CA2013396A1 - Diagnostic method for alzheimer's disease: examination of non-neural tissue - Google Patents

Abstract

A method for diagnosing Alzheimer's Disease is provided, in which a non-neural tissue biopsy sample is contacted with an antibody capable of identifying .beta.-amyloid protein, or fragment or precursor thereof, and the resulting reaction is monitored. A
kit for conducting the method is also provided.

Description

2nl3.)!~
r DIAGN08TIC H~THOD FOR ALZ~EI~R'8 DI9XA8E
INATION or NON-NEURAL TI88UE

The instant invention generally relate~ to the d-agnosis and monitoring of Alzheimer's Disease.
Alzheimer's Di~ease (AD) is the most common cause of progressive intellectual failure in aged humans, and is typified by gradual neurological ~'eterioration including memory loss, disorientation, and general intellectual regression. It eventually produce~ additional symptoms of physical deterioration, such a~ ~otor disability, incontinence and weight 1088, and is often fatal. It has been observed in races and ethnic groups worldwide and presents a present Ind future major - publ'c health and economic problem. The disease is estimated to affect two to thre~ million individual~ in the United States ;lone.
The problem is compounded by substantial difficulties in making a definitive diagnosis o~ the disease during the patient~s lifetime. Typically, Alzheimer's Disease is definitively diagnosed at autopsy, ~rom an examination of the condition of ~he brain. Alzheimer's Disease i5 earmarked by structural changes in the form of filamentous lesions occurring in particular resions of the brain. These brain lesions are referred to as (1) senile (amyloid) plaques; (2) amyloid angiopathy (deposition of amyloid in small blood vessel~); (3) neurofibrillary tangles; and (4) altered nerve cell fibers (neuritic dystrophy).

-201339~

. I
The brains of virtually all very old humans contain at least some small quantities of the ~ilamentou~ brain lesions which are characteri~tiC of Alzheimer's Disease (AD) and occur within neurons (neurofibrillary tangles), in extracellular cerebral deposits (senile or amylo~d plaques) and in meningeal and --:ebra~ blood vess~31s (amyloid ~nglopathy). Chemical, immunochemical and molecular biological analvses during recent years have shown that a 4.2 kiloDalton (kDa) protein tof about 38 to about 43 amino acids), designated ~-amyloid protein (~-AP) or A4, is a subunit Or the vascular and cerebral amyloid filaments that occur in the br~ins of individuals with AD or with trisomy 21 (Down's syndrome), as well as (to a lesser extent) in individuals undergoing the normal aging process. See Glenner and Wong, Bio~he~ BioDhYs. ~ç~. Commun. 120: 885-890 (1984); Glenner ¦ and Wong, Bioch~. BioDhys. ~ç~. Commun. 122: 1131-1135 (1984);
Masters et al., Proc. ~1- Acad. ~~i- ~A 82: 4245-4249 (1985);
Selkoe et al., J. N~uro~ . 46: 1820-1834 (1986); Kang et al., ~ture 325: 733-736 (1987); and Coria et al., ~ . Pathol. 129:
422-428 (1988). These analyses have also shown that the ~-AP is a small fragment o~ a much larger precvrsor protein that is normally produced by c~lls in many tissues of various animals.
The gene coding for this ~-AP precursor protein in humans is located on the long ar~ o~ human chr. -s ~ 21.
The first purification and partial anino acid sequencing of the ~-amyloid protein of AD was reported by Glenner and Wong in 1984 (~lenner and Wong, Bioche~. ~ioDhys. Res. Commun. 120: 885-890, (1984)). They isolated ~-AP fron the meningeal blood vessels of patients who died with AD, using a modification of a long-established protocol for purifying a~yloids from tissues.

They were able to establish the amino ac d gequence of the first 28 amino acids of what was later shown to be a 38-43 amino acid protein. The isolation procedure that they used and their L

~U133~t~

r sequence data are described in U.S. Pat. No. 4,666,829 issued toGlenner et al. on May 19, 1987. In 1985, they made synthetic peptides includlng the first 10 amino acids of the 28-amino acid sequence ldentifled by them. These synthetic peptides were then injected into either rabbits or mice to produce, respectively, polyclonal or monoclonal antibodies. These antibodies detected (i.e., reacted with) both the senile (amyloid) plaques and the vascular amyloid deposits in AD and normal aged brains.
PCT application having Int. Pub. No. Wo 88/03951 assigned to California ~iotechnology, Inc., is directed to certain DNA
sequences encoding the ~-amyloid precursor protein, and to the use Or their recombinant proteins for generating antibodies for cerebrospinal fluid or serum diagnosis of Alzheimer's Disease.
Throughout the more than 80 years since Alzheimer's Disease was first identl~ied as a disease> the pathological lesions of AD
have been found only in the brain. Since the examination of a patient's brain tissue by biopsy is particularly invasive and, therefore, very rarely carried out, a need exi~ts for an effective, less invasive diagnostic method that can be used simply and safely on a living subject.

According t~ : ~irst aspect of the inventior., a method is disclosed for diagnosing Alzheimer's Disease. The method com-prises the steps of obtaining a non-neural tissue biopsy sample, contacting at least a portion of the sample with at least one antibody capable Or identifying ~-AP, a ~-amyloid precursor protein ~ragment comprising ~-AP, or a ~-AP peptide fragment of about 8 or morQ amino acids, and monitoring the extent of the reaction between the sample and the antibody.
According to another aspect of the invention, a kit is disclosed for diagnosing Alzheimer's Disease, comprising one or more antibodies a~ noted above, and means for detectinq the ~
extent and specificity of the reaction of the antibody with a non-neural tissue sample.

'~.
These and other aspects of the invention and advantages attendant thereto will best be understood by a reading of the folloving definitions, detailed description, examples and appended claims.

The term "amyloid" is defined herein as a proteinaceous ~~ filament of about SO-100 Angstrom diameter, having certain dye-binding properties (e.g., the binding of molecules of Congo red in a particular orderly array, such that when polarized light is passed through the amyloid, a green birefringent color is seen;
and the binding of thioflavin S, which produces a green fluorescent signal when viewea in a microscope outfitted with epifluorescence), and accumulating progressively in the extracellular spaces of a tissue. An additional characteristic of amyloid is the presence of extensive ~-pleated sheet protein confor~ation, as established by X-ray diffraction analysis.
Amyloid is a generic term that refers to proteinaceous filaments as defined in this paragraph, regardless of the tissue in which they appear or the protein of which they are composed, and occurs in a wide variety of diseases. Notwithstanding the above definition, the instant invention is directed to the type of amyloid that is characteristic of Alzheimer's Disease and Down's Syndrome, the subunit protein of which is the ~-amyloid protein defined below.
The term "~-amyloid precursor protein fragment comprising ~-AP" as used herein i~ defined as a polypeptide that is a portion of a ~-amyloid precursor protein which is larger than, but includes within it, the ~-amyloid protein.

r The term "~-amyloid precursor protein" as used herein is defined as a polypeptide that is produced by a gene localized in humans on the long arm of chromosome 21 and that includes within its carboxyl third the ~-amyloid protein t~-AP). Examples o.
specific forms of the B-amYloid precursor protein which are currently known to exist in humans are the 695-amino acid polypeptide described by Kanq et al., Nature 325: 733-736 (1987);
the 751-amino acid polypeptide described by Ponte, et al., Nature 331: 525-527 (1988) and Tanzi et al., Na~ur_ 331: 528-530 (1988);
and the 770-amino acid polypeptide described by Kitaguchi et al., - ature 331: 530-532 (1988).
The term ~-amyloid protein" (~-AP) as used herein is an approximately 4.2 kDa protein which, in the brains of AD and Down's syndrome patients, forms the subunit of the amyloid filaments comprising the center of the senile (amyloid) plaque and is depositsd in small vessels and in amorphous, non-filamentous parenchymal (tissue) deposits. The ~-AP can occur in a filamentous polymeric form (in this form it exhibits the Congo-red and thioflavin-S dye-binding characteristics of amyloid described in connectlon therewith). It can also occur in a non-filamentous forc ("preamyloid" or "amorphous" or "diffuse"
deposits) in tissue, in which form no detectable birefringent staining by Congo red occurs. A portion of this protein in the form obtained from the meninges or meningeal blood vessels is described in U.S. Pat. No. 4,666,829 issued to Glenner et al. on May 19, 1987. B-Ap when used in connection with this invention refers ~o an approximately 38-43 amino acid peptide that is substantially homologous to the form of the protein produced by the method described in the Glenner et al. patent, but which, according to the instant invention, is preferably extracted from the cerebral cortex of AD patients or, alternatively, from the skin (or other non-neural tissue) of AD subjects, although other sources of immunoreactive material including synthetic peptides 2~ ~ "3!3~

having all or part (e.g., about 8 or more amino acid residues) of the sequence of ~-AP can also be used in conjunction with this invention. In whatever form, ~-AP i~ a hydrophobic 38-43 amino a_id frag~ent of a large glycoprotein, referred to as the ~-amyloid precursor protein, encoded by a gene on the long arm of human chromosome 21. It is further characterized by its relative mobility in SDS-polyacrylamide gel electrophoresis and its propensity to aggregate into polymeric filaments having the ultra-structural and tinctorial properties of amyloid. Its 43-amino acid sequence ig Asp Ala Glu Phe Arg His Asp Ser Gly Tyr Glu Val ~is Gln Lys Leu Val Phe Ala Glu Asp Val Gly Ser Asn Lys Gly Ala Ile Gly Leu Met Val Gly Val Ile Ala Thr or is substantially homologous thereto. It can be extracted from the cerebral cortex by the method described in Selkoe et al, Journal of NeurochemistrY 46: 1820-1834, 1986, which is hereby fully incorporated into this disclosure by reference as though fully set forth herein (see also Example IA), or frsm the skin by the method set forth in Example IB below.
The term "~-AP peptide fragments of about 8 or more amino acids" as used herein means a fragment of ~-AP (or a peptide substantially homologous thereto), capable of producing antibodies that will react with ~-AP or a form thereof present in the skin or other non-neural tissues of AD subjects.
The term "brief fixation" as used herein in conjunction with the preparation of a biopsied skin sample or other non-neural tissue sample for AD diagnosis means treating the sample with an agent (termed herein as a "fixative") that stabilizes the molecules in the sample by cross-linking them or other~ise producing strong intramolecular and intermolecular bonds, this treat~ent occurring in a period of time that i5 long enough to produce structural stability and resistance to degradation of the proteins in the sample but not so long as to markedly alter the molecular form or conformation of ~-AP-immunoreactive antigens in the sample. Typically, brief fixation in accordance with this invention will involve treatment with an aldehyde-containing fixative, such as 10% neutral buffered formalin, Bouin's solution (70% (vol/vol) saturated picric acid, 10% (vc;/vol) formalin, 5 (vol/vol) glacial acetic acid) in water or PLP fixative (periodate-lysine-paraformaldehyde in relative concentrations (weight/vol) of 0.2~: 1.4~: 2% in water). Treatment time will preferably be from about 15 to about 60 minutes, although these times can vary from as little as about 10 minutes to as long as about 3 hours depending on the nature of the fixative. However, conventional fixation for prolonged periods (e.g., for weeks to months) can also be carried out.
The term "intestine" includes small intestine, large intestine (colon), and rectal tissue.
The term "skin biopsy sample" is defined herein for the purpose of the appended claims as ircluding either or both of cutaneous and subcutaneous tissues.

The subject invention results from the unexpected and entirely novel detection of ~-AP deposits in tissues of non-neural organs in AD patients, including skin, subcutaneous connective tissue, and intestine. This discovery provides the first demonstration that the ~-amyloid protein, which is invariably and progressively deposited in the brain in AD, is also present in non-neural tissues, particular y the skin and intestine. Accordingly, the invention is generally directed to a method for diagnosing Alzheimer~s Disease (AD) by examining a 2al3~3~

r: non-neural tissue sample and detecting the presence and quantityo~ ~-AP (or a ~-amyloid precursor protein fraqment comprising ~-AP) therein. The detection and quantitative measurement of ~-AP
in such tissue in accordance with this invention is useful in confirminq a clinical diagnosis of AD in demented patients and in following the course of the illness. The $nvention can be applied, for example, to monitor the disease during a period of treatment with agents that stabilize, decrease or prevent the depo~ition of ~-AP into tissues, including the brain.
According to the present invention, monoclonal or polyclonal antibcdies produced to various forms of ~-amyloid protein (or a ~-amyloid precursor protein fragment comprising ~-AP) can be used in an immunoassay on biopsied samples of skin, intestine, or other non-neural tissue to diagnose patients with Alzheimer';
Disease (AD). A preferred embodiment involves testing such samples with antibodies produced by using the ~-AP extracted from AD cerebral cortex as immunogen. The extraction can be carried out according to (or in a mar..ner similar to) the method of Selkoe et al., Journal Q~ Nouro~h~mistrv 46: 1820-1834 (1986) and of Selkoe and Abraham, Methods in Enz~moloqY 137: 37-44 (1386), which articles are hereby fully incorporated by reference herein.
The method of car~ying out the preferred extraction is taught in Example IA belov. Alternatively, the antibodies used for detection of the ~-AP can be produced by the immunization with ~-AP antigen extracted from the skin (or othsr non-neural tissue) of Alzheimer Disease victims, as described in Example IB. ~-AP
extracted from AD meningeal blood vessels or synthetic peptides having part or all of the sequence of ~-AP can also be used as immunogens, although the resulting assays are generally less sensitive (see Table I).
The antibodies used in the diagnosis of the skin, intestine, or other non-neural tissue sample can be polyclonal or monoclonal, and can be made by methods now well-known in the art.

,, 2û~339F;

r For example, polyclonal antibodies made to the extract of Example IA or IB can be produced by immunizinq an animal such as a rabbit and then purifying the antibodies as described in Example IC, while monoclonal antibodies to said extracts can be made, for example, according to the method of ~ohler and Milstein (Nature - (1975) 256:495) by immunizing an animal such as a mouse, extracting splenocyte~ from the spleen thereof, fusing them with mouse myeloma cells to make hybridomas, and screening and subcloning the hybridomas, all as described in Example ID.
- It i3 interesting to note that non-neural ~-AP deposits (like many so-called preamyloid deposits now known to occur in brain) are veakly or not reactive with the classical ; histochemical stains for amyloid, Congo red and thioflavin S; the skin, which demonstrates such deposit~ readily and is most easily sampled, has not been thought to be a site for ~-amyloidosis in AD; and generally, antibodie~ to ~-AP synthetic peptides are currently being used by AD investigators other than the inventor and his coworkers, but these detect non-neur~l deposits such as those in the skin ~uch less sensitively than antisera to native cerebral ~-AP such as those made by the inventor and his co-workers and shown in Table I. The instant discovery is particularly unexpacted because of the~e and cther factors.
Antibodies produced from synt~etic peptide immunogens which are generally used in the art for staining AD brain tissue include those listed in Table I as L and Y, which are similar to those described in U.S. Patent No. 4,666,829 by Glenner and Wong.
Moreover, as shown in Table I, the ~-AP deposits are generally less detectable when using antibodies produced against imrunogen extracted from the meninges of AD victims than when using antibodies produced against immunogen extracted from the cerebral cortex of AD victims. These observations suggest that the form of ~-AP found in skin and other perivascular connective tissues 'Vl~;~"t;

r resembles the native ~-AP molecules in brain tissue more closely than it doe~ synthetic ~-AP pept$des.
Interestingly, ~-AP im~unoreactivity ~ound in the skin and intestine has certain properties that dif~er from ~-AP found in - the meninges or brain tissue. For example, the reactivity with antibodies to synthetic ~-AP is generally destroyed by pretreatment with formic acid, in contrast to the enhancement of ~-AP staining by synthetic peptide antib~ies that this reagent generally produces in the brain deposits. Accordingly, in the prererred embodlment Or the invention, skin and intestine samples should not be pretreated ~ith formic acid, and should be preserved by brief ~ixation (e.g., from about 15 to about 120 minutes, pre~erably for about 30 minutes) in one of several possible flxatives, as described in full detail in Example III
below.
In practicing the invention, a skin, intestine, or other non-neural tissue sample is removed from the patient by a biopsy procedure. Preferably, one or more punch biopsies are taken from an area o~ the skin wlth a lo~ density Or nerve endings and littl~ sensory importanc-, such as the inside surface of the arm, to minimize discomfort, although other types of biopsy, such as an elliptically-shaped incisional biopsy, can be taken from any area of the skin. An intestinal biopsy can similarly be obtained by methods well-known in the art.
Samples of skin, intestine or other non-neural tissue obtained by a biopsy (punch biopsy or incisionzl biopsy) from patients suspected of having Alzheimer's Disease (or having another disorder charac~eerized Dy ~-AP deposition in brain, e.g., Down's syndrome or advanced aging) can be quick-frozen for subsequent ~-AP extraction, solubilization and quantitative immunoassay. Alternatively, the non-neural tissue samples can be denatured in a fixative and then examined i~munohistochemically.

2013~

For the purposes of extraction, solubilization and quantitation of the ~-AP in skin, intestine, or other non-neural tissue, the samples should not be fixed but can be extracted in a buffeL (e.g., those described in Example IB) or stored frozen until used.
For the purposes of immunohistochemical analysis, the preferred embodiment involves the fixation of the biopsied skin, intastine, or other non-neural tissue samples from the patients for brief periods of time. ~he purpose of such brief fixation is to avoid conformationa. alterations from prolonged fixation that -could destroy the antigenic sites on the ~-AP-containing molecules in the samples. For example, the samples can be fixed for 1'-60 minutes in 10% neutral buffered formalin, follo~ing ~hich the sample i5 removed to a physiological buffer (e.g., pho~phate buffered normal saline, pH 7.6, with 0.02% sodium azide) and stored at 4-C. Alternatively, a longer period of fixatlon can be used, but in the preferred embodiments of this invention, brief fixation (i.e., fixation for generally less than three hours) is called for. Other fixatives can be used including a picric acid-containing fixative (e.g., Bouin's method) or the PLP fixative (periodate-lysine-paraformaldehyde).
Af~er this treatment, the fixed sample can be further prepar~d for immunohistolo~ic staining. For example, the sample can be embedced in a medium such as paraffin and sectioned in a microtome, and typically, 5-15 ~m-thick sections can be mounted on glass microscope slides for immunocytochemistry. Alter-natively, a microtome in a cryostat can be used to prepare frozen sections from eit~.er a fresh-frozen or a fixed-frozen sample of skin, in~estine, or other non-neural tissue. The inclusion of some subcutaneous tissus in the skin sample is often useful diagnostically, since perivascular ~-AP deposits can be seen to advantage in this reqion.

20~33~J

The skin ~or other non-neural tissue) samples, whether mounted on slides or used free-floating, are then contacted (i.e., reacted; with an antiserum or monoclonal antibody, such as those prepared as described herein and set forth in Table I. The primary antibody, or a control aliquot following absorp~ion of the primary antibody with a form of the ~-AP immunogen, can then be placed on the sample and incubated for about 2 to about 4 hours, or overnight. Thereafter, tha samples are washed and incubated with a secondary antibody reactive against the first, e.g., goat anti-rabbit ~gG antibGdy or goat anti-mouse IgG or IgM
antibody. Ihe secondary antibody can be labelled. For example, it can be coupled directly to an enzyme marker such as pcroxidase-antiperoxidase or ~-an be coupled with biotin followed by reaction vith an avidin-biotir. complex coupled with peroxidase. Such a reaction product can be detected colorimetrically by incubation with 3,3'-diaminobenzidine (DAP).
The stained sample is viewed through a light microscope.
Alternative method~ for analysis Or immunohistochemical sections include using an alkaline pho~phatase-coupled seccndary antibody; or using a fluorochrome-coupled second antibody (e.g., fluorescein), in which the reaction of the primary antibody with the fixed skin, intestine, (or other non-neural tissue) section is assessed u~ing a light micro~cope outfitted with epifluorescence.
_n the preferred embodiment, several ~-AP antibodies (or antibodies to amyloid precursor protein fragment3 comprising ~-AP
or ~-AP peptide fragments of 8 or more amino acids) prod~ced as described in the rollowing examples (and as further exemplified in Table I) can be used 3S primary antibodies to detect ~-AP
deposits in AD skin, intestine, or other non-neural ti-sue sampl2s. ~wo typeQ of controls can be provided by comparing the primary antibody-stained samples to: (1) like samples from the AD patient reacted with the immunogen-absorbed primary antibody ~1.,3~

r. or with normal rabblt serum, and (2) samples of normal tissue (free of ~-AP) staln-d vith the primary (i.e., anti-~-AP) antibody. In the pr~ferred embodiments, cor.trols are carried out to deter~ine whether the suspected ~-AP depo~it that has been identified by a primary ~i.e., anti-~-AP) antibody: (1) will produce a positive a~ay signal following absorption of the primary antibody wlth a form of ~-AP, or (2) will be recognized by an irrelevant antiserum, an irrelevant monoclonal antibody or an irrelevant preimmune serum. The completeness of the absorption of the pri~3ry B-AP antibody by a ~-AP antigen can be confirmed by reactin- . iquots of both the unabsorbed and absorbed antib~dy wlth sections of AD brain tissue and demonstrating that known ~-AP-containlng lesions (e.g., senile - plaques ar~ amyloldotic blood vessels) in the brain are r,o longer stained by the ab~orbed antibody.
- Once the depo~ltion of ~-AP in a skin, intestine, or other non-neural tissue sample i3 detected, the stained sample should be compared to standard immunohistochemical sections (both negative and posltlv- controls), preferably provided with the kit of the present invention, in the case of _he fixed samples; or to a standard antigen curve in the case of a quantitative immunoassay, or otherwise evaluated to determine whether the sample has tested positive for AD. An example of a quantitative competitive immunoa-say i~ described in Example VII.
In an immunohistochemical analysis, the amount of ~-AP
immunoreactive deposlts in the skin, intestine, or other non-neural tissue can bo assessed semi-quantitatively (e.q., O =
absent; 1 ~ veak stalnlng; 2 = moderate staining; and 3 ~ strong s~aining) by the microscopist using a relative scale based on experience or simultaneously-assessed standards of known AD

patients having ~-AP skin ~or other non-neural tissue) deposits.
Al~ernatively, the approximate cross-sectional area of microscopic section~ that contain ~-AP immunoreactive material 2~ 33~i~J

r can be determined through the microscope using an ocular graticule (or other quantitative microscopy device) and compared ~; to a standard range Or the areas o~ such B-Ap deposits previously determined in a series of AD versus like, age-matched, non-AD
samples.
In addition to the use of immunohistochemistry as described above, another preferred embodiment involves the extraction of p-AP (and/or ~-amyloid precursor protein fragments comprising ~-AP) from fr~sh or fresh-frozen biopsied s~in, intestine, or other non-neural tissua samples obtained from putative AD
patients and from age-~atched controls. The extraction of ~-AP
(or B-AP-containing molecules) and its solubilization can be carried out as described in Example IB. A ~-AP antibody such as those described in the Examples or exemplified in Table I can be reacted in a diagnostic immunoassay using standard methods (see ~or example, Campbell, et al., Hethods çf Immunoloqy (~964)). A
radioi ~nuassay (RIA) or enzyme-linked immunosorbent assay (ELISA) can be employ~d for immunodetection of an extracted, prefQrably solubilized, for~ of ~-AP. A variety of labels can be used, such as radioisotopes that include t~I, ~C, sH, enzymes, fluorophores or chemilu~inescent molecules. The amount of ~-AP-immUnoreaCtiVQ moleculQs extracted from the biopsied samples can bQ compared to a standard curve, whether using a competitive or non-competitiv~ assay.
Table I below sets forth six distinct, well-characterized antisera to nati~e or synthetic p-AP which detect p-amyloid protein tissue deposits: namely, antisera A, C, F, Y Ph and L.
Antisera A and C were produced against AD cortical ~-AP
immunoqen, as described in Selkoe et al., J. ~eurochem. 46: 1820-1834 (1986) and Selkoe Qt al., Science 235: 8~3-877 (1987), while antiserum L was produced against synt~letic ~-AP immunoqen, as described in the latter. Ar.tisera F and Ph were produced against AD meningovascula- p-AP i~munogen, as described in Joachim et 2~i., ''~t~

r~ I
al., ~ n Research 474: 100-111 tl987)- Antiserum Y was prepared as described in Example II below.
In addition, Table I sets forth six irrelevant antisera for comparison, namely, antibodies P, DJ, G, T, AA and C~7.
Antiserum P is described in Ihara, Abraham, Selkoe, Nature 304, 727-730 (1983). Antiserum W is described in Selkoe, BQ11~
Podlisny, Price, Cork, Science ~1~, 873-877 (1987). Antiserum G
is described in Dahl, ~ln Res. 57, 343-360 (1973). Antiserum T
was obtained from Calbiochem. Antiserum AA is described in Shirahama, Cohen & Skinner, in Advance~ Ln Immunohistochemistry, 277-302, (1984). Monoclonal antibody CB7 is described in Ju, Skinner, Shirahama & Cohen, Fed. Proc. 46, 1326 (1987). The references cited in this paragraph are hereby fully incorporated into this disclosure by reference.
To confirm the specificity of antibodies to ~-AP, they were absorbed with highly purified ~-AP. While the ~-AP-specific antibodies detected the ti~sue deposits before this absorption, they did not do ~o aft-rwards, confirming that they were in fact detecting ~-AP as deposited in the skin or intestine. Similarly, the antibodies staine~ ~-amyloid deposits in the brain before, but not a~ter~ the antibodies were absorbed with ~-AP specific antigen. Antigen-absorbed and "control-absorbed" aliquots (both in lS0 mM NaCl, 50 mM Tris, pH 7.6) were reacted simultaneously with AD brain sections, AD skin or intestine sections and dot blots of the ~-AP-containing antigen. In the case of antiserum A, for example, a partially purified fraction of detergent-extracted senile plaque cores from AD cerebral cortex (as described in Example IA, below) was used as the absorbent; a "control absorption" employed a fraction prepared identically from normal aged human cerebral cortex which contained the contaminating particles of core fractions (e.g., lipofuscin granules, collagen, and microvessel fragments). In each such experiment the AD plaque core absorption markedly diminished or -2 .. ' t ~, ~, 3 i abolished the staining of plaque and vascular amyloid in AD
brain, the dermal and subdermal amorphous ~-AP deposits, and the ~-AP-containing antigen on dot blots. The control absorptions produced no change in these immunoreactions. Similar results were obtained with each of the other ~-AP antisera listed in Table I after absorption with relevant antigen. Preimmune sera, secondary antibody alone and various hyperimmune sera (see Table I) to puri'ied paired helical filaments (PHF), tau protein, human glial fibrillary acidic protein, and human transthyretin (prealbumin), all failed to label specifically the skin ~-AP
deposits. The six ~-amyloid antisera varied greatly in the sensitivity of their detection and precise staining patterns cn I non-neural ~-AP deposits. Antiserum A, produced to the native 4.2 kDa ~-AP purified by HPLC from amyloid filaments extracted from AD cerebral cortex, was the most sensitive detector.
Antisera to HPLC-purified meningovascular ~-AP (F and Ph in Table I) and to the synthetic ~ 38 peptide (Y in Table I) were considerably weaker, although the intensity of labeling at a standard dilution Or any one ~articular antibody varied from case to case and even among the ~-AP deposits found within one skin section of a particular AD case.
~ Tne sensitivity ratings set forth in ~able I are based upon i the relative detecrion of ~-AP in adjacent skin sections from the same AD patient. The ratings are based upon a peroxidase/
diamino-benzidine (DAB) immunodetection system and are intended as an example of the variability among different ~-AP antibodies in detecting skin ~-AP deposits. A rating of high means moderately intense or intense brown (DAB-positive) immunoreaction prcduct; a rating of moderate means medium or medium-low intensity of brown react.on product; and a rating of minimal means barely detectable brown reaction product, under the specific immunohis'ochemical conditions used. A high intensit~

2n~
~~
rating for a given antibody i5 preferred for ease of interpretation of results.
Table II summarizes the results of an examination of skin, subcutaneous tissue or colon from 4. human subjects: 11 with AD, 26 without AD and 4 with Down's Syndrome. All but 4 had been autopsied, establishing precise neuropathological diagnoses.
Eight of the 11 AD cases showed definite B-AP-reactive non-neural deposits and two showed slight, equivocal reactivity. In contrast, only 3 of 26 non-AD subjects showed definite non-neural ~AP reactivity; all of these were ~ 77 years old. Of 4 Down's syndrome skin samples, 2 were posi.ive and 2 were equivocal. In addition to autopsy-proven cases, fresh punch biopsies of forearm skin were examined from one AD and one normal subject: the former showed specific ~AP-staining, whereas the latter did not.
~ AP-immunoreactive deposits were located in the dermis in a patchy distribution beneath the epidermal/dermal junction, sometimes near small blood vessels or glandular elements. The stained material appeared to be localized extracellularly among connective tissue fibers in the reticular dermis. Some deposits were found circumferentially around small vessels, particularly arterioles. Most vessels and other skin structures (sweat and sebaceou~ glands; hair follicles; hair shafts; smooth muscle;
epidermis; keratin) were unstained, attesting further to the specificity of the reaction. In subcutaneous tissue and the submucosal tissue of the colon, ~-AP-reactivity was found around microvessels and in perivascular connective tissue, apparently extracellular. Deposits of ~-AP were most easily visualized in the intestine (both ~mall and large), because they often occurrred in a distinct vascular locus rese~bling that found in the brain and meninges. Almcst all of the dermal deposits observed had an amorphous, non-fibril'.ar appearance, resembling somewhat the diffuse ~-AP deposits abundantly present in AD

brains and the brains of some patients with hereditary "~1 ,3'.3~
r cerebrovascular ~-amyloidosis of Dutch type. The non-neural deposits lost their reactivity wi'h antibodies to synthetic ~-AP
(y in Table I) when sections were pretreated with formic acid, a reagent that can enhance the staining of some brain ~-AP
deposits. The deposits in skin were not detected by Congo red or thioflavin S.
These results provide the first demonstration that the ~-amyloid protein, which is invariably and proqressively deposi~ed in the brain in AD, is also deposited in extracerebral tissues.
Tests on four specific antibodies (A,C,F and Y, Table I) to native or synthetic ~-AP showed detection of the tissue deposits before but not after ~AP antigen absorption; staining of ~-amyloid ln brain was simultaneously abolished. Antisera to proteins other than ~-AP tested to date do not label the non-neural deposits.
Of course, it should be understood that the foreqoing detailed description Of Freferred embodiments and the following Examples are merely illustrative and not limiting. Many variations will be apparent to one of ordinary skill in the art, and accordingly, the invention is intended to be limited only in accordance with the claims and all legal equivalents thereto.

C~ ~'J~ ~ 3 ~
?
r TABT~ r: SUMMARY OF ANTIBODIES
.

Antiserum T~muno~en AssaY SensitivitY
; A -4 kD ~AP purified by HPLC
from AD cortex high C -4 kD ~AP purified by HPLC
from AD cortex moderate F -4 kD ~AP purified by HPLC
from AD meninges moderate Y synthetic BAP~ ~ peptide moderate Ph -4 kD ~AP purified by HPLC
from AD meninges minimal L synthetic ~APl2t peptide minimal ~P isolated paired helical filaments negative DJ heat-stable MAPs (includinq tau) negative from bovine brain G glial fibrillary acidic pro'ein negative T human transthyretin negative AA isolated AA amyloid fibrils negative Monoclonal AntibodY
CB7 isolated AA amyloid fibrils negative Series of irrelevant hyperiD une sera (see Examples II and III), presented ~or comparison only.
Antisera A and C are described in Selkoe, Abraham, Podlisny, Duffy, J. Neuroche~ 46, 1820-1834 (1986) and Selkoe, Bell, Podlisny, Price, Cork, Science ~, 873-877 (1987).
Antiserum L is described in and Selkoe, Bell, Podlisny, Price, Cork, Science ~, 873-877 (1987).
Antiserum F and Ph are described in Joachim, Duffy, Morris, Selkoe, ~L~n ~Ç~. 474, 100-111 (1988).
Antiserum Y is de~cribed in Example I.C. herein.
Antiserum P is described in Ihara, Abraham, Selkoe, Nature 304, 727-730 (1983).
Antiserum DJ is described in Selkoe, Bell, Podlisny, Price, Cork, Science 235, 873-877 (1987).
Antiserum G is described in Dahl, ~in Res. S7, 343-360 (1973). Antiserun T was obtained from Calbiochem.
Antiserum AA is described in Shirahama, Cohen & Skinner, in Advances ~n i~l1nohistochcm~str~, 277-302, (1984).
Monoclonal antibody CB7 is described in Ju, Skinner, Shirahama & Cohen, EÇ~. ~L~- 46, 1326 (19P7).

2~3 .~'~ 6 r Table II, set forth belov, demonstrates the ~ethod of the present invention for the diagnosis of AD. Demonstrable imounoreactivity i~ generally diagnositic of AD, and is indicated by thQ + valuer shown in the Table.
- TABL~
~AP IMMU~OPT~CTIVITY I~ ~O~-NEURAT TISSUE SA~.PLES
FROM 41 HUMU~I SU3JECTS

BAP I ~0 -I$~rolog/c-l r~ c~ y Subiect ~ ~ ~'~ Yeuror~tholocY 51in In~eS~ine ALDE 1~ s DI ~ISE
Slin:
A~$~$-~6 'J9 f AO (~S ~l/"_o~ c CVA A0, ~1 AA; r_o~e CV~
X1330 f~5 r A0 (6)~l~vSrl9 p t~ ntl Ao~S - 77 t~ JS AD ( t ~ )AS~ ~ AA
Afl~$-213 S}2 Sl A0 (2~A0 ~1 AA, cut- NA
Af5-196 7~1 r A0 (7~ AD, ~r~~S AA
A66 55 7r f ADfPO tll ADtPD~, ocS AA
Atr 237 76 ~ ASJ ~7) AD, ~i AA
Ao7-~s 7 ' AD (-) AD, ~1 Ight AA s AfJ~ 10 71 St AD ts) AD, ~S AA
At9-1~ 62 t AD <LS~ A0, ~ AA s Int tln-AfJts-~6 69 f A0 ~>5)~ r_ot- S V~ AD, ~1 AA; r_ot- CV~ s A~$~$-lS9 ~S7 f A0 (>S~ AS~, _rlc-cS AA
AOED D I SEUED
SS~ln Ao7-1~ ~Cs r Po, ~_nti- IS~SSo S'D, ~1 ~D, no AA
AUS - 2 3 ~12 f PD ( a ~ nt i ~ ( 1 ) IS l gro l /$~r l 1 ' d~ 1 cl~n, no p, ~light AA
Aat 52 ol r l_ot- poll~lltl~ I_ot- polio, no p, AU-130 79 r R_ot- CV~ ot- CV~ p, Ja9--a 7~1 r ~nt-l r-t-rd tlon lI ot- contu lon no p, no AA
A~1ti- 125 77 f R_ot- CV~ _ot CvA OCi ~P;
Int t~n-:
Aoti-125 77 F R_ot- CV~ I--ot CVA, ~ ~p, JU-l9~ 62 t C-rcino_ ~tcrtic c-rcin~
AOED ~ no p, no AA
Sl~ln A~i7-~S ~10 11 Yor_l ll-r- w, rlight AA s Aô7-1~7 76 11 Yor~ O p ~ AA
Ali7-302 n r Yo~l xO p no AA
Aat-3 7 n r yOr_l Yo p, ~light AA s X1362 6~ 1~ Yor_I ~ I Ivlrg p t i ntl Il~-AOED ~L
S~in A~~7-26ti 61 r yo~_l llor_l Aa9-3 60 ri Creln tu r ClioblA-to_ AUo 376 56 ~ Yor_l Yo~_l A~~o 259 55 t Y~_l Y0 Aoti-25~~ ~9r Yor_l Y~_l -AU-35a ~5 ~ llor~ or_I -A~~ti-367 ~5 t Yor_l "O'~' Ja3 Z56 '.2 r spln l corci ~ra_ Tr-~_tk ~lopsthy A~~-2~10 35 r 9reln tu r Clioblc~t~
AU-30~ 32 ~1 Iio~_l Yor_l A~~-300 30 ~1 ~or_l ~ID
AU-159 r-tu- p,_turitr C r~n l ~tr~JI he~norr Inte~tine Aati-la3 3a r Yoc_l Y~'--l ~a-160 z- t Yor_l YOr_l Approximate duration (in ~ears) of disease.
Typical neuropathological findings of both AD and PD.

20~33~6 r DC~ ' S mo~
Stin ~o ~o ~ ~ Oo~s ~ndr~ llod ~p, sl AA
A~5 1-0 3J 11 Do~'~ ~vndr~ lod sp, r~ ~ t ~S-~O ~ r Dc~'~ ~~rdr~ ~ p, slig~t A~
1~-61 25 ~ Do~n's s~ndrc~ p, ra M
AD, A~ r~ ; P~\, V-rtin~ s dl~ yloid ~iop-th~; CVA, c-r~ro~-scul-r ~cci~, sp ry~ortic-l ~nil- pl~; ~d, ~r-t-; ~ ht; ~ lnit-, ~pKit~C ~t~iAi~; ', no s~-~nir~; ~, sli~hl or i voc-l ~ ~ i ni n~ IlD "~ot d t~

~ LF8 Example I: Immunoaen and Antibodv Production A. Extraction of ~-Amyloid Protein from the Cerebral Cortex Human brains of patients who died with Alzheimer's Disease were removed at autopsy and frozen at -70-C. For extraction of ~-AP from brain, only brain tissue markedly enriched in ~-AP
deposits (i.e., in both plagues and microvascular ~-AP) was used.
A rapid, quantitative light microscopic assay that relies on the insolubility of neurofibrillary tangles and senile plague amyloid cores in sodium dodecylsulfate (SDS) was developed and used to identify the ~-AP enriched brain tissue, as follows. Frozen cerebral cortex (50-lO0 mg) was dissQcted from lO representative brain areas, homogenized briefly in 2% SDS buffer (2~ SDS/O. 1 M
~-mercaptoethanol (~-ME)/50 mM Tris-HCl, pH 7.6), heated to lOO C
for 5 min., and pelleted at 10,000 g for 5 min. The pellets were re-suspended in 50 ~l of 2% SDS buffer, placed on albumin-coated slides, and stained with 1% Congo red in water. Using a polarizing microscope, the number of birefringent cores and/or tangles can be counted or a semiquantitative estimate (e.g., 0-4+, where 0 means no cores and 4+ means a large number of cores) can be made. During amyloid core purification, a modification of this assay was used in which 5 ~l of 0.2% Congo red were mixed with 5 ~l of the fraction and placed in a hemocytometer to count precisely the yield and purity of the cores at each step.
Frozen cerebral cortex (20-lO0 g) was selected from areas found to be ~arkedly enriched in ~-AP-containing amyloid deposits by the assay just described. The tissue was dissected free of visib~e blood vesselg and meningeal fragments, minced with a -' 3 !~ ~
r scalpel and incubated for 2 hours in five volumes of 2% SDS
buffer (see FIG. 1 of Selkoe et al., J. Neurochem. 46:'820-1834 (1986)). The suspenslon was homogenized (Dounce, pestle B, 20 stro~es), heated at lOO C for 10 min., and sieved through 100 ~
nylon ~esh (Nitex). The filtrate was centrifuqed at 300 g for 30 min. The supernatant (containing abundant neurofibrillary tangles among other contaminant3) was frozen for a separate purification of neurofibrillary tangles. The pellet was washed and pelleted (300 q, 10 min.) three times in 0.1~ SDS solution (0.1% SDS/150 mM NaCl/0.2% NaN~). This pellet was homogenized;
sieved through 35 ~m Nitex; loaded on a gradient composed of layers of 1.2 M, 1.4 M, 1.6 M, and 1.8 H sucrose in 1% SDS/50 mM
Tris; and centrifuged in an SW 28 Beckman rotor at 72,000 g for 60 min. Each interface was collected, diluted five-fold in 0.1%
SDS solution, and pelleted at 300 g for 30 min. A small aliquot fro~ each pellet w~s assayed as above to assess core enrichment.
In most AD cases, the 1.6/1.8 and 1.4/1.6 M sucrose interfaces showed the greatest enrlchment. Pellets fro~ these layers were stored at 4-C until further use. The 1.4/1.6 H and the 1.6/1.8 M
sucrose gradient interfaces can be used without further purification ag i nogens to raise polyclonal and monoclonal antibodies, following their extensive sa~ication and their combination with co~plete or incomplete Freund's adjuvant, or can be used as immunogens following their solubilization in concentrated formic zcid followed by lyophilization, resuspension in a buffer and further purification of the ~-AP by HPLC.
Alternatively, the following further purification of the immunogen can be carried out.
The post-gradient core fractions were Dounce-homogenized (pestle B, 10-20 strokes) and sieved through 35 ~m Nitex just prior to loading on a Becton Dickinson FACS 440 cell sorter.
Each fraction was analyzed on the FACS as to forward angle light-scattering (particle size) and fluorescence intensity. Based on -"~133'i6 r the size range Or isolated cores, the instrument was programmed to separate particles of -5-30 ~m diameter from smaller and larger particles and simultaneously to separate the 5-30 ~m partlcles into two subpopulations displaying bright or dull autofluorescence. The samples were sorted in 0.1~ SDS solution at a flow rate of 2,000-5,000 particles per second using an 80 ~m nozzle and an arqon laser with an excitation maximum at 488 nm, and emission picked up by a 580 nm dichroic filter followed by a 580 nm long pass filter.
Following the first FACS sort, the fractions were assayed for cores, and the dull autofluorescent fraction, containing the large majority of sorted cores, was fluorescently immunolabeled.
For this purpose, any of the ~-AP antibodies li:ted in Table I or a similar antibody can be used. Here, for example, partially purified cores (prepared as above through the sucrose gradient step) were used as an immunogen in a rat or a rabbit to raise crude amyloid core antibodies. After an initial injection (-200~g protein in complete Freund's adjuvant) and four boosts (50-lOO~g each in incomplete Freund's adjuvant), the rat serum labeled the SDS-isolated cores at -1:2,000 dilution. The serum wa~ absorbed overnight at 1:1,000 dilution with an homogenate of the SDS-insoluble fraction of normal aged human cerebellum. The post-first-FACS core fraction was incubated (4-C, 18 hr.) with the absorbed antiserum (or with a ~-AP antiserum from Table I) in TBS (150 mM NaCl, 50 mM Tris, pH 7.6). The labeled cores were pelleted, washed twice in T~S, labeled witn rhodamine-conjugated goat anti-rabbit IgG antibodies (1:20, 2 hr.), ar,d washed twice.
The cores were examined by fluorescence microscopy, briefly homogenized, and sorted in the FACS for a second time. Analysis usually indicated two discrete populations of fluorescent, 5-30 ~m particles (or a single peak with a more brightly fluorescent shoulder), and these were separated. The brighter fraction contained highly purified cores, which were heated ~lOO C, 5 2~133 r min.) in 1% SDS to remove IgG, washed three or four times (300 g, 10 min.), and stored at 4 C. Part$cle purity was determined by light and electron microscopy.
An alternative method of preparing isolated amyloid filaments from AD cerebral cortex can be used. This is as described by Selkoe and Abraham in Methods in Enzymoloqv 134: 3~-4-4 (1386). The method yields mixed fractions of neuronal-derived paired helical filaments and extraneuronal amyloid filaments. Since the latter elements can be quite enriched in the fractions produced by this method, it can be used to advantage to obtain ~-AP for use 25 immunogen in the pract,ce o~
this invontion. Indeed, HPLC-purified ~-AP extracted from fractions prepared by this method were used as the immunogen to produce antiserum A ~Table I), which sensitively detects deposits of ~-AP in AD skin and other non-neural tissues. Cerebral cortex (20-100 g) containing abundant vascular and plaque amyloid deposits is d1ssected from frozen coronal brain sections based on the results of the semiquantitative Congo red assay dèscribed previously. MeningQs, white matter, and visible blood vessels are carefully removed and the cortex is fihely minced with a scalpel bl~de. The tissue is incubated for 2 hours in 5 vol. of 2% SDS buffer (2% SDS/0.1 M ~-ME/50 mM Tris-HCL, pH 7.6) and homoaenized in a Dounce glass homogenizer using 10-20 strokes with a loose-fi~ting (B) pestle. The homogenate is heated to lOO C for 5 min. and sieved once through 110 ~m nylon mesh (Nitex). The filtrate is spun at 300 g for 30 min. The E.~llet, which contains the vast majority of senile plaque amyloid cores in the homogenate as well as larqe NFT (-15-20 ~m diameter), some blood vessel fragments, some lipofuccin granules, and other larqer contaminants, is stored at -70 C for a separate purification of amyloid cores. The supernatant is aspirated with a Pasteur pipet and centrifuqed at 100,000 g for 30 min. in a fixèd-angle rotor. The pellets are resuspended in 1% SDS buffer (1% SDS/O-l M ~-ME/50 ~H Tris-HCl, pH 7.6) at a ratio of approximately 1 ml/l.S-2.0 g starting cortex and re-homoqenized with a tight-fitting (A) Dounce pestle.
The resultir.g suspension is made 0.4 M in sucrcse and layered on top of a discontinuous sucrose gradient in 38 ml Beckman ultraclear ultr&centr'ruge tubes: S ml of 2.o M sucrose, 8 ml of 1.4 M sUCroSQ, 8 ml of 1.2 H sucrose, and 7 ml of 1.0 M
sucrose (all in 1% SDS/50 mH Tris-HCl buffer). The gradients are immediately spun at 145,000 g~l (27,000 rpm) for 240 min. in an SW28 rotor. Relatively well-defined bands are observed as follows: brown bands at the top of the 1.0 M sucrose layer and at the 1.0/1.2 M interface, hazy white bands at the 1.2/1.4 and 1.4/2.0 M interfaces, and a very small pellet. Each fraction is collected with a Pasteur pipet, diluted at least S times with 1%
SDS buffer, and pelleted at 100,000 g for 60 min. in a fixed angle rotor.
To purify such sucrose gradient fractions further, the pelle~s are rc-suspended in ml of 1% SDS huffer, sonicated for 30 sec. (Rontes sonicator; power setting 8; tune 4~, and spun at 14,00Q g for 30 min. (Beckman Type 40 rotor). The peliet, containing larger aggregates of lipofuscin plus PH}~ fragments, is discarded; the supernatant is centrifuged at 176,000 g for 60 min. Electron microscopy o~ this final pellet reveals relatively homogeneous fields of randomly oriented small fragments of PHF
and amyloid filaments. Small fragments of lipofuscin and fine, dense granules remain as contaminants, but lipofu_cin contamination is less than in the unsonicated sucrose gradient fraction. The sonicated PHF an~ amyloid fiber fragments retain their green birefringence when stained with Congo red after hign-speed pellet.ng.

B. Extraction of ~-Am~loid Protein from the SXin or Other Non-~eural Tissue The immuncgen used to produce the antibodies to ~-AP that recognize the sKin deposits in AD patients can also be prepared 20133~6 r from the skin itself. Frozen or fixed sections of skin samples are first examined immunohistocnemically to determine the extent of ~-AP deposition using antibodies such as those in Table I.
Fresh or fresh-frozen samplQs of B-AP-rich ~kin collected at autopsy from patie~ts who died with pathologically confirmed AD
can be extensively minced and then homogenized in a physiologic buffer (e.g., phosphate buffered normal saline, pH 7.4).
Following differential centrifugation, the supernatant and the insoluble material Or skin are separated. The supernatant is assayed for the presence of ~-AP using, for example, an enzyme -linked immunosorbent a~ay (ELISA) or radioimmunoassay (RIA) incorporating antibodies such as those in Table I. If the supernatant is found to contain ~-AP immunoreactivity, it can be further ~ractionated by both conventional chromatographic procedures and affinity binding methods to obtain an enriched fraction of the ~-AP molecule from s~in. If the insoluble pellet of the first centrifugation demonstrates ~-AP reactivity by the immunoassay, the pellet can be differentially extracted in certain detergents and/or salts (for example, 1-2% triton X-100, 1-2% sodium-dodecylsul~ate, 6 ~ guanidine hydrochloride, or 6 M
urea). Each extract of the insoluble pellet is then tested for ~-AP immunoreact~vity by immunoassay. Once solubilized ~-AP is detected in a particular extract, this extract can ~e further purified chromatographic~lly (e.g., by HPLC) and/or by affinity binding methods. Fractions enriched in ~-AP derived in ~his way can be injected into rabbits for production o~ polyclonal antibodies or into mice ~or production of monoclonal antibodies by hybridoma techniques.
C. Production o~ Pol~clonal Antibodies To prepare the ~-AP immunogen for production of the preferred polyclonal antibodies for this invention, the partially purified or the purified amyloid plaque cores or the paired helical filament/amyloid filament-enriched fractions, each - 2~13~6 i, ~ -prepared from AD cerebeum as described above, were then treated with 6.8 M guanidine SCN followed by dialysis and lyophilization or ~ith 88% foroic acid followed by lyophilization. The - lyophilates were dissolved in 1% SDS, incubated at 37-C for 30 min. and centrifuged at 10,000 g. The supernatant was applied to a Spectra-Physlcs 8700 HPLC system consisting of two TSK G3000 SW
- columns (Bio-Rad) connected in series and equilibrated with 0.1%
SDS/150 mM Na phosphate, pH 6.8, and run at 0.5 ml/min. Proteins were detected at 220 nm and peaks were collected with a Gilford 201 fraction collector.
Both types of ~-amyloid fiber preparations (purified amyloid cores or isolated PHF/amyloid filament preparations) - yield similar HPLC chromatograms usually showing two mzjor protein peaks having approximate molecular weights of 3-7 kDa and 11-15 kDa (see Figure 5 of Selkoe et al., Journal of ~eurochemistrY 46: 1820-1834 (1986), incorporated fully by reference herein). Either or both HPLC fractions (or other ~-AP
contair.ing chromatographic fractions) can be used as immunogens in either rabbits (to produce polyclonal antibodies) or in mice (to produce monoclonal antibodies by conventional hybridoma techni~tes). For example, 250 ~g of protein were used, from the 11-15 kDa HPLC fraction of either formic acid-solubilized -myloid cores or formic acid solubilized PHF/amyloid filament fractions, to immunize rabbits. The antisera tha~ resulted are represented by antisera A and C in Table I.
An alternate method of preparing the immunogen for antibody production involves the production of synthetic peptides having partial or complete sequences of the ~-AP, their coupling to various carriers (e.g., keyhole limpet hemocyanin, rabbit serum albumin, edestin, etc.) and then immunization into rabbits by conventional procedures in complete or incomplete Freund's ad~uvant. Antisera that resulted from this procedure are exemplified by Antisera Y and L in Table I.

2~,' ~

r D. Prod~c'ion of Monoclonal Antibodies ~he injectio. schedule used to generate monoclonal antlbodies is ~s follows. Approximately 10 ~g/ml of the antigen is emulslf~ed in complete Preunds Ad~uvant ar.d injected intraperitoneally. On day 15 ~nd day 28 post initial injection, approximately 5 ~g of antigen is emulsified in incomplete Freunds Ad~uvant and injected intraperitoneally. The mouse is rested for 21 days and then injected intraperitoneally with approximately 10 ~g of antigen. The mouse spleen is fused three days later.
The fusion protocol uses the following materials:
Dulbecco's Hodified Eagle's Medium (DMEM) with high glucose;
Polyethylene Glycol 1500, screened for fusions, obtained from Boehringer Mannheim, West Gel~any: Fetal Bovine Serum (FBS), obtained from JR Scientific, Woodland, California; 200 mM tissue culturs grade Glutamine; 10 2M tissue culture grade hypoxanthine;
200 ~g/ml of tissue culture grade Azaserine; 0.93 percent ammoniu~ chloride; 3% Dextran (the high molecular weight fraction); lM tissue culture grade HEPES Buffer, pH 7.2; a 35 ml storlle pQtri dish; sterile forceps and dissecting scissors;
fifteen sterile 96-well flat-bottom tissue culture plates with lids; a twelve channel pipet and sterile tips therefor; two sterile microscope slides with frosted ends; SP2/0 cells obtained from American Type Culture Collection (Roc~ville, Maryland).
~n the day of the fusion, two different media are prepared.
The first is a growth medium containing 400 ml of DMEM, 100 ml of FBS, 5 ml of 200mM glutamine, 7.5 ml of lM HEPES buffer, pH 7.2, and 5 ml of 10ZM hypoxanthine. The other medium-that is prepared is the selection media, comprising 75 ml of growth media to which is added 750 ~1 of 200 ~g/ml azaserine.
The mouse is killed and then immersed in 70% ethanol for 10 seconds. The mouse is placed in a flow hood, and the spleen is aseptically removed, and placed in a 3smm petri dish containing 5ml of growth media. The spleen is dissociated between the 2a -2 ~ 3 ~3 fi r,frosted ends of the slide. The spleen cell suspension is placed ln a 15 ml 5terile tube. The petri dish is rinsed With lO ml of qrowth media and placed in the same 15 ml tube.
,The SP2/0 cells are counted. ~The density is generally greater than about 0.5 million cells per ml and less than about o.g million per ml.) The viability of the cells is generally greater than about 95% as estimated by trypan blue exclusion.
Approximately fifty million cells are then pla~ed into sterile 50 ml tubes. Both the spleen cells and the SP2/0 cells are then centrifuged in a clinical centrifuge for lO minutes at 500 x g.
~ he supernatant is removed from the spleen cells, and they are resuspended in 8 ml of ammonium chloride at about 4 C. This is allowed to incubate on ice for 5 minutes.
The SP2/0 cells are resuspended in 15 ml of plain DMEM.
When the five minutes incubation of the spleen cells is complete, the suRpended spleen cellR are removed from debris pellet and added to the SP2/0 cells. DMEM is added to this cell mixture until the total volume is 25 ml. 25 ml of the 3% dextran is added to the spleen - SP2/0 cell mixture and this is incubated for 5 minutes. The cells are centrifuged at 500 x g for 10 minutes.
The supernatant is removed from the pellet, and the pellet is then loosened by vortexing. 1 ml of the PEG 1500 is added.
The tube i~ then vortexed and slowly rotated for one minute.
This step is the actual fusion step.
25 ml of DMEM is added to the tube and this is incubated for one minute. After this incubation, 25 ml of growth media is added and incubated for one minute. The cells are centrifuged at 500 x g.
The pellet is resuspended in 7S ml of selection media.
12 channel pipet is used to place 50 ~l into each well of the fifteen 96-well plates.

r ~ O 1 3 ~ 5 t , r ~ Two days after the fusion, each well is fed 50 ~1 of grolth media. On day fiv~ post fusion, 50 ~1 arQ removed and the fusion is fed l00 ~1 of growth media. Visible hybrids are observed between day seven post-fusion and day ten post-fusion. At tnl-tlme, the fusion i8 fed 100 ~1 of growth media. Screening generally begins around day ten. If hybrids are not ready to be screened on day ten, 150 ~1 o~ media are removed and 150 ~l of - fresh growth media are added thereto.
Viable hybridoma supernatants are screened for ~-AP skin monoclonal antibodies by immunohlstochemistry as described above.

EY~le II
Observaticns in Patholoqic~llY Confirmed AD Victims In an initial experiment, non-keratinized perianal skin (including cutaneous and s-lbcutaneous tissue) was taken at autopsy from a 78-year-old man with a seven year history of clinically typical AD and severe AD brain les$ons, including marked ~-AP-positive amyloid angiopathy in cortical and meningeal arterioles. Reaction of formalin-fixed sections with an antiserum (antibody Y in Table I) raised to a synthetic peptide comprising residues 1-38 of ~-AP (residues 597-634 of the ~-AP
precursor described by ~ang et al., Na~ure 325:733-736 (1987)) showed specific, peptide-inhibitable labeling of amorphous material deposited multifocally in the connective tissue and periarteriolar spaces of the dermis. Ad~acent sections from this block and skin sa~ples from several other pathologically confiroed AD patients (Table II) were then examined with certain well-characterized antibodies to purified native B-Ap (A,C, and/or F in Table I), synthetic ~-AP peptide (Y in ~ble I) or unrelated proteins (Table I). The ~-AP antibo~.e, te~t d showed specific immunolabeling o~ amorp.~ous proteinaceous deposits in the dermis and subcutaneous tissue, of AD patients (rable II).
The sensitivity of detection differed consideraDly among the ~ ) 2~133~r~

anti~era. In seven of ten confirmed AD patients, clear ~-AP
i~munoreactivity was ob~erved in the s~in. Two AD patients showed equivocal staining and, in one, staining was absent.
To determine the spQcificity of the immunoreactions, antiqen-absorbed and ~control-absorbed" aliquot~ o~ each ~-AP
antiserum tested were reacted simultaneou~ly with AD brain, AD
skin and dot blots of the antigen. In th- ca~e Or antiserum A, for example, partially purified, SDS-extracted senile plaque cores were used as the absorbent; a control absorption employed a fraction prepared identically from normal cortex which contained the contaminating particles of core fraction~ ~lipofuscin granules, collagen and microvessel fragments). In each such experiment, the plaque core absorption markedly diminished or abolished the staining of plaque and vascular amyloid in AD
brain, the dermal and subdermal amorphous deposits, and the antigen on dot blots; the~e reactions were all carried out simultaneously wlth the same batch of absorbed antiserum.
Control absorptions produced no change. Similar results were obtained with the other ~-AP antisera. Preim~une sera, secondary antibody alone and hyperlmmune sera to purlfied PHF (P), ~au protein ~DJ), glial fibrillary acid protein (G), transthyretin (prealbumin) (T) or amylold AA protein (as well as a mouse monoclonal antibody to amyloid AA protein) all failed to label the skin deposits.
The ~-AP-immunoreactive deposits detected in the dermis appeared in a patchy, multifocal distribution beneath the epidermal/der~al ~unction, ~ometimes near small blood vessels or glandular elements such as sweat and sebaceous glands. The stained material appeared to be localized extracellularly among connective tissue fibers in the reticular dermis. It was found amongst the normal collagen and elastin fibers of the dermis.

Some deposits were found circumferentially around small blood vessels. Such deposits involved a minority of microvessels; most 2 3 ~ 3 !j fj r r vessels and other skin structures (sweat and sebaceous glands;
hair follicles; hair shafts; smooth mu5cle, epidermis) were entirely unstained, attesting further to the specificity of the iDunoreaction. In subdermal tissues, amorphous ~-AP-reactive material was occasionaily found around microvessels, in connectiv~ tissues or around bundles of muscle fibers, apparently extracellular. Almost all of the deposit~ that were observed had an amorphous, non-fibrillar appearance resembling somewhat the diffuse, preamyloid deposits of ~-AP present in AD brains, in aged normal human brain~, and in the brains of some patients with hereditary cerebrovascular ~-amyloidosis of Dutch type (Van Duinen et al, Proc. Natl Acad sci, USA 84:5991-5994 (1987)).
However, the deposits lost their reactivity with antibodies to - synthetic ~-AP (e.g., Y in Table I) if the sections were pretreated with formic acid, a reagent th~t generally enhances the immunostaining of diffuse ~-AP deposits in the brain by synthetic ~-AP antibodies. The immunoreactive ~-AP deposits in skin were not detectQd by (or relcted very weakly with) the cla~sical a~yloid staln~, Congo red and thioflavin S. Recently, Yamada, et al. reported that skin of AD patients does not show staining with Congo Red. _ç~ Neuropathologica 77:136-141 (1988).

Example III
Observation in Living Patient A 3 mm-diameter punch biopsy of skin from the forearm of an 85-year-old man with a ix-year history of clinically typical AD
was briefly fixed (30 minutes) in 10% neutral buffered formalin and exa~ined immunohistochemically.
The skin sa~ple was obtained with a punch biopsy bore from the inside of the forearm, under sterile technique following a small injection of local anaesthetic (1% lidocaine with epinephrine). Punch biopsies ranging in size from 3mm to 7mm 2~3396 i ~

r ~ diameter are generally suitable for this purpose. The biopsied sample included approximately 80% cutaneous and 20% subcutaneous tissue. The skin sample was fixed for about 30 minutes in 10~
neutral bu~fered formalin. The sample was then removed from the formalin and placed in a physiological bu~er (sodium phosphate buf~ered normal saline, pH 7.6, with 0.02~ sodium azide) and stored at 4 C. The skin sample was then embedded in paraffin and - sectioned in a microtome. Sections of approximately 5 ~o approximately 15 micrometers in diameter were mounted on glass microscope slides using a conventional histological adhering reagent.
After reaction with antiserum A (Table I) and a secondary goat anti-rabbit IgG antibodyjavid-n-biotin peroxidase detection system, this skin sample from a living patient showed readily detectable, specific ~-AP immunostaining in the reticular dermis, when viewed through a briqhtfield light microscope. Using the panel Or antibodies to native or synthetic ~-AP (set forth in Table I), antibodies A, C and F produced moderate or stronq staining of amorphous, apparently extracellular proteinaceous deposits having the hi~tological characteristics described in Example II above for autopsied AD skin sections. Antibody Y
produced mild to moderate staining. Abolition of the dermal staining by antibodies A, C and F occurred when these antisera were absorbed with a partially purified fraction of AD amyloid cores, whereas no change in the staining occurred when the absorption was carried out with a control brain function lacking amyloid cores. A similar abolition was obtained when antibody Y
was absorbed with synthetic ~-AP. Pretreatment (5-10 minutesj of the skin sections with concentrated (about 88%) formic acid abolished the ~-AP immunoreaction. In order to insure that the ~-AP deposits in the skin vere being adequately immunoreacted, the sections were examined throuqh the microscope to check the extent of the colorimetric reaction at frequent intervals during L

~ ~ S ~) ?, ~

r the final development step that uses DA8: as positive controls, simultaneously-stained sections of autopsied AD skin, previously proven to contain imnunoreactive B-Ap deposits, and autopsied AD
brain were used.
As noted in Table I, the degree of ~-AP detection was generally strongest when the skin sample was treated with antiserum A produced to ~-AP antigen from AD cortex, and weakest v when the sample was treated with antiserum L, produced to antigen from a synthetic Bl2~ peptide. However, the latter synthetic fragment of the ~-AP ~comprising amino acids 1-28) or smaller ~-AP fragment~ of 8 or more amino acids, can be used as immunogens to produce peptide antlbodies that can be used to detect skin ~-AP deposits in AD patlents. Various discrete regions of ~-AP (or a ~-amyloid precursor protein fragment comprising ~-AP) can serve as userul and specific antigens in the production of antibodies for immuno-detection of ~-AP in skin.
A series Or irrelevant hyperim~une sera to various antigens (listed in Table I and Example II) other than ~-AP, as well as secondary antibody (goat anti-rabbit IgG) alone, produced no positive, speclfic staining of the B-AP-reactive deposits in the biopsied skin sample.

Example IV
Examination of Other ~on-~eural Tissues Certain other non-neural tissues of AD patients were eYA~i~ed i~munocytochemlcally using one or more of the ~-AP
antibodies in Table I, as well as the unrelated control antisera.
Deposits of ~-AP-reactlve material were detected in the submucosal connective tissue of the intestine in two neu_opathologically confirmed AD patients, particularly in and around the walls of small blood vessels. Such vascular deposits were also found in the ~mall intestine o~ an aged (77-year-old) patient without overt dementia, but whose brain showed many , -2~133"~
;~, , . ~., senile plaques characteristic of AD. Some of the immunoreactive material appeared to be localized extracellularly among connectiv~ tissue fibers. The vascular staining in the intestinal submucosa resembled the vaccular and perivascular staining co~monly ~een in AD brain and meninges. This intestinal staining uas markedly decreaned or eliminated by absorption of the ~-AP antibodies with AD brain amyloid cores, whereas absorption with an i~olated fraction of control brain particles did not alter the reaction.

mmunochemical Exarination Procedure for the Presence of ~-AP DeDosits in Intestine sections of colon or small intestine obtained from patients who died with Alzheimer's Disease, with other neurological disorders, or of advanced age, were removed at autopsy and fixed in conventional fixatives (e.g., 10% neutral-buffered formalin).
Following embedding of the sections in para4fin (or an alternative embedding m-diu~), microscopic samples (i.e., sections) thereof were prepared using a microtome and placed on glass microscope slide~. Following conventional deparaffinization and hydration of these samples by standard methods well-known in the art, the samples were contacted (reacted) with an antibody to either native or synthetic amyloid protein, as exemplified in Table I. For example, antibody A was used at 1:250 dilution in Tris-buffered or phosphate-buffered normal saline solution. This diluted primary antibody, or a control aliquot following absorption of the primary antibody with a form of the ~-AP immunogen, was placed on the colon or intestine scction and incubated for about 2-4 hours or overnight. Thereaft~r, the sections were washed with buffer and incubated with a secondary antibody reactive with the prima~v antibody, e.g., goat anti-rabbit IgG antibody. The secondary antibody was obtained co~mercially an~ had been coupled directly to biotin. This reaction was followed by washing the sections and then exposing them to a co~mercially obtained avidin-biotin 2~13~

complex coupled with peroxida5e (Vector Laboratories, aurlingame, CA). Following this reaction, the reaction product was Aetected colorimetrically by incubation with 3, 3'-diamino-benz dine and hydrogen peroxide. The sample was viewed through a briqhtfield light microscope during the colorimetric reaction in order to determine the intensity of immunostaining, and therefore the proper duration, of this reaction. Thereafter, the sample was washed and dried, and a glass coverslip was mounted over the sample.
For analysis of immunoreactive ~-AP in such immunostained intestinal sections, a slide was viewed in a standard light microscope using brightfield optics. Sections stained with the primary antibody were compared to an immediately adjacent section that had been stained with the primary antibody following its absorption with a ~-AP immunogen. (For immunogens, see second column of Table I.) B-AP deposits in the colon were o'oserved in the walls of small blood vessels znd surrounding connective tissue in the submucosa. The brown immunoreaction product often resembled a serpentine cuff in the outer wall of the small blood vesRel (e.g., arterioles). In the intestinal sections taken from two patients who died with autopsy-proven Alzheimer's disease, such vascular and perivascular submucosal colonic deposits were seen in both cases examined (see Table II). In intestinal sections taken from two patients dying with other disorders, only patients of advanced age (e.g., 77 years old or older) have thus far showed some intestinal submucosal blood vessels containina ~-AP-immunoreactive deposits. Intestinal samples from patients under age 63 showed no perivascular ~-AP deposits in these experiments.
In further immunohistochemical experimer.ts on colonic and small intestinal tissues from AD and non-AD humans, each immunostained section was compared to a known positive colonic sample tha~ had previously been identified as described above.

r Each new section was also compared to a known negative colonic sample so previously identified.
The location o~ the ~-AP immuno reactive material in only selected vessels in this submucosal connective tissue of colon and small intestine made it easy to identi~y when viewed by light microscopy. In the ~-AP-positive sections from AD subjects and certAln aged nor~al patients, numerou~ intestinal blood vessels were devoid of ~-AP reactivity whereas some contained peroxidase-positlve (brown) immunoreactive deposits. Some immunoreactive ~-AP deposits were al~o noted in the connective tissue surrounding blood vessels and in the deeper portion of mucosa.
In addition to the experiments just described, further experlments on sectlon~ of AD or non-AD lntestine were performed using antibody Y ~see Table I) at a dilution of 1:500, and antlbodles C and F at dllutionc of 1:250. Otherwise, the method was the same, and the same positive and negative controls were used in these experlment~.
The rPsults Or th- ~amples o~ lntestlne from AD and non-AD
patl~nts examlned ln the experiments described herein are exempllfled ln Table rr .
ExamDle V
Exarlnatlon of Controls Slnce some deposltlon of cerebrovascular and cerebral ~-AP
occur~ commonly durlng normal aging, non-neural ti~ue samples were compared from 41 human subjects: eleven with ~D, twenty-six without AD, and four wlth Trisomy 21 (Down's Syndrome) (Table II). All but four had been autopsied, establishing precise neurologlcal dlagnose~ based on detailed neuropathological examinations. Eight of the eleven AD cases showed B-AP-reactive dermal deposits upon lmmunohistochemical staining. In contrast, none Or rourteen non-demented subject~ less than 63 years old showed ~AP-immunoreactlve material in skin. Of seven aged, neurologically diseased patients who did noe have AD, four were 2C~ G

r negative and three were positive. Of five aged t2 54 year-old) normal sub~ects, three vere negative, whereas tvo showed faintly staining, equivocal ~AP-reactive material in skin. The deposits in the aged diseased and aged normal sUb~eCts had similar histological characteristics to those observed in the AD
subject~, as described in Example II.

ExamDle VI
Diagnostic Xit The data in Table II demonstrate the diagnostic utility of this invention for AD. A kit for diagnosing Alzheimer's Disease can be prepared by separately packaging, for example, the following reagents (typical quantities and concentrations are listed or a 20-assay kit):
Reaqent OuantitY Concentration Source Formalin 5 mls 10% ------AntiB-AP 5 ml~ 10 mg/ml murine antibody monoclonal IgG

Anti-mouse IgG 5 ml~ 50 ~g/ml goat (with biotin label~
Avidin HRP 5 mls o.5 ~g/ml ------Phosphate 25 mls ------ ------buffered saline The amounts and concentrations cf reagents described herein can vary considerably depending on the specific application, as will be apparent to one of ordinary skill in the art.

~ u ~ 3 3 ~3 f;
i,,~, r~

-Exa~rle vII

Quantitatlon of ~-a~Yloid Reactivity in Skin ~ or Other Non-Neural Tissues ~~ A quantitative competitiv~ immunoassay for ~-AP reac~ivity - can comprlse the ~ollowing: 1) a solid pha~e consisting o~ bound ~-AP-containing ~aterial (e.g., native, purified ~-AP or synthetic ~-AP or ~-AP filaments isolated from tissue) 2) a labeled anti-~-AP antibody and 3) a series of ~-AP protein standards for prcJuctlon of a standard curve. Pre~rably, these - components would be packaged in separate con~ainers. The "label"
on the antibody may be, for example, an enzyme (horseradish peroxide or alkaline phosphatase), fluorophore cr radioisotope such as those typically used in the art.
A pre-determined amount of anti-~-AP antibody is pre-~ incubated with various concentrations of ~-AP protein followed by incubation of the supernatant with the ~-AP bcund to the solid pha~e. Following detection of the bound label, a competitive standard curve will bQ generated and used to quantify the concentration of solubilized ~-AP (or a fragment of ~-AP
conprising ~-~AP) in unknown skin (or other non-neural tissue) extracts.
This quantitative immunoassay may also be carried out using an extract or fine suspension of the skin sample without the ~-AP
actually being in ~olution (i.e., with the ~-AP still in an insoluble form~.

Claims (30)

I CLAIM:

1. A method for diagnosing Alzheimer's Disease, comprising the steps of:
obtaining a non-neural tissue biopsy sample;
contacting at least a portion of the sample with at least one antibody capable of identifying .beta.-amyloid protein, a .beta.-amyloid precursor protein fragment comprising .beta.-amyloid protein, or a .beta.-amyloid peptide fragment of about 8 or more amino acids;
and monitoring the extent of reaction between the contacted sample and the antibody.

2. The method of Claim 1 wherein the non-neural tissue biopsy sample is selected from skin and intestine and the antibody is to .beta.-amyloid protein or to a .beta.-amyloid precursor protein fragment comprising .beta.-amyloid protein.

3. The method Or Claim 2, wherein the antibody is to .beta.-amyloid protein or to a .beta.-amyloid precursor protein fragment comprising .beta.-amyloid protein isolated from an Alzheimer Disease or Down's Syndrome subject.

4. The method of Claim 3, wherein the antibody is to .beta.-amyloid protein or to a .beta.-amyloid precursor protein fragment comprising .beta.-amyloid protein isolated from the cerebral cortex of said subject.

5. The method of Claim 3, wherein the antibody is to .beta.-amyloid protein or to a .beta.-amyloid precursor protein fragment comprising .beta.-amyloid protein isolated from the skin or other non-neural tissue of said subject.

6. The method of Claim 2, wherein the contacting and monitoring steps are carried out by extracting .beta.-amyloid protein from the sample and conducting a standard assay to determine the quantity of .beta.-amyloid protein therein.

7. The method of Claim 2, wherein the contacting and monitoring steps are carried out by immunohistochemically reacting the skin sample and the antibody and then detecting the reaction in the sample.

8. The method of Claim 7 further comprising carrying out a control test comprising the additional steps of: pre-absorbing the antibody with an antigen comprising .beta.-amyloid protein;
contacting a portion of the skin sample not previously contacted with antibody, or contacting a second skin sample from the same subject with the pre-absorbed antibody; and measuring the extent of reaction between the portion of the skin sample or the second skin sample and the pre-absorbed antibody.

9. The method of Claim 3 where the antibody is polyclonal.

10. The method of Claim 3 where the antibody is monoclonal.

11. A method for diagnosing Alzheimer's Disease, comprising the steps of:
obtaining a non-neural tissue biopsy sample;
performing an extraction suitable for extracting .beta.-amyloid protein, or a .beta.-amyloid precursor protein fragment comprising .beta.-amyloid protein, from the biopsy sample, and thereby obtaining an extract sample;
optionally solubilizing the .beta.-amyloid protein or the amyloid precursor protein fragment comprising .beta.-amyloid protein present in the extract sample, and thereby obtaining a solution sample;
contacting the extract sample or the solution sample with a n antibody to .beta.-amyloid protein or to a precursor protein fragment comprising .beta.-amyloid protein or to a peptide fragment of .beta.-AP of eight or more amino acids; and monitoring the extent of reaction in the extract or solution sample between the antibody and the .beta.-amyloid protein or the .beta.-amyloid precursor protein fragment comprising .beta.-amyloid protein.

12. The method of Claim 11, wherein the non-neural tissue is selected from skin and intestine and the antibody is to .beta.-amyloid protein or to a .beta.-amyloid precursor protein fragment comprising .beta.-amyloid protein isolated from an Alzheimer Disease or Down's Syndrome subject.

13. The method of Claim 12, wherein the antibody is to protein isolated from the cerebral cortex of said subject.

14. The method of Claim 12, wherein the antibody is to protein isolated from the skin or other non-tissue of said subject.

15. The method of Claim 12 where the antibody is polyclonal.

16. The method of Claim 12 where the antibody is monoclonal.

17. The method of Claim 13 where the antibody is polyclonal.

18. The method of Claim 13 where the antibody is monoclonal.

19. The method of Claim 14 where the antibody is polyclonal.

20. The method of Claim 14 where the antibody is monoclonal.

21. A kit for diagnosing Alzheimer's Disease, comprising:
an antibody capable of identifying .beta.-amyloid protein, or a .beta.-amyloid precursor protein fragment comprising .beta.-amyloid protein, or a .beta.-amyloid protein fragment of at least about eight amino acids; and means for detecting the extent of the reaction of the antibody with a non-neural tissue sample.

22. The kit of Claim 21, wherein the antibody is to .beta.-amyloid protein isolated from an Alzheimer Disease subject.

23. The kit of Claim 22, wherein the antibody is to .beta.-amyloid protein isolated from the cerebral cortex of said subject.

24. The kit or Claim 22, wherein the antibody is to .beta.-amyloid protein or to a .beta.-amyloid precursor protein fragme~
comprising .beta.-amyloid protein, isolated from the skin or non-neural tissue of said subject.

25. The kit of Claim 23 further comprising .beta.-amyloid protein, an amyloid precursor protein fragment comprising .beta.-amyloid protein, or a .beta.-amyloid protein fragment of at least about 8 amino acids.

26. The kit of Claim 24 further comprising .beta.-amyloid protein or an amyloid precursor protein fragment comprising .beta.-amyloid protein, or a .beta.-amyloid protein fragment of at least 8 amino acids.

27. The kit of claim 25 where the kit is designed for use in a competitive immunoassay, and further comprises a solid phase, and the protein, protein fragment or peptide fragment is bound to the solid phase.

28. The kit of claim 26 where the kit is designed for use .
in a competitive immunoassay, and further comprises a solid phase, and the protein, protein fragment or peptide fragment is bound to the solid phase.

29. The kit of claim 25 wherein the protein, protein fragment, or peptide fragment is provided as a control in a known quantity.

30. The kit of claim 26 wherein the protein, protein fragment, or peptide fragment is provided as a control in a known quantity.