Cell. Mol. Life Sci. DOI 10.1007/s00018-013-1542-7
Cellular and Molecular Life Sciences
Visions and reflections
Alzheimer’s disease and CADASIL are heritable, adult‑onset dementias that both involve damaged small blood vessels Vincent T. Marchesi
Received: 2 October 2013 / Revised: 26 November 2013 / Accepted: 12 December 2013 © Springer Basel 2013
Abstract This essay explores an alternative pathway to Alzheimer’s dementia that focuses on damage to small blood vessels rather than late-stage toxic amyloid deposits as the primary pathogenic mechanism that leads to irreversible dementia. While the end-stage pathology of AD is well known, the pathogenic processes that lead to disease are often assumed to be due to toxic amyloid peptides that act on neurons, leading to neuronal dysfunction and eventually neuronal cell death. Speculations as to what initiates the pathogenic cascade have included toxic abeta peptide aggregates, oxidative damage, and inflammation, but none explain why neurons die. Recent high-resolution NMR studies of living patients show that lesions in white matter regions of the brain precede the appearance of amyloid deposits and are correlated with damaged small blood vessels. To appreciate the pathogenic potential of damaged small blood vessels in the brain, it is useful to consider the clinical course and the pathogenesis of CADASIL, a heritable arteriopathy that leads to damaged small blood vessels and irreversible dementia. CADASIL is strikingly similar to early onset AD in that it is caused by germ line mutations in NOTCH 3 that generate toxic protein aggregates similar to those attributed to mutant forms of the amyloid precursor protein and presenilin genes. Since NOTCH 3 mutants clearly damage small blood vessels of white matter regions of the brain that lead to dementia, we speculate that both forms of dementia may have a similar pathogenesis, which is to cause ischemic damage by blocking blood flow or by
V. T. Marchesi (*) Department of Pathology, Yale University, New Haven, CT 06536‑0812, USA e-mail: [email protected]
impeding the removal of toxic protein aggregates by retrograde vascular clearance mechanisms. Keywords Blood vessel damage · Alzheimer’s dementia · CADASIL · NOTCH 3
Alzheimer’s dementia is the result of the uncoupling of the neuro‑transmission machinery from the neuro‑vascular blood supply chain The amyloid hypothesis dominates the mainstream approach to Alzheimer’s therapy and focuses on attempts to either block or reduce amyloid abeta accumulations in the brains of patients in what we now know to be the late stages of Alzheimer’s dementia. This approach is inspired by the conviction that amyloid is the prime toxic element that leads to neuronal dysfunction and eventually neuronal death. Since the chemical nature of toxic amyloid and its biological targets still remain undefined, the treatments in question are in effect focusing on the pathology of the late stages of the disease without having a clear understanding of the pathogenic processes that initiate the beginning stages of the disease. This essay explores alternative pathways to Alzheimer’s dementia that focus on pathogenic processes that act in concert with amyloid, to disrupt the finely balanced interactions between neurons and their neuro-vascular blood supply. If small blood vessels rather than their cognate neurons are the first targets of a toxic process that compromises the microcirculation, the pathogenic processes that damage them represent an entirely new set of therapeutic targets that remain to be explored. While there is little doubt that amyloid, in some form, contributes to dementia, advocates of the amyloid hypothesis assume that abeta peptides are THE initiating
13
V. T. Marchesi
toxic factors that initiates the devastating cascade depicted below. Toxic factors > pathogenic processes > biologic targets > pathology However, amyloid abeta peptides are not the only agents that have the potential to impair neuronal activity. Other potential pathogens must also be considered, including reactive oxygen and nitrogen species (ROS, NOS), cytokines, chemokines, heavy metals, and mutant forms of APP, Tau and the presenilins. If abeta peptides are the prime suspects, it is critical to find out which pathogenic processes are caused by abeta peptides since these are likely to be more promising therapeutic targets than the peptides themselves. We have to ask the same question with regard to other potential neurotoxins. What types of oxidative damage result from reactive oxygens or nitrogens? Are inflammatory reactions provoked? Do damaged blood vessels lead to ischemia/hypoxia or localized vascular occlusions? Is apoptosis a consequence? To determine which pathologic processes are the most critical for brain dysfunction, it will be necessary to identify the relevant biological targets. Critical neurons in specific parts of the brain are obviously the final targets, but what happens at the earliest stages of the disease, that 10 to 15-year asymptomatic phase? During this pre-symptomatic period, small blood vessel damage has been detected by high-resolution MRI, and reduced neurotransmission, documented by reduced connectome activity, both leading to axonal damage. Activated microglia reflect an ongoing inflammatory process. None of these potential targets would be protected or reversed by therapies that focus on end-stage disease. By focusing on the most likely biological targets in the brain we have a better chance to link specific toxic factors to pathogenic processes, which in the end are most likely to be the therapeutic targets of choice. The toxic factor problem Amyloid abeta peptides are clearly a feature of advanced AD dementia, but two questions limit their potential as effective therapeutic targets. The first, unanswerable at present, is whether they are the toxic element that initiates the disease process from the very beginning. Recent PET scan data indicate that PIB stainable material is present in the brains of asymptomatic people. This indicates that detectable amyloid can accumulate in the human brain without obvious damage. Since it accumulates as people age and higher levels are often (but not always) correlated with dementia, many consider the association correlative, but not necessarily causative. The second concern is the molecular form of the toxic component. Abeta peptides, whether
13
derived from human brains or synthetically created, are technically demanding to study because of their perplexing tendency to assemble into many physical forms, which range from small 4-K monomers to aggregates that exceed 100-K Daltons. Despite this uncertainty, there is a quasiconsensus in the field that toxic “oligomers” are probably the most important pathogenic agents, although some experienced observers doubt their biological significance [1]. Compounding the problem is the lack of a consensus as to their biological targets. They have been accused of blocking synaptic activity, modifying neuronal dendrites, and reducing axonal flow in addition to damaging both large and small blood vessels. Which, if any, of these targets contributes to irreversible dementia remains to be determined. To summarize: fibrillar forms of abeta peptides make up the bulk of the late-stage amyloid plaques but their role as the prime instigators of disease remains an unresolved question. Antibody treatments that bind to abeta peptides may in some cases reduce the amount of amyloid in AD brains, but it remains to be seen whether such treatments impact the demented state.
The most relevant biologic targets Many consider the human brain the most complex organ in the body, which it undoubtedly is, but to simplify our search for answers to the dementia problem, we can more profitably focus on two functional units as the most likely pathogenic targets: the neuro-vascular blood supply and the neuronal-dominated neurotransmission machinery. Together, they represent complex but exceedingly specific interactions between blood vessel elements, various glial elements, and highly specialized neurons. The brain, like every other metabolically active organ, needs a reliable supply of oxygen and nutrients and a means of flushing out unwanted catabolites, all of which are provided by an elaborately designed neuro-vascular unit that has been well described by others [2]. Much attention has focused on the blood–brain barrier (BBB) as the unique feature of the brain vasculature that very effectively insulates the brain from the systemic circulation under normal conditions. However, when vessels are damaged by inflammatory reactions, thrombosis, or amyloid infiltrations, the tissue nourished by these vessels can undergo ischemic/hypoxic damage. Brain blood vessels of all sizes, including large leptomeningeal vessels, smaller arterioles, and the tiniest capillaries, are also sites of abeta amyloid accumulation. Why they accumulate at these sites and how they contribute to vascular injury and dysfunction is still not clear. Strickland and Grammas propose that they generate thrombotic occlusions [3, 4], but Iadacola and Zlokovic propose that they interfere with vasomotor control of local blood flow
Alzheimer’s disease and CADASIL
[5] or modify pericyte actions [2]. Whatever the mechanisms that damage them, small blood vessels could be the earliest target of the neurodegeneration process. The alternative primary target is of course the neurons themselves. There are many claims that abeta peptides of various compositions damage dendrites or interfere with synaptic activity. Most of these are in vitro studies or of model systems. Amyloid-related peptides could be involved in blocking neurotransmission, but it is not obvious whether and how they kill neurons. If abeta peptide are able to damage both blood vessels and the neurons they nourish, it is reasonable to assume that the damaged blood vessels lead to neuronal dysfunction rather than the reverse. This speculation places small blood vessels as the prime target of whatever toxic factors might be involved.
Pathogenic processes If neurons and their blood vessels are the obvious targets that lead, eventually, to dementia, we have to ask which pathogenic processes, which may act over a decades-long duration, are most likely to uncouple the neuro-transmission machinery from the neuro-vascular supply chain? It is essential to determine which are primary first causes and which are later consequences of a progressive disease. The thesis proposed here is that multiple factors which might include oxidative damage, a modified form of inflammation, and related circulatory failures act synergistically to uncouple the blood supply from selective neurons, resulting in localized oxygen and metabolite deprivation. This leads to impaired executive functions, the earliest sign of incipient dementia, which Bartzokis attributes to reduced cognitive processing speed as a consequence of myelin breakdown [6]. These changes are consistent with recent NMR findings that show white matter lesions and a disrupted connectome [7]. Progressive disease leads to reduced dendritic and synaptic function, amyloid dysregulation, and, finally, neuronal cell death, and irreversible dementia. This is the stage when amyloid plaques and neurofibrillary tangles dominate the cortical landscape. Attempting to remove them at this stage is not likely to reverse the course of the disease.
Oxidative damage and the failure of antioxidant therapy Vast literature documents the damage that ROS and NOS can inflict on the human species. Too voluminous to summarize here, a few facts emphasize why oxidative damage is considered by many to be one of the primary causes of a disease process that culminates in irreversible dementia.
Reactive oxygens can modify all classes of biological molecules, but much has been made of the damage to DNA, RNA, and related nucleotides that have been found in the brains of late-stage AD [8, 9]. Modified nucleic acids can potentially have a life of their own, unlike oxidized lipids or proteins, which can be replaced or eliminated rapidly. Guanine is considered the nucleotide base most sensitive to oxidation but it is not the only one that can be modified, and oxo-guanine in either DNA or RNA can miscode and create aberrant m-RNAs and eventually mutated proteins. Significant amounts of oxo-guanine and oxo-trinucleotides are found in AD brains, but as yet, no one has identified any mutant proteins that might be generated by modified m-RNAs, although there is now an ongoing search. It has been suggested that such oxidative modifications, which happen with alarming frequency in humans, may be more pathogenic in AD patients who lack effective DNA repair mechanisms [10] or who lack sufficient antioxidant capacity. This latter speculation has led to a massive search for agents that can scavenge unwanted radicals. The general public, less concerned with scientific rigor, has enthusiastically embraced the idea that anti-oxidants, of whatever form, are desirable, if not necessary, for a healthy functioning brain. Resveratrol, popularized by its presence in red wine, is one of many nutraceuticals that are consumed by the public with or without advice from their attending physicians. If we look at this trend critically, we have to admit that no one can claim, or disclaim, that these agents work as advertised, since there is no reliable way to measure their impact, either positively or negatively.
Oxidative damage is a fact of aerobic life, but oxidative regulation is far more biologically relevant A more in-depth look at the many roles that oxygen radicals play in daily life reveals why the search for effective antioxidant therapy has so far met with disappointing failure. We now realize that reactive oxygens such as superoxide and hydrogen peroxide are as important to post-translational modifications of critical proteins as are phosphorylation and other metabolic regulators. Cellular metabolism is under tight redox control through a complex network of sulfhydryl-regulated reactions involving glutathione and multiple peroxidases. The reducing environment of cells is regulated in part by the supply of NADPH that is generated by glycolysis and aerobic oxidative reactions, which together maintain an appropriate SH/SS ratio of critical cysteines of a vast number of proteins [11, 12]. These include numerous protein phosphatases that regulate protein phosphorylation and many other metabolic pathways. The dual action of these oxidizing agents complicates attempts to block their potentially toxic actions by
13
V. T. Marchesi
ingesting large doses of antioxidants, which, as we now know, probably never reach into the cellular compartments that generate the oxidants, since ingested antioxidants are not taken up by most cells and are rapidly excreted. One suspects that massive doses may be more likely to block normal physiologic functions than to deter unwanted damage, consistent with recent reports of toxic effects of excessive vitamin ingestion. The long-accepted notion that oxidative damage results from superoxide leaking out of mitochondria and converted via SOD to radials after exposure to heavy metals has now been replaced by a more nuanced redox concept that involves multiple NADPH oxidases that act locally and are under tight control by a glutathione/cysteine-based network. Given these many uncertainties, we have no way of evaluating the significance of failed antioxidant clinical trials.
Inflammation and the failure of anti‑inflammatory therapies More than a decade ago, inflammation was considered an important pathogenic mechanism that led to dementia, stimulated in part by anecdotal claims that individual taking NSAIDS for arthritis had a lower incidence of AD in later life. Two factors contributed to the downplaying of this idea. Enthusiasm was dampened when clinical trials of different NSAIDs failed to confirm the anecdotal claims, but its almost total rejection by opinion leaders was inspired by what seemed to be a much better idea: the amyloid hypothesis. In retrospect, failed clinical trials could have failed, and may well continue to fail, for many reasons. The simplest interpretation of failed anti-inflammation trials is that classical inflammation is either not the problem or only a contributing factor. Agents that block the acute stages in inflammation, such as blocking neutrophil accumulations, may be ineffective because that stage of the process is either truncated or bypassed by other events, such as reduced blood flow due to unregulated vasomotor activity or the lack of a classical inflammatory response in the damaged brain. One of the surprising features of late-stage Alzheimer’s dementia is the almost complete absence of connective tissue scarring in the brain despite massive neuronal cell death. In contrast to an infarct in the heart, where dead myocardial cells are replaced by angiogenesis and eventually fibrosis, dead neurons are replaced by diffuse gliosis, with very little if any neo-vascularization or scarring. Earlier studies of ischemic-induced necrosis, both experimental and naturally occurring human, reported that neutrophil accumulations were present in infarcted areas of the brain, but recent studies dispute this claim and instead suggest that the neurovascular sheath that surrounds brain
13
micro-vessels blocks neutrophil transit into the brain parenchyma [13]. Circulating blood monocytes do enter the brain in large numbers and become the population of blood-derived microglia. In contrast to the monocytes that infiltrate other ischemic organs and become tissue macrophages, these monocytes do not stimulate angiogenesis or fibrosis. It is claimed that some monocytes emerge from the blood marrow as “genetically naked” cells that are much more sensitive to genetic and epigenetic modifications than other circulating leukocytes [14]. They have a lesser capacity to repair DNA modifications that are caused by oxidative damage, and their presumed genetic flexibility allows them to home to the cerebral vasculature where they cross the neuro-vascular barrier and take up residence in the brain in the form of blood-born microglia. In this environment they “mature” into macrophages that do not stimulate extensive fibrosis. How this works is unknown. It makes sense that a connective tissue-dominated repair process would not be needed or wanted in the brain. Without regenerating neurons to replace neural functions, collagenous scar tissue, not usually a problem in other organs, could disrupt the functional linkages of the brain. A dramatic change in attitude of some AD investigators has been stimulated by recent GWAS studies, which implicate genes that may have pro-inflammatory properties. The idea that inflammation may be a significant factor in AD pathogenesis has been given a big boost by the recent discovery that mutant forms of the TREM2 gene, clearly a pro-inflammatory factor, have been found in a subset of AD patients [15]. The obvious targets of inflammatory reactions are small blood vessels, consistent with the idea that vascular injury sets the stage for neuronal dysfunction, as described in more detail below. It remains to be seen whether anti-inflammatory agents have a role in treating AD. The choice of agents, the specific inflammatory targets, and time and duration of treatment all have to be considered, but equally important, we have to find out what activates the damage of cerebral blood vessels in the first instance.
Ischemic brain damage leads to apoptosis If damaged small blood vessels represent the first step of a decades-long pathogenic process that leads ultimately to neuronal cell death and irreversible cognitive impairment, the pathogenesis of Alzheimer’s dementia shares many features with ischemic strokes that also plague the adult human population. Both result from occlusions of blood vessels that feed the brain leading to apoptosis, although each operates on a distinctly different temporal and anatomical scale. Strokes have a rapid onset caused by a
Alzheimer’s disease and CADASIL
massive loss of the blood supply to a sizable segment of the brain with debilitating consequences. Microvascular injury has a limited, localized impact and is almost always asymptomatic, but it is often distributed throughout the brain to varying degrees. Its impact can be felt when neurotransmission is affected long before apoptosis is a factor. The distinctive feature of microvascular injury in AD that leads to dementia is the presence of abeta amyloid deposits in the affected blood vessels which we assume affect the microvasculature long before they accumulate as dense plaques in the late-stage Alzheimer’s brain [16–18]. If this hypothesis is correct, the most effective therapeutic approach to AD will be to block the earliest stages of the disease, before apoptosis is reached, since blocking or reversing apoptosis following strokes is still a distant goal.
Lessons learned from CADASIL To explore how small blood vessel damage is initiated in the earliest stages of a disease it is instructive to consider the pathogenesis of CADASIL a well-recognized heritable disease of small blood vessels. CADASIL (cerebral autosomal dominant arteriopathy with sub-cortical infarct and leukoencephalopathy) is described as the most common heritable cause of stroke and vascular dementia in the adult human population. Several reviews document its clinical course and presumed pathogenesis [19–21]. Although it affects people of all ages, most are affected in the 40–60-age range and suffer from dementia that progresses to complete incapacitation by the end of the sixth decade. MRI studies detect leukoaraiosis before the onset of cognitive symptoms, and at autopsy destructive changes are found in the penetrating arteries that feed the white matter regions housing the infarcted sites. The damaged blood vessels contain what are considered pathognomonic deposits of granular osmiophilic material (GOM) that includes polypeptide segments of the extracellular domain of the trans-membrane NOTCH 3 protein and other matrix-related proteins [22]. The genetic basis of this condition is believed to be due to germline mutations in NOTCH 3, which so far include close to 200 unique nonsense mutations that modify cysteine residues in the EGF-rich segments of its extracellular domain. Genetic studies in mice and man support the idea that NOTCH 3 mutants in affected patients are gain–offunction mutations that generate toxic factors that damage smooth muscle cells of blood vessels rather than compromised NOTCH 3 functions, but it is also possible that such mutations might represent hypomorphic receptor activities [23]. It is notable that of the four known human NOTCH genes, only NOTCH 3 is expressed exclusively in vascular smooth muscle cells and brain pericytes, and
only NOTCH 3 knockouts in mice develop with minimal problems and do not develop CADASIL. It is worth pointing out how remarkably similar CADASIL and early onset Alzheimer’s dementia are to each other. • Both are caused by multiple, dominantly inherited mutations that generate potentially toxic protein aggregates • Both have a 10 to 15-year “asymptomatic” course. • Both involve damaged small blood vessels in white matter regions of the brain. • Both have protein deposits in damaged blood vessels. • Both protein deposits are generated in part by gamma secretase activities. • Both affected peoples develop symptoms in the 40– 60-age range and eventually suffer from debilitating dementia that progresses to complete incapacitation by the end of the sixth decade. While the similarities between early onset AD and CADASIL are striking, CADASIL damage is largely localized to sub-cortical white matter regions of the brain in its earliest stages while “classical” Alzheimer’s dementia is generally considered a cortical problem, dominated by massive deposits of amyloid abeta peptides. Numerous clinical trails designed to remove or reduce amyloid are based on the assumption that this is the primary pathogenic mechanism. However, many recent studies, described earlier, demonstrate that both early onset and sporadic AD dementia also involve damage to small blood vessels that feed the sub-cortical white matter regions. The small blood vessels in CADASIL brains contain mutant forms of NOTCH 3 that are expressed in vascular smooth muscle cells and generate disulphide-linked protein aggregates that are believed to be toxic to smooth muscle cells and possibly even brain capillary pericytes. Amyloid deposits are also found in small blood vessels in both forms of AD, but why they develop near vessels and how they are toxic are still unknown. It is not at all clear how mutant forms of APP or the presenilins contribute to blood vessel damage or how 95 % of AD patients without these germ line mutations develop disease. Both sets of mutations seem to influence the amount or the quality of beta peptides that are generated, but there is not the obvious connection to the vessel damage that mutant NOTCH 3 provides. Since aberrant over-production of abeta peptides in AD patients without obvious mutations is not a problem, it has been suggested that amyloid accumulations develop because of reduced turnover or defects in clearing mechanisms. Among the latter include the idea that abeta peptides, and possibly other protein aggregates, are eliminated from the brain by passing in a reverse-flow
13
V. T. Marchesi
direction along channels surrounding brain capillaries and small arterioles, an idea proposed many years ago [24] and recently updated to include the possibility that aggregates of mutant NOTCH 3 might behave in the same way [25].
Speculations If small blood vessel damage leading to localized ischemia is indeed the first step in a decades-long process that involves amyloid dysregulation and, eventually, neuronal cell death and irreversible dementia, we have to answer two critical questions: What initiates damage to small blood vessels in white matter regions of the brain? How does ischemia cause amyloid dysregulation? The second question is easier to address than the first. It is well recognized that localized ischemia can influence the metabolism of affected tissues as well as increasing amyloid abeta production by modifying the enzymes that selectively cleave the APP molecule [26]. Localized accumulations of abeta might be generated that overwhelm the drainage machinery that is also compromised by local blood flow and/or damaged blood vessels. Too little is known to be more specific. If CADASIL and AD dementia do indeed share a common pathogenic mechanism as so many shared properties imply, we have to consider the possibility that one or more NOTCH 3-like mutations, yet to be identified, damage small blood vessels in the earliest asymptomatic stages of AD that initiate the virulent cascade that ensues. Among the possibilities, I would include low levels of somatic mutations of APP, the presenilins, tau, and even NOTCH 3 itself. The large number of novel germ line mutations in APP, presenilin 1, and NOTCH 3 that appear to be pathogenic suggests that these genes might have mutational hotspots that might also be subject to additional somatic mutations during an individual’s lifetime.
References 1. Benilova I, Karran E, De Strooper B (2012) The toxic Aβ oligomer and Alzheimer’s disease: an emperor in need of clothes. Nat Neurosci 15(3):349–357 2. Sagare AP, Bell RD, Zlokovic BV (2012) Neurovascular dysfunction and faulty amyloid β-peptide clearance in Alzheimer disease. Cold Spring Harb Perspect Med 2(10). doi:10.1101/cshperspect.a011452 3. Cortes-Canteli M, Paul J, Norris EH, Bronstein R, Ahn HJ, Zamolodchikov D, Bhuvanendran S et al (2012) Fibrinogen and altered hemostasis in Alzheimer’s disease. J Alzheimers Dis 32(3):599–608 4. Tripathy D, Sanchez A, Yin X, Luo J, Martinez J, Grammas P (2013) Thrombin, a mediator of cerebrovascular inflammation in AD and hypoxia. Front Aging Neurosci 5:19
13
5. Park L, Zhou P, Koizumi K, El Jamal S, Previti ML, Van Nostrand WE, Carlson G, Iadecola C (2013) Brain and circulating levels of Aβ1-40 differentially contribute to vasomotor dysfunction in the mouse brain. Stroke 44(1):198–204 6. Lu PH, Lee GJ, Tishler TA, Meghpara M, Thompson PM, Bartzokis G (2013) Myelin breakdown mediates age-related slowing in cognitive processing speed in healthy elderly men. Brain Cogn 81(1):131–138 7. Moghekar A, Kraut M, Elkins W, Troncoso J, Zonderman AB, Resnick SM, O’Brien RJ (2012) Cerebral white matter disease is associated with Alzheimer pathology in a prospective cohort. Alzheimers Dement 8(5 Suppl):S71–S77 8. Castellani RJ, Nunomura A, Rolston RK, Moreira PI, Takeda A, Perry G, Smith MA (2008) Sublethal RNA oxidation as a mechanism for neurodegenerative disease. Int J Mol Sci 9(5):789–806 9. Ohno M, Miura T, Furuichi M, Tominaga Y, Tsuchimoto D, Sakumi K, Nakabeppu Y (2006) A genome-wide distribution of 8-oxoguanine correlates with the preferred regions for recombination and single nucleotide polymorphism in the human genome. Genome Res 16(5):567–575 10. Liu D, Croteau DL, Souza-Pinto N, Pitta M, Tian J, Wu C, Jiang H, Mustafa K, Keijzers G, Bohr WA, Mattson MP (2010) Evidence that OGG1 glycosylase protects neurons against oxidative DNA damage and cell death under ischemic conditions. J Cereb Blood Flow Metab 31:680–692 11. Thamsen M, Jakob U (2010) The redoxome proteomic analysis of cellular redox networks. Curr Opin Chem Biol 15(1):113–119 12. Cremers C, Jakob U (2013) Oxidant sensing by reversible disulfide bond formation. J Biol Chem 288:26463 13. Enzmann G, Mysiorek C, Gorina R, Cheng YJ, Ghavam pour S, Hannocks MJ, Prinz V, Dirnagl U, Endres M, Prinz M, Beschorner R, Harter PN, Mittelbronn M, Engelhardt B, Sorokin L (2013) The neurovascular unit as a selective barrier to polymorphonuclear granulocyte (PMN) infiltration into the brain after ischemic injury. Acta Neuropathol 125(3):395–412 14. Bauer M, Goldstein M, Christmann M, Becker H, Heylmann D, Kaina B (2011) Human monocytes are severely impaired in base and DNA double-strand break repair that renders them vulnerable to oxidative stress. PNAS 108(52):21105–21110 15. Guerreiro R, Wojtas A, Bras J, Carrasquillo M, Rogaeva E, Majounie E, Cruchaga C, Sassi C, Kauwe JS, Younkin S, Hazrati L, Collinge J, Pocock J, Lashley T, Williams J, Lambert JC, Amouyel P, Goate A, Rademakers R, Morgan K, Powell J, St George-Hyslop P, Singleton A, Hardy J, Alzheimer Genetic Analysis Group (2013) TREM variants in Alzheimer’s disease. N Engl J Med 368(2):117–127 16. Thal DR, Capetillo-Zarate E, Larionov S, Staufenbiel M, Zurbruegg S, Beckmann N (2009) Capillary cerebral amyloid angiopathy is associated with vessel occlusion and cerebral blood flow disturbances. Neurobiol Aging 30(12):1936–1948 17. Hunter JM, Kwan J, Malek-Ahmadi M, Maarouf CL, Kokjohn TA et al (2012) Morphological and pathological evolution of the brain microcirculation in aging and Alzheimer’s disease. PLoS One 7(5):36893. doi:10.1371/journal.pone.0036893 18. Austin BP, Nair VA, Meier TB, Xu G, Rowley HA, Carlsson CM, Johnson SC, Prabhakaran V (2011) Effects of hypoperfusion in Alzheimer’s disease. J Alzheimers Dis 26(Suppl 3):123–133 19. Charlton RA, Morris RG, Nitkunan A, Markus HS (2006) The cognitive profiles of CADASIL and sporadic small vessel disease. Neurology 66(10):1523–1526 20. Joutel A (2011) Pathogenesis of CADASIL: transgenic and knock-out mice to probe function and dysfunction of the mutated gene, Notch3, in the cerebrovasculature. Bioessays 33(1):73–80 21. Yamamoto Y, Craggs LJ, Watanabe A, Booth T, Attems J, Low RW, Oakley AE, Kalaria RN (2013) Brain microvascular accumulation and distribution of the NOTCH3 ectodomain and
Alzheimer’s disease and CADASIL granular osmiophilic material in CADASIL. J Neuropathol Exp Neurol 72(5):416–431 22. Monet-Leprêtre M, Haddad I, Baron-Menguy C, Fouillot-Panchal M, Riani M, Domenga-Denier V, Dussaule C, Cognat E, Vinh J, Joutel A (2013) Abnormal recruitment of extracellular matrix proteins by excess Notch3 ECD: a new pathomechanism in CADASIL. Brain 136(Pt 6):1830–1845 23. Louvi A, Artavanis-Tsakonas S (2012) Notch and disease: a growing field. Semin Cell Dev Biol 23:473–480 24. Hawkes CA, Härtig W, Kacza J, Schliebs R, Weller RO, Nicoll JA, Carare RO (2011) Perivascular drainage of solutes is impaired
in the ageing mouse brain and in the presence of cerebral amyloid angiopathy. Acta Neuropath 121:431–443 25. Carare RO, Hawkes CA, Jeffrey M, Kalaria RN, Weller RO (2013) Review: cerebral amyloid angiopathy, prion angiopathy, CADASIL and the spectrum of protein elimination failure angiopathies (PEFA) in neurodegenerative disease with a focus on therapy. Neuropathol Appl Neurobiol 39(6):593–611 26. Bulbarelli A, Lonati E, Brambilla A, Orlando A, Cazzaniga E, Piazza F, Ferrarese C, Masserini M, Sancini G (2012) Aβ42 production in brain capillary endothelial cells after oxygen and glucose deprivation. Mol Cell Neurosci 49(4):415–422
13
Cellular and Molecular Life Sciences
Visions and reflections
Alzheimer’s disease and CADASIL are heritable, adult‑onset dementias that both involve damaged small blood vessels Vincent T. Marchesi
Received: 2 October 2013 / Revised: 26 November 2013 / Accepted: 12 December 2013 © Springer Basel 2013
Abstract This essay explores an alternative pathway to Alzheimer’s dementia that focuses on damage to small blood vessels rather than late-stage toxic amyloid deposits as the primary pathogenic mechanism that leads to irreversible dementia. While the end-stage pathology of AD is well known, the pathogenic processes that lead to disease are often assumed to be due to toxic amyloid peptides that act on neurons, leading to neuronal dysfunction and eventually neuronal cell death. Speculations as to what initiates the pathogenic cascade have included toxic abeta peptide aggregates, oxidative damage, and inflammation, but none explain why neurons die. Recent high-resolution NMR studies of living patients show that lesions in white matter regions of the brain precede the appearance of amyloid deposits and are correlated with damaged small blood vessels. To appreciate the pathogenic potential of damaged small blood vessels in the brain, it is useful to consider the clinical course and the pathogenesis of CADASIL, a heritable arteriopathy that leads to damaged small blood vessels and irreversible dementia. CADASIL is strikingly similar to early onset AD in that it is caused by germ line mutations in NOTCH 3 that generate toxic protein aggregates similar to those attributed to mutant forms of the amyloid precursor protein and presenilin genes. Since NOTCH 3 mutants clearly damage small blood vessels of white matter regions of the brain that lead to dementia, we speculate that both forms of dementia may have a similar pathogenesis, which is to cause ischemic damage by blocking blood flow or by
V. T. Marchesi (*) Department of Pathology, Yale University, New Haven, CT 06536‑0812, USA e-mail: [email protected]
impeding the removal of toxic protein aggregates by retrograde vascular clearance mechanisms. Keywords Blood vessel damage · Alzheimer’s dementia · CADASIL · NOTCH 3
Alzheimer’s dementia is the result of the uncoupling of the neuro‑transmission machinery from the neuro‑vascular blood supply chain The amyloid hypothesis dominates the mainstream approach to Alzheimer’s therapy and focuses on attempts to either block or reduce amyloid abeta accumulations in the brains of patients in what we now know to be the late stages of Alzheimer’s dementia. This approach is inspired by the conviction that amyloid is the prime toxic element that leads to neuronal dysfunction and eventually neuronal death. Since the chemical nature of toxic amyloid and its biological targets still remain undefined, the treatments in question are in effect focusing on the pathology of the late stages of the disease without having a clear understanding of the pathogenic processes that initiate the beginning stages of the disease. This essay explores alternative pathways to Alzheimer’s dementia that focus on pathogenic processes that act in concert with amyloid, to disrupt the finely balanced interactions between neurons and their neuro-vascular blood supply. If small blood vessels rather than their cognate neurons are the first targets of a toxic process that compromises the microcirculation, the pathogenic processes that damage them represent an entirely new set of therapeutic targets that remain to be explored. While there is little doubt that amyloid, in some form, contributes to dementia, advocates of the amyloid hypothesis assume that abeta peptides are THE initiating
13
V. T. Marchesi
toxic factors that initiates the devastating cascade depicted below. Toxic factors > pathogenic processes > biologic targets > pathology However, amyloid abeta peptides are not the only agents that have the potential to impair neuronal activity. Other potential pathogens must also be considered, including reactive oxygen and nitrogen species (ROS, NOS), cytokines, chemokines, heavy metals, and mutant forms of APP, Tau and the presenilins. If abeta peptides are the prime suspects, it is critical to find out which pathogenic processes are caused by abeta peptides since these are likely to be more promising therapeutic targets than the peptides themselves. We have to ask the same question with regard to other potential neurotoxins. What types of oxidative damage result from reactive oxygens or nitrogens? Are inflammatory reactions provoked? Do damaged blood vessels lead to ischemia/hypoxia or localized vascular occlusions? Is apoptosis a consequence? To determine which pathologic processes are the most critical for brain dysfunction, it will be necessary to identify the relevant biological targets. Critical neurons in specific parts of the brain are obviously the final targets, but what happens at the earliest stages of the disease, that 10 to 15-year asymptomatic phase? During this pre-symptomatic period, small blood vessel damage has been detected by high-resolution MRI, and reduced neurotransmission, documented by reduced connectome activity, both leading to axonal damage. Activated microglia reflect an ongoing inflammatory process. None of these potential targets would be protected or reversed by therapies that focus on end-stage disease. By focusing on the most likely biological targets in the brain we have a better chance to link specific toxic factors to pathogenic processes, which in the end are most likely to be the therapeutic targets of choice. The toxic factor problem Amyloid abeta peptides are clearly a feature of advanced AD dementia, but two questions limit their potential as effective therapeutic targets. The first, unanswerable at present, is whether they are the toxic element that initiates the disease process from the very beginning. Recent PET scan data indicate that PIB stainable material is present in the brains of asymptomatic people. This indicates that detectable amyloid can accumulate in the human brain without obvious damage. Since it accumulates as people age and higher levels are often (but not always) correlated with dementia, many consider the association correlative, but not necessarily causative. The second concern is the molecular form of the toxic component. Abeta peptides, whether
13
derived from human brains or synthetically created, are technically demanding to study because of their perplexing tendency to assemble into many physical forms, which range from small 4-K monomers to aggregates that exceed 100-K Daltons. Despite this uncertainty, there is a quasiconsensus in the field that toxic “oligomers” are probably the most important pathogenic agents, although some experienced observers doubt their biological significance [1]. Compounding the problem is the lack of a consensus as to their biological targets. They have been accused of blocking synaptic activity, modifying neuronal dendrites, and reducing axonal flow in addition to damaging both large and small blood vessels. Which, if any, of these targets contributes to irreversible dementia remains to be determined. To summarize: fibrillar forms of abeta peptides make up the bulk of the late-stage amyloid plaques but their role as the prime instigators of disease remains an unresolved question. Antibody treatments that bind to abeta peptides may in some cases reduce the amount of amyloid in AD brains, but it remains to be seen whether such treatments impact the demented state.
The most relevant biologic targets Many consider the human brain the most complex organ in the body, which it undoubtedly is, but to simplify our search for answers to the dementia problem, we can more profitably focus on two functional units as the most likely pathogenic targets: the neuro-vascular blood supply and the neuronal-dominated neurotransmission machinery. Together, they represent complex but exceedingly specific interactions between blood vessel elements, various glial elements, and highly specialized neurons. The brain, like every other metabolically active organ, needs a reliable supply of oxygen and nutrients and a means of flushing out unwanted catabolites, all of which are provided by an elaborately designed neuro-vascular unit that has been well described by others [2]. Much attention has focused on the blood–brain barrier (BBB) as the unique feature of the brain vasculature that very effectively insulates the brain from the systemic circulation under normal conditions. However, when vessels are damaged by inflammatory reactions, thrombosis, or amyloid infiltrations, the tissue nourished by these vessels can undergo ischemic/hypoxic damage. Brain blood vessels of all sizes, including large leptomeningeal vessels, smaller arterioles, and the tiniest capillaries, are also sites of abeta amyloid accumulation. Why they accumulate at these sites and how they contribute to vascular injury and dysfunction is still not clear. Strickland and Grammas propose that they generate thrombotic occlusions [3, 4], but Iadacola and Zlokovic propose that they interfere with vasomotor control of local blood flow
Alzheimer’s disease and CADASIL
[5] or modify pericyte actions [2]. Whatever the mechanisms that damage them, small blood vessels could be the earliest target of the neurodegeneration process. The alternative primary target is of course the neurons themselves. There are many claims that abeta peptides of various compositions damage dendrites or interfere with synaptic activity. Most of these are in vitro studies or of model systems. Amyloid-related peptides could be involved in blocking neurotransmission, but it is not obvious whether and how they kill neurons. If abeta peptide are able to damage both blood vessels and the neurons they nourish, it is reasonable to assume that the damaged blood vessels lead to neuronal dysfunction rather than the reverse. This speculation places small blood vessels as the prime target of whatever toxic factors might be involved.
Pathogenic processes If neurons and their blood vessels are the obvious targets that lead, eventually, to dementia, we have to ask which pathogenic processes, which may act over a decades-long duration, are most likely to uncouple the neuro-transmission machinery from the neuro-vascular supply chain? It is essential to determine which are primary first causes and which are later consequences of a progressive disease. The thesis proposed here is that multiple factors which might include oxidative damage, a modified form of inflammation, and related circulatory failures act synergistically to uncouple the blood supply from selective neurons, resulting in localized oxygen and metabolite deprivation. This leads to impaired executive functions, the earliest sign of incipient dementia, which Bartzokis attributes to reduced cognitive processing speed as a consequence of myelin breakdown [6]. These changes are consistent with recent NMR findings that show white matter lesions and a disrupted connectome [7]. Progressive disease leads to reduced dendritic and synaptic function, amyloid dysregulation, and, finally, neuronal cell death, and irreversible dementia. This is the stage when amyloid plaques and neurofibrillary tangles dominate the cortical landscape. Attempting to remove them at this stage is not likely to reverse the course of the disease.
Oxidative damage and the failure of antioxidant therapy Vast literature documents the damage that ROS and NOS can inflict on the human species. Too voluminous to summarize here, a few facts emphasize why oxidative damage is considered by many to be one of the primary causes of a disease process that culminates in irreversible dementia.
Reactive oxygens can modify all classes of biological molecules, but much has been made of the damage to DNA, RNA, and related nucleotides that have been found in the brains of late-stage AD [8, 9]. Modified nucleic acids can potentially have a life of their own, unlike oxidized lipids or proteins, which can be replaced or eliminated rapidly. Guanine is considered the nucleotide base most sensitive to oxidation but it is not the only one that can be modified, and oxo-guanine in either DNA or RNA can miscode and create aberrant m-RNAs and eventually mutated proteins. Significant amounts of oxo-guanine and oxo-trinucleotides are found in AD brains, but as yet, no one has identified any mutant proteins that might be generated by modified m-RNAs, although there is now an ongoing search. It has been suggested that such oxidative modifications, which happen with alarming frequency in humans, may be more pathogenic in AD patients who lack effective DNA repair mechanisms [10] or who lack sufficient antioxidant capacity. This latter speculation has led to a massive search for agents that can scavenge unwanted radicals. The general public, less concerned with scientific rigor, has enthusiastically embraced the idea that anti-oxidants, of whatever form, are desirable, if not necessary, for a healthy functioning brain. Resveratrol, popularized by its presence in red wine, is one of many nutraceuticals that are consumed by the public with or without advice from their attending physicians. If we look at this trend critically, we have to admit that no one can claim, or disclaim, that these agents work as advertised, since there is no reliable way to measure their impact, either positively or negatively.
Oxidative damage is a fact of aerobic life, but oxidative regulation is far more biologically relevant A more in-depth look at the many roles that oxygen radicals play in daily life reveals why the search for effective antioxidant therapy has so far met with disappointing failure. We now realize that reactive oxygens such as superoxide and hydrogen peroxide are as important to post-translational modifications of critical proteins as are phosphorylation and other metabolic regulators. Cellular metabolism is under tight redox control through a complex network of sulfhydryl-regulated reactions involving glutathione and multiple peroxidases. The reducing environment of cells is regulated in part by the supply of NADPH that is generated by glycolysis and aerobic oxidative reactions, which together maintain an appropriate SH/SS ratio of critical cysteines of a vast number of proteins [11, 12]. These include numerous protein phosphatases that regulate protein phosphorylation and many other metabolic pathways. The dual action of these oxidizing agents complicates attempts to block their potentially toxic actions by
13
V. T. Marchesi
ingesting large doses of antioxidants, which, as we now know, probably never reach into the cellular compartments that generate the oxidants, since ingested antioxidants are not taken up by most cells and are rapidly excreted. One suspects that massive doses may be more likely to block normal physiologic functions than to deter unwanted damage, consistent with recent reports of toxic effects of excessive vitamin ingestion. The long-accepted notion that oxidative damage results from superoxide leaking out of mitochondria and converted via SOD to radials after exposure to heavy metals has now been replaced by a more nuanced redox concept that involves multiple NADPH oxidases that act locally and are under tight control by a glutathione/cysteine-based network. Given these many uncertainties, we have no way of evaluating the significance of failed antioxidant clinical trials.
Inflammation and the failure of anti‑inflammatory therapies More than a decade ago, inflammation was considered an important pathogenic mechanism that led to dementia, stimulated in part by anecdotal claims that individual taking NSAIDS for arthritis had a lower incidence of AD in later life. Two factors contributed to the downplaying of this idea. Enthusiasm was dampened when clinical trials of different NSAIDs failed to confirm the anecdotal claims, but its almost total rejection by opinion leaders was inspired by what seemed to be a much better idea: the amyloid hypothesis. In retrospect, failed clinical trials could have failed, and may well continue to fail, for many reasons. The simplest interpretation of failed anti-inflammation trials is that classical inflammation is either not the problem or only a contributing factor. Agents that block the acute stages in inflammation, such as blocking neutrophil accumulations, may be ineffective because that stage of the process is either truncated or bypassed by other events, such as reduced blood flow due to unregulated vasomotor activity or the lack of a classical inflammatory response in the damaged brain. One of the surprising features of late-stage Alzheimer’s dementia is the almost complete absence of connective tissue scarring in the brain despite massive neuronal cell death. In contrast to an infarct in the heart, where dead myocardial cells are replaced by angiogenesis and eventually fibrosis, dead neurons are replaced by diffuse gliosis, with very little if any neo-vascularization or scarring. Earlier studies of ischemic-induced necrosis, both experimental and naturally occurring human, reported that neutrophil accumulations were present in infarcted areas of the brain, but recent studies dispute this claim and instead suggest that the neurovascular sheath that surrounds brain
13
micro-vessels blocks neutrophil transit into the brain parenchyma [13]. Circulating blood monocytes do enter the brain in large numbers and become the population of blood-derived microglia. In contrast to the monocytes that infiltrate other ischemic organs and become tissue macrophages, these monocytes do not stimulate angiogenesis or fibrosis. It is claimed that some monocytes emerge from the blood marrow as “genetically naked” cells that are much more sensitive to genetic and epigenetic modifications than other circulating leukocytes [14]. They have a lesser capacity to repair DNA modifications that are caused by oxidative damage, and their presumed genetic flexibility allows them to home to the cerebral vasculature where they cross the neuro-vascular barrier and take up residence in the brain in the form of blood-born microglia. In this environment they “mature” into macrophages that do not stimulate extensive fibrosis. How this works is unknown. It makes sense that a connective tissue-dominated repair process would not be needed or wanted in the brain. Without regenerating neurons to replace neural functions, collagenous scar tissue, not usually a problem in other organs, could disrupt the functional linkages of the brain. A dramatic change in attitude of some AD investigators has been stimulated by recent GWAS studies, which implicate genes that may have pro-inflammatory properties. The idea that inflammation may be a significant factor in AD pathogenesis has been given a big boost by the recent discovery that mutant forms of the TREM2 gene, clearly a pro-inflammatory factor, have been found in a subset of AD patients [15]. The obvious targets of inflammatory reactions are small blood vessels, consistent with the idea that vascular injury sets the stage for neuronal dysfunction, as described in more detail below. It remains to be seen whether anti-inflammatory agents have a role in treating AD. The choice of agents, the specific inflammatory targets, and time and duration of treatment all have to be considered, but equally important, we have to find out what activates the damage of cerebral blood vessels in the first instance.
Ischemic brain damage leads to apoptosis If damaged small blood vessels represent the first step of a decades-long pathogenic process that leads ultimately to neuronal cell death and irreversible cognitive impairment, the pathogenesis of Alzheimer’s dementia shares many features with ischemic strokes that also plague the adult human population. Both result from occlusions of blood vessels that feed the brain leading to apoptosis, although each operates on a distinctly different temporal and anatomical scale. Strokes have a rapid onset caused by a
Alzheimer’s disease and CADASIL
massive loss of the blood supply to a sizable segment of the brain with debilitating consequences. Microvascular injury has a limited, localized impact and is almost always asymptomatic, but it is often distributed throughout the brain to varying degrees. Its impact can be felt when neurotransmission is affected long before apoptosis is a factor. The distinctive feature of microvascular injury in AD that leads to dementia is the presence of abeta amyloid deposits in the affected blood vessels which we assume affect the microvasculature long before they accumulate as dense plaques in the late-stage Alzheimer’s brain [16–18]. If this hypothesis is correct, the most effective therapeutic approach to AD will be to block the earliest stages of the disease, before apoptosis is reached, since blocking or reversing apoptosis following strokes is still a distant goal.
Lessons learned from CADASIL To explore how small blood vessel damage is initiated in the earliest stages of a disease it is instructive to consider the pathogenesis of CADASIL a well-recognized heritable disease of small blood vessels. CADASIL (cerebral autosomal dominant arteriopathy with sub-cortical infarct and leukoencephalopathy) is described as the most common heritable cause of stroke and vascular dementia in the adult human population. Several reviews document its clinical course and presumed pathogenesis [19–21]. Although it affects people of all ages, most are affected in the 40–60-age range and suffer from dementia that progresses to complete incapacitation by the end of the sixth decade. MRI studies detect leukoaraiosis before the onset of cognitive symptoms, and at autopsy destructive changes are found in the penetrating arteries that feed the white matter regions housing the infarcted sites. The damaged blood vessels contain what are considered pathognomonic deposits of granular osmiophilic material (GOM) that includes polypeptide segments of the extracellular domain of the trans-membrane NOTCH 3 protein and other matrix-related proteins [22]. The genetic basis of this condition is believed to be due to germline mutations in NOTCH 3, which so far include close to 200 unique nonsense mutations that modify cysteine residues in the EGF-rich segments of its extracellular domain. Genetic studies in mice and man support the idea that NOTCH 3 mutants in affected patients are gain–offunction mutations that generate toxic factors that damage smooth muscle cells of blood vessels rather than compromised NOTCH 3 functions, but it is also possible that such mutations might represent hypomorphic receptor activities [23]. It is notable that of the four known human NOTCH genes, only NOTCH 3 is expressed exclusively in vascular smooth muscle cells and brain pericytes, and
only NOTCH 3 knockouts in mice develop with minimal problems and do not develop CADASIL. It is worth pointing out how remarkably similar CADASIL and early onset Alzheimer’s dementia are to each other. • Both are caused by multiple, dominantly inherited mutations that generate potentially toxic protein aggregates • Both have a 10 to 15-year “asymptomatic” course. • Both involve damaged small blood vessels in white matter regions of the brain. • Both have protein deposits in damaged blood vessels. • Both protein deposits are generated in part by gamma secretase activities. • Both affected peoples develop symptoms in the 40– 60-age range and eventually suffer from debilitating dementia that progresses to complete incapacitation by the end of the sixth decade. While the similarities between early onset AD and CADASIL are striking, CADASIL damage is largely localized to sub-cortical white matter regions of the brain in its earliest stages while “classical” Alzheimer’s dementia is generally considered a cortical problem, dominated by massive deposits of amyloid abeta peptides. Numerous clinical trails designed to remove or reduce amyloid are based on the assumption that this is the primary pathogenic mechanism. However, many recent studies, described earlier, demonstrate that both early onset and sporadic AD dementia also involve damage to small blood vessels that feed the sub-cortical white matter regions. The small blood vessels in CADASIL brains contain mutant forms of NOTCH 3 that are expressed in vascular smooth muscle cells and generate disulphide-linked protein aggregates that are believed to be toxic to smooth muscle cells and possibly even brain capillary pericytes. Amyloid deposits are also found in small blood vessels in both forms of AD, but why they develop near vessels and how they are toxic are still unknown. It is not at all clear how mutant forms of APP or the presenilins contribute to blood vessel damage or how 95 % of AD patients without these germ line mutations develop disease. Both sets of mutations seem to influence the amount or the quality of beta peptides that are generated, but there is not the obvious connection to the vessel damage that mutant NOTCH 3 provides. Since aberrant over-production of abeta peptides in AD patients without obvious mutations is not a problem, it has been suggested that amyloid accumulations develop because of reduced turnover or defects in clearing mechanisms. Among the latter include the idea that abeta peptides, and possibly other protein aggregates, are eliminated from the brain by passing in a reverse-flow
13
V. T. Marchesi
direction along channels surrounding brain capillaries and small arterioles, an idea proposed many years ago [24] and recently updated to include the possibility that aggregates of mutant NOTCH 3 might behave in the same way [25].
Speculations If small blood vessel damage leading to localized ischemia is indeed the first step in a decades-long process that involves amyloid dysregulation and, eventually, neuronal cell death and irreversible dementia, we have to answer two critical questions: What initiates damage to small blood vessels in white matter regions of the brain? How does ischemia cause amyloid dysregulation? The second question is easier to address than the first. It is well recognized that localized ischemia can influence the metabolism of affected tissues as well as increasing amyloid abeta production by modifying the enzymes that selectively cleave the APP molecule [26]. Localized accumulations of abeta might be generated that overwhelm the drainage machinery that is also compromised by local blood flow and/or damaged blood vessels. Too little is known to be more specific. If CADASIL and AD dementia do indeed share a common pathogenic mechanism as so many shared properties imply, we have to consider the possibility that one or more NOTCH 3-like mutations, yet to be identified, damage small blood vessels in the earliest asymptomatic stages of AD that initiate the virulent cascade that ensues. Among the possibilities, I would include low levels of somatic mutations of APP, the presenilins, tau, and even NOTCH 3 itself. The large number of novel germ line mutations in APP, presenilin 1, and NOTCH 3 that appear to be pathogenic suggests that these genes might have mutational hotspots that might also be subject to additional somatic mutations during an individual’s lifetime.
References 1. Benilova I, Karran E, De Strooper B (2012) The toxic Aβ oligomer and Alzheimer’s disease: an emperor in need of clothes. Nat Neurosci 15(3):349–357 2. Sagare AP, Bell RD, Zlokovic BV (2012) Neurovascular dysfunction and faulty amyloid β-peptide clearance in Alzheimer disease. Cold Spring Harb Perspect Med 2(10). doi:10.1101/cshperspect.a011452 3. Cortes-Canteli M, Paul J, Norris EH, Bronstein R, Ahn HJ, Zamolodchikov D, Bhuvanendran S et al (2012) Fibrinogen and altered hemostasis in Alzheimer’s disease. J Alzheimers Dis 32(3):599–608 4. Tripathy D, Sanchez A, Yin X, Luo J, Martinez J, Grammas P (2013) Thrombin, a mediator of cerebrovascular inflammation in AD and hypoxia. Front Aging Neurosci 5:19
13
5. Park L, Zhou P, Koizumi K, El Jamal S, Previti ML, Van Nostrand WE, Carlson G, Iadecola C (2013) Brain and circulating levels of Aβ1-40 differentially contribute to vasomotor dysfunction in the mouse brain. Stroke 44(1):198–204 6. Lu PH, Lee GJ, Tishler TA, Meghpara M, Thompson PM, Bartzokis G (2013) Myelin breakdown mediates age-related slowing in cognitive processing speed in healthy elderly men. Brain Cogn 81(1):131–138 7. Moghekar A, Kraut M, Elkins W, Troncoso J, Zonderman AB, Resnick SM, O’Brien RJ (2012) Cerebral white matter disease is associated with Alzheimer pathology in a prospective cohort. Alzheimers Dement 8(5 Suppl):S71–S77 8. Castellani RJ, Nunomura A, Rolston RK, Moreira PI, Takeda A, Perry G, Smith MA (2008) Sublethal RNA oxidation as a mechanism for neurodegenerative disease. Int J Mol Sci 9(5):789–806 9. Ohno M, Miura T, Furuichi M, Tominaga Y, Tsuchimoto D, Sakumi K, Nakabeppu Y (2006) A genome-wide distribution of 8-oxoguanine correlates with the preferred regions for recombination and single nucleotide polymorphism in the human genome. Genome Res 16(5):567–575 10. Liu D, Croteau DL, Souza-Pinto N, Pitta M, Tian J, Wu C, Jiang H, Mustafa K, Keijzers G, Bohr WA, Mattson MP (2010) Evidence that OGG1 glycosylase protects neurons against oxidative DNA damage and cell death under ischemic conditions. J Cereb Blood Flow Metab 31:680–692 11. Thamsen M, Jakob U (2010) The redoxome proteomic analysis of cellular redox networks. Curr Opin Chem Biol 15(1):113–119 12. Cremers C, Jakob U (2013) Oxidant sensing by reversible disulfide bond formation. J Biol Chem 288:26463 13. Enzmann G, Mysiorek C, Gorina R, Cheng YJ, Ghavam pour S, Hannocks MJ, Prinz V, Dirnagl U, Endres M, Prinz M, Beschorner R, Harter PN, Mittelbronn M, Engelhardt B, Sorokin L (2013) The neurovascular unit as a selective barrier to polymorphonuclear granulocyte (PMN) infiltration into the brain after ischemic injury. Acta Neuropathol 125(3):395–412 14. Bauer M, Goldstein M, Christmann M, Becker H, Heylmann D, Kaina B (2011) Human monocytes are severely impaired in base and DNA double-strand break repair that renders them vulnerable to oxidative stress. PNAS 108(52):21105–21110 15. Guerreiro R, Wojtas A, Bras J, Carrasquillo M, Rogaeva E, Majounie E, Cruchaga C, Sassi C, Kauwe JS, Younkin S, Hazrati L, Collinge J, Pocock J, Lashley T, Williams J, Lambert JC, Amouyel P, Goate A, Rademakers R, Morgan K, Powell J, St George-Hyslop P, Singleton A, Hardy J, Alzheimer Genetic Analysis Group (2013) TREM variants in Alzheimer’s disease. N Engl J Med 368(2):117–127 16. Thal DR, Capetillo-Zarate E, Larionov S, Staufenbiel M, Zurbruegg S, Beckmann N (2009) Capillary cerebral amyloid angiopathy is associated with vessel occlusion and cerebral blood flow disturbances. Neurobiol Aging 30(12):1936–1948 17. Hunter JM, Kwan J, Malek-Ahmadi M, Maarouf CL, Kokjohn TA et al (2012) Morphological and pathological evolution of the brain microcirculation in aging and Alzheimer’s disease. PLoS One 7(5):36893. doi:10.1371/journal.pone.0036893 18. Austin BP, Nair VA, Meier TB, Xu G, Rowley HA, Carlsson CM, Johnson SC, Prabhakaran V (2011) Effects of hypoperfusion in Alzheimer’s disease. J Alzheimers Dis 26(Suppl 3):123–133 19. Charlton RA, Morris RG, Nitkunan A, Markus HS (2006) The cognitive profiles of CADASIL and sporadic small vessel disease. Neurology 66(10):1523–1526 20. Joutel A (2011) Pathogenesis of CADASIL: transgenic and knock-out mice to probe function and dysfunction of the mutated gene, Notch3, in the cerebrovasculature. Bioessays 33(1):73–80 21. Yamamoto Y, Craggs LJ, Watanabe A, Booth T, Attems J, Low RW, Oakley AE, Kalaria RN (2013) Brain microvascular accumulation and distribution of the NOTCH3 ectodomain and
Alzheimer’s disease and CADASIL granular osmiophilic material in CADASIL. J Neuropathol Exp Neurol 72(5):416–431 22. Monet-Leprêtre M, Haddad I, Baron-Menguy C, Fouillot-Panchal M, Riani M, Domenga-Denier V, Dussaule C, Cognat E, Vinh J, Joutel A (2013) Abnormal recruitment of extracellular matrix proteins by excess Notch3 ECD: a new pathomechanism in CADASIL. Brain 136(Pt 6):1830–1845 23. Louvi A, Artavanis-Tsakonas S (2012) Notch and disease: a growing field. Semin Cell Dev Biol 23:473–480 24. Hawkes CA, Härtig W, Kacza J, Schliebs R, Weller RO, Nicoll JA, Carare RO (2011) Perivascular drainage of solutes is impaired
in the ageing mouse brain and in the presence of cerebral amyloid angiopathy. Acta Neuropath 121:431–443 25. Carare RO, Hawkes CA, Jeffrey M, Kalaria RN, Weller RO (2013) Review: cerebral amyloid angiopathy, prion angiopathy, CADASIL and the spectrum of protein elimination failure angiopathies (PEFA) in neurodegenerative disease with a focus on therapy. Neuropathol Appl Neurobiol 39(6):593–611 26. Bulbarelli A, Lonati E, Brambilla A, Orlando A, Cazzaniga E, Piazza F, Ferrarese C, Masserini M, Sancini G (2012) Aβ42 production in brain capillary endothelial cells after oxygen and glucose deprivation. Mol Cell Neurosci 49(4):415–422
13
