Background of the Invention This invention relates to an in vitro method for th detection of cellular pathology, and more specifically to a assay for monitoring cytoskeleton breakdown products t determine cell necrosis.

This invention was made with government support unde AFOSR Contract No. 86-0099 (P.I.: Lynch), NIH Grants Nos. NS-18427 and NIA Grant No. AG00538. The government ha certain rights in this invention.

The structural integrity of cells is maintained in par by the cytoskeleton, a mesh-like structure compose primarily of proteins, which lies adjacent to the inner cel surface. The cytoskeleton of many cell types (a partia list includes neurons, lymphocytes, kidney, liver, cardia and smooth muscle, and blood platelets) contain larg amounts of a protein either identical to or closely relate to brain spectrin (also known as fodrin) . Spectrin bind F-actin, and together they are generally associated with th inner face of the cell membrane, where they form filamentous meshwork.

Brain and many other tissues have been known for som time to express calcium-stimulated proteolytic activity Studies of degradation in peripheral nerves have indicate that a calcium activated neutral protease, calpain, i critically involved in the degradation of neurofilamen proteins following denervation or injury. Two forms o this protease have been identified in brain and othe tissues. The two forms are differentiated by thei threshold for activation by calcium: calpain I require micromolar calcium while calpain II is activated b concentrations of calcium between 0.1 and 0.5 mM. The tw forms are differentially distributed in the brain. Whil calpain II is mainly localized in the cytoplasmic fractio

of brain cells, the highest activity of calpain I is foun in small processes. While the two forms of calpain diffe in these and other ways, the term "Calpain" shall be use herein to refer to calcium activated neutral protease generally, including both forms of calpain.

A variety of cellular insults (e.g., toxins, anoxia etc.) and disease states (e.g., Alzheimer's, Parkinson's HIV-induced neuropathy, muscular dystrophy) cause th degeneration and death of cells. Often, however, it is no possible to determine that injury has occurred unti degenerative effects are irreversible. There thus exists need for reliable methods to detect degenerative events a soon as possible, preferably before the onset o pathological symptoms. Preferably such methods also hav high sensitivity, wide ranging applicability and ease o administration.

Summary of the Invention Briefly, the present invention provides a method o detecting cellular death or degradation in a subject, suc as a mammal, comprising analyzing a biological sample fro the subject for the presence of spectrin breakdown product and comparing the quantity of spectrin breakdown products t the quantity of spectrin breakdown products in a norma subject, wherein an increased level of spectrin breakdow products indicates cellular death or degradation in th subject. In many cases, the quantity of spectrin breakdow products in the normal subject is substantiall undetectable. The biological sample can be any sample fro an organism, especially cerebrospinal fluid or a componen of blood. The cellular death or degradation detected can for example, be due to a non-pathological cellular insult such as a trauma, ischemia, lesions, or exposure to toxins or may be due to a pathology, including those of th nervous system, such as Alzheimer's disease, Parkinson' disease and muscular dystrophy. Biological samples for th detection of cellular death or degradation in the nervou system can include neural tissue or cerebrospinal fluid.

The step of analyzing the biological sample for the presence of spectrin breakdown products can comprise, for example, contacting a spectrin breakdown product in the sample with a detectably labeled antibody, and can also include the steps of exposing the sample to an electrical gradient so as to separate the components in such a way that spectrin breakdown products are separated from spectrin, contacting the separated components with a detectably labeled antibody that binds to a spectrin breakdown product, and determining the presence of antibody binding, wherein the presence of the antibody binding indicates the presence of spectrin breakdown products. The step of analyzing the biological sample can also comprise the step of staining a separated product with a stain which visualizes the product, and determining the presence of stain binding, wherein the presence of the stain indicates the presence of spectrin breakdown products.

In another embodiment of the invention, a method of detecting cellular death or pathology in a sample from a subject, such as a mammal, is provided comprising obtaining a biological sample from the subject, analyzing the biological sample for the presence of spectrin breakdown products, determining a basal level of spectrin breakdown products, and comparing the determined level to the basal level, wherein a higher level than the basal level indicates cellular death or degrada-.ion. The basal level of this method is usually the basal level of spectrin breakdown products in a normal subject, and in many instances can be assumed to be zero. The sample can be any biological sample from the subject, including cerebrospinal fluid, a tissue sample, or blood or any component of blood.

In still another embodiment, there is provided a method of detecting cellular death or degradation in a subject, comprising obtaining a biological sample from the subject; determining the total amount of spectrin, including intact spectrin and spectrin breakdown products, in the sample; determining a basal quantity of the total

amount of spectrin; and comparing the determined quantit of the total amount of spectrin to the basal quantity o the total amount of spectrin; wherein a quantity of th total amount of spectrin in the sample greater than th basal quantity indicates cellular death or degradation The total amount of spectrin can be measured as tota spectrin immunoreactivity through such means as an ELIS assay or a Western Blot assay. The sample can be an biological sample from the subject, including cerebrospina fluid, a tissue sample, or blood or any component of blood.

Another aspect of the present invention provides method of enriching a biological sample for spectri breakdown products relative to intact spectrin in biological sample, comprising precipitating intact spectri in said sample, leaving spectrin breakdown products i solution in said sample by altering conditions whic differentially affect the solubility of the intact spectri molecule and the spectrin breakdown products; an collecting the precipitated or soluble spectrin breakdow products. The step of precipitating intact spectrin o spectrin breakdown products can comprise altering the pH o ionic strength of the biological sample solution.

Further objects, features and other advantages of th present invention will become apparent from the ensuin detailed description, considered together with the appende figures.

Brief Description of the Figures Figure 1 A. Spectrin immunoreactivity in blotte samples of the contralateral (left lane) and ipsilatera (right lane) dentate gyrus two days after a unilatera lesion of the entorhinal cortex. Arrows indicate the alph and beta spectrin subunits with apparent M _r s of about 24 and 230 kilodaltons ("kD") , respectively, and tw additional immunoreactive peptides (BDPl and BDP2) wit apparent M _r s of about 155 and 150 kD, respectively.

B. Purified brain spectrin incubated unde the following conditions: Lane 1: no additions; Lane 2

l M CaCl2, 1.8 μg/ml calpain I, 10 minutes; Lane 3: ImM CaCl2, 3 μg/ml calpain I, 30 minutes; Lane 4: 10μg of dentate gyrus protein homogenate two days post-lesion; Lane 5: ImM CaCl2, 13 μg/ml calpain, 1.25 μg/ml cal odulin, 3 minutes; Lane 6: ImM CaCl ₂ , 13 μg/ml calpain, 1.25 μg/ml calmodulin, 30 minutes.

Figure 2. BDPl (filled circles) and BDP2 (open circles) levels are expressed as a percentage of the total spectrin immunoreactivity, as determined by scanning reflective densitometry.

Figure 3. Levels of BDP's in regions of the brain of control and Brindled mice, showing the effect of treatment with copper.

Figure 4. Levels of BDP's in the dentate gyrus and the CAl region of the hippocampus of rats receiving trimethyltin.

Figure 5. Levels of BDP's in the dentate gyrus and the CAl region of the hippocampus from a gerbil following ischemia. Figure 6. Standard curve showing absorbance at 405 nm for various levels of spectrin in an ELISA assay.

Figure 7. Left (rostral to caudal, 1-4) and right (5- 8) hippocampi of a control animal and the left (9-12) and right (13-16) hippocampi from a kainate injected animal. Top panel, western blot assay; bottom panel, ELISA assay.

Figure 8. Spectrin immunoreactivity of CSF samples, as measured by ELISA, from two patients of group 1 (unruptured aneurisms) and four patients of group 2 (ruptured aneurisms) . Figure 9. Spectrin immunoreactivity of CSF samples, as measured by ELISA for: ANEU (unruptured aneurism patients (n * 2)); SAH (subarachnoid hemorrhage patients (n = 12)); AD (Alzheimer's Disease patients (n = 3)); STROKE (stroke patient (n = 1)); IVH (intravascular hemorrhage patients (n - 3)); and PICK'S (Pick's Disease Patient (n = 1)) .

Detailed Description of the Preferred Embodiment

The present invention relates to sensitive and

efficient methods for the early detection of cellular deat and degradation. The methods of the invention detec cellular death and/or degradation through an assay fo spectrin or the breakdown products of spectrin. The activation of Calpain leads to the proteolysis o many proteins including spectrin. Accordingly, Calpain i believed to cause production of spectrin breakdown product ("BDP's") from spectrin in dead and degenerating cells Thus, the detection of BDP's is believed to advantageousl serve as an indicator of the activation of Calpain.

Spectrin BDP's are unusually stable polypeptides BDP's can be detected in vivo for up to as long as tw months or more after release from the cytoskeleton. Thus BDP's can advantageously remain to serve as indicators o cellular death or degradation during this period o stability.

There is evidence that the activation of Calpain is a early event in cell death. This is in contrast to othe known proteases which are believed to be activated only i the late stages of cell death. The activation of Calpai is believed to often occur before the onset of pathologica symptoms associated with cell death. Thus, the detectio of BDP's is believed to advantageously be useful as method for the early detection of cell death, potentiall prior to the onset of pathological symptoms.

The present invention advantageously provides a metho for the detection of cellular pathology by means of a immunoassay to determine the presence of BDP's of spectrin Two major BDP's are known, BDPl and BDP2. In one aspect o the invention, the components from a sample of spectrin containing cells are physically separated, as by exposur to an electric field, in such a way that BDP's and spectri are separated. The separated components can then b visualized, as by staining with a stain such as Coomassi Blue. Alternatively, antibodies reactive with BDP's ca then be contacted with the separated sample, and antibod binding to the portion of the sample containing BDP'

determined. The determined amount of BDP's can then be compared with a basal level of BDP's in similar samples from normal patients. An increase in the level of BDP's is indicative of cellular death or degradation. In many cases, the basal level will be a level which is below the detection threshold of the methods herein described. Thus, in these cases, the detection of any immunoreactivity is an indication of cellular death or degradation.

The invention provides an additional method for the detection of cellular pathology to determine the presence of intact spectrin itself or spectrin immunoreactivity. In one embodiment of this aspect of the invention, an enzyme- linked immunosorbent assay (ELISA) for spectrin immuno¬ reactivity in biological samples, such as tissue extracts, cerebro-spinal fluid (CSF) , or blood serum is provided. One particular application of the method is the detection of spectrin or spectrin immunoreactivity in the CSF as an indication of neurodegenerative conditions such as subarachnoid hemorrhage, Alzheimer's Disease, HIV-induced neuropathy and/or stroke.

EXMTOE I BDP's were identified from a sample of purified spectrin, by exposing the sample to Calpain for varying lengths of time and exposing the treated samples to SDS-

PAGE. Brain spectrin was purified to greater than 90% purity by the method of Davis and Bennett, J. Biol. Chem.

258:7757-7766 (1983), which is incorporated herein by reference. Calpain was purified to a similar level of purity from rat erythrocyte cytosol according to the method of Seubert, et al., Svnaose 1:20-24 (1987), which is incorporated herein by reference.

Spectrin at a concentration of 75 μg/ml was incubated at 30 ^* C with 100 μM CaCl , 3 μg/ml calpain I, 20mM Tris-Cl, 5mM 9-mercaptoethanol and 150 mM NaCl at pH 7.5. Aliquots were withdrawn at 10 minutes and at 30 minutes. The aliquots were added to one-third volume of 3X SDS-PAGE

buffer (150mM Tris-Pθ4, 6%, SDS, 30% glycerol, 3.75mM EDT 3% ø-mercaptoethanol, pH 6.8). The samples were heated in 90*C water bath for 3-10 minutes, and subjected to SDS-PA on 3 to 10% gradient gels. The gels were stained wi Coomassie blue and destained with 7% acetic acid. T foregoing method is described in Seubert, et al.. Synap 1:20-24 (1987), which is incorporated herein by reference. The amount of peptide in two peptide bands of approximatel 150 kilodaltons (kD) and 155 kD respectively (referred jointly as the "150 kD bands") were found to increase wi exposure time to Calpain. Correspondingly, the amount peptide in the two peptide bands representing the α and subunits of spectrin at 240 kD and 230 kD respectivel decreased with exposure time to Calpain. The peptides o the two 150 kD bands were termed BDPl and BDP2.

Thus, Example I shows that spectrin produces BDP's i the presence of Calpain I. Example I also shows tha staining after SDS-PAGE can be used to detect spectrin o BDP's in samples having the relatively high level necessary for polypeptide bands to be visible to the nake eye upon exposure to stains such as Coomassie Blue. Th method of Example I is also well suited to relatively pur samples where the bands corresponding to the BDP's an intact spectrin can be easily identified. Samples havin suitably high levels and purity of spectrin and/or BDP' are, for example, obtained from homogenized neural tissue after purification. See Davis and Bennett, supra.

A more sensitive method of detecting spectrin o BDP's, even in complex mixtures of polypeptides ca advantageously be obtained by exposing the separated sampl to antibodies reactive with spectrin or BDP's. One assa which is suitable for this purpose has come to be known a a Western blot assay.

BDP's exhibit apparent stability towards furthe degradation, suggesting that antibodies directed agains spectrin can recognize the BDP's in biological samples such as tissues, fluids, etc. Both BDP's and intac

spectrin can be recognized by antibodies directed against spectrin. Accordingly, anti-spectrin antibodies will detect both intact spectrin and BDP's when used in a Western blot assay. The following example shows such a Western blot assay, using the gel obtained from Example I and anti-spectrin antibodies to detect the presence of both intact spectrin and BDP's.

Western Blot Assay for Spectrin and BDP's Antibodies to brain spectrin were raised in rabbits using well known procedures (see, for example. Hum, B.A.L. and Chantler, S.M., Meth. Enz. 10:104-135 (1988), which is incorporated herein by reference) . The anti-brain spectrin antibodies were purified from serum by brain spectrin- sepharose affinity chromatography. Briefly, antibodies to brain spectrin were isolated from the serum by adsorption to brain spectrin coupled to 6-amino hexanoic acid activated sepharose 4B (Sigma Chemical Co., St. Louis, MO). The specifically bound antibodies were then eluted in 0.2 M glycine, pH 2.8. These affinity purified antibodies were then equilibrated to pH 7.4 and frozen until use. The antibodies were found to be reactive against BDPl and BDP2 as well as to intact spectrin. Thus, the 150 kD bands which appeared upon exposure of spectrin to Calpain comprised polypeptides which were cross reactive with spectrin.

Purified brain spectrin was incubated as described in Example I. After SDS-PAGE, the proteins were electrophoretically transferred to a nitro-cellulose membrane using a Trans-Blotter (Bio-Rad, Richmond, CA) according to the manufacturer's recommendations for the transfer of high molecular weight proteins. The nitro- cellulose sheets were incubated with anti-spectrin antibodies and the bound antibodies detected using an Immuno-Blot assay kit (also available from Bio-Rad) according to the manufacturer's directions. Briefly, anti- rabbit IgG (Bio-Rad, Richmond, CA) conjugated to alkaline

phosphatase was used in a 5-bromo-4-chloro-3-indoly phosphate/nitro blue tetrazolium substrate system o detection according to the manufacturer's recommendations Affinity purified anti-spectrin antibody was diluted 1/75 (in a volume of 50 ml) and incubated overnight with th blot during the primary antibody step. The immuno reactivity of the bands is shown in Figure IB, lanes 1-3 Lane 1 shows spectrin without exposure to Calpain; Lane shows spectrin after treatment with ImM CaCl2 and 1.8 μg/m calpain I for 10 minutes; and Lane 3 shows spectrin afte treatment with ImM CaCl2 and 3 μg/ml calpain I for 3 minutes. It can be seen that in the presence of Calpain the degradation of spectrin produces BDP's, primarily BDP and BDP2. The findings of Example II and other simila experiments led us to the discovery that the increase Calpain activity following denervation or injury results i significant levels of BDP's being generated in injure tissues. In accordance with the foregoing discovery, in on aspect of the present invention, there is provided a metho for detecting cellular pathology comprising the steps o extracting a sample from a subject mammal and analyzing th sample for the presence of spectrin BDP's. In order t obtain greater sensitivity, the analyzing step can involv an immunoassay using antibodies which recognize spectrin o stable breakdown products of spectrin.

In this aspect of the invention, the amount of BDP' is determined and this amount is compared to a basal leve of BDP; any increase of BDP's being indicative of cellula death or degradation. Normally, the basal level is th level of BDP's from healthy cells. The basal level can b taken from a corresponding sample in a healthy subjec mammal. Alternatively, the basal level can be obtaine from a sample from the same subject at a point in tim prior to the insult. Thus, a series of samples can b taken from a single subject over time and analyzed for th

presence of BDP's, thereby advantageously providing an indication of the course of cellular death or degradation in the subject. In many samples taken from healthy subjects, the level of BDP's is below the detection threshold of the assays described herein. Accordingly, the basal level against which the detected amount of BDP's are compared is often zero. Therefore, in many samples, the detection of any BDP's is indicative of cellular death or degradation. The following example illustrates the establishment of a basal level for human CSF.

EXAMPLE in Establishment of a Basal Level of BDP's in Human CSF CSF samples are obtained from a healthy human subject. All CSF samples are concentrated by ultrafiltration. Two marker samples are also obtained to identify bands corresponding to intact spectrin and BDP's in completed gels. The first marker sample is a sample of purified spectrin without exposure to Calpain as in Lane 1 of Example I to show the position of intact spectrin. The other marker sample is of purified spectrin after exposure to Calpain as in Lane 3 of Example I to show the position of BDP's. Protein concentration of the samples and marker samples was determined by the method of Bradford, Anal. Biochem. 72:248-254 (1976), the disclosure of which is hereby incorporated by reference. Ten μg, of each sample and marker sample is subjected to SDS polyacrylamide gel electrophoresis on a 3-10% gradient gel until a bromophenol marker dye reaches the front of the gel. The proteins are then transferred to nitrocellulose membrane using a transblot apparatus (Bio-Rad, Richmond, CA) according to the manufacturer's instructions for high molecular weight proteins. Antibodies are produced as in Example II and used to detect spectrin immunoreactivity on the nitrocellulose membrane. Spectrin immunoreactivity is found for each of the CSF samples at positions corresponding to the position of the intact spectrin as

determined by the marker sample. However, no detectabl spectrin immunoreactivity is found in the CSF samples at position corresponding to the position of the BDP' determined by the marker sample. Thus, a basal level o BDP's for CSF of this human subject is determined to b zero.

The methods described herein can be used to measur BDP's in a variety of tissues and fluids because spectri is found in a variety of tissues. For example, BDP's o spectrin have been observed in blood platelets (Fox, e al., Blood 69:537-545 (1987)) and intestinal brush borde cells (Glenney, et al., PNAS 79:4002-4005 (1982)). Th following tissue taken from rats have been examined by th present inventors and others using the methods describe herein and found to exhibit spectrin and BDP's submandibular gland, brush border, testes, thymus, skeleta muscular, heart muscle, lung, liver, spleen, adrenal gland kidney, brain. Additionally, humans, gerbils and mice hav been determined by the present inventors and others t contain spectrin and BDP's, suggesting that spectrin an BDP's are common to all mammals.

Injury in the mammalian Central Nervous System (CNS) results in both the degeneration of damaged neurons an growth responses of undamaged neuronal elements. A well documented paradigm for investigating the mechanism underlying these processes involves lesioning th entorhinal cortex, resulting in the production of a well defined dendritic zone in the dentate gyrus deprived of th majority of its axonal inputs. The anatomical consequence of denervation include dendritic atrophy, glial hypertroph and atrophy, and a growth response in undamaged axons.

Thus, to show the ability of a preferred method of th present invention to detect the well-defined dendritic zon in the dentate gyrus after lesioning the entorhinal corte of rats, Examples IV through VIII are provided, showing th detection of cellular death or degradation in the expecte tissues.

EXAMPLE IV Preparation of Dentate Gyrus -?aιn 7^ Stereotaxically-placed unilateral electrolytic lesions of the entorhinal cortex were made in Sprague-Dawley rats. Animals were sacrificed after postoperative survival times of 0.2, 0.4, 1, 2, 4, 7, 14 and 27 days.

Immediately after sacrificing the animals by decapitation, brains were rapidly dissected in ice-cold homogenization buffer consisting of 0.32 M sucrose, lOmM Tris, 2mM EDTA, ImM ethylene glycol bis ( -amino- ethylester) N,N,N',N'-tetraacetic acid (EGTA) , 100 μM leupeptin, 1 μg/ml N-tosyl-L-phenylalanine chloromethyl ketone (TPCK) , pH 7.4. Each hippocampus was dissected free and cuts were made with a scalpel blade to isolate the dentate gyrus. With the hippocampus resting on the alvear surface, one cut was made longitudinally along the hippocampal fissure to separate the subiculum and another longitudinal cut removed most of the CA3 field. A third cut was then made in CAl, parallel to the fissure, to remove the majority of CAl. The remaining tissue (10-20 mg) served as the dentate gyrus sample which was used as a tissue sample as in Example V. Contralateral and ipsilateral samples of the dentate gyrus were obtained.

EXAMPLE V Preparation of Tissue Samples for Electrophoresis

The contralateral dentate gyrus tissue sample and the ipsilateral dentate gyrus tissue sample were each homogenized in 500 μl of dissection buffer. An aliquot of each was added to one-third volume of 3X SDS-PAGE sample buffer (consisting of 150mM Tris-P04, 6% SDS, 30% glycerol, 3.75mM EDTA, 3% β-mercaptoethanol, pH 6.8) and placed in a 90*C water bath for three minutes. The protein concentration of each homogenate was determined by the method of Bradford, supra. The concentration of proteins in each homogenate sample was then adjusted to 0.33 mg/ml with additinal samplebuffer. EXAMPLE V

Separation of Sample Proteins and Transfer t-n ff _? τni^ _π <=.g

Ten μg of protein from each of the samples fro Example V, were subjected to SDS polyacrylamide ge electrophoresis on a 3-10% gradient gel until a bromopheno marker dye reached the front of the gel. The proteins wer then transferred to nitrocellulose membrane using transblot apparatus (Bio-Rad, Richmond, CA) according t the manufacturer's instructions for high molecular weigh proteins.

EXAMPLE V I Preparation of Anti-Soectrin Antibodies

Antibodies were produced by the following method: Fo each rabbit, approximately 200 μg of purified brai spectrin was excised from SDS-polyacrylamide gels (afte electrophoretic separation) and emulsified with Freund' complete adjuvant. Multiple subdermal injections were mad and the procedure repeated again after two to four weeks using Freund's incomplete adjuvant. After an additiona two weeks, subcutaneous injections of an emulsio containing approximately 100μg of spectrin were made. Thi procedure was repeated approximately one month later. Te days following this series of injections, approximately 2 ml of blood was drawn from each rabbit and the seru collected after allowing the blood to clot overnight a 4 ^β C. Antibodies to brain spectrin were then affinit purified by adsorption to brain spectrin coupled to δ-amin hexanoic acid activated sepharose 4B, as described i Example II. The affinity purified antibodies were the equilibrated to pH 7.4 and frozen until use. E M LE V II

Determination of BDP's Resulting from Brain Lesions To determine the amount of spectrin immunoreactivit on the membrane of Example VI, the membrane was exposed t the antibody of Example VII as part of a Western Blo assay. Procedures for blocking, primary and secondar antibody incubations and color development were a described in Example II. Quantitation of th

immunoreactive species was made using a soft laser scanning densitometer (Model #SLR504-XL, BioMed Instruments, Fullerton, CA) . An integrator (Model 4270, Varian, Sunnyvale, CA) summed the amount of reaction product in each band and expressed them as a percentage of the total in that sample.

The anti-spectrin reactive species present in the contralateral (lane 1) and ipsilateral dentate gyrus (lane 2) two days after a unilateral entorhinal lesion are shown in Figure IA. The homogenates of the ipsilateral dentate gyrus exhibited a marked increase in the amount of two peptides, termed BDPl and BDP2, with apparent M _r s of about 155,000 and 150,000 Daltons, respectively.

The procedures of Examples IV through VIII were repeated, allowing various lengths of time to elapse between the lesion and sacrifice of the animals of Example IV. The time course of the changes in BDPl and BDP2 in the dentate gyrus following a unilateral entorhinal cortex lesion is shown in Figure 2. BDP's are usually below the limit of detection in samples from unoperated animals. A significant elevation of BDP's in the ipsilateral sample is evident as early as four hours post-lesion. The increase is maximal two days after the lesion, where the BDP's represent 25% of the total immunoreactivity. Two and even four weeks after the lesion, the amounts of BDP's were still significantly increased; the contralateral dentate gyri at two and four weeks showed no detectable BDP's and average BDP2 levels were less than 0.1% of total spectrin immunoreactivity. Small increases in the amounts of BDP's were observed in the contralateral region during the first week following the lesion as compared to non-operated animals.

The results indicate that removal of the main input to the dentate gyrus is followed by a rapid and long-lasting increased degradation of the cytoskeletal protein brain spectrin. It is known that aberrations in cytoplasmic calcium levels occur in the dendritic zone of the dentate

gyrus after lesioning. See, for example, Baudry, et al., Neurosci.. 3:252-259 (1983), the disclosure of which i hereby incorporated by reference. Thus, we believe that th elevated levels of BDP's in these tissues is the result o the activation of Calpain in these tissues by these aberran levels of calcium. The results of Example VIII, therefore confirm the ability of the present invention to detect th well-defined dendritic zone in the dentate gyrus afte lesioning of the entorhinal cortex in rats. Examples IX through XI are provided in order to sho that the methods utilized in Examples V through VIII hav widespread utility in detecting cellular death o degradation. These examples show the detection of cellula death or degradation from a variety of causes and in variety of cellular tissues through methods of the presen invention. As such, these examples are intended t illustrate, not to limit the invention. While th procedures described herein, such as those of Examples through VIII, are typical of those that might be used other alternative procedures known to those skilled in th art can be alternatively employed.

EXAMPLE IX

Assay for BDP in the Brindled Mouse.

A Hereditary Degenerative Condition Samples from the hippocampus, cortex and striatu o 12 day mouse pups were processed as described for th dentate gyrus samples in Examples V-VIII. The experimenta groups were control mice, Brindled mice, and Brindled mic receiving supplemental copper, a treatment which prevent the premature death which otherwise occurs. Brindled mic are characterized in that they have a copper deficit whic untreated causes normal degradation. As can be seen i Figure 3, spectrin BDP's are elevated in the pathologica condition and this elevation is blocked by the coppe supplement.

EXAMPLE X Assay for BDP after Exposure to the Industrial Toxin T-rJm thylt n (TMT) Three rats were injected intraparitoneally (i.p.) with 10 mg TMT/kg body weight. A fourth rat was injected i.p. with saline alone to serve as control. At 3, 7 and 14 days post-treatment, the dentate gyrus and CAl regions of the hippocampus of the three test rats were removed and analyzed for BDP's, as described in Examples V-VII. Massive increases in BDP's were noted in both the dentate gyrus and CAl regions, as depicted in Figure 4. These regions of the brain have been identified as the regions most at risk to TMT toxin (see Balabin, et al., Neurosa 26:337-361 (1988), which is incorporated herein by reference) .

EXAMPLE XI

Assay for BDP Following Ischemia

Carotid arteries were clamped for 10 minutes to interrupt the principal blood flow to the cortex in each of two groups of eight Mongolian gerbils. Two control groups of gerbils were also analyzed. Samples of the CAl hippocampal region and the cerebellum were taken at 4 hours after ischemia from one group of control gerbils and one group of test gerbils. Samples were also taken at 24 hours after ischemia from the second control and test groups of gerbils. The test gerbils showed elevated BDP's in the CAl region compared to control animals, as shown in Figure 5.

The blood supply to the cerebellum was not interrupted and this structure showed no such increase. Analysis of BDP levels was as described in Example V - VIII.

Thus, the foregoing examples show that the methods of the present invention can advantageously be used to detect cellular death or degradation from a variety of causes in a variety of samples. The present invention advantageously provides an additional method for the detection of cellular pathology without the necessity of separating the sample into BDP and

intact spectrin. This additional method is by means of immunoassay to determine the presence of intact spectri itself or spectrin immunoreactivity regardless of source Therefore, in this embodiment of the invention, tota spectrin immunoreactivity, including immunoreactivity t spectrin and to BDP's, can be measured. In one embodimen of this aspect of the invention, an enzyme-linke immunosorbent assay (ELISA) for spectrin immunoreactivit in biological samples, such as tissue extracts, cerebro spinal fluid (CSF) , or blood serum is provided.

In preparation for the competitive ELISA assay of preferred embodiment, a spectrin sample is immobilized t polystyrene microliter plates. We have found that spectri desorbs from conventional activated polystyrene plate after immobilization, resulting in an unexpected bell shaped relationship between the amount of antibody boun and the amount of spectrin in the sample which is measured While not wishing to be bound by any particular explanatio for this unexpected result, it is believed that th desorbed spectrin forms polymers with still immobilize spectrin in the presence of accessory proteins present i the sample. The spectrin polymers are believed to be mor accessible to binding of anti-spectrin antibody. I addition, spectrin in the sample is believed to bind to th plate through further polymerization of the spectrin.

In order to prevent the unexpected relationshi between the amount of antibody bound and the spectrin i the sample, polystyrene plates can be treated wit glutaraldehyde prior to the immobilization of spectrin t the plates. Glutaraldehyde forms covalent bonds to bot the polystyrene of the plates and to the spectri molecules. The use of buffers with conditions, includin salinity and pH, unfavorable to the polymerization o spectrin has also been found to prevent the unexpecte results. High ion concentration has been found to inhibi formation of spectrin polymers, however, suc concentrations also interfere with immunoreactivity

Addition of various other agents has also been found t prevent the unexpected results, including EGTA, sucrose an detergents. Thus, in a preferred method of the presen invention, buffers with a pH slightly greater than 7.0 i physiological saline with EGTA, sucrose and detergent is used. Chaotropic salts, such as NaBr or KI, can also be used to inhibit formation of polymers.

When an unknown sample is tested, a limiting amount o anti-spectrin antibody is added to each well along with th sample. Spectrin in the sample competes for antibody wit the spectrin immobilized to the plate. Thus, the more spectrin in the sample, the less antibody will bind to the spectrin immobilized to the plate. Accordingly, the amount of antibody binding to the spectrin on the plate provides an indication of the amount of spectrin in the sample. The "ttount of antibody can be detected by a colorimetric . _^ action as in a standard ELISA procedure, or can be detected in any known manner, such as through a radio- immune assay. EXAMPLE m

ELISA Assay for Spectrin A spectrin preparation was prepared from rat brains by the method of Davis and Bennett (J. Biol. Chem. 258:7757-7766, 1983). Antibodies to spectrin were prepared by subjecting the spectrin preparation to SDS-PAGE (see Seubert, et al., Synapse 1:20-34, 1987), excising the region of the gel containing the spectrin, homogenizing the gel and immunizing rabbits with the homogenized gel according to established procedures (see, e.g.. Hum and Chantler, Methods Enzy ol. 70:104-135, 1980).

Microtiter plates having immobilized spectrin were prepared by first treating microtiter plates (unmodified polystyrene, such as those sold under the trade mark "Corning Easy-Wash") with glutaraldehyde 0.2% in 0.1M sodium phosphate pH=5.0 for 4 hours at room temperature. After removal of glutaraldehyde, 100 μl of a solution of spectrin (10 μg/ml) in 0.1M sodium phosphate pH=8.0 was

added to each well and the plates incubated an additional hours at room temperature. The plates were rinsed wit 0.1M lysine in 0.1M sodium phosphate pH=8.0, and 100 μl o this lysine solution was added to each well. The plate were then incubated for 4 hours at room temperature Lysine serves to react with unreacted glutaraldehyd binding sites to prevent the further binding of spectrin t the plates.

For the measurement of spectrin immunoreactivity of a unknown, the lysine solution in each well was discarded an a sample of the unknown was placed in each well. Th volume was then adjusted to 50 μl with 20mM Tris, 0.8 NaCl, 0.02% KC1, 0.5% bovine serum albumin, 0.05% Tween 20 2mM EGTA, 0.2% sodium azide pH=7.2 ("assay buffer"). T this was added 50 μl of a 1:50,000 dilution of anti spectrin antiserum in assay buffer. The plates were mixe and incubated overnight at 4'C. The plates were the rinsed 4 times in lOmM Tris, 0.9% NaCl pH=7.2 ("rins buffer") and 100 μl of biotinylated goat anti-rabbi antiserum (available from Vector Laboratories) , diluted i assay buffer at the concentration recommended by th manufacturer was added and the plates incubated on rocking platform at room temperature for 4 hr. The plate were rinsed 4 times with rinse buffer and 100 μl of AB (alkaline phosphatase) reagent (also available from Vecto Laboratories) prepared according to the manufacturer' instructions in assay buffer was added. The plates wer incubated for 2 hrs on a rocking platform at roo temperature and rinsed 6 times with rinse buffer. Colo was developed by adding 100 μl of alkaline phosphatas substrate solution (available from Bio-Rad) made accordin to the manufacturer's directions and incubating for 3 minutes to 4 hours at room temperature..

In parallel with the measurements of the unknow samples, measurements of spectrin immunoreactivity of well initially containing known concentrations of spectrin wer also performed.

The absorbance at 405 nm of the wells containing these standard concentrations was read using a plate reader and the standard curve shown in Figure 6 was produced from this data. The absorbance at 405 nm of the wells containing unknown samples was also read and the concentration of spectrin determined by comparing the absorbance of the unknown wells to the standard curve. The concentration of spectrin immunoreactivity correlates well with measurements of the same samples which are subjected to the Western Blot assay of Example VIII. The following example demonstrates the correlation between the Western blot and the ELISA assay in neurodegenerating rats.

EX P E XIII Comparison of Western Assay and ELISA Assay

Adult rats were given intracerebral ventricle injections of 75 ng of kainate, a compound known to cause neurodegeneration within the hippocampus. A second set of rats were given equal volume injections of saline. The rats were allowed to recover for four days. The hippocampi were then removed and divided into four sections, rostral to caudal. Each section was analyzed using both the Western blot assay as in Example II, and using the ELISA assay of Example XII. Results are shown in Figure 7. Figure 7 shows the left (rostral to caudal, 1-4) and right (5-8) hippocampi of a control animal and the left (9- 12) and right (13-16) hippocampi from a kainate injected animal. The top panel shows the Western Assay and the bottom panel shows the ELISA assay. It can be seen that the increase in the total amount of immunoreactivity measured by the ELISA correlates well with the increase in BDP's measured by the Western Blot assay.

One particular application of the ELISA of the present invention is the detection of spectrin or spectrin immuno- reactivity in the cerebrospinal fluid (CSF) as an indication of neurodegenerative conditions, including subarachnoid hemorrhage, stroke, multiple infarction

dementia, HIV-induced neuropathy and Alzheimer's Disease.

Although it is possible to detect small quantities o spectrin in normal CSF concentrated by ultrafiltration; i normal unconcentrated CSF, virtually no spectrin i muno reactivity is detected using the ELISA assay of the presen invention. Therefore, the detection of either spectrin o spectrin BDP's in unconcentrated CSF would be indicative o cellular death or degradation within the nervous system Although spectrin is present in isodermal cells lining th ventriculus and in particular in specific cells standin between the CSF and the blood, the number of such cells i very small compared to neural cells. The death o degradation of glial cells, the support cells for neurons would also be capable of contributing spectrin or BDP's t the CSF. However, after death or degradation of glia cells, the death or degradation of the neural cells suc cells support would follow shortly thereafter. Therefore the vast majority of spectrin immunoreactivity found in th CSF would be indicative of breakdown of neural cells. The cerebrospinal fluid of humans can be assaye either directly or after concentration using lyophilizatio or centrifugal ultrafiltration (using materials such a those sold under the trademarks "Centricon-10" o "Amicon"). The following example illustrates one typica method of the present invention for assaying human CSF fo the presence of spectrin immunoreactivity.

EXAMPLE XIV Assay of Spectrin Immunoreactivity -in Hnwan r.-v Cerebrospinal fluid samples were obtained from patients diagnosed as having unruptured aneurysms (group 1) and from 4 patients in which the aneurysm has burs producing subarachnoid hemorrhage (group 2) . Two ml of eac sample was lyophilized, resuspended in 100 μl water, and 1 μl of the resulting solution was analyzed for spectri immunoreactivity using the ELISA assay of Example XII. Results are shown in Figure 8. The CSF samples from th subarachnoid hemorrhage group all showed spectrin immuno

reactivity while the unruptured aneurism group had no detectable spectrin immunoreactivity.

Thus, it can be seen from Example XIV that the presence of detectable quantities of spectrin immuno- reactivity in the CSF is indicative of cellular death or degeneration in neural tissue.

In summary, the foregoing examples clearly show that neurodegeneration in vivo dramatically elevates total spectrin immunoreactivity, as measured by the ELISA. To demonstrate the widespread applicability of the ELISA assay in detecting neurodegeneration, the following example was performed.

EXAMPLE XV Levels of Spectrin Immunoreactivity in CSF Total spectrin immunoreactivity was measured, using the ELISA method of Example XII, in CSF taken from a number of different patients suffering from a variety of conditions known to be associated with neurodegeneration. Results are shown in Figure 9. The first column of Figure 9, labeled "ANEU" shows the results of two patients who had brain aneurisms detected and surgically corrected before the aneurisms burst, as in Group 1 of Example XIV. Thus, significant neurodegeneration would not be expected in these patients. As can be seen in Figure 9, no spectrin immunoreactivity was found in CSF taken from these patients. The data confirms that no spectrin immuno¬ reactivity is detected in CSF in non-neurodegenerating mammals using the ELISA assay of the present invention.

The next column of Figure 9, labeled "SAH", shows spectrin immunoreactivity measurements in CSF of 12 patients with subarachnoid hemorrhage who had had CSF drains installed. The CSF from all 12 patients shows spectrin immunoreactivity, indicating that neurodegeneration has occurred. The third column of Figure 9, labeled "AD" shows spectrin immunoreactivity measurements in CSF from 3 Alzheimer's disease sufferers. All three patients show

spectrin immunoreactivity in their CSF, indicatin neurodegeneration has occurred.

The fourth column of Figure 9 shows spectrin immuno reactivity in the CSF from one stroke victim. It can b seen that spectrin immunoreactivity is quite high in thi patient, indicating significant neurodegeneration.

The fifth column of Figure 9, labeled "IVH" show spectrin immunoreactivity from three premature infant suffering from intraventricular hemorrhage. Results sho that two out of three of these patients show spectri immunoreactivity in their CSF, indicatin neurodegeneration.

The last column of Figure 9 shows spectrin immuno reactivity in one victim of Pick's disease. The result show high levels of spectrin immunoreactivity in thi patient, indicative of the neurodegeneration whic accompanies this disease. Pick's disease is clinicall very difficult to distinguish from Alzheimer's disease Presently, Pick's can only be readily distinguished fro Alzheimer's upon autopsy. It can be seen from the presen data, that the Pick's sufferer had significantly highe levels of spectrin immunoreactivity than any of th Alzheimer's sufferers. Thus, it is believed that th present method will provide a diagnostic tool i distinguishing between these two diseases by the generall higher levels of spectrin immunoreactivity found in the CS of Pick's patients.

Thus, it can be seen from the foregoing example tha measurements of spectrin immunoreactivity in CSF are usefu indicators of neurodegeneration from a wide spectrum o clinical causes.

The spectrin immunoreactivity detected in all of th foregoing examples is, of course, due to a large number o different antigenic epitopes. It is believed that afte the proteolysis of spectrin to BDP's, additional or occul epitopes are exposed which are not present in intac spectrin. Thus, when performing the ELISA assay usin

polyclonal antibodies raised against BDP's, BDP's can give a stronger signal than the intact spectrin. In such assays, treating the spectrin in such a way to expose the occult epitopes, can also give a stronger signal than intact spectrin.

The anti-spectrin antibodies used in the Western Blot assays of examples II - XI, were affinity purified using the affinity purification method described in Example II. This affinity purification step with intact spectrin, resulted in the purification of antibodies to epitopes of spectrin present and exposed in intact spectrin. However, the raw serum contained at least two other classes of antibodies which react against spectrum. One class of antibodies are to epitopes of spectrin not exposed in the intact tetrameres, but exposed in cleaved spectrin. Another class of antibodies would be antibodies specific to spectrin-SDS complexes. This class of antibodies is expected because the spectrin used to immunize the rabbits producing the antibodies in Example II was purified from SDS-PAGE, resulting in the formation of these SDS-spectrin complexes. In order to demonstrate that raw, not affinity purified, anti-spectrin antibodies raised against denatured rat spectrin react more efficiently with denatured spectrin than with intact spectrin, and that, therefore, occult epitopes exist and that antibodies against the occult epitopes can be used to distinguish native from denatured spectrin, the following example was performed.

EXAMPLE XVI Immunoreactivity of Denatured and Native Spectrin Spectrin was immobilized onto polystyrene plates as in Example XII. Each well was incubated with one of four denaturing treatments for one hour at room temperature and then washed six times in wash buffer (50mM tris, 150 mM NaCl pH=7.5). The four treatments were: control (wash buffer), 1% SDS, 1 M acetic acid, and 1 M KI. Four wells for each treatment were analyzed. The amount of spectrin immunoreactivity on the plate was determined by incubating

plates with raw serum from immunized rabbits at 1:10,00 dilution in assay buffer (lOOμl/well) overnight at 4"C rinsing the plates four times with wash buffer, an detecting bound antibody using the Vector ABC-AP kit as i Example XII. Results are shown in Table 1.

TABLE 1

TREATMENT ABSORBANCE f405nm) ± S.D.

Control 0.997 ± 0.115 1% SDS 2.540 ± 0.281

1 M Acetic Acid 1.110 ± 0.087

1 M KI 1.835 ± 0.117

It can be seen from the results of Table 1, that ra serum from rabbits immunized with SDS-treated spectri recognizes denatured spectrin more effectively than th native, control-treated spectrin. Not unexpectedly, th SDS-treated spectrin reacts the most strongly with thi serum. Thus, it is clear from these results tha denaturing spectrin exposes occult epitopes not present i intact spectrin molecules. It is expected that othe denaturing treatments, such as TCA, organic solvents, ethanol and guanidine, will produce similar increases i immunoreactivity.

It is also believed that the cleavage of intac spectrin into BDP's exposes hidden epitopes. In order t demonstrate that cleavage of spectrin in solution expose hidden epitopes and that antibodies directed against thes epitopes can be used to distinguish intact spectrin fro cleaved spectrin, the following example was performed.

EXAMPLE XV Cleavage of Spectrin to Increase Immunoreactivity Rat brain was homogenized in 10 mM HEPES, ImM EGTA, mM DTT pH = 7.2. The homogenate was centrifuged at 12,00 x g for ten minutes and the supernatant split into tw fractions. CaCl2 was added to Fraction 2 in order t activate Calpain. The final concentration of Ca ⁺⁺ i Fraction 2 was 50 mM. No CaCl ₂ was added to Fraction 1.

Both fractions were incubated at 37"C for 30 minutes. Th immunoreactivity of the samples was taken both before an after this incubation period using the ELISA assay o Example XII. Extensive proteolysis of the second sample, leading to formation of BDP's, was confirmed by Wester Blot analysis as in Example II. No proteolysis was detected in any of the other samples. No precipitate was observed in any samples. Results of the ELIS determination are shown in Table 2.

TABLE 2

APSQRBA CE AT -. Q5m

FACTION ■ Q p ) _τ t ^■ ?o mint

1 (- Ca ⁺⁺ ) 0.316 ± 0.014 0.381 ± 0.023 2 (+ Ca ⁺⁺ ) 0.334 ± 0.033 0.451 ± 0.022

It can be seen from Table 2 that in Fraction 2 where Calpain is activated that total immunoreactivity increased from an average of 0.381 to an average of 0.451, an increase of 18%. In no instance did the immunoreactivity of Fraction 1 exceed the immunoreactivity of Fraction 2. Thus, the foregoing example shows that cleavage of spectrin into BDP's in vitro increases the immunoreactivity towards raw serum of rabbits immunized with SDS-spectrin complexes.

Referring back to Figure 7, where the Western Blot assay is compared to the ELISA assay, it can be seen that the amount of spectrin immunoreactivity is most dramatically increased in those sections of rat hippocampus in which increased BDP's are found. Thus, the data of Figure 7 confirms that immunoreactivity is enhanced by cleavage into BDP's in vivo, as well as in vitro.

The availability of hidden epitopes in intact spe trin suggests an ELISA assay or other immunoassay for specifically detecting BDP's as opposed to intact spectrin. Such an assay could use antibodies directed solely to these hidden epitopes, obtained through methods known in the art.

such as through affinity purification or the production o a monoclonal antibody directed to a hidden epitope. It i expected that the affinity purification of the class o antibodies directed to epitopes present and exposed i intact spectrin as in Example II results in a fraction no bound to the spectrin-sepharose containing antibodie directed to hidden epitopes.

Using a source of antibodies directed to hidde epitopes, it is expected that a determination of the amoun of total spectrin immunoreactivity and the amount of BD immunoreactivity could be separately made.

Alternatively, it is believed to be possible t separate intact spectrin from BDP's, including the BDPl an BDP2 polypeptides visualized by Western blot and othe spectrin fragments, by altering the conditions of th solution to affect the solubility of the intact spectri molecule. By altering the pH, ionic strength, or othe such factors, it is believed possible to solubilize th BDP's while precipitating the intact spectrin molecules. It is believed that treatment of samples containin spectrin immunoreactivity with an agent selected from th group consisting of detergents, agents which produce a acidic or basic pH (preferably a pH of greater than 8.5 o less than 5.5), chaotropic agents and organic solvents o lowered dielectric will result in the altered solubilit conditions required to precipitate intact spectrin withou precipitating some or all of the BDP's. By removing th precipitated intact spectrin molecules, a determination o the amount of BDP's present can be obtained. Fo commercial utility, an enrichment for BDP's of at least te fold is preferable, more preferably on the order of on hundred fold.

Although the invention has been described wit reference to the presently preferred embodiments, it shoul be understood that various modifications can be made

without departing from the spirit of the invention. Accordingly, the invention is limited only by the following claims.