Definitions

• the invention relates to methods for identifying proteolytic enzymes with specificity for processing the precursor to the Alzheimer's Disease (hereinafter "AD") beta-amyloid protein; methods for identifying inhibitors of proteases specific for the precursor to the beta-amyloid protein; and methods for regulating formation of beta-amyloid protein with inhibitors of proteases specific for the precursor to the beta-amyloid protein, such as inhibitors of aspartic protease, cathepsin D, and a chymotryptic-like serine protease.
• AD Alzheimer's Disease
• the present assays have utility in the identification of the proteases which control the rate of formation of amyloidic peptides in the brains of AD patients. As such, they can be used to isolate such proteases, and can also be used to identify protease inhibitors which can be used as therapeutics for AD. Described hereinbelow is the application of the assays to identify the aspartic protease, cathepsin D as a major amyloidogenic protease for processing Amyloid Precursor Protein (hereinafter "APP"). Also provided is a partial characterization of a second, serine protease which can form amyloidic precursors for the APP holoprotein.
• APP Amyloid Precursor Protein
• AD is a progressive, degenerative disorder of the brain, characterized by progressive atrophy, usually in the frontal, parietal and occipital cortices.
• the clinical manifestations of AD include progressive memory impairments, loss of language and visuospatial skills, and behavioral deficits (McKhan et al., 1986, Neurology, 34: 939).
• Overall cognitive impairment is attributed to degeneration of neuronal cells located throughout the cerebral hemispheres (Price, 1986, Annu. Rev. Neurosci, 9: 489).
• the primary distinguishing features of the post-mortem brain of an AD patient are: (1) pathological lesions comprised of neuronal perikarya containing accumulations of neurofibrillary tangles; (2) cerebrovascular amyloid deposits; and (3) neuritic plaques.
• Both the cerebrovascular amyloid (Wong et al., 1985, PNAS, 82: 8729) and the neuritic plaques (Masters et al., 1985, PNAS, 82: 4249) contain a distinctive peptide simply designated "A4" or "beta-amyloid".
• Beta-amyloid is an insoluble, highly aggregating, small polypeptide of relative molecular mass 4,500, and is composed of 39 to 42 amino acids.
• Several lines of evidence support a role of beta-amyloid in the pathogenesis of AD lesions. For instance, beta-amyloid and related fragments have been shown to be toxic for PC-12 cell lines (Yanker et al., 1989, Science, 245: 417); toxic for primary cultures of neurons (Yanker et al., 1990, Science, 250: 279); and cause neuronal degeneration in rodent brains and corresponding amnestic response in the rodents (Flood et al., 1991, PNAS, 88: 3363; Kowall et al, 1991, PNAS, 88: 7247).
• APP holo-amyloid precursor protein
• Kang et al., 1987, Nature, 325: 733 described the beta-amyloid protein as originating from and as a part of a larger precursor protein. To identify this precursor, a full-length complementary DNA clone coding for the protein was isolated and sequenced, using oligonucleotide probes designed from the known beta-amyloid sequence. The predicted precursor contained 695 residues and is currently designated "APP 695" (Amyloid Precursor Protein 695).
• APP 695 is the most abundant form of APP found in the human brain, but three other forms exists, APP 714, APP 751 and APP 770 (Tanzi et al., 1988, Nature, 351: 528; Ponte et al, 1988, Nature, 331: 525; and Kitaguchi et al, 1988, Nature, 331: 530).
• the different length isoforms arise from alternative splicing from a single APP gene located on human chromosome 21 (Goldgaber et al., 1987, Science, 235: 877; and Tanzi et al, 1987, Science, 235: 880).
• APP 751 and APP 770 contain a 56 amino acid Kunitz inhibitor domain, which shares 40% homology with Bovine Pancreatic Trypsin Inhibitor. Both these forms of APP have protease inhibitory activity (Kitaguchi et al., 1988, Nature, 311: 530; and Smith et al., 1990, Science, 248: 1126), and at least one of these forms is probably what was previously identified as Protease Nexin II (Oltersdorf et al., 1989, Nature, 341: 144; Van Nostrand et al., 1989, Nature, 341: 546).
• amyloid precursor proteins The physiological role for the amyloid precursor proteins has not yet been confirmed. It has been proposed to be a cell surface receptor (Kang et al., 1987, Nature, 325: 733); an adhesion molecule (Schubert et al., 1989, Neuron, 3: 689); a growth or trophic factor (Saitoh et al., 1989, Cell, 58: 615; Araki et al, 1991, Biochem. Biophys. Res.
• C-100 fragment APP fragments extending from the N- terminus of A4 to the C-terminus of the full length APP (referred to hereinafter as the "C-100 fragment", because it is comprised of approximately 100 amino acids) are also capable of aggregation both in vitro (Dyrks et al., 1988, EMBO ⁇ ., 7: 949), and in transfected cells (Wolf et al, 1990, EMBO ⁇ ., 9: 2079; and Maruyama et al., 1990, Nature, 347: 566). Over-expression of the C-100 fragment in transfected P19 cells has been shown to cause cellular toxicity (Fuckuchi et al., 1992, Biochem. Biophys. Res. Comm., 182: 165).
• C-terminal fragments containing both the beta-amyloid and the C-terminal domains have been shown to exist in human brain (Estus et al., 1992, Science, 255: 726), and studies in transfected cell lines suggest that these fragments may be produced in the endosomal-lysosomal pathway (Golde et al., 1992, Science, 255: 728).
• CSF cerebral spinal fluid
• secretases The enzymes responsible for the normal, non-pathological processing of APP have been termed "secretases". C-terminal fragments resulting from secretase action are smaller than the C-100 fragments (defined above) by 17 amino acids, and will hereinafter be referred to as the "physiological C-terminal fragment.”
• amyloidogenic APP processing was initially suggested to be an endosomal-lysosomal event (Golde et al, 1992, Science, 255: 728; and C. Haas et al., 1992, Nature, 357: 500), there is recent evidence that beta-amyloid is released by cultured cells (C. Haas et al, 1992, Nature, 359: 322; and Shoji et al., 1992, Science, 258: 126), along with an alternatively processed form of secreted APP (Suebert et al. 1993, Nature, 361: 260), consistent with participation of protease within the secretory pathway or at the plasma membrane in beta-amyloid formation.
• transfected cell lines expressing the APP 695 associated with the Swedish form of FAD were shown to release beta-amyloid like fragments 6-8 times faster than cells transfected with wild type APP (Citron et al., 1992, Nature, 360: 672; and Cai et al., 1993, Science, 259: 514; and 1992, Neuroscience Lett., 144: 42), although similar studies of the effect of the London (V to I) mutation showed no effect on amyloid release (See, Cai et al., Id.).
• the amyloid released by cultured cells contains an unusual form of beta-amyloid with an N-terminus starting at valine 594 (numbering according to reference 1) of the APP precursor (C.
• Beta-amyloid must be formed by the direct action of protease(s).
• the identification of the so-called "pathologic" brain protease(s) responsible for the C- 100 or beta-amyloid formation is an essential step in an effort to develop therapeutic protease inhibitors designed to block amyloid accumulation. Identification of such enzymes requires the development of specific, assays for the activity of such proteases which would allow one to specifically measure the activity of the proteases in the presence of other brain proteolytic enzymes which are present in brain extracts.
• Such assays are then used to detect the protease during protease purification. Finally, the assays can be used to measure the effect of potential inhibitors of the enzyme such as is required in pharmaceutical screening for lead therapeutic compounds.
• the pathologic protease is: lysosomal in origin (Cataldo et al., 1990, Proc. Natl. Acad. Sci. USA, 87: 3861; and Haas et al., 1992, Nature, 357: 500); a calcium dependent cathepsin G-like serine protease or a metal dependent cysteine protease (Razzaboni et al., 1992, Brain Res., 589: 207; and Abrahams et al., 1991, An. N.Y. Acad. Sci., 640: 161); Calpain I (Siman et al., 1990, /. Neuroscience, 10: 2400); a multicatalytic protease (Ishiura et al., 1989, FEBS. Lett., 257: 388); a serine protease
• SUBSTITUTE SHEET (Nelson et al., 1990, /. Biol. Chem., 265: 3836); thrombin (Igarashi et al., 1992, Biochem. Biophys. Res. Comm., 185: 1000); or a zinc metallo-peptidase (WIPO application, WO 92/07068 by Athena Neurosciences, Inc.).
• the present disclosure describes a method which identifies some of the APP processing enzymes with specific assays based on the proteolytic degradation of recombinant APP in combination with immunochemical detection of the reaction products.
• the assays of the present invention identify human brain proteases that possess the correct specificity and appropriate localization to play a role in the formation of beta-amyloid from the APP.
• One goal of the presently disclosed invention is to provide a method for discovering drugs that can be used to treat AD patients. As stated previously, the
• SUBSTITUTE SHEET proteolytic degradation of APP to yield the 39 to 42 amino acid peptide beta- amyloid is the first step in the pathophysiological process of amyloid plaque formation.
• lines of evidence point to a causative role of beta-amyloid and the amyloid plaques in the neuro-degeneration characteristics found in the AD brain. These include:
• proteolytic conversion of APP to beta-amyloid appears to be an essential step in the pathogenesis of AD and, as such, an important target for therapeutic intervention.
• Identification of the relevant protease activities, as well as the development of suitable in vitro screening assays, are therefore essential prerequisites for the development of therapeutic protease inhibitors that could be used as treatments to block amyloid plaque formation in AD patients.
• the present invention relates to two developments which can be used to discover inhibitors of proteolytic beta-amyloid formation:
• An in vitro assay comprising a holo-APP substrate and either a highly purified protease that degrades APP or a crude biological extract containing unidentified proteases that can degrade APP;
• the assay enables the detection of in vitro APP degradation activity to yield C-terminal APP fragments.
• the assay can be used to monitor the purification of, or to characterize the protease responsible for the detected activity.
• the assay when used with either a purified protease or a crude biological extract containing unidentified APP degrading enzyme activities, the assay can be used to measure the inhibition of the APP processing activity by chemical or biological compounds that are co-incubated in the assay mixture. Inhibitory compounds thereby identified can have application as therapeutic inhibitors of the in vivo amyloid plaque formation characteristic of AD patients.
• Proteases identified according to (2) above include the aspartic protease, cathepsin D and a chymotryptic-like serine protease distinct from cathepsin G and inhibited by N-tosyl-L-phenylalanine-chloromethylketone ("TPCK”) and alpha-2 antiplasmin and chymotrypsin inhibitor II from potato.
• TPCK N-tosyl-L-phenylalanine-chloromethylketone
• alpha-2 antiplasmin and chymotrypsin inhibitor II from potato.
• the identification of cathepsin D is particularly significant. We show that cathepsin D is able to form C-100-like and beta-amyloid-like fragments of 10.0 kDa and 5.6 kDa size, respectively.
• cathepsin D as well as the design of specific aspartic protease inhibitors
• identification of cathepsin D as an amyloidogenic protease enables both the development of specific cathepsin D inhibitors using established methods, as well as the utilization of established cathepsin D inhibitors.
• cathepsin D unexpectedly, hydrolyzes APP at the peptide bond between Glu(593)-Val(594) (numbering according to Kang et al., supra).
• the preferred specificity of cathepsin D is, ordinarily, between hydrophobic residues. This information can be used further in the design of
• inhibitory compounds thereby identified have application as therapeutic inhibitors of the in vivo amyloid plaque formation characteristic of AD patients.
• APP degrading enzymes identified by the use of the present invention can be purified and used to:
• APP degrading enzyme inhibitors identified by the use of the present invention are also useful, for example, as ligands in the purification of the APP degrading enzymes by affinity chromatography.
• the column will normally be packed with an inert matrix, e.g., agarose, to which the enzyme inhibitors have been attached, if necessary indirectly through a hydrocarbon spacer arm.
• the composition containing the enzyme is then applied to the column, and the enzyme is trapped by the inhibitors while all other proteins pass through and are discarded.
• the enzyme can then be liberated from the column either by eluting with a deforming buffer at a pH which changes the characteristics of the enzyme and no longer allows it to bind to the inhibitor, or by the use of a competitive counter-ligand, which displaces the inhibitor.
• Assays incorporating synthetic peptide substrates are useful for in vitro enzymological studies of highly purified protease preparations, but are generally of insufficient specificity to enable the selective detection of a desired protease activity in crude biologic extracts containing a plethora of proteases.
• brain tissue is abundant with a wide and varied range of peptide processing and degrading enzymes, which may explain why efforts to isolate specific brain APP
• Example 3 it is shown that synthetic peptide assays lead to the identification of several peptidases which are unable to degrade APP to yield C-terminal fragments under the specified assay conditions, and that the pattern of APP degrading proteases does not resemble in any way the corresponding pattern of brain peptidases.
• a more definitive approach to this problem is the utilization of holo-APP as a substrate, in conjunction with a method of assessing its specific degradation following incubation with protease containing fractions.
• the present disclosure describes such a method, wherein the enzymic degradation of recombinant APP by brain protease fractions is monitored by immunoblot using antibodies to the C-terminal region of APP.
• Our assay procedure focuses on the formation of C-terminal fragments from APP of size sufficient to include the full length beta-amyloid peptide (a process requiring endoproteolysis, N-terminal to the A4 region).
• Non-AD control or AD Human brain tissue (non-AD control or AD) is homogenized and then sub-fractionated into a soluble fraction (hereinafter "S"), a post 15,000 g pellet (hereinafter “P-2”), and a microsomal fraction (hereinafter “M”) using conventional ultracentrifugation.
• S soluble fraction
• P-2 post 15,000 g pellet
• M microsomal fraction
• the membranous M and P-2 fractions are solubilized with a Triton X-100 preparation.
• the resulting soluble fractions from M and P-2, as well as the S fraction are then separately subjected to chromatography on a Mono-Q strong anion exchange column which results in separation of different brain proteases.
• peptidase activity of individual mono-Q fractions from the purification of M, soluble and P-2 fractions is assessed. Contiguous pools of column fractions are made based on the recovery of discrete peaks of peptidase activity.
• the pools of peptidase activity are used to establish assay conditions for the detection of proteolytic degradation of highly pure recombinant APP purified
• SUBSTITUTE SHEET from a transfected CHO cell line.
• An immunoblot assay is developed in which antibodies directed either to the APP C-terminal domain or the beta-amyloid region are used to locate C-terminal APP fragments.
• the assay is used to identify six potentially different proteolytic activities capable of forming APP C-terminal fragments of a size large enough to potentially contain full length beta-amyloid.
• the recovery of APP degrading activity amongst the mono-Q pools is not found to correlate well with the peptidase activity profiles established in step 2.
• Inhibitor studies reveal that the APP degrading activities include both serine and aspartic protease activities.
• peptidase assay for monitoring enzyme purification is abandoned. Larger supplies of recombinant APP are obtained by expression in a baculovirus directed insect cell system, enabling use of the APP degradation assay as the primary method to monitor APP degrading activity during protein purification. A major aspartic protease activity is identified in fractions from the mono-Q purification of the P-2 fraction.
• cathepsin D is shown to hydrolyze holo-APP forming a beta-amyloid-like fragment of 5.6 kDa.
• Aprotinin sepharose affinity chromatography is used to attempt to isolate aprotinin sensitive APP degrading activities identified above.
• a chymotrypsin- like serine protease activity is partially purified that can degrade APP to form specific C-terminal fragments of 11, 14 and 18 kDa, that are shown by immunochemical means to contain full length beta-amyloid.
• APP substrate shall mean full length APP, whether derived by isolation or purification from a biological source or by expression of a cloned gene encoding APP or its analogs, and fragments of any such protein, including fragments obtained by digestion of the protein or a portion thereof,
• APP substrates for the assays of the present invention can be provided as a test reagent in a variety of forms. Although preferably derived from, or corresponding at least in part with the amino acid sequence of, APP 695, derivatives or analogs of other APP isoforms (supra) are contemplated for use in the present method as well.
• APP 695 can be obtained by biochemical isolation or purification from natural sources such as described in Schubert et al., 1989, Proc. Natl. Acad. Sci. USA, 86: 2066; or by expression of recombinant DNA clones encoding the protein or a functional portion thereof (Knops et al., 1991, /. Biol. Chem., 266: 7285; and Bhasin et al., 1991, Proc. Natl. Acad. Sci. USA, 88: 10307).
• the fragments of the APP protein will comprise a sequence of amino acids sufficient for recognition and cleavage by the pertinent proteolytic test sample activity (supra).
• Isolation of APP from biological material usually will involve purification by conventional techniques such as chromatography, particularly affinity chromatography. Purified APP or fragments thereof can be used to prepare monoclonal or polyclonal antibodies which can then be used in affinity purification according to conventional procedures. Resulting purified APP material can be further processed, e.g., fragmented, by chemical or enzymatic digestion. Useful fragments will be identified by screening for desired susceptibility to the pertinent proteolytic test sample activity (supra).
• the APP substrate can also be prepared by expression of recombinant DNA clones coding for APP or a portion thereof.
• the cloned APP gene may itself be natural or synthetic, with the natural gene obtainable from cDNA or genomic libraries using degenerate probes based on known amino acid sequences (Kang et al., 1987, Nature, 325: 733). Other techniques for obtaining suitable recombinant DNA clones, as well as methods for expressing the cloned gene, will be evident to the worker in the field.
• SUBSTITUTE SHEET be applied to this step without departing from the inventive features hereof.
• any method can be used for this purpose which is capable of detecting the occurrence of proteolytic cleavage of the APP substrate.
• Such can be afforded by appropriate design of the APP substrate such that cleavage produces a signal producing species, e.g., an optically responsive product such as a colored or fluorescent dye.
• cleavage product is preferentially a C-terminal fragment of the APP substrate; however, any fragment which appears upon incubation with samples can be the object of detection.
• the detection of one or more cleavage products characteristic of the pathologic proteolytic activity can be accomplished in many ways.
• One such method involves the procedure commonly known as Western blot.
• gel electrophoresis is performed to separate the components resulting in the reaction mixture.
• the separated protein components are then transferred to a solid matrix such as a nitrocellulose membrane.
• An antibody specific to a fragment characteristic of APP degradation is then reacted with the components fixed to the membrane and detected by addition of a secondary enzyme-labeled antibody conjugate. The location of the resulting bound conjugate is developed with a chromogenic substrate for the enzyme label.
• a variety of immunoassay formats which are amenable to currently available test systems can also be applied to the detection of APP fragments.
• the APP substrate will be incubated with the test sample and resulting intact APP rendered immobilized (such as by capture onto a solid phase), or alternatively, the test sample is incubated with an immobilized form of the APP substrate.
• Proteolytic cleavage is then detected by reacting the immobilized APP substrate with an antibody reagent directed to a portion of the APP substrate which is cleaved from the APP substrate or which defines the cleavage site.
• the antibody reagent can comprise whole antibody or an antibody fragment comprising an antigen combining site such as Fab or Fab', and can be of the monoclonal or polyclonal type.
• the detection of antibody reagent bound to the immobilized phase is indicative of the absence of the characteristic proteolytic cleavage. Conversely, loss of antibody binding to the immobilized phase is indicative of APP cleavage.
• the detection of binding of the antibody reagent will generally involve use of a labeled form of such antibody reagent or the use of a second, or anti-(antibody), antibody which is labeled.
• Capture or immobilization of APP can be accomplished in many ways.
• An antibody can be generated specific to an epitope of APP which is not on the cleavable fragment. Such an antibody can be immobilized and used to capture or immobilize intact APP.
• a ligand or hapten can be covalently attached to APP and a corresponding immobilized receptor or antibody can be used to capture or immobilize APP.
• a typical ligand:receptor pair useful for this purpose is biotin:avidin. Examples of haptens useful for this purpose are fluorescein and digitoxigenin.
• the solid phase on which the APP substrate is immobilized or captured can be composed of a variety of materials including microtiter plate wells, test tubes, strips, beads, particles, and the like.
• a particularly useful solid phase is magnetic or paramagnetic particles.
• Such particles can be derivatized to contain chemically active groups that can be coupled to a variety of compounds by simple chemical reactions.
• the particles can be cleared from suspension by bringing a magnet close to a vessel containing the particles. Thus, the particles can be washed repeatedly without cumbersome centrifugation or filtration, providing the basis for fully automating the assay procedure.
• Labels for the primary or secondary antibody reagent can be selected from those well known in the art. Some such labels are fluorescent or chemiluminescent labels, radioisotopes, and, more preferably, enzymes for this purpose are alkaline phosphatase, peroxidase, and ⁇ -galactosidase. These enzymes are stable under a variety of conditions, have a high catalytic turnover rate, and can be detected using simple chromogenic substrates.
• Proteolytic cleavage of the APP substrate can also be detected by chromatographic techniques which will separate and then detect the APP
• the present invention is also directed to a method of treating a patient suffering from AD comprising administering to such patient an amount effective therefor of an inhibitor of an aspartic protease alone or in admixture with a non-toxic, inert pharmaceutically acceptable excipient.
• the present invention further relates to pharmaceutical formulations which contain such inhibitors of an aspartic protease in admixture with a non- toxic, inert pharmaceutically acceptable excipient.
• the present invention also includes such pharmaceutical formulations in dosage units.
• the formulations are present in the form of individual parts, for example, tablets, dragees, capsules, caplets, pills, suppositories and ampoules, the inhibitor content of which corresponds to a fraction or a multiple of an individual dose.
• the dosage units can contain, for example, 1, 2, 3 or 4 individual doses or 1/2, 1/3 or 1/4 of an individual dose.
• An individual dose preferably contains the amount of active compound which is given in one administration and which usually corresponds to a whole, 1/2, 1/3 or 1/4 of a daily dose.
• non-toxic, inert pharmaceutically acceptable excipients there are to be understood solid, semi-solid or liquid diluents, fillers and formulation auxiliaries of all types.
• Preferred pharmaceutical formulations which may be mentioned are tablets, dragees, capsules, caplets, pills, granules, suppositories, solutions, suspensions and emulsions, pastes, ointments, gels, creams, lotions, dusting powders and sprays.
• Tablets, dragees, capsules, caplets, pills and granules can contain the inhibitor in addition to the customary excipients, such as (a) fillers and extenders, for example, starches, lactose, sucrose, glucose, mannitol and silicic acid, (b) binders, for example, carboxymethylcellulose, alginates, gelatin and polyvinylpyrrolidone, (c) humectants, for example, glycerol, (d)
• fillers and extenders for example, starches, lactose, sucrose, glucose, mannitol and silicic acid
• binders for example, carboxymethylcellulose, alginates, gelatin and polyvinylpyrrolidone
• humectants for example, glycerol
• SUBSTITUTE SHEET disintegrating agents for example, agar-agar, calcium carbonate and sodium carbonate, (e) solution retarders, for example, paraffin and (f) absorption accelerators, for example, quaternary ammonium compounds, (g) wetting agents, for example, cetyl alcohol and glycerol monostearate, (h) absorbents, for example, kaolin and bentonite and (i) lubricants, for example, talc, calcium stearate, magnesium stearate and solid polyethylene glycols, or mixtures of the substances listed under (a) to (i) directly hereinabove.
• solution retarders for example, paraffin
• absorption accelerators for example, quaternary ammonium compounds
• wetting agents for example, cetyl alcohol and glycerol monostearate
• absorbents for example, kaolin and bentonite
• lubricants for example, talc, calcium stearate, magnesium stearate and solid polyethylene glyco
• the tablets, dragees, capsules, caplets, pills and granules can be provided with the customary coatings and shells, optionally containing opacifying agents and can also be of such composition that they can release the inhibitor only or preferentially in a certain part of the intestinal tract, optionally in a delayed manner.
• embedding compositions which can be used are polymeric substances and waxes.
• the inhibitor can also be present in microencapsulated form, if appropriate, with one or more of the abovementioned excipients.
• Suppositories can contain, in addition to the inhibitor, the customary water-soluble or water-insoluble excipients, for example, polyethylene glycols, fats, for example, cacao fat and higher esters (for example, Ci 4 -alcohol with Ci 6 - fatty acid), or mixtures of these substances.
• the customary water-soluble or water-insoluble excipients for example, polyethylene glycols, fats, for example, cacao fat and higher esters (for example, Ci 4 -alcohol with Ci 6 - fatty acid), or mixtures of these substances.
• Ointments, pastes, creams and gels can contain, in addition to the inhibitor, the customary excipients, for example, animal and vegetable fats, waxes, paraffins, starch, tragacanth, cellulose derivatives, polyethylene glycols, silicones, bentonites, silicic acid, talc and zinc oxide, or mixtures of these substances.
• the customary excipients for example, animal and vegetable fats, waxes, paraffins, starch, tragacanth, cellulose derivatives, polyethylene glycols, silicones, bentonites, silicic acid, talc and zinc oxide, or mixtures of these substances.
• Dusting powders and sprays can contain, in addition to the inhibitor, the customary excipients, for example, lactose, talc, silicic acid, aluminum hydroxide, calcium silicate and polyamide powder, or mixtures of these substances.
• Sprays can additionally contain customary propellants, for example, chlorofluorocarbons.
• Solutions and emulsions can contain, in addition to the inhibitor, customary excipients, such as solvents, solubilizing agents and emulsifiers, for
• SUBSTITUTE SHEET example water, ethyl alcohol, isopropyl alcohol, ethyl carbonate, ethyl acetate, benzyl alcohol, benzyl benzoate, propylene glycol, 1,3-butylene glycol, dimethylformamide, oils, in particular, cottonseed oil, peanut oil, corn germ oil, olive oil, castor oil and sesame oil, glycerol, glycerol formal, tetrahydrofurfuryl alcohol, polyethylene glycols and fatty acid esters of sorbitan, or mixtures of these substances.
• oils in particular, cottonseed oil, peanut oil, corn germ oil, olive oil, castor oil and sesame oil, glycerol, glycerol formal, tetrahydrofurfuryl alcohol, polyethylene glycols and fatty acid esters of sorbitan, or mixtures of these substances.
• solutions and emulsions can also be in a sterile form which is isotonic with blood.
• Suspensions can contain, in addition to the inhibitor, customary excipients, such as liquid diluents, for example, water, ethyl alcohol or propylene glycol and suspending agents, for example, ethoxylated isostearyl alcohols, polyoxyethylene sorbitol and sorbitan esters, microcrystalline cellulose, aluminum metahydroxide, bentonite, agar-agar and tragacanth, or mixtures of these substances.
• customary excipients such as liquid diluents, for example, water, ethyl alcohol or propylene glycol and suspending agents, for example, ethoxylated isostearyl alcohols, polyoxyethylene sorbitol and sorbitan esters, microcrystalline cellulose, aluminum metahydroxide, bentonite, agar-agar and tragacanth, or mixtures of these substances.
• the formulation forms mentioned can also contain coloring agents, preservatives and smell- and taste-improvement additives, for example, peppermint oil and eucalyptus oil and sweeteners, for example, saccharine or aspartame.
• the inhibitor should be present in the abovementioned pharmaceutical formulations in a concentration of about 0.1 to 99.5%, preferably about 0.5 to 95% by weight of the total mixture.
• the abovementioned pharmaceutical formulations can contain multiple inhibitors, in which case, the total amount of inhibitors in the abovementioned pharmaceutical formulations is about 0.1 to 99.5%, preferably about 0.5 to 95% by weight of the total mixture.
• the inventive formulations can contain other active ingredients in addition to the inventive inhibitors.
• the aforementioned pharmaceutical formulations are prepared in the customary manner by known methods, for example, by mixing the active compound or compounds with the excipient or excipients.
• formulations mentioned can be administered orally, rectally,
• SUBSTITUTE SHEET buccally parenterally (intravenously, intramuscularly or subcutaneously), intracisternally, intravaginally, intraperitoneally or locally (dusting powder, ointment or drops).
• Suitable formulations are injection solutions, solutions and suspensions for oral therapy, gels, pour-on formulations, emulsions, ointments or drops.
• Ophthalmological and dermatological formulations, silver salts and other salts, ear drops, eye ointments, powders or solutions can be used for local therapy.
• inhibitors can furthermore be incorporated into other carrier materials, such as, for example, plastics (e.g., chains of plastic for local therapy), collagen or bone cement.
• the inhibitors Since the site of action is the brain, the inhibitors must pass the blood- brain barrier. This may require in some cases that the lipophilicity of the inhibitor be increased, for example, by conjugation to a lipophilic carrier or by the introduction of lipophilic substituents, e.g., hydrocarbons, e.g., long chain alkyl groups, alkenyl groups, e.g., vinyl, etc. Such modification to increase lipophilicity is conventional and within the skill of the ordinary practitioner in the art. See, e.g., R. B. Silverman, "The Organic Chemistry of Drug Design and Drug Action", 1992, Academic Press, San Diego, particularly pages 361-364, the entire contents of which are incorporated herein by reference.
• any conventional method of accomplishing the increased lipophilicity is contemplated.
• An example of a suitable lipophilic carrier is the reversible redox drug delivery system devised by N. Bodor et al., which is discussed in Silverman, Id., at page 362. See also, N. Bodor et al., 1983, Pharmacol. Ther., 19: 337; and N. Bodor, 1987, Ann. N.Y. Acad. Sci., 507: 289, the entire contents of both of which are incorporated herein by reference.
• the inhibitors in total amounts of about 0.5 to 500, preferably 5 to 100 mg/kg of body weight every 24 hours, if appropriate, in the form of several individual doses, to achieve the desired results.
• An individual dose preferably contains the inhibitor in amounts of about 1 to about 80, in particular 3 to 30 mg/kg of body weight.
• SUBSTITUTE SHEET sufficient to manage with less that the above-mentioned amount of inhibitor, while in other cases, the above-mentioned amount of inhibitor can easily be determined by any expert on the basis of his/her expert knowledge.
• Figures la-lf show two dimensional contour plots of peptidase activities of control compared to AD human cortex subfractions.
• Subfractions were prepared according to Example 1, by ion-exchange (mono-Q) separation of P-2, S and M fractions.
• Enzymatic cleavage of N-dansyl- Ile-Ser-Glu-Val-Lys-Met-Asp-Ala-Glu-Phe-Arg-His-Asp- Asp-Asp-Asp (SEQ ID NO: 1) by Mono-Q fractions was performed as described in Example 3.
• Each plot shows the relative amounts of each fluorescent product (abscissa) obtainable by incubation of each mono-Q fraction (ordinate) under the same incubation conditions.
• the amount of product is represented vertically by contour lines. Greater numbers of contour lines indicate greater amounts of a particular product.
• Figures 2a-2f depict immunoblot analysis of the APP 695 degrading activity associated with selected Mono-Q pools from the ion-exchange separation of M, S or P-2 fractions derived from AD cortex.
• FIG. 2a Activity associated with P-2 pool V: APP was present in each of lanes 2 to 6. C-100 from PMTI 73 (lane 1), no P2-V blank (lane 2), P2-V (lane 3), P2-V plus EDTA (lane 4), P2-V plus methanol (lane 5), and P2-V plus pepstatin A in methanol (lane 6).
• FIG. 2b Activity associated with M pool HI: APP was present in lanes 2 to 7. C-100 from PMTI 73 (lane 1), M-III plus cystatin C (lane 2), M-III plus aprotinin (lane 3), M-III plus captopril (lane 4), M-III plus EGTA (lane 5), M-III plus N-dansyl-Ile-Ser-Glu-Val-Lys-Met-Asp-Ala-Glu-Phe-Arg-His-Asp-Asp-Asp-Asp-Asp-
• Figure 2c Activity associated with S pool I: APP was present in each of lanes 2 through 6.
• Lane 1 contains prestained molecular weight markers.
• Figure 2d Activity recovered in individual mono-Q fractions from the separation of AD P-2: Mono-Q fractions 38 to 43 corresponding to the conductance region in which P-2 pool VII is otherwise observed were individually examined for APP degrading activity. For each fraction, the incubation was carried out both in the absence (-) or presence (+) of recombinant APP 695. The fraction numbers are located on Figure 2. The C-100 standard used was from PMTI 100. Mr indicates molecular size markers.
• FIG. 2e Comparison of the position of migration of C-100 products directed either by PMTI 73 or PMTI 100: The protein product of PMTI 73 (lanes 1 and 3) and PMTI 100 (lanes 2 and 4) are shown in comparison with molecular markers (lane 5).
• Figures 3a and 3b depict results from further purification of P-2 VII pool by gel filtration.
• FIG. 23 Figure 3a: P2-pool VII fractions from Mono-Q 10/10 chromatography were pooled, concentrated to 0.25 ml and applied to a tandem arrangement of two Superose 6HR 10/30 columns equilibrated in 10 mM tris HC1 buffer pH 7.5 containing 150 mM NaCl. Elution was performed at a flow rate of 0.3 ml/min, and column eluent was monitored at 280 run. Fractions (0.24 ml) were collected and subjected to both peptidase activity, and APP degradation assay using the immunoblot. The arrows locate the region of the chromatogram in which the APP degrading activity was recovered. Also shown are the peptidase activities associated with both K-M (closed circles) and M-D (open circles) bond cleavage. Chromatography was performed at 22°C.
• Figure 3b The migration of the APP degrading activity relative to the indicated standard proteins of known molecular weight was used to calculate an Mr apparent of the APP degrading protease which is listed in Example 8.
• FIG. 4 shows Peptide Epitope Mapping of Murine Monoclonal Antibody C286.8A Raised against the Beta-amyloid Peptide.
• Micro-litre plates were coated with 50 ng of synthetic APP 695 (597-638) (beta-amyloid 1-42), blocked, then incubated with 100 ⁇ l of C286.8A (80 ng of IgG) which had been preincubated (60 min at room temperature) in the presence or absence of the indicated concentrations of competitor peptide.
• the peptide 1-7 refers to peptide SEQ ID NO: 1.
• %C 1-0- O.D. (+ competitor) - P.P. blank x lOO O.D. (- competitor) - O.D. blank
• Figure 5 APP 695 processing activity recovered in ion-exchange fractions from the purification of human brain P-2 subfraction. A total of 123 fractions were collected from the column. The first 32 fractions corresponded to the load
• Incubations were performed as follows: Baculo- derived holo-APP 695 (80 nm) was incubated with 5 ⁇ l of each column fraction in a total of 15 ⁇ l containing 100 mM Mes buffer pH 6.5, 0.008 % (v/v) Triton X- 100, 160 mM NaCl, 6.7 mM tris (from the APP stock). Reactions were terminated after 24 h by addition of SDS-PAGE sample buffer. Immunoblots were developed using the C-terminal polyclonal antiserum of Example 6, as described in Example 8. The arrows locate the product fragments. Fractions 45 to 86 were also tested but showed relatively little activity (therefore not shown). Peaks A and B locate the major activities.
• Figure 6 shows results of purification of P-2 derived APP degrading activity on gel filtration: correlation with the elution of cathepsin D.
• Panel (a) a 280 nm elution profile for the purification of P-2 peak B on a superose 6HR column.
• Panels (b) and (c) corresponding APP C-terminal processing activity in the eluted fractions 49 to 60, determined essentially as described in Example 5. Arrows locate major product bands.
• Panels (d) and (e) immunoblot analysis of eluted fractions using a rabbit polyclonal antibody to cathepsin D (1/300, dilution). The arrows locate the position of migration of immunoreactive bands. Human liver cathepsin D was also analyzed for comparison.
• Figure 7 depicts protease inhibitor specificity of protease activities isolated from the P-2 subfraction. Reactions (32 ⁇ l) were initiated at 37°C by APP addition to achieve the following initial component conditions: P-2 enzyme (2.54 ⁇ g/ml) fraction from the 15-25 kDa region of gel filtration ( Figure 6); APP (168 nM), in 96 mM Mes buffer pH 6.5. Reactions were terminated after 26 hr by addition of 15 ⁇ l of 3X sample buffer, and subjected to immunoblot analysis using a 1/1000 dilution of the rabbit polyclonal antiserum to the APP C-terminus.
• P-2 enzyme (2.54 ⁇ g/ml) fraction from the 15-25 kDa region of gel filtration ( Figure 6); APP (168 nM), in 96 mM Mes buffer pH 6.5. Reactions were terminated after 26 hr by addition of 15 ⁇ l of 3X sample buffer, and subjected to immunoblot analysis using
• Figure 8 shows time course of Cathepsin D catalyzed APP cleavage monitored using an antibody to the APP C-terminal domain.
• Panel (a) shows time course of APP proteolysis by cathepsin D in the absence (lanes 10-14) or presence (lanes 4-8) of 86 ⁇ M pepstatin A. APP was also incubated alone (lanes 1- 3). The numbers indicate the time (hr) after initiation of reactions. Reactions were initiated at 37°C by the addition of APP to achieve the following initial component concentrations: APP (82 nM), cathepsin D (9.2 ⁇ g/ml), in 89 Mes buffer pH 6.5.
• Figure 9 depicts pH and ionic strength dependence of APP C-terminal processing by the P-2 derived enzyme or cathepsin D.
• Panel (a) shows pH dependence observed with cathepsin D and the P-2 enzyme (peak B, Figure 5 following gel filtration chromatography, Figure 6).
• Panel (b) is ionic strength dependence for both enzymes at pH 6.5. Reactions were initiated at 37°C by enzyme addition to achieve the following initial component concentrations: cathepsin D (9.2 ⁇ g/ml) or P-2 enzyme from Figure 6 (11.7 ⁇ g/ml), 100 mM in each of sodium acetate, Mes, and Tris-HCl, and purified APP (79 nM).
• Figure 10 shows cleavage of N-Dansyl-Ile-Ser-Glu-Val-Lys-Met-Asp-Ala- Glu-Phe-Arg-NH 2 (SEQ ID NO: 7) by cathepsin D and the P-2 enzyme (peak B, Figure 5) following further purification on Superose 6HR.
• Figure 11 depicts pH dependence of N-Dansyl-Ile-Ser-Glu-Val-Lys-Met- Asp-Ala-Glu-Phe-Arg-NH 2 (SEQ ID NO: 7) cleavage by cathepsin D and the P-2 enzyme (peak B). Reaction conditions were essentially as described in Figure 10, except that the cocktail buffer was adjusted to the indicated pH values and the incubation times were 23 hours for the P-2 enzyme and five hours for cathepsin D. (a) Cleavage by cathepsin D, and (b) cleavage by P-2 enzyme. Rates of cleavage at the -Glu-Val- (closed circles) and -Met-Asp- (open circles) bonds are shown in each case.
• Figure 12 is SDS-PAGE analysis of reaction products from the preparative digestion of APP by cathepsin D.
• Panel (a) is a photograph of the coomassie stained electroblot prior to excision of bands
• panel (b) is the corresponding blot after band excision
• panel (c) is the corresponding immunoblot analysis (1/100 dilution of monoclonal 286.8A) of a parallel series of reactions to those depicted in panels (a) and (b).
• Figure 13 shows a time course of cathepsin D catalyzed APP degradation monitored using a monoclonal antibody to the N-terminus of beta-amyloid.
• Figure 14 shows the effect of amino acid substitution on the time course of hydrolysis of synthetic peptides by cathepsin D and the P-2 derived enzyme (peak B). Reactions were initiated at 37°C by substrate addition to achieve the following initial component concentrations: Cathepsin D (2.5 ⁇ g/ml) or P-2 peak B enzyme (7.5 ⁇ g/ml); N-dansyl-peptide (58 ⁇ M), captopril (0.3 mM), with or without pepstatin A (213 ⁇ M), sodium chloride (75 mM), in 135 mM buffer in each of tris, Mes and acetate buffer pH 5.0. At various times, aliquots (30 ⁇ l) were removed, adjusted to 12.5% (v/v) in TFA and subjected to RP-HPLC analysis. Time course of hydrolysis for the following substrate/protease combinations are shown:
• Figure 15 shows purification of solubilized P-2 fraction on aprotinin
• Figure 16 shows purification of P-2 derived aprotinin binding protease on mono-Q.
• (a) shows A280 nm elution profile
• (b) activity of eluted fractions using a rabbit anti-C-terminal APP antiserum (1/1000 dilution) for detection
• (c) activity of eluted fractions using Mab 286.8A. (1/100 dilution of 1.6 mg/ml pure IgG) for detection.
• the three arrows indicate the migration of the 11, 14 and 18 kDa C-terminal APP degradation products.
• Figure 17 shows pH and ionic strength dependence of APP degradation catalyzed by the pool Y serine protease.
• Lanes 1 and 12 contain prestained Mr markers.
• Lane 9 contained C-100 and lanes 10 and 11 contain APP incubated for 0 and 3 hr respectively at pH 6.5 in the absence of pool Y.
• Panel (b) shows ionic strength dependence. Reactions were performed essentially as described in (a) except the buffer was 95 mM Mes pH 6.5 containing the indicated molar concentrations of sodium chloride. APP cleavage in the absence (lanes 2 to 7) or presence (lanes 9 to 14) of pool Y are shown for each concentration of sodium chloride.
• Lanes 1 and 8 contain Mr markers and C-100 standard respectively. The arrows indicate the migration of the 11, 14 and 18 kDa APP fragments.
• Figure 18 depicts inhibitor selectivity of the pool Y protease. Reactions were initiated at 37°C by enzyme addition to achieve the following initial component concentrations in a 16 ⁇ l volume: pool Y # 3-5 (14 ⁇ g/ml) after purification on a superdex 75 column, APP (35 nM), in 30 mM Mes buffer pH 6.5. Reactions were terminated by addition of 7.5 ⁇ l of 3X sample buffer. Immunoblots were developed using the rabbit antiserum to the APP C-terminal
• SUBSTITUTE SHEET domain according to Example 8. Data are shown for the effect of the following inhibitors: Panel (a) 860 ⁇ M PMSF in methanol (lane 4), 400 ⁇ M pepstatin in methanol (lane 6), 5 mM benzamidine (lane 8), 350 ⁇ M E-64 (lane 9), 7.7 mM EDTA (lane 10), 15 ⁇ M aprotinin (lane 11), and 0.1% (w/v) deoxycholate (lane 15). The following controls were also run: no inhibitor (lane 12), ethanol (lane 3), methanol (lane 5), pool Y only (lane 2), APP only at time zero (lane 13) and 4 hr (lane 14).
• Lanes 1 and 7 respectively show prestained Mr markers and the C-100 standard.
• Panel (b) 1.8 ⁇ M alpha-1-antichymotrypsin (lane 2), 156 ⁇ M TLCK (lane 3), 46 ⁇ M chymotrypsin inhibitor I (lane 4), 119 ⁇ M chymotrypsin inhibitor II (lane 5), 4 ⁇ M alpha-2-antiplasmin (lane 6), 51 ⁇ M alpha-1-antitrypsin (lane 7), 98 ⁇ M chymostatin administered in DMSO (lane 10), 153 ⁇ M methanolic TPCK (lane 12). Controls included: no inhibitor (lane 8), DMSO (lane 11), and methanol (lane 13). Pre-stained molecular weight markers and C-100 standards were applied to lanes 1 and 9 respectively. The arrows indicate the migration of the 11, 14 and 18 kDa APP degradation products.
• Figure 19 shows 1) that neither DMSO nor DMSO plus 10 uM pepstatin A effect growth of HEK 293 cells (panel A), 2) conditioned media harvested from late log phase cultures treated with DMSO plus pepstatin A shows significantly lower levels of a 15 kDa APP C-terminal fragment than cultures treated with DMSO only (panel B).
• Panel A HEK 293 cells (ATCC CRL 1573, adapted for suspension culture) were seeded (1 X 10 5 /ml final) in roller bottles containing 400 ml of MoAb medium (JRH Biosciences, Lenexa, Kansas) plus 0.2 % v/v fetal calf serum (closed squares).
• Additional rollers also contained 0.01 % v/v DMSO (open squares) or 0.01 % DMSO plus 10 uM pepstatin A (closed diamonds).
• Cell growth was at 37°C in 5 % C0 2 /95% air. Viable cell numbers in trypan blue treated samples were quantified with a hemocytometer, and were a constant percentage (60%) with time.
• Conditioned medium was harvested at the end of log phase (located by the arrow at day 7), centrifuged at 1500 RPM in a Beckman Gs-6 bench top centrifuge and subjected to chromatography on columns of immobilized anti-beta amyloid monoclonal antibody (C286.8A).
• Panel B Immunoblot with Rabbit anti-APP C-terminal antiserum.
• Lane 1 prestained molecular weight markers;
• lanes 2-7 inclusive are the respective analyses of fractions 1-6 inclusive from the purification of medium from DMSO treated cells;
• lanes 9-14 inclusive are the respective analyses of fractions 1-6 inclusive from the purification of medium from the DMSO /pepstatin A treated
• Lane 8 contains C-100 standard (example 5, from PMTI 100), and lane 15 contains recombinant holo-APP695 purified according to example 7, method 2. Note that fractions 5 and 6 from the DMSO only treatment contain a 15 kDa band that is absent from the corresponding fractions in the DMSO /pepstatin A treatment. Further details are given in the text to example 12.
• Figure 20 summarizes the purification from human brain of an aspartic protease with APP processing activity, a) Elution profile for the purification of solubilized P-2 fraction (140 mg protein) on a mono-Q HR 10/10 column. Protein concentration (open circles), conductance (hatched line) and rates of formation of N-dansyl-ISE from N-dansyl-ISEVKMDAEFR-NH 2 in the absence (closed circles) or presence (open triangles) of 10 uM pepstatin A, are shown for each collected fraction, b) Immunoblot assay for APP hydrolysis using MAb 286.8A: lane 1, C- 100 standard; lane 2, pooled fractions 24 to 45 (2.0 to 5.3 mmho at 4°C) from 1(b) incubated alone; lane 3, pooled fractions incubated with APP; lane 4, pooled fractions plus APP plus 10 uM final pepstatin A.
• Beta-amyloid 1-42, 1-28, 1-16 and N-dansyl-ISEVKMDAEFRHDDDD inhibited C286.8A binding dose dependently, whereas 12-28, 25-35 and 645- 695 did not, thus localizing the reactive epitope for C286.8A to the N-terminal 7 amino acids of the A4 region (APP 597-603).
• APP Processing 31 ⁇ g/ml of homogeneous holo-APP 695 (prepared according to Example 7, method 2) was incubated with 5ul of each column fraction in a total of 15 ul containing 40 mM in each of acetate, Mes and tris pH 4.0, 0.002 % (v/v) triton X-100 and 30 mM exogenous NaCl. Reactions were terminated after 24 h by addition of SDS-PAGE sample buffer to IX final, electrophoresed on 10-20 % acrylamide gradient Tricine gels (Novex) at 100 V constant and then electroblotted onto Problott membrane (Applied Biosystems). Blots were developed with monoclonal antibody C286.8A at 50 ⁇ g/ml final by standard sandwich procedures using appropriate second antibodies coupled to alkaline phosphatase. Recombinant C-100 was prepared as described in the methods.
• SUBSTITUTE SHEET Peptidase Aliquots of column fractions (20 ⁇ l), were added to 10 ⁇ l of reaction mixture to achieve the following initial component concentrations: synthetic N-dansyl-ISEVKMDAEFR-NH 2 (58 ⁇ M), captopril (0.2 mM), methanol (0.3 v/v) and when included pepstatin A (10 ⁇ M) in methanol, in a cocktail buffer comprising 50 mM sodium acetate pH 5.. Reactions were terminated after 24 hr by addition of 10 ⁇ l of 40 ⁇ M pepstatin A. Enzymic products were detected and quantified by RP-HPLC (Example 2).
• Figure 21 shows that immobilized antibodies to human cathepsin D selectively remove the human brain APP degrading activity from solution, a)
• APP processing activities are shown for selected void fractions (1 to 27) from the anti-CD (+) or control (-), as well as for the applysate (Q-pool), purified cathepsin D (CD) or APP alone.
• APP processing Homogeneous holo-APP 695 (31 ⁇ g/ml final) was incubated with 5 ⁇ l of each column fraction in a total of 15 ⁇ l containing 40 mM in each of sodium acetate, Mes and Tris adjusted to pH 4.0, 40 mM NaCl, and 0.002 % (v/v) exogenous triton X-100. Reactions were terminated after 24 hr by addition of SDS-PAGE sample buffer, and analyzed by immunoblot developed with the monoclonal antibody 286.8A. Arrows locate the product fragments. Highly pure human cathepsin D was present in enzymic incubations at 1.27 ⁇ g/ml final.
• Figure 22 shows immobilized antibodies to cathespin D adsorb a human brain peptidase that degrades APP mimetic peptides.
• a) and b) respective peptidase activities in the flow through fractions and pooled glycine /triton eluent pool (#80-89, figure 21a, concentrated 21 fold) in the degradation of N- dansyl-ISEVKMDAEFR-NH 2 .
• E-V hydrolyzing activity (circles) of flow through fractions is shown.
• c) L-D hydrolysis is shown.
• Activities are depicted from fractions recovered from columns containing coupled control (closed symbols) or anti-cathepsin D IgG (open symbols). Triangles show effect of 10 ⁇ M pepstatin A on amount of uncontroverted substrate remaining at the end of incubations. In b) and d) activities were measured in the absence (open histograms) or presence (closed histograms) of pepstatin A.
• Figure 23 shows an updated summary of sequence assignment of beta- amyloid fragments formed from the hydrolysis of APP 695 by cathepsin D made since initial assignments reported in Table 5.
• cathepsin D Example 9
• Example 9 One of the six activities was subsequently identified as cathepsin D (Example 9), and was further characterized according to its chromatography on gel filtration.
• Example 2 In an alternate approach to attempt to affinity purify some of the human brain serine proteases described in Example 8 (Table 4), "Method 2" was implemented. This procedure was based on the affinity purification of serine proteases using aprotinin sepharose at an early step, and lead to the identification of a serine protease(s) (Example 10), which also exhibited the capacity for APP C- terminal processing.
• the loose pellet was removed, re-homogenized and centrifuged as described above.
• the supernatant for each extraction was combined and centrifuged at 15,000 g x 30 min in the Sorval SS-34 rotor.
• the resultant "P-2" pellet was resuspended in 100 ml of ice cold 0.32 M sucrose by vortexing and stored at -70°C.
• the supernatant from the last spin was centrifuged at 105,000 g x 60 min to yield the supernatant or soluble fractions ("S”), and the microsomal fraction ("M”) which was resuspended in 60 ml of 0.32 M sucrose. Both S and M were stored at -70°C.
• Protein fractions of P-2, microsomal (M), or soluble (S) were loaded onto the column and equilibrated with 50 mM Tris HCl, pH 7.5 (conductivity 1.8 mU at 4°C) at a flow rate of 2 ml/min. The column was then washed with equilibration buffer until the A280 nm in the eluent decreased to zero whereupon the column flow rate was increased to 4 ml/min.
• microsomal fraction (47 mg) described above was applied to a column of aprotinin sepharose (Sigma, catalog # 42268, 1.5 x 10 cm), previously equilibrated with 20 mM Tris HCl buffer pH 7.0. Once loaded the column was washed with equilibration buffer (100 ml), and then eluted with 60 ml of 50 mM sodium acetate buffer pH 5.0 containing 500 mM sodium chloride. The flow rate was 1.0 ml/min throughout.
• Eluted fractions (4 ml) were monitored at 280 nm, analyzed using the Bradford protein assay, and examined for APP C-terminal processing activity as described in Example 8, using recombinant APP derived by expression in a baculo virus system. Active fractions were capable of forming 11, 14 and 18 kDa (approx.) fragments which were detectable on immunoblots using an anti-APP C-terminal antibody (see Example 6 for method of antibody production).
• the pools comprised the following conductance
• STITUTE SHEET ranges: pool X (12.2 to 14.4 mmho), pool Y (14.9 to 18.9 mmho) and pool Z (20.2 to 22.9 mmho).
• the gel filtration was calibrated by chromatography of each of the following standard proteins: Thyroglobulin (670 kDa), bovine gamma globulin (158 kDa), ovalbumin (44 kDa), myoglobin (17.5 kDa) and vitamin B i2 (1.35 kDa).
• a peptidase assay was developed to enable the high throughput detection of endoproteases in human brain tissues which might possess a specificity appropriate for APP hydrolysis at the junction between the "extracellular" domain(s) and the N-terminus of the beta-amyloid peptide region.
• the technology selected utilized dansylated peptide substrates, in conjunction with subsequent detection of fluorescent peptide products by RP-HPLC separation, and post column fluorescent detection.
• the peptides were cleaved and deprotected in 90% trifluoroacetic acid, 4% thioanisole, 2% ethanedithiol, and 4% liquefied phenol for 2 h at room temperature.
• This latter peptide was relatively insensitive to carboxy peptidase digestion even in the presence of crude tissue fractions and was used in the peptidase profiling studies of Example 3.
• the C-terminal alpha amide substrate (SEQ ID NO: 7) was used in the peptidase studies of Example 9 using more purified enzyme fractions. The degradation of either of the peptides was monitored using the HPLC protocol of Example 3, below.
• HPLC quantification of proteolytic products HPLC analysis was performed using a Hewlet-Packard HP1090 complete with binary solvent delivery, heated column compartment, and auto injector. Fluorescence detection (post column) was performed with an in-line Gilson model 121 filter fluorometer (excitation at 310-410 nm, emission at 480-520 nm) in conjunction with an HPLC chem-station (DOS series) and suitable software for data analysis.
• the retention time of certain metabolites listed in Table 2, above differ to those quoted in Example 2 for the cleavage of N-dansyl-Ile-Ser- Glu-Val-Lys-Met-Asp-Ala-Glu-Phe-Arg-His (SEQ ID NO: 2) due to variables in the HPLC set up.
• the studies which reflect data listed in Table 2 have relatively longer retention times because chromatography was performed using a guard column in line with the HPLC column.
• HPLC column was calibrated for day to day variation in the retention times of the enzymatically generated products by analysis of synthetic product standards in parallel with the experimental samples.
• Data for the proteolytic metabolite profile of individual ion-exchange fractions was collected using the HP CHEM station data acquisition software.
• SpyGlass takes this array and transforms it into a three-dimensional surface in which Mono-Q fraction number, cleavage site and area % for the product formed are the three axes.
• Contours are defined according to the following criteria set manually within the SpyGlass Program: the first contour line connects contiguous regions of the plot where 1.5% substrate conversion to the particular product was observed. Similarly successive contour lines connecting regions of 5%, 10%, 20%, 30%, 40% and 50%, substrate conversion were also displayed.
• the resulting contour plots represent brain peptidase maps in which fraction numbers span the ordinate, and peptide bond cleavage sites are on the abscissa, and the amount of product formed is represented by the contours.
• Figure 1 shows a comparison of the peptidase profiles obtained for the cleavage of the N-dansyl peptide substrate by both control and AD P-2, S and M fractions subjected to further subfractionation by ion-exchange chromatography. The analysis enables the identification of a high number of potentially different peptidase activities throughout the subfractions of control and AD brain.
• each peptidase pool was concentrated using an Amicon Centriprep-10 membrane. Prior to concentrating, each Centriprep membrane was washed in 50 mM Tris/HCl, pH 7.5. A 15 ml Centriprep was used for peptidase pools that contained a volume of 5 ml or more. These pools were concentrated on a Du Pont Sorval tabletop centrifuge at 2700 rpm for approximately 40 min. Pools that contained less than 4 ml were concentrated with a 2 ml Centriprep on a Du Pont
• each concentrated pool was washed with an equal volume of 50 mM Tris/HCl, pH 7.5. Pools were maintained at 4°C throughout the concentrating process and stored at -70°C. The pools are listed below in Table 3 with the conductivities and the final concentration volumes.
• This example describes the method for expressing holo-APP 695 which was then purified as described in Example 7 and then used as the recombinant substrate for the APP degradation assay described in Example 8 through 10. Two approaches were used.
• APP 695 was obtained by expression using a baculovirus directed system.
• the greater amounts of APP 695 thereby generated made it feasible to perform the more detailed studies outlined in Examples 9 and 10, leading to the identification of certain APP degrading enzymes.
• Method 1 Development of a Chinese Hamster Ovary (CHO) cell line expressing holo-APP 695
• APP 695 cDNA from FC-4 was filled in by the large fragment of E. coli DNA polymerase I and blunt-end inserted into the Smal cloning site of KS Bluescript M13+ (Stratagene Cloning Systems, La Jolla, CA) creating pMTI-5 (APP 695 under the T3 promoter).
• KS Bluescript M13+ Stratagene Cloning Systems, La Jolla, CA
• pMTI-5 APP 695 under the T3 promoter.
• a new optimal Kozak consensus DNA sequence was then created using site-specific mutagenesis (Kunkel et al., 1987, Methods in Enzymology, 154: 367) with the oligo:
• the full length APP cDNA containing the optimal Kozak consensus sequence and Val to Glu mutation was then cut out of PMTI-39 with NotI and a Hindlll partial digest.
• the 2.36 Kb APP 695 fragment was then gel purified and ligated into Notl/Hindlll cut pcNAINeo (Invitrogen Corp., San Diego, CA) to create PMTI 90 in which the APP 695 expression is placed under the control of the CMV promoter.
• the Val to Glu mutation was sequence confirmed and the vector used to stably transform CHO cells.
• Chinese Hamster Ovary K-l cells (ATCC CCL 61) were used for transfection with the APP 695 construct. Twenty micrograms of an expression plasmid containing APP 695 and a neomycin drug resistance marker was transfected into 1 x 10 7 CHO cells in 0.5 ml PBS by electroporation using a Bio-Rad Gene Apparatus (Bio-Rad Laboratories, Richmond, CA). A single pulse of 1200 V at 25 ⁇ f capacitance was administered to the cells.
• cells were incubated in ice for 10 minutes and collected by centrifugation. The cell pellet was resuspended in Alpha MEM, 10% fetal calf serum at a density of 5 x 10 4 cells/ml, and 1 ml aliquots were distributed into each well of five 24-well tissue culture cluster plates. After 48 hours incubation, cells containing the neomycin drug resistance marker were selected by addition of 1 ml of media containing 1 mg/ml Geneticin (GIBCO-BRL, Grand Island, NY) and incubation was continued and bi-weekly changes of drug containing media.
• Geneticin GEBCO-BRL, Grand Island, NY
• Drug resistant cells were tested for APP 695 expression by Western blotting.
• Cells positive for APP 695 expression were cloned by limiting dilution, and individual clones were isolated and tested for APP 695 expression.
• a clone positive for APP 695 expression was subcultured and expanded into roller bottles for large scale production of APP 695 expressing cells and subsequent isolation of recombinant protein.
• the C-100 peptide fragment contains the C-terminal portion of APP which spans from the N-terminus of the A4 peptide to the C-terminus of full length APP (see above, BACKGROUND section).
• the C-100 fragment is the purported initial degradation product leading to the ultimate formation of the A4 peptide.
• cell lysates from Hela S3 cells (ATCC CCL 2.2) expressing recombinant C-100 are analyzed in the immunoblot assay in parallel with the recombinant APP samples that have been incubated with brain fractions, sub-fractionated by Mono-Q chromatography (See Example 3).
• the migration and detection of the C-100 fragments serves both as a
• Comparison of the size of enzymatically generated products with the size of the C-100 fragment can provide insights into whether or not the enzymatically generated fragments result from cleavage close to the N-terminus of the A4 peptide or alternatively within the A4 segment as would be catalyzed by secretase.
• Plasmid construction Two methods were used to make plasmids for C-100 expression. Each plasmid shall be identified separately as either PMTI 73 or PMTI 100.
• PMTI 73 construction The commercially available plasmid PUC-19 was digested with EcoRI to eliminate its polylinkers. Commercially available PWE16 was then inserted into the digested PUC-19 to create PMTI 2300.
• PMTI 2301 was derived from PMTI 2300 following BamHI/Hind III digestion using an oligonucleotide adapter. The EcoRI promoter fragment of APP was inserted into the Hindlll site of PMTI 2301 by blunt end ligation to produce PM ⁇ 2307.
• PMTI 2311 was generated by ligating the BamHI fragment from PC-4 (Kang et al, supra) into the BamHI site of PMTI 2307.
• the Xhol fragment from FC-4 was inserted into the Xhol site of PM ⁇ 2311 to generate PMTI 2312.
• PMTI 2323 was generated by insertion of the 2.2 kb Bglll/EcoRI fragment from the EcoRI genomic clone of the mouse metallothionine-I gene into the Clal site of PMTI 2312.
• minigene PMTI 2337 the sequences between the Kpnl and Bglll sites of PMTI 2323 were deleted and the clone was ligated using synthetic oligonucleotide adaptor, sp-spacer-A4.
• PMTI 2337 was cut with Bam Hl/Spel and the fragment ligated into the Bam Hl/Xbal restriction sites of Bluescript KS (+) (Stratagene) to create PMTI 2371.
• PMTI 2371 was cut Hindlll/NotI to release a 0.7 kb fragment coding for the terminal 100 amino acids of APP 695. Also encoded was the sequence of signal peptide. This insert was ligated into the Hindlll/NotI site of pcDNAINEO (Invitrogen Corp.) to create the plasmid PM ⁇ 73.
• PMTI 100 Construction PMTI 90 (see Example 4, method 1) was cut Xbal/HindlH to release a 0.6 kb fragment again coding for the terminal 100 amino acids of APP 695 and this was ligated to the Xbal/Hindl ⁇ site of pcDNAINEO to create PM ⁇ 100.
• vectors, inserts and plasmids were purified by methods known to those skilled in the art.
• Sufficient vaccinia virus vTF7-3 was trypsin treated to infect at a multiplicity of 20 plaque forming units per cell, mixing an equal volume of crude virus stock and 0.25 mg/ml trypsin, then vortexed vigorously. The trypsin treated virus was incubated at 37°C for 30 minutes, with vortexing at 10 minute intervals. Where clumps persisted, the incubation mixture was chilled to 0°C and sonicated for 30 seconds in a sonicating water bath. The chilled sonication was repeated until no more clumps were detected.
• the trypsin treated virus was then diluted with sufficient serum free DMEM for each well with Hela SI cells to have 0.5 ml of virus. Medium was aspirated away, then the cells were infected with virus for 30 minutes, with rocking at 10 minute intervals to distribute the virus.
• transfection mixture was prepared as follows: To each well was added 0.015 ml lipofection reagent (Bethesda Research Labs, Gaithersburg, MD) to 1 ml OPTIMUM (Bethesda Research Labs, Gaithersburg, MD) in a polystyrene tube, mixing gently. Vortex was avoided. Then, 3 ⁇ g CsCl purified DNA was added and mixed gently.
• Virus mixture was aspirated from cells, then the transfection solution was introduced. The resulting mixture was incubated for three hours at 37°C. Each well was then overlaid with 1 ml of OPTIMUM and incubated at 37°C in a C0 2 incubator overnight.
• SUBSTITUTE SHEET contained 1% Triton X-100, 10 ⁇ g/ml BPTI, 10 ⁇ g/ml Leupeptin, 200 mM NaCl, 10 mM HEPES, 1 mM CaCl 2 , 1 mM MgCl 2 , 1 mM EDTA, adjusted to pH 7.5. Complete lysis was monitored by light microscopy, and harvested immediately. Lysis took less than 1 minute to complete, with delay at this step causing lysis of nuclei resulting in a gelatinous mass.
• Recombinant lysates were stored at -20°C for later use.
• recombinant lysates should be diluted 1:50, in (3X) SDS-PAGE sample buffer which is devoid of 2-mercaptoethanol prior to freezing.
• the largest of the three bands produced by PM ⁇ 73 was slightly larger than the single band observed with PMTI 100. Amino acid sequence analysis of the largest band from PM ⁇ 73 expression showed that the signal peptide sequence was cleaved from the initial translation product to yield a C-100 fragment containing 5 extra amino acids at the N-terminus.
• an affinity purified antibody which recognized the C-terminus of APP was prepared and used to synthesize an immunoaffinity column for the affinity purification of APP expressed in a baculo virus directed system (see Example 7);
• ⁇ APP 645-694 was used to immunize rabbits to elicit polyclonal antibodies.
• Sera were screened by immunoblot analysis of lysates of E. coli that expressed a fusion protein including the amino acids 19 through 695 of human APP 695.
• Sera which were immunoreactive against the recombinant fusion protein were further screened for immunoprecipitating activity against [ 35 S] methionine- labeled APP 695, which was produced from ⁇ APP 645-694 cDNA by successive in vitro transcription (kit purchased from Stratagene, La Jolla, CA) and translation (reticulocyte lysate kit purchased from Promega Corp., Madison, WI).
• the pooled void volume from the gel filtration column (2.9 mg P-142 conjugated to BSA/ 12.5 mis) was coupled to 1 gm (3.5 mis) of CNBr activated sepharose (>90% peptide conjugate coupled by standard Pharmacia protocol). Remaining sites were blocked with ethanolamine.
• the sepharose affinity matrix was packed into a 1.0 x 3.5 cm glass column.
• mice were immunized by multiple injections of a mixture of the following two synthetic peptides: 1) APP amino acids 597 to 638 of holo-APP 695 (numbering according to Kang et al., Id.) containing beta amyloid, and 2) APP 295 amino acids 645-695 containing the C- terminal domain.
• Splenoytes from immunized animals were fused with X63/Ag 8.653 mouse myeloma cells using standard procedures (Herzenberg et al, 1978, In: D.M. Weir (Ed.), Handbook of Experimental Immunology, pp 25.1-25.7, Blackwell Scientific Publications, Oxford, UK).
• Synthetic peptides containing amino acids 597-612, 597-624, 597-638, 608-624, 621-631 and 645-695 of human APP 695 (numbering according to Kang et al, Id.-) as well as N-dansyl-Ile- Ser-Glu-Val-Lys-Met-Asp-Ala-Glu-Phe-Arg-His-Asp-Asp-Asp-Asp (SEQ ID NO: 1) were tested for the ability to block binding of C286.8A to APP 597-638.
• peptides APP 597-612, 597-624, 597-638, and N-dansyl-Ile-Ser-Glu-Val- Lys-Met- Asp- Ala-Glu-Phe-Arg-His-Asp- Asp-Asp-Asp (SEQ ID NO: 1) were able to inhibit C286.8A binding in a dose-dependent fashion.
• These peptides contain respectively amino acids 1-16, 1-28, 1-42 and 1-7 of the beta-amyloid sequence (numbering from the N-terminal aspartate residue).
• Example 7 Purification of recombinant holo-APP 695.
• Holo-APP 695 was detected by immunoblot analysis using an anti-human APP 695 C- terminal antibody essentially as described in Example 8, below.
• the cells were homogenized using a teflon potter (10 return strokes), then layered (25 ml per centrifuge tube) onto 10 ml of homogenization buffer containing 41% sucrose and devoid of the protease inhibitors EDTA, PMSF, E-64 and pepstatin. Following centrifugation (26,800 RPM X 60 min, in a Beckman SW-28 rotor, the interfacial layer was carefully removed (approximately 150 ml in combined volume), diluted with an equal volume of homogenization buffer (minus protease inhibitors), resuspended with a teflon potter (3 return strokes), and recentrifuged as described above to yield a tightly packed pellet.
• Proteins eluting between a conductivity range of 17 to 22 mmho (4°C) contained the majority of immunoreactive APP 695, and were combined and dialyzed for 4 hours versus 2L of 5 mM tris-HCl pH 8.0 containing 0.025% triton X-100, and clarified to remove slight turbidity by centrifugation (26,800 x 60 min in a Beckman SW 28 rotor).
• the APP recovered at 300 mM and 600 mM NaCl were collected separately and stored in aliquots at -80°C.
• the APP used in the following studies were from the 300 mM fraction.
• the yield of partially pure APP from the 300 mM heparin agarose eluent was 5.5 ⁇ g (Bradford assay) per gram of wet CHO cell pellet.
• the APP in the preparation was judged to be about 25% pure based upon SDS PAGE analysis.
• the homogenate was centrifuged (105,000 g X 1 h in a Beckman 70 Ti rotor) and the pellet was then resuspended by teflon potter (10 return strokes) in 160 ml of 10 mM Tris-HCl buffer pH 7.5 containing 0.5 M NaCl and the same inhibitors and concentrations as listed above. After brief sonication (Branson Sonifier Cell, 2 min power level 4), Triton X-100 was then added to a final concentration of 5 % (v/v), and the suspension was gently stirred for 20 min at 4°C.
• the mixture was centrifuged (50,000 RPM X 60 min, in a Beckman Ti 70 rotor), and the first supernatant (574 mg of protein) carefully removed for heparin-agarose chromatography.
• the pellet was resuspended by teflon potter (20 return strokes) in 160 ml of 10 mM tris-HCl buffer pH 7.5 containing 0.5 M NaCl, and each of the inhibitors at the concentrations listed above. Solubilization with 5% (v/v) triton X-100, and subsequent centrifugation was performed as described above to yield a second solubilized supernatant (683 mg of protein).
• Fractions (5 ml) were collected into 0.5 ml each of IM tris-HCl pH 8.0, and monitored for A280 nm, total protein (Bradford assay), and the presence of immunodetectable APP as above. Fractions containing significant APP were combined. The combined heparin agarose eluent was cycled through the affinity purification procedure a total of five times. The APP pool recovered from each successive purification was combined for a total of 9 mg of APP.
• the final preparation of APP was >95% pure based on SDS-PAGE developed with coomassie brilliant blue stain, exhibited an amino acid composition that was within 86% agreement with the theoretical composition, and exhibited the following N-terminal sequence for the mature protein: Leu- Glu-Val-Pro-Thr-Asp-Gly-Asn-Gly-Leu-. 5.6 mg of purified protein was obtained from the pellet from the two 5 L fermentation runs.
• SUBSTITUTE SHEET Example 8 The immunoblot assay for the detection of the degradation of APP 695 catalyzed by human brain protease subfractions.
• the proteolytic reaction was terminated by addition of 7.5 ⁇ l, of the following 3X Laemlie SDS-PAGE sample buffer: 1.5 M Tris HCl, pH 8.45, containing 36% (v/v) glycerol and 12% (v/v) SDS, 10% (v/v) 2-mercaptoethanol, and trace bromophenol blue tracking dye. Samples were heated (100°C X 87 min), and then cooled.
• the reaction mixtures (15 ⁇ l) were applied to the wells of a 10 to 20% acrylamide gradient Tricine gel (routinely a 1.0 mm thick, 15 well Novex precast gel, Novex Experimental Technology, San Diego, CA).
• the gel was run under constant voltage conditions, and at 50 V until the sample enters the gel whereupon the voltage was raised to 100 V. Electrophoresis was discontinued when the tracking dye reaches to within 0.5 cm of the gel bottom.
• the gels were calibrated using prestained Mr markers ranging in Mr from 3 to 195 kDa (Bethesda Research Laboratories, Gaithersburg, MD.).
• kits containing high and low molecular weight markers were mixed with 10 ⁇ l of 3X sample buffer, and treated as described in section (a) (ii).
• the following molecular weight marker proteins were present in the kit as pre-stained markers: Myosin H-chain (196 kDa); phosphorylase B (106 kDa); bovine serum albumin (71 kDa); ovalbumin (45.3 kDa); carbonic anhydrase (29.1 kDa); betalactoglobulin (18.1 kDa); lysozyme (14.4 kDa); bovine trypsin inhibitor (5.8 kDa); and insulin A and B chains (3 kDa).
• the membrane was transferred to 15 ml of blocking buffer containing a 1:1000 dilution of rabbit polyclonal antiserum elicited to a synthetic human APP 695 C-terminal peptide immunogen and incubated at 4°C overnight.
• the membrane was rinsed with three successive 15 ml volumes of blocking buffer with gentle shaking for 5 minutes.
• the membrane was then transferred to 15 ml of blocking buffer containing a 1:1000 dilution of alkaline phosphatase-coupled Goat anti-Rabbit IgG (Fisher Scientific, Pittsburgh, PA.), and incubated at room temperature for 90 minutes.
• the membrane was then rinsed with three successive 15 ml volumes of blocking buffer with gentle shaking for 10 minutes.
• the membrane was next washed with three consecutive 15 ml volumes of alkaline phosphatase buffer for 5 minutes each, comprising: 100 mM Tris HCl pH 9.5, containing 100 mM NaCl and 5 mM MgCl 2 .
• the gel was next incubated in the dark with 15 ml of 100 mM Tris HCl pH 9.5, containing 100 mM NaCl, 5 mM MgCl 2 and 50 ⁇ l of BCIP substrate (50 mg/ml, Promega, Madison, WI.) and 99 ⁇ l of NBT substrate (50 mg/ml, Promega). Incubation was continued until there was no apparent further intensification of low Mr immunoreactive bands
• SUBSTITUTE SHEET typically 3 hours at room temperature. The gel was then rinsed with deionized water and dried.
• M-III, M-VIII, S-I, S-III, P-2 V, and P-2 VII Six selected pools (designated "M-III, M-VIII, S-I, S-III, P-2 V, and P-2 VII") were found to contain significant APP degrading activity.
• Corresponding control brain pools also contained some of the above activities, but it was not possible to determine whether the levels of the activities were different or not, between control and AD pools.
• Each of the above six pools had an enzyme activity capable of forming an 11. 5 kDa APP C-terminal fragment.
• the proteolytic product of MR 11.5 kDa was of particular interest because
• T TE in further studies it was usually the major immuno-detectable C-terminal product, and was found to co-migrate with a recombinant C-terminal fragment of APP comprising an open reading frame that would start with the n-terminal aspartate of the beta-amyloid peptide and extend to the C-terminus of the full length molecule (the C-100 fragment).
• This co-migration is exemplified in Figure 2d.
• the 11.5 kDa band is the product of endoproteolysis of APP at or near the N-terminus of the A4 region, and that the above protease activities capable of forming this fragment might play a role in vivo, in the genesis of amyloidogenic peptides.
• Figure 2d shows that at least in the case of P2 pool VII, the 11.5 kDa C- terminal enzymatic product of APP proteolysis is capable of aggregation.
• the 11.5 kDa C- terminal enzymatic product of APP proteolysis is capable of aggregation.
• the peptide band at Mr 11.5 kDa which comigrates with the PMTI 100 driven C-100 standard, and the 18 kDa fragment, there appear other bands at Mr 24.3, 27.4 and 35.5 kDa.
• the 24.3 and 35.5 kDa bands are of a Mr expected for dimers and trimers, respectively, of the C-100 fragment, and roughly comigrate with the corresponding faint bands in the C-100 which are due to aggregation (see Example 5 for details).
• Figures 2a-2c also help to show that the assay can be used to examine the effect of classical protease inhibitors.
• P-2 V is inhibited partially by methanol and completely by methanolic pepstatin A
• M-III Figure 2b
• S-I Figure 2c
• the assay in one embodiment, is applied to the search for novel in vitro inhibitors of the APP degrading enzymes.
• the potent compounds thereby identified are tested for in vivo efficacy using a suitable animal model such as a transgenic animal designed to overexpress APP or a beta-amyloid-containing fragment thereof.
• Table 4 summarizes some of the properties of the six main pools of APP degrading activity recovered from the Mono-Q fractions, including peptide product sizes, apparent pH dependence for product formation, and the effects of commercially available protease inhibitors.
• Trial B EGTA (1 mM), cystatin (20 ⁇ M), captopril (300 ⁇ M), aprotinin (15 ⁇ M), N-dansyl-Ile-Ser-Glu-Val-Lys-Met- Asp-Ala-Glu-Phe-Arg-His-Asp-Asp-Asp-Asp (SEQ ID NO: 1 mM), cystatin (20 ⁇ M), captopril (300 ⁇ M), aprotinin (15 ⁇ M), N-dansyl-Ile-Ser-Glu-Val-Lys-Met- Asp-Ala-Glu-Phe-Arg-His-Asp-Asp-Asp-Asp (SEQ ID
• N.D. not determined due to low or inconsistent levels of activity; N.I. no inhibition observed
• M-III and S-I are highly similar by all the listed criteria, and probably represent the same enzyme cross contaminating each of the S and M fractions. It is probable, therefore, that human brain contains a minimum of five different protease activities capable of degrading APP to yield a 11.5 kDa C-100-like product fragment.
• Table 4 shows that some of the activities were insensitive to inhibition by any of the inhibitors tested, and these enzymes may represent members of an unusual group.
• the activities involved in the formation of 11.5 kDa C-100 fragments in M-III and S I, are either of serine or cysteine type, or represent members of an unusual group.
• these fractions may contain both a serine and cysteine protease with both enzymes playing an obligatory (sequential) role in the production of C-100.
• P2 V contains both an aspartic protease activity and a serine protease activity.
• P2 VII contains a serine protease activity based upon its sensitivity toward PMSF.
• the present assay is of a sufficient specificity to enable the isolation of specific APP degrading enzymes from human brain.
• APP degrading enzyme activity is shown in Figure 5 for individual mono-Q fractions from the purification of solubilized P-2 fraction according to the method of Example 1.
• the relative staining intensity for enzymatic C-terminal APP fragments was consistently greatest in the P-2 subfraction from Mono-Q.
• APP degrading activity in the P-2 was recovered from Mono-Q as two distinct migration peaks (A and B, Figure 5).
• Peak A eluted in the loading and low ionic strength wash, i.e. in a region roughly corresponding to the recovery of P-2 V, seen in our initial studies (Table 4), whereas peak B overlapped with the pooled region in which P-2 VII activity was previously observed (Table 4), and shown to comprise both serine and aspartic protease activities. Similar sized degradation products were observed with both the peak A and B activities at Mr approx. 28, 18 and 14 and ⁇ 11 kDa, although the relative staining intensity of the 18 kDa band was much greater in peak B than in peak A. Peak B was pooled and subjected to purification on superose 6HR as described in Example 1, Method 1.
• Eluted fractions contained two qualitatively distinct types of activity which overlapped in their elution profiles.
• the activity which produced an APP breakdown pattern most closely resembling that observed with the original peak B fractions (figure 5) was recovered in fractions 51 through 56 from gel filtration (figure 6 b and c), consistent with an apparent Mr of 15 to 25 kDa.
• This elution peak was preceded by elution of an activity which predominantly formed an 18 kDa breakdown product, and is presumably catalyzed by a protease of larger Mr apparent. This latter activity probably corresponds to the serine protease activity previously
• cathepsin D selectively cleaved the APP so as to produce a similar pattern of C-terminal degradation products to those observed with P-2 peak B ( Figure 5) described above from Mono-Q.
• P-2 peak B Figure 5
• commercial cathepsin D preparations degraded holo-APP in a time dependent fashion to produce major C-terminal products of approximate Mr 18 and 28 kDa.
• Inhibition of the activity by pepstatin A confirmed the involvement of cathepsin D in the reaction ( Figure 8).
• a commercial polyclonal antibody to human cathepsin D was obtained (Dako Corp, Carpinteria, CA, catalog # A561), and found to be reactive toward human cathepsin D on immunoblots, generating an immunoreactive band of Mr 27-28 kDa.
• the antibody was used in an immunoblot assay to examine if chromatography fractions from the mono-Q purification of either P-2, soluble or microsomal fractions contained immunoreactive cathepsin D. The antibody did not cross-react with human renin on immunoblots.
• the pKa of the Asp side chain is more acidic than the Glu residue, and would be protonated to a lesser degree than the Glu residue throughout the pH range examined in Figure 11. This may explain the lower cleavage rates at the -Met-Asp- bond with cathepsin D, and the hint at a lower pH optimum for cleavage ( ⁇ pH 3) at this site when compared with the- Glu-Val- bond.
• the activity of the pooled fractions was inhibited completely by 10 ⁇ M pepstatin A, an aspartic protease inhibitor ( Figure 20b), but was unaffected (not shown) by inhibitors of other protease classes such as EDTA (1 mM), PMSF (0.4 mM), E-64 (0.1 mM), and aprotinin (10 ⁇ g/ml).
• the APP-degrading activity coincided with the elution of a pepstatin sensitive protease which hydrolyzed the APP mimetic N-dansyl-ISEVKMDAEFR-NH 2 at the -E-V- bond ( Figure 20a).
• Cathepsin D co- eluted with the holo-APP and peptide degrading activities as judged by immunoblot detection of the 20 kDa cathepsin D light chain ( Figure 20d).
• Cathepsin D immunoreactivity was recovered from the anti-cathepsin D column but not from the control column by elution with 100 mM glycine pH 2.5 when 0.5 % triton X-100 was included. Notice that no protein bands other than those corresponding to immunoreactive cathepsin D were detected in this eluent ( Figure 21c). This renders unlikely the possibility that the adsorbed APP degrading activity resulted from an immunologically cross reacting protease other than cathepsin D itself. Unfortunately, only trace amounts of APP degrading enzyme activity were recovered from the anti- cathepsin D column (not shown). This is probably because the activity was inhibited by the neutralized elution buffer: the composition of the elution buffer also quantitatively inhibited the degradation of holo-APP by purified cathepsin D (not shown).
• Figure 12 shows both a coomassie stained blot as well as an immunoblot (using the anti-beta-amyloid monoclonal antibody) of such a reaction mixture.
• incubations were also performed in the absence of cathepsin D (wherein cathepsin D would be added back to the incubation mixture after addition of SDS-PAGE sample buffer), or in the absence of APP 695 substrate.
• Eight main product bands were observed by coomassie staining ( Figure 12a) of the complete incubation mixture, and which were also absent from either of the controls.
• the cleavage of the -Glu-Val bond at APP 593-594 is consistent with the observed capacity of the cathepsin D to cleave the corresponding bond in the peptide substrate (as described above) in a pepstatin inhibitable reaction.
• cathepsin D generated a 5.6 kDa product (band 3, Table 5), by atypical hydrolysis at the -Glu-Val- bond three amino acid residues N- terminal to the purported N-terminal -Asp- residue of the common form of beta- amyloid.
• the fragment was absent in the equivalent sections of the blot taken from the incubation without cathepsin D.
• the fragment is of the right size (5.6 kDa) to contain full length beta-amyloid peptide, and its generation
• SUBSTITUTE SHEET suggests that cathepsin D must also cleave the APP at a second site close to the C- terminal region of the beta-amyloid peptide.
• APP 695 contains numerous other peptide bonds that would seem to have been ideal substrates for cathepsin D cleavage yet were not cleaved by cathepsin D.
• the fact that they were not hydrolyzed reflects the high degree of sequestration of these sites away from access to cathepsin D within the folded APP structure: most of the hydrophobic pairs would be expected to locate to the hydrophobic APP protein core.
• SUBSTITUTE SHEET migrated in the same position as band 3 in Figure 12a (Table 5) a doublet between Mr 9 to 10 kDa comigrating with band 5 in Figure 12a (Table 5), and a doublet at Mr 14 kDa, a doublet at 16 to 18 kDa comigrating with band 6, Figure 12a, and a band at Mr 40 kDa.
• bands 3 and 6 comigrated with bands detected by immunoblot in Figure 12c. Consistent with this, only these same three bands in Table 5 were of the appropriate N-terminal sequence and size to contain the beta-amyloid epitope.
• SUBSTITUTE SHEET Figure 23 provides an update of the sequences of APP fragments formed by cathepsin D to include those identified since Table 5 was prepared. It also relates the sequences to the particular C286.8A immunoreactive bands observed in Figure 12c. For each C286.8A immunoreactive band, the corresponding segments of a sequencing blot yielded a fragment(s) of a size and sequence sufficient to contain the C286.8A epitope and thus account for the immunoblot band ( Figure 23a).
• the fragmentation pattern suggests that formation of the 5.5 kDa fragment occurs by progressive N- and C-terminal nibbling of a larger precursor such as the 38 kDa peptide, or even the un-characterized transient 28 kDa fragment of Figure 13. It is probable that the 5.5 kDa immunoreactive fragment derives directly from the 10-12 kDa fragment with the same sequence. Both of these fragments and those of Mr 15-16 and 18-19 seem to be of a size sufficient to contain a full length copy of ⁇ AP. Obviously, other products such as those resulting from further processing of the 10-12 kDa immunoreactive fragment may have gone undetected, perhaps due to further degradations, or to loss during electroblotting.
• Amyloid deposition is favored at the acid pH of the lysosome (Burdick et al., 1992, /. Biol. Chem., 267: 546).
• cathepsin D is a lysosomal protease, it has also been shown by histochemistry to be present in significant levels associated with amyloid deposits in Alzheimer's brain (Cataldo et al., 1990, Proc. Natl. Acad. Sci USA 87: 3861).
• beta-amyloid released by cells in culture comprises a minor N- terminal sequence starting at residue Val 594 (Haas et al., 1992, Nature, 359: 322) which is three amino acids N-terminal to the more abundant sequence beginning at the Asp 597 residue commonly seem in beta-amyloid 1-42.
• the minor sequence probably arises by direct endoproteolysis at the -Glu-Val bond at position 593-594, ie. the same site as shown presently to undergo specific proteolysis by cathepsin D.
• cathepsin D can hydrolyse both the -Glu-Val- and -Met-Asp- bonds, it has the necessary specificity to form both of the beta-amyloid fragments sequenced by Haas et al.
• cysteine masculinease inhibitors E-64 and leupeptin were without effect on the release of beta-amyloid by LC-99 cells while general lysosomal inhibitors blocked the release (Shoji et al., 1992, Science, 258: 126), showing that beta-amyloid formation by these cells was catalyzed by a lysosomal enzyme other than a cysteine protease.
• a remaining candidate protease for such a reaction would be lysosomal cathepsin D which is not inhibited by the cysteine protease inhibitors used in their studies.
• APP contains a stretch of hydrophobic residues between the C- terminus of beta-amyloid and the membrane anchor sequence.
• SUBSTITUTE SHEET peptide bonds in this region could be hydrolysed by cathepsin D.
• the- Leu-Val- peptide bond at position 645-646 is highlighted by the PEPTIDESORT computer program as being a probable cathepsin D recognition site. This site is close to the position of three of the point mutations shown to co-segregate with certain forms of Familial Alzheimer's Disease (FAD). Cleavage within this region as well as the -Glu-Val- bond at positions 593-594 could account for the size of band 3 in Table 5. The FAD mutations at this site could augment the rates of APP cleavage within this region by cathepsin D.
• a double mutation of APP from -Lys-Met- to -Asn-Leu- at positions 595-596 also cosegregates with FAD.
• This mutation makes it unlikely that a specific amyloidogenic protease exists with specificity for cleavage about the -Lys-Met- bond, since the mutation would be expected to abrogate rather than augment cleavage by such an enzyme. Rather, the likelihood is increased that the primary endoproteolysis yielding beta-amyloid occurs adjacent to and preferably N-terminal to this dipeptide.
• the -Glu-Val- bond represents the closest N-terminal site left unaffected in the SI and SI' positions.
• the -Asn-Leu- mutation at S2' and S3' sites to the -Glu-Val- scissile bond could augment cleavage by cathepsin D.
• This cleavage by cathepsin D also liberates a peptide with an N-terminus that is the same as that found in the major form of beta-amyloid, and provides a mechanism by which the -NL- mutation observed in this particular early onset FAD causes enhanced rates of beta-amyloid formation by providing a site that is more rapidly cleaved by the amyloidogenic protease cathepsin D.
• the increased rates of beta-amyloid accumulation that could result, could trigger the early onset form of Alzheimer's Disease linked to this APP mutation.
• cathepsin D as a serious candidate for the primary amyloidogenic protease of Alzheimer's Disease, significantly aids the effort of development of therapeutic inhibitors for the disease.
• specific cathepsin D inhibitors could provide therapeutic benefit by inhibiting the toxic accumulation of beta-amyloid.
• the new information provided herein makes it comparatively straightforward to rationally design tight-binding inhibitors as has been accomplished for the design of novel inhibitors of other aspartic proteases such as renin and HIV-protease.
• cathepsin D can now be adapted for use in a high
• a suitable assay for such purposes could include the N-Dansyl-peptide assay described in Examples 2 and 3 of the present invention.
• Example 1 shows that human brain contains serine proteases capable of C- terminal processing of recombinant APP, and that in some cases these serine proteases were inhibitable with aprotinin.
• Example 1 An alternate isolation scheme was devised (Example 1, Method 2) incorporating affinity purification on aprotinin-sepharose as an early step.
• One or more of these product peptides could be amyloid or give rise to beta-amyloid by further processing of these peptides C-terminal to the beta- amyloid region.
• the serine protease activity involved in formation of these products could therefore play a role in amyloidosis.
• Example 11 Design of therapeutic cathepsin D inhibitors.
• the PI and PI' amino acid side chains correspond to the amino acids involved in formation of the peptide bond which is to be cleaved.
• the side chains PI to Pn and PI' to Pn' are envisioned to form specific interactions with a corresponding series of enzyme subsites SI to Sn and SI' to
• peptidomimetic inhibitors could utilize either the n-dansyl peptide substrate assays of Examples 2 and 3, or the assay of holo-APP degradation described in Example 8, to make enzymologic measurements, in conjunction with purified cathepsin D.
• Peptidomimetic compounds would be synthesized containing essential amino acid sequences necessary for optimal cleavage (from 1 a, b above), and the appropriate spacer.
• the amino acid sequences in these peptides could be the same as those observed around the cleavage site in the APP substrate, e.g. Glu-Ile-Ser-Glu-Val-Lys-Met-Asp (SEQ ID NO: 4) and Trp-His-Ser- Phe-Gly-Ala-Asp-Ser (SEQ ID NO: 5) or alternatively selected from those sequences found to confer optimal binding to cathepsin D based on studies of their potency for in vitro inhibition of cathepsin D.
• Glu-Ile-Ser-Glu-Val-Lys-Met-Asp SEQ ID NO: 4
• Trp-His-Ser- Phe-Gly-Ala-Asp-Ser SEQ ID NO: 5
• Peptidic inhibitors would be synthesized that contain either the above sequences or sequences exhibiting optimal cathepsin D inhibition (including shorter variants perhaps containing N- and/or C- substitutions), in which the- CO-NH- atoms of the peptide bond between PI and PI' are replaced with any of the following standard spacer groups and using appropriate synthetic routes so as to obtain any possible stereo-chemical configuration thereof: reduced amide, hydroxy isostere, ketone isostere, dihydroxy isostere, statine analogs, phosphonates or phosphonamides, reversed amides. Most of these inhibitors would function as transition state analogs.
• N- and C-terminal substitution with blocking groups such as Boc or acetyl (N-terminally), or O-Me, O-benzyl, N-benzyl (C-terminally).
• cathepsin D inhibitors either in whole or in part could be used as therapeutic inhibitors for Alzheimer's Disease, or as starting points for optimization of inhibitory potency and the development of new derivatives for Alzheimer's Disease.
• SUBSTITUTE SHEET therapy of Alzheimer's Disease Such inhibitors include: 1-Deoxynojirimicin (Lemansky et al., 1984, /. Biol Chem., 259: 10129); Diazoacetyl-norleucine methyl ester (Keilova et al., 1970, Febs Lett, 9: 348); Gly-Glu-Gly-Phe-Leu-Gly-Asp-Phe- Leu (SEQ ID NO: 6) (Gubenseck et al, 1976, Febs Lett, 71: 42); Pepsin inhibitor from Ascaris (Keilova et al., 1972, Biochem Biophys Ada., 284: 461); pepstatin (Yamamoto et al., 1978, European Journal of Biochemistry, 92: 499).
• Example 12 The effect of pepstatin A, an inhibitor of cathepsin D on the formation of APP C-terminal fragments by Human Embryonal Kidney (HEK) 293 cells maintained in culture.
• HEK Human Embryonal Kidney
• pepstatin A an inhibitor of cathepsin D activity in vitro inhibits the capacity of HEK 293 cells to form and release into the tissue culture medium APP C-terminal fragments of the same size (15 kDa) as those shown to be formed from APP695 by cathepsin D in vitro (example 9).
• HEK cells are known to release beta-amyloid from transfected APP695, and so contain the proteases necessary for amyloidogenic APP processing [C Haass et al., 1992, Nature, 359: 322]. These cells therefore provide an accepted cellular model for the study of beta-amyloid formation.
• Fractions detected by this procedure would have to contain the N-terminal heptapeptide sequence of beta-amyloid (to explain binding to immobilized C286.8A) as well as the C-terminal domain or a portion thereof (to explain reactivity with the anti-C-terminal antibody).
• Figure 19 b compares the amounts of C-terminal fragments recovered in the elution fractions from a column of immobilized C286.8A that had been loaded with the media from cells grown either in the presence of DMSO only or DMSO plus pepstatin A. Chromatography was performed in parallel under identical conditions. As can be seen, treatment with pepstatin A significantly reduced the amount of an eluted 15 to 16 kDa APP-derived fragment that could be detected by immunoblot. This fragment is the same size as the fragment formed in vitro by cathepsin D with the N-terminal sequence G-A-D-S-V-P-A-
• SUBSTITUTE SHEET (Table 5 and Figure 23), and could represent an intermediate in the cellular formation of the smaller 5.6 kDa fragment with an N-terminus corresponding to a form of beta-amyloid.
• Other APP fragments that are formed by cathepsin D in vitro were not detected in this experiment. The undetected fragments may have been present below the detection limit or further degraded in the cells by other proteases.
• This experiment shows that HEK 293 cells, an accepted cell line for the characterization of cellular amyloid formation make and release at least one APP fragment that resembles the APP695 fragments formed in vitro by cathepsin D and that formation of this fragment is inhibited by a non toxic dose of a cathepsin D inhibitor.
• peptidic based inhibitors of cathepsin D have utility in altering cellular APP processing.
• Inhibitors 1, 2, 3, 4, 10 and 20 have been described in German application DE 4,215,874, filed on May 14, 1992, which corresponds to U.S. Serial No. 08/059,488, filed on May 10, 1993. The disclosures of both of these applications are incorporated herein by reference.
• the inhibitors are disclosed to be antiviral agents, specifically HIV protease inhibitors.
• Inhibitor 1 was prepared as follows:
• Second step A solution of 2.41 g (3.28 mmol) of compound 21, above, in 17 ml of a 4 N solution of gaseous hydrogen chloride in anhydrous dioxane is stirred at 0°C for 30 min. 15 ml of toluene are then added and the mixture is concentrated in vacuo. This process is repeated a further two times, and the residue is then triturated with ether, filtered off with suction and dried in a high vacuum over potassium hydroxide (KOH) to yield 2.29 g (98% of theory) of the compound:
• SUBSTITUTE SHEET Third step A stirred solution, cooled to 0°C, of 0.80 g (2.40 mmol) of (2S)- 3-tert-butylsulphonyl-2-(l-naphthylmethyl)-propionic acid [prepared according to H. Buhlmayer et al., 1988, /. Med. Chem., 31: 1839] and 0.40 g (2.64 mmol) of HOBT in 20 ml of anhydrous dichloromethane is treated with 0.52 g (2.52 mmol) of DCC and stirred for 5 min.
• Inhibitors 2, 3, 4, 10 and 20 can be prepared analogously by coupling the appropriate acids with the appropriate amine hydrochlorides, which are known or can be prepared by conventional means. In the case of inhibitors 3 and 20, it will be necessary to start from the compound having the formula: 23/
• Inhibitor 10 will require the starting material of the formula:
• compound 24 is analogous to that of compound 21, but starting from 258 mg (0.92 mmol) of (2R)-N-(tert-butoxycarbonyl)-2-amino-2-[2- (l,3-dithiolan-2-yl)]acetic acid [EP 412 350] and 500 mg (0.84 mmol) of 1- ⁇ (2R, S, 4S, 5S)-[5-amino-6-cyclohexyl-4-hydroxy-2-(2-phenyl)-hexanoyl] ⁇ -S-isoleucinyl-2-
• Inhibitor 7 is known from published European application EP 0 472 077, which was published on February 26, 1992, and the entire contents of which are incorporated herein by reference. The inhibitor is disclosed therein as an inhibitor of HIV protease activity.
• Inhibitors 8 and 13 are known from published European application EP 0 441 912, which was published on August 14, 1991, and the entire contents of which are incorporated herein by reference. The inhibitors are disclosed therein as inhibitors of renin.
• Inhibitors 9, 15, 16 and 19 are known from published European application EP 0 472 078, which was published on February 26, 1992, which is equivalent to U.S. Patent No. 5,147,865, which issued September 15, 1992. The entire contents of both publications are incorporated herein by reference. The inhibitors are disclosed therein as inhibitors of HIV protease activity.
• Inhibitors 11 and 12 are described in German application DE 41 26 485, which was filed on August 10, 1991, and corresponds to U.S. Serial No.
• Second step A stirred solution, cooled to 0°C, of 4.81 g (22.13 mmol) of N- (tert-butoxycarbonyl)-L-valine and 3.29 g (24.35 mmol) of HOBT in 40 ml of
• SUBSTITUTE SHEET is obtained as a colorless foam.
• inhibitor 5 and 6 are within the generic teachings of EP 0 441 912, supra, and can be prepared following the preparation schemes taught therein.
• inhibitor 5 can be prepared as follows:
• SUBSTITUTE SHEET Second step 265 g (1.0 mol) of compound 30 are dissolved in 2 1 of methanol and hydrogenated on 20 g of 5% Rh/C under 40 atm for 5 hr. The catalyst is filtered off through celite with suction and washed with methanol, and the resulting solution is concentrated. 271 g (100%) of the compound:
• Third step 163.0 g (0.601 mol) of compound 31 and 40.3 g (0.661 mol) of N,0-dimethylhydroxylamine are dissolved in 2 1 of methylene chloride at room temperature. At 0°C, 303.5 g (3.005 mol) of triethylamine are added dropwise (pH ⁇ 8). At max. -10°C, 390.65 ml of a 50% strength solution (0.601 mol) of n-PPa in methylene chloride are added dropwise. The mixture is warmed to 25°C overnight and is stirred for 16 hr. The reaction is then concentrated, 500 ml of saturated bicarbonate solution are added to the residue, and the mixture is stirred at 25°C for 20 min.
• Compound 33 is either further processed immediately or stored at
• Inhibitor 6 is obtained analogously to inhibitor 5, except that in the eighth step a solution of 3-methylpentanoic anhydride [prepared from 21 mmol of 3- methylpentanoic acid and 2.16 g (10.5 mmol) of dicyclohexylcarbodiimide in 50 ml of methylene chloride, filtration] in methylene chloride is used.
• 3-methylpentanoic anhydride prepared from 21 mmol of 3- methylpentanoic acid and 2.16 g (10.5 mmol) of dicyclohexylcarbodiimide in 50 ml of methylene chloride, filtration
• Inhibitor 17 is within the generic teachings of EP 0 472 077, supra, and can be prepared following the preparation schemes taught therein. Thus, inhibitor 17 can be prepared as follows:
• Second step A stirred solution, cooled to 0°C, of 4.63 g (21.3 mol) of N- (tert-butoxycarbonyl)-L-valine and 3.29 g (24.35 mmol) of HOBT in 40 ml of anhydrous dichloromethane is treated with 5.29 g (25.65 mmol) of DDC and the mixture was stirred for 5 min. A solution of 3.60 g (19.00 mmol) of compound 39 and 8.85 ml (80.48 mmol) of N-methylmorpholine in 30 ml of dichloromethane is then added dropwise. The cooling bath is removed and the reaction mixture stirred at room temperature for 2 hr. The end of the reaction is determined by thin layer chromatography.
• Inhibitor 14 is known from published European application EP 0 437 729, which was published on July 24, 1991, and corresponds to U.S. Patent No. 5,145,951, which issued on September 8, 1992. The entire contents of both publications are incorporated herein by reference. The inhibitor is disclosed therein as an inhibitor of HIV protease activity.
• Inhibitor 18 is known from published European application EP 0 403 828, which was published on December 27, 1990, and corresponds to U.S. Serial No. 07/876,697, filed April 28, 1992, still pending, which is a continuation of U.S. Serial No. 07/524,779, filed May 16, 1990, now abandoned. The complete disclosures of these three applications are incorporated herein by reference. The inhibitor is disclosed therein as an inhibitor of HTV protease activity.
• SUBSTITUTE SHEET lie Ser Glu Val Lys Met Asp Ala Glu Phe

Landscapes

• Pharmaceuticals Containing Other Organic And Inorganic Compounds (AREA)
• Acyclic And Carbocyclic Compounds In Medicinal Compositions (AREA)
• Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)

Applications Claiming Priority (3)

Publications (2)

Family

ID=26787106

Family Applications (1)

Country Status (2)

Families Citing this family (1)

Family Cites Families (3)

• 1993
  • 1993-11-12 AU AU57264/94A patent/AU5726494A/en not_active Withdrawn
  • 1993-11-12 EP EP94903250A patent/EP0694076A1/de not_active Withdrawn

Also Published As

Similar Documents

Legal Events