated by a ninefold excess (v/v) of ice cold methanol/acetic acid solution (33% methanol, 4% acetic acid) for 1 h at 4'C. The 38/48 kDa Ap(1-42) globulomer is then pelleted (10 min at 16200 g), resuspended in 5 mM 15 NaH 2

PO

4 , 35 mM NaCl, pH 7.4, and the pH adjusted to 7.4. b) Cross-linked Ap~(1-42) globulomer The Ap(1-42) synthetic peptide (H-1368, Bachem, Bubendorf, Switzerland) was suspended in 100% 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP) at 6 mg/mL and incubated for complete solubi 20 lization under shaking at 37 0C for 1.5 h. The HFIP acts as a hydrogen-bond breaker and was used to eliminate pre-existing structural inhomogeneities in the AP peptide. HFIP was removed by evaporation in a SpeedVac and AP(1-42) resuspended at a concentration of 5 mM in di methylsulfoxide and sonicated for 20 s. The HFIP-pre-treated AP(1-42) was diluted in phos phate-buffered saline (PBS) (20 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4) to 400 pM and 1/10 25 volume 2% sodium dodecyl sulfate (SDS) (in H 2 0) added (final concentration of 0.2% SDS). An incubation for 6 h at 37 *C resulted in the 16/20-kDa AP~(1-42) globulomer (short form for globular oligomer) intermediate. The 38/48-kDa AP(1-42) globulomer was generated by a fur ther dilution with three volumes of H 2 0 and incubation for 18 h at 37 *C. Cross-linking of the 38/48-kDa Ap(1-42) globulomer was now performed by incubation with 1 mM glutaraldehyde 30 for 2 h at 21 0C room temperature (RT) followed by ethanolamine (5 mM) treatment for 30 min at RT. c) Ap(20-42) globulomer 1.59 ml of Ap1(1-42) globulomer preparation prepared according to example 1a were admixed 35 with 38 ml of buffer (50 mM MES/NaOH, pH 7.4) and 200 pl of a 1 mg/ml thermolysin solution (Roche) in water. The reaction mixture was stirred at RT for 20 h. Then 80 pl of a 100 mM EDTA solution, pH 7.4, in water were added and the mixture was furthermore adjusted to an SDS content of 0.01% with 400 pl of a 1% strength SDS solution. The reaction mixture was concentrated to approx. 1 ml via a 15 ml 30 kDa Centriprep tube. The concentrate was ad- 66 mixed with 9 ml of buffer (50 mM MES/NaOH, 0.02 % SDS, pH 7.4) and again concentrated to 1 ml. The concentrate was dialyzed at 6*C against 1 1 of buffer (5 mM sodium phosphate, 35 mM NaCl) in a dialysis tube for 16 h. The dialysate was adjusted to an SDS content of 0.1% with a 2% strength SDS solution in water. The sample was centrifuged at 10000 g for 10 min 5 and the Ap(20-42) globulomer supernatant was withdrawn. d) Ap(12-42) globulomer 2 ml of an Ap(1-42) globulomer preparation prepared according to example la were admixed with 38 ml buffer (5 mM sodium phosphate, 35 mM sodium chloride, pH 7.4) and 150 pl of a 1 10 mg/ml GluC endoproteinase (Roche) in water. The reaction mixture was stirred for 6 h at RT, and a further 150 pl of a 1 mg/ml GluC endoproteinase (Roche) in water were subsequently added. The reaction mixture was stirred at RT for another 16 h, followed by addition of 8 pl of a 5 M DIFP solution. The reaction mixture was concentrated to approx. 1 ml via a 15 ml 30 kDa Centriprep tube. The concentrate was admixed with 9 ml of buffer (5 mM sodium phosphate, 15 35 mM sodium chloride, pH 7.4) and again concentrated to 1 ml. The concentrate was dialyzed at 6 0 C against 1 1 of buffer (5 mM sodium phosphate, 35 mM NaCl) in a dialysis tube for 16 h. The dialysate was adjusted to an SDS content of 0.1% with a 1% strength SDS solution in wa ter. The sample was centrifuged at 10000 g for 10 min and the Apr(12-42) globulomer super natant was withdrawn. 20 Example 2: Size-exclusion chromatography of different AP(1-42) monomer and Ap(1-40) monomer preparations Ap(1-42), 0.1% NH 4 0H: 25 1 mg of Ap(1 -42) (Bachem, catalogue no. H-1368) were dissolved in 500 pl of 0.1% NH 4 0H in

H

2 0 and agitated for 1 min at ambient temperature. The sample was centrifuged for 5 min at 10'000 g. The supernatant was collected. AP(1-42) concentration in the supernatant was de termined according to Bradford's method (BIO-RAD). 30 5 min sample: 20 pl of Ap(1-42) in the 0.1% NH 4 0H containing supernatant were diluted with 20 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4 to an Ap(1-42) concentration of 0.2 mg/ml. The sample was incubated for 5 min at ambient temperature. Then 100 pl were analyzed by size exclusion chromatography (SEC). 35 1 hour sample: 20 pl of AP(1-42) in the 0.1% NH 4 0H containing supernatant were diluted with 20 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4 to an Apr(1-42) concentration of 0.2 mg/ml. The sample was 67 incubated for 1 hour at ambient temperature. Then 100 pl were analyzed by size exclusion chromatography (SEC). AP(1-42), 70% HCOOH: 5 1 mg of AP(1-42) were dissolved in 50 pl 70% HCOOH in H 2 0 and agitated 1 min at ambient temperature. The sample was centrifuged for 5 min at 10'000 g. The supernatant was col lected. AP(1-42) concentration in the supernatant is determined according to Bradford's method (BIO-RAD). 10 5 min sample: 2 pl of Ap(1-42) in 70% HCOOH were diluted to a concentration of 0.2 mg/ml AS3(1-42) with 20 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4 and adjusted to pH 7.4 with 1 M NaOH. The sample was incubated for 5 min at ambient temperature. Then 100 pl were analyzed by size exclusion chromatography. 15 1 hour sample: 2 pl of Ap~(1-42) in 70% HCOOH were diluted to a concentration of 0.2 mg/ml A1(1-42) with 20 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4 and adjusted to pH 7.4 with 1 M NaOH. The sample was incubated for 1 hour at ambient temperature. Then 100 pl were analyzed by size exclusion 20 chromatography. A(1-42), 0.1% NaOH: 1 mg of A1(1-42) (Bachem, catalogue no. H-1368) were dissolved in 500 pl of 0.1% NaOH in

H

2 0 and agitated 1 min at ambient temperature. The sample was centrifuged for 5 min at 25 10'000 g. The supernatant was collected. Ap(1-42) concentration in the supernatant is deter mined according to Bradford's method (BIO-RAD). 5 min sample: 20 pl of A1(1-42) in 0.1% NaOH were diluted to a concentration of 0.2 mg/mI A@3(1-42) with 20 30 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4. The sample was incubated for 5 min at ambient tem perature. Then 100 pl were analyzed by size exclusion chromatography. 1 hour sample: 20 pl of AP(1-42) in 0.1% NaOH were diluted to a concentration of 0.2 mg/ml A@(1-42) with 20 35 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4. The sample was incubated for 1 hour at ambient tem perature. Then 100 pl were analyzed by size exclusion chromatography. AP(1-40). 0.1% NaOH : 1 mg of AP(1-40) (Bachem, catalogue no. H-1 194) were dissolved in 500 pl of 0.1% NaOH in 40 H 2 0 and agitated 1 min at ambient temperature. The sample is was centrifuged for 5 min at 68 10'000 g. The supernatant was collected. AP1(1-42) concentration in the supernatant was de termined according to Bradford's method (BIO-RAD). 5 min sample: 5 20 pl of AP(1-40) in 0.1% NaOH were diluted to a concentration of 0.2 mg/mI Ap1(1-40) with 20 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4. The sample was incubated for 5 min at ambient tem perature. Then 100 pl were analyzed by size exclusion chromatography. 1 hour sample: 10 20 pi of Ap(1-40) in 0.1% NaOH were diluted to a concentration of 0.2 mg/mI Ap(1-40) with 20 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4. The sample was incubated for 1 hour at ambient tem perature. Then 100 pl were analyzed by size exclusion chromatography. Conditions for size exclusion chromatography (SEC): 15 SEC column: Superose 12 HR 10/300 GL (Amersham, catalogue no. 17-5173-01) Flow: 0.5 ml/min Paper feed: 0.2 cm/min Extinction at 214 nm: 0 - 0.2 absorption units Mobile phase: 20 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4 20 Results are shown in Figure 1. The preparation of a purely monomeric Ap-solution is a great challenge due to the strong ten dency of the As peptide, especially the AD(1 -42) monomer, to aggregate into fibrils. Neverthe less, for the screening and characterization of anti-Ap(20-42) globulomers that discriminate 25 Ap(1-42)-monomers and AP(1-40)-monomers the best technically achievable Ap-monomeric preparation should be used. Here the effect of the initial solvent of the As peptide on the ag gregation effect after further dilution into 20 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4 was tested. The As peptide supplier (Bachem) states in their technical information that the Ap(1-42) should be solubilized in 0.1% NH 4 0H. Five minutes at room temperature (RT) after solubilizing the 30 Ap(1-42) in NH 4 0H and immediate further 1:10 dilution in 20 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4 a size-exclusion chromatography shows first signs of Ap(1-42) aggregation to fibrillary pre cursors with a minor peak at 74kD. Monomeric AP(1-42) runs at a major peak with 11 kD and a shoulder at 6kD. After incubation for one hour at room temperature (RT) the Ap(1-42) peptide in NH 4 0H has already aggregated to a high extent to Ap(1-42) fibrils leading to a loss of de 35 tectable material that did not enter the size-exclusion chromatographic column. If 70% formic acid is used as the initial solvent for AP(1-42) peptide a high extent of aggregation after 1 h at RT occurs with only a minor fraction AP(1-42) monomer left (note that the formic acid itself leads to a high background absorption at the protein detection wavelength). The best initial solvent for APf(1-42) to prevent aggregation is 0.1% NaOH which even after 1 h incubation of 69 solubilization and further dilution shows only a minor fraction of aggregated AP(1-42) with the majority of Ap(1-42) being still monomeric. Ap(1-40) solubilized initially in 0.1% NaOH shows no signs at all of aggregation even after 1 h at RT incubation. 5 Example 3: Semi-quantitative analysis visualized by SDS-PAGE of the discrimination of AP(20 42) globulomer selective antibodies for Ap(1-42) fibrils. AP(1-42) fibril preparation: 1 mg of AP(1-42) (Bachem, Cat. no.: H-1368) were dissolved in 500 pl 0.1% NH 4 0H in H 2 0 10 and agitated for 1 min at ambient temperature. The sample was centrifuged for 5 min at 10'000 g. The supernatant was collected. AP(1-42) concentration in the supernatant was determined according to Bradford's method (BIO-RAD). 100 pl of APf(1-42) in 0.1% NH 4 0H were mixed with 300 pl of 20 mM NaH 2

PO

4 , 140 mM NaCl, 15 pH 7.4 and adjusted to pH 7.4 with 2% HCl. The sample was then incubated at 37 *C for 20 hours. Following which the sample was centrifuged for 10 min at 10'000 g. The supernatant was discarded, and the residue was mixed with 400 p1 of 20 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4, resuspended by vigorous agitation ("vortexing") for 1 min and centrifuged for 10 min at 10'000 g. The supernatant was discarded and the residue was mixed with 400 pl of 20 mM 20 NaH 2

PO

4 , 140 mM NaCl, pH 7.4, resuspended by vigorous agitation ("vortexing") for 1 min and centrifuged for 10 min at 10'000 g once more. The supernatant was discarded. The residue was resuspended in 380 pl of 20 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4 and prompted by vigor ous agitation ("vortexing"). 25 Binding of anti-As antibodies to Ap(1-42) fibrils: 40 pl of Ap(1-42) fibril preparation were diluted with 160 pl of 20 mM NaH 2

PO

4 , 140 mM NaCl, 0.05% Tween 20, pH 7.4 and agitated 5 min at ambient temperature, then the sample was centrifuged for 10 min at 10'000 g. The supernatant was discarded, and the residue was re suspended in 95 pl of 20 mM NaH 2

PO

4 , 140 mM NaCl, 0.05% Tween 20, pH 7.4. Resuspen 30 sion was prompted by vigorous agitation ("vortexing"). Aliquots of 10 pl of the fibril preparation were each mixed with: a) 10 pl 20 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4 b) 10 p1 0.5 pg/pl of 5F7 in 20 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4 35 c) 10 p1 0.5 pg/pl of 6E10 (Signet Nr.: 9320) in 20 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4 The samples were incubated at 37 'C for 20 hours, then centrifuged for 10 min at 10'000 g. The supernatants were collected and mixed with 20 p1 of SDS-PAGE sample buffer. The resi dues were mixed with 50 pl of 20 mM NaH 2

PO

4 , 140 mM NaCl, 0.025% Tween 20, pH 7.4 and 70 resuspended by "vortexing", then the samples were centrifuged for 10 min at 10'000 g. The supernatants were discarded, and the residues were mixed with 20 pl 20 mM NaH 2

PO

4 , 140 mM NaCl, 0.025% Tween 20, pH 7.4, then with 20 pl of SDS-PAGE sample buffer. The sam ples were applied to a 4 - 20% Tris/glycine gel for electrophoresis. 5 Parameters for SDS-PAGE: SDS sample buffer: 0.3 g SDS 4 ml 1 M Tris/HCI pH 6.8 8 ml glycerine 10 1 ml 1% bromphenol blue in ethanol Fill with H 2 0 ad 50 ml 4-20% Tris/Glycine Gel: (Invitrogen, Cat. no.: EC6025BOX) 1 5 Electrophoresis buffer: 7.5 g Tris 36 g Glycine 2.5 g SDS Fill with H 2 0 ad 2.5 I 20 The gel is run at a constant current of 20 mA. Staining of the gels: Coomassie Blue R250 Results are shown in Figure 2. 25 Semiquantitative analysis of different anti-A3 antibodies and their discrimination of AP(1-42) fibrils. Positions of antibodies, Ap(1-42) fibrils and Ap(1-42) monomers are marked at the edge of the gel. Due to their size, Ap(1-42) fibrils cannot enter the SDS-PAGE gel and can be seen in the gel slot. 30 1. Marker 2. Ap(1-42) fibril preparation; control 3. AP(1-42) fibril preparation; + mAb SF7; 20h 37*C; supernatant 4. Ap(1-42) fibril preparation; + mAb 5F7; 20h 37*C; pellet 5. APf(1-42) fibril preparation; + mAb 6E10; 20h 37*C; supernatant 35 6. Ap(1-42) fibril preparation; + mAb 6E10; 20h 37"C; pellet The relative binding to fibril type As was evaluated from SDS-PAGE analysis by measuring the Optical Density (OD) values from the Heavy Chain of the antibodies in the fibril bound (pellet fraction) and the supernatant fractions after centrifugation. Antibodies that have bound to the 40 As fibrils should be co-pelleted with the Ap-fibrils and therefore are found in the pellet fraction 71 whereas non-Ap-fibril bound (free) antibodies are found in the supernatant. The percentage of antibody bound to A3-fibrils was calculated according to the following formula: Percent antibody bound to Ap-fibrils = 5 ODrIbriI fraction x1 00%/(ODfirit fraction + OD supernatant fraction). This procedure was performed for the mAbs 6E1 0 (Signet, Cat. no.: 9320), 5F7, 2F2, 6A2, 4D10, 10F1, 3B10, 7C6, 7E5 and 10C1. 10 In the Alzheimer disease brain the As fibrils are a major component of the total A3 peptide pool. By attacking these fibrils by anti Ap-antibodies the risk of negative side effects is elevated due to a liberation of high amounts of As which subsequently may increase the risk of micro haemorrhages. An increased risk for microhaemorrhages was observed in an active immuniza tion approach with fibrillar aggregates of the As peptide (Bennett and Holtzman, 2005, Neurol 15 ogy, 64, 10-12; Orgogozo J, Neurology, 2003, 61,46-54; Schenk et al., 2004, Curr Opin Im munol, 16, 599-606). In contrast to the commercially available antibody 6E10 (Signet 9320) which recognizes a lin 20 ear Ap-epitope between AA1-17, the AO(20-42) globulomer selective antibody 5F7 (which ac tually has the lowest selectivity for Ap(20-42) globulomers over other Ap-forms) does not bind to Ap(1-42) fibrils in an co-pelleting experiment. This is shown by the fact that the 5F7 antibody after an incubation with AP(1-42) fibrils remains after a pelleting step in the supernatant and is not co-pelleted because of being bound to the Ap(1-42) fibrils. 25 Example 4: Analysis of cognitive performance in mice by means of an object recognition test after active immunization with AP~(1-42) monomer (0.1% NH 4 0H), AP(1-42) globulomer or Ap(20-42) globulomer in comparison to wild type. 30 In these experiments mice overexpressing human APP with a point mutation were used. The point mutation refers to amino acid 717 (substitution of isoleucine for valine) and has been found in a London family where it leads to onset of AD before the beginning of the sixth dec ade of life (Mullan et al., Nature Genetics 2 (1992) 340-342). The transgenic mice, herein re ferred to as APP/L, were created by and first described in Leuven (Moechars et al., J. Biol. 35 Chem. 274 (1999) 6483-6492). Female APP/L mice were subjected to active immunization at 6 weeks of age. The mice received either 100 pg of Apf(1-42) monomer (0.1% NH 4 0H), Ap(1 -42) globulomer or AP(20-42) globulomer in phosphate-buffered saline (PBS) mixed with an equal amount of 40 complete Freund's adjuvant intraperitoneally, followed by booster injections with the same 72 amount of antigene in incomplete Freund's adjuvant every third week for three months. Throughout the time course of the experiment the animals were kept under standard condi tions in a reverted day/night cycle (14 hours of light beginning at 7 pm/10 hours of darkness). Body weight gain over the time of the experiment was as expected and did not differ from a 5 control group which received PBS / adjuvant alone, suggesting that the antigen treatments were well tolerated. At 4.5 month of age cognitive ability of the mice was tested by an object recognition test as described in the art (Dewachter et al. Journal of Neuroscience 22 (2002) 3445-3453). To this 10 end, mice were accustomed to an arena and then exposed for 10 minutes to an acquisition phase during which they were individually placed in the arena which now contained two identi cal elements (blue pyramid, green cube, yellow cylinder of similar size, ca. 4 cm). The duration and frequency with which the mouse explored the objects were recorded. During retention phase, 2.5 h later, mice were returned to the arena which now contained, in addition to the 15 known object, an unknown object randomly selected from the other objects. Recognition of the new object was recorded as the time during which the mouse was exploring the old object rela tive to total time (exploration of old and new object). The ,recognition index" expresses this relation (time for new object/total time). A mouse which does not remember the known object will consider it as equally interesting as the new object and spend an equal amount of time on 20 exploring it, in other words, will show a recognition index of 50%. A mouse which remembers the known object will consider it as not interesting and therefore show a significantly higher recognition index. APP/L mice are known to be cognitively deficient at 4.5 months of age and exhibit a recognition index in the dimension of the random level, i.e. 50%. 25 Results are shown in figure 3. Object recognition test in mice. The test reports recognition of a known object in comparison to an unknown one, measured in terms of explorative behaviour during a 10 minute test phase. The recognition index is defined as the percentage of time which the mouse spends on explor 30 ing the unknown object relative to the time spent on exploring both objects. The known object was explored by the mouse during a 10 minute acquisition phase three hours before the test phase. Five groups of mice (number n given below the columns) were compared. Normal C57B1/6 mice (wild type) show a high RI significantly different from random level (50%, i.e. equal times of exploration spent on both the known and the unknown object) (*** = p < 0.001; 35 Student's t-test). The other four groups of APP transgenic mice were subjected to active im munisation three months before. The immunogens used were Ap(1-42) monomer, AP1(1-42) globulomer and AP(20-42) globulomer. Phosphate-buffered saline (PBS) was used as control. Significant differences between PBS and the other groups are indicated with circles: * p < 0.05; 00 = p < 0.01 (post-hoc t-test after p < 0.05 in ANOVA). 40 73 APP/L mice are known to show, in contrast to non-transgenic mice, a cognitive deficiency at 4.5 months of age, scoring results close to random level (i.e. 50% recognition index). In fact, the PBS-treated mice showed random behaviour, in contrast to non-transgenic mice (wild type). Immunization with native Ap(1-42) globulomer as well as with Ap(20-42) globulomer 5 resulted in significantly improved object recognition in APP/L mice. As both globulomer preparations (native and truncated) resulted in memory improvement in APP transgenic animals and even superior recognition in animals treated with AP(20-42) globulorner it is reasonable to assume that induction of antibodies against truncated AP(20-42) 10 globulomer will produce the best result, and that passive immunisation with antibodies reacting specifically with this species represents the optimal strategy of treatment. Example 5: Dot-Blot analysis of the antibody profile for different Ap-forms after an active im munization of APP/L Tg mice with Ap(20-42) globulomer. 15 After immunization of mice (compare example 4) of APP/L mice (Moechars et al., 1999, J. Biol. Chem. 274, 6483-6492) with different forms of AP, plasma samples were assessed for anti-Ap antibodies. To this end, dilution series of the individual AP(1-42) forms ranging from 100 pmol/pl to 0.01 pmol/pl in PBS supplemented with 0.2 mg/ml BSA were made. 1 pl of each 20 sample was blotted onto a nitrocellulose membrane. For detection the corresponding mouse plasma samples were used (diluted 1:400). Immunostaining was done using alkaline phos phatase conjugated anti-mouse-IgG and the staining reagent NBT/BCIP. 25 Ap-standards for dot-blot: 1. Ap(1-42) globulomer The preparation of the AP(1-42) globulomer is described in example la. 30 2. HFIP pretreated Ap(1-42) monomer in Pluronic F68 3 mg of Ap1(1-42), (Bachem Inc.; cat. no. H-1 368 ) were dissolved in 0.5 ml HFIP (6 mg/ml sus pension) in an 1.7 ml Eppendorff tube and was shaken (Eppendorff Thermo mixer, 1400 rpm) for 1.5 h at 37*C till a clear solution was obtained. The sample was dried in a SpeedVac con centrator (1.5 h) and resuspended in 13.2 pi DMSO, shaken for 10 sec., followed by sonifica 35 tion (20 sec), and shaking (e.g. in Eppendorff Thermo mixer, 1400 rpm) for 10 min. 6 ml of 20 mM NaH 2

PO

4 ; 140 mM NaCI; 0.1% Pluronic F68; pH 7.4 were added and stirred for 1 h at room temperature. The sample was centrifuged for 20 min at 3000g. The supernatant was discarded and the precipitate solved in 0.6 ml 20 mM NaH 2

PO

4 ; 140 mM NaCl; 1% Pluronic F68, pH 7.4. 3.4 ml of H 2 0 were added and stirred for 1 h at room temperature followed by 20 74 min centrifugation at 3000g. Eight aliquots of each 0.5ml of the supernatant were stored at 20* for further use. 3. Ap(20-42) globulomer 5 The preparation of the Ap(20-42) globulomer is described in example 1c. 4. A(12-42) globulomer The preparation of the Ap(1 2-42) globulomer is described in example 1 d. 10 5. HFIP pretreated AP(1-40) monomer, 5 mM in DMSO 1 mg AP(1-40), (Bachem Inc, cat. no. H-1194) were suspended in 0.25 ml HFIP (4mg/ml sus pension) in an Eppendorff tube. The tube was shaken (e.g. in Eppendorff Thermo mixer, 1400 rpm) for 1.5 h at 37 *C to get a clear solution and afterwards dried in a speed vac concentrator (1.5 h). The sample was redissolved in 46 pl DMSO (21.7 mg/mI solution = 5 mM), shaken for 15 10 sec and subsequently sonicated for 20sec. After 10 min shaking (e.g. in Eppendorff Thermo mixer, 1400 rpm) the sample is stored at -20*C for further use . 6. Ap(1-42) monomer, 0.1% NH 4 0H 1 mg Ap(1-42) (Bachem Inc., cat. no. H-1368) were dissolved in 0.5 ml 0.1% NH 4 0H in H 2 0 20 (freshly prepared) ( = 2 mg/ml) and immediately shaken for 30 sec. at room temperature to get a clear solution. The sample was stored at -20 *C for further use. 7. Ap(1 -42) fibrils 1 mg AS(1-42) (Bachem Inc. Catalog Nr.: H-1368) were dissolved in 500 pl aqueous 0.1% 25 NH 4 0H (Eppendorff tube) and the sample was stirred for 1 min at room temperature. 100 pl of this freshly prepared AO(1-42) solution were neutralized with 300 pi 20 mM NaH 2

PO

4 ; 140 mM NaCl, pH 7.4. The pH was adjusted to pH 7.4 with 1% HCI. The sample was incubated for 24 h at 370C and centrifuged (10 min at 10000g). The supernatant was discarded and the fibril pel let resuspended with 400 pl of 20 mM NaH 2

PO

4 ; 140 mM NaCl, pH 7.4 by vortexing for 1 min. 30 8. sAPPa Supplied from Sigma (cat.no. S9564; 25 pg in 20 mM NaH 2

PO

4 ; 140 mM NaCl; pH 7.4). The sAPPa was diluted with 20 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4, 0.2mg/mi BSA to 0.1 mg/ml (= lpmol/pl). 35 Materials dot blot: AP-standards: Serial dilution of AP-antigens in 20 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4+ 40 0.2 mg/ml BSA 75 1) 100 pmol/pl 2) 10 pmol/pl 3) 1 pmol/pl 5 4) 0,1 pmol/pl 5) 0,01 pmol/pl Nitrocellulose: Trans-Blot Transfer medium, Pure Nitrocellulose Membrane (0.45 pm); 10 BIO-RAD Anti-Mouse-AP: AQ330A (Chemicon) 15 Detection reagent: NBT/BCIP Tablets ( Roche) Bovine Serum Albumin, (BSA): A-7888 (SIGMA) 20 Blocking reagent: 5 % low fat milk in TBS Buffer solutions: 25 TBS 25 mM Tris / HCI - buffer pH 7.5 + 150 mM NaCl TTBS 30 25 mM Tris / HCI - buffer pH 7.5 + 150 mM NaCl + 0.05 % Tween 20 PBS + 0.2 mg/ml BSA 35 20 mM NaH 2

PO

4 buffer pH 7.4 + 140 mM NaCl + 0.2 mg/ml BSA Antibody solution 1: 76 Mouse plasma samples from an active immunization study with Ap(20 42) globulomer (1:400 diluted in 20 ml 1 % low fat milk in TBS) Antibody solution 11: 5 1:5000 dilution Anti-Mouse-AP in 1 % low fat milk in TBS Dot-blot-procedure: 10 1) 1 pl each of the different Ap-standards (in their 5 serial dilutions) were dotted onto the nitrocellulose membrane in a distance of approximately 1 cm from each other. 2) The AP-standards dots were allowed to dry on the nitrocellulose membrane on air for at least 10 min at room temperature (RT) (= dot blot) 15 3) Blocking: The dot blot was incubated with 30 ml 5 % low fat milk in TBS for 1.5 h at RT. 4) Washing: 20 The blocking solution was discarded and the dot blot was incubated under shaking with 20 ml TTBS for 10 min at RT. 5) Antibody solution 1: The washing buffer was discarded and the dot blot was incubated with antibody solu 25 tion I overnight at RT. 6) Washing: The antibody solution I was discarded and the dot blot was incubated under shaking with 20 ml TTBS for 10 min at RT. The washing solution was discarded and the dot blot 30 was incubated under shaking with 20 ml TTBS for 10 min at RT. The washing solution was discarded and the dot blot was incubated under shaking with 20 ml TBS for 10 min at RT. 7) Antibody solution 11: 35 The washing buffer was discarded and the dot blot was incubated with antibody solu tion 11 for 1 h at RT 8) Washing: 77 The antibody solution 11 was discarded and the dot blot was incubated under shaking with 20 ml TTBS for 10 min at RT. The washing solution was discarded and the dot blot was incubated under shaking with 20 ml TTBS for 10 min at RT. The washing solution was discarded and the dot blot was incubated under shaking with 20 ml TBS for 10 min 5 at RT. 9) Development: The washing solution was discarded. 1 tablet NBT/BCIP was dissolved in 20 ml H 2 0 and the dot blot was incubated fo-5 mih vith this solution. The development was 10 stopped by intensive washing with H 2 0. Results are shown in figure 4. Dot blot analysis of anti-A3 antibodies produced after active immunization of mice with Ap(20 15 42) globulomer to assess their specificity towards different forms of As. The individual forms of As were blotted in serial dilutions and incubated with the corresponding mouse plasma con taining anti Ap-antibodies produced during immune reaction. The individual dot blots corre spond to different individuals of immunized mice. 20 1. Apr(1-42) globulomer 2. AP(1-42) monomer, HFIP pretreated, in 0.1% Pluronic F68 3. Ap(20-42) globulomer 4. Ap~(12-42) globulomer 5. Ap(1-40) monomer, HFIP pretreated, 5 mM in DMSO 25 6. Ap(1-42) monomer, 0.1% NH 4 0H 7. Ap(1-42) fibril preparation 8. sAPPa (Sigma); (first dot: 1 pmol) In the active immunization study of APP/L Tg-mice it was shown that immunization with Ap(20 30 42) globulomer leads to the best result in alleviating a cognitive impairment in these mice com pared to PBS treatment. Plasma samples from APP/IL Tg mice after being actively immunized with Ap(20-42) globulomer exhibit an antibody profile (predominant recognition of Ap(20-42) globulomer and Ap1(12-42) globulomer) which resembles that of the Ap(20-42) globulomer mAbs claimed herein. 35 Example 6: Concentration of soluble and insoluble Ap(1-42) and AP1(1-40) peptide in brain ex tracts of actively immunized APP/PS1 Tg mice with either APf(1-42) monomer, AP(1-42) globu lomer, Ap(20-42) globulomer or vehicle as control.

78 40 female mice of a double transgenic mouse model of Alzheimer's Disease (APP/PS1 Tg mice) in FVBxC57BI/6J background with an age of 4 months were used for this study. The APP/PS1 Tg mice contain the 695 amino acid form of human APP with the V7171 mutation (position refering to the longest APP-isoform) and in addition the human Presenilin 1 gene with 5 the A264E mutation. Both genes are under control of the Thyl promotor. The mice were gen erated and characterized in one of the founding laboratories of reMYND, the Experimental Ge netics Group, Campus Gasthuisberg, Catholic University Leuven, by Prof. Fred Van Leuven et al. 10 All mice were genotyped by polymerase chain reaction (PCR) at the age of 3 weeks and re ceived a unique identity number, once the PCR results were known. Mice had free access to pre-filtered and sterile water (UV-lamp) and standard mouse chow. The food was stored under dry and cool conditions in a well-ventilated storage room. The 15 amount of water and food was checked daily, supplied when necessary and by default re freshed twice a week. Mice were housed under a reversed day-night rhythm: 14 hours light/10 hours darkness start ing at 7 p.m., in standard metal cages type RVS T2 (area of 540 cm 2 ). The cages are equipped 20 with solid floors and layer of bedding litter. The number of mice per cage was limited in accor dance with legislation on animal welfare. Five days before the onset of the behaviour test, mice were re-caged in macrolon Type 2 cages and transported to the laboratory in order to adapt to the laboratory environment in preparation to the behaviour test. 25 The mice received either 100 pg of AP(1-42) monomer (0.1% NH 4 0H), AP1(1-42) globulomer or AP(20-42) globulomer in phosphate-buffered saline (PBS) mixed with an equal amount of complete Freund's adjuvant intraperitoneally, followed by booster injections with the same amount of antigene in incomplete Freund's adjuvant every third week for four months. 30 Biochemistry The AP~(1-40) and AP(1-42) in the soluble fraction of 1 hemisphere was determined by ELISA. In addition the AP~(1-40) and AP(l-42) of the insoluble membrane fraction of 1 hemisphere was determined by ELISA. 35 The mice were anaesthetized with a 2:1:1 mixture of Ketalar (ketamin), Rompun (xylazin 2%) and atropin and flushed trans-cardially with physiological serum at 4'C. This was performed to remove blood from the brain vessels, a procedure which has no influence on organ integrity.

79 Cerebrospinal fluid (CSF) was collected via an incision in the neck muscles, between the skull and the first cervical vertebrae. A puncture into the cisterna magna was given with a 26 gauge needle and 10-20 pl of CSF was collected with a fine glass pipette. 5 Blood was collected via a heart puncture and a 1 ml syringe into heparin-coated Eppendorf tubes. The blood was centrifuged at 14.000 rpm at 4 0 C for 5 minutes. The serum was stored at -70 0 C. The mice were flushed transcardially with physiological serui-n at 4 0 C. 10 The brain was removed from the cranium and hindbrain and forebrain were separated by a cut in the coronal/frontal plane. The cerebellum was discharged. The forebrain was divided evenly into left and right hemisphere by using a midline sagittal cut. 15 One hemisphere was immediately immersed in liquid nitrogen and stored at -70 0 C until bio chemical analysis. Homogenization and fractionation of one brain hemisphere 20 Brains were homogenized using a Potter, a glass tube (detergent free, 2 cm 3 ) and a mechani cal homogenizer (650 rpm). A volume of 6,5 x % brain weight of freshly prepared 20 mM Tris/HCI buffer (pH 8.5) with Proteinase Inhibitors (1 tablet per 50 ml Tris/HCI buffer, Com pleteTM, Roche, Mannheim, Germany) was used as homogenization buffer. 25 Samples were transferred from -70*C into a sample holder with liquid nitrogen and each indi vidual sample was pre-warmed by incubation on the bench for a few seconds prior to homog enization. The homogenates were collected in Beckman centrifuge tubes TLX and collected on ice prior to centrifugation. Between two samples, the Potter and the glass tube were rinsed carefully with distilled water (AD) without detergents and dried with absorption paper. 30 Samples were centrifuged in a pre-cooled ultracentrifuge (Beckman, Mannheim, Germany) for 1 hour and 20 minutes at 48000 rpm (135.000 x g) at 4 0 C. Due to a limited number of centri fuge holders (N=8), samples were sorted by brain weight (to equilibrate the centrifuge) and randomized in order to divide the different treatment groups over the different centrifugation 35 sessions. The supernatant (soluble fraction containing secreted APP and amyloid peptides) was sepa rated from the pellet (membrane fraction containing membrane-bound APP-fragments and plaque-associated amyloid peptides in case of aged mice). The supernatant was divided over 80 two tubes of which one was stored at -20*C as back-up and the other was processed further for column chromatography to concentrate the amyloid peptides. Brain weights, volume Tris/HCI buffer used, centrifugation sessions (marked by colour) and 5 volume soluble fraction used for column chromatography are given exemplary in the following table. Weight V Tris 50% V Tris Sample Treatment Mouse ID N* brain (W) (=W x 6,5) N * (m g) (p1) (pI1) 19 N TAB.TPF 1305 157,8 1026 513 21 1 ?9N. T A B. T PF 1335 160,2 1041 521 Small reversed phase columns (C18-Sep-Pack Vac 3cc cartridges, Waters, Massachusetts, MA) were mounted on a vacuum system and washed with 80% acetonitrile in 0.1% 10 trifluoroacetic acid (A-TFA) followed with 0.1% TFA twice. Then the samples were applied and the columns were washed successively with 5% and 25% A-TFA. Amyloid peptides were eluted with 75% A-TFA and the eluates were collected in 2 ml tubes on ice. Eluates were freeze-dried in a SpeedVac concentrator (Savant, Farmingdale, NY) overnight and resolved in 330 pl of the sample diluent furnished with the ELISA kits. 15 The pellets were further fractionated into different membrane fractions: membrane fraction A (MFA), membrane fraction B (MFB) containing full length APP and membrane fraction C (MFC) containing plaque associated amyloid. Therefore the pellets were dissolved in TBS buffer with proteinase inhibitors (1 tablet per 50 ml TBS buffer, CompleteTM, Roche, Mannheim, 20 Germany) and the MFA was divided over two tubes of which one was stored at -20oC as back up. 60% of MFA was further processed with addition of NP40 (2% of final volume) and Triton X-100 (2% of final volume) in TBS with proteinase inhibitors and centrifuged for one hour at 27.000 rpm (98'000 x g) in a Beckman ultracentrifuge at 4*C using a swing-out rotor (SW60). The supernatant (MFB) was separated from the pellet (MFC) and both were stored at -20*C. 25 Brain weights, 60% of the brain weight, the volumes TBS + PI + NP40 + Triton X-100 buffer used and the centrifugation sessions (marked by colour) are given exemplary in the following table.

81 3/5 x Volume Weight Weight buffer = Sample Treatment Mouse ID N* brain (W) brain 3/5 Wx15 N' (m (m g) (p) 19 TAB.TPF 1305 157,8 95 1420 21 U TAB.TPF 1335 160,2 96 1442 ELISA of human As in the soluble fraction of one hemisphere 5 To quantify the amount of human Apr(1-40) and human AP(1-42) in the soluble fraction of the brain homogenates and/or in cerebrospinal fluid (CSF), commercially available Enzyme Linked-1mmunosorbent-Assay (ELISA) kits were used (h Amyloid p40 or p42 ELISA high sen sitive, The Genetics Company, Zurich, Switzerland). The ELISA was performed according to the manufacturer's protocol. Briefly, standards (a dilution of synthetic Ap3(1-40) or Ap~(1-42)) 10 and samples were prepared in a 96-well polypropylene plate without protein binding capacity (Greiner bio-one, Frickenhausen, Germany). The standard dilutions with final concentrations of 1000, 500, 250, 125, 62.5, 31.3 and 15.6 pg/ml and the samples were prepared in the sample diluent, furnished with the ELISA kit, to a final volume of 60 pl. Since amyloid levels increase with the age of the mouse and since the actual evaluation requires that the readings of the 15 samples are within the linear part of the standard curve, the samples for Ap~(1-40) analysis were diluted 1:3, the samples for Ap(1-42) analysis were diluted 1:6. Samples, standards and blanks (50 pl) were added to the anti-Ap-coated polystyrol plate (cap ture antibody selectively recognizes the C-terminal end of the antigen) in addition with a selec 20 tive anti-Ap-antibody conjugate (biotinylated detection antibody) and incubated overnight at 4*C in order to allow formation of the antibody-Amyloid-antibody-complex. The following day, a Streptavidine-Peroxidase-Conjugate was added, followed 30 minutes later by the addition of a TMB/peroxide mixture, resulting in the conversion of the substrate into a coloured product. This reaction was stopped by the addition of sulfuric acid (1 M) and the colour intensity was 25 measured by means of photometry with an ELISA-reader with a 450 nm filter. Quantification of the As content of the samples was obtained by comparing absorbance to the standard curve made with synthetic AP(1-40) or Ap(1-42). ELISA human As in the insoluble fraction of one hemisphere 30 To quantify the amount of human Ap(1-40) and human Ap(1-42) in the insoluble membrane fraction of the brain homogenates, the MFC samples were further processed and dissolved in 8M Guanidine in 80 mM Tris/HCI. Subsequently samples were incubated for 3 hours in a ther momixer at 25 0 C and pipetted up and down with a 100 pl pipette every hour to dissolve the 35 MFC pellet into the guanidine buffer. Finally samples were centrifuged for only 1 minute at 4000 rpm to remove debris.

82 Brain weight, the weight of the MFC pellet and the volume of 8M guanidine buffer are given examplary in the following table. Weight pellet of volume Sample Treatment Mouse ID N' Weight brain MFC (WMFC) 8M guanidine N'- (W) (m g) (40% brain) (WMFCx1,6) (p1) 19 KTAB.TPF 1305 157.8 61 21 f TAB.TPF 1335 160,2 64 103 5 To quantify the amount of human AP(1-40) and human A1(1-42) in the final samples, commer cially available Enzyme-Linked-immunosorbent-Assay (ELISA) kits were used (h Amyloid s40 or 042 ELISA high sensitive, The Genetics Company, Zurich, Switzerland). The ELISA was performed according to the manufacturer's protocol, except for the preparation of the stan 10 dards (a dilution of synthetic AP(1-40) or APf(1-42)). The samples were prepared in the sample diluent, furnished with the ELISA kit, to a final volume of 60 pl. Since guanidine influences the OD-values of the standard curve, the standard dilutions with final concentrations of 1000, 500, 250, 125, 62.5, 31.3 and 15.6 pg/ml were prepared in sample diluent with the same concentra tion guanidine as for the samples. This was performed in a 96-well polypropylene plate without 15 protein binding capacity (Greiner bio-one, Frickenhausen, Germany). Since amyloid levels increase with the age of the mouse and since the actual evaluation re quires that the readings of the samples are within the linear part of the standard curve, the samples for insoluble AP(1-40) and insoluble AP(1-42) analysis were diluted 1:500. 20 Samples, standards and blanks (50 pl) were added to the anti-Ap-coated polystyrol plate (cap ture antibody selectively recognizes the C-terminal end of the antigen) in addition with a selec tive anti-Ap-antibody conjugate (biotinylated detection antibody) and incubated overnight at 4 0 C in order to allow formation of the antibody-Amyloid-antibody-complex. The following day, a 25 streptavidin-peroxidase conjugate was added, followed 30 minutes later by the addition of a TMB/peroxide mixture, resulting in the conversion of the substrate into a coloured product. This reaction was stopped by the addition of sulfuric acid (1M) and the colour intensity was measured by means of photometry with an ELISA-reader with a 450 nm filter. Quantification of the AP content of the samples was obtained by comparing absorbance to the standard curve 30 made with synthetic A1(1-40) or Ap(1-42). Results are shown in figure 5 Concentration of soluble and insoluble APf(1-42) and Ap(1-40) peptide in brain extracts of ac 35 tively immunized APP/PS1 Tg-mice with either Ap(1-42) monomer (0.1% NH 4 0H), Ap(1-42) globulomer, Ap(20-42) globulomer or vehicle as control.

83 In the soluble and insoluble fraction of a brain extract of APP/PS1 Tg-mice actively immunized with Ap(20-42) globulomer the level of Ap(1-40)- and Ap(1-42)-peptide is not significantly dif ferent to the vehicle control. In contrast, immunization with Ap(1-42) globulomer and Ap(1-42) 5 monomer leads to a reduction in brain Ap(1-40)- and Ap(1-42)-levels. This shows that an Ap(20-42) globulomer directed immunization approach does not alter the total Ap-brain levels significantly but nonetheless is effective in alleviating the Ap-peptide related cognitive impair ments (see example 4). 10 Example 7: Analysis of cognitive performance by object recognition test in APP/L transgenic mice after passive immunization with anti-AP(20-42) globulomer antibodies In these experiments mice overexpressing human APP with a point mutation were used. The point mutation refers to amino acid 717 (substitution of isoleucine for valine) and has been 15 found in a London family where it leads to onset of AD before the beginning of the sixth dec ade of life (Mullan et al., Nature Genetics 2 (1992) 340-342). The transgenic mice, herein re ferred to as APP/L, were created by and first described in Leuven (Moechars et al., J. Biol. Chem. 274 (1999) 6483-6492). Female APP/L mice were subjected to passive immunization at 3 months of age. Mice received 250 pg of any of the monoclonal mouse antibodies 5F7, 1OF1 1 20 or 7C6 in 100 pl of phosphate-buffered saline (PBS). Throughout the time course of the ex periment the animals were kept under standard conditions in a reverted day/night cycle (14 hours of light beginning at 7 pm/1 0 hours of darkness). They tolerated passive immunization well, without any signs of adverse effects. 25 After the third injection (day 15 of experiment) cognitive ability of the mice was tested by an object recognition test as described in the art (Dewachter et al. Journal of Neuroscience 22 (2002) 3445-3453). To this end, mice were accustomed to an arena and then exposed for 10 minutes to an acquisition phase during which they were individually placed in the arena which now contained two identical elements (green cube or orange cylinder of similar size, ca. 4 cm). 30 The duration and frequency with which the mouse explored the objects were recorded. During retention phase, 2.5 h later, mice were returned to the arena which now contained, in addition to the known object, the other object. Recognition of the new object was recorded as the time during which the mouse was exploring the old object relative to total time (exploration of old and new object). The ,,recognition index" expresses this relation (time for new object/total 35 time). A mouse which does not remember the known object will consider it as equally interest ing as the new object and spend an equal amount of time on exploring it, in other words, will show a recognition index of 50%. A mouse which remembers the known object will consider it as not interesting and therefore show a significantly higher recognition index. APP/L mice are known to be cognitively deficient at 4.5 months of age and exhibit a recognition index in the 40 dimension of the random level, i.e. 50%.

84 Results are shown in figure 6. Object recognition test in mice. The test reports recognition of a known object in comparison to 5 an unknown one, measured in terms of explorative behaviour during a 10 minute test phase. The recognition index is defined as the percentage of time which the mouse spends on explor ing the unknown object relative to the time spent on exploring both objects. The known object was explored by the mouse during a 10 minute acquisition phase 2.5 hours before the test phase. 10 a) APP transgenic mice were immunized once a week for three weeks by intraperitoneal injec tion of 250 pg of the antibody 5F7 (n=9), the antibody 1OF11 (n=11) or the antibody 7C6 (n=11); control animals received PBS (n=6). Significant differences from random level (50%, i.e. equal time of exploration spent on the known and the unknown object) are indicated with 15 asterisks. * = p < 0.05 (t-test) b) Comparison of all mice treated with antibodies (5F7, 1OF1 1 and 7C6; (n=31)) and mice treated with phosphate-buffered saline (PBS; n=6). The RI of the antibody-treated group differ ent significantly from random level (** = P<0.01; t-Test). 20 APP/L mice are known to be cognitively deficient at 4.5 months of age and exhibit a recogni tion index in the dimension of the random level, i.e. 50%. Indeed the PBS-treated mice showed random behaviour. Passive immunization with all three 25 antibodies (5F7, 1OF 11 and 7C6) resulted in a markedly increased recognition index. When compared as a pooled group against the controls, the recognition index is significantly in creased. This beneficial effect on memory performance of APP/L mice after administration of all three antibodies suggests that an antibody against truncated A@(20-42) globulomer is suffi cient to achieve cognitive improvement. 30 Example 8: Dot-Blot profile of the selectivity of the anti-AP(20-42) globulomer antibodies. In order to characterize the selectivity of the monoclonal anti As(20-42) globulomer antibodies they were probed for recognition with different Ap-forms. To this end, serial dilutions of the 35 individual A@(1-42) forms ranging from 100 pmol/pl to 0.01 pmol/pl in PBS supplemented with 0.2 mg/mI BSA were made. 1 pl of each sample was blotted onto a nitrocellulose membrane. For detection the corresponding antibody was used (0.2 pg/ml). Immunostaining was done using peroxidase conjugated anti-mouse-IgG and the staining reagent BM Blue POD Substrate (Roche). 40 85 AP-standards for dot-blot: 1. Ap(1-42) monomer, 0.1% NH 4 0H 1 mg Ap(1-42) (Bachem Inc., cat. no. H-1368) were dissolved in 0.5 ml 0.1% NH 4 0H in H 2 0 5 (freshly prepared) (= 2 mg/ml) and immediately shaken for 30 sec at room temperature to get a clear solution. The sample was stored at -20 *C for further use. 2. Ap1(1-40) monomer, 0.1% NH 4 0H 1 mg As(1-40) (Bachem Inc., cat. no. H-1368) were dissolved in 0.5 ml 0.1% NH 4 0H in H 2 0 10 (freshly prepared) (= 2 mg/ml) and immediately shaken for 30 sec. at room temperature to get a clear solution. The sample was stored at -20*C for further use. 3. Ap(1-42) monomer, 0.1% NaOH 2.5 mg Ap(1-42) (Bachem Inc., cat. no. H-1368) were dissolved in 0.5 ml 0.1% NaOH in H 2 0 15 (freshly prepared) ( = 5 mg/ml) and immediately shaken for 30 sec. at room temperature to obtain a clear solution. The sample was stored at -20*C for further use. 4. Ap(1-40) monomer, 0.1% NaOH 2.5 mg Ap(1-40) (Bachem Inc., cat. no. H-1368) were dissolved in 0.5 ml 0.1% NaOH in H 2 0 20 (freshly prepared) (= 5 mg/ml) and immediately shaken for 30 sec. at room temperature to ob tain a clear solution. The sample was stored at -20*C for further use. 5. AP(1-42) globulomer The preparation of the Ar3(1-42) globulomer is described in example 1a. 25 6. Ap(12-42) globulomer The preparation of the Ap(12-42) globulomer is described in example 1d. 7. AO(20-42) globulomer 30 The preparation of the A(20-42) globulomer is described in example 1c. 8. AO(1-42) fibrils 1 mg AP(1-42) (Bachem Inc. cat. no.: H-1368) were solved in 500 pl aqueous 0.1% NH 4 0H (Eppendorff tube) and the sample was stirred for 1 min at room temperature. 100 pl of this 35 freshly prepared AO(1-42) solution were neutralized with 300 pl 20 mM NaH 2

PO

4 ; 140 mM NaCl, pH 7.4. The pH was adjusted to pH 7.4 with 1% HCI. The sample was incubated for 24 h at 37*C and centrifuged (10 min at 10000g). The supernatant was discarded and the fibril pel let resuspended with 400 pl 20 mM NaH 2

PO

4 ; 140 mM NaCl, pH 7.4 by vortexing for 1min. 40 9. sAPPa 86 Supplied by Sigma (cat.no. S9564; 25 pg in 20 mM NaH 2

PO

4 ; 140 mM NaCl; pH 7.4). The sAPPa was diluted to 0.1 mg/mI ( = 1pmol/pl) with 20 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4, 0.2 mg/ml BSA. 5 Materials for dot blot: Ap-standards: Serial dilution of As antigens in 20 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4 + 0.2 mg/mI BSA 10 1) 100 pmol/pl 2) 10 pmol/pl 3) 1 pmol/pl 4) 0,1 pmol/pl 15 5) 0,01 pmol/pl Nitrocellulose: Trans-Blot Transfer medium, Pure Nitrocellulose Membrane (0.45 pm); BIO-RAD 20 Anti-Mouse-POD: Cat no: 715-035-150 (Jackson Immuno Research) Detection reagent: 25 BM Blue POD Substrate, precipitating (Roche) Bovine Serum Albumin, (BSA): Cat no: A-7888 (SIGMA) 30 Blocking reagent: 5 % low fat milk in TBS Buffer solutions: TBS 35 25 mM Tris / HCI buffer pH 7.5 + 150 mM NaCl TTBS 25 mM Tris / HCl - buffer pH 7.5 87 + 150 mM NaCl + 0.05 % Tween 20 PBS + 0.2 mg/ml BSA 5 20 mM NaH 2

PO

4 buffer pH 7.4 + 140 mM NaCl + 0.2 mg/ml BSA 10 Antibody solution 1: 0.2 pg/ml antibody diluted in 20 ml 1 % low fat milk in TBS Antibody solution 1l: 15 1:5000 dilution Anti-Mouse-POD in 1 % low fat milk in TBS Dot blot procedure: 20 1) 1 pl each of the different A-standards (in their 5 serial dilutions) were dotted onto the nitrocellulose membrane in a distance of approximately 1 cm from each other. 2) The AO-standards dots were allowed to dry on the nitrocellulose membrane on air for at 25 least 10 min at room temperature (RT) (= dot blot) 3) Blocking: The dot blot was incubated with 30 ml 5% low fat milk in TBS for 1.5 h at RT. 30 4) Washing: The blocking solution was discarded and the dot blot was incubated under shaking with 20 ml TTBS for 10 min at RT. 5) Antibody solution I: 35 The washing buffer was discarded and the dot blot was incubated with antibody solu tion I for 2 h at RT 6) Washing: 88 The antibody solution I was discarded and the dot blot was incubated under shaking with 20 ml TTBS for 10 min at RT. The washing solution was discarded and the dot blot was incubated under shaking with 20 ml TTBS for 10 min at RT. The washing solution was discarded and the dot blot was incubated under shaking with 20 ml TBS for 10 min 5 at RT. 7) Antibody solution 1l: The washing buffer was discarded and the dot blot was incubated with antibody solu tion Il dvsriiht~aftRT 10 8) Washing: The antibody solution 1I was discarded and the dot blot was incubated under shaking with 20 ml TTBS for 10 min at RT. The washing solution was discarded and the dot blot was incubated under shaking with 20 ml TTBS for 10 min at RT. The washing solution 15 was discarded and the dot blot was incubated under shaking with 20 ml TBS for 10 min at RT. 9) Development: The washing solution was discarded. The dot blot was developed with 10 ml BM Blue 20 POD Substrate for 10 min. The development was stopped by intense washing of the dot blot with H 2 0. Quantitative evaluation was done using a densitometric analysis (GS800 densitometer (BioRad) and software package Quantity one, Version 4.5.0 (BioRad)) of the dot-intensity. Only dots were evaluted that had a relative density of greater than 20% of the relative density of the last optically unambiguously identified dot of the 25 Ap(20-42) globulomer. This threshold value was determined for every dot-blot inde pendently. The calculated value indicates the relation between recognition of AP(20-42) globulomer and the respective As form for the antibody given. Results are shown in figure 7. 30 Dot blot analysis of the specificity of different anti-Ap antibodies (6E10, 5F7, 4B7, 10F11, 6A2, 4D10, 2F2; 3B10, 7C6, 7E5, 10Ci) towards different forms of Ap. The monoclonal antibodies tested were obtained (except for 6E10) by active immunization of mice with Ap(20-42) globu lomer, followed by selection of the fused hybridoma cells. The individual As forms were applied 35 in serial dilusions and incubated with the respective antibodies for immune reaction. 1. Ap(1-42) monomer, 0.1% NH 4 0H 2. AP(1-40) monomer, 0.1% NH 4 0H 3. Ap(1-42) monomer, 0.1%NaOH 89 4. Ap~(1-40) monomer, 0.1% NaOH 5. Ap~(1-42) globulomer 6. Ap(1 2-42) globulomer 7. Ap(20-42) globulomer 5 8. AP(1 -42) fibril preparation 9. sAPPa (Sigma); (first dot: 1pmol) The anti Ap(20-42) globulomer selective mAbs can be divided in 3 classes with respect to the discriminationof Ap(1 -42)~globulomer and-Ap(12-42) globulomer. The first class comprising 10 the antibodies 6A2, 5F7 and 2F2 recognizes preferentially Ap(20-42) globulomer and to some extent Ap(1-42) globulomer (and also AP1(12-42) globulomer). The second class comprising the antibodies 1 OF1 1, 4D1 0 and 38130 recognizes preferentially Ap(20-42) globulomer and also recognizes AP(12-42) globulomer but to a lesser extent and do not significantly recognize Ap(1-42) globulomer. The third class comprising the antibodies 7C6, 4B7, 7E5 and 1OC1 rec 15 ognizes Ap(20-42) globulomer but shows no significant recognition of the others. All three classes do not significantly recognize monomeric Ap(1-42), monomeric AP(1-40), AP(1-42) fibrils or sAPPa. The selectivity profile of the anti-Ap(20-42) globulomer antibodies shows that the significantly 20 elevated recognition index in the passive immunization (in figure 6) must mainly be due to a selective recognition of truncated AP(20-42) globulomer and Ap(12-42) globulomer and to a much lesser extent of Ap(1-42) globulomer and not monomeric AP(1-42), monomeric AP(1-40), Ap(1-42) fibrils or sAPPo. 25 Example 9: In situ analysis of the specific reaction of AP(20-42) selective antibodies to fibrillary As peptide in the form of AP plaques in old TG2576 mice and A0 amyloid in meningeal ves sels. For these experiments brain material of 19 month old TG2576 mice (Hsiao et al., 1996, Sci 30 ence; 274(5284), 99-102) or 9 month old APP/LxPS1 mice (description as above; ReMYND, Leuven, Belgium) or autopsy material of two Alzheimer's disease patients (RZ16 and RZ55; obtained from BrainNet, Munich) was used. The mice overexpress human APP with the so called Swedish mutation (K670N/M671 L; Tg2576) or with the so-called London mutation (V7171) in addition with the human Presenilin 1 gene with the A264E mutation (APP/LxPS1) 35 and formed p amyloid deposits in the brain parenchyma at about 7-11 months of age and p amyloid deposits in larger cerebral vessels at about 18 months of age (Tg2576). The animals were deeply anaesthetized and transcardially perfused with 0.1 M phosphate-buffered saline (PBS) to flush the blood. Then the brain was removed from the cranium and divided longitudi nally. One hemisphere of the brain was shock-frozen, the other fixated by immersion into 4% 40 paraformaldehyde. The immersion-fixated hemisphere was cryoprotected by soaking in 30% 90 sucrose in PBS and mounted on a freezing microtome. The entire forebrain was cut into 40 pm section which were collected in PBS and used for the subsequent staining procedure. The hu man brain material was an about 1 cm 3 deep-frozen block of the neocortex. A small part of the block was immersion-fixated in 4% paraformaldehyde and further treated like the mouse brain 5 material. Individual sections were stained with Congo Red using the following protocol: Material: 10 - Amyloid dye Congo Red kit (Sigma-Aldrich; HT-60), consisting of alcoholic NaCl solution, NaOH solution and Congo Red solution - staining cuvettes - microscope slides SuperfrostPlus and coverslips - Ethanol, Xylol, embedding medium 15 Reagents: - NaOH diluted 1 : 100 with NaCI solution yields alkaline saline - alkaline saline diluted 1 : 100 with Congo Red solution yields alkaline Congo Red solution (prepare no more than 15 min before use, filtrate) 20 - mount sections on slide and allow them to dry - incubate slide in staining cuvette, first for 30 - 40 minutes in alkaline saline, then for 30 - 40 minutes in alkaline Congo Red solution - rinse three times with fresh ethanol and embed over xylol 25 Staining was first photographed using a Zeiss Axioplan microscope and evaluated qualita tively. Red colour indicated amyloid deposits both in the form of plaques and in larger men ingeal vessels. These Results are shown in Fig. 8A. Later on, evaluation of antibody staining focused on these structures. 30 Antibody staining was performed by incubating the sections with a solution containing 0.07-7.0 pg/ml of the respective antibody in accordance with the following protocol: Materials: - TBST washing solution (Tris Buffered Saline with Tween 20; 10x concentrate; DakoCytoma 35 tion; S3306 1:10 in Aqua bidest) - 0.3% H 2 0 2 in methanol - donkey serum (Serotec), 5% in TBST - monoclonal mouse-anti-globulomer antibody diluted in TBST - secondary antibody: biotinylated donkey-anti-mouse antibody (Jackson Immuno; 715-065 40 150; diluted 1:500 in TBST) 91 - StreptABComplex (DakoCytomation; K 0377) - Peroxidase Substrate Kit diaminobenzidine (=DAB; Vector Laboratories; SK-4100) - SuperFrost Plus microscope slides and coverslips - xylol free embedding medium (Medite; X-tra Kitt) 5 Procedure: - transfer floating sections into ice-cold 0.3% H 2 0 2 and incubate for 30 min - wash for 5 min in TBST buffer - incubate-with donkey serum / TBST for-20 minutes 10 - incubate with primary antibody for 24 hours at room temperature - wash in TBST buffer for 5 minutes - incubate with blocking serum from the Vectastain Elite ABC peroxidase kit for 20 minutes - wash in TBST buffer for 5 minutes - incubate with secondary antibody for 60 minutes at ambient temperature 15 - wash in TBST buffer for 5 minutes - incubate with StreptABComplex for 60 minutes at ambient temperature - wash in TBST buffer for 5 minutes - incubate with DAB from the Vectastain Elite ABC peroxidase kit for 20 minutes - mount the section on slides, air-dry them, dehydrate them with alcohol and embed them 20 Besides visual inspection of the staining, plaque staining was additionally quantified by graphi cally excising 10 randomly selected plaques from the histological images using the ImagePro 5.0 image analysis system and determining their average greyscale value. Optical density val ues were calculated from the greyscale values by subtracting the mean background density of 25 the stained material from the density of amyloid plaques (0% - no plaque staining above sur rounding background, 100% - no transmission / maximal staining), and the differences be tween control and antibodies and between 6G1 and the antibodies selective for Ap(20-42), respectively, were tested for statistical significance with ANOVA. 30 Results of the staining in Tg2576 and APP/LxPS1 mice are shown in Figure 8 B-D and H. Binding of different antibodies at a concentration of 0.7 pg/ml in transversal section of the neo cortices of transgenic TG 2576 mice at 19 months of age: C) Parenchymal Ap deposits (amyloid plaques) were stained only with 6G1 and 6E10 but 35 not with the globulomer selective antibodies (i.e. 5F7, 2F2, 6A2, 4D10, 1OF1 1, 3B10, 7C6, 7E5 and 10C1). D) All globulomer selective antibodies (i.e. 5F7, 2F2, 6A2, 4D10, 1OF11, 3B10, 7C6, 7E5 and 10C1) showed significantly less parenchymal plaque staining compared to the com mercially available antibodies 6E10 and 4G8.

92 Binding of different antibodies at a concentration of 0.07-7.0 pg/ml in transversal section of the neocortices of transgenic APP/LxPS1 mice at 11 months of age: E) Parenchymal Apdeposits (amyloid plaques) were significantly more and at lower con centrations stained with 6G1, 6E10 and 4G8 than with the globulomer selective antibodies 5 (i.e. 5F7, 2F2, 6A2, 4D10, 1OF11, 3B10, 7C6, 7E5 and 10C1). All amyloid deposits had been verified by congophilic staining before (Congo Red; see Fig. 8A). Bar = 100 pm. 10 Evaluation of brown DAB deposits showed that the Ap-unselective 6G1 and 6E10 antibodies stained plaques and meningeal vessels, whereas the Ap(20-42) globulomer selective antibod ies 5F7, 2F2, 6A2, 4D10, 1OF11, 3B10, 7C6. 7E5 and 10C1 did not. This finding demonstrates that there is no or markedly less binding of these antibodies to As fibrils or other As species present in the amyloid structures in vivo. This reduced binding is supposed to reduce the dan 15 ger of side effects induced by too fast dissolution of plaques and subsequent increase in solu ble A3 or neuroinflammation due to the interaction of plaque-bound antibodies with microglia. Results of the staining in human Alzheimer's disease brain are shown in Figure 8 B, F-H. 20 Binding of different antibodies at a concentration of 0.7 pg/mi in transversal section of the neo cortex of patient RZ55: B) Parenchymal AP deposits (amyloid plaques) were stained only with 6G1 and 6E10 but not with the globulomer selective antibodies (i.e. 5F7, 2F2, 6A2, 4D10, 1OF1 1, 3B10, 7C6, 7E5 and 10C1). 25 F) All globulomer selective antibodies (i.e. 5F7, 2F2, 6A2, 4D10, 1OF1 1, 3B10, 7C6, 7E5 and 10C1) showed significantly less staining compared to the commercially available anti bodies 6E10 and 4G8. H) Vascular As deposits (arrows) were stained only with 6G1 and 6E10 but not with globu lomer selective antibodies (i.e. 5F7, 2F2, 6A2, 4D10, 10F11, 3B10, 7C6, 7E5 and 10C1). 30 Binding of different antibodies at a concentration of 0.07-7.0 pg/ml in transversal section of the neocortices of transgenic APP/LxPS1 mice at 11 months of age: G) Parenchymal AP deposits (amyloid plaques) were significantly more and at lower con centrations stained with 6G1, 6E10 and 4G8 than with the globulomer selective antibodies (i.e. 5F7, 2F2, 6A2, 4D10, 1OF11, 3B10, 7C6, 7E5 and 10C1). 35 All amyloid deposits had been verified by congophilic staining before (Congo Red; see Fig. 8A). Evaluation of brown DAB deposits showed that the A-unselective 6G1 and 6E10 antibodies stained plaques and meningeal vessels, whereas the AP(20-42) globulomer selective antibod 40 ies 5F7, 2F2, 6A2, 4D10, 1OF1 1, 3B10, 7C6, 7E5 and 10C1 did not. Commercially available 93 antibodies 6E1 0 and 4G8 showed stronger staining compared to globulomer selective antibod ies, but less staining than 6G1. This finding confirms the staining pattern in APP transgenic mice where there is no or markedly less binding of the globulomer selective antibodies to AP fibrils or other AP species present in the amyloid structures in vivo. This reduced binding to 5 human amyloid is supposed to reduce the danger of side effects induced by too fast dissolu tion of plaques and subsequent increase in soluble As or neuroinflammation due to the inter action of plaque-bound antibodies with microglia. Example 10: Anti-Ap-antibody titer and dot-blot selectivity profile in plasma of TG2576 mice 10 approximately one year after active immunization. Approximately one year after the last immunization (with Ap(20-42) globulomer, Ap1(12-42) globulomer, Ap1(1-42) monomer and vehicle) of Tg 2576 mice (from example 9) plasma sam ples were assessed for anti-Ap antibodies produced and still present. To this end, dilution se 15 ries of the different forms of Ap(1-42) in the concentration range from 100 pmol/pl to 0.01 pmol/pl in PBS + 0.2 mg/ml BSA were made. Of each sample, 1 pl was applied to a nitrocellu lose membrane. Detection was performed with suitable mouse plasma samples (diluted 1:400). Staining was done with anti-mouse-IgG conjugated alkaline phosphatase and addition of the staining reagent NBT/BCIP. 20 Ap-standards for dot-blot: 1. Apf(1-42) globulomer The preparation of the Ap~(1-42) globulomer is described in example 1a. 25 2. HFIP pretreated AP(1-42) monomer in Pluronic F68 3mg AP(1-42), (Bachem Inc.; cat. no. H-1368 ) were dissolved in 0.5ml HFIP (6mg/ml suspen sion) in an 1.7 ml Eppendorff tube and was shaken (Eppendorff Thermo mixer, 1400 rpm) for 1.5 h at 37 0 C till a clear solution was obtained. The sample was dried in a SpeedVac concen 30 trator (1.5 h) and resuspended in 13.2 pl DMSO, shaken for 10 sec., followed by sonification (20sec), and shaking (e.g. in Eppendorff Thermo mixer, 1400 rpm) for 10 min. 6ml 20 mM NaH 2

PO

4 ; 140 mM NaCl; 0.1% Pluronic F68; pH 7.4 was added and stirred for 1 h at room temperature. The sample was centrifuged for 20min at 3000g. The supernatant was discarded and the precipitate solved in 0.6ml 20 mM NaH 2

PO

4 ; 140 mM NaCl ; 1% Pluronic F68, pH 7.4. 35 3.4ml H 2 0 was added and stirred for 1 h at room temperature followed by 20 min centrifuga tion at 3000g. Eight aliquots of each 0.5ml of the supernatant were stored at -20* for further use. 3. Ap(20-42) globulomer 94 The preparation of the Ap(20-42) globulomer is described in example 1c. 4. A(12-42) globulomer The preparation of the Ap(1-42) globulomer is described in example 1d. 5 5. AO(1-40) monomer, HFIP pretreated, 5 mM in DMSO 1 mg A1(1-40), (Bachem Inc, cat. no. H-1 194) were suspended in 0.25 ml HFIP (4 mg/ml sus pension) in an Eppendorff tube. The tube was shaken (e.g. in Eppendorff Thermo mixer, 1400 rpm) for 1.5 h at 37 *C to get a clear solution and afterwards dried in a Sp-eedVac-Concentrator 10 (for 1.5 h). The sample was redissolved in 46 pl DMSO (21.7 mg/ml solution = 5 mM), shaken for 10 sec and subsequently sonicated for 20 sec. After shaking (e.g. in Eppendorff Thermo mixer, 1400 rpm) for 10 min the sample is stored at -20*C for further use . 6. AP(1-42) monomer, 0.1% NH 4 0H 15 1 mg Ap(1-42) (Bachem Inc., cat. no. H-1368) were dissolved in 0.5ml 0.1% NH 4 0H in H 2 0 (freshly prepared) ( = 2mg/ml) and immediately shaken for 30 sec. at room temperature to get a clear solution. The sample was stored at -20*C for further use. 7. AP(1-42) fibrils 20 1 mg Ap~(1-42) (Bachem Inc. Catalog Nr.: H-1368) were solved in 500 pl aqueous 0.1%

NH

4 0H (Eppendorff tube) and the sample was stirred for 1 min at room temperature. 100 pI of this freshly prepared AP(1-42) solution were neutralized with 300 pl 20 mM NaH 2

PO

4 ; 140 mM NaCl, pH 7.4. The pH was adjusted to pH 7.4 with 1% HCI. The sample was incubated for 24h at 37*C and centrifuged (10 min at 10000g). The supernatant was discarded and the fibril pel 25 let resuspended with 400 pl 20 mM NaH 2

PO

4 ; 140 mM NaCl, pH 7.4 by vortexing for 1min. 8. sAPPa Supplied from Sigma (cat.no. S9564; 25 pg in 20 mM NaH 2

PO

4 ; 140 mM NaCl ; pH 7.4). The sAPPa was diluted with 20 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4, 0.2mg/mI BSA to 0.1 mg/ml ( 30 = 1pmol/pl). Materials for dot blot: A0-standards: 35 Serial dilution of Ap-antigens in 20 mM NaH 2

PO

4 , 140 mM NaCl, pH 7.4 + 0.2 mg/ml BSA 1) 100 pmol/pl 2) 10 pmol/pl 40 3) 1 pmol/pl 95 4) 0.1 pmol/pl 5) 0.01 pmol/pl Nitrocellulose: 5 Trans-Blot Transfer medium, Pure Nitrocellulose Membrane (0.45 pm); BIO-RAD Anti-Mouse-AP: AQ330A (Chemicon) 10 Detection reagent: NBT/BCIP Tablets (Roche) Bovine Serum Albumin, (BSA): 15 A-7888 (Fa. SIGMA) Blocking reagent: 5 % low fat milk in TBS 20 Buffer solutions: TBS 25 mM Tris / HCI - buffer pH 7.5 + 150mM NaCI 25 TTBS 25 mM Tris / HCI - buffer pH 7.5 + 150 mM NaCl + 0.05 % Tween 20 30 PBS + 0.2 mg/ml BSA 20 mM NaH 2

PO

4 buffer pH 7.4 + 140 mM NaCl + 0.2 mg/mI BSA 35 Antibody solution 1: Plasma of the TG2576 mice actively immunized 1/400 diluted in 20 ml 1 % low fat milk in TBS Antibody solution 11: 96 1:5000 dilution Anti-Mouse-AP in 1 % low fat milk in TBS Dot blot procedure: 5 1) 1 pl each of the different AP standards ( in their 5 serial dilutions) were dotted onto the nitrocellulose membrane in approximately 1cm distance from each other. 2) The AP standards- dots are allowed to dry on the nitrocellulose membrane on air for at 10 least 10 min at room temperature (RT) (= dot blot) 3) Blocking: The dot blot is incubated with 30 ml 5% low fat milk in TBS for 1.5 h at RT. 15 4) Washing: The blocking solution is discareded and the dot blot incubated under shaking with 20 ml TTBS for 10 min at RT. 5) Antibody solution I: 20 The washing buffer is discarded and the dot blot incubated with antibody solution I overnight at RT 6) Washing: The antibody solution I is discarded and the dot blot incubated under shaking with 20 25 ml TTBS for 10 min at RT. The washing solution is discarded and the dot blot incubated under shaking with 20 ml TTBS for 10 min at RT. The washing solution is discarded and the dot blot incubated under shaking with 20 ml TBS for 10 min at RT. 7) Antibody solution 11: 30 The washing buffer is discarded and the dot blot incubated with antibody solution 11 for 1 h at RT. 8) Washing: The antibody solution |1 is discarded and the dot blot incubated under shaking with 20 35 ml TTBS for 10 min at RT. The washing solution is discarded and the dot blot incubated under shaking with 20 ml TTBS for 10 min at RT. The washing solution is discarded and the dot blot incubated under shaking with 20 ml TBS for 10 min at RT. 9) Development: 97 The washing solution is discarded. 1 tablet NBT/BCIP is dissolved in 20 ml H 2 0 and the dot blot is incubated for 5 min with this solution. The development is stopped by inten sive washing with H 2 0. 5 Results are shown in figure 9. Sera of different immunization groups: a) Ap(20-42) globulomers; b) AP(12-42) globulomers; c) AP(1-42) monomer, 0.1% NH 4 0H; d) vehicle control were tested against different As forms in a dot blot for differing antibody-profiles. 10 1. Ap(1-42) globulomer 2. Ap(1-42) monomer, HFIP pretreated, in 0.1% Pluronic F68 3. Ap(20-42) globulomer 4. Ap(12-42) globulomer 15 5. Ap(1-40) monomer, HFIP pretreated, 5 mM in DMSO 6. Ap(1-42) monomer, dissolved in 0.1% NH 4 0H 7. Ap(1-42) fibril preparation 8. sAPPa (Sigma); (first dot: 1 pmol) 20 In contrast to the active immunizations with either vehicle as control or Ap(1-42) monomer the immunization with AP(20-42) globulomer or Ap(12-42) globulomer exhibits even after approxi mately one year of the last immunization a high titer of antibodies. These antibodies are selec tive for the Ap(20-42) globulomer in the case of the AP(20-42) globulomer immunization or A(20-42) globulomer and Ap~(12-42) globulomer selective in the case of the AP(12-42) globu 25 lomer immunization. This shows that the truncated Ap(20-42) globulomer and AP(12-42) globulomer represent a very good antigen and that antibodies directed against them persist very long in vivo. (Note that on some dot-blots an unspecific staining signal is observed which is most likely a 30 cross reaction of murine antibodies to the BSA used in the serial dilutions of the As peptides.) Example 11: Brain-levels of Ap(20-42) globulomer epitopes in Alzheimer's disease patients. SDS-DTT-brain extract: 35 AD brain samples: RZ 16; RZ 52 und RZ 55 (obtained from Brain Net, Munich) Control sample: RZ 92 (obtained from Brain Net, Munich) 98 One tablet of Complete Protease Inhibitor (Roche, Cat. No. 1697 498) is dissolved in 1ml H 2 0 ( = protease inhibitor solution). 100 mg of AD brain sample are homogenized in 2.5 ml NaH 2

PO

4 , 140 mM NaCl, 0.05% Tween 20, 0.5% BSA (supplemented with 25 pl protease in hibitor solution) with 20 strokes in a glass potter. The suspension is sonified for 30 sec on ice, 5 then incubated at 37 *C for 16 h. The suspension is centrifuged at 100'000 g and 8 *C for one hour, then the supernatant is collected. The residue is dissolved in 5 mM NaH 2

PO

4 , 35 mM NaCl, pH 7.4 and homogenized with 10 strokes in a glass potter. 75 pl of 10% SDS and 125 pl of 0.16 mg/ml DTT are added and stirred for 20 minutes at ambient temperature. The sample was centrifuged for 10 minutes at 1000g-, andthe-supeinatait is stored overnight at -20 *C. 10 Before use the supernatant is thawed and centrifuged for another 10 min at 10'000 g. The su pernatant (= SDS/DTT brain extract) is used for ELISA. a) Sandwich-ELISA for AP(20-42) globulomer epitope 15 Reagent list: 1. F96 Cert. Maxisorp NUNC-Immuno Plate (Cat.No.:439454) 2. Binding antibody: 5F7, 7C6, 1OF11 3. Coupling buffer: 100 mM sodium hydrogen carbonate, pH9.6 20 4. Blocking reagent for ELISA (Roche Diagnostics GmbH Cat.No.: 1112589) 5. PBST buffer: 20 mM NaH 2

PO

4 , 140 mM NaCl, 0.05% Tween 20, pH 7.4 6. Ap(20-42) calibration standard 7. Primary antibody: 25 anti-Ap pRAb BA199; affinity purified (by Ap(1-42) globulomer-Sepharose) IgG so lution in PBS; Konz.: 0.22mg/ml 8. Secondary antibody: anti-rabbit-POD conjugate; (Jackson ImmunoResearch, Cat.No.: 111-036-045) 9. Development: 30 - TMB; (Roche Diagnostics GmbH Cat.No.: 92817060) 42 mM in DMSO - 3% H 2 0 2 in H 2 0 - 100 mM sodium acetate, pH4.9 - Stop solution: 2M sulfuric acid 35 Preparation of reagents: 1. Binding antibody: The individual binding antibodies 5F7, 7C6 and 10F1 1 are diluted to a final concentra tion of 0.7 pg/ml in coupling buffer. 40 99 2. Blocking reagent: For preparation of the blocking stock solution the blocking reagent is dissolved in 100 ml H 2 0 and stored at -20 "C in aliquots of 10 ml each. 3 ml of the blocking stock solution are diluted with 27 ml H 2 0 for blocking one ELISA 5 plate. 3. A(20-42) calibration standard (CS1) The preparation of the AP(1-42) globulerris described i ieximple -. 10 The As(20-42) globulomer protein concentration was determined (6.81 mg/ml) after Bradford (BioRad). 14.68 pl Ap(20-42) globulomer (6.81 mg/ml) are diluted in 1Oml 20 mM NaH 2

PO

4 , 140 mM NaCl, 0.05% Tween20, pH 7.4, 0.5% BSA (= 10 pg/ml). 10 pl of the 10 pg/ml solution are further diluted in 1Oml 20 mM NaH 2

PO

4 , 140 mM NaCl, 0.05% Tween20, pH 7.4, 0.5% BSA ( = 1Ong/mi = CS1) 15 Calibration standards for AP(20-42): Calibration volume of calibration PBST + 0.5% BSA final concentration standard standard AP(20-42) (pg/ml) CS1.1 1ml of CS1 OmI 10000 CS1.2 0.316ml of CS1.1 0.684ml 3160 CS1.3 0.316ml of CS1.2 0.684ml 1000 CS1.4 0.316ml of CS1.3 0.684ml 316 CS1.5 0.316ml of CS1.4 0.684ml 100 CS1.6 0.316ml of CS1.5 0.684ml 31.6 CSI.7 0.316ml of CS1.6 0.684ml 10 CS1.8 0.0 1.0ml 0.0 20 SDS/DTT-brain extracts: SDS/DTT-brain extracts = E# (# represents the 4 human brain samples (1) RZ 16; (2) RZ 52; (3) RZ 55; (4) RZ 92) extraction sample volume of extraction sample PBST + 0.5% dilution BSA factor E#.1 1mI of E# 0.Oml direct E#.2 0.316ml of E#.1 0.684ml 1:3.16 E#.3 0.316ml of E#.1 0.684ml 1:10 100 E#.4 0.316ml of E#.1 0.684ml-- 1:31.6 4. Primary antibody: The anti-Ap pRAb stock solution is diluted to 0.05 pg/ml in PBST + 0,5 %BSA. The an 5 tibody solution is used immediately. 5. Secondary antibody: Lydphilized anti-rabbit-POD conjugate is dissolved in 0.5 ml H2O and mied-iWith 500 pl glycerol. The antibody concentrate is then stored at -20 *C in aliquots of 100 pl. 10 The concentrate is diluted 1:10'000 in PBST buffer. The antibody solution is used im mediately. 6. TMB solution: 20 ml of 100 mM sodium acetate, pH 4.9, are mixed with 200 pl TMB solution and 29.5 15 pi of 3% hydrogen peroxide. This solution is used immediately. ELISA-plate for AP(20-42): Calibration standards (CS1.1-CS1.8) and SDS/DTT-brain extracts of the 4 human brain samples (1) RZ 16; (2) RZ 52; (3) RZ 55; (4) RZ 92 ( = E1-E4 in their 4 serial dilutions 20 E#.1-E#.4) are determined in double: 1 2 3 4 5 6 7 8 9 10 11 12 A CSI. CS1. E1.1 E1.1 E3.1 E3.1 1 1 B CS1. CS1. E1.2 E1.2 E3.2 E3.2 2 2 C CS1. CS1. E1.3 E1.3 E3.3 E3.3 3 3 D CS1. CS1. E1.4 E1.4 E3.4 E3.4 4 4 E CS1. CS1. E2.1 E2.1 E4.1 E4.1 5 5 F CS1. CS1. E2.2 E2.2 E4.2 E4.2 6 6 G CS1. CS1. E2.3 E2.3 E4.3 E4.3 7 7 H CS1. CS1. E2.4 E2.4 E4.4 E4.4 8 8 101 This ELISA is performed with each of the binding monoclonal antibodies 5F7, 7C6, 1 OF1 1. Procedure 1. Add 100 pl mAb solution per well. Incubate the ELISA plate overnight at +60C 5 (fridge). 2. Decant the antibody solution and wash wells three times with 250 pl PBST buffer each. 3. Add 250 pl/well of blocking solution. Incubate for 2 hours at ambient tempera ture: 10 4. Decant the blocking solution and wash wells three times with 250 pl PBST buffer each. 5. Add 100 pl/well each of calibration standards and SDS/DTT brain extracts. In cubate plate for 2 hours at ambient temperature, then overnight at 6 0C. 6. Decant the calibration standards and SDS/DTT brain extracts solution and wash 15 wells three times with 250 pl PBST buffer each. 7. Add 200 pl/well of primary antibody solution and incubate for 1 hour at ambient temperature. 8. Decant the primary antibody solution and wash wells three times with 250 pI PBST buffer each. 20 9. Add 200 pl/well of secondary antibody solution and incubate for 1 hour at ambi ent temperature. 10. Decant the secondary antibody solution and wash wells three times with 250 pl PBST buffer each. 11. Add 100 pl/well of TMB solution. 25 12. Monitor plate colour during development (5 - 15 min at ambient temperature) and terminate reaction by adding 50 pl/well of stop solution when an appropriate colour has developed. 13. Measure extinction at 450 nm. 14. Calculate results using calibration. 30 15. Evaluation: If the extinctions of the samples are beyond the linear calibration range, dilute them again and repeat. Results are shown in figure 10. 35 Brain levels of AD(20-42) globulomer epitopes in brain extracts from AD patients and control subjects A sandwich ELISA was used to assess brain extracts for their truncated AO(20-42) globulomer epitope content. ELISAs with the respective antibodies against the Ap(20-42) globulomer were used for calibration. 40 102 Extraction of Alzheimer's disease brain tissue shows that the Ap(20-42) globulomer epitope content is significantly elevated compared to a control patient. This shows that indeed the AP(20-42) globulomer epitope is a relevant Ap-species in human Alzheimer's disease brain and not only relevant for Alzheimer's disease animal models. Antibodies directed against the 5 Ap(20-42) globulomer epitope therefore are highly desirable for the treatment of Alzheimer's disease. Example 12: Development of anti-Ap(20-42) globulomer hybridoma cell lines 10 Monoclonal antibodies can be prepared using a wide variety of techniques known in the art including the use of hybridoma, recombinant, and phage display technologies, or a combina tion thereof. For example, monoclonal antibodies can be produced using hybridoma tech niques including those known in the art and taught, for example, in Harlow et al., Antibodies: A Laboratory Manual, (Cold Spring Harbor Laboratory Press, 2nd ed. 1988); Hammerling, et al., 15 in: Monoclonal Antibodies and T-Cell Hybridomas 563-681 (Elsevier, N.Y., 1981) (said refer ences incorporated herein by reference in their entireties). The term "monoclonal antibody" as used herein is not limited to antibodies produced through hybridoma technology. The term "monoclonal antibody" refers to an antibody that is derived from a single clone, including any eukaryotic, prokaryotic, or phage clone, and not the method by which it is produced. 20 The particular protocol used to produce the antibodies described herein is as follows: Immunization of mice: Balb/c and A/J mice (6-8week old) were immunized subcutaneously with 50 ug of antigen in CFA. Animals were boosted every three weeks with 50 ug of antigen in 25 lmmuneasyTM (Qiagen) for a total of three boosts. Four days prior to fusion, mice were boosted with 10 ug of antigen intravenously. Cell fusion and hybridoma screening: Spleen cells from immunized animals were fused with SP2/0-Ag14 myeloma cells at a ratio of 5:1 using standard techniques. Seven to ten days post 30 fusion, when macroscopic colonies were observed, SN were tested by ELISA for antibody to Ap(20-42) globulomer. Cells from ELISA positive wells were scaled up and cloned by limiting dilution. Antibody isotype determination: The isotype of the anti-Ap(20-42) globulomer mAbs was de 35 termined using the Zymed EIA isotyping kit.

103 Scale up and purification of monoclonal antibodies: Hybridomas were expanded into media containing 5% Low IgG. Fetal bovine serum (Hyclone). Supernatant was harvested and con centrated. mAb was purified using Protein A chromatography and dialyzed into PBS. 5 Serum titers: Ten mice were immunized with the AP(20-42) globulomer. All mice seroconverted with ELISA titers (1/2 Max OD 450 nm) of 1:5000-10,000. 10 DESIGNATIONS OF HYBRIDOMAS PRODUCING MONOCLONAL ANTIBODIES Internal designations of Abbott Laboratories used for the deposits. Deposited Cell Lines: 15 1) ML13-7C6.1D4.4A9.5G8 (also referred to herein as "7C6") 2) ML15-5F7.5B10 (also referred to herein as "5F7") 3) ML15-1OF11.3D9 (also referred to herein as "1OF11") 4) ML15-4B7.3A6 (also referred to herein as "4B7") 5) ML15-2F2.3E12 (also referred to herein as "2F2") 20 6) ML15-6A2.4B10 (also referred to herein as "6A2") 7) ML13-4D10.3F3 (also referred to herein as '4D10") 8) ML15-7E5.5E12 (also referred to herein as "7E5") 9) ML15-10C1.5C6.3H4 (also referred to herein as "10C1") 10) ML15-3B10.2D5.3F1 (also referred to herein as "3810") 25 104 ATCC 10801 Universtyllvd -Haannas, VA 20110429 -Tecphoom 703-3GS-700 -VAX: 7034652745 BUDAPEST TREATY ON THE INTERNATIONAL RECOGNITION OF THE DEPOSIT OF MICROORGANISMS FOR THE PURPOSES OF PATENT PROCEDURE INTERNATIONAL FORM RECEIPT IN THE CASE OF AN ORIGINAL DEPOSIT ISSUED PUBSUANT TO RULE 7.3 AND VIABILITY STATEMENT ISSUED PURSUANT TO RULE 10. To; (Name and Address of Depositor or Attorney) Abbott Laboratories Attn: Boris Labkovsky 100 Resear Drive --Wercester, MA-Q4Q- Deposited on BehAIf oft Abbou Laboratories Identiilcation Reference by Depositon Patent Deposit Designation Ce line: MLIS-10FIIL3D9 PTA-7239 Cellin: ML137C6.1D4.4A9.508 PTA-7240 Coll line: ML15-5P7.5B10 PTA-7241 Call lin: MLIS-4B73A6 PTA-7242 The deposits were accouipanied byi , a sclentllet description . a proposed taxonomic description Indicated above, The deposits were received December L. ZOOS by this Intenational Depository Authority and have been accepted. AT YOUR REQUEST: . We will inform you of requests for the strains for 30 years. The strais will be made available If a patent office signatory to the Budapest Treaty certifies one's right to receive, or if a U.S. Patent Is issued citing the strains, and ATCC Is instructed by the United States Patent & Trademark Office or the depositor to release said strains. If the cultures should die or be destroyed during the effective term of the deposit, It shall be your respomibility to replace them with Itring cultures of the same. The strains wIi be maintained for period of at least 30 years from date of deposit, or five years after the most recent request for a sample, whichever is longer. The United States and many other conutries are signatory to the Budapest Treaty. The viabhilty of the cultures cited above was tested February 2. 2006. On that date, the cultures were viable. International Depository Authority: American Type Culture Collection, Manassas, VA 20110-2209 USA. Signature of person having authority to represent ATCC: 427-til .4&, Date: february 22, 2006 Marie arris, Patent Specialist, ATCC Patent Depository cc: Cheryl L. Becker 105 ATCC BUDAFES TREATY ON THE INTERNATIONAL RECOGNITION 0OF im DxPOSfT OF MROORGANIMS Y01 THE PURPOSES OVP ATENT PROCEDURE MELIN4TIoNAL FORMf RsCEIPT MN THE CASE OFAN ORIGINAL DEPOSIT ISSUED PURSUANT TO RULE 7.3 AND VLABILITY STATEMENT ISSUED PURSUJANT TO RULE 10. Tat (Namae and Address of Depositor or AUorUsy) Ab~bott Laboratories Anno: Boths Labkovaky 100 RusnrahDive Worcester, MA 01605 Deposited on *behf of; Abbott Laboratories IdentiWication Reference by Depositor:, ratoar Deposit Desdguaton Ilybridome; ML13-MDOM PTA44405 Hybridoam 1MLIS-2FS3E12 PrA-7408 Hybrldozw; D1L15-6AS-4310 ITAo.7409 The deposits wcro accompanied bys _ a slentlc descriptionl* a proposed t*aCounuac description Indica ted above. The dlepoits were received Paboay 2, 00 by tis Internationai Depository Authority nod hove been accepted. AT YOUR EUEST: -E We will Inform you of vequests for the strains for 30 yeams The strains will be made available Its paten i ce signatory to the Budapest Treuly eertifies one's right to receive. or If a U.S. Patent Isied citing the Miun, and ATCC Is Instructed by the United Stater Patent & Trademark Office or the depositor to release said strains. If the cultures should die or be destroyed during the effectiv, ferm of the deposit, It sball be your respotuibillty to replace them with Ulvizg Cuture @1 thme r. The strains will be maintained for a poflod of at Ueat 30 yeas from date of deposit, or five years after the moost recent request ror a sainple, whichever I longer. The Uited States and many other countries are signatory to thme Budapest Traty. The viability of the cultures cited above was tested March 31I. 2f6 On that date, the cultures wete viable. internaloli Depouitosy Authority: American Type Culture Collection, MdanassasVA 20110-2209 USA. Siguslure of person having auitbority to represent ATCC: Jay u afyPaetSped ATCC Patent DepositaryDa: nL .14 cc: Cheryl L. Bcker 106 ATCC 10801 UttveIdl B0v4 - Ks*flkU VA 10110-U"0 - Tedeplase: 103444700 * FAX: 703-4U4745 DUDAPBST TRZATY ON THE U47R1UNATIONAL RECOGNM~ON 0F THE DEPOSIT 0OF MICROORGANISMS FOR THE PURPOSES OF PATENT PROCEDURE &YFFAPA4TIN4L FAMR8 RECEIPT IN THE CASE OF AN OIGINAL DEPOSIT ISSUE PURISUAN4T TO RUL '7.3 AND VIARZVY STAT1WENT ISSUED PURSUANT TO RULE 10, Tot (Name and Address of Depositor or Attorney) Boris Labkovsiq' 109 Research Dr. Worceftur, MA 01605 Deposited on Behalf oft Abbott Laboratories Identification Reference by Depositor: Patent Deposit Desiguatiop House bybridoma cell line- ?LI-73&CE1 PTA-7809 Mows hybridema cell line- M4L15-IOCI.5C6-3H4 PTA-7810 Tne deposits were accozapaniled byt _A-L a scientific description a proposed taxonowic description indicated above. The deposits were received Auwum 16. 2006 by "h Intornittlona.l Depository Authority atad have been accepted. AT VOUR REQUEST: 2L We will inform you of requests for the stras for 30 years. The strains w4I be elide available, if a patnt office pipaatory to the Budapest Treaty certifies one's rigbt to receive, or ff a U-5. Patent Is Issued citing the mtains, and ATCC Is Instructed by the United States patent& Tradentarlt Ofice or the depositor to release said strains. If dhe cultures should die or be destroyed during the effective term of the deposit, it shall be your responuibility to replace themo with living cultures of the Beate. The stritus will be maintained ior a period of at least 30 years from date of deposit, or &sv years after the most recent request for a sample, whichever Is longer. The United States and many other countries are stgnanor to the Budapest Trefity. toe viability orthe cultures cited above was tested Aucfust 2§-2006 . On thAt date, the cultures were viable. International Depository Authoritr. American Type Culture CoillectIon, Manaass VA 20110-2209 USA. Signature cot pespon having authority to represent ATCC Q~io, A CCP Dpo~stor Date; August 31, 2006 cc Cheryl L. Bocicer, Abbot Laboretories *ATCC 10101 UsmversttylUd -MM1 A111.0 tootr7336709 A 0414 TBE DRpOSnF OF MICRtOORGANISMS FOR TMZ PURPOSES OFP ATWtIT PROCEDURE F% ATZTONAL FO"~ RECEIPT W THE CASE OF AN ORIGINAL DIPOSIT ISSUED) PURSUANT TO RULEg 7.3 AND vIABILITY STA7bMNT ISSUED rvRSUANT TO RULE 10.2 Tot (Name and Addtim of Depositor or Attorney) Abbott Laboratorles ATTK{; Boris Labitoviky Joe Reueareb Dr. Worcester, MA 01605 Deiiohtod on Behalf pf., Abbott Labtoratoles Identificati Refenc. by Pepositori Patent Deposit Designation Mouse hybridoma cell ifne;, MLI5-3B 10.2fl.3P1 PTA-785 1 The deposit was accompanied bys _&~ a scientific description a, prulposed tatonoinle description indiated above. The dleposit wu received Sep!MhpXt 1. 2006 by thii International Depository Authority and has been accepted. AT YOUR REQUESTi We XZg~lnform you of requests for the strain foV 30 yeam~ The strain will be made available if a patent offic signatory to the Budapest Thaty certiie one's right to recelve or If a U.S. Patent is isued citing the strain, and ATCC bs Instructed by the United States Patent & Trademak Officc: or the depositor to release iaid strain. lIfthe culture should die or be destroyed during thoeffetive terma of t deposit, it shall be your responsibly to replace it with living culture of the sins. The stral willibe Maintlained for a period of at leant30 years from dAts otdepcult, or &e yeasafter the most recent request for a sample, whichever Is longer. The United Sta ad many other countries are signatory to tho Budapest Treaty. 7h. viability of the culture cited above wras tested September. 11,.2W0. On that date, dhe culture was viable. International Depository Autitoittys Ainericsii Type Culture Collection, Maues, VA 20110-2209 USA. Signature of son having authority to repruseat ATCV: "r 7,onDate! SeutgnberlS2.201 Dec Dishl~p, ArcatenatDEpository cc-, Cheryl Becker

Claims (14)

1. 2. The antibody of claim 1, wherein the antibody is humanized.
2. 3. The antibody of claims I or 2 wherein the antibody comprises a human constant region.
3. 4. The antibody of anyone of claims I to 3 wherein the antibody possesses a human gly cosylation pattern.
4. 5. Antigen-binding moiety of an antibody as defined in anyone of claims I to 4.
5. 6. The antigen-binding moiety of claim 5, wherein the moiety is a Fab fragment, a F(ab') 2 fragment or a single chain Fv fragment of the antibody.
6. 7. Isolated nucleic acid encoding an antibody amino acid sequence of any one of claims I to 6. (6091042 1):GGG 110
7. 8. Vector comprising an isolated nucleic acid of claim 7.
8. 9. Host cell comprising a vector of claim 8.
9. 10. Method of producing an antibody, comprising culturing a host cell of claim 9 in culture medium under conditions suitable to produce the antibody.
10. 11. Antibody obtainable by a method of claim 10.
11. 12. Pharmaceutical composition comprising an antibody or an antigen-binding moiety as defined in anyone of claims I to 6 and a pharmaceutically acceptable carrier.
12. 13. The antibody or an antigen-binding moiety as defined in anyone of claims I to 6 for use in treating or preventing, or diagnosing an amyloidosis such as Alzheimer's disease or Down's syndrome.
13. 14. Method of diagnosing an amyloidosis such as Alzheimer's disease or Down's syndrome which comprises providing a sample from the subject suspected of having the amyloidosis, contacting the sample with an antibody or an antigen-binding moiety as defined in anyone of claims I to 6 and detecting the formation of a complex comprising the antibody or the antigen-binding moiety with an antigen, the presence of the complex indicating an amyloidosis in the subject.
14. 15. Antibody having a binding affinity to an AP(20-42) globulomer that is greater than the binding affinity of the antibody to an AP (1-42) globulomer. Dated 6 March, 2012 Abbott Laboratories and Abbott GmbH & Co. KG Patent Attorneys for the Applicants/Nominated Persons SPRUSON & FERGUSON (6091042_1):GGG