Microvasculature changes and cerebral amyloid angiopathy in ...

7 downloads 0 Views 1MB Size Report
Feb 22, 2009 - elastic lamina, a tunica media and a fibrous adventitia. In young individuals, there .... in older people exhibit subintimal fibrosis, Xattening of the.
Acta Neuropathol (2009) 118:87–102 DOI 10.1007/s00401-009-0498-z

R EV IE W

Microvasculature changes and cerebral amyloid angiopathy in Alzheimer’s disease and their potential impact on therapy Roy O. Weller · Delphine Boche · James A. R. Nicoll

Received: 24 December 2008 / Revised: 8 February 2009 / Accepted: 9 February 2009 / Published online: 22 February 2009 © Springer-Verlag 2009

Abstract The introduction of immunotherapy and its ultimate success will require re-evaluation of the pathogenesis of Alzheimer’s disease particularly with regard to the role of the ageing microvasculature and the eVects of APOE genotype. Arteries in the brain have two major functions (a) delivery of blood and (b) elimination of interstitial Xuid and solutes, including amyloid- (A), along perivascular pathways (lymphatic drainage). Both these functions fail with age and particularly severely in Alzheimer’s disease and vascular dementia. Accumulation of A as plaques in brain parenchyma and artery walls as cerebral amyloid angiopathy (CAA) is associated with failure of perivascular elimination of A from the brain in the elderly and in Alzheimer’s disease. High levels of soluble A in the brain correlate with cognitive decline in Alzheimer’s disease and reXect the failure of perivascular drainage of solutes from the brain and loss of homeostasis of the neuronal environment. Clinically and pathologically, there is a spectrum of disease related to functional failure of the ageing microvasculature with “pure” Alzheimer’s disease at one end of the spectrum and vascular dementia at the other end. Changes in the cerebral microvasculature with age have a potential impact on therapy with cholinesterase inhibitors and especially on immunotherapy that removes A from plaques in the brain, but results in an increase in severity of CAA and no clear improvement in cognition. Drainage of A along perivascular pathways in ageing artery walls may need to be improved to maximise the potential for improvement of cognitive function with immunotherapy.

R. O. Weller (&) · D. Boche · J. A. R. Nicoll Clinical Neurosciences, University of Southampton School of Medicine, LD74, South Laboratory & Pathology Block, Southampton General Hospital, Southampton SO16 6YD, UK e-mail: [email protected]

Keywords Structure and functions of normal cerebral arteries · Perivascular drainage of A · Cerebral amyloid angiopathy · Microvascular disease · Arteriosclerosis · Arteriolosclerosis · Vascular dementia · Alzheimer’s disease · Brain homeostasis · Cholinesterase inhibitors · Immunotherapy

Introduction Alzheimer’s disease is a disorder of elderly individuals and is characterised by the failure of elimination of hyperphosphorylated tau protein from neurons and the failure of elimination of amyloid- (A) from brain parenchyma and blood vessel walls [48]. NeuroWbrillary tangles (NFT) within neurons in the brain are composed largely of ubiquitin and the microtubule-associated protein tau [48]. The accumulation of neuroWbrillary tangles appears to be associated with the failure of the ubiquitin–proteasome system to dispose of hyperphosphorylated tau from ageing neurons [11, 46]. A accumulates mainly in the extracellular spaces of the brain parenchyma as insoluble plaques and in the walls of arteries and capillaries as cerebral amyloid angiopathy (CAA) [48]. The accumulation of insoluble A in the brain and eventually the rise in levels of soluble A in Alzheimer’s disease [49, 51] reXect the failure of mechanisms by which A is normally eliminated from the brain [92]. Advancing age is also a major risk factor for cerebrovascular disease that aVects large and small arteries supplying the brain [31]. Cerebrovascular disease aVects two of the major functions of cerebral arteries viz: (a) the supply of blood to the brain and (b) the perivascular drainage of interstitial Xuid and solutes that constitutes the lymphatic drainage of the brain [90]. Both the Xow of blood and the drainage of Xuid and solutes fail with age and these failures

123

88

Acta Neuropathol (2009) 118:87–102

contribute to the pathogenesis of vascular dementia, CAA and Alzheimer’s disease. This review concentrates on how structural and functional changes in the microvasculature in the elderly interfere with the blood supply of the brain and contribute to the failure of elimination of A from the brain in the pathogenesis of Alzheimer’s disease. We explore the concept that age changes in cerebral arteries are a major factor not only in (a) the pathogenesis of ischaemic lesions in the brain, but also in (b) the pathogenesis of CAA and Alzheimer’s disease. Finally, we examine the interface between Alzheimer’s disease and vascular dementia and how age changes in the microvasculature have an impact on current therapies for Alzheimer’s disease, especially on immunotherapy [16].

Structure and functions of the cerebral vasculature

Fig. 1 Normal young leptomeningeal artery in transverse section showing a convoluted internal elastic lamina (IEL), a media of smooth muscle cells and a thick collagenous adventitia (Adv). Klüver–Barrera stain, bar 20 m

The cerebral vasculature has several functions that are reXected in the structure of blood vessels within the brain. Arteries and capillaries supply blood, nutrients and inXammatory cells to the brain but their walls are also the pathways for the drainage of interstitial Xuid from the brain [17, 90]. Branches of the internal carotid and vertebral arteries form major cerebral arteries that have relatively thin, translucent and malleable walls in young individuals [31]. Leptomeningeal arteries derived from the cerebral arteries branch and penetrate the superWcial and deep surfaces of the brain to supply structures in the cerebrum, cerebellum and brain stem. Leptomeningeal arteries Medium-sized leptomeningeal arteries have an internal elastic lamina, a tunica media and a Wbrous adventitia. In young individuals, there is little Wbrous tissue separating the endothelium from the internal elastic lamina (Fig. 1) and very little Wbrous tissue in the media. The internal elastic lamina is highly convoluted (Fig. 1) which reXects a lack of stiVness in young artery walls and their ability to recoil following expansion by the pulse wave. There is a relatively thick collagenous adventitia around leptomeningeal arteries even in children and young adults (Fig 1). Electron microscopy reveals a thin layer of leptomeningeal (arachnoid) cells coating the outer aspects of leptomeningeal arteries separating the adventitia from the CSF in the subarachnoid space [66, 98] (Fig. 2). Tracer studies in experimental animals suggest that interstitial Xuid and solutes drain from the brain along the tunicae media et adventitia of leptomeningeal arteries to cervical lymph nodes [17, 80, 90].

123

Fig. 2 Normal young leptomeningeal artery in transverse section showing the outer coating of leptomeninges (Lep) enclosing the collagenous perivascular space (PVS). Smooth muscle cells of the media (SMC). Transmission electron micrograph, bar 20 m

Cortical arteries Branches of leptomeningeal arteries penetrate the cerebral cortex perpendicular to its pial surface to form arterioles that supply cortex and subcortical white matter [24]. Those arterioles that supply the subcortical white matter pass through the cortical layers without branching [24]. The walls of arterioles in the cerebral cortex are compact and have no natural perivascular space [66, 98] (Fig. 3). Vascular endothelium surrounds the lumina of cortical arterioles and, as there is no internal elastic lamina, the endothelial basement membrane is fused with basement membranes of smooth muscle cells in the tunica media. A thin sheath of

Acta Neuropathol (2009) 118:87–102

Fig. 3 A portion of the wall of a normal young cortical artery. Endothelium (Endo) lines the lumen and the tunica media is composed of smooth muscle cells separated by basement membrane (BM). A layer of leptomeningeal cells (Lep) separates the tunica media from the astrocytes (Ast) of the perivascular glia limitans. There is no adventitia or perivascular space. Transmission electron micrograph, bar 1 m (reproduced with permission from reference [66])

89

Fig. 4 Scanning electron micrograph of an artery (Art) in the basal ganglia to show the two layers of leptomeniniges (Lep 1, Lep 2) surrounding the artery and separated by a perivascular space (PVS). Basement membrane of the glia limitans (BM), bar 250 m (reproduced with permission from reference [65])

leptomeningeal cells, derived from the pia mater [98], separates the smooth muscle cells of the tunica media from the astrocytes of the perivascular glia limitans [66, 98] (Fig. 3). Arteries supplying the deep grey matter of the cerebral hemispheres Deep penetrating arteries arise from the circle of Willis and its major branches at the base of the brain to supply the deep grey matter structures comprising the basal ganglia and thalamus [31]. These arteries diVer in their structure from those in the cortex as they are invested by a double layer of leptomeninges and an expandable perivascular space (Figs. 4, 5) [65, 89]. In older individuals, the perivascular spaces in the basal ganglia are often expanded (état lacunaire) [64]; the expansion may be great enough to result in space occupying lesions [70]. Capillaries

Fig. 5 Diagram comparing the structure of the walls of (a) a cortical artery with a single layer of leptomeninges—solid line (1)—that separates the glia limitans—dotted line (2) from the tunica media (3). There is no perivascular space. b An artery from the basal ganglia in which there is a second layer of leptomeninges (4). A perivascular space separates the two layers of leptomeninges (reproduced with permission from reference [89])

Capillaries are the site of the blood–brain barrier for the exchange of Xuid and solutes between the blood and the brain and this is a major source of interstitial Xuid [1]. Capillary basement membranes also form the initial part of the perivascular pathway for the drainage of interstitial Xuid and solutes from the brain [17]. Endothelial cells of cerebral capillaries are almost devoid of vesicles and joined by tight junctions (Fig. 6) [1, 66], reXecting the presence of the blood–brain barrier [1, 10]. Pericapillary basement membranes are in direct

contact with the narrow and restricted extracellular (interstitial) spaces between the neuronal and glial processes of the grey matter (Fig. 6). Interstitial Xuid and solutes secreted by the capillary endothelium are distributed by bulk Xow along pericapillary basement membranes to supply nutrients to the neuropil [1]. Waste products and soluble metabolites return in the interstitial Xuid to drain out of the brain along basement membranes in the walls of capillaries and arteries [17, 90]. The continuity of the

123

90

Acta Neuropathol (2009) 118:87–102

Perivascular drainage of interstitial Xuid and solutes from the brain

Fig. 6 A transmission electron micrograph of a capillary in the cerebral cortex from a young individual shows endothelial cells lining the capillary lumen and joined by tight junctions (TJ). A basement membrane coats the abluminal surface of the endothelial cells and is direct contact (asterisks) with the narrow extracellular space of the cortex, bar 1 m (reproduced with permission from reference [66])

extracellular space with capillary basement membranes (Fig. 6) seems to be well suited for the perivascular bulk Xow of interstitial Xuid and solutes from the brain [1, 17]. Capillary endothelium is also a site for the absorption of solutes from the brain into the blood. Particularly relevant to Alzheimer’s disease is the absorption of A by receptormediated pathways involving lipoprotein receptor-related protein-1 (LRP-1) [12, 78], p-glycoprotein [20] and receptor for advanced glycation end products (RAGE) [22]. Although absorption of A into the blood is six times faster than drainage of A along perivascular lymphatic pathways, it appears to fail with age [12, 78]. Veins and venules Veins in grey and white matter have a relatively large lumen and thin walls that lack smooth muscle cells [17, 98]. These characteristics distinguish veins from arteries and capillaries. The leptomeningeal cells around veins do not form a complete sheath and there is a perivascular space around veins containing a few collagen Wbres [98]. Postcapillary venules are the site for the receptor-mediated entry of inXammatory cells into the brain [10, 25]. Once through the endothelium, inXammatory cells traverse the glia limitans [47, 61] into brain tissue or accumulate in the dilated perivascular spaces around venules and veins. Lymphocytes that enter the CNS appear to undergo apoptosis and there is no deWned pathway for the migration of lymphocytes from CNS parenchyma to regional lymph nodes [90]. Lymphocytes and dendritic cells in the CSF, however, may migrate to lymph nodes via nasal lymphatics [34, 35].

123

Approximately 140 ml of CSF and 280 ml of interstitial Xuid comprise the extracellular Xuids associated with the human CNS [13]. CSF in the ventricles and subarachnoid spaces drains from the intracranial and spinal compartments via arachnoid villi and granulations and to some extent via nasal lymphatics and along nerve roots [90]. Interstitial Xuid and solutes drain from the brain along the walls of capillaries and arteries and this function of the cerebral vasculature is eVectively the lymphatic drainage of the CNS parenchyma [17, 90]. The drainage pathway for interstitial Xuid is largely separate from the CSF in the normal CNS [80, 90]. Experimental studies have shown that perivascular drainage of ISF and solutes out of the brain is along the basement membranes of capillaries and arteries [17]. When Xuorescent 3 kDa dextran or 40-kDa ovalbumin is injected as tracer into the grey matter of the mouse brain, they initially spread diVusely through the brain parenchyma but almost immediately enter the basement membranes of capillaries and arteries to drain out of the brain. Fluorescent tracers do not appear to enter the walls of venules and veins within the brain, so these blood vessels do not seem to be part of the perivascular drainage system [17]. Other studies have shown that radioactive tracers and horseradish peroxidase injected into animal brains drain along the media and adventitia of leptomeningeal arteries to cervical lymph nodes [80]. The rate of lymphatic drainage of interstitial Xuid and solutes from the brain is comparable to that of lymphatic drainage from other organs [80]. Experimental evidence in rats has shown that only 10–15% of ISF draining from the brain leaks into the CSF [80]. This suggests that drainage of interstitial Xuid and solutes via the perivascular route is separate from the CSF [90].

Lymphatic drainage of the human brain In humans, A acts as a natural tracer for perivascular drainage [90]. The distribution of A in the basement membranes of capillary and artery walls and in the adventitia of leptomeningeal arteries in CAA [86, 92] corresponds exactly with the pathways outlined by tracers in the experimental studies [17, 80]. Figure 7a summarises the structure of the carotid arterial blood supply to the cerebral cortex and identiWes the tissue components of the artery and capillary walls that appear to form the lymphatic drainage pathways of the human brain [2, 38, 89, 98]. The distribution of A in CAA (Fig. 7b) outlines the drainage pathways for solutes from the brain, initially along the basement membranes in the walls of capillaries and arteries [17, 66] and then out of the skull along

Acta Neuropathol (2009) 118:87–102

91

the adventitia of the carotid artery to lymph nodes in the neck. Neither tracers injected into the interstitial Xuid of the brain nor A is detectable in the walls of the carotid artery in the neck [79, 80] which suggests that solutes draining from the brain leave the carotid artery wall at the base of the skull for adjacent cervical lymph nodes [80].

and the deposition of A in the tunicae media et adventitia of cerebral arteries as CAA [81, 92]. Focal Wbrous thickening and deposition of cholesterol and other lipids in the walls of larger cerebral arteries and occasionally of leptomeningeal arteries constitutes atherosclerosis [31]. Functional failure of the cerebral circulation associated with atherosclerosis results in ischaemic damage and infarction due to thrombotic–embolic occlusion of large or small cerebral arteries. Infarcts may be large or small and result in strokes or in multiple small infarcts associated with vascular dementia [31]. Arteriosclerosis is characterised by the generalised Wbrosis and stiVening of artery walls [55] and this may result not only in a reduction of cerebral blood [26] but also in the failure of perivascular drainage of interstitial Xuid and solutes from the brain in elderly individuals and in Alzheimer’s disease [92]. Medium-sized leptomeningeal arteries in older people exhibit subintimal Wbrosis, Xattening of the internal elastic lamina (Fig. 8a, b) and an increase in the amount of Wbrous tissue in the tunica media when compared with similar arteries in the young (Fig. 1). As a result, arteries in the elderly do not recoil in the same way as younger vessels [55] and this may interfere with the motive force for the perivascular lymphatic drainage of the brain [72]. Arteriolosclerosis (lipohyalinosis, hyalinosis [31]) aVects the arterioles in the grey and white matter of the cerebral hemispheres [81, 84]. Arterioles do not posses an internal elastic lamina but they do show an increase in Wbrous tissue in the tunica media with age resulting in sclerosis and stiVening of the vessel walls [31, 81]. DiVerent regions of the brain are aVected by arteriolosclerosis in the ageing population and in Alzheimer’s disease [81]. Arteriolosclerosis initially aVects arterioles in the basal ganglia, then those in the cerebral white matter, cerebral and cerebellar cortices and thalamus. Arterioles in the brainstem are aVected later [81]. Arterioles also show an increase in tortuosity with age, particularly in the white matter [54]. Capillaries in the cerebral cortex and white matter show an increase in thickness of basement membranes with associated deposition of collagen Wbres with age and in Alzheimer’s disease [27, 28]. These changes in capillary basement membranes may aVect blood Xow, transfer of nutrients from blood to brain [27, 28] and the perivascular drainage of Xuid and solutes from the brain [90, 92].

Structural changes and failure of function in the cerebral vasculature with age

Pathophysiology of the microvasculature in Alzheimer’s disease

The major changes that occur in the cerebral vasculature with advancing age and in Alzheimer’s disease are Wbrosis and stiVening of the walls of arteries and arterioles [55, 81]

The changes that occur in the cerebral microvasculature with age have two major eVects on the brain (a) reduction in blood supply (hypoperfusion) with resulting ischaemia,

Fig. 7 A diagrammatic summary of blood Xow and perivascular lymphatic drainage of the cerebral cortex. a Blood Xows into the brain (red arrow) and interstitial Xuid and solutes Xow out along capillary and artery walls (green arrow). As the carotid artery penetrates the base of the skull it acquires an outer leptomeningeal (arachnoid) coating (dark blue) which is reXected onto the surface of the brain as the pia mater but also forms a sheath around arteries in the cerebral cortex. Capillaries lack a leptomeningeal coat and the capillary endothelial basement membrane is in direct contact with the extracellular space and interstitial Xuid of the brain. Cortical arteries have no adventitia. The relatively thick collagenous adventitia around leptomeningeal arteries (light blue layer) is continuous with the adventitial of the carotid artery in the neck. b Cerebral amyloid angiopathy (CAA) outlines the perivascular lymphatic pathway by which interstitial Xuid and solutes (including A) drain from the brain. A (shown as a green line) is deposited in basement membranes of (i) capillaries, (ii) the tunica media of cortical and (iii) leptomeningeal arteries. A also accumulates in the adventitia of leptomeningeal arteries (iii) as part of the drainage pathway to cervical lymph nodes

123

92

Acta Neuropathol (2009) 118:87–102

Fig. 8 Age changes in leptomeningeal arteries. a A leptomeningeal branch of the anterior cerebral artery from an elderly individual. The lumen has not collapsed post mortem (cf a young artery in Fig. 1) due to stiVness of the wall. A layer of leptomeninges (Lep) separates the

vessel wall from the CSF in the subarachnoid space. b Enlargement of the artery wall showing Xattening of the internal elastic lamina (IEL) and the thickness of the Wbrous adventitia (Adv). Klüver–Barrera stain. a Bar 40 m, b Bar 25 m

and (b) failure of perivascular drainage of interstitial Xuid associated with reduced elimination of A, the development of CAA and loss of homeostasis of the extracellular Xuid in cerebral grey and white matter [92].

least frequently aVected. CAA appears to occur in arteries in diVerent areas of the brain in a hierarchical pattern that follows closely the deposition of A in plaques in brain parenchyma in Alzheimer’s disease [81]. Leptomeningeal arteries and arterioles of the cerebral neocortex are involved in the early stages of CAA followed by extension of CAA to arterioles in the allocortex and the midbrain. CAA less frequently involves arteries and arterioles in the basal ganglia, and thalamus; the lower brainstem is least frequently aVected by deposition of A in CAA [81]. The relative infrequency of A plaques and CAA in the basal ganglia and their late appearance in Alzheimer’s disease [81] may be related to the structure of the artery walls. Whereas there is no perivascular space around cortical arterioles (Fig. 3) [66, 98], there is an expandable perivascular space around arteries in the basal ganglia (Figs. 4, 5) [65] that may facilitate the drainage of interstitial Xuid and A from the basal ganglia. Immunocytochemistry shows the pattern of involvement of artery and capillary walls by CAA. Smaller leptomeningeal arteries in the cerebral sulci (Fig. 9) show deposition of A initially in the basement membranes in the tunica media [66, 92, 95] (Fig. 10). In larger leptomeningeal arteries, A may be deposited in the adventitia (Fig. 11) as well as in the tunica media [86]. Deposits of A are occasionally associated with the walls of veins (Fig. 10) and arachnoid trabeculae in the subarachnoid space, especially when CAA in the arteries is severe in patients who have received immunotherapy for Alzheimer’s disease [16]. The mechanism for such deposition is not known, but as amyloid deposits in artery walls

Hypoperfusion The main pathological consequences of hypoperfusion are lacunar infarcts in the basal ganglia, microinfarcts in the cerebral cortex, especially in the water-shed zones, and leukoaraiosis of the cerebral white matter [26, 48]. Lacunar infarcts of the subcortical grey matter and ischaemia of the white matter are major pathological features of vascular dementia but they also occur in cases of mixed dementia in which the features of Alzheimer’s disease and vascular dementia are present in the same brain [41]. Failure of perivascular drainage The major consequences for the elderly brain of failure of perivascular drainage of interstitial Xuid and solutes appear to be (a) CAA, (b) accumulation of insoluble and soluble A in cerebral grey matter and (c) retention of Xuid in the subcortical cerebral white matter (leukoaraiosis).

Cerebral amyloid angiopathy Cerebral amyloid angiopathy is a feature of the ageing brain and patients with Alzheimer’s disease [5]. It aVects the leptomeningeal arteries most commonly; cortical arteries are also frequently involved and capillaries are

123

Acta Neuropathol (2009) 118:87–102

Fig. 9 Cerebral amyloid angiopathy (CAA). A section of Alzheimer brain from a patient immunised with A42 showing extensive CAA of small leptomeningeal arteries (Lep A) within a sulcus, and of cortical arteries (Cort A). Numerous plaques of A are present in the cortex. Immunohistochemistry for A, bar 80 m

Fig. 10 Cerebral amyloid angiopathy (CAA). Enlargement of part of Fig. 9 showing small leptomeningeal arteries laden with A. In an oblique section of an artery wall, A is in the basement membranes (BM) of the tunica media. Fragments of A attached to the wall of a vein (AV) have probably become detached from leptomeningeal arteries. Immunohistochemistry for A, bar 80 m

tend to be brittle (Preston and Weller, unpublished observation), it is possible that fragments of A break from the surface of leptomeningeal arteries and adhere to the arachnoid and to the walls of veins in the subarachnoid space. Amyloid- accumulates in the basement membranes of the tunica media of cortical arteries in the initial stages of CAA [92]. However, in the later stages of CAA, smooth muscle cells and their basement membranes are replaced by A (Fig. 12). Deposition of A is seen in a linear distribution in capillary basement membranes (Fig. 12) and excrescencies of A may form on the capillary walls as globular or Wlamentous “Drusen” [66, 73].

93

Fig. 11 Cerebral amyloid angiopathy (CAA) of a medium-sized leptomeningeal artery on the surface of a gyrus. A is deposited in the adventitia (CAA Adv). Arteries in the underlying cortex (Cort A) also have CAA. Immunohistochemistry for A1-40, bar 40 m

Fig. 12 Cerebral amyloid angiopathy (CAA) involving a cortical artery (Cort A) and its capillary bed (Cap CAA). Immunohistochemistry for A, bar 10 m

Pathogenesis of cerebral amyloid angiopathy The major evidence now indicates that CAA is due to the deposition of A in the perivascular pathways by which interstitial Xuid and solutes drain from the brain [36, 92]. A is derived from the enzymic cleavage of the transmembrane amyloid precursor protein (APP) [76, 77] resulting in a small pool of soluble A in normal brain [43]. Several pathways have been identiWed for the elimination of A from the normal brain. Enzymes, such as neprilysin and insulin degrading enzyme that degrade A are produced by neurons and glia and are also expressed in cerebral vessel

123

94

walls [52]. A is also absorbed into the blood by mechanisms involving LRP-1 [12, 78], and p-glycoprotein [20]. However, enzymic degradation by neprilysin and absorption of A into the blood fail with age [53, 78]. The decrease in neprilysin activity is associated with an increase in CAA [53]. The perivascular route for the elimination of A appears to function throughout life. A has been detected biochemically in the walls of cerebral arteries in individuals at the age of 20 years and not just in the elderly [79]. Although A is present in the walls of intracranial arteries, it is undetectable in the walls of the extracranial portion of the internal carotid artery [79]. This vascular distribution of A supports the suggestion that A drains along perivascular lymphatic drainage pathways in the walls of intracranial arteries to lymph nodes in the neck [90]. Entrapment of A in the perivascular drainage pathways in the pathogenesis of cerebral amyloid angiopathy Evidence for entrapment of A in the perivascular drainage pathways comes from a number of observations [36, 58, 91, 92]. 1. The distribution of A in the basement membranes of capillaries and arteries and its presence in the adventitia of leptomeningeal arteries corresponds exactly to the distribution of tracers injected into the brain to outline perivascular lymphatic drainage pathways for the elimination of Xuid and solutes from the brain [17, 21, 80, 90]. 2. Transgenic mice that produce excessive amounts of A only in neurons develop CAA [36]. This suggests that A deposits in the perivascular drainage pathways are derived from the brain [36]. 3. It is not only A that accumulates in vessel walls in CAA. A number of amyloidogenic proteins are involved in diVerent types of CAA [67]. They include cystatin, transthyretin, gesolin, prion protein [67] and the amyloid deposited in the vessel walls in the British and Danish types of dementia [45]. The variety of proteins involved suggests a common pathogenesis for the diVerent types of CAA. Thus, there appears to be the phenomenon of protein-elimination failure arteriopathy (PEFA) that is common to all types of CAA [92]. In familial CAA associated with intracerebral haemorrhage [36, 85] there are mutations in the APP gene that involve the A peptide itself resulting in an A that is resistant to breakdown by neprilysin [85]. In these patients, the lack of elimination of A by neprilysin appears to result in an overload of the perivascular drainage system and the development of severe CAA at an early age [85].

123

Acta Neuropathol (2009) 118:87–102

Other theories for the pathogenesis of CAA Other mechanisms for the pathogenesis of CAA have been proposed. Scholz [73] published the Wrst comprehensive account of the pathology of CAA in 1938 and suggested that the amyloid in the vessel walls and the surrounding brain was derived from the blood. This was also proposed by Glenner [33] who Wrst isolated A from leptomeningeal arteries with CAA. Evidence against this proposal comes mainly from observations in transgenic mice that produce human A only in neurons; they develop CAA [36] and a marked increase in A in the brain is associated with decreased clearance of A and the development of CAA [42]. Most cells in the body produce A [77] including smooth muscle cells in the walls of cerebral arteries [58, 95]. Although vascular smooth muscle cells may contribute to the A load in CAA [95] they are probably not the sole source for a number of reasons. (1) Transgenic animals develop CAA when the A is only produced by neurons in the brain [36], (2) patients with the British type of dementia deposit the amyloid protein, ABri, in vessel walls as CAA but no mRNA for ABri has been detected in cerebral vascular smooth muscle cells [45]. (3) A is deposited in the basement membranes of capillaries that posses no smooth muscle cells [66], (4) the highest concentration of immunocytochemically detectable insoluble A in CAA in humans is in the smaller cerebral and leptomeningeal artery walls and not in the larger intracranial artery walls [91]. Furthermore, A is not detectable in the walls of the extracranial carotid arteries in the neck even in cases of CAA [79]. If the majority of A were produced by vascular smooth muscle cells in CAA, the larger vessels with more smooth muscle calls would be expected to develop the most severe CAA and this is not the case. Age changes in cerebral arteries and the pathogenesis of CAA Although the evidence outlined above suggests that the pathogenesis of CAA involves the deposition of A and other amyloids in the perivascular pathways by which interstitial Xuid and solutes drain from the brain, it does not explain why the majority of cases of CAA occur in the elderly. Age is a major risk factor for CAA and for arteriosclerosis and other forms of cerebrovascular disease, so the changes in arteries that occur with age may play a role in the pathogenesis of CAA. Theoretical studies modelling the motive force for perivascular drainage of interstitial Xuid and solutes from the brain suggest that the contrary (reXection) wave that follows each pulse wave drives interstitial Xuid and solutes along artery walls in the reverse direction to the Xow of

Acta Neuropathol (2009) 118:87–102

blood in the artery lumen [72]. This mechanism would require a valve-like action to prevent back Xow of Xuid during passage of the pulse wave [72]. The nature of the valvelike action is not known. However, one possibility is that the conformational changes in vascular basement membranes that occur as the artery walls expands during the passage of the pulse wave and recoils during diastole may provide the valve-like eVect. StiVening of artery walls with age reduces the amplitude of the pulse wave and may thus diminish the motive force for drainage of interstitial Xuid and solutes from the brain [72]. Slowing of perivascular drainage probably induces the formation of the A amyloid Wbrils in the vascular basement membranes that further impairs the drainage of Xuid and solutes from the brain [92].

Capillary amyloid angiopathy Capillary CAA, in which A is deposited in capillary basement membranes in Alzheimer’s disease (Fig. 12) is much less common than CAA in cerebral and leptomeningeal arteries. Even in cases of severe CAA aVecting leptomeningeal and cortical arteries, capillary CAA is only abundant in about one-third of cases [60]. Capillary CAA has been associated with a high frequency apolipoprotein E 4 genotype [82] and with Alzheimer’s disease pathology [4], but does not correlate with CAA of leptomeningeal and cortical arteries [3, 5, 82]. The mechanisms by which the A is deposited in the capillary walls in the pathogenesis of capillary CAA are unclear but may be due to an overload of A42 draining into capillary basement membranes from degraded plaques in the brain parenchyma [16, 93]. Capillary CAA is frequently seen as clusters of A-laden capillaries in regions of cerebral cortex that are devoid of A plaques [60]. Serial transverse sections of cortical arteries feeding areas of capillary CAA has revealed an association between capillary CAA and thrombotic–embolic occlusion of cortical arteries [93, 96]. The inverse relationship between capillary CAA and A plaques suggests that occlusion of cortical arterioles results in focal ischaemia and destruction of the A plaques by activated microglia. A released from the plaques then accumulates in perivascular drainage pathways in capillary walls. A42 in Alzheimer’s disease is predominantly in the plaques [69] and the more soluble A40 is located in artery walls in CAA (Fig. 11). In capillary CAA it is predominantly A42 that is deposited in the capillary walls [4]. This suggests that solubilised plaque A may be overloading the perivascular drainage system in capillary CAA [96]. Capillary CAA is also seen in the brains of patients with Alzheimer’s disease who have been immunised with A42.

95

The destruction of A plaques by microglia that occurs in these cases may also result in overloading the perivascular drainage system [16].

Relationship of CAA to Alzheimer’s disease Several complications are associated with CAA. The most devastating is acute intracerebral haemorrhage associated with sporadic or familial CAA [97]. Granulomatous inXammation may also occur in relation to amyloid laden vessels in CAA [75]. The relationship of CAA to Alzheimer’s disease is less certain. In community studies there is an association between severe CAA and dementia [18] but both positive [81] and negative [83] correlations between CAA and A plaques in Alzheimer’s disease have been made. Most quantitative studies of CAA are performed on histological sections of cerebrum cut in a coronal plane. These investigations do not take account of the long drainage pathways along artery walls. Blockage of the drainage pathways at one point along the wall of an artery by the deposition of A may have an eVect on the eYciency of drainage of A along substantial lengths of the pathway. Examination of isolated leptomeningeal and cortical arteries [91] has shown that amyloid deposition does not necessarily involve the whole length of an artery and gaps in the amyloid deposition are present. This may mean that quantitation of CAA in coronal histological sections underestimates both the amount of A in the vessel walls and its eVect on the elimination of A from the brain. Smaller leptomeningeal arteries are more severely involved by CAA than the larger arteries in humans [91]. However, some transgenic mice show a pattern of involvement in which A is initially deposited in the walls of the larger arteries near the circle of Willis [23]; then there is a progressive spread of CAA to involve the smaller branches. This suggests a damming back process in which the drainage channels for A in the walls of larger arteries are blocked with subsequent deposition of A in the more proximal parts of the drainage pathways in the walls of the smaller arteries. Until the dynamics of CAA are fully understood, quantitative estimates of CAA should perhaps include the examination of isolated cerebral and leptomeningeal arteries [91]. The full eVects of CAA on the brain are poorly understood. One of the major roles of A in Alzheimer’s disease may be the obstruction of lymphatic drainage of interstitial Xuid and solutes from the brain. This would lead to (1) the early deposition of insoluble plaques of A in the brain parenchyma, (2) the eventual rise in soluble A in the cortex that correlates with cognitive decline in Alzheimer’s disease [49, 51] and (3) the accumulation of Xuid in the

123

96

white matter in leukoaraiosis [68]. It has been suggested that an increase in oligomeric A species is associated with cognitive decline in Alzheimer’s disease [87]. However, it is also possible that the rise in the level of soluble A related to dementia in Alzheimer’s disease is a reXection of a general failure of drainage of metabolites from the brain, blocked by CAA. Such a general failure of drainage may result in loss of homeostasis of the neuronal environment and failure of neuronal function in Alzheimer’s disease.

Acta Neuropathol (2009) 118:87–102

Vascular Factors in Alzheimer’s disease Atherosclerosis Arteriosclerosis

Failure of perivascular drainage of ISF and solutes including A from the brain

Fibrosis and stiffening of artery walls and hypoperfusion of grey and white matter Thromboembolic occlusion of arteries and infarction

Cerebral Amyloid Angiopathy

White matter disease in Alzheimer’s disease and in vascular dementia Changes in the white matter with increased signal in T2 weighted MRI images (leukoaraiosis) are frequently seen in the elderly, in Alzheimer’s disease and in vascular dementia; such changes are associated with low cognitive performance [14, 29]. Abnormal signal suggesting an increase in Xuid content of the white matter is seen in the periventricular regions and in the subcortical white matter [29] (see Fig. 16). The periventricular white matter lesions may indicate a disturbance of CSF drainage and the infusion of CSF into periventricular white matter [88, 90]. Leukoaraiosis in the subcortical white matter, however, has been correlated with arteriolosclerosis in arteries in the white matter and with markers of tissue hypoxia [29]. This suggests that there is ischaemia of the white matter due to the age changes in the arteries supplying the white matter. Leukoaraiosis has also been correlated with amyloid load in the brain parenchyma [19] and with severe CAA in the leptomeningeal arteries from which the arterial supply of the white matter arises [68]. This raises the possibility that white matter changes in Alzheimer’s disease result from a combination of ischaemia and a failure of Xuid drainage along perivascular pathways blocked by the deposition of A and by Wbrosis in the walls of arteriolosclerotic arteries in the white matter [29, 68].

Relationship between Alzheimer’s disease and vascular dementia Microvascular changes in the elderly brain are associated with a spectrum of dementias ranging from vascular dementia to Alzheimer’s disease with many cases showing a mixed dementia with features of both Alzheimer’s disease and of vascular dementia [41]. At one end of the spectrum (Fig. 13), demented patients exhibit pathological changes characteristic of Alzheimer’s disease with the accumulation of neuroWbrillary tangles and the deposition of A as plaques in brain parenchyma and as CAA. At the other end of the spectrum, there are the heterogeneous pathological

123

Intracellular accumulation of Tau and NFTs

White Matter Lesions (Leukoaraiosis)

Ischaemic damage to the brain

Accumulation of insoluble and soluble A in the brain in Alzheimer’s disease

Vascular Dementia

Mixed dementia Fig. 13 Relationship between age changes in the microvasculature, Alzheimer’s disease and vascular dementia. Arteriosclerosis (and arteriolosclerosis) has two major eVects. The left hand side of the diagram depicts the failure of perivascular drainage along the ageing arteries resulting in cerebral amyloid angiopathy and accumulation of A in the Alzheimer brain. The other eVect of arteriosclerosis, on the right, is hypoperfusion. Atherosclerosis mainly has its eVect through thromboembolism of cerebral arteries and cerebral infarction. Features of both Alzheimer’s disease and vascular dementia are present in mixed dementia

features of vascular dementia with multifocal lacunar infarcts in the basal ganglia and thalamus with or without larger cerebral infarcts. Ischaemic lesions are also seen in the hippocampus and in the white matter [40]. In mixed dementia the features of Alzheimer pathology and ischaemic lesions appear to act synergistically [40, 41]. The presence of lesions of vascular dementia results in a signiWcant deterioration in cognitive function in the earliest stages of Alzheimer’s disease [26]. Figure 13 illustrates how microvascular changes in the ageing brain play an important role throughout the spectrum of Alzheimer’s disease, mixed and vascular dementias with varying proportions of ischaemia and failure of perivascular lymphatic drainage.

Impact on therapy Treatment of Alzheimer’s disease with cholinesterase inhibitors is well established [32] and immunotherapy for the clearance of A plaques from the brain is under development [37]. The changes in the microvasculature that occur with age potentially have an impact on both forms of therapy.

Acta Neuropathol (2009) 118:87–102

Treatment with cholinesterase inhibitors Alzheimer’s disease is characterised not only by the accumulation of neuroWbrillary tangles and by the deposition of A in brain parenchyma and in blood vessel walls but also by a reduction in the cholinergic system in the brain [7, 44]. This has led to the introduction of acetylcholinesterase inhibitors for the treatment of Alzheimer’s disease [32]. The cholinergic deWcit appears be related to the accumulation of A in the brain [7, 8, 32]. Patients with dementia with Lewy bodies who were treated with cholinesterase inhibitors were found to have signiWcantly reduced levels of parenchymal A deposition [6], whereas blockade of muscarinic receptors in patients with Parkinson’s disease resulted in a 2.5-fold increase in the density of A plaques [63]. One of the mechanisms by which the accumulation of A may be associated with a cholinergic deWcit is illustrated by experimental ablation of the nucleus basalis in rabbits by a selective immunotoxin [9]. This results in cortical cholinergic deaVerentation, and leads to deposition of A in cerebral blood vessels (CAA) and in the perivascular neuropil. Biochemical measurements revealed a 2.5- and 8fold elevations of cortical A40 and A42, respectively [9]. These results suggest that cholinergic deaVerentation in Alzheimer’s disease may reduce the elimination of A along perivascular pathways resulting in CAA and plaque formation in the brain. So part of the eVect of treatment of Alzheimer’s disease with cholinesterase inhibitors may be enhanced elimination of A along perivascular drainage pathways and their eYcacy in this way may be aVected by age changes in the microvasculature.

Impact of age changes in the cerebral vasculature on A immunotherapy for Alzheimer’s disease In pursuing the amyloid hypothesis for the pathogenesis of Alzheimer’s disease, it was found that immunization of APP transgenic mice with A peptide could prevent or reverse age-related A plaques accumulation [71]. Subsequent studies have conWrmed that clearance of A plaques can occur as a result of immunotherapy in humans with Alzheimer’s disease (Elan Pharmaceuticals AN1792) [16, 30, 37, 50, 56, 57], although there is little evidence yet that this beneWts cognitive function [37]. The results of immunotherapy in humans are summarised in Fig. 14. In non-immunised patients with Alzheimer’s disease (Fig. 14a), there are numerous plaques of A and neuroWbrillary tangles within neurons. Dystrophic neurites are associated with A plaques (neuritic plaques) and neuropil threads are present in the parenchyma of the grey matter. Following immunisation with A42 (Fig. 14b), the

97

plaques of A have been removed and their surrounding dystrophic neurites have disappeared, but the intraneuronal neuroWbrillary tangles and the neuropil threads remain. Figure 15 summarises the eVects of immunisation with A42. Plaques of A42 are removed by activated microglia and A42 accumulates in the perivascular drainage pathways resulting in a signiWcant increase in severity of CAA. In investigating the potential mechanisms of removal of plaque A by immunotherapy, we hypothesised that plaque A might be solubilised by binding of anti-A antibodies and subsequently drain to the perivascular pathways, thus increasing the severity of CAA [15, 16, 58]. Of relevance to this hypothesis is the observation that A42 is predominantly localised in plaques whereas A40 is predominantly localised in the vessel walls as CAA (Fig. 11). We therefore proposed that immunisation might result in an increase in the amount of plaque-derived A42 in the blood vessel walls which would be detectable by immunohistochemistry using an A42-speciWc antibody [16]. Our observations conWrmed that there is a substantial increase in the quantity of A42, accompanied by A40, in the cerebral vasculature in Alzheimer’s disease patients immunised with A compared with unimmunised Alzheimer’s disease controls [16]. This increase in CAA aVects arteries and arterioles in the cerebral cortex and overlying leptomeninges. In some cases, prominent capillary angiopathy was observed. A range of post-immunization time-points were available for study and these were consistent with a dynamic sequence of events involving solubilisation of plaque A followed by A accumulation in the vessel walls. A striking appearance was observed in sections immunostained using A42-speciWc antibodies, in which plaques were absent but there was full thickness and full circumference A accumulation within the blood vessel walls [16], i.e. the converse of the pattern usually seen in Alzheimer’s disease with such antibodies. The increase in the severity of CAA [16] and of soluble A in the brain parenchyma [62] raises the possibility that the eVect of immunization is merely redistributing A within the brain rather than removing A from the brain. However, in two cases with prolonged survival of about 5 years after immunisation, there was extensive clearance of A plaques and very little CAA, suggesting that, given suYcient time, the plaque-derived A can be cleared from the perivascular pathways [16]. Although the evidence has yet to emerge, age and APOE genotype related alterations in the blood vessels walls, are likely to inXuence the process of vascular and perivascular clearance of A after immunization. Studies of A immunotherapy in APP transgenic mice with carefully selected time-points have conWrmed that CAA severity increases as plaques are removed from the parenchyma [94]. Passive immunisation using anti-A

123

98

Fig. 14 Changes that occur in the brain in Alzheimer’s disease following immunisation with A42. a Non-immunised patient with Alzheimer’s disease showing rounded plaques of A and associated dystrophic neurites. Neurons contain neuroWbrillary tangles and there

Acta Neuropathol (2009) 118:87–102

are neuropil threads in the brain parenchyma. b Following immunisation with A42, the plaques of A and the dystrophic neurites have disappeared but the neuroWbrillary tangles and neuropil threads remain. Bielschowsky stain, bars 50 m

Fig. 15 Summary diagram of the results of immunisation with A42 in Alzheimer’s disease. On the left is the pathological picture in a non-immunised patient with Alzheimer’s disease with plaques of A42 (green) surrounded by dystrophic neurites and microglia. Following immunisation with A42 (right hand panel) the plaques of A have been removed by activated microglia, the dystrophic neurites have disappeared, but there is a signiWcant rise in the amount of A42 (green) in the walls of capillaries and arteries as the severity of CAA increases (diagram modiWed from reference [15])

antibodies that speciWcally bind diVerent epitopes on the A peptide have diVerent eVects on CAA. Antibodies that bind the N-terminal portion on the A peptide cause a reduction in CAA severity [74]. It is unclear whether this is a direct eVect on the A already in the vessel walls or whether it is an eVect on the time-course of plaque-derived A tracking through the perivascular pathway. The APOE gene polymorphism, which is the basis of the major genetic risk factor for sporadic Alzheimer’s disease, seems likely to have an inXuence on the success with which A can be removed from the brain. Although the precise mechanism by which APOE genotype inXuences the risk of Alzheimer’s disease remains unclear, there is evidence that APOE chaperones A and inXuences its accumulation and

123

removal. It is already known that the relative distribution of A between the parenchymal plaques and the cerebral vasculature is inXuenced by APOE genotype; patients with APOE 4 have more severe CAA. Putatively, this could be because of the known association of APOE 4 with atherosclerosis and arteriosclerosis, resulting in more marked agerelated impairment of perivascular drainage. Currently there are few data on the role of APOE and the relevance of the APOE gene polymorphism in post-immunotherapy A mobilisation. However, the observation that APOE colocalises with the marked vascular accumulation of A following immunotherapy [56], is consistent with the mobilised A being transported or chaperoned by apoE, as it is in Alzheimer’s disease.

Acta Neuropathol (2009) 118:87–102

99

cortex, the function of which is impaired by CAA, and changes in the white matter. It is, therefore, possible to hypothesise a sequence of events that occurs after immunotherapy as follows: (1) Anti-A antibodies enter the brain and bind to plaques resulting, in the solubilisation of A. (2) The solubilised A diVuses through the brain parenchyma and enters the perivascular drainage pathways, manifesting as increased CAA. (3) AVected blood vessels are dysfunctional and, whether through a mechanism of ischaemia or impaired drainage of extracellular Xuid, result in alteration of the white matter. According to this hypothesis, the patches of white matter abnormality underlie regions of cortex in which there has been recent or rapid solubilisation of plaques. The hypothesis is now testable because of the recent ability to image amyloid plaques in vivo using PIB PET imaging [39] and it is likely that the relevant data with which to test the hypothesis will be available soon.

Conclusions

Fig. 16 T2 weighted MRI of a patient with Alzheimer’s disease following immunisation with A42 showing high signal in the white matter of one temporal lobe. This may be due to accumulation of Xuid resulting from blockage of perivascular Xuid drainage by an increase in the severity of CAA (reproduced with permission from reference [57])

A further point hinting at the role of the vasculature and of APOE genotype in the process of post-immunisation plaque removal comes from consideration of the side eVects of immunotherapy that have been a cause for concern in the human clinical trials [59]. Brain imaging studies performed on patients in both the active and the more recent passive vaccine trials have shown focal changes in the cerebral white matter (Fig. 16). How can this be explained when the plaques that are being targeted by the immunotherapy are located in the cerebral cortex? Although the answer to this question is far from clear at this stage, one intriguing possibility of relevance to the processes under discussion in this review relates to the known association of severe CAA and cerebral white matter alterations [15, 68]. Much of the cerebral white matter receives its blood supply from vessels that penetrate the brain surface from the leptomeninges, supply and pass through the cortex into the underlying white matter [24]. According to the perivascular drainage hypothesis [17, 90, 92] extracellular Xuid in the white matter, can drain retrogradely along the walls of these blood vessels. This provides a potential link between blood vessels in the leptomeninges and

This review has exposed a complex association between microvascular changes and Alzheimer’s disease ranging from ischaemia to impaired elimination of A from the brain. Further understanding of such vascular factors will clarify their impact on therapies for Alzheimer’s disease. Acknowledgments We thank Dr. Anton Page of the Biomedical Imaging Unit Southampton University Hospitals for preparing the Wgures for this paper. This study was supported by the Medical Research Council and the Alzheimer Research Trust. Research Ethics Committee Approval reference 07/H0505/86.

References 1. Abbott NJ (2004) Evidence for bulk Xow of brain interstitial Xuid: signiWcance for physiology and pathology. Neurochem Int 45:545–552. doi:10.1016/j.neuint.2003.11.006 2. Alcolado R, Weller RO, Parrish EP, Garrod D (1988) The cranial arachnoid and pia mater in man: anatomical and ultrastructural observations. Neuropathol Appl Neurobiol 14:1–17. doi:10.1111/ j.1365-2990.1988.tb00862.x 3. Attems J, Jellinger K (2004) Only cerebral capillary amyloid angiopathy correlates with Alzheimer pathology—a pilot study. Acta Neuropathol 107:83–90. doi:10.1007/s00401-003-0796-9 4. Attems J, Lintner F, Jellinger KA (2004) Amyloid beta peptide 142 highly correlates with capillary cerebral amyloid angiopathy and Alzheimer disease pathology. Acta Neuropathol 107:283– 291. doi:10.1007/s00401-004-0822-6 5. Attems J, Quass M, Jellinger KA, Lintner F (2007) Topographical distribution of cerebral amyloid angiopathy and its eVect on cognitive decline are inXuenced by Alzheimer disease pathology. J Neurol Sci 257:49–55. doi:10.1016/j.jns.2007.01.013 6. Ballard CG, Chalmers KA, Todd C, McKeith IG, O’Brien JT, Wilcock G, Love S, Perry EK (2007) Cholinesterase inhibitors reduce cortical Abeta in dementia with Lewy bodies. Neurology 68:1726– 1729. doi:10.1212/01.wnl.0000261920.03297.64

123

100 7. Beach TG (2008) Physiologic origins of age-related beta-amyloid deposition. Neurodegener Dis 5:143–145. doi:10.1159/00011 3685 8. Beach TG, Kuo YM, Spiegel K, Emmerling MR, Sue LI, Kokjohn K, Roher AE (2000) The cholinergic deWcit coincides with Abeta deposition at the earliest histopathologic stages of Alzheimer disease. J Neuropathol Exp Neurol 59:308–313 9. Beach TG, Potter PE, Kuo YM, Emmerling MR, Durham RA, Webster SD, Walker DG, Sue LI, Scott S, Layne KJ, Roher AE (2000) Cholinergic deaVerentation of the rabbit cortex: a new animal model of Abeta deposition. Neurosci Lett 283:9–12. doi:10.1016/S0304-3940(00)00916-2 10. Bechmann I, Galea I, Perry VH (2007) What is the blood–brain barrier (not)? Trends Immunol 28:5–11. doi:10.1016/j.it. 2006.11.007 11. Bedford L, Hay D, Paine S, Rezvani N, Mee M, Lowe J, Mayer RJ (2008) Is malfunction of the ubiquitin proteasome system the primary cause of alpha-synucleinopathies and other chronic human neurodegenerative disease? Biochim Biophys Acta 1782:683–690 12. Bell RD, Sagare AP, Friedman AE, Bedi GS, Holtzman DM, Deane R, Zlokovic BV (2007) Transport pathways for clearance of human Alzheimer’s amyloid beta-peptide and apolipoproteins E and J in the mouse central nervous system. J Cereb Blood Flow Metab 27:909–918 13. Bergsneider M (2001) Evolving concepts of cerebrospinal Xuid. Neurosurg Clin N Am 36:631–638 14. Black S, Gao F, Bilbao J (2008) Understanding white matter disease. Imaging–pathological correlations in vascular cognitive impairment. Stroke [Epub ahead of print] 15. Boche D, Nicoll JA (2008) The role of the immune system in clearance of Abeta from the brain. Brain Pathol 18:267–278. doi:10.1111/j.1750-3639.2008.00134.x 16. Boche D, Zotova E, Weller RO, Love S, Neal JW, Pickering RM, Wilkinson D, Holmes C, Nicoll JA (2008) Consequence of Abeta immunization on the vasculature of human Alzheimer’s disease brain. Brain 131:3299–3310. doi:10.1093/brain/awn261 17. Carare RO, Bernardes-Silva M, Newman TA, Page AM, Nicoll JAR, Perry VH, Weller RO (2008) Solutes, but not cells, drain from the brain parenchyma along basement membranes of capillaries and arteries. SigniWcance for cerebral amyloid angiopathy and neuroimmunology. Neuropathol Appl Neurobiol 34:131–144. doi:10.1111/j.1365-2990.2007.00926.x 18. NGoM CFAS (2001) Pathological correlates of late-onset dementia in a multicentre, community-based population in England and Wales. Lancet 357:169–175. doi:10.1016/S0140-6736(00)03589-3 19. Chalmers K, Wilcock G, Love S (2005) Contributors to white matter damage in the frontal lobe in Alzheimer’s disease. Neuropathol Appl Neurobiol 31:623–631. doi:10.1111/j.1365-2990.2005. 00678.x 20. Cirrito JR, Deane R, Fagan AM, Spinner ML, Parsadanian M, Finn MB, Jiang H, Prior JL, Sagare A, Bales KR, Paul SM, Zlokovic BV, Piwnica-Worms D, Holtzman DM (2005) P-glycoprotein deWciency at the blood–brain barrier increases amyloid-beta deposition in an Alzheimer disease mouse model. J Clin Invest 115:3285–3290. doi:10.1172/JCI25247 21. Cserr HF, Knopf PM (1992) Cervical lymphatics, the blood–brain barrier and the immunoreactivity of the brain: a new view. Immunol Today 13:507–512. doi:10.1016/0167-5699(92)90027-5 22. Deane R, Zlokovic BV (2007) Role of the blood–brain barrier in the pathogenesis of Alzheimer’s disease. Curr Alzheimer Res 4:191–197. doi:10.2174/156720507780362245 23. Domnitz SB, Robbins EM, Hoang AW, Garcia-Alloza M, Hyman BT, Rebeck GW, Greenberg SM, Bacskai BJ, Frosch MP (2005) Progression of cerebral amyloid angiopathy in transgenic mouse models of Alzheimer disease. J Neuropathol Exp Neurol 64:588– 594

123

Acta Neuropathol (2009) 118:87–102 24. Duvernoy HM, Delon S, Vannson JL (1981) Cortical blood vessels of the human brain. Brain Res Bull 7:519–579. doi:10.1016/ 0361-9230(81)90007-1 25. Engelhardt B (2008) Immune cell entry into the central nervous system: involvement of adhesion molecules and chemokines. J Neurol Sci 274:23–26. doi:10.1016/j.jns.2008.05.019 26. Esiri MM, Nagy Z, Smith MZ, Barnetson L, Smith AD (1999) Cerebrovascular disease and threshold for dementia in the early stages of Alzheimer’s disease. Lancet 354:919–920. doi:10.1016/ S0140-6736(99)02355-7 27. Farkas E, De Jong GI, de Vos RA, Jansen Steur EN, Luiten PG (2000) Pathological features of cerebral cortical capillaries are doubled in Alzheimer’s disease and Parkinson’s disease. Acta Neuropathol 100:395–402. doi:10.1007/s004010000195 28. Farkas E, de Vos RA, Donka G, Jansen Steur EN, Mihály A, Luiten PG (2006) Age-related microvascular degeneration in the human cerebral periventricular white matter. Acta Neuropathol 111:150–157. doi:10.1007/s00401-005-0007-y 29. Fernando MS, Simpson JE, Matthews F, Brayne C, Lewis CE, Barber R, Kalaria RN, Forster G, Esteves F, Wharton SB, Shaw PJ, O’Brien JT, Ince PG (2006) White matter lesions in an unselected cohort of the elderly: molecular pathology suggests origin from chronic hypoperfusion injury. Stroke 37:1391–1398. doi:10.1161/01.STR.0000221308.94473.14 30. Ferrer I, Boada Rovira M, Sánchez Guerra ML, Rey MJ, CostaJussá F (2004) Neuropathology and pathogenesis of encephalitis following amyloid-beta immunization in Alzheimer’s disease. Brain Pathol 14:11–20 31. Ferrer I, Kaste M, Kalimo H Vascular diseases. In: Love S, Louis DN, Ellison DW, editors. GreenWeld’s Neuropathology. Eighth ed. London: Hodder Arnold; 2008. p. 121–240. 32. Fisher A (2008) Cholinergic treatments with emphasis on m1 muscarinic agonists as potential disease-modifying agents for Alzheimer’s disease. Neurotherapeutics 5:433–442. doi:10.1016/ j.nurt.2008.05.002 33. Glenner GG, Wong CW, Quaranta V, Eanes ED (1984) The amyloid deposits in Alzheimer’s disease: their nature and pathogenesis. Appl Pathol 2:357–369 34. Goldmann J, Kwidzinski E, Brandt C, Mahlo J, Richter D, Bechmann I (2006) T cells traYc from brain to cervical lymph nodes via the cribroid plate and the nasal mucosa. J Leukoc Biol 80:797– 801. doi:10.1189/jlb.0306176 35. Hatterer E, Davoust N, Didier-Bazes M, Vuaillat C, Malcus C, Belin MF, Nataf S (2006) How to drain without lymphatics? Dendritic cells migrate from the cerebrospinal Xuid to the B-cell follicles of cervical lymph nodes. Blood 107:806–812. doi:10.1182/blood-2005-01-0154 36. Herzig MC, Van Nostrand WE, Jucker M (2006) Mechanism of cerebral beta-amyloid angiopathy: murine and cellular models. Brain Pathol 16:40–54. doi:10.1111/j.1750-3639.2006. tb00560.x 37. Holmes C, Boche D, Wilkinson D, Yadegarfar G, Hopkins V, Bayer A, Jones RW, Bullock R, Love S, Neal JW, Zotova E, Nicoll JA (2008) Long-term eVects of Abeta42 immunisation in Alzheimer’s disease: follow-up of a randomised, placebo-controlled phase I trial. Lancet 372:216–223. doi:10.1016/S0140-6736(08) 61075-2 38. Hutchings M, Weller RO (1986) Anatomical relationships of the pia mater to cerebral blood vessels in man. J Neurosurg 65:316– 325 39. Ikonomovic MD, Klunk WE, Abrahamson EE, Mathis CA, Price JC, Tsopelas ND, Lopresti BJ, Ziolko S, Bi W, Paljug WR, Debnath ML, Hope CE, Isanski BA, Hamilton RL, DeKosky ST (2008) Post-mortem correlates of in vivo PiB-PET amyloid imaging in a typical case of Alzheimer’s disease. Brain 131:1630–1645. doi:10.1093/brain/awn016

Acta Neuropathol (2009) 118:87–102 40. Jellinger KA (2008) The pathology of “vascular dementia”: a critical update. J Alzheimers Dis 14:107–123 41. Jellinger KA, Attems J (2007) Neuropathological evaluation of mixed dementia. J Neurol Sci 257:80–87. doi:10.1016/ j.jns.2007.01.045 42. Kuo YM, Crawford F, Mullan M, Kokjohn TA, Emmerling MR, Weller RO, Roher AE (2000) Elevated A beta and apolipoprotein E in A betaPP transgenic mice and its relationship to amyloid accumulation in Alzheimer’s disease. Mol Med 6:430–439 43. Kuo YM, Emmerling MR, Vigo Pelfrey C, Kasunic TC, Kirkpatrick JB, Murdoch GH, Ball MJ, Roher AE (1996) Water-soluble Abeta (N-40, N-42) oligomers in normal and Alzheimer disease brains. J Biol Chem 271:4077–4081. doi:10.1074/jbc.271.8.4077 44. Lahiri DK, Rogers JT, Greig NH, Sambamurti K (2004) Rationale for the development of cholinesterase inhibitors as anti-Alzheimer agents. Curr Pharm Des 10:3111–3119. doi:10.2174/13816120 43383331 45. Lashley T, Revesz T, Plant G, Bandopadhyay R, Lees AJ, Frangione B, Wood NW, de Silva R, Ghiso J, Rostagno A, Holton JL (2008) Expression of BRI2 mRNA and protein in normal human brain and familial British dementia: its relevance to the pathogenesis of disease. Neuropathol Appl Neurobiol 34:492–505. doi:10.1111/j.1365-2990.2008.00935.x 46. LayWeld R, Lowe J, Bedford L (2005) The ubiquitin–proteasome system and neurodegenerative disorders. Essays Biochem 41:157– 171. doi:10.1042/EB0410157 47. Lossinsky AS, Shivers RR (2004) Structural pathways for macromolecular and cellular transport across the blood–brain barrier during inXammatory conditions. Histol Histopathol 19:535–564 48. Lowe J, Mirra SS, Hyman BT, Dickson DW et al (2008) Ageing and dementia. In: Love S, Louis DN, Ellison DW (eds) GreenWeld’s neuropathology, 8th edn. Hodder Arnold, London, pp 1031–1152 49. Lue LF, Kuo YM, Roher AE, Brachova L, Shen Y, Sue L, Beach T, Kurth JH, Rydel RE, Rogers J (1999) Soluble amyloid beta peptide concentration as a predictor of synaptic change in Alzheimer’s disease. Am J Pathol 155:853–862 50. Masliah E, Hansen L, Adame A, Crews L, Bard F, Lee C, Seubert P, Games D, Kirby L, Schenk D (2005) Abeta vaccination eVects on plaque pathology in the absence of encephalitis in Alzheimer disease. Neurology 64:129–131 51. McLean CA, Cherny RA, Fraser FW, Fuller SJ, Smith MJ, Beyreuther K, Bush AI, Masters CL (1999) Soluble pool of Abeta amyloid as a determinant of severity of neurodegeneration in Alzheimer’s disease. Ann Neurol 46:860–866. doi:10.1002/15318249(199912)46:63.0.CO;2-M 52. Miners JS, Baig S, Palmer J, Palmer LE, Kehoe PG, Love S (2008) Abeta-degrading enzymes in Alzheimer’s disease. Brain Pathol 18:240–252. doi:10.1111/j.1750-3639.2008.00132.x 53. Miners JS, Van Helmond Z, Chalmers K, Wilcock G, Love S, Kehoe PG (2006) Decreased expression and activity of neprilysin in Alzheimer disease are associated with cerebral amyloid angiopathy. J Neuropathol Exp Neurol 65:1012–1021 54. Moody DM, Brown WR, Challa VR, Ghazi-Birry HS, Reboussin DM (1997) Cerebral microvascular alterations in aging, leukoaraiosis, and Alzheimer’s disease. Ann N Y Acad Sci 826:103–116. doi:10.1111/j.1749-6632.1997.tb48464.x 55. Nagasawa S, Handa H, Okumura A, Naruo Y, Moritake K, Hayashi K (1979) Mechanical properties of human cerebral arteries. Part 1: EVects of age and vascular smooth muscle activation. Surg Neurol 12:297–304 56. Nicoll JA, Barton E, Boche D, Neal JW, Ferrer I, Thompson P, Vlachouli C, Wilkinson D, Bayer A, Games D, Seubert P, Schenk D, Holmes C (2006) Abeta species removal after Abeta42 immunization. J Neuropathol Exp Neurol 65:1040–1048. doi:10.1097/ 01.jnen.0000240466.10758.ce

101 57. Nicoll JA, Wilkinson D, Holmes C, Steart P, Markham H, Weller RO (2003) Neuropathology of human Alzheimer disease after immunization with amyloid-beta peptide: a case report. Nat Med 9:448–452. doi:10.1038/nm840 58. Nicoll JAR, Yamada M, Frackowiak J, Mazur Kolecka B, Weller RO (2004) Cerebral amyloid angiopathy plays a direct role in the pathogenesis of Alzheimer’s disease. Pro-CAA position statement. Neurobiol Aging 25:589–597. doi:10.1016/j.neurobiolaging.2004.02.003 Discussion 603-604 59. Orgogozo JM, Gilman S, Dartigues JF, Laurent B, Puel M, Kirby LC, Jouanny P, Dubois B, Eisner L, Flitman S, Michel BF, Boada M, Frank A, Hock C (2003) Subacute meningoencephalitis in a subset of patients with AD after Abeta42 immunization. Neurology 61:46–54 60. Oshima K, Akiyama H, Tsuchiya K, Kondo H, Haga C, Shimomura Y, Iseki E, Uchikado H, Kato M, Niizato K, Arai H (2006) Relative paucity of tau accumulation in the small areas with abundant Abeta42-positive capillary amyloid angiopathy within a given cortical region in the brain of patients with Alzheimer pathology. Acta Neuropathol 111:510–518. doi:10.1007/s00401-006-0070-z 61. Owens T, I. B, Engelhardt B (2008) Perivascular spaces and the two steps to neuroinXammation. J Neuropathol Exp Neurol 67:1113–1121. doi:10.1097/NEN.0b013e31818f9ca8 62. Patton RL, Kalback WM, Esh CL, Kokjohn TA, Van Vickle GD, Luehrs DC, Kuo YM, Lopez J, Brune D, Ferrer I, Masliah E, Newel AJ, Beach TG, Castano EM, Roher AE (2006) Amyloidbeta peptide remnants in AN-1792-immunized Alzheimer’s disease patients: a biochemical analysis. Am J Pathol 169:1048– 1063. doi:10.2353/ajpath.2006.060269 63. Perry EK, Kilford L, Lees AJ, Burn DJ, Perry RH (2003) Increased Alzheimer pathology in Parkinson’s disease related to antimuscarinic drugs. Ann Neurol 54:235–238. doi:10.1002/ana.10639 64. Poirier J, Derouesne C (1985) Le concept de lacune cerebrale de 1838 a nos jours. Rev Neurol 141:3–17 65. Pollock H, Hutchings M, Weller RO, Zhang ET (1997) Perivascular spaces in the basal ganglia of the human brain: their relationship to lacunes. J Anat 191:337–346. doi:10.1046/j.1469-7580. 1997.19130337.x 66. Preston SD, Steart PV, Wilkinson A, Nicoll JAR, Weller RO (2003) Capillary and arterial amyloid angiopathy in Alzheimer’s disease: deWning the perivascular route for the elimination of amyloid beta from the human brain. Neuropathol Appl Neurobiol 29:106–117. doi:10.1046/j.1365-2990.2003.00424.x 67. Revesz T, Ghiso J, Lashley T, Plant G, Rostagno A, Frangione B, Holton JL (2003) Cerebral amyloid angiopathies: a pathologic, biochemical, and genetic view. J Neuropathol Exp Neurol 62:885– 898 68. Roher AE, Kuo Y-M, Esh C, Knebel C, Weiss N, Kalback W, Luehrs DC, Childress JL, Beach TG, Weller RO, Kokjohn TA (2003) Cortical and leptomeningeal cerebrovascular amyloid and white matter pathology in Alzheimer’s disease. Mol Med 9:112–122 69. Roher AE, Lowenson JD, Clarke S, Wolkow C, Wang R, Cotter RJ, Reardon IM, Zurcher Neely HA, Heinrikson RL, Ball MJ et al (1993) Structural alterations in the peptide backbone of beta-amyloid core protein may account for its deposition and stability in Alzheimer’s disease. J Biol Chem 268:3072–3083 70. Salzman KL, Osborn AG, House P, Jinkins JR, DitchWeld A, Cooper JA, Weller RO (2005) Giant tumefactive perivascular spaces. AJNR Am J Neuroradiol 26:298–305 71. Schenk D, Barbour R, Dunn W, Gordon G, Grajeda H, Guido T, Hu K, Huang J, Johnson Wood K, Khan K, Kholodenko D, Lee M, Liao Z, Lieberburg I, Motter R, Mutter L, Soriano F, Shopp G, Vasquez N, Vandevert C, Walker S, Wogulis M, Yednock T, Games D, Seubert P (1999) Immunization with amyloid-beta attenuates Alzheimer-disease-like pathology in the PDAPP mouse. Nature 400:173–177. doi:10.1038/22124

123

102 72. Schley D, Carare-Nnadi R, Please CP, Perry VH, Weller RO (2006) Mechanisms to explain the reverse perivascular transport of solutes out of the brain. J Theory Biol 238:962–974. doi:10.1016/j.jtbi.2005.07.005 73. Scholz W (1938) Studien zur Pathologie der Hirngefässe II. Die drusige Entartung der Hirnarterien und -capillaren (Eine Form seniler Gefässerkrankung). Zeitschrift für die gesamte. Neurol Psychiatr (Bucur) 162:694–715 74. Schroeter S, Khan K, Barbour R, Doan M, Chen M, Guido T, Gill D, Basi G, Schenk D, Seubert P, Games D (2008) Immunotherapy reduces vascular amyloid-beta in PDAPP mice. J Neurosci 28:6787–6793. doi:10.1523/JNEUROSCI.2377-07.2008 75. Scolding NJ, Joseph F, Kirby PA, Mazanti I, Gray F, Mikol J, Ellison D, Hilton DA, Williams TL, MacKenzie JM, Xuereb JH, Love S (2005) Abeta-related angiitis: primary angiitis of the central nervous system associated with cerebral amyloid angiopathy. Brain 128:500–515. doi:10.1093/brain/awh379 76. Selkoe DJ (2001) Alzheimer’s disease: genes, proteins, and therapy. Physiol Rev 81:741–766 77. Selkoe DJ (2006) Amyloid beta-peptide is produced by cultured cells during normal metabolism: a reprise. J Alzheimers Dis 9:163–168 78. Shibata M, Yamada S, Kumar SR, Calero M, Bading J, Frangione B, Holtzman DM, Miller CA, Strickland DK, Ghiso J, Zlokovic BV (2000) Clearance of Alzheimer’s amyloid-beta(1-40) peptide from brain by LDL receptor-related protein-1 at the blood-brain barrier. J Clin Invest 106:1489–1499. doi:10.1172/JCI10498 79. Shinkai Y, Yoshimura M, Ito Y, Odaka A, Suzuki N, Yanagisawa K, Ihara Y (1995) Amyloid beta -proteins 1-40 and 1-42(43) in the soluble fraction of extra- and intracranial blood vessels. Ann Neurol 38:421–428. doi:10.1002/ana.410380312 80. Szentistvanyi I, Patlak CS, Ellis RA, Cserr HF (1984) Drainage of interstitial Xuid from diVerent regions of rat brain. Am J Physiol 246:F835–F844 81. Thal DR, Ghebremedhin E, Orantes M, Wiestler OD (2003) Vascular pathology in Alzheimer disease: correlation of cerebral amyloid angiopathy and arteriosclerosis/lipohyalinosis with cognitive decline. J Neuropathol Exp Neurol 62:1287–1301 82. Thal DR, Ghebremedhin E, Rub U, Yamaguchi H, Tredici KD, Braak H (2002) Two types of sporadic cerebral amyloid angiopathy. J Neuropathol Exp Neurol 61:282–293 83. Tian J, Shi J, Bailey K, Mann DM (2003) Negative association between amyloid plaques and cerebral amyloid angiopathy in Alzheimer’s disease. Neurosci Lett 352:137–140. doi:10.1016/ j.neulet.2003.08.048 84. Tian J, Shi J, Bailey K, Mann DM (2004) Relationships between arteriosclerosis, cerebral amyloid angiopathy and myelin loss from cerebral cortical white matter in Alzheimer’s disease. Neuropathol Appl Neurobiol 30:46–56. doi:10.1046/j.0305-1846.2003. 00510.x 85. Tsubuki S, Takaki Y, Saido TC (2003) Dutch, Flemish, Italian, and Arctic mutations of APP and resistance of Abeta to physiolog-

123

Acta Neuropathol (2009) 118:87–102

86.

87.

88.

89.

90.

91.

92.

93.

94.

95.

96.

97.

98.

ically relevant proteolytic degradation. Lancet 361:1957–1958. doi:10.1016/S0140-6736(03)13555-6 Vinters HV, Wang ZZ, Secor DL (1996) Brain parenchymal and microvascular amyloid in Alzheimer’s disease. Brain Pathol 6:179–195. doi:10.1111/j.1750-3639.1996.tb00799.x Walsh DM, Selkoe DJ (2007) Abeta oligomers—a decade of discovery. J Neurochem 101:1172–1184. doi:10.1111/j.1471-4159. 2006.04426.x Weller RO (1998) Pathology of cerebrospinal Xuid and interstitial Xuid of the CNS: signiWcance for Alzheimer disease, prion disorders and multiple sclerosis. J Neuropathol Exp Neurol 57:885– 894. doi:10.1097/00005072-199810000-00001 Weller RO (2005) Microscopic morphology and histology of the human meninges. Morphologie 89:22–34. doi:10.1016/S12860115(05)83235-7 Weller RO, Djuanda E, Yow HY, Carare RO (2009) Lymphatic drainage of the brain and the pathophysiology of neurological disease. Acta Neuropathol 117:1–14. doi:10.1007/s00401-008-0457-0 Weller RO, Massey A, Newman TA, Hutchings M, Kuo YM, Roher AE (1998) Cerebral amyloid angiopathy: amyloid beta accumulates in putative interstitial Xuid drainage pathways in Alzheimer’s disease. Am J Pathol 153:725–733 Weller RO, Subash M, Preston SD, Mazanti I, Carare RO (2008) Perivascular drainage of amyloid-beta peptides from the brain and its failure in cerebral amyloid angiopathy and Alzheimer’s disease. Brain Pathol 18:253–266. doi:10.1111/j.1750-3639.2008.00133.x Weller RO, Yow HY, Preston SD, Mazanti I, Nicoll JAR (2002) Cerebrovascular disease is a major factor in the failure of elimination of Abeta from the aging human brain: implications for therapy of Alzheimer’s disease. Ann N Y Acad Sci 977:162–168 Wilcock DM, Rojiani A, Rosenthal A, Subbarao S, Freeman MJ, Gordon MN, Morgan D (2004) Passive immunotherapy against Abeta in aged APP-transgenic mice reverses cognitive deWcits and depletes parenchymal amyloid deposits in spite of increased vascular amyloid and microhemorrhage. J NeuroinXamm 1:24. doi:10.1186/1742-2094-1-24 Wisniewski HM, Wegiel J (1994) Beta-amyloid formation by myocytes of leptomeningeal vessels. Acta Neuropathol 87:233– 241. doi:10.1007/BF00296738 Yow HY, Weller RO (2002) A role for cerebrovascular disease in determining the pattern of beta amyloid deposition in Alzheimer’s disease. Neuropathol Appl Neurobiol 28:149. doi:10.1046/j.13652990.2002.39286_4.x Zhang-Nunes SX, Maat-Schieman ML, van Duinen SG, Roos RA, Frosch MP, Greenberg SM (2006) The cerebral beta-amyloid angiopathies: hereditary and sporadic. Brain Pathol 16:30–39. doi:10.1111/j.1750-3639.2006.tb00559.x Zhang ET, Inman CB, Weller RO (1990) Interrelationships of the pia mater and the perivascular (Virchow–Robin) spaces in the human cerebrum. J Anat 170:111–123