Portal de Periódicos da CAPES

Familial Encephalopathy with Neuroserpin Inclusion Bodies

1999; Elsevier BV; Volume: 155; Issue: 6 Linguagem: Inglês

10.1016/s0002-9440(10)65510-1

1525-2191

Richard L. Davis, P D Holohan, Antony E. Shrimpton, Arthur H. Tatum, John Daucher, George H. Collins, Robert B. Todd, Charles B. Bradshaw, Paul F. Kent, David Feiglin, Arthur L. Rosenbaum, Mark S. Yerby, Cheng‐Mei Shaw, Felicitas Lacbawan, Daniel A. Lawrence,

Alzheimer's disease research and treatments

We report on a new familial neurodegenerative disease with associated dementia that has presented clinically in the fifth decade, in both genders, and in each of several generations of a large family from New York State—a pattern of inheritance consistent with an autosomal dominant mode of transmission. A key pathological finding is the presence of neuronal inclusion bodies distributed throughout the gray matter of the cerebral cortex and in certain subcortical nuclei. These inclusions are distinct from any described previously and henceforth are identified as Collins bodies. The Collins bodies can be isolated by simple biochemical procedures and have a surprisingly simple composition; neuroserpin (a serine protease inhibitor) is their predominant component. An affinity-purified antibody against neuroserpin specifically labels the Collins bodies, confirming their chemical composition. Therefore, we propose a new disease entity—familial encephalopathy with neuroserpin inclusion bodies (FENIB). The conclusion that FENIB is a previously unrecognized neurodegenerative disease is supported by finding Collins bodies in a small kindred from Oregon with familial dementia who are unrelated to the New York family. The autosomal dominant inheritance strongly suggests that FENIB is caused by mutations in the neuroserpin gene, resulting in intracellular accumulation of the mutant protein. We report on a new familial neurodegenerative disease with associated dementia that has presented clinically in the fifth decade, in both genders, and in each of several generations of a large family from New York State—a pattern of inheritance consistent with an autosomal dominant mode of transmission. A key pathological finding is the presence of neuronal inclusion bodies distributed throughout the gray matter of the cerebral cortex and in certain subcortical nuclei. These inclusions are distinct from any described previously and henceforth are identified as Collins bodies. The Collins bodies can be isolated by simple biochemical procedures and have a surprisingly simple composition; neuroserpin (a serine protease inhibitor) is their predominant component. An affinity-purified antibody against neuroserpin specifically labels the Collins bodies, confirming their chemical composition. Therefore, we propose a new disease entity—familial encephalopathy with neuroserpin inclusion bodies (FENIB). The conclusion that FENIB is a previously unrecognized neurodegenerative disease is supported by finding Collins bodies in a small kindred from Oregon with familial dementia who are unrelated to the New York family. The autosomal dominant inheritance strongly suggests that FENIB is caused by mutations in the neuroserpin gene, resulting in intracellular accumulation of the mutant protein. Degenerative diseases of the central nervous system (CNS) are a heterogeneous group of disorders with diverse clinical presentations, neuropathological findings, and pathogenic bases. In recent years much progress has been made in understanding the biology of these diseases. Genetic analysis of human populations, large families, and transgenic-animal models has advanced our understanding by identifying mutations in specific disease-associated genes.1Hardy J Gwinn-Hardy K Genetic classification of primary neurodegenerative disease.Science. 1998; 282: 1075-1079Crossref PubMed Scopus (307) Google Scholar, 2Price DL Sisodia SS Borchelt DR Genetic neurodegenerative disease: the human illness and transgenic models.Science. 1998; 282: 1079-1083Crossref PubMed Scopus (231) Google Scholar, 3Muller U Graeber MB Neurogenetic diseases: molecular diagnosis and therapeutic approaches.J Mol Med. 1996; 74: 71-84Crossref PubMed Scopus (14) Google Scholar, 4Giannakopoulos P Hof PR Savioz A Guimon J Antonarakis SE Bouras C Early-onset dementias: clinical, neuropathological and genetic characteristics.Acta Neuropathol. 1996; 91: 451-465Crossref PubMed Scopus (20) Google Scholar Yet the final diagnosis of most of these conditions still rests with the identification of specific gross and microscopic changes in central nervous tissue.5Hart MN Contributions of autopsy to modern neurologic science.in: Hill RB Anderson RE The Autopsy: Medical Practice and Public Policy. Butterworths, Boston1988: 91-105Google Scholar Cellular inclusions, affecting both neurons and glia, are prominent features of many of these disorders; they can be recognized by histomorphology and ultrastructure, by histochemical staining properties, and by immunohistochemical characterization.6Schochet SS Neuronal inclusions.in: Bourne GH The Structure and Function of Nervous Tissue. vol IV. Academic Press, New York, London1972: 129-177Google Scholar, 7Leigh PN Probst A Dale GE Power DP Brion JP Dodson A Anderton BH New aspects of the pathology of neurodegenerative disorders as revealed by ubiquitin antibodies.Acta Neuropathol. 1989; 79: 61-72Crossref PubMed Scopus (83) Google Scholar, 8Cooper PN Jackson M Lennox G Lowe J Mann DMA Tau, ubiquitin, and α B-crystallin immunohistochemistry define the principal causes of degenerative fronto temporal dementia.Arch Neurol. 1995; 52: 1011-1015Crossref PubMed Scopus (120) Google Scholar We have used these strategies together with elementary biochemical techniques to investigate a novel neurodegenerative disease characterized by unusual neuronal inclusions in a large kindred from New York State. This disease typically manifests itself in the fifth decade of life and is characterized by an insidious onset of cognitive decline, impairment of attention and concentration, and perseveration, and loss of daily living skills exemplified by poor judgment and lack of insight. Learning and memory are also affected but to a lesser degree than is typically seen in Alzheimer's disease. The principle neuropathological finding is the presence of round, eosinophilic, periodic acid/Schiff reagent (PAS)-positive, but diastase-resistant neuronal inclusion bodies distributed throughout the deeper layers of the cerebral cortex and in many subcortical nuclei, especially the substantia nigra. They are rarely seen in white matter. Extensive histochemical, immunohistochemical, and electron microscopic analyses show that these inclusions are distinct from any described previously.9Davis RL Daucher JW Welker DM Barcza MA Collins GH A familial dementia with unusual neuronal inclusions.J Neuropathol Exp Neurol. 1996; 55 (abstr.): 636Crossref Google Scholar The finding that these histologically unique inclusion bodies are also chemically distinct, being composed predominately of neuroserpin (a serine protease inhibitor), leads us to conclude that we have discovered a new disorder—familial encephalopathy with neuroserpin inclusion bodies (FENIB). Evidence is presented showing that FENIB also occurs in an unrelated family from Oregon. The significance of our findings is that we can now study the mechanism of self-aggregation and tissue deposition of neuroserpin, how this causes neurodegeneration, and why this presents as dementia. Ultimately this information should advance our understanding of the biology of this group of disorders, because, despite their diversity, some common principles are emerging, particularly, the role that aberrant protein expression and processing play in producing the characteristic disease phenotype. Entire brains of two affected individuals from the New York family were available for study; they were fixed in 10% neutral buffered formalin for 2 weeks before gross examination and sectioning. Blocks were obtained from representative cortical and subcortical areas, embedded in paraffin, and stained with hematoxylin and eosin (H&E), according to routine histological procedures. Prepared glass slides from another case, stained with H&E/Luxol fast blue, as well as flash-frozen sections of one hemisphere were provided by the Harvard Brain Tissue Resource Center (PHSMH31862). Portions of fixed brain tissue from an additional case, as well as fixed tissue from other organs from two cases, were obtained from regional tissue archives. Microscopic glass slides of formalin-fixed and paraffin-embedded cerebral biopsy tissue from the Oregon family were supplied by the University of Washington School of Medicine (Seattle, WA) and were the only samples available for histological study of this family. Enzyme digestion studies were conducted with porcine pancreas α-amylase, 10 mg/ml, pH 6.8, at 37°C; barley β-amylase, 0.1 mg/ml, pH 4.8, at 37°C; and Aspergillus nigra amyloglucosidase, 0.1 mg/ml, pH 4.8, at 37°C. Dimedone (5,5-dimethyl-1,3 cyclo-hexanedione) was obtained from Sigma Chemical Co., St. Louis, MO, and was used as a 5. ethanolic solution at 60°C by the procedure of Bulmer.10Bulmer D Dimedone as an aldehyde blocking reagent to facilitate the histochemical demonstration of glycogen.Stain Technol. 1959; 34: 95-98Crossref PubMed Scopus (67) Google Scholar All other histochemical procedures were performed by routine histological laboratory protocols. All light microscopy was done on an Olympus microscope. Immunostaining was performed on tissue fixed in 10% buffered formalin for at least 2 weeks and embedded in paraffin before staining. Commercially available antibodies were obtained and used as follows, for standard epitope retrieval techniques, strepavidin-biotin methodology, and 3–3′ diaminobenzidine as chromogen; ubiquitin (polyclonal, 1:400; Vector-NovoCastra, Burlingame, CA); α B-crystallin (polyclonal, 1:80; Vector-NovoCastra). amyloid-β (Aβ) (monoclonal, 1:80; Vector-NovoCastra); tau (monoclonal, 1:250; Labvision, Fremont, CA); neurofilament protein (monoclonal, 1:200; DAKO, Carpenteria, CA); phosphorylated neurofilament protein (SM1–31) (monoclonal, 1:200; Sternberger Monoclonals, Baltimore, MD); nonphosphorylated neurofilament protein (SM1–32) (monoclonal, 1:2000; Sternberger Monoclonals); synaptophysin (polyclonal, 1:500; DAKO); neuron-specific enolase (polyclonal, 1:40. Biogenex, San Ramon, CA); lysozyme (polyconal, 1:1000; Biogenex). antichymotrypsin (polyclonal, prediluted; Signet, Dedham, MA). antitrypsin (polyclonal, prediluted; Signet); actin (monoclonal, 1:400. DAKO); GFAP (polyclonal, 1:2000; DAKO); PGP9.5 (polyclonal, 1:3000. Vector); HSP27 (monoclonal, 1:100; Biogenesis, Portsmoth, NH); HSP70 (monoclonal, 1:200; Stressgen, Victoria, BC, Canada); superoxide dismutase (monoclonal, 1:8000; Boehringer Mannheim, Indianapolis, IN). amyloid-β precursor protein (APP) (monclonal, 1:100; Boehringer Mannheim); and α-synuclein (1:500; Chemicon, Temecula, CA). An affinity-purified, polyclonal rabbit antibody to recombinant human neuroserpin was produced11Hastings GA Coleman TA Haudenschild CC Stefansson S Smith EP Barthlow R Cherry S Sandkvist M Lawrence DA Neuroserpin, a brain-associated inhibitor of tissue plasminogen activator is localized primarily in neurons: implications for the regulation of motor learning and neuronal survival.J Biol Chem. 1997; 272: 33062-33067Crossref PubMed Scopus (203) Google Scholar, 12Shermann PM Lawrence DA Yang AY Vandenberg ET Paielli P Olson ST Shore JD Ginsburg D Saturation mutagenesis of the plasminogen activator inhibitor-1 reactive center.J Biol Chem. 1992; 267: 7588-7595PubMed Google Scholar and was used at a 2000-fold dilution. Lectin histochemistry studies were performed on unfixed, frozen sections of cerebral cortex from a case and an age-matched control. A total of 21 fluorescein-labeled lectins available in three kits (Vector) were applied and viewed by fluorescence microscopy on a Leitz microscope. Tissue from the brain of one affected individual, including cerebral cortex, cingulate gyrus, substantia nigra, and subcortical white matter, was processed for electron microscopy (EM). Tissue was fixed initially for at least 2 weeks in 10% buffered formalin and was subsequently fixed overnight in 2.5% glutaraldehyde, postfixed for 1 hour in 1% osmium tetroxide, and embedded in Araldite 502 from which 1-mm-thick sections were cut and stained with toluidine blue. Ultra-thin sections of selected areas were stained with lead citrate and uranyl acetate and examined with a Jeol 100 SX electron microscope. Cortical tissue (2 g) was homogenized in 10 ml of 250 mmol/L sucrose and 10 mmol/L ethylenediaminetetraacetic acid, buffered to pH 7.4 with 10 mmol/L 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid containing 10 μl of Sigma protease inhibitor cocktail. The homogenate was filtered through gauze, and the retentate was washed with an additional 5 ml of homogenization medium. Verification that the inclusion bodies withstood the homogenization was accomplished by visualization using light microscopy after H&E staining by air drying a sample on a glass slide followed by fixing for 1 minute in absolute ethanol before H&E. Similarly, the assay used to monitor the fractionation of the inclusions throughout the isolation procedure was light-microscopic visualization after H&E staining. The inclusion bodies were isolated from the homogenate as follows: the homogenate was centrifuged at 1000 × g for 10 minutes, the supernatant decanted, and the soft pellet resuspended in 5 ml of homogenization medium; 5 ml of 1% sarcosyl in homogenization medium was added, the suspension was incubated at 37°C for 30 minutes with frequent mixing and centrifuged at 45,000 × g for 20 minutes; the pellet was resuspended in 10 ml of the 1% sarcosyl solution and incubated at 37°C for 30 minutes with frequent mixing and again subjected to centrifugation at 45,000 × g for 20 minutes; the pellet was resuspended in 5 ml of homogenization medium, 5 ml of collagenase (2 mg/ml) was added, and the suspension was incubated at 37°C for 60 minutes followed by centrifugation at 45,000 × g for 20 minutes; and the resultant pellet was resuspended in 0.5 ml (1/20 original volume) of 4% sodium dodecyl sulfate (SDS), 125 mM Tris-HCl (pH 6.8), 20% glycerol, 10. mercaptoethanol and heated at 75°C for various periods of time. After heating, the samples were subjected to centrifugation at 14,000 × g for 10 minutes, and the supernatant was removed and analyzed by SDS-polyacrylamide gel electrophoresis (PAGE). The samples were diluted 2:1 in 4% SDS, 20% glycerol, 10. mercaptoethanol, 50 mmol/L Tris-HCl (pH 6.8) before being subjected to SDS-PAGE in a Bio-Rad 7.5% Redi-gel in Tris/glycine/SDS buffer at 150 V for 1.25 hours. Fifty microliters of the various samples were analyzed. After electrophoresis the gels were stained with Coomassie blue and destained by standard procedures. After SDS-PAGE the samples were electrophoretically transferred to a polyvinylidene difluoride membrane at 250 mA for 1.5 hours and stained with Coomassie blue. The membrane was sent to the Biotech Center at Cornell University, Ithaca, NY, where the 57-kd band was excised and either sequenced directly for N-terminal analysis or first subjected to endoproteinase C digestion (EC 3.4.99.30). The resulting peptides were separated by reverse-phase chromatography, and selected peaks were then sequenced. All of the routine chemicals (Tris, SDS, glycine, N-lauroylsarcosine), as well as the enzymes, were obtained from Sigma. The classification of FENIB as a novel neurodegenerative disorder is based on a battery of criteria including neuropathological, biochemical, and immunohistochemical analyses. A clinical investigation of the New York family is continuing. Thus far we have found the disease in two unrelated families. The findings, taken in total, describe a new neurological disorder that is associated with familial presenile dementia. Two cases from the New York family were available for a complete neuropathological investigation of the brain. Both individuals, siblings aged 64 and 57 with clinical dementias of approximately 10 years' duration, showed normal appearing sulci and gyri without atrophy and with transparent leptomeninges. Brain weights were 1530 and 1400 g, respectively. Examination of the bases of the brains disclosed no vascular malformations or significant atherosclerotic changes. Multiple coronal sections demonstrated slight ventricular dilatation together with normal appearing cortical gray, subcortical white, and deep gray matter structures free of encephalomalacia, hemorrhage, or discoloration. In both cases the substantia nigra, brainstem, and cerebellum were grossly unremarkable (data not shown). Sections of brain obtained from these two individuals and from a limited amount of autopsy material obtained from two additional family members were studied using standard histopathological methods. The neuropathological findings are presented in Figure 1, Figure 2 and in Table 1, Table 2, Table 3. Sections of brain stained with H&E were examined by light microscopy. At scanning magnification the neocortex shows mild microvacuolar change and gliosis affecting cortical layer II, principally in sections of the frontal, temporal, and parietal cortices (Figure 1A). Numerous, eosinophilic, sharply defined, round-to-oval inclusion bodies of variable size (approximately 5–50 μm) are scattered throughout the deeper layers of the cerebral cortex primarily in layers III to V (Figure 1A). They occur both singly and in clusters of three or more (Figure 1, B and C). Many of these inclusions are homogeneous in appearance, although others have either a dark core with a lighter halo or a variegated speckled appearance; a thin rim of darker material surrounds a few of them (Figure 1, B and C). Some of the inclusions are found within neuronal cell bodies, but many appear to lie free in the neuropil within vacuoles. The majority of cortical neurons are free of this involvement. Inclusions are very infrequently noted in the cerebral white matter. Deep gray matter is variably affected with marked involvement of the substantia nigra where many of the pigmented neurons show one or more inclusion bodies within their perikarya. However, neither neuronal loss nor gliosis are apparent. Many of the neurons affected with these bodies show no other abnormalities; however, others are grossly distorted, having no apparent cytoplasm, a displaced and compressed nucleus, and possibly an overall reduction in cell size. Neuronal loss is evident only in cortical layer II, and glial involvement is not seen except for the reactive change in layer II, as well as a low-grade subcortical gliosis. Histological examination of other organs disclosed only extensive myocardial fibrosis, severe coronary atherosclerosis, and pulmonary edema in one case, and hepatic steatosis in another case (data not shown).Figure 2Togographic distribution of Collins bodies. Coronal sections of the cerebral hemispheres: frontal (A), frontotemporal (B and C), and parieto-occipital (D). Transverse sections: midbrain (E), pons and cerebellum (F and G), and medulla oblongata and cerebellum (H). The composite figure summarizes data obtained from four autopsy cases of FENIB. (Templates redrawn and modified with permission from the authors of 52Esiri MM Oppenheimer DR Diagnostic Neuropathology. Blackwell Scientific, London1989: 12-40Google Scholar).View Large Image Figure ViewerDownload Hi-res image Download (PPT)Table 1Anatomical Distribution of Inclusion BodiesAreaOccurrenceSpecific location/commentsCerebral cortex+++In small to medium-sized neurons primarily in layers III through VInsular cortex+++Cingulate gyrus+++In small to medium size neurons in deeper layersClaustrum+Substantia innominata−Hippocampal formationAmmon's horn+Dentate gyrus−Subiculum and parahippocampus++Amygdala++Scattered inclusionsThalamus and hypothalamus+Scattered inclusions in medial structures extending from mammillary nuclei to the region of the parafascicular nucleusBasal gangliaCaudate+In small neuronsPutamen++In small neuronsGlobus pallidus+In small neuronsMidbrainSubstantia NigraParsCompacta+++Large or multiple intraneuronal inclusions in pigmented neuronsPar Retuculta−Red Nucleus++Nucleus of CN III+PonsLocus ceruleus+Motor nucleus of CN V+Medulla oblongata+Rare inclusion noted, including nucleus of CN XII and dorsal vagal nucleusCerebellum−Spinal CordNot examined Open table in a new tab Table 2Histochemical Study of Inclusion BodiesStainResultInterpretationH&EPink/deep redEosinophilicPASHomogeneous, magenta stainingVicinyl hydroxylsPAS after α-amylase+Vicinyl hydroxyls—not glycogenPAS after dimedone−Not glycogenBest's carmine−Not glycogenHigh-iron diamine−Not sulfatedMucopolysaccharideAlcian blue−Not acidic mucopolysaccharideCresyl violetBluePrussian blue−No ironVon Kossa−No calciumAlizarin red S−No calciumOil red-O (unfixed, frozen tissue)−Not neutral lipidSudan black B (unfixed, frozen tissue)−Not neutral lipidBodianGray-blackBielschowskyBrown (diffuse plaques noted in one case)Inconsistent with Alzheimer's diseaseCongo redOrangeNo birefringence under polarized lightNot amyloid depositsFormalin-fixed, paraffin-embedded tissue was used, except where indicated. Open table in a new tab Table 3Immunohistochemistry of Inclusion BodiesAntigenImmunohistochemistry resultRemarksTau−Not Pick bodiesAlpha-synuclein−Not Lewy bodiesUbiquitin−No evidence of ubiquitin degradation pathwayPGP9.5−No evidence of ubiquitin degradation pathwayAlpha B-crystallin−Heat shock proteins not detectedHSP 27−Heat shock proteins not detectedHSP 70Heat shock proteins not detectedSuperoxide dismutase−No evidence of ALS-related inclusionsβ-Amyloid−Inconsistent with Alzheimer's diseaseβ-Amyloid precursor protein−Inconsistent with Alzheimer's diseaseGFAP−Mild cortical and subcortical gliosisActin−Not Hirano bodiesNeurofilament proteins−Synaptophysin−Neuron-specific enolase−Lysozyme−Antitrypsin−Antichymotrypsin−Formalin-fixed, paraffin-embedded tissue was used for all the following experiments. Open table in a new tab Formalin-fixed, paraffin-embedded tissue was used, except where indicated. Formalin-fixed, paraffin-embedded tissue was used for all the following experiments. Sections of brain were examined by a variety of histochemical methods to elucidate the nature of the inclusion bodies. The results are shown in Figure 1 and in Table 2. The inclusions are eosinophilic (Figure 1, A-D; Table 2) and are strongly and uniformly stained by the PAS method (Figure 1E). Furthermore, they retain their PAS reactivity after digestion by α-amylase or other glycosidases for up to 48 hours at 37°C (Table 2). Conversely, preincubation in 5% ethanolic dimedone, an aldehyde blocker, inhibits the PAS reaction (Table 2). The inclusions fail to stain with Best's carmine (Table 2). These findings suggest that the inclusion bodies contain a carbohydrate component other than glycogen. Lack of staining with either the Alcian blue or the high-iron diamine method shows that the carbohydrate is not an acidic or a sulfated mucopolysaccharide (Table 2). Stains for neutral lipid, including Sudan black B and oil red O, are negative on unfixed frozen tissue (Table 2). Additional histochemical studies show that the inclusions are negative for iron and for calcium (Table 2). The inclusions were further analyzed using a screening panel of fluorescein isothiocyanate-conjugated lectins (Figure 1, G and H). By fluorescence microscopy, the presence of neuronal lipofuscin was apparent by its bright autofluorescence. Lectins showing moderate to strong affinity for the bodies included those with relative specificity for mannose (Figure 1G) and for N-acetylglucosamine (Figure 1H). In summary, these findings suggest that the inclusions are composed of a glycoprotein or glycoproteins containing N-linked oligosaccharides. Immunohistochemical methods were used to probe for known components of other well-characterized neuronal inclusions. The results are presented in Figure 1 and in Table 3. As shown (Table 3; Figure 1F) the inclusions are negative for both ubiquitin and α-synuclein, and thus they are distinguishable from Lewy bodies13Lennox G Lowe J Landon M Byrne EJ Mayer RJ Godwin-Austen RB Diffuse Lewy body disease: correlative neuropathology using anti-ubiquitin immunocytochemistry.J Neurol Neurosurg Psychiatry. 1989; 52: 1236-1247Crossref PubMed Scopus (180) Google Scholar, 14Fukuda T Tanaka J Watabe K Numoto RT Minamitani M Immunohistochemistry of neuronal inclusions in the cerebral cortex and brain-stem in Lewy body disease.Acta Pathol Jpn. 1993; 43: 545-551PubMed Google Scholar, 15Spillantini MG Schmidt ML Lee VML Trojanowski JQ Jakes R Goedert M Alpha-synuclein in Lewy bodies.Nature. 1997; 388: 839-840Crossref PubMed Scopus (6414) Google Scholar; they contain neither Aβ nor other abnormal proteins or peptides characteristic of Alzheimer's disease. Furthermore, none of the following proteins are detected: intermediate filaments or other cytoskeletal proteins, PGP9.5, α B-crystallin, heat shock proteins 27 or 70, superoxide dismutase, or tau (Table 3). Neither Pick bodies nor ballooned neurons are seen. Tau-positive glial inclusions are not identified. Rare tau-positive neurofibrillary tangles are observed in the hippocampal formation. An antibody to synaptophysin strongly labels the cortical neuropil but not the inclusions. Thus, the inclusion bodies described here appear to be distinctly different from any described previously; therefore we propose the name Collins bodies to identify these unique inclusions. Their neuroanatomical mapping is shown in Figure 2. By transmission EM, Collins bodies appear as moderately osmiophilic round globules with an amorphous to finely granular texture (Figure 3). They have little internal structure, aside from some speckling with darker material, and a few demonstrate a darker inner core contrasting with a lighter outer shell. Most are well delimited at their periphery, sometimes crowding aside other cytoplasmic organelles; others appear to be enclosed within a limiting membrane of rough endoplasmic reticulum (Figure 3). The Collins bodies were isolated from unfixed, frozen cerebral cortex tissue by homogenization in a buffered isotonic sucrose solution, washed with detergent (N-lauroylsarcosine), digested with collagenase, washed, and collected by centrifugation (Figure 4; see Materials and Methods for complete details). The presence of Collins bodies was monitored throughout the isolation procedure by light microscopy after H&E staining; as noted, they are eosinophilic (Figure 5, A-C). Clearly, a highly enriched preparation is obtained (the H&E stain of the enriched fraction that is labeled fraction P5 in Figure 4 is presented in Figure 5 C), although the bodies appear to be damaged somewhat (compare Figures 5A and 5C).Figure 5Visualization of inclusions by histochemistry and immunohistochemistry. The various fractions obtained during the isolation scheme were assayed for the presence of the inclusions by H&E staining (original magnification, ×400) (A–C). A: Starting material (unfixed, frozen brain tissue). B: Fraction labeled P1 in Figure 4. C: Fraction identified as P5 in Figure 4. The presence of neuroserpin was detected with a neuroserpin-specific polyclonal antibody (D–F. original magnification, ×400). D: Cortical tissue from an age- and gender-matched control; E: Cortical tissue from an affected individual; F: Fraction labeled P5 in Figure 4. Tissue from an affected individual from the Oregon family. G: Cortical tissue stained with H&E (original magnification, ×400). H: Cortical tissue stained with neuroserpin antibody (original magnification, ×400).View Large Image Figure ViewerDownload Hi-res image Download (PPT) In preliminary experiments the fraction enriched with Collins bodies (fraction P5) was subjected to routine SDS-PAGE analysis, ie, heating for 5 minutes in boiling water in a 2. SDS-reducing buffer. However, this treatment released very little protein detectable by Coomassie blue staining (Figure 6, lane 2 represents SDS-PAGE of fraction S6 obtained by heating for 5 minutes). Moreover, microscopic examination revealed that Collins bodies remained intact and could be collected by centrifugation (fraction P6). Therefore, the Collins body-enriched fraction (P5) was treated a second time and subjected to harsher conditions, heating for 2 hours at 75°C in 4. SDS. By this treatment the Collins bodies were disrupted, and SDS-PAGE analysis showed one prominent protein band at 57 kd (Figure 6, lane 3, is fraction S6 obtained after prolonged heating). A 57-kd protein was not found in an identically treated specimen obtained from an age- and gender-matched control (Figure 6, lane 4). The results show that Collins bodies are composed primarily of a single protein. The identity of the 57-kd protein was determined by transferring it to a polyvinylidene difluoride membrane and subjecting it to amino acid sequence analysis before and after proteolytic digestion (see Materials and Methods). The sequence information was analyzed using the BLAST program and gave convincing evidence that the 57-kd protein is neuroserpin (Figure 7), a serine protease inhibitor synthesized primarily in the CNS. Differences between the amino acid composition of the isolate and the wild-type neuroserpin were found: 1) asparagine 100 in the wild type is identified as aspartic acid in the isolate (most likely, this difference is an artifact because, when an asparagine residue is followed by a glycine in the primary structure, the asparagine frequently becomes deamidated during the sequencing procedure), and 2. the N-terminal amino acid of the isolate corresponds to residu