Carmela R. Abraham

Immunochemical identification of the serine protease inhibitor α1-antichymotrypsin in the brain amyloid deposits of Alzheimer's disease

Two approaches--molecular cloning and immunochemical analysis--have identified one of the components of Alzheimers disease amyloid deposits as the serine protease inhibitor alpha 1-antichymotrypsin. An antiserum against isolated Alzheimer amyloid deposits detected immunoreactivity in normal liver. The antiserum was then used to screen a liver cDNA expression library, yielding three related clones. DNA sequence analysis showed that these clones code for alpha 1-antichymotrypsin. Antisera against purified alpha 1-antichymotrypsin stained Alzheimer amyloid deposits, both in situ and after detergent extraction from brain. The anti-amyloid antiserum recognizes at least two distinct epitopes in alpha 1-antichymotrypsin, further supporting the presence of this protein in Alzheimer amyloid deposits. In addition to being produced in the liver and released into the serum, alpha 1-antichymotrypsin is expressed in Alzheimer brain, particularly in areas that develop amyloid lesions. Models by which alpha 1-antichymotrypsin could contribute to the development of Alzheimer amyloid deposits are discussed.

Isolation of Low-Molecular-Weight Proteins from Amyloid Plaque Fibers in Alzheimer's Disease

During aging of the human brain, and particularly in Alzheimers disease, progressive neuronal loss is accompanied by the formation of highly stable intra‐ and extraneuronal protein fibers. Using fluorescence‐activated particle sorting, a method has been developed for purifying essentially to homogeneity the extracellular amyloid fibers that form the cores of senile plaques. The purified plaque cores each contain 60–130 pg of protein. Their amino acid composition shows abundant glycine, trace proline, and ∼50% hydrophobic residues; it resembles that of enriched fractions of the paired helical filaments (PHF) that accumulate intraneuronally in Alzheimers disease. Senile plaque amyloid fibers share with PHF insolubility in numerous protein denaturants and resistance to proteinases. However, treatment of either fiber preparation with concentrated (88%) formic acid or saturated (6.8 M) guanidine thiocyanate followed by sodium dodecyl sulfate causes disappearance of the fibers and releases proteins migrating at 5–7.000 and 11–15.000 Mr which appear to be dimerically related. Following their separation by size‐exclusion HPLC, the proteins solubilized from plaque amyloid and PHF‐enriched fractions have highly similar compositions and, on dialysis, readily aggregate into higher Mr polymers. Antibodies raised to the major low‐Mr protein selectively label both plaque cores and vascular amyloid deposits in Alzheimer brain but do not stain neurofibrillary tangles, senile plaque neurites, or any other neuronal structure. Thus, extraneuronal amyloid plaque filaments in Alzheimers disease are composed of hydrophobic low‐Mr protein(s) which are also present in vascular amyloid deposits. Current evidence suggests that such protein(s) found in PHF‐enriched fractions may derive from copurifying amyloid filament srather than from PHF.

α1-Antichymotrypsin is associated solely with amyloid deposits containing the β-protein. Amyloid and cell localization of α1-antichymotrypsin

Our recent studies demonstrated that alpha 1-antichymotrypsin (ACT), a serine protease inhibitor, was associated with the beta-protein in the brain amyloid deposits of Alzheimers disease, aged human controls and aged monkeys, suggesting a role for the inhibitor in the amyloid deposition. In the present study we used immunohistochemistry to test for the presence of ACT in the amyloid deposits which contain, as their major component, a protein different from the beta-protein. ACT was not found in the amyloid deposits in primary or secondary amyloidosis, familial and amyloidotic polyneuropathy or Creutzfeldt-Jakob disease (non-beta-protein amyloidoses), but was found (together with beta-protein) in Alzheimers disease, Downs syndrome, normal aging, and hereditary cerebral hemorrhage with amyloidosis of Dutch origin. These results suggest a specific association of ACT with beta-protein amyloid. We next examined the distribution of the inhibitor in normal human brain and in various human neuropathological states in order to identify cells that express this protein during brain degeneration. In addition to its association with amyloid, ACT immunoreactivity was also located in astrocytes near areas of neuronal or tissue loss, in a few neurons and pericytes and in the epithelium of the choroid plexus.

Alzheimer's disease: Immunoreactivity of neurofibrillary tangles with anti-neurofilament and anti-paired helical filament antibodies

The origin of the paired helical filaments (PHF) that accumulate in human neurons during aging and in Alzheimers disease and their relationship to normal neurofilaments remain unclear. The observation that a rabbit antiserum to highly enriched PHF fractions specifically labeled PHF in Alzheimer neurofibrillary tangles but showed no reaction with neurofilaments or other normal cytoskeletal proteins led us to compare this antiserum to two monoclonal antibodies, RT97 and BF10, previously found to cross-react with tangles and with the 210,000 and 155,000 mol. wt. neurofilament proteins, respectively. Both alpha-PHF serum and the neurofilament monoclonals strongly immunolabel almost all neurofibrillary tangles in Alzheimer cortical sections. Double-immunolabeling studies show that both reagents recognize the same tangles and usually show identical patterns of staining of intraneuronal fibrous material. Following prolonged extraction of cortex in sodium dodecyl sulfate, a step which removes normal neurofilaments but leaves PHF intact, almost all isolated tangles retain strong immunoreactivity with alpha-PHF serum at an intensity which is slightly reduced from that in cortical sections. In contrast, only a small number of isolated tangles are stained strongly by RT97 and BF10; most show much decreased or no reactivity with these monoclonal neurofilament antibodies. This differential immunoreactivity was confirmed by double-labeling studies. Tangles prepared under gentle extraction conditions show strong reactivity with alpha-PHF antibodies but again only a small number are strongly labeled by RT97 and BF10. We conclude that neurofibrillary tangles in Alzheimers disease are heterogeneous as regards their filamentous content and contain both antigens cross-reacting with neurofilaments and antigens which are apparently unique to PHF and not shared with normal neurofilaments.

α1-Antichymotrypsin is present together with the β-protein in monkey brain amyloid deposits

Abstract The recent finding that the serine protease inhibitor,α1-antichymotrypsin, is tightly associated with the amyloid deposits in brains of normal aged individuals and patients with Alzheimers disease [Abraham C.R., Selkoe D.J. and Potter H. (1988)Cell52, 487–501], suggests a role for this inhibitor in the progressive deposition of brain amyloid in humans. We have used immunocytochemistry to detectα1-antichymotrypsin in the amyloid that accumulates in brains of aged monkeys, a naturally occurring animal model of Alzheimer-like neuropathology. In monkeys of increasing age, the earliestα1-antichymotrypsin immunoreactivity was found in cortical perivascular cells, before the appearance of either Thioflavin S-detectable amyloid deposits or β-protein reactivity in the vessel walls. Subsequently, amyloid deposits appeared in small meningeal blood vessels and cortical neuritic plaques. The oldest monkeys also showed microvascular amyloid in the cortical gray matter. Amyloid was never seen in white matter. The amyloid deposits in meningeal vessels were always positive for both β-protein andα1-antichymotrypsin, whereas in the cortex,α1-antichymotrypsin immunoreactivity seemed to appear somewhat later than that of β-protein. These findings demonstrate that two of the brain amyloid components of human senescence and Alzheimers disease—the β-protein and the protease inhibitorα1-antichymotrypsin—are also present in the amyloid deposits of normal aged monkey brain. The extended molecular parallels between normal brain aging and Alzheimers disease suggest that similar biochemical mechanisms may underlie progressive amyloid deposition in both situations.

Huntington's disease: Changes in striatal proteins reflect astrocytic gliosis

Huntingtons disease is an autosomal dominant neuronal degeneration characterized by age-related neuronal loss principally affecting caudate and putamen and, to a lesser extent, cerebral cortex. In order to identify selective polypeptide alterations in HD brain, we analyzed unfractionated homogenates and purified neuronal perikarya from striatum and cortex of 12 control and 14 HD brains by gel electrophoresis and immunochemical techniques. SDS polyacrylamide gel electrophoresis (SDS-PAGE) revealed a 3- to 8-fold increase in a 50,000 MW (50K) protein in HD striatal homogenates. Neuronal fractions isolated from the same tissue almost never showed this change. In cortex, 50 K protein was either normal or minimally increased. The increase at 50 K in striatal homogenates was often associated with variable increases of proteins at 40 K and 43 K. No other consistent polypeptide changes in HD brain tissue were found by one-dimensional SDS-PAGE. The increased 40 K protein in HD striatum extracts showed a strong immunoprecipitant line with an antiserum to GFA. This antiserum also produced greater immunofluorescent staining of HD than control striatum. Direct immunostaining of polypeptides in gels demonstrated selective staining of the 50 K, 43 K and 40 K proteins in HD striatum. The pattern was highly similar to that reported by Dahl et al. 6 for glial filament preparations that underwent postmortem proteolysis. We conclude that these polypeptide changes are related to increased glial filaments in affected HD tissue, and that similar protein changes reported in other human neuronal degenerations also reflect secondary astrocytic gliosis rather than the primary gene product.

A calcium-stimulated serine protease from monkey brain degrades the β-amyloid precursor protein

Amyloid deposition, a histopathological feature of Alzheimers disease brain, may be the underlying cause of this disease. The isolation of enzymes involved in both the normal and aberrant or alternative processing of the beta-amyloid precursor protein may lead to an understanding of how beta-protein, the major component of amyloid deposits, is formed in the brain parenchyma and vasculature of Alzheimers disease patients and aged humans. As the same kind of deposits is also found in aged primates, the use of primates will undoubtedly help to understand the mechanisms of amyloid deposition, both spatially and temporally. Here we report the partial purification from adult monkey brain of a calcium-activated serine protease that is immunoreactive with antibodies against cathepsin G and is potentially involved in the abnormal degradation of the beta-amyloid precursor protein. Moreover, immunoreactivity with cathepsin G antibodies was localised to astrocytes in both adult and aged monkey cortex, suggesting that our protease may be expressed in astrocytes.

The Protease Inhibitor, α1-Antichymotrypsin, Is a Component of the Brain Amyloid Deposits in Normal Aging and Alzheimer's Disease

The purpose of this study was to characterize the nature and the origin of the Alzheimers disease amyloid deposits. We used an amyloid antiserum to screen a human liver expression library. A positive clone was sequenced and found to code for the serine protease inhibitor α1-antichymotrypsin, an acute phase serum protein. Thus, this protein is a second component of the brain amyloid in addition to the β-protein. In order to determine whether the inhibitor originated from the serum or was made in the brain, we performed Northern blots on tissue from control and Alzheimer brain and found that α1-antichymotrypsin RNA is present in the brain and that the diseased brain contained larger amounts than the controls. Immunocytochemistry and in situ hybridization show the astrocytes to produce the inhibitor, mainly around senile plaques, α1-antichymotryp-sin is only associated with the amyloid deposits of the β-protein kind in normal aging of man and monkeys, Alzheimers, Downs syndrome and hereditary cerebral hem...

Potential roles of protease inhibitors in Alzheimer's disease.

Recently, protease inhibitors have been recognized as potential contributors to the pathogenesis of Alzheimers disease. In this role, they could mediate an exaggerated regenerative response in the brain, participate as acute phase reactants, or be involved in the aberrant proteolytic processing of the amyloid proteins. Protease inhibitors are, therefore, attractive targets for drug intervention in Alzheimers disease.

Production and Characterization of Monoclonal Antibodies to Alzheimer Paired Helical Filaments

The involvement of intermediate filaments in the pathological human neuronal structures called neurofibrillary tangles (NFT) has been documented in several laboratorie~. -~ We have recently shown that a determinant found on the 200K neurofilament subunit can have various degrees of expression on tangles that have been partially purified by extraction in 1% sodium dodecyl sulfate (SDS); Because of these immunocytochemical results with antibodies produced against neurofilaments, it was important to produce antibodies against the major protein component of NFT, that is, paired helical filaments (PHF). In the past, polyclonal antisera against PHF isolated from Alzheimers disease brains, which had been produced in rabbits, stained isolated tangles.6 However, when we attempted to make antibodies in BALB/c mice using SDS-isolated PHF as the immunogen, we had difficulty getting a titer high enough for the production of monoclonal antibodies. (Only one mouse out of four had a titer higher than 1:1,ooO.) Only recently have monoclonal antibodies to NFT been reported to have been produced from mice.. To circumvent the low titer problem, we decided to use Lewis rats as our immunized animal. The use of rats had two major advantages: (1) preliminary screening could be accomplished by a simple ELISA procedure to test for the production of rat IgG by our hybridomas and (2) if successful, a rat monoclonal antibody could be used for double-labeling studies in conjunction with a series of mouse antineurofilament monoclonals now in use.4~~ A Lewis rat initially received an i.p. injection of PHF, isolated from Alzheimer postmortem brain, which was emulsified in complete Freunds adjuvant. The rat was boosted monthly with a similar amount of PHF for four months. The PHF used as antigen was prepared from cortical tissue from four patients with the neurological and histopathological findings of Alzheimers disease. The ages of the patients at time of death were 77, 59, 65, and 91 years, respectively. After the fourth boost, the rat serum recognized both isolated SDS tangles and tangles in formalin-fixed sections at 1:2,OOO dilution. Five days after a final immunization, spleen cells from the rat were fused with mouse NS1 myeloma cells using the method of Kennett and distributed among six 96-well microtiter plates. Initial screening of hybridomas was performed using the ELISA method to determine the presence of rat IgG in the supernatant of the clones. Rat IgG (+) clones