Journal of the Neurological Sciences, 112 (1992) 46-50 © 1992 Elsevier Science Publishers B.V. All rights reserved 0022-510X/92/$05.00
46
JNS 03856
Ubiquitin in cerebral amyloid angiopathy F61ix F. Cruz-Sanchez
a,
Concepci6 Marin b, Marco L. Rossi c, A. Cardozo and Eduardo Tolosa b
a,
Isidro Ferrer d
a Neurological Tissue Bank, Hospital Clinic i Provincial, Universityof Barcelona, Barcelona, Spain, b Service of Neurology, Hospital Clinic i Provincial, Barcelona, Spain, c Department of Neuropathology, Midland Centrefor Neurology & Neurosurgery (MCNN), Birmingham, UK, and d Neuropathology Unit, Hospital Principes de Espa~a, Hospitalet de Llobregat, Spain
(Received 26 August, 1991) (Revised, received 8 May, 1992) (Accepted 20 May, 1992) Key words: Cerebral amyloid angiopathy; Ubiquitin; Alzheimer's disease; Neurodegeneration
Summary Immunohistological findings in cerebral blood vessels of 4 cases with cerebral amyloid angiopathy (CAA) were compared with those of 4 Alzheimer's (AD) cases. A panel of antibodies against 2 neurofilament subunits (BF10 and RT97), a microtubule-associated protein (TAU) and ubiquitin were used. CAA cases showed a strong immunoreactivity for ubiquitin in blood vessel wall. Senile plaques (SPs) in CAA cases showed strong ubiquitin positivity but the central amyloid core was negative. AD brains showed immunoreactivity with all antibodies in SPs and neurofibriilary tangles (NFTs); blood vessels were consistently negative for ubiquitin. Control brains showed few SPs and NFTs; these were positive for ubiquitin, but blood vessels were negative. These results indicate that vascular amyloid deposition in CAA and AD may have different pathophysiological mechanisms.
Introduction
Cerebral amyloid angiopathy (CAA) is a microangiopathy of unknown cause featuring non-traumatic and non-hypertensive cerebral hemorrhage (Gilbert and Vinters 1983; Vinters 1987; Vonsattel et al. 1991) associated with progressive dementia (Glenner et al. 1981). Cases of CAA without hemorrhages have also been described (Vonsattel et al. 1991). Morphologically, most cases of CAA consist in the deposition of a type of amyloid (A4beta protein) in the wall of cerebral blood vessels, which can be recognized using Congo red staining. Other changes such as senile plaques (SPs) with a prominent amyloid core and few neurofibrillary tangles (NFTs) have also been described (Vinters et al. 1988). The finding of SPs and NFTs may point towards a close relationship of this condition with .Alzheimer's disease (AD). Results obtained with a
Correspondence to: F.F. Cruz-Sanchez, MD, Banco de Tejidos Neurolbgicos, Servicio de Neurologia, Hospital Clinico i Provincial, Villarroel 170, 08036 Barcelona, Spain. Fax: (3) 4515272; Phone: (3) 454 60 00.
number of antibodies against these structures support this contention (Vinters et al. 1988; Maruyama et al. 1990). Immunohistological studies using antibodies for ubiquitin, a 8.6 kDa polypeptide related to normal short-lived, and abnormal ATP-dependent protein degradation (Mori et al. 1987), have demonstrated similar antigenic determinants in senile plaques (SPs) of CAA and AD (Dickson et al. 1989, 1990). The present study describes immunohistochemical findings in the cerebral blood vessels in cases of CAA. Our results may give a new perspective on the pathogenesis of both CAA and AD. Material and methods
Formalin-fixed, paraffin-embedded brain tissue from 12 patients pathologically diagnosed as suffering from AD (4 cases), sporadic non-familial CAA (4 cases), 2 elderly patients without neurological disorder, and 2 patients with intracerebral non-CAA-associated hemorrhage were studied. The mean age of the CAA patients was 60 years (range 54-73) and that of AD cases was 71.5 years (range 58-73). The mean age of
47 control cases was 71 years (range 70 to 72). Female and male patients were equally represented. Patients with a pathological diagnosis of CAA were not hypertensive and had not sustained head injury. Two of these patients had shown mild to moderate symptoms of cognitive deterioration which was not sufficient for a diagnosis of AD to be made. They died because of intracerebral hemorrhage or hemorrhagic cerebral infarct. Pa-
tients with AD showed a progressive dementia with a mean evolution span of 8 years. Patients died of nonneurological complications. Brain tissue examined included samples from all regions of cerebral cortex, hippocampus, all of the basal ganglia nuclei and multiple cerebellum and brain stem samples. Brain tissue from hemorrhagic areas from cases of CAA and 2 control cases with intracere-
Fig. 1. Sections of CAA brain immunostained with and antibody ~ ubiquitin showing strong positivity in the media and to a lesser extent in the adventitia of arterlo,~-~ ~-,~,-~i~,~-~. ~, ~ •
48 bral non-CAA-associated hemorrhage was also examined. Criteria for the diagnosis of AD were not fulfilled in 2 cases of CAA with minor cognitive deterioriation (Khachaturian 1985; Wisniewski et al. 1989). These, together with the other 2 CAA cases without cognitive impairment were called "pure" CAA. Seven #~m thick paraffin sections were stained with hematoxylin and eosin (H&E), alkaline Congo red method (CR) and Bielschowsky's silver impregnation (Khachaturian 1985). CR sections were examined by conventional light microscopy and polarized light. The avidin biotin complex method was used for the polyclonal antibodies to ubiquitin (Dakopatts) and TAU (Sigma) and for the monoclonal antibodies to the 155 kDa (BF10) and 210 kDa (RT97) neurofilament subunits (Boehringer, Mannheim). Substantia nigra sections from two cases of idiopathic Parkinson's disease showing conspicuous Lewy bodies were used as positive control for ubiquitin.
Results
Macroscopically, brains from patients with CAA showed hemorrhagic infarction and intracerebral hemorrhage in atypical locations such as the cortical and subcortical areas in occipital, frontal a n d / o r parietal lobes. Histologically, these lesions consisted of necrotic areas at different stages of evolution. The most notable feature was the presence of obvious microangiopathy
-
of medium and small size subarachnoid and neocortical blood vessels resulting in thickening of the walls These blood vessels were also observed around ischemic/hemorrhagic areas. CR sections showed moderate staining of blood vessel walls and the classical apple-green birefringence when examined under polarized light. CR and Bielschowsky's stains also showed cortical and hippocampal SPs with an amorphous core. Macroscopically, AD brains showed ventricular enlargement and moderate to severe diffuse cortical atrophy especially posteriorly. Histologically, cerebral cortex showed severe neuronal loss, gliosis and large numbers of SPs of different types and NFTs, both in superficial and deep layers. These changes were found in frontal, temporal and occipital lobes as well as in the hippocampus. These findings are in agreement with the histological criteria for AD (Khachaturian 1985; Wisniewski et al. 1989). Some subarachnoid and cortical medium size blood vessels showed thickening of the wall and strong birefringence under polarized light with CR stains. Control brains showed mild to moderate diffuse cortical atrophy. Hi=tologically, there was moderate neuronal loss and very few SPs and NFTs in cerebral cortex on CR staining. CAA cases showed strong positivity for ubiquitin in leptomeningeal and cortical blood vessel walls; these vessels were Congo red-positive in serial sections. Several patterns of immunostaining were observed. In most cases, the positivity was located in the media of blood vessels (Fig. 1A and B) and in the subendothelial
W
o Q ~
b0 q lt
O O
i.
q
B
Ir
O
Fig. 2. Section of CAA brain immunostained with an antibody to ubiquitin. Note the granular positive material surrounding blood vessels (arrows) (ubiquitin, × 150).
49 space. In some cases, marked immunoreactivity was seen in foci in the perivascular space and in the neuropil around blood vessels (Fig. 2). These findings were seen in all CAA samples studied including those from the cerebral cortex far away from hemorrhagic or necrotic lesions. Positivity for neurofilaments and TAU was not observed. SPs in cases of CAA also showed strong positivity for ubiquitin. The pattern was granular and fibrillary and most plaques showed a negative central core which was composed of amyloid component. Very few NFTs were ubiquitin positive in cases of CAA. Both SPs and NFTs were also positive for TAU and both neurofilament subunits. In AD and control brains including two cases with intracerebral non-CAA-associated hemorrhage, cerebral blood vessels were consistently negative with all antibodies. In AD cases, SPs and NFTs showed immunoreactivity for ubiquitin, TAU and neurofilaments. Control brains showed very few SPs and NFTs which were positive with both neurofilament subunits, TAU and ubiquitin. Positivity for ubiquitin in SPs from CAA, AD and control cases was localized to the neurites, but not in the amyloid component.
Discussion
The clinicopathologic entity named CAA has been recognized as a distinct nosological entity on the basis of its pathology. It has been recognized that there are both sporadic and familial types of CAA with the Dutch type of A4beta amyloid associated with a gene mutation and a variant of CAA due to the deposition of cystatin C (Icelandic familial cerebral hemorrhage) as a result of gene mutation (Benson and Cohen 1977; Glenner 1980; Goren et al. 1980; Lofberg et al. 1987; Vinters 1987; Haan et al. 1990a,b). Microscopically, CAA may share some of the features of AD (Vinters 1987). Amyloid infiltrates the media and adventitia of the microvasculature, preferentially of leptomeningeal and superficial cortical blood vessels (Vinters 1987; Vinters et al. 1990). In the present study of sporadic non-familial CAA cases, amyloid deposits were found in the media and adventitia of leptomeningeal and superficial cortical vessels (Vinters and Gilbert 1983). These deposits showed strong positivity for ubiquitin, a finding which, to our knowledge, has not been reported. Dickson et al. (1989) found strongly perivascular ubiquitin positive structures surrounding cortical vessels. In agreement with these authors, we thought that these elements were of neuritic derivation. Antibodies to A4 amyloid protein have been used to study histological lesions in CAA and AD (Masters et al. 1985; Wong et al. 1985). Cases of "pure" CAA were
distinguishable from cases with AD in that the former did not show the number of plaques necessary to make a diagnosis of AD on Bielschowsky's stain. We cannot exclude that the 2 cases with CAA and mild cognitive impairment could have been early cases of AD but we followed established clinical and pathological criteria (Khachaturian 1985; Wisniewski et al. 1989) which had to be met before a case was assigned to the AD group. Reactivity with antibodies to neurofilament subunits, TAU and ubiquitin has been found in SPs and NFTs of AD and other neurodegenerative disorders of the CNS (Rasool et al. 1984; Delacourte and Defossez 1986; Papolla et al. 1989; Suenaga et al. 1990). These findings suggest that these disorders may have similar physiopathogenic mechanisms (Shankar et al. 1989). However, NFTs from patients with progressive supranuclear palsy appear to be negative for ubiquitin thus suggesting that a different mechanism is involved in their formation (Lennox et al. 1988; Leigh et al. 1989). Amyloid deposits found in vessels from AD cases in the present study did not show reactivity for ubiquitin which suggests differing mechanisms of formation in CAA and AD. However, it could also be possible that our findings reflect the overall less prominent (and/or advanced) degree of blood vessel compromise (amyloid deposits) in the small number of AD patients studied. The ubiquitin-dependent metabolism in vessels of CAA cases may play a significant role in the pathogenesis of the disease. While blood vessel amyloid deposits in AD may be considered to be a secondary phenomenon, some authors (Alizadeh-Khiavi et al. 1991; Chronopoulos et al. 1991) have demonstrated ubiquitin bound to extracellular amyloid and proposed that in AD ubiquitin has enhancing activity With respect to the deposition of amyloid. We did not think that positivity for ubiquitin in CAA blood vessel wall could be attributed to a stress response secondary to vascular insult because such positivity was seen in vessels which were not directly related to hemorrhage or necrosis. Also, cases with intracerebral non-CAA-associated hemorrhage did not show ubiquitin positivity in blood vessels. These observations may suggest that in "pure" CAA cases there is a primary pathological process inducing ubiquitin production which may lead to enhanced amyloid deposition. We conclude that this mechanism is unlikely to operate in AD as blood vessels were ubiquitin-negative. The anti-ubiquitin polyclonal antibody (Dakopatts) used in the present study reacts in a broad zone corresponding to the conjugates of ubiquitin and intracellular proteins when tested on extracts of human, mouse and rat cells in immunoblotting. Further work is needed to establish whether the immunoreactive material demonstrated in CAA blood vessel walls in this study has identity with ubiquitin protein conjugate af-
50 ter extraction. Ultrastructural studies of the localization of the antigen in CAA blood vessel walls would also be of interest. In this study we were not able to co-localize A4 protein and ubiquitin, but this is of obvious importance. Acknowledgements gaci6n Sanitaria de Trustees of MCNN. Interdepartamental fellowship) and Ms. assistance.
This work was supported by Fondo de lnvestila Seguridad Social (FISss) 90/E0593 and the We are grateful to Ms. M. Mencia (on Comissi6 de Recerca i Innovacio Tecnologica, CIRIT, J.R. Buller (MCNN) for tile excellent technical
References Alizadeh-Khiavi, K., J. Normand, S. Chronopoulos and Z. All-Khan (1991) Alzheimer's disease brain-derived ubiquitin has amyloidenhacing factor activity: behavior of ubiquitin during accelerated amyloidogenesis, Acta Neuropathol. (Berl.), 81: 280-286. Benson, M.D. and A.S. Cohen (1977) Generalized amyloid in a family of Swedish origin. A study of 426 family members in seven generations of a new kinship, nephropathy and central nervous system involvement, Ann. Intern. Med., 86: 419-424. Chronopoulus, S., K. Alizadeh-Khiavi, J. Normand, and Z. All-Khan (1991) Binding of ubiquitin to experimentally induced murine AA amyloid, J. Pathol., 163: 199-203. Delacourte, A. and A. Defossez (1986) Alzheimer disease: tau proteins the promoting factors of microtubule assembly, are major components of paired helical filament, J. Neurol. Sci., 76: 173186. Dickson, D.W., M. C.,T.stal, L.A. Mattiace, Y. Kress, A. Schwagerl, H. Ksiezak-Reding, P. Davies and S.H. Yen (1989) Diffuse Lewy body disease: light and electron microscopic immunocytochemistry of senile plaques, Acta Neuropathol., 78: 572-584. Dickson, D.W., A. Wertkin, L.A. Mattiace, E. Fier, Y. Kress, P. Davies and S.H. Yen (1990) Ubiquitin immunoelectron microscopy of dystrophic neurites in cerebellar senile plaques of Alzheimer's disease, Acta Neuropathol., 79: 486-493. Gilbert, J.J. and H.V. Vinters (1983) Cerebral amyloid angiopathy: incidence and complications in the aging brain. !. Cerebral hemorrhage, Stroke, 14: 915-923. Glenner, G.G. (1980) Amyloid deposits and amyloidosis. The Beta fibrilloses. I1. N. Engl. J. Med., 24: 1333-1343. Glenner, G.G., J.H. Henry and S. Fujihara (1981) Congophilic angiopathy in the pathogenesis of Alzheimer's degeneration, Ann. PathoL, 1: 120-129. Goren, H., M.C. Steinber and G.H. Farboody (1980) Familial o c u Ioleptomeningeal amyloidosis, Brain, 103: 473-495. Haan, J., P.R. Algra and R.A.C. Roos (1990a) Hereditary cerebral hemorrhage with amyloidosls.Dutch type, Arch. Neurol., 47: 649-653. Haan, J., J.B.K. Lanser, I. Zijderveld, G.F. van der Does and R.A.C. Roos (1990b) Dementia in hereditary cerebral hemorrhage with amyloidosis-Dutch type, Arch. Neurol., 47: 965-967. KhachaturJan, Z.S. (1985) Diagnosis of Alzheimer's disease, Arch. Neurol., 42: 1097-1105. Leigh, P.N., A. Probst, G.E. Dale, D.P. Power, J.P. Brion, A. Dodson and B.H. Anderton (1989) New aspects of the pathology
of neurodegenerative disorders as revealed by ubiquitin antibodies, Acta Neuropathol., 79: 61-72. Lennox, G., 3. Lowe, K. Morrell, M. Landon and R.J. Mayer (1988) Ubiquitin is a component of neurofibrillary tangles in a variety of neurodegenerative diseases, Neurosci. Lett., 94: 211-217. Lot'berg, H., A.O. Grubb, E.K. Nilsson, O. Jensson, G. Gadmundsson, H. Bl6ndal, A. Arnason and L. Thorsteinsson (1987) Immunocytochemical characterization of the amyloid deposits and quantitation of pertinent cerebrospinal fluid proteins in hereditary cerebral henm~'rhage with amyloidosis, Stroke, 18: 431-440. Maruyama, K., S. Ikeda, T. Ishihara, D. Allsop and N. Yanagisawa (1990) Immunohistochemical characterization of cerebrovascular amyloid in 46 autopsied cases using antibodies to beta-protein and cystatin C, Stroke, 21: 397-403. Masters, C.L., G. Simms, N.A. Weinman, G. Mulraup, B.L. McDonald and K. Beyreuther (1985) Amyloid plaque core protein in Alzheimer's disease and Down syndrome, Proc. Natl. Acad. Sci. USA, 82: 4245-4249. Mori, H., J. Kondo and Y. lhara (1987) Ubiquitin is a component of paired helical filaments in Alzheimer's disease, Science, 235: 1641-1644. Pappolla, M.A., R. Omar and B. Saran (1989) The normal brain. Abnormal ubiquitinated deposits highlight an age-related protein change, Am. J. Pathol., 135: 585-591. Rasool, C.G., C. Abraham, B.H. Andeton, M. Haugh, J. Khan and D.J. Salkoe (1984) Alzheimer's disease: immunoreactivity of neurofibrillary tangles with antineurofilament and anti-paired helical filament antibodies, Brain Res., 310: 249-260. Shankar, S.K., R. Yanagihara, R.M. Garruto, I. Grundke-Iqbal, K.S. Kosik and D.C. Gajdusek (1989) lmmunocytochemical characterization of neurofibrillary tangles in amyotrophic lateral sclerosis and Parkinson-dementia of Guam, Ann. Neurol., 25: 146-151. Suenaga, T., A. Hirano, J.F. Llena, H. Ksiezak-Reding, S.H. Yen and D.W. Dickson (1990) Modified Bielschowsky and immunocytochemical studies on cerebellar plaques in Alzheimer's disease, J. Neuropathoi. Exp. Neurol., 49: 31-40. Vinters, H.V. (1987) Cerebral amyloid angiopathy. A critical review, Stroke, 18: 311-324. Vinters, H.V. and J.J. Gilbert (1983) Cerebral amyloid angiopathy; incidence and complications in the aging brain. II. The distribution of amyloid vascular changes, Stroke, 14: 924-928. Vinters, H.V., W.M. Pardridge and J. Yang (1988) Immunohisto. chemical study of cerebral amyloid angiopathy: use of an antiserum to a synthetic 28 amino-acid peptide fragment of the Alzheimer's disease amyloid precursor, tlum. Pathol., 19: 214222. Vinters, H.V., D.L. Secor, W.M. Pardridge and F. Gray (1990) lmmunohistochemical study of cerebral amyloid angiopathy. III. Widespread Alzheimer A4 peptide in cerebral microvessei walls colocalizes with gamma trace in patients with leukoencephalopathy, Ann. Neurol., 28: 34-42. Vonsattel, J.P.G., R.H. Meyers, E.T. Hedley-Whyte, A.H. Ropper, E.D. Bird and E.P. Richardson (1991) Cerebral amyloid angiopathy without and with cerebral hemorrhages: a comparative histological study. Ann. Neurol., 30: 637-649. Wisniewski, H.M., A. Rake, W. Zigman and W. Silverman (1989) Neuropathological diagnosis of Alzheimer's disease, J. Neuropathol., 48: 606-609. Wong, C.W., V. Quaranta and G.G. Glenner (1985) Neuritic plaques and cerebrovascular amyloid in AIzheimer's disease are antigenically related, Proc. Natl. Acad. Sci. USA, 82: 8279-8732.
46
JNS 03856
Ubiquitin in cerebral amyloid angiopathy F61ix F. Cruz-Sanchez
a,
Concepci6 Marin b, Marco L. Rossi c, A. Cardozo and Eduardo Tolosa b
a,
Isidro Ferrer d
a Neurological Tissue Bank, Hospital Clinic i Provincial, Universityof Barcelona, Barcelona, Spain, b Service of Neurology, Hospital Clinic i Provincial, Barcelona, Spain, c Department of Neuropathology, Midland Centrefor Neurology & Neurosurgery (MCNN), Birmingham, UK, and d Neuropathology Unit, Hospital Principes de Espa~a, Hospitalet de Llobregat, Spain
(Received 26 August, 1991) (Revised, received 8 May, 1992) (Accepted 20 May, 1992) Key words: Cerebral amyloid angiopathy; Ubiquitin; Alzheimer's disease; Neurodegeneration
Summary Immunohistological findings in cerebral blood vessels of 4 cases with cerebral amyloid angiopathy (CAA) were compared with those of 4 Alzheimer's (AD) cases. A panel of antibodies against 2 neurofilament subunits (BF10 and RT97), a microtubule-associated protein (TAU) and ubiquitin were used. CAA cases showed a strong immunoreactivity for ubiquitin in blood vessel wall. Senile plaques (SPs) in CAA cases showed strong ubiquitin positivity but the central amyloid core was negative. AD brains showed immunoreactivity with all antibodies in SPs and neurofibriilary tangles (NFTs); blood vessels were consistently negative for ubiquitin. Control brains showed few SPs and NFTs; these were positive for ubiquitin, but blood vessels were negative. These results indicate that vascular amyloid deposition in CAA and AD may have different pathophysiological mechanisms.
Introduction
Cerebral amyloid angiopathy (CAA) is a microangiopathy of unknown cause featuring non-traumatic and non-hypertensive cerebral hemorrhage (Gilbert and Vinters 1983; Vinters 1987; Vonsattel et al. 1991) associated with progressive dementia (Glenner et al. 1981). Cases of CAA without hemorrhages have also been described (Vonsattel et al. 1991). Morphologically, most cases of CAA consist in the deposition of a type of amyloid (A4beta protein) in the wall of cerebral blood vessels, which can be recognized using Congo red staining. Other changes such as senile plaques (SPs) with a prominent amyloid core and few neurofibrillary tangles (NFTs) have also been described (Vinters et al. 1988). The finding of SPs and NFTs may point towards a close relationship of this condition with .Alzheimer's disease (AD). Results obtained with a
Correspondence to: F.F. Cruz-Sanchez, MD, Banco de Tejidos Neurolbgicos, Servicio de Neurologia, Hospital Clinico i Provincial, Villarroel 170, 08036 Barcelona, Spain. Fax: (3) 4515272; Phone: (3) 454 60 00.
number of antibodies against these structures support this contention (Vinters et al. 1988; Maruyama et al. 1990). Immunohistological studies using antibodies for ubiquitin, a 8.6 kDa polypeptide related to normal short-lived, and abnormal ATP-dependent protein degradation (Mori et al. 1987), have demonstrated similar antigenic determinants in senile plaques (SPs) of CAA and AD (Dickson et al. 1989, 1990). The present study describes immunohistochemical findings in the cerebral blood vessels in cases of CAA. Our results may give a new perspective on the pathogenesis of both CAA and AD. Material and methods
Formalin-fixed, paraffin-embedded brain tissue from 12 patients pathologically diagnosed as suffering from AD (4 cases), sporadic non-familial CAA (4 cases), 2 elderly patients without neurological disorder, and 2 patients with intracerebral non-CAA-associated hemorrhage were studied. The mean age of the CAA patients was 60 years (range 54-73) and that of AD cases was 71.5 years (range 58-73). The mean age of
47 control cases was 71 years (range 70 to 72). Female and male patients were equally represented. Patients with a pathological diagnosis of CAA were not hypertensive and had not sustained head injury. Two of these patients had shown mild to moderate symptoms of cognitive deterioration which was not sufficient for a diagnosis of AD to be made. They died because of intracerebral hemorrhage or hemorrhagic cerebral infarct. Pa-
tients with AD showed a progressive dementia with a mean evolution span of 8 years. Patients died of nonneurological complications. Brain tissue examined included samples from all regions of cerebral cortex, hippocampus, all of the basal ganglia nuclei and multiple cerebellum and brain stem samples. Brain tissue from hemorrhagic areas from cases of CAA and 2 control cases with intracere-
Fig. 1. Sections of CAA brain immunostained with and antibody ~ ubiquitin showing strong positivity in the media and to a lesser extent in the adventitia of arterlo,~-~ ~-,~,-~i~,~-~. ~, ~ •
48 bral non-CAA-associated hemorrhage was also examined. Criteria for the diagnosis of AD were not fulfilled in 2 cases of CAA with minor cognitive deterioriation (Khachaturian 1985; Wisniewski et al. 1989). These, together with the other 2 CAA cases without cognitive impairment were called "pure" CAA. Seven #~m thick paraffin sections were stained with hematoxylin and eosin (H&E), alkaline Congo red method (CR) and Bielschowsky's silver impregnation (Khachaturian 1985). CR sections were examined by conventional light microscopy and polarized light. The avidin biotin complex method was used for the polyclonal antibodies to ubiquitin (Dakopatts) and TAU (Sigma) and for the monoclonal antibodies to the 155 kDa (BF10) and 210 kDa (RT97) neurofilament subunits (Boehringer, Mannheim). Substantia nigra sections from two cases of idiopathic Parkinson's disease showing conspicuous Lewy bodies were used as positive control for ubiquitin.
Results
Macroscopically, brains from patients with CAA showed hemorrhagic infarction and intracerebral hemorrhage in atypical locations such as the cortical and subcortical areas in occipital, frontal a n d / o r parietal lobes. Histologically, these lesions consisted of necrotic areas at different stages of evolution. The most notable feature was the presence of obvious microangiopathy
-
of medium and small size subarachnoid and neocortical blood vessels resulting in thickening of the walls These blood vessels were also observed around ischemic/hemorrhagic areas. CR sections showed moderate staining of blood vessel walls and the classical apple-green birefringence when examined under polarized light. CR and Bielschowsky's stains also showed cortical and hippocampal SPs with an amorphous core. Macroscopically, AD brains showed ventricular enlargement and moderate to severe diffuse cortical atrophy especially posteriorly. Histologically, cerebral cortex showed severe neuronal loss, gliosis and large numbers of SPs of different types and NFTs, both in superficial and deep layers. These changes were found in frontal, temporal and occipital lobes as well as in the hippocampus. These findings are in agreement with the histological criteria for AD (Khachaturian 1985; Wisniewski et al. 1989). Some subarachnoid and cortical medium size blood vessels showed thickening of the wall and strong birefringence under polarized light with CR stains. Control brains showed mild to moderate diffuse cortical atrophy. Hi=tologically, there was moderate neuronal loss and very few SPs and NFTs in cerebral cortex on CR staining. CAA cases showed strong positivity for ubiquitin in leptomeningeal and cortical blood vessel walls; these vessels were Congo red-positive in serial sections. Several patterns of immunostaining were observed. In most cases, the positivity was located in the media of blood vessels (Fig. 1A and B) and in the subendothelial
W
o Q ~
b0 q lt
O O
i.
q
B
Ir
O
Fig. 2. Section of CAA brain immunostained with an antibody to ubiquitin. Note the granular positive material surrounding blood vessels (arrows) (ubiquitin, × 150).
49 space. In some cases, marked immunoreactivity was seen in foci in the perivascular space and in the neuropil around blood vessels (Fig. 2). These findings were seen in all CAA samples studied including those from the cerebral cortex far away from hemorrhagic or necrotic lesions. Positivity for neurofilaments and TAU was not observed. SPs in cases of CAA also showed strong positivity for ubiquitin. The pattern was granular and fibrillary and most plaques showed a negative central core which was composed of amyloid component. Very few NFTs were ubiquitin positive in cases of CAA. Both SPs and NFTs were also positive for TAU and both neurofilament subunits. In AD and control brains including two cases with intracerebral non-CAA-associated hemorrhage, cerebral blood vessels were consistently negative with all antibodies. In AD cases, SPs and NFTs showed immunoreactivity for ubiquitin, TAU and neurofilaments. Control brains showed very few SPs and NFTs which were positive with both neurofilament subunits, TAU and ubiquitin. Positivity for ubiquitin in SPs from CAA, AD and control cases was localized to the neurites, but not in the amyloid component.
Discussion
The clinicopathologic entity named CAA has been recognized as a distinct nosological entity on the basis of its pathology. It has been recognized that there are both sporadic and familial types of CAA with the Dutch type of A4beta amyloid associated with a gene mutation and a variant of CAA due to the deposition of cystatin C (Icelandic familial cerebral hemorrhage) as a result of gene mutation (Benson and Cohen 1977; Glenner 1980; Goren et al. 1980; Lofberg et al. 1987; Vinters 1987; Haan et al. 1990a,b). Microscopically, CAA may share some of the features of AD (Vinters 1987). Amyloid infiltrates the media and adventitia of the microvasculature, preferentially of leptomeningeal and superficial cortical blood vessels (Vinters 1987; Vinters et al. 1990). In the present study of sporadic non-familial CAA cases, amyloid deposits were found in the media and adventitia of leptomeningeal and superficial cortical vessels (Vinters and Gilbert 1983). These deposits showed strong positivity for ubiquitin, a finding which, to our knowledge, has not been reported. Dickson et al. (1989) found strongly perivascular ubiquitin positive structures surrounding cortical vessels. In agreement with these authors, we thought that these elements were of neuritic derivation. Antibodies to A4 amyloid protein have been used to study histological lesions in CAA and AD (Masters et al. 1985; Wong et al. 1985). Cases of "pure" CAA were
distinguishable from cases with AD in that the former did not show the number of plaques necessary to make a diagnosis of AD on Bielschowsky's stain. We cannot exclude that the 2 cases with CAA and mild cognitive impairment could have been early cases of AD but we followed established clinical and pathological criteria (Khachaturian 1985; Wisniewski et al. 1989) which had to be met before a case was assigned to the AD group. Reactivity with antibodies to neurofilament subunits, TAU and ubiquitin has been found in SPs and NFTs of AD and other neurodegenerative disorders of the CNS (Rasool et al. 1984; Delacourte and Defossez 1986; Papolla et al. 1989; Suenaga et al. 1990). These findings suggest that these disorders may have similar physiopathogenic mechanisms (Shankar et al. 1989). However, NFTs from patients with progressive supranuclear palsy appear to be negative for ubiquitin thus suggesting that a different mechanism is involved in their formation (Lennox et al. 1988; Leigh et al. 1989). Amyloid deposits found in vessels from AD cases in the present study did not show reactivity for ubiquitin which suggests differing mechanisms of formation in CAA and AD. However, it could also be possible that our findings reflect the overall less prominent (and/or advanced) degree of blood vessel compromise (amyloid deposits) in the small number of AD patients studied. The ubiquitin-dependent metabolism in vessels of CAA cases may play a significant role in the pathogenesis of the disease. While blood vessel amyloid deposits in AD may be considered to be a secondary phenomenon, some authors (Alizadeh-Khiavi et al. 1991; Chronopoulos et al. 1991) have demonstrated ubiquitin bound to extracellular amyloid and proposed that in AD ubiquitin has enhancing activity With respect to the deposition of amyloid. We did not think that positivity for ubiquitin in CAA blood vessel wall could be attributed to a stress response secondary to vascular insult because such positivity was seen in vessels which were not directly related to hemorrhage or necrosis. Also, cases with intracerebral non-CAA-associated hemorrhage did not show ubiquitin positivity in blood vessels. These observations may suggest that in "pure" CAA cases there is a primary pathological process inducing ubiquitin production which may lead to enhanced amyloid deposition. We conclude that this mechanism is unlikely to operate in AD as blood vessels were ubiquitin-negative. The anti-ubiquitin polyclonal antibody (Dakopatts) used in the present study reacts in a broad zone corresponding to the conjugates of ubiquitin and intracellular proteins when tested on extracts of human, mouse and rat cells in immunoblotting. Further work is needed to establish whether the immunoreactive material demonstrated in CAA blood vessel walls in this study has identity with ubiquitin protein conjugate af-
50 ter extraction. Ultrastructural studies of the localization of the antigen in CAA blood vessel walls would also be of interest. In this study we were not able to co-localize A4 protein and ubiquitin, but this is of obvious importance. Acknowledgements gaci6n Sanitaria de Trustees of MCNN. Interdepartamental fellowship) and Ms. assistance.
This work was supported by Fondo de lnvestila Seguridad Social (FISss) 90/E0593 and the We are grateful to Ms. M. Mencia (on Comissi6 de Recerca i Innovacio Tecnologica, CIRIT, J.R. Buller (MCNN) for tile excellent technical
References Alizadeh-Khiavi, K., J. Normand, S. Chronopoulos and Z. All-Khan (1991) Alzheimer's disease brain-derived ubiquitin has amyloidenhacing factor activity: behavior of ubiquitin during accelerated amyloidogenesis, Acta Neuropathol. (Berl.), 81: 280-286. Benson, M.D. and A.S. Cohen (1977) Generalized amyloid in a family of Swedish origin. A study of 426 family members in seven generations of a new kinship, nephropathy and central nervous system involvement, Ann. Intern. Med., 86: 419-424. Chronopoulus, S., K. Alizadeh-Khiavi, J. Normand, and Z. All-Khan (1991) Binding of ubiquitin to experimentally induced murine AA amyloid, J. Pathol., 163: 199-203. Delacourte, A. and A. Defossez (1986) Alzheimer disease: tau proteins the promoting factors of microtubule assembly, are major components of paired helical filament, J. Neurol. Sci., 76: 173186. Dickson, D.W., M. C.,T.stal, L.A. Mattiace, Y. Kress, A. Schwagerl, H. Ksiezak-Reding, P. Davies and S.H. Yen (1989) Diffuse Lewy body disease: light and electron microscopic immunocytochemistry of senile plaques, Acta Neuropathol., 78: 572-584. Dickson, D.W., A. Wertkin, L.A. Mattiace, E. Fier, Y. Kress, P. Davies and S.H. Yen (1990) Ubiquitin immunoelectron microscopy of dystrophic neurites in cerebellar senile plaques of Alzheimer's disease, Acta Neuropathol., 79: 486-493. Gilbert, J.J. and H.V. Vinters (1983) Cerebral amyloid angiopathy: incidence and complications in the aging brain. !. Cerebral hemorrhage, Stroke, 14: 915-923. Glenner, G.G. (1980) Amyloid deposits and amyloidosis. The Beta fibrilloses. I1. N. Engl. J. Med., 24: 1333-1343. Glenner, G.G., J.H. Henry and S. Fujihara (1981) Congophilic angiopathy in the pathogenesis of Alzheimer's degeneration, Ann. PathoL, 1: 120-129. Goren, H., M.C. Steinber and G.H. Farboody (1980) Familial o c u Ioleptomeningeal amyloidosis, Brain, 103: 473-495. Haan, J., P.R. Algra and R.A.C. Roos (1990a) Hereditary cerebral hemorrhage with amyloidosls.Dutch type, Arch. Neurol., 47: 649-653. Haan, J., J.B.K. Lanser, I. Zijderveld, G.F. van der Does and R.A.C. Roos (1990b) Dementia in hereditary cerebral hemorrhage with amyloidosis-Dutch type, Arch. Neurol., 47: 965-967. KhachaturJan, Z.S. (1985) Diagnosis of Alzheimer's disease, Arch. Neurol., 42: 1097-1105. Leigh, P.N., A. Probst, G.E. Dale, D.P. Power, J.P. Brion, A. Dodson and B.H. Anderton (1989) New aspects of the pathology
of neurodegenerative disorders as revealed by ubiquitin antibodies, Acta Neuropathol., 79: 61-72. Lennox, G., 3. Lowe, K. Morrell, M. Landon and R.J. Mayer (1988) Ubiquitin is a component of neurofibrillary tangles in a variety of neurodegenerative diseases, Neurosci. Lett., 94: 211-217. Lot'berg, H., A.O. Grubb, E.K. Nilsson, O. Jensson, G. Gadmundsson, H. Bl6ndal, A. Arnason and L. Thorsteinsson (1987) Immunocytochemical characterization of the amyloid deposits and quantitation of pertinent cerebrospinal fluid proteins in hereditary cerebral henm~'rhage with amyloidosis, Stroke, 18: 431-440. Maruyama, K., S. Ikeda, T. Ishihara, D. Allsop and N. Yanagisawa (1990) Immunohistochemical characterization of cerebrovascular amyloid in 46 autopsied cases using antibodies to beta-protein and cystatin C, Stroke, 21: 397-403. Masters, C.L., G. Simms, N.A. Weinman, G. Mulraup, B.L. McDonald and K. Beyreuther (1985) Amyloid plaque core protein in Alzheimer's disease and Down syndrome, Proc. Natl. Acad. Sci. USA, 82: 4245-4249. Mori, H., J. Kondo and Y. lhara (1987) Ubiquitin is a component of paired helical filaments in Alzheimer's disease, Science, 235: 1641-1644. Pappolla, M.A., R. Omar and B. Saran (1989) The normal brain. Abnormal ubiquitinated deposits highlight an age-related protein change, Am. J. Pathol., 135: 585-591. Rasool, C.G., C. Abraham, B.H. Andeton, M. Haugh, J. Khan and D.J. Salkoe (1984) Alzheimer's disease: immunoreactivity of neurofibrillary tangles with antineurofilament and anti-paired helical filament antibodies, Brain Res., 310: 249-260. Shankar, S.K., R. Yanagihara, R.M. Garruto, I. Grundke-Iqbal, K.S. Kosik and D.C. Gajdusek (1989) lmmunocytochemical characterization of neurofibrillary tangles in amyotrophic lateral sclerosis and Parkinson-dementia of Guam, Ann. Neurol., 25: 146-151. Suenaga, T., A. Hirano, J.F. Llena, H. Ksiezak-Reding, S.H. Yen and D.W. Dickson (1990) Modified Bielschowsky and immunocytochemical studies on cerebellar plaques in Alzheimer's disease, J. Neuropathoi. Exp. Neurol., 49: 31-40. Vinters, H.V. (1987) Cerebral amyloid angiopathy. A critical review, Stroke, 18: 311-324. Vinters, H.V. and J.J. Gilbert (1983) Cerebral amyloid angiopathy; incidence and complications in the aging brain. II. The distribution of amyloid vascular changes, Stroke, 14: 924-928. Vinters, H.V., W.M. Pardridge and J. Yang (1988) Immunohisto. chemical study of cerebral amyloid angiopathy: use of an antiserum to a synthetic 28 amino-acid peptide fragment of the Alzheimer's disease amyloid precursor, tlum. Pathol., 19: 214222. Vinters, H.V., D.L. Secor, W.M. Pardridge and F. Gray (1990) lmmunohistochemical study of cerebral amyloid angiopathy. III. Widespread Alzheimer A4 peptide in cerebral microvessei walls colocalizes with gamma trace in patients with leukoencephalopathy, Ann. Neurol., 28: 34-42. Vonsattel, J.P.G., R.H. Meyers, E.T. Hedley-Whyte, A.H. Ropper, E.D. Bird and E.P. Richardson (1991) Cerebral amyloid angiopathy without and with cerebral hemorrhages: a comparative histological study. Ann. Neurol., 30: 637-649. Wisniewski, H.M., A. Rake, W. Zigman and W. Silverman (1989) Neuropathological diagnosis of Alzheimer's disease, J. Neuropathol., 48: 606-609. Wong, C.W., V. Quaranta and G.G. Glenner (1985) Neuritic plaques and cerebrovascular amyloid in AIzheimer's disease are antigenically related, Proc. Natl. Acad. Sci. USA, 82: 8279-8732.
