Hereditary cystatin C (γ-trace) amyloid angiopathy of the CNS causing cerebral hemorrhage

Hereditary CNS amyloid angiopathy occurring in Icelanders is the first human disorder known to be caused by deposition of cystatin C amyloid fibrils in the walls of the brain arteries leading to single or or multiple strokes with fatal outcome. One or more affected members have been verified by histological examination in 8 families containing 127 affected. These originated from the same geographic area. Abnormally low value of cystatin C found in the cerebrospinal fluid of those affected can be used to support or make diagnosis of this disease, also in asymptomatic relatives. By amino acid sequence analysis the amyloid fibrils in the patients are found to be a variant of cystatin C (gamma-trace), a major cysteine proteinase inhibitor. The variant protein has an amino acid substitution (glutamine for leucine) at position 58 in the amyloid molecule. It is postulated that a point mutation has occurred leading to production of amyloidogenic protein causing the disorder.     

Dementia of the Alzheimer type and related dementias in the aged: DAT subgroups and senile dementia of the neurofibrillary tangle type

The aims of this study are to elucidate variations in the neuropathology of dementia of the Alzheimer's type (DAT) and related dementias in the elderly, and to delineate a senile dementia characterized by abundant neurofibrillary tangles (NFT) in the hippocampal region and by a scarcity of senile plaques (SP) throghout the brain (senile dementia of the NFT type; SD-NFT) in comparison with usual DAT cases. One hundred and five autopsied patients who developed dementia at age 70–100 years were investigated. The autopsy series included 57 cases of DAT, five of diffuse Lewy body disease (DLBD) with Alzheimer-tyupe neuropathological changes, and five of SD-NFT. The 57 DAT patients were classified into three subgroups depending onthe severity of the neocotical neuronal degeneration (NFT and neuronal loss): the diffuse, severe type (DAT-DS, extensive and severe involvement of the neocortex; n= 18); the diffuse, mild type (DAT-DM, mild involvement of the neocortex; n= 29); and the localized type (DAT-L, neuronal degeneration almost localized to the hippocampal region; DAT without neocortical NFT; n= 10). The frequencies of the DAT subgroups, DLBD, and SD-NFT in each age group at the onset were DAT-DS (41%) > DAT-DM (38%) > DLBD (16%) > DAT-L (6%) of patients in their 70s, DAT-DM (52%) > DAT-L (24%) > DAT-DS (16%) > SD-NFT (8%) of those in their 80s, and DAT-DS (40%) > SD-NFT (30%) > DAT-L (20%) > DAT-DS (10%) of those in their 90s and older. In a morphometric comparison with age-mathced DAT cases, the SD-NFT showed a significantly higher density of hippocampal NFT and a significant scarcity of SP and cerebral amyloid angiopathy in the brain. Analysis of genotypes of apolipoprotein E gene revealed that no patient with the SD-NFT had ε4 allele which was shown to be frequently associated with DAT. Our results have indicated (i) that the later onset of DAT and related dementias is linked with the milder neuronal degeneration in the neocortex, and (ii) that the SD-NFT is a common neuropathological condition that causes dementia in the very aged, in which the pathogenetic process may be different from that in DAT.     

Cerebral amyloid angiopathy in the aged

Summary Cerebral amyloid angiopathy (CAA) was found in 57% of 123 autopsy brains removed from patients aged 59–101 years. The incidence of CAA increased with age. CAA was seen most frequently in the occipital cortex. Immunohistochemically, amyloid of CAA was positive for amyloid P component and negative for human AA protein and human prealbumin. The presence and severity of CAA were significantly correlated with the number of senile plaques and neurofibrillary tangles. The incidence of CAA in 17 patients with dementia of Alzheimer type (DAT) was estimated to be 88% and was significantly higher than that in 26 patients with dementia of non-Alzheimer type. CAA had a pathogenetic relationship with both brain ageing and DAT. Lobar cerebral haemorrhage was found in 3 patients with CAA of marked or moderate degree. Lobar cerebral haemorrhage in the aged and in patients with DAT suggest the presence of CAA.     

Co-localization of β/A4 and cystatin C in cortical blood vessels in Dutch, but not in Icelandic hereditary cerebral hemorrhage with amyloidosis

Based on the recent discovery of co-localization of β/A4 and cystatin C in cortical blood vessels of patients with cerebral hemorrhages due to sporadic amyloid angiopathy and patients with Alzheimer's disease we investigated the presence of these two proteins in the cortical blood vessels of patients suffering from hereditary cerebral hemorrhage with amyioidosis of the Dutch (n = 11) and the Icelandic (n = 2) type. The brains of three patients with sporadic cerebral amyloid angiopathy were also investigated. Blood vessels of the Dutch patients clearly showed immunostaining with β/A4 as well as with cystatin C antibodies, whereas the blood vessels of Icelandic patients showed only staining with cystatin C. In one of the three sporadic amyloid angiopathy patients co-localization was shown as well. The co-localization of mutated β/A4 with normal cystatin C in the Dutch patients suggests that cystatin C deposition occurs secondarily to β/A4 deposition. This is probably also the case in sporadic amyloid angiopathy and Alzheimer's disease. Cystatin C deposition may play a role in the development of cerebral hemorrhages and leukoencephalopathy.     

Cerebral amyloid angiopathy in a 95+ cohort: complement activation and apolipoprotein E (ApoE) genotype

There is growing evidence that in Alzheimer's disease (AD) amyloid beta-protein (Abeta) triggers a chronic inflammatory reaction in cerebral amyloid plaques, including complement proteins. Abeta also accumulates cerebrovascularly in age- and AD-associated cerebral amyloid angiopathy (CAA). We investigated complement proteins in CAA in a population-based series using histological and immunohistochemical staining methods. The 74 subjects, aged 95 years or more, had undergone clinical neurological examination and apolipoprotein E (ApoE) genotyping. The brains had been studied for AD post-mortem, allowing us to relate the histopathological findings to clinical and genetic conditions. CAA with congophilic amyloid was found in 36/74 individuals (48.6%). The vascular amyloid deposits immunoreacted with antibodies to Abeta and complements 3d (C3d) and 9 (C9). The positivity in complement stains increased with growing severity of CAA (P = 0.001). The presence of CAA associated with ApoE epsilon4 (P = 0.0005) and overrepresentation of epsilon4 among those with moderate or severe vs. mild CAA (P = 0.03) was demonstrated. The presence of CAA associated with dementia (P = 0.01), which was contributed by both epsilon4+ (P = 0.02) and epsilon4- (P = 0.06) subjects. Our study shows that complement proteins are deposited in the affected vessels in Abeta-associated CAA. They may solely represent the cerebral Abeta- burden associated to inflammatory stimuli, or signal a contribution in the clearance of cerebral Abeta, thereby contributing to the events associated with evolution of clinical dementia. Our results demonstrate a strong association between CAA and ApoE epsilon4 as well as dementia and suggest that the contribution of CAA to dementia is largely independent of ApoE epsilon4.     

Cerebral amyloid angiopathy: An overview

Cerebral amyloid angiopathy (CAA) is characterized by amyloid deposition in cortical and leptomeningeal vessels. Several cerebrovascular amyloid proteins (amyloid β‐protein (Aβ), cystatin C (ACys), prion protein (AScr), transthyretin (ATTR), gelsolin (AGel), and ABri (or A‐WD)) have been identified, leading to the classification of several types of CAA. Sporadic CAA of Aβ type is commonly found in elderly individuals and patients with Alzheimer’s disease. Cerebral amyloid angiopathy is an important cause of cerebrovascular disorders including lobar cerebral hemorrhage, leukoencephalopathy, and small cortical hemorrhage and infarction. We review the clinicopathological and molecular aspects of CAA and discuss the pathogenesis of CAA with future perspectives.     

Relationships in Alzheimer's disease between the extent of Abeta deposition in cerebral blood vessel walls, as cerebral amyloid angiopathy, and the amount of cerebrovascular smooth muscle cells and collagen

The relationship between degree of cerebral amyloid angiopathy (CAA) and the amount of smooth muscle cells (SMCs) and deposition of collagen IV fibres (COL IV) was investigated in the frontal and occipital cortex of 70 patients with autopsy confirmed Alzheimer's disease (AD). The extent of CAA was significantly greater in occipital than in frontal cortex, although SMC loss was greater in frontal than in occipital cortex. COL IV staining was significantly higher in occipital than in frontal cortex. The degree of SMC loss correlated with CAA, as Abeta40 but not as Abeta42 or total Abeta, in frontal cortex, but not in occipital cortex. Leptomeningeal arteries within occipital cortex showed significantly greater external diameter, greater wall thickness and greater luminal area than those in frontal cortex. The degree of CAA correlated with thickness of blood vessel wall and external diameter in frontal cortex, whereas extent of SMC loss correlated with thickness of blood vessel wall in occipital cortex. There were significant negative correlations between duration of disease and thickness of vessel wall, external diameter and luminal area. In patients with disease durations exceeding 10 years, external vessel diameter and thickness of the vessel wall were both halved compared with patients with durations less than 5 years; luminal area was reduced by about 75%. Blood vessels in AD undergo degenerative changes involving deposition of Abeta and COL IV with loss of SMC. SMC loss may relate to increasing Abeta deposition in early stages of disease, but this relationship may be lost with disease progression.     

Observations of amyloid angiopathy and senile plaques by the scanning electron microscope

Summary Blood vessels with amyloid angiopathy and senile plaques in the cortices of the brains with Alzheimer's disease and senile dementia were observed by means of a scanning electron microscope.The results obtained were as follows: The blood vessels with amyloid angiopathy were surrounded by solid substances. The senile plaques consisted of rough solid substances, contained degenerated cell processes, and almost all plaques existed around the degenerated capillaries with amyloid angiopathy.From the above described findings, we suggest that the senile plaque has an extremely close relationship to the capillary which had undergone amyloid angiopathy.     

Unexpectedly low prevalence of intracerebral hemorrhages in sporadic cerebral amyloid angiopathy

In a retrospective study of a consecutive autopsy series of 2060 elderly subjects (mean age 78.5 ± 6.8 SD years), sporadic cerebral amyloid angiopathy (CAA) of various degrees was detected in 73.2% and in 98.5% of autopsy-confirmed cases of typical (plaque and tangle) Alzheimer disease (AD). Spontaneous (non-traumatic) intracerebral hemorrhages (ICH) (excluding microbleeds) were seen in 5.6% of the total cohort and in 7.2% of definite AD cases; CAA was found in 49% of brains without and in 48.7% with ICH which was not significantly different. The latter groups showed a significantly higher frequency of severe degrees of CAA than those without ICH (80.4 vs 30.9%, p < 0.001). Patients with CAA were older than those without CAA, showing a higher frequency of clinical dementia and pathologically confirmed AD, but signs of hypertension (history and/or autopsy) were seen in 41 and 33.6% of these cases, respectively, compared to 70–75% in patients with non-CAA related ICHs. CAA-related ICH much more frequently involved cerebral lobes or hemispheres, while non-CAA related lesions were more often located in basal ganglia and brainstem. The data of a lower prevalence of CAA in cases without than with ICH, but a similar prevalence of ICH with and without CAA do not support the concept that CAA represents the most evident risk factor for ICH in the aged. While severe degrees of CAA were indeed associated with ICH, the general prevalence of large ICH in this autopsy cohort was much higher in cases without CAA, probably due to other risk factors including hypertension, which was documented in around 40% of cases with CAA-related ICH. APOE ε3/4 and ε4/4 were significantly more frequent in AD (n = 163) than in age-matched controls (n = 47) and were associated with more severe degrees of CAA, but no general genotyping in ICHs with and without CAA was performed. Hence, the role of APOE in the pathogenesis of ICH with and without CAA needs further elucidation.     

Caveolin-1 and -2 and their relationship to cerebral amyloid angiopathy in Alzheimer's disease

Cerebral amyloid angiopathy (CAA) affects over 90% of patients with Alzheimer's disease (AD) and increases the risk of cerebral haemorrhage and infarction. Caveolae--cholesterol-enriched plasmalemmal microinvaginations--are implicated in the production of amyloid beta peptide (Abeta). Caveolin-1 (CAV-1) is essential for the formation of caveolae. Caveolin-2 (CAV-2) is expressed at the plasma membrane only when in a stable hetero-oligomeric complex with CAV-1. CAV-1 and CAV-2 are highly co-expressed by endothelium and smooth muscle. Recent studies suggest that down-regulation of CAV-1 causes a reduction in alpha-secretase activity and consequent accumulation of Abeta. We have used quantitative immunohistochemical techniques to assess the relationship between CAV-1 and CAV-2 with respect to Abeta accumulation in the cerebral vasculature in a series of post mortem brains. CAV-1 and CAV-2 were co-expressed within the tunica media and endothelium of cerebral blood vessels. There were regional differences in CAV-1 immunolabelling, which was significantly greater in the frontal cortex and white matter than in the parietal lobe (in both control and AD cases) or the temporal lobe (in AD alone). However, CAV-1 labelling in AD did not differ from that in controls in any of the three lobes examined. Assessment of CAV-1 labelling in relation to the severity of CAA showed CAV-1 to be significantly increased in the frontal white matter in a subgroup of AD cases with absent/mild CAA compared with controls with absent/mild CAA and to AD cases with moderate/severe CAA, but the latter groups did not show significant differences from one another. CAV-1 labelling did not vary with age, gender, APOE genotype, post mortem delay or brain weight. Only segments of blood vessels with particularly abundant Abeta and extensive loss of smooth muscle actin showed loss of CAV-1 and CAV-2 from the tunica media. Within these vessels endothelial CAV-1 was preserved and discontinuous CAV-2 labelling was noted along the outer aspect of the vessel wall. Our findings suggest that alterations in the expression of vascular CAV-1 and CAV-2 are unlikely to play a role in the development of CAA in AD.     

Relationship of neurofibrillary pathology to cerebral amyloid angiopathy in Alzheimer's disease

Over 90% of patients with Alzheimer's disease (AD) develop cerebral amyloid angiopathy (CAA). Severe dyshoric CAA, in which amyloid extends into the surrounding brain parenchyma, may be associated with adjacent clustering of tau-immunopositive neurites but the relationship of CAA to neurofibrillary pathology has not been systematically investigated. In the present study this relationship was examined in sections of frontal, temporal and parietal cortex from 25 AD patients with moderate to severe CAA and 26 with mild or absent CAA. We measured immunolabelling of abnormally phosphorylated tau adjacent to A beta-laden and non-A beta-laden arteries and arterioles, and in cortex away from arteries and arterioles. We also analysed the possible influence of APOE genotype on these measurements. There were no significant differences between the lobes in measurements of tau labelling, either around blood vessels or elsewhere in the cortex. However, tau labelling around A beta-laden arteries and arterioles significantly exceeded that around non-A beta-laden blood vessels (P<0.001) and this, in turn was greater than the labelling of cortex away from blood vessels (P<0.001). There was no association between APOE epsilon 4 and the immunolabelling density for tau, whether around amyloid- or non-amyloid-laden arteries and arterioles, or in the cerebral cortex away from these. We propose that both CAA and peri-vascular accumulation of hyperphosphorylated tau may be a consequence of elevated levels of soluble A beta around cortical arteries and arterioles.     

Cytoskeletal pathology in familial cerebral amyloid angiopathy (British type) with non-neuritic amyloid plaque formation

The histological features of familial cerebral amyloid angiopathy (British type) with non-neuritic amyloid plaque formation (FAB) include deposition of amyloid, (supposedly associated with the C-terminal fragments of both α- and β-tubulin), in small cerebral and spinal arteries, hippocampal amyloid plaques and neurofibrillary tangles (NFTs) as well as ischaemic white matter changes. In the present study we report on the cytoskeletal pathology that occurs in association with FAB. Sections from the hippocampus and cerebellum of three cases from three unrelated families were stained with silver impregnation methods and antibodies to antigens including tau, neurofilaments, ubiquitin and glial fibrillary acidic protein. Electron microscopic examination of the hippocampus was carried out in one case. All hippocampal subregions contained large numbers of NFTs and neuropil threads (NT), which were stained with both phosphorylation-dependent and phosphorylation-independent tau antibodies and ultrastructurally were found to be composed of paired helical filaments (PHFs). Although the majority of the amyloid plaques were of the non-neuritic type, distended PHF-containing and tau-positive neurites were seen in close proximity of a minority of the hippocampal plaques. The perivascular amyloid deposits of the cerebellum contained numerous ubiquitin-positive granular elements similar to those seen in cerebellar Aβ amyloid plaques in Alzheimer’s disease. In FAB severe cytoskeletal pathology is present in areas most affected by amyloid plaque deposits, thus suggesting a localised neurotoxic effect of the poorly characterised amyloidogenic peptide characteristic of this condition. Received: 2 June 1998 / Revised, accepted: 11 August 1998     

Immunohistochemical study of constituents other than β-protein in canine senile plaques and cerebral amyloid angiopathy

To clarify the significance of the constituents of canine senile plaques (SPs) or cerebrovascular amyloid deposits, paraffin and cryostat sections of canine brains were examined by immunohistochemistry using antibodies against cathepsin B (CB), cathepsin D (CD), cystatin C (CC), α-1-antichymotrypsin (ACT), heat shock protein 70 (HSP70), ubiquitin (Ubq), and apolipoprotein E (Apo E). On the cryostat sections, all types of canine SPs and cerebrovascular amyloid deposits in both arterioles and capillaries were positive for Apo E. On paraffin sections, the Apo E immunoreactivity of diffuse plaques was weak and varied according to the method of fixation or pretreatment before immunostaining. Moreover, amyloid plaques were found to contain several elements that were positive for CC, ACT, CD, and Ubq, and a subset of vascular amyloid deposits around cortical capillaries showed significant immunoreactivity for CD, CC, and ACT. In addition, vascular amyloid deposits in the arterioles showed moderate CD immunoreactivity and were intensely Apo E positive. No significant labeling of canine SPs or vascular amyloid deposits was detected when the antibodies against CB and HSP 70 were applied to the cryostat and paraffin sections. These results indicated that, of the constituents examined, Apo E might be most closely related to canine β-amyloidosis in the early stage of this brain disorder. Received: 16 June 1996 / Revised: 28 May 1996, 24 June 1996, 8 August 1996 / Accepted: 8 August 1996     

Sporadic cerebral amyloid angiopathy is not a frequent cause of spontaneous brain hemorrhage

The retrospective study of a consecutive autopsy series of 1100 elderly subjects (mean age 78.3 ± 6.8 SD years), revealed sporadic cerebral amyloid angiopathy (CAA) in 50.0% and in 95.7% of autopsy-confirmed cases of Alzheimer disease (AD). Apolipoprotein (APOE) ɛ 3/4 and ɛ 4/4 were significantly more frequent in AD than in controls, and were associated with more severe degrees of CAA. Spontaneous (non-traumatic) intracerebral hemorrhages (ICH) (excluding microbleeds and hemorrhagic infarctions) were seen in 5.4% and only in 3.3% of AD cases. CAA was found in 50.6% of brains without and in 42.4% with ICH, the latter showing a significantly higher frequency of severe degrees of CAA. ICH was related to CAA in 42.4%, whilst no such relation was seen in 57.6%. Patients with CAA were older, showed a higher frequency of clinical dementia and pathologically confirmed AD, but signs of hypertension (history and/or autopsy) occurred in 40%, compared with 80% in those with non-CAA-related ICHs. CAA-related ICH more frequently involved in cerebral lobes or hemispheres, whilst non-CAA-related ones were more often located in the basal ganglia and brainstem. The data of a lower prevalence of CAA in cases with than without ICH and of ICH with and without CAA do not support the concept that CAA represents the most important risk factor for ICH in the aged, probably because of other risk factors including hypertension.     

The impact of cerebral amyloid angiopathy on the occurrence of cerebrovascular lesions in demented patients with Alzheimer features: a neuropathological study

Objective: The aim of this neuropathological study was to determine the prevalence of the different cerebrovascular lesions to be attributed to cerebral amyloid angiopathy (CAA) and of those associated with the severity of the Alzheimer dementia (AD) itself. Patients and methods: The cerebrovascular lesions were compared separately in 40 brains of patients with mild and 50 with severe AD features. In the two groups, the number of lesions were compared between the brains with severe and those with mild of absent CAA. Results: The age of the patients, the vascular risk factors and antithrombotic treatment were similar in all the compared groups. The brains with mild and severe AD features and with CAA contained more haematomas, cortical micro‐infarcts and micro‐bleeds, and more severe white matter changes, and cortico‐subcortical and white matter mini‐bleeds. In the CAA brains with severe AD features, also more cortical territorial infarcts were observed, compared to those with mild AD features. Conclusions: The increase in cortical infarcts cannot be attributed to the CAA alone, but also to the severity of the degenerative features, implying additional vascular factors in the pathogenesis of AD.     

Dementia in cerebral amyloid angiopathy: a clinicopathological study

Summary Dementia is in addition to cerebral haemorrhage major sympton of cerebral amyloid angiopathy (CAa). In order to explore the pathological basis for dementia in CAa-related conditions, we made a clinicopathological analysis of CAa, with special attention to dementia. Among 150 patients (mean age 78.6 years) with autopsy-proven intracranial haemorrhage in Tokyo Metropolitan Geriatric Medical Center, CAa with cerebral haemorrhage accounted for 8.0% (12 cases), associated with hypertension and metastatic brain tumour. Among 38 patients with lobar haemorrhage, CAa represented the second most common cause (21.1%) of intracranial haemorrhage after hypertension. A total of 20 patients with CAa (mean age 82.5 years) were studies clinically and pathologically. Hypertension was present in 50%. Thirteen had a history of stroke and others had either ill-defined or no strokes. The average number of strokes 2.9. Fifteen patients (75%) had dementia. Based on the clinicopathological grounds for dementia, CAa-related conditions could be divided into three subtypes: haemorrhagic, dementia-haemorrhagic and dementia type. Haemorrhagic type (30%, 6 cases) showed multiple recurrent lobar haemorrhages caused by CAa. Hypertension was present in only 1 patient. The incidence of senile plaques and neurofibrillary tangles was generally correlated with age. Only 1 patient had dementia. The dementia-haemorrhagic type (40%, 8 patients) had recurrent strokes with cerebral haemorrhage after preceding dementia. There were two different neuropathological subsets: CAa with atypical senile dementia of Alzheimer type (SDAT) and CAa with diffuse leucoencephalopathy. Patients with CAa with atypical SDAT had multiple cerebral haemorrhages caused by CAa combined with atypical Alzheimer-type pathology. Patients with CAa with diffuse leucoencephalopathy had cerebral haemorrhages in combination with diffuse white matter damage like Binswanger's subcortical vascular encephalopathy (BSVE). The incidence of senile changes correlated with age. Patients with the dementia type (30%, 6 patients) showed progressive dementia with or without haemorrhage. All had hypertension. They had a combined condition of Alzheimer-type pathology with conspicuous CAa with BSVE. Dementia in CAa-related conditions may be responsible for multiple factors including not Alzheimer-type degeneration, but also diffuse leucoencephalopathy like Binswanger's disease. We also found an asymptomatic type, an ischaemic type, a vasculitis type and an hereditary type in this condition.     

Cerebral amyloid angiopathy in Lewy body disease

Summary. While Alzheimer and Lewy body pathologies are discussed as major substrates of dementia in Parkinson’s disease (PD/Lewy body disease of brainstem type), the incidence and impact of cerebral amyloid angiopathy (CAA) and its association with cognitive decline in PD and dementia with Lewy bodies (DLB) are unknown. The severity of CAA and other Alzheimer lesions were assessed in 68 cases of autopsy-confirmed PD, 32 of them with dementia (PDD), and in 20 cases of DLB. PDD patients were significantly older than those without dementia (mean age 84.5 vs 77.6 years; p < 0.01), the age of DLB patients was in between both groups (mean 80.0 years), while duration of disease was DLB < PDD < PD (mean 6.5 vs 8.5 and 14.3 years). PDD patients had a significantly higher neuritic Braak stage (mean 4.2 vs 2.4, p < 0.01), significantly higher cortical amyloid β (Aβ) load, capillary cerebral amyloid angiopathy (CapCAA) and generalized CAA than those without dementia (mild CapCAA in 22% vs moderate to severe CapCAA in 87%; mild generalized CAA in 5.5% vs moderate to severe generalized CAA in 82%). Mean PD stage was higher in both DLB and PDD than in PD (mean 5.2 vs 4.5 and 4.0, respectively): Mean neuritic Braak stage in DLB was 3.4, severe Aβ plaque load was seen in 95%, moderate to severe CapCAA in 90% and mild to severe generalized CAA in 70%. This and other recent studies imply an association of CAA with cognitive decline in both PD/PDD and DLB, particularly in cases with concomitant AD-type pathology. Correspondence: Kurt A. Jellinger, Institute of Clinical Neurobiology, Kenyongasse 18, 1070 Vienna, Austria