Cerebral amyloid angiopathy

Cerebral amyloid angiopathy (CAA) is a feature of ageing and Alzheimer's disease (AD); it is also associated with intracerebral hemorrhage and stroke. Here, the pathogenesis of CAA and its effects on the brain are reviewed and the possible effects of CAA on therapies for Alzheimer's disease are evaluated. Tracer experiments in animals and observations on human brains suggest that peptides such as A[beta] are eliminated along the peri-arterial interstitial fluid drainage pathways that are effectively the lymphatics of the brain. In CAA, A[beta] becomes entrapped in drainage pathways in the walls of cerebral arteries, reflecting a failure of elimination of A[beta] from the ageing brain. One consequence of failure in clearance of A[beta] is accumulation of soluble and insoluble A[beta] associated with cognitive decline in AD. Replacement of vascular smooth muscle cells by A[beta] occurs in severe CAA with weakening of artery walls and increased risk of vessel rupture and intracerebral hemorrhage. Risk factors for CAA include mutations of the amyloid precursor protein (APP) gene and possession of the [epsilon]4 allele ofapolipoprotein E. There is also evidence that cerebrovascular disease may be a factor in the failure of elimination of A[beta] along perivascular pathways in sporadic AD; this would link ageing in cerebral arteries with the pathogenesis of Alzheimer's disease. If therapeutic agents, including anti-A[beta] antibodies, are to be used to eliminate A[beta] in the treatment of Alzheimer's disease, the effects of CAA on the treatment and the effects of the treatment on the CAA need to be considered. [Neurol Res 2003; 25: 611-616]

Keywords: Cerebral amyloid angiopathy; Alzheimer's disease; cerebral hemorrhage; stroke; interstitial fluid; cerebrospinal fluid; perivascular spaces

INTRODUCTION

Cerebral amyloid angiopathy (CAA) is a major pathological feature of Alzheimer's disease (AD) and a significant cause of intracerebral hemorrhage in the elderly. Although CAA was described in detail in the 1930s1, it has been recognized only relatively recently as a significant factor in the pathogenesis of AD. An understanding of the pathogenesis of CAA is essential, firstly so that it can be prevented and secondly so that the effects of CAA on therapies designed to clear amyloid from the AD and ageing brain can be thoroughly evaluated.

This review examines the evidence that amyloid [beta] (A[beta]) produced by the brain is normally cleared with interstitial fluid along perivascular drainage pathways and that in CAA, A[beta] becomes entrapped in the fluid drainage pathways in artery walls. There are two major effects of this entrapment: reduced drainage of A[beta] from brain tissue potentially leading to an eventual increase in the level of soluble A[beta] in the brain that is associated with cognitive decline in Alzheimer's disease2,3, and weakening of artery walls with increased risk of rupture and intracerebral hemorrhage.

ALZHEIMER'S DISEASE

Alzheimer's disease is the commonest dementia affecting the elderly population. Sporadic AD affects individuals over the age of 65 years and increases in incidence with age. Familial Alzheimer's disease is rare4 and normally has a lower age of onset than the sporadic form. Clinically, AD is characterized by loss of memory, visuospatial disorientation and disorders of language. Major risk factors for sporadic AD include increasing age and possession of the c4 allele of the apolipoprotein E gene (APOE)5. Mutations in the genes encoding the cell membrane associated proteins, amyloid precursor protein (APP)4 and presenilins 1 and 2(6), account for the majority of cases of familial AD.

The pathology of AD is characterized by the presence of intraneuronal neurofibrillary tangles (NFT) and by the deposition of A[beta] as plaques in cerebral gray matter and in blood vessel walls as CAA (Figure 7). These features suggest that major factors in the pathogenesis of AD are the failure of disposal of intracellular proteins (NFT) from neurons7 combined with failure of disposal of A[beta] from the extracellular compartments of the brain1,8.

Neurofibrillary tangles are composed of the microtubule-associated protein, tau, and other proteins bound to ubiquitin. Recent studies suggest that tau in AD and the protein synuclein in Parkinson's disease accumulate in neurons due to a failure of the intracellular ubiquitin-proteasome disposal system7. The reasons why A[beta] accumulates in the brain and vessel walls are complex but there may be a combination of overproduction of A[beta] especially in familial diseases4, or a failure of disposal of A[beta] from the brain9. A[beta] is deposited in the extracellular spaces of gray matter as diffuse plaques or as more compact, usually spherical, neuritic plaques (Figure 1) composed of a series of A[beta] peptides up to 42 or 43 amino acids in length and with a molecular weight of some 4 kD. A[beta] peptides are derived from APP that is encoded by a gene on chromosome 214. A number of secretases cleave APP to form the different A[beta] peptides4. The longer A[beta]42 peptide is less soluble than the shorter A[beta]40 and forms amyloid fibrils more readily10.

Various mutations in the APP gene on chromosome 21 have been identified in familial AD resulting either in an excessive production of A[beta] or in the production of abnormal forms of A[beta]4. However, there is little evidence for either over-production of A[beta] or abnormal forms of A[beta] in the 90% of cases of AD that are sporadic. This emphasizes that a disturbance of elimination of A[beta] from the ageing brain may be a factor in the pathogenesis of sporadic AD.

ELIMINATION OF A[beta] FROM THE BRAIN

A[beta] peptides in the brain parenchyma of young animals are degraded by the metalloproteinase, insulin-degrading enzyme (IDE) and by the catabolic peptidase, neprilysin11. However, neprilysin activity has been shown to decline significanlty with age in regions of the mouse brain, such as the hippocampus12. A[beta]1-40 is also cleared from the brains of young mice via the endothelium into the blood. This clearance is mediated by a low density lipoprotein (LDL) receptor-related protein-1 (LDLR-1) and by [alpha]2-macroglobulin; apolipoprotein E also appears to play a role in this clearance pathway (Figure 2)13. Elimination of A[beta] by this route is substantially reduced in older mice, and probably in older humans; A[beta] is then cleared with interstitial fluid along the perivascular route13. In transgenic APP mice, in which there is an overproduction of A[beta] by neurons, levels of A[beta] in the brain are low at 2 months of age, but by 12-14 months, A[beta] levels in the brain rise, blood levels fall and the mice develop CAA14. This suggests that the failure of elimination of A[beta] from brain tissue coincides with the development of CAA.

A[beta] deposited in the extracellular spaces as insoluble amyloid in neuritic plaques is surrounded by microglia and by astrocytes. In vitro experiments suggest that microglia actively ingest and degrade A[beta] particularly in the presence of anti-A[beta] antibodies15. This process is enhanced in mice and humans that have been treated with antibodies against A[beta]16,17. The deleterious effects of the microglial inflammatory reaction induced by A[beta] amyloid deposits have been extensively discussed in relation to neuronal damage in AD as a bystander effect of the microglial activation18. Astrocytes play an important role in the elimination of soluble proteins from the extracellular spaces of the brain particularly in cerebral edema, and a role for astrocytes in the clearance of soluble or insoluble A[beta] from the brain had just been elucidated19.

From the discussion above it seems that in older animals and humans, A[beta] is mainly eliminated from the brain via perivascular interstitial fluid drainage pathways. In order to understand the significance of this mode of elimination it is necessary to briefly review the route by which extracellular fluid drains from the brain both in rodents and in humans.

EXTRACELLULAR FLUID OF THE BRAIN

Extracellular fluid in the CNS is composed of two major components, namely, CSF and interstitial fluid (ISF)9. The functions of the two fluids are rather different and in humans their drainage pathways appear to be largely separate. Many of the data on the physiology of CSF and ISF are derived from studies in experimental animals, most of which have concentrated on the CSF9. However, experiments using tracers give a clear insight into the pathways by which ISF and A[beta] peptides are cleared from the human brain and a clear indication of the pathogenesis of CAA.

Cerebrospinal fluid is produced, both in experimental animals and in man, by the choroid plexuses within the ventricles of the brain (Figure 3). The choroid plexus is composed of a stroma derived from arachnoid mater, a rich vascular supply and an epithelium derived from ependyma. Fluid and proteins pass from the blood into the stroma of the choroid plexus and the tight intercellular junctions of the choroid plexus epithelium form the blood-CSF barrier that regulates the entry of protein and other substances in the CSF9.

Interstitial fluid in the brain is also derived from the blood and from cells within the brain. The blood-brain barrier is located at the capillary endothelium (Figure 3) and depends upon the presence of tight intercellular junctions between adjacent endothelial cells20. Entry of water is regulated by the entry of electrolytes that are themselves regulated in their passage into the brain by the presence of ATP-dependent ion channels. Substances such as glucose are actively transported through the glucose transporter system and little, if any, protein enters the brain through the normal blood-brain barrier. Lipid soluble substances enter the brain more readily than nonlipid soluble material. When brain tissue is damaged, there is breakdown of the blood-brain barrier, mediated by a number of factors including cytokines released at the time of brain damage20. This results in an increased flow of water, proteins and other soluble materials (such as contrast media used in CT and MRI) into brain tissue and the development of vasogenic edema. Inflammatory cells such as lymphocytes and monocytes enter the brain at sites of tissue injury through post-capillary venules. Receptors on lymphocytes match with receptors on endothelial cells and inflammatory cells enter the perivenular spaces by passing between the endothelial cells21. The inflammatory cells may then infiltrate the surrounding brain or may remain in the perivenular or perivenous spaces as inflammatory cell cuffs so typical of viral encephalitis and multiple sclerosis. There is no direct evidence that lymphocytes leave the brain.

ELIMINATION OF INTERSTITIAL FLUID AND CSF FROM THE BRAIN IN EXPERIMENTAL ANIMALS

Proteins and other substances derived from the blood and produced by cells of the CNS enter CSF and ISF, and are eliminated by bulk clearance mechanisms from both CSF and ISF22.23.

Experiments in the rat and rabbit using tracer molecules have shown that CSF and ISF drain to deep cervical lymph nodes in these species and that such lymphatic drainage of the brain is highly significant for immunological reactions within the brain9,21. It has been known for over 150 years that injection of tracers such as Indian ink into the CSF of the cisterna magna of the rat results in drainage of the tracer to deep cervical lymph nodes within approximately one minute. Tracers, such as isotope-labeled human serum albumin, injected into the ISF of the central gray matter of the rat brain also drain to cervical lymph nodes, but this takes a few hours22,23.

Routes of CSF drainage from the brain to cervical lymph nodes have been elucidated using colored dyes or Indian ink as tracers. Indian ink injected into the CSF in the cisterna magna passes forward through the basal cisterns to the inferior surface of the olfactory bulbs. Ink then enters lymphatic channels that pass through the cribiform plate and nasal submucosa to drain to deep cervical lymph nodes24. Although there are arachnoid villi associated with venous sinuses in the rat, they are very small and may play a secondary role to lymphatic drainage of CSF and ISF in these animals22-24.

ISF drainage pathways have also been defined by a series of tracer studies. ISF and soluble substances pass through the narrow convoluted interstitial spaces in gray matter and enter bulk flow pathways in the walls of brain capillaries25. Thence fluid and solutes drain out of the brain along the walls of arteries (Figure 3). At the surface of the brain, ISF passes along the walls of major arteries to the circle of Willis and then joins CSF to drain into lymphatics of the nasal mucosa en route to cervical lymph nodes9,24,26.

The drainage of ISF and CSF from the brain to cervical lymph has very significant implications for immunological reactions and inflammation in the rat brain involving both B lymphocytes and T lymphocytes. Cervical lymph nodes are the site of antibody formation to antigens derived from the brain22,23 and may be a major source of T lymphocytes that target the brain as shown in experimental autoimmune encephalomyelitis, a model for multiple sclerosis27,28.

CSF AND INTERSTITIAL FLUID DRAINAGE PATH WAYS IN THE HUMAN BRAIN

Pathways for drainage of CSF from the human brain have been extensively studied, but drainage of interstitial fluid has received less attention. As a consequence of evolution, the human brain is the largest in size, relative to body mass, of all mammals. One of the major functions of the cerebrospinal fluid for the human CNS appears to be as a buoyancy fluid whereby the brain and spinal cord are stabilized by the presence of intraventricular and subarachnoid CSF. It is estimated that CSF is secreted at 2.1 [mu]l min^sup -1^ in the rat but at a considerably greater rate of 350 [mu]l min^sup -1^, or more than 500 ml day^sup -1^, in man9. The human brain has evolved an efficient bulk-flow drainage system of CSF back into the blood via arachnoid villi and granulations invaginating the walls of the major venous sinuses29. Although arachnoid villi do pass through the human cribiform plate into the nasal mucosa, such villi are not very numerous nor do they have significant connections to the lymphatic vessels of the nasal mucosa. It seems unlikely, therefore, that the nasal lymphatics are of significance for the drainage of CSF in humans9.

Drainage of ISF from the human brain has been less extensively studied than elimination of CSF, despite the potential of ISF significance for immunological diseases such as multiple sclerosis, and for AD9. Most of the data on ISF drainage are derived first from ultrastructural studies showing homology of perivascular compartments in the human brain with ISF drainage pathways in the cat1,9,29,30, and secondly from the distribution of A[beta] in CAA. A[beta] effectively acts as a natural tracer that outlines perivascular ISF drainage pathways. Deposition of A[beta] in the walls of blood vessels in the human and transgenic APP mouse brain is in a pattern that corresponds closely to the ISF drainage pathways outlined by the injection of tracers in the rat1,8,31.

Electron microscope studies of human cerebral cortex show that the narrow extracellular spaces of the brain are in continuity with the perivascular compartment around capillaries. In the normal brain, the pericapillary space is filled with basement membrane and is continuous with peri-arterial compartments1. Sheaths of pia mater encase arteries in the cortex and separate peri-arterial spaces from the surrounding brain tissue (Figure 4)9,29,30. At the surface of the brain, the sheath of pia mater ensures continuity between peri-arterial spaces of cortical and leptomeningeal arteries and separates the ISF drainage pathways from the CSF in the subarachnoid space. Pia mater is reflected from the surface of the brain to coat the outside of leptomeningeal vessels, thus separating the CSF from perivascular spaces and from the underlying brain (Figure 4)30. As a consequence, blood in subarachnoid hemorrhage and subarachnoid pus in leptomeningitis are largely excluded from brain tissue32.

As will be seen from the description of CAA below, ISF appears to enter the perivascular drainage pathways only at the level of capillaries and thence drains out of the brain along the walls of cortical and leptomeningeal arteries1. The sheath of pia mater around arteries appears to block direct entry of ISF from the brain into the perivascular drainage pathways of cortical arteries. Perivascular spaces surrounding leptomeningeal arteries are much larger than those in the cortex and are filled with loosely packed collagen fibers1,30. Such spaces are readily expandable and are coated with a layer of piaarachnoid that may impede passage of perivascular ISF into the CSF of the subarachnoid space. Perivascular spaces around veins do not appear to be involved in the drainage of ISF as veins in the cortex possess no pial sheath to maintain continuity of perivascular spaces with veins in the subarachnoid space (Figure 4)29,30. A[beta] is infrequently associated with the walls of veins in the subarachnoid space, and when it is, this may be due to spread of A[beta] to the meninges and vein walls from heavy deposits of A[beta] in the walls of leptomeningeal arteries.

One of the major differences between the ISF drainage pathways in humans and rodents is the apparent separation of ISF from CSF in the human as opposed to the confluence of ISF and CSF as they drain to cervical lymph nodes in rodents. One unanswered question is whether lymphatic drainage of ISF from the brain occurs in humans. It is possible that ISF drains from the human brain to cervical lymph nodes along the perivascular spaces that extend through the base of the skull around the carotid and vertebral arteries9.

CEREBRAL AMYLOID ANGIOPATHY

Cerebral amyloid angiopathy (CAA) can be defined as the deposition of A[beta] and other amyloid peptides in the walls of leptomeningeal and cortical arteries and less frequently in the walls of cortical capillaries8,33,34. CAA is a feature of the ageing brain and is seen in the majority of patients with AD. Severe CAA is strongly related to dementia35. There are also a number of hereditary disorders in which CAA is a major pathological feature33,34. Some involve the deposition of A[beta], but in others, peptides such as cystatin or ABri (the British type of amyloid)36 are deposited. This supports a common pathogenesis for CAA whereby chemically different peptides are deposited in vessel walls by a similar mechanism.

When capillaries are involved by CAA, A[beta] is deposited in the basement membrane of the endothelial cells and may accumulate as small hemispherical or filiform pericapillary deposits extending into the surrounding brain tissue1. These are the 'Drusen' that were noted in the original descriptions of CAA and account for the name 'drusige Entartung' given to CAA by Scholz in 1938(37). In the early stages of CAA in cortical and leptomeningeal arteries A[beta] is deposited at the periphery of the artery walls and then spreads to involve the basement membranes between the smooth muscle cells of the media8. In advanced CAA, smooth muscle cells of the media are lost and are replaced totally by A[beta]5, but the endothelium is preferentially preserved in these vessels38. Immunocytochemistry shows that A[beta] is mainly located in the cortical artery and capillary walls and in the smaller, distal branches of the leptomeningeal arteries8. Biochemical studies, however, have shown that A[beta] is present in the walls of arteries as large as the basilar and middle cerebral arteries, not only in the elderly but also in individuals as young as 20 years39. A[beta] is greatly increased in amount in these vessels in the nondemented elderly and in AD. No A[beta] was detected in the walls of extracranial arteries; this strengthens the hypothesis that A[beta] drains only from the brain along walls of arteries, probably throughout life.

PATHOGENESIS OF CEREBRAL AMYLOID ANGIOPATHY

The original suggestion from Scholz37 in the 1930s was that amyloid in vessel walls was derived from the blood. It was then proposed that A[beta] in CAA was derived from smooth muscle cells in the media of cerebral arteries40. Although both these sources may contribute to the deposition of A[beta] in CAA, more recent evidence from transgenic mice and human studies now suggests that entrapment of A[beta] in ISF fluid drainage pathways is the main cause of CAA1,8,31. In transgenic mice that over-express mutated human APP, the over-production of A[beta] is by neurons and only in certain areas of the brain14. It is these areas that develop CAA suggesting that brain rather than blood is the source of the A[beta]31. Although smooth muscle cells in cerebral vessel walls produce A[beta]33, the distribution of CAA in smaller arteries and capillaries with least smooth muscle cells militates against smooth muscle cells as the major source of A[beta] in CAA1,8.

There are a number of predisposing or risk factors for CAA. First there are genetic diseases in humans and in transgenic mice in which there is over-production of A[beta] or other amyloid peptides associated with the development of CAA, often at an earlier age than in sporadic AD; in some cases, intracerebral hemorrhage is a major complication33,34. Polymorphisms of the APOE gene are also related to the development of CAA. Three common alleles [epsilon]2, [epsilon]3 and [epsilon]4, encode corresponding isoforms of the protein, designated E2, E3 and E4. APOE [epsilon]4 is the most important genetic risk factor for the development of sporadic AD and has a strong link with CAA both in AD and in nondemented elderly individuals5. As apoE co-localized with A[beta] in plaques in the brain and in cerebral blood vessels, apoE may act as a chaperone molecule for A[beta]5. There may be enhanced binding to A[beta] and deposition of A[beta] as amyloid fibrils in vessel walls in APOE [epsilon]4 carriers. The initial event in CAA may be seeding of the more insoluble A[beta]1-42, and progression of CAA may be due to further deposition of A[beta]1-40, the amount of which increases with APOE [epsilon]4 allele dose41. An alternative or additional possibility, that would support the hypothesis presented below, is that cerebrovascular disease, promoted by ageing and possession of APOE [epsilon]4, could impede the elimination of A[beta] along the perivascular ISF drainage pathways and promote deposition of A[beta] as CAA. Secondly, age is a major risk factor for both CAA and cerebrovascular disease. This suggests that functional changes may occur in ageing cerebral arteries that inhibit the elimination of A[beta] along perivascular ISF drainage pathways and result in CAA42. There is some experimental evidence for this hypothesis and some observations in the human brain that support this concept. Experimental ablation of the nucleus basalis, the source of cholinergic innervation of blood vessels in the brain, results in CAA in cortical arteries in the rabbit43, presumably by interfering with vessel tone. In the human brain, thrombotic occlusion of cortical arteries is associated with accumulation of A[beta] in the walls of capillaries arising from those arteries44. These observations suggest that reduced vessel pulsation in cerebrovascular disease may inhibit perivascular drainage of A[beta] from the human brain.

Complications of cerebral amyloid angiopathy

Intracerebral hemorrhage is the most extensively documented acute complication of CAA, particularly in patients over the age of 75 years5,33. Both the [epsilon]4 and [epsilon]eles of APOE appear to act as risk factors for hemorrhage in CAA5,45,46. The current suggestion is that whereas possession of APOE [epsilon]4 promotes deposition of A[beta] in the cerebral vasculature, possession of [epsilon]2 promotes the development of specific secondary vasculopathic features, including fibrinoid necrosis5 and vessel wall splitting47 which are preludes to hemorrhage in vessels already laden with amyloid. The development of hemorrhages in CAA is a particularly common feature in the Dutch ([beta]APP) and Icelandic (cystatin C) hereditary syndromes33. Intracerebral hemorrhage has also been identified as a complication in some strains of [beta]APP transgenic mice48. The effects CAA has upon the blood supply of the brain and on ISF drainage have also been emphasized particularly in relation to white matter disease49,50.

IMPLICATIONS OF CAA FOR THE THERAPY OF AD

Transgenic mice that over-express mutant human APP and accumulate A[beta] both in the brain parenchyma and in blood vessel walls have been treated with antibodies against A[beta] induced either by active or passive immunization. Plaques of A[beta] are rapidly cleared from the brain parenchyma in these animals, possibly through the action of microglia, but the CAA remains largely unaffected16. Clearance of A[beta] from the brain into the blood in APP transgenic mice is also increased by administration of anti-A[beta] antibodies51. Initial reports on one human patient dying from an unrelated cause while on similar antibody treatment also shows reduction in plaque A[beta] but only minor effects on A[beta] in CAA17. There is also one report in which the occurrence of intracerebral hemorrhage is increased in APP transgenic mice immunized against A[beta]52. These observations suggest that CAA should be seriously considered when devising therapeutic strategies for the elimination of A[beta] from the ageing human brain. The possible role of cerebrovascular disease in impeding the elimination of A[beta] from the ageing human brain42 suggests that therapeutic strategies aimed at preventing cerebrovascular disease may also be valuable in the prevention of AD.

REFERENCES

1 Preston SD, Steart PV, Wilkinson A, Nicoll JAR, Weller RO. Capillary and arterial amyloid angiopathy: Deliniation of the perivascular route for elimination of amyloid-beta from the human brain. Neuropath Appl Neurobiol 2003; 29: 106-117

2 Lue LF, Kuo YM, Roher AE, et al. Soluble amyloid beta peptide concentration as a predictor of synaptic change in Alzheimer's disease. Am J Pathol 1999; 155: 853-862

3 McLean CA, Cherny RA, Fraser FW, et al. Soluble pool of a beta amyloid as a determinant of severity of neurodegeneration in Alzheimer's disease. Ann Neurol 1999; 46: 860-866

4 Selkoe DJ. Alzheimer's disease: Genes, proteins, and therapy. Physiol Rev 2001; 81: 741-766

5 McCarron MO, Nicoll JA, Stewart J, et al. The apolipoprotein E epsilon2 allele and the pathological features in cerebral amyloid angiopathy-related hemorrhage. J Neuropathol Exp Neurol 1999; 58: 711-718

6 Mann DM, Pickering Brown SM, Takeuchi A, Iwatsubo T. Amyloid angiopathy and variability in amyloid beta deposition is determined by mutation position in presenilin-1-linked Alzheimer's disease. Am J Pathol 2001; 158: 2165-2175

7 Layfield R, Alban A, Mayer RJ, Lowe J. The ubiquitin protein catabolic disorders. Neuropathol Appl Neurobiol 2001; 27: 171-179

8 Weller RO, Massey A, Newman TA, Hutchings M, Kuo YM, Roher AE. Cerebral amyloid angiopathy: Amyloid beta accumulates in putative interstitial fluid drainage pathways in Alzheimer's disease. Am J Pathol 1998; 153: 725-733

9 Weller RO. Pathology of cerebrospinal fluid and interstitial fluid of the CNS: Significance for Alzheimer disease, prion disorders and multiple sclerosis. J Neuropathol Exp Neurol 1998; 57: 885-894

10 Roher AE, Lowenson JD, Clarke S, et al. beta-Amyloid-(1-42) is a major component of cerebrovascular amyloid deposits: Implications for the pathology of Alzheimer disease. Proc Natl Acad Sci USA 1993; 90: 10836-10840

11 Evin G, Weidemann A. Biogenesis and metabolism of Alzheimer's disease Abeta amyloid peptides. Peptides 2002; 23: 1285-1297

12 Iwata N, Takaki Y, Fukami S, Tsubuki S, Saido TC. Region-specific reduction of A beta-degrading endopeptidase, neprilysin, in mouse hippocampus upon aging. J Neurosci Res 2002; 70: 493-500

13 Shibata M, Yamada S, Kumar SR, et al. Clearance of Alzheimer's amyloid-ss(1-40) peptide from brain by LDL receptor-related protein-1 at the blood-brain barrier. J Clin Invest 2000; 106: 1489-1499

14 Kuo YM, Crawford F, Mullan M, et al. Elevated A beta and apolipoprotein E in A betaPP transgenic mice and its relationship to amyloid accumulation in Alzheimer's disease. Mol Med 2000; 6: 430-439

15 Bard F, Cannon C, Barbour R, et al. Peripherally administered antibodies against amyloid beta-peptide enter the central nervous system and reduce pathology in a mouse model of Alzheimer disease. Nat Med 2000; 6: 916-919

16 Janus C, Pearson J, McLaurin J, et al. A beta peptide immunization reduces behavioural impairment and plaques in a model of Alzheimer's disease. Nature 2000; 408: 979-982

17 Nicoll JA, Wilkinson D, Holmes C, Steart P, Markham H, Weller RO. Neuropathology of human Alzheimer disease after immunization with amyloid-beta peptide: A case report. Nat Med 2003; 9: 448-452

18 Akiyama H, Barger S, Barnum S, et al. Inflammation and Alzheimer's disease. Neurobiol Aging 2000; 21: 383-421

19 Wyss-Coray T, Loike JD, Brionne TC, Lu E, Anankov R, Yan F, Silverstein SC, Husemann JA. Adult mouse astrocytes degrade amyloid-beta in vitro and in situ. Nat Med 2003; 9: 453-457

20 Stanimirovic D, Satoh K. Inflammatory mediators of cerebral endothelium: A role in ischemic brain inflammation. Brain Pathol 2000; 10: 113-126

21 Westermann J, Engelhardt B, Hoffmann JC. Migration of T cells in vivo: Molecular mechanisms and clinical implications. Ann Intern Med 2001; 135: 279-295

22 Cserr HF, Knopf PM. Cervical lymphatica, the blood-brain barrier and the immunoreactivity of the brain: A new view. Immunol Today 1992; 13: 507-512

23 Knopf PM, Cserr HF, Nolan SC, Wu TY, Harling Berg CJ. Physiology and immunology of lymphatic drainage of interstitial and cerebrospinal fluid from the brain. Neuropathol Appl Neurobiol 1995; 21: 175-180

24 Kida S, Pantazis A, Weller RO. CSF drains directly from the subarachnoid space into nasal lymphatics in the rat. Anatomy, histology and immunological significance. Neuropathol Appl Neurobiol 1993; 19: 480-488

25 Nicholson C, Chen KC, Hrabetova S, Tao L. Diffusion of molecules in brain extracellular space: Theory and experiment. Prog Brain Res 2000; 125: 129-154

26 Zhang ET, Richards HK, Kida S, Weller RO. Directional and compartment-alised drainage of interstitial fluid and cerebrospinal fluid from the rat brain. Acta Neuropathol 1992; 233-239

27 Phillips MJ, Needham M, Weller RO. Role of cervical lymph nodes in autoimmune encephalomyelitis in the Lewis rat. J Pathol 1997; 182: 457-464

28 Lake J, Weller RO, Phillips MJ, Needham M. Lymphocyte targeting of the brain in adoptive transfer cryolesion-EAE. J Pathol 1999; 187: 259-265

29 Weller RO. Fluid compartments and fluid balance in the central nervous system. In: Williams PL, ed. Gray's Anatomy, 38th edn, Edinburgh: Churchill Livingstone, 1995: pp. 1202-1224

30 Zhang ET, Inman CB, Weller RO. Interrelationships of the pia mater and the perivascular (Virchow-Robin) spaces in the human cerebrum. J Anat 1990; 170: 111-123

31 Van Dorpe J, Smeijers L, Dewachter I, et al. Prominent cerebral amyloid angiopathy in transgenic mice overexpressing the london mutant of human APP in neurons. Am J Pathol 2000; 157: 1283-1298

32 Hutchings M, Weller RO. Anatomical relationships of the pia mater to cerebral blood vessels in man. J Neurosurg 1986; 65: 316-325

33 Vinters HV, Wang ZZ, Secor DL. Brain parenchymal and microvascular amyloid in Alzheimer's disease. Brain Pathol 1996; 6: 179-195

34 Yamada M. Cerebral amyloid angiopathy: An overview. Neuropathology 2000; 20: 8-22

35 CFAS NGoM. Pathological correlates of late-onset dementia in a multi-centre, community-based population in England and Wales. Lancet 2001; 357: 169-175

36 Holton JL, Ghiso J, Lashley T, et al. Regional distribution of amyloid-Bri deposition and its association with neurofibrillary degeneration in familial British dementia. Am J Pathol 2001; 158: 515-526

37 Scholz W. Studien zur Pathologie der Hirngefasse II. Die drusige Entartung der Hirnarterien und -capillaren. (Ein Form seniler Gefaesserkrankung). Zeilshrift fur die gesamte Neurologie und Psychiatrie 1938; 162: 694-715

38 Subash M, Weller RO. Cerebral amyloid angiopathy: Differential preservation of endothelial cell basement membranes. Neuropathol Appl Neurobiol 2003; 29: 184

39 Shinkai Y, Yoshimura M, Ito Y, et al. Amyloid beta-proteins 1-40 and 1-42(43) in the soluble fraction of extra- and intracranial blood vessels. Am Neurol 1995; 38: 421-428

40 Wisniewski HM, Wegiel J. Beta-amyloid formation by myocytes of leptomeningeal vessels. Acta Neuropathol (Berl) 1994; 87: 233-241

41 Alonzo NC, Hyman BT, Rebeck GW, Greenberg SM. Progression of cerebral amyloid angiopathy. Accumulation of amyloid-beta 40 in affected vessels. J Neuropathol Exp Neurol 1998; 57: 353-359

42 Weller RO, Yow HY, Preston SD, Mazanti I, Nicoll JAR. Cerebrovascular disease is a major factor in the failure of elimination of A-beta from the human brain: Implications for therapy of Alzheimer's disease. Ann NY Acad Sci 2002; 977: 162-168

43 Roher AE, Kuo YM, Potter PE, et al. Cortical cholinergic denervation elicits vascular A beta deposition. Ann NY Acad Sci 2000; 903: 366-373

44 Yow HY, Weller RO. A role for cerebrovascular disease in determining the pattern of beta amyloid deposition in Alzheimer's disease. Neuropathol Appl Neurobiol 2002; 28: 149

45 Greenberg SM, Briggs ME, Hyman BT, et al. Apolipoprotein E epsilon 4 is associated with the presence and earlier onset of hemorrhage in cerebral amyloid angiopathy. Stroke 1996; 27: 1333-1337

46 Nicoll JA, McCarron MO. APOE gene polymorphism as a risk factor for cerebral amyloid angiopathy-related hemorrhage. Amyloid 2001; 8 (Suppl. 1): 51-55

47 Greenberg SM, Vonsattel JP, Segal AZ, et al. Association of apolipoprotein E epsilon2 and vasculopathy in cerebral amyloid angiopathy. Neurology 1998; 50: 961-965

48 Winkler DT, Bondolfi L, Herzig MC, et al. Spontaneous hemorrhagic stroke in a mouse model of cerebral amyloid angiopathy. J Neurosci 2001; 21: 1619-1627

49 Vinters HV. Cerebral amyloid angiopathy: A microvascular link between parenchymal and vascular dementia? Ann Neurol 2001; 49: 691-693

50 Roher AE, Kuo Y-M, Esh C, Knebel C, Weiss N, Kaiback W, Lueurs DC, Childress JL, Beach TG, Weller RO, Korjohn TA. Cortical and leptomeningeal cerebro-vascular amyloid and white matter pathology in Alzheimer's disease. Mol Med 2003; 9: 112-122

51 DeMartos RB, Bales KR, Cummins DJ, Steven MP. Holtzman DM. Brain to plasma amyloid-beta efflux: A measure of brain amyloid burden in a mouse model of Alzheimer's disease. Science 2002; 295: 2264-2267

52 Pfeifer M, Boncristiano S, Bondolfi L, et al. Cerebral hemorrhage after passive anti-A-beta immunotherapy. Science 2002; 298: 1379

Roy O. Weller and James A.R. Nicoll

Neuropathology, Division of Clinical Neurosciences, University of Southampton School of Medicine, Southampton, UK

Correspondence and reprint requests to: Roy O. Weller, BSc, MD, PhD, FRCPath, Neuropathology, Division of Clinical Neurosciences, Mail Point 813, Southampton General Hospital, Southampton SO16 6YD, UK. [row@soton.ac.uk] Accepted for publication April 2003.

Copyright Forefront Publishing Group Sep 2003
Provided by ProQuest Information and Learning Company. All rights Reserved