Cerebral small vessel disease (SVD) is a major contributor to stroke, cognitive impairment and dementia with limited therapeutic interventions. There is a critical need to provide mechanistic insight and improve translation between pre-clinical research and the clinic. A 2-day workshop was held which brought together experts from several disciplines in cerebrovascular disease, dementia and cardiovascular biology, to highlight current advances in these fields, explore synergies and scope for development. These proceedings provide a summary of key talks at the workshop with a particular focus on animal models of cerebral vascular disease and dementia, mechanisms and approaches to improve translation. The outcomes of discussion groups on related themes to identify the gaps in knowledge and requirements to advance knowledge are summarized.

Introduction

Cerebral small vessel disease (SVD) is a major health challenge. Therapeutic approaches remain limited, hampered by the lack of mechanistic understanding and identification of therapeutic targets. Relevant animal models could provide a cornerstone to basic scientific studies of disease mechanisms and pre-clinical studies of potential therapies, but there is a critical need to improve the current translational gap that exists between pre-clinical research and treatments in patients. The Medical Research Council Dementias Platform UK (MRC DPUK) Vascular Experimental Medicine Theme identified that a comprehensive assessment of the latest developments in animal models and of their contribution to understanding of cerebral microvascular disease would reduce the translational gap. In response to this, a 2-day workshop took place in late January 2017 at the British Heart Foundation Centre of Research Excellence in Glasgow, Scotland in conjunction with MRC DPUK and brought together experts from several disciplines in cerebrovascular disease, dementia and cardiovascular biology, to highlight current advances in these fields, explore synergies and scope for development. There were presentations from UK and international researchers and a specific focus on animal models of cerebral microvascular disease and dementia, considering vascular biology, neurogliovascular coupling, blood–brain barrier (BBB) function, neuroinflammation, cerebral drainage pathways, and methodological and translational challenges (see Figure 1 for the general organization of the meeting including the key topics and themes discussed). This overview provides a summary of the key talks, with a particular focus on mechanisms of cerebral vascular disease (see Figure 2) and improving translation. These talks were followed by related themed discussion groups on the gaps in knowledge and requirements to advance knowledge, the outcomes of which are highlighted in Table 1; in addition, articles related to the topic published in Clinical Science have been gathered together in a special themed collection at http://www.portlandpresspublishing.com/cc/small-vessels.

General organization of the meeting

Mechanisms relevant to cerebral vascular disease.

Figure 2
Mechanisms relevant to cerebral vascular disease.

An outline of mechanisms relevant to cerebral vascular disease and VCI. These mechanisms were highlighted during the workshop and described in more detail in the text. In general, downstream of risk factors, a cascade of key events are indicated, of which some may be considered as primary events (e.g. endothelial changes, inflammation, oxidative stress) and may also be bidirectional, leading to cognitive impairment. Abbreviations: BBB, blood–brain barrier; MMP, matrix metalloproteinase.

Figure 2
Mechanisms relevant to cerebral vascular disease.

An outline of mechanisms relevant to cerebral vascular disease and VCI. These mechanisms were highlighted during the workshop and described in more detail in the text. In general, downstream of risk factors, a cascade of key events are indicated, of which some may be considered as primary events (e.g. endothelial changes, inflammation, oxidative stress) and may also be bidirectional, leading to cognitive impairment. Abbreviations: BBB, blood–brain barrier; MMP, matrix metalloproteinase.

Table 1
Gaps in knowledge and requirements to advance knowledge
Mechanism Gaps in knowledge Requirements for early advances in knowledge 
Common to all mechanisms • The most relevant model for human SVD is unclear. • Use different models to recapitulate components of SVD appropriate to the specific research question. 
• Regional variation in CBF, in vulnerability to low flow, hypoxia or inflammation, and in flow-metabolism coupling, in grey and white matter (periventricular and deep), in normal tissue and in SVD, is not well established. • Better in vitro and in vivo models of BBB function and dysfunction, including under conditions of shear stress, in cell co-cultures (3D >2D) and including the glycocalyx. 
• Influence of regional differences (e.g. in arteriolar morphology between lenticulostriate and cortical perforating arterioles) in microvessel and perivascular structures on lesion development is poorly understood. • A reliable ‘NGVU on a chip’ would help accelerate understanding of pathophysiological mechanisms, identify targets and allow high throughput testing of potential interventions. 
 • Better molecular imaging probes to study different NGVU cell types, determine molecular tissue changes and probe inflammatory processes. 
 • Use larger mammalian models, e.g. well characterized porcine models, since rodents have low white matter:grey matter ratio. 
BBB • A dynamic complex structure whose dysfunction is a common feature of neurological diseases including cerebral SVD and related dementia syndromes (e.g. AD and VCI/VaD). • Core set of reference standard techniques for preclinical, neuropathological and clinical studies. 
• Lack of understanding of: In vivo and in vitro methods to model small and transient BBB changes. 
a) how cellular morphology relates to function; • More sensitive neuropathological methods to identify WMH as seen on MRI. 
b) the temporal changes and sequence of damage; • New molecular imaging probes for pericytes and endothelial cell function status. 
c) how BBB dysfunction leads to brain parenchymal lesions. • More use of the retina routinely in rodent models (as in human SVDs) could help visualize arteriolar/venular, retinopathic and nerve fibre layer changes that could advance understanding of brain changes in rodent models (as suggested in human SVDs). 
 • Alternatives to gadolinium for human in vivo BBB imaging. 
Hypoperfusion and ischaemia • Often poor characterization in models. • Longitudinal investigations in models. 
• Poor understanding of the influence of structural differences in perforating arterioles in different brain regions and arterial territories. • Better characterization of tissue metabolic changes (e.g. pO2, glucose, metabolites) (particularly white matter). 
• The primary trigger, whether inflammation, altered flow, other, and their interactions. • More standard use of terminology for SVD in in vivo models and neuropathology, and for studying components of NVU in vitro would help, as has been implemented in clinical studies [3]. 
• Is flow-metabolism coupling altered in SVD?  
Inflammation • How do microglia, pericytes, perivascular macrophages, fibroblasts, oligodendrocytes all interact to maintain homoeostasis? • Determine the contribution of inflammation in SVD and define whether there is a classic innate response and/or contribution of systemic inflammation or both. 
• Are soluble inflammatory mediators (complement, chemokines, cytokines) involved in perivascular inflammation in SVD? • Undertake longitudinal studies across models to profile microglial proliferation and migration and the broader molecular/cellular neuroinflammatory environment including measures of circulating immune cells. 
• What is the role of fibrosis in SVD? • In relation to above a wider investigation of the ‘matrisome’. 
• How is the matrix environment perturbed to influence cell function and signalling?  
White matter: axons, myelin sheaths • The role of oligodendrocytes and myelinated axons in the pathogenesis of SVD is unclear as is their interaction with other cells, such as endothelial cells. • Comparative studies of potential SVD mechanisms across models, species, and diseases of white matter that have potential similarities to SVD (e.g. multiple sclerosis) using the most innovative molecular, cellular and imaging approaches. 
• The extent of damage to myelinated axons in cortical grey matter in SVD is poorly defined. • Use MRI/DTI of white matter status as routinely as possible to link in vivo to tissue pathology more clearly. 
• The relationship of demyelination and axonal loss to CBF changes in SVD is unclear. • Link white matter pathology stages and burden to vascular pathology load and CBF alterations more clearly. 
 • Investigate integrity of architecture of myelinated axons in cortical grey matter and white matter. 
Perivascular and interstitial fluid drainage • To what extent are rodents relevant models of human drainage pathways and mechanisms? • Determine how CSF and ISF and clearance pathways all relate to each other. 
• What is the driving force for ISF clearance: • Determine if ISF and perivascular drainage vary with sleep-wake cycle. 
a) vasomotion of cortical arteries?; • Better methods to detect perivascular and ISF drainage in humans. 
b) perivascular innervation?; • Can CSF or ISF drainage pathways be targetted to inform treatment for SVD? 
c) does the smooth muscle action in veins have a functional role? • Does targetting of matrix proteins (TIMPs or MMPs) or vasomotion increase drainage? 
• How does ISF clearance relate to SVD development or progression?  
• Does extracellular remodelling vary during the sleep-wake cycle?  
Neuro-glial-vascular mechanisms • How do vascular changes drive glial and neuronal dysfunction and damage? • Target white matter more in rodents. 
 • Standardize anaesthetics and stimulation procedures. 
 • Identify better ways to measure neuronal responses than blood oxygen dependent (BOLD) and related imaging methods (arterial spin labelling, blood volume measures, O2 metabolism, EEG and magnetoencephalography). 
 • Expand cohort imaging similar to that achieved in the AD Neuroimaging Initiative. 
 • Make more use of optogenetics and genetically encoded calcium indicators. 
Mechanism Gaps in knowledge Requirements for early advances in knowledge 
Common to all mechanisms • The most relevant model for human SVD is unclear. • Use different models to recapitulate components of SVD appropriate to the specific research question. 
• Regional variation in CBF, in vulnerability to low flow, hypoxia or inflammation, and in flow-metabolism coupling, in grey and white matter (periventricular and deep), in normal tissue and in SVD, is not well established. • Better in vitro and in vivo models of BBB function and dysfunction, including under conditions of shear stress, in cell co-cultures (3D >2D) and including the glycocalyx. 
• Influence of regional differences (e.g. in arteriolar morphology between lenticulostriate and cortical perforating arterioles) in microvessel and perivascular structures on lesion development is poorly understood. • A reliable ‘NGVU on a chip’ would help accelerate understanding of pathophysiological mechanisms, identify targets and allow high throughput testing of potential interventions. 
 • Better molecular imaging probes to study different NGVU cell types, determine molecular tissue changes and probe inflammatory processes. 
 • Use larger mammalian models, e.g. well characterized porcine models, since rodents have low white matter:grey matter ratio. 
BBB • A dynamic complex structure whose dysfunction is a common feature of neurological diseases including cerebral SVD and related dementia syndromes (e.g. AD and VCI/VaD). • Core set of reference standard techniques for preclinical, neuropathological and clinical studies. 
• Lack of understanding of: In vivo and in vitro methods to model small and transient BBB changes. 
a) how cellular morphology relates to function; • More sensitive neuropathological methods to identify WMH as seen on MRI. 
b) the temporal changes and sequence of damage; • New molecular imaging probes for pericytes and endothelial cell function status. 
c) how BBB dysfunction leads to brain parenchymal lesions. • More use of the retina routinely in rodent models (as in human SVDs) could help visualize arteriolar/venular, retinopathic and nerve fibre layer changes that could advance understanding of brain changes in rodent models (as suggested in human SVDs). 
 • Alternatives to gadolinium for human in vivo BBB imaging. 
Hypoperfusion and ischaemia • Often poor characterization in models. • Longitudinal investigations in models. 
• Poor understanding of the influence of structural differences in perforating arterioles in different brain regions and arterial territories. • Better characterization of tissue metabolic changes (e.g. pO2, glucose, metabolites) (particularly white matter). 
• The primary trigger, whether inflammation, altered flow, other, and their interactions. • More standard use of terminology for SVD in in vivo models and neuropathology, and for studying components of NVU in vitro would help, as has been implemented in clinical studies [3]. 
• Is flow-metabolism coupling altered in SVD?  
Inflammation • How do microglia, pericytes, perivascular macrophages, fibroblasts, oligodendrocytes all interact to maintain homoeostasis? • Determine the contribution of inflammation in SVD and define whether there is a classic innate response and/or contribution of systemic inflammation or both. 
• Are soluble inflammatory mediators (complement, chemokines, cytokines) involved in perivascular inflammation in SVD? • Undertake longitudinal studies across models to profile microglial proliferation and migration and the broader molecular/cellular neuroinflammatory environment including measures of circulating immune cells. 
• What is the role of fibrosis in SVD? • In relation to above a wider investigation of the ‘matrisome’. 
• How is the matrix environment perturbed to influence cell function and signalling?  
White matter: axons, myelin sheaths • The role of oligodendrocytes and myelinated axons in the pathogenesis of SVD is unclear as is their interaction with other cells, such as endothelial cells. • Comparative studies of potential SVD mechanisms across models, species, and diseases of white matter that have potential similarities to SVD (e.g. multiple sclerosis) using the most innovative molecular, cellular and imaging approaches. 
• The extent of damage to myelinated axons in cortical grey matter in SVD is poorly defined. • Use MRI/DTI of white matter status as routinely as possible to link in vivo to tissue pathology more clearly. 
• The relationship of demyelination and axonal loss to CBF changes in SVD is unclear. • Link white matter pathology stages and burden to vascular pathology load and CBF alterations more clearly. 
 • Investigate integrity of architecture of myelinated axons in cortical grey matter and white matter. 
Perivascular and interstitial fluid drainage • To what extent are rodents relevant models of human drainage pathways and mechanisms? • Determine how CSF and ISF and clearance pathways all relate to each other. 
• What is the driving force for ISF clearance: • Determine if ISF and perivascular drainage vary with sleep-wake cycle. 
a) vasomotion of cortical arteries?; • Better methods to detect perivascular and ISF drainage in humans. 
b) perivascular innervation?; • Can CSF or ISF drainage pathways be targetted to inform treatment for SVD? 
c) does the smooth muscle action in veins have a functional role? • Does targetting of matrix proteins (TIMPs or MMPs) or vasomotion increase drainage? 
• How does ISF clearance relate to SVD development or progression?  
• Does extracellular remodelling vary during the sleep-wake cycle?  
Neuro-glial-vascular mechanisms • How do vascular changes drive glial and neuronal dysfunction and damage? • Target white matter more in rodents. 
 • Standardize anaesthetics and stimulation procedures. 
 • Identify better ways to measure neuronal responses than blood oxygen dependent (BOLD) and related imaging methods (arterial spin labelling, blood volume measures, O2 metabolism, EEG and magnetoencephalography). 
 • Expand cohort imaging similar to that achieved in the AD Neuroimaging Initiative. 
 • Make more use of optogenetics and genetically encoded calcium indicators. 

A summary of outcomes of discussion groups at the workshop are outlined. Several key mechanisms, the gaps in knowledge and requirements for early advances in knowledge are highlighted. Abbreviations: AD, Alzheimer’s disease; BBB, blood–brain barrier; CBF, cerebral blood flow; CSF, cerebrospinal fluid; DTI, diffusion tensor imaging; EEG, electroencephalogram; ISF, interstitial fluid; MRI, magnetic resonance imaging; MMP, matrix metalloproteinase; NGVU, neuro-glial-vascular unit; SVD, small vessel disease; TIMP, tissue inhibitor of metalloproteinase; VCI, vascular cognitive impairment; VaD, vascular dementia.

Cerebral vascular disease: clinical context

Of the approximate 17 million strokes per year worldwide, 20–25% are due to SVD. Additionally, a large burden of silent SVD contributes to up to 45% of the 35–42 million new cases of dementia globally each year. SVD is difficult to research, small vessel stroke being neglected in stroke research, and vascular causes are neglected in dementia research. Human pathology often reflects late disease stages with few relevant experimental models identified [1,2] Consequently, treatment is empirical and probably suboptimal. The individual SVD features on neuroimaging are now well-described with standardized terminology [3]. These include acute (recent) small subcortical infarcts and clinically covert features (lacunes, white matter hyperintensities (WMH), microbleeds, perivascular spaces, microinfarcts and cerebral atrophy), and many haemorrhagic strokes. From early neuropathology studies onwards, SVD has been the vascular lesion most strongly associated with vascular cognitive impairment (VCI) [4], an observation confirmed consistently through more recent large cohort studies of donated brain tissue, showing that neuropathological SVD is associated with cognitive impairment [5].

SVD features are associated with age and several vascular risk factors (hypertension, smoking, diabetes, hypercholesterolemia), but vascular risk factors explain only a small proportion of the variance [6] leaving a large proportion unexplained. Potential pathophysiological mechanisms include vessel stiffening, blood–brain barrier (BBB) failure, poorer premorbid white matter integrity [7,8] a cross-sectional association with reduced cerebral blood flow (CBF) [9,10], and several underlying genetic contributors [11].

Progress in reducing SVD requires questioning of conventional wisdom. Firstly, the effects of SVD on the brain are both diffuse and more dynamic than appreciated previously [12,13]. Subcortical lesions have remote effects; lesions can appear silently or cause symptoms, or can resolve completely, cavitate or form a WMH [14]. WMH, lacunes and microbleeds may increase, but also decrease [15,16] which suggests that these lesions do not indicate permanent brain damage only. Perivascular spaces are crucial for cerebral interstitial fluid (ISF) drainage and waste clearance [17]. They are a site of inflammation and amyloid protein build-up, linking to Alzheimer's disease (AD). Other cell types and structures are involved in SVD, not just the vascular endothelium and smooth muscle cells, but also microglia, astrocytes, oligodendrocytes, pericytes, matrix proteins all play a role. Finally, data from humans and experimental models (see below) indicate a vulnerability to SVD in part [18] explaining why vascular risk factor exposure in some people is far more devastating than in others.

Cerebral vascular disease: rodent models

Whilst there is no one model that captures the complexities of SVD and VCI, different animal models have been studied that encompass many hypothesized causes (e.g. SVD); hypoxic hypoperfusion injury secondary to large artery disease; or genetic influences (see recent review [2] and previous systematic review [19]).

A widely studied model is the spontaneously hypertensive stroke prone (SHRSP) rat which develops malignant hypertension. This model has been reported to develop SVD-like features including white matter changes, lipohyalonisis and fibrinoid necrosis [20,21]. Endothelial and the earliest measured myelin changes have also been shown to predate hypertension [22]. Analysis of gene and protein expression before the onset of hypertension highlighted several genes and proteins associated with SVD and pathways that may underlie these changes (e.g. matrix and vascular integrity, inflammation), and suggests a potential predisposition of SHRSPs to development of SVD-like pathology which is later exacerbated by hypertension [23]. However these findings vary and others have failed to identify white matter damage in older SHRSP in the absence of stroke [24] and in other models of hypertension such as an inducible transgenic model [25]. Additional factors, such as added dietary salt, hypoperfusion and high fat diet have been shown to exacerbate white matter pathology in SHRSP [26] similar to such factors accelerating human SVD.

Alternative models of hypertension have been developed including the effects of angiotensin (ANG) administration. Low doses of angiotensin-II (ANG-II) in rodents results in a slow-developing rise in blood pressure which impairs the coupling of the neuro-glial-vascular unit (NGVU) and endothelial responses, before the onset of hypertension and in the absence of SVD-like changes [27] . More recently, simultaneous administration of ANG-II and L-NAME (NG-nitro-L-arginine methyl ester) led to chronically elevated blood pressure and produced early SVD-like features [28] . Key features relevant to human SVD were reported in this model including hypertrophy of the cerebral vessel wall, structural impairment of the NGVU, hippocampal atrophy, BBB impairment and occasional microinfarcts and microbleeds.

To investigate vascular insufficiency as a central mechanism in the pathophysiology of VCI, models of cerebral hypoperfusion were developed [29]. These models have been refined over the years to induce hypoperfusion whilst avoiding severe reductions in blood flow that would cause immediate ischaemia. The recent models involve stenosis of the common carotid arteries which immediately restricts blood flow to the forebrain on application of microcoils [30], or gradually restricts blood flow by the use of constrictor devices [31]. These models have been widely studied with behavioural measures of cognition which demonstrate particular impairments in spatial working memory reminiscent of the disrupted frontal cortical circuitry found in VCI [29,32]. In these models, reduced perfusion causes a slowly evolving diffuse white matter pathology that can be detected using a variety of approaches such as immunohistochemistry and in vivo magnetic resonance imaging (MRI). The oxygen levels are reduced to hypoxic levels in white matter, which is thought to be the main driver of the ensuing pathological changes [29]. These models have been widely studied to understand the pathophysiology of hypoperfusion-induced white matter changes that are often accompanied by indices of inflammation. Carotid stenosis induced using microcoils in the mouse has also been shown to culminate in marked changes to microvascular structure, BBB breakdown, glial-vascular damage and microinfarcts/microbleeds relevant to human SVD [33]. These studies are predominantly conducted in C57Bl/6J mice which have poor collateral flow through the Circle of Willis and the resultant effects of microcoil/constrictor application assigned to reduced cerebral perfusion. However other, albeit ill-defined, effects on vascular stiffness/pulsatility and cerebrospinal fluid (CSF) drainage may also contribute to the pathology. In a longitudinal study in humans, there is an absence of a direct association between carotid stenosis and WMH or cognitive decline and instead a link with vascular stiffness [34].

Notably, these models of carotid stenosis and hypertension often lack the overt white matter lesions (WML) characteristic of SVD. However, WML are difficult to detect pathologically and a combination of different risk factors may be required to produce WML. As indicated above, the effects of hypoperfusion and diet have been studied in the SSP model [26] which exacerbate white matter changes, as compared with individual factors, and precipitate overt WML. Models with co-morbidities may more accurately reflect clinical SVD and be considered as a basis for testing therapeutic interventions.

A number of different mouse models that express highly penetrant gene mutations linked to familial SVD (NOTCH3, COL4A1, COL4A2, HTRA1) have provided important insights into causal pathways related to these gene mutations. Interestingly, investigation of these models has now indicated that these different gene mutations may lead to common convergent pathways involving the impairment of extracellular matrix (ECM) function [35]. Further genome wide association studies (GWAS) have shown common variants associated with familial SVD, such as COL4A1/4A2, are also risk factors for common sporadic SVD [36]. Thus, an emerging view is that these models may be useful for probing mechanisms important not only in rare monogenic forms of SVD, but also in the more common sporadic forms of SVD, opening up avenues of potential therapeutic intervention.

Both vascular and AD pathologies appear to work synergistically to promote neurodegeneration and cognitive impairments [37]. Further, there is considerable overlap between risk factors for cerebral microvascular disease, cardiovascular disease and AD, and yet the synergies between these areas are so far largely unexplored in models. Animal models could provide a powerful basis to tease out molecular interactions. Amyloid is believed to be a key driver of the pathophysiology of AD so most work has focussed on models that accumulate amyloid in the brain. Rodents do not naturally develop and deposit amyloid, but genetically modified mice were developed that harbour mutations in human amyloid precursor protein associated with familial forms of AD (TgAPP) and that lead to age-dependent accumulation of amyloid. The study of these TgAPP models revealed that a prominent cerebrovascular dysfunction occurs prior to the onset of amyloid deposition, which would be consistent with impaired vasoreactivity and reduced CBF seen in AD patients [37]. TgAPP models are now being probed to study interactions with risk factors such as hypertension and high fat diet which in general have been shown to exacerbate the cerebrovascular dysfunction and progression of amyloid pathology [37,38].

Across the various models, vascular haemodynamics and NGVU coupling are impaired (see [28]) and related to breakdown in cell-cell communication and signalling within the NGVU and supporting ECM. Whilst animal models are limited in their ability to fully recapitulate human SVD, they remain pertinent to tease out specific questions that are impossible to address in human studies and to identify molecular targets.

Mechanisms relevant to cerebral vascular disease

Endothelial-related mechanisms

The exact molecular and cellular mechanisms underlying cerebral microvascular disease remain elusive, but endothelial dysfunction seems to be especially important. Cerebral endothelial cells regulate vascular tone and blood flow, protect against thrombosis, inflammation and fibrosis, control exchange across the BBB, modulate innate immunity and influence clearance of amyloid peptides [39,40]. In addition, and unique to the cerebrovascular system, the microvascular endothelium is critically involved in blood–brain function and neuroprotection, through autoregulatory mechanisms that protect against barotrauma. Consequences of cerebrovascular autoregulatory dysfunction include structural injury to capillaries, microvascular rarefaction, exaggerated disruption of the BBB, neuroinflammation, neurodegeneration and increased susceptibility to intracerebral haemorrhage. This link between impaired autoregulation and downstream microvascular injury has been demonstrated experimentally and clinically.

Endothelial-cell (EC) signalling is mediated through production of numerous endothelial cell-derived vasoactive agents, including nitric oxide (NO), endothelium-derived hyperpolarizing factors, endothelium-derived relaxing factors (EDRF), eicosanoid mediators (EETs, HETEs), prostaglandins and endothelin-1 (ET-1) [41,42]. Endothelial-derived NO is not only a potent vasodilator, but also inhibits platelet aggregation inflammation, apoptosis, fibrosis and vascular smooth muscle cell proliferation as well as modulates mitochondrial function, neuronal metabolism and synaptic transmission [43]. Cerebrovascular endothelial cells possess reactive oxygen species (ROS)-generating enzymes, such as NADPH oxidases (Nox) and mitochondrial oxidases that produce superoxide and hydrogen peroxide, important in redox signalling and endothelial function [44,45] . Because the endothelium is in direct contact with blood flow, it ‘senses’ haemodynamic changes and adapts accordingly through production of EC-derived factors, such as NO, EDRF and prostacyclins to induce vasodilation and prevent vascular damage or ET-1 and 20- hydroxyeicosatetraenoic acid to promote vasoconstriction. At the capillary level, endothelial cells are in direct contact with neurons via astrocytes and as such there is ‘cross-talk’ between vascular and neural cells through NGVU coupling, important in regulating neural activity and cerebral function [46].

Ageing, cardiovascular disease and ischaemia/hypoxia are associated with endothelial damage that impacts microvascular function, moment-to-moment adjustment of regional blood flow and neuronal function [47]. Molecular mechanisms involved include reduced microvascular eNOS-derived NO production with consequent decline in cerebrovascular, neuronal, astrocyte, microglial and angiogenic function. Mitochondrial damage, production of arachidonic metabolites and increased oxidative stress are also important. Decreased NO bioavailability in microvascular endothelial cells increases expression of amyloid precursor protein (APP) and β-site APP cleaving enzyme 1 (BACE-1) as well as amyloid (Aβ) formation, suggesting a role for endothelial dysfunction in neurodegenerative pathologies, such as AD [48,49]. Moreover chronic reduction of NO generation and increased production of pro-inflammatory cytokines (TNF-α, IL-1β, interferon γ) increases endothelial permeability, transmigration of pro-inflammatory cells and BBB dysfunction, promoting neuroinflammatory and neurodegenerative processes [50,51].

Increased ROS production in the cerebral microvasculature is due primarily to hyperactivation of Noxs (particularly Nox2 and Nox4) [52–54], but also to disruption of mitochondrial respiratory chain oxidases, decreased antioxidant capacity, reduced nuclear factor erythroid 2-related factor 2 (Nrf-2) activity, uncoupled eNOS, and increased levels of asymmetric dimethylarginine (ADMA), an endogenous NOS inhibitor [55,56] . Vascular oxidative stress, activation of endothelial cell cation channels, such as transient receptor potential (TRP) melastatin 2 [57] and TRP vanilloid 4 cation channels [58], and induction of calcium and redox-sensitive signalling pathways promote oxidative modification of lipids, proteins and DNA causing endothelial, pericyte and neuronal cell injury [59]. Overexpression of the antioxidant superoxide dismutase-1 rescues cerebral endothelial dysfunction associated with neurodegeneration [60], while resveratrol treatment improves cerebromicrovascular endothelial function, reduces Nox activity and rescues NGVU coupling in aged mice [61].

Thus, endothelium-mediated mechanisms of cerebromicrovascular dysfunction involve a complex network of factors and systems. There is important interplay between molecular and cellular elements that regulate CBF, blood brain function and the microvascular endothelium. Perturbations of these interconnections promote cerebral hypoperfusion, BBB damage and endothelial dysfunction, with consequent neurodegeneration and clinical manifestations of cognitive decline and vascular dementia (VaD). To date there are still no therapies to target the microvascular endothelium and hence management of these patients remains sub-optimal.

White matter, inflammation and hypoxia

Demyelinating lesions in the CNS white matter are a feature common to both SVD and multiple sclerosis (MS). MS has been studied in much greater detail than SVD, raising the possibility that knowledge gained in MS may help in understanding the demyelination in SVD. Demyelinating MS lesions have been classified into four types depending on their morphological features during lesion development [62]. Two types, Patterns I and II, have morphological features consistent with an autoimmune aetiology, but a third type, Pattern III, has features suggesting an important role for tissue hypoxia, based on the similarity with demyelination in the penumbra around ischaemic foci. Since demyelination in SVD is likely to involve tissue hypoxia, it may be instructive to consider the factors determining the formation of Pattern III lesions in MS.

Our understanding of Pattern III demyelination has been greatly advanced by the introduction of a model lesion formed in the rodent spinal white matter by the direct intraspinal injection of the pro-inflammatory agent lipopolysaccharide (LPS) [63]. The lesion has been validated as an accurate model for Pattern III lesions in MS [64], and it results from innate, rather than autoimmune, mechanisms [65]. A surprising feature of the experimental Pattern III lesion is its location, because it does not form precisely at the site of injection (in this case in the middle of the dorsal columns), as is usual for demyelinating agents [66,67] but rather it forms near the base of the dorsal columns, and after a delay of 7–10 days from the time of injection [68]. It is as if the LPS does not directly cause the demyelination, but rather sets in motion a train of events that culminates in the lesion forming at a nearby site. This peculiarity focusses attention on the features at the base of the dorsal columns that make it vulnerable, and these include a paucity of blood vessels, and location at a watershed between three arterial trees formed by long, fine, end arteries. All these features render the site particularly vulnerable to hypoxia, especially if the blood supply is impaired. The hypoxia kills the most vulnerable cells first, and these include the oligodendrocytes, resulting in the demyelination observed. The suspected key role of tissue hypoxia in causing Pattern III demyelination has been confirmed by the observation that breathing raised oxygen avoids tissue hypoxia, and prevents the demyelination [68].

In summary, impaired vascular perfusion, as occurs in cerebral SVD and MS, appears to cause tissue hypoxia that is most severe at sites rendered vulnerable by the vascular architecture, notably at poorly vascularized sites located at vascular watersheds, such as those in the periventricular white matter. The most vulnerable cells, oligodendrocytes, die first, resulting in demyelination, but the tissue can be protected from demyelination, in rodents, if hypoxia is avoided by breathing air enriched with oxygen.

Microvascular extracellular matrix and small vessels diseases of the brain

The ECM of cerebral blood vessels, and especially the basement membrane, is a key component of the NGVU that occupies a very strategic location, at the interface between the cerebral microcirculation and astrocytes [35]. Proteins constituting or associated with the ECM, referred to as the ‘matrisome’, have not only structural but also biochemical and signalling roles [69]. Hence, any change in the microvascular matrisome could have profound impact on the brain parenchyma. A common feature in both monogenic and sporadic forms of SVDs of the brain is the presence of major remodelling of the brain microvascular ECM, with prominent fibrosis, associated with severe degeneration of vascular smooth muscle cells [35]. Over the past 10 years, genetic studies have revealed that most monogenic forms of SVDs are caused by mutations in genes encoding matrisome proteins, namely dominant mutations in the coding region of α1 (COL4A1) or α2 (COL4A2) chains of collagen type IV, a major component of vascular basement membranes, in COL4A1/COL4A2-related haemorrhagic stroke [70,71], dominant gain-of-function mutations in the 3′UTR of COL4A1 in Pontine Autosomal Dominant Microangiopathy with Leukoencephalopathy (PADMAL) [72], recessive or dominant loss-of-function mutations in HTRA1, a serine protease regulating the TGFβ pathway, in Cerebral Autosomal Recessive Arteriopathy with Subcortical Infarcts and Leukoencephalopathy (CARASIL) and HTRA1-related SVDs, respectively [73–75], and dominant mutation in Cathepsin A in Cathepsin A-related arteriopathy with strokes and leukoencephalopathy (CARASAL) [76]. Moreover, recent work established that cerebrovascular manifestations of CADASIL, the most frequent hereditary SVD and the most aggressive vasculopapthy, are driven by perturbations of the microvascular matrisome, as a consequence of an abnormal accumulation of the extracellular domain of Notch3 receptor [77,78]. Specifically, elevated levels of tissue inhibitor of metalloproteinases 3 (TIMP3), an ECM regulator, blunts the activity of the ADAM17/HB-EGF/(ErbB1/ErbB4) pathway leading to the up-regulation of voltage-gated potassium channels subfamily 1 in cerebral arterial myocytes, thereby attenuating myogenic responses in brain arteries and compromising CBF regulation in CADASIL [79].

Yet, much work remains to be done to understand in detail the mechanisms linking perturbations of the microvascular ECM and cerebrovascular manifestations in these hereditary SVDs. Remarkably, these recent findings suggest that similar mechanisms may be at work in sporadic SVDs. Therefore, work is needed to characterize the ‘prefibrotic/transitional changes’ of the brain microvascular matrisome in response to age or hypertension, two major risk factors of sporadic SVDs, and investigate their contribution to disease initiation. Also, with respect to fibrosis of small vessels in the brain, major gaps remain in our understanding of the mechanisms initiating the fibrosis response, in identifying the molecular and cellular mediators as well as the effector cells, and in understanding the relationship with SMC degeneration. Improvement of animal models of SVDs are needed to address these gaps.

Intramural periarterial drainage pathways

Arteries that supply the brain are vital for CBF but they also play a major role in the drainage of ISF from the brain. As arteries penetrate the brain from the subarachnoid space, they acquire a layer of pia mater that separates the arterial wall from the glia limitans. After injections of particulate and soluble tracers into the subarachnoid space, tracers are entering by glymphatic/convective influx within the arterial pial-glial basement membranes of cortical arteries [80]. When soluble fluorescent tracers are injected into the brain parenchyma, they rapidly drain out of the brain with ISF along basement membranes in the walls of cerebral capillaries and basement membranes surrounding smooth muscle cells of arteries, a process named intramural periarterial drainage [81]. Changes in arterial basement membranes related to genotype (apolipoprotein E4), age or blockage due to immune complexes lead to a failure of intramural perivascular drainage pathways [82,83]. A recent study on a canine brain using electron microscopy demonstrated that all arterioles are enveloped in one complete layer of leptomeningeal cells, often with a second incomplete layer in the white matter, whereas venules in the white matter only occasionally possessed one layer only of leptomeningeal cells [84]. It is possible that dilated perivascular spaces observed in the white matter are created between the two adventitial layers of pia mater as a result of the blocked drainage pathways in the overlying grey matter [85]. The extensive fibrosis and hyaline change seen in SVD disrupts not only the basement membrane pathways carrying ISF out of the brain parenchyma but is also likely to contribute to the stiffening of the artery walls that would also impair drainage of ISF from cerebral white matter. Clarifying the key factors involved in the impairment of perivascular drainage of ISF such as the role of vascular smooth muscle cells and modulators of extracellular matrix are essential therapeutic strategies. It is currently unknown whether there is a pattern of anatomical or topographical changes in the cerebrovascular basement membrane and perivascular compartment architecture in brains susceptible to different degrees or subtypes of VaD.

The glymphatic fluid transport system (and interactions with amyloid)

Lymphatic circulation is essential in peripheral organs for the removal of excess fluid and metabolic waste products; but the brain, in spite of having the highest metabolic activity of all tissues, has no conventional lymphatic system. A fundamentally novel pathway for fluid transport has recently been identified: the glymphatic system consisting of a periarterial CSF influx path and a perivenous ISF clearance route, which are coupled through convective interstitial bulk flow supported by astrocytic aquaporin4 (AQP4) water channels [86]. It is designated the ‘glymphatic system’ because it operates analogously to the peripheral lymphatic system and the CSF/ISF fluxes depend on astroglial AQP4 channels [87] . This brain-wide pathway serves as a waste disposal system of fluid, as well as larger solutes and proteins, such as amyloid and tau, from the CNS. Recent work has extended the glymphatic concept to show that outflow occurs along both cervical and meningeal/pial lymphatic vessels [88,89] . An interesting observation is that the sleep-wake cycle regulates glymphatic activity, which is primarily active during sleep, or the use of some anaesthesthetic agents [90]. Glymphatic drainage is similar to lymphatic depending upon position, and the lateral and supine positions, which are the most popular sleep positions among human and most animal species, appear to be superior for amyloid-β clearance compared with the prone position [91]. Ageing is linked to a sharp decline in glymphatic activity in wildtype mice [92]. Moreover, APP/PS1 mice, a murine model of AD, exhibited reduced glymphatic activity in young mice without amyloid plaques as well as in older mice with plaques. Mice with multiple mini-strokes induced by intracarotid injection of cholesterol crystals similarly showed a global decrease in glymphatic function [93]. Also, traumatic brain injury triggered a sustained suppression of glymphatic activity [94], whereas middle-aged rats with diabetes exhibit increased influx, but reduced clearance, of contrast agents in an MRI study [95]. The latter observation can best be explained by trapping of the contrast agent within the neuropil due to pathological changes in the extracellular matrix and reactive gliosis, since the volume of fluid influx and efflux must match each other. Very recent small studies, one with Gadolinium injected into the CSF, the other using a novel MR technique (MREG), provided interesting evidence for the existence of glymphatic fluid transport in the human brain [96,97]. Altogether, glymphatic function has been shown to be altered in all disease states studied so far.

Gaps in knowledge and requirements to advance knowledge

Following the focussed lectures the participants assembled in small groups and discussed the gaps in the different research areas and opportunities for development which are summarized in Table 1. There were several key themes to emerge from the discussion groups which are highlighted.

Methodological approaches

Modelling the neuro-glial-vascular unit in vitro

The neuro-glial-vascular (NGVU) is arguably the most elementary representative functional unit of the brain and specific alterations in the NGVU underpin much of the pathophysiology of cerebrovascular disease. The ideal in vitro model will depend upon the particular use and research question to be addressed. However, to probe disease mechanisms, attempts should be made to recapitulate the complex micro-environment of the NGVU, including the key cell types and the extracellular matrix. The constituent cells should also display mature properties of the NGVU such as the specialization of endothelial cells to impart effective BBB function, specific transport mechanisms and the requisite cell–cell interactions including those that underpin NGVU coupling. NGVU models should be robust, yield reproducible results and be scalable if they are to be incorporated into drug screening programmes. Current models can be broadly divided into easily adopted static compartment models such as Transwell cultures and more sophisticated flow models based upon a microfluidic design [98]. The latter have emerged in recognition of the importance of flow and shear stress of maintaining NGVU properties.

More than 80% of drug candidates tested in animal models for stroke, and a large proportion so far of drugs tested for AD, have failed in clinical trials. Better in vitro mimics of human systems might help increase the likelihood of successful drug discovery. Notably there have been few trials to date of any drugs for SVD, therefore it would be premature to level the same criticism at human microvascular disease. Nonetheless, human stem cell technologies offer the opportunity to generate patient specific NGVU unit models for improved drug discovery and personalized medicine [99]. Differentiation protocols have been described that would produce all the constituent cells of the NGVU [100–102]. However further work is required to demonstrate the reproducibility of these methods and the ideal co-culture conditions.

NGVU in vitro models should enable assessment of physiological and pathophysiological mechanisms that would be difficult to dissect in vivo. Molecular 'omics' technologies in particular provide a means to evaluate disease processes in an unbiased approach and identify new drug targets. However, molecular studies require careful planning as they can be subject to significant technical confounds and non-relevant biological processes which are only beginning to be fully understood. Conducting studies with sufficient samples to allow data normalization methods and remove nuisance variance is important. Single cell 'omics' are also a powerful technique to deal with heterogeneity in cell culture models as well as reveal cell-specific disease processes. These recent advances in tissue engineering, stem cell methods and novel 'omics' platforms are anticipated to enable unprecedented mechanistic insights into NGVU brain disorders such as dementia [103].

Measuring behavioural outcomes in animal models

In human studies, different cognitive domains can be assessed with a battery of visual touchscreen-based tasks using similar stimuli and responses, thus facilitating between-task comparison and potentially reducing confounds [104–106]. Conversely, in the rodent, comprehensive cognitive batteries are seldom employed, and different cognitive domains are assessed by tasks that typically vary widely in the nature of stimuli, responses, reinforcers, and testing environment. Moreover, these tasks often bear no resemblance to the tests used with humans. Finally, the measurement of rodent behaviour is notoriously variable across labs and experimenters [107]. These factors reduce the likelihood of rodent and human studies assessing consistent and comparable cognitive functions, thus compromising the efficacy of translation. There is a recognized need to share standardized operating procedures for common outcomes and to improve translation.

One approach to addressing these issues is by using a touchscreen-based cognitive testing method for rodents, which has the potential to achieve more accurate, efficient and reproducible phenotyping of rodents, and help bridge the translational divide between animal and human studies of cognition. This method uses an automated operant chamber with a computer monitor for the presentation of visual stimuli and an infrared touchscreen assembly to record the animal's responses. The apparatus allows for the flexible presentation of visual stimuli at any location on the screen. Rats or mice respond directly to the stimuli by breaking the infrared beams overlaying the touchscreen with their nose. Appetitive reinforcers such as strawberry milkshake are delivered in a reward magazine to the rear of the chamber.

An extensive battery of tests using this touchscreen cognitive testing apparatus has been developed and validated. The tasks closely parallel human tests and have high translational face validity. Although face validity does not guarantee translational neurocognitive validity (i.e. the same cognitive constructs and circuits mediating the tasks across species), minimizing methodological differences enables back- and forward-translational opportunities and improves the likelihood of neurocognitive validity of touchscreen tasks [108–111]. Touchscreen-based tests now form a core component of recommended cognitive test batteries for models of schizophrenia (e.g. NEWMEDS and CNTRICS) and AD (e.g. PHARMACOG). The system is presented as a key method in the standard guide to behavioural testing for mouse researchers [112], and protocols for an extensive battery of tests have been published [113–115].

Experimental design

Findings from in vivo research may be less reliable if those studies do not adopt and report measures to reduce the risk of bias in the experiments reported. The experimental stroke community have been at the forefront of implementing changes to improve the conduct and reporting of studies, but the recent impact of these efforts is not clear. A systematic identification of recent literature was conducted describing animal experiments inducing middle cerebral artery occlusion or lacunar stroke, and text mining approaches developed automatically to ascertain risk of bias reporting from full text articles [116]. There were substantial improvements in the reporting of middle cerebral artery occlusion studies since the first systematic report in 2007 [117]. However, in reports of experiments in lacunar stroke there was no substantial improvement since the first systematic report of the field to 2012 [118]. This may in part reflect that there has been less work on SVD models. The accuracy (true positive plus true negative as a percentage of the total) of automated risk of bias annotation ranged from 67% (randomization, lacunar stroke) to 100% (sample size calculation, middle cerebral artery occlusion). There therefore remains substantial opportunity for improvement in the reporting, and probably the conduct, of animal research modelling stroke, particularly lacunar stroke. Automated tools are sufficiently accurate to identify whether studies report the blinded assessment of outcome, but improvements are required in the tools to ascertain whether randomization and a sample size calculation were reported.

Multi-centre pre-clinical studies

There are many steps in the translational pathway from original mechanistic discoveries through to successful translation of a therapy to the clinic. Progress in this pathway has been poor across a range of CNS diseases despite increased knowledge of the mechanistic basis of diseases, identification of drug targets and positive drug studies using in vitro and in vivo models. This failure to translate led to criticism of the quality of in vivo drug study design and reporting, and the publication and uptake of guidelines (e.g. STAIR, ARRIVE, RIGOR, IMPROVE) [119–123] to improve the translational potential of pre-clinical studies. However, single centre studies, although useful for proof of concept of a particular therapy, are unlikely to be rigorous enough to predict the therapies which will also be effective in man. What is needed is an additional step in the pathway where promising therapies which demonstrate efficacy in a number of single centre studies are then tested more rigorously on a multi-site platform. Multi-PART (Multicentre Preclinical Animal Research Team, http://www.dcn.ed.ac.uk/multipart/) is such a platform, set up with funding from EU FP7 for ischaemic stroke. However, the structure of the platform could easily be adapted for multi-site studies for any CNS disease.

The key objective of Multi-PART was ’to implement and establish a platform for international multicentre preclinical stroke trials using the repertoire of randomized clinical trial design and the complexities of a multicentre, multimodel paradigm‘. This was made possible by the experience within the Multi-PART team of scientists and clinicians with expertise in a range of stroke models and outcome measures, regulation and ethics of in vivo research, precision, reproducibility and external validity in animal research, design and running of clinical trials, statistics, database and web site management, good laboratory practice (GLP), and standard operating procedures (SOPs), development and costing structures etc. The design of the Multi-PART platform also benefitted from improvements in clinical trial design by incorporating strategies to minimize bias, biostatistical advances, data monitoring and auditing etc.

Multicentre pre-clinical randomized control trials (pRCTs) in models relevant to SVD and VCI could be performed after single centre laboratory studies and a systematic review of all existing data, prior to clinical testing. The need for pRCTs reflects the repeated failure of translation from preclinical to clinical development and lack of reproducibility between laboratories. Since new healthcare interventions require one or more positive large multicentre clinical RCTs prior to introduction, pRCTs should largely follow the design of phase III clinical RCTs involving randomization, treatment and outcome blinding [124–126]. Coordination would need a Trial Steering Committee and central database for management of randomization and data storage. Outcomes would need central adjudication of imaging (e.g. MRI, histology for lesion volume, microbleeds) and videos (behavioural and functional testing). Studies can utilize advanced trial techniques such as adaptive design, randomization, and statistical analysis, and the concept extended to designing observational and characterization studies across multiple laboratories.

Key differences exist between preclinical and clinical RCTs, many of which could be ameliorated by adopting specific design elements in preclinical RCTs. Since patients are very heterogeneous, a variety of SVD models would be needed. Preclinical RCTs may need to incorporate both positive and neutral control comparator groups. Classically, trials have a single primary outcome but co-primary outcomes, composite outcomes, or statistical integration of several outcomes are alternatives that may be preferred. The Data Monitoring Committee will assess futility during the course of the study with the aim of stopping further development if data are not positive, thus preventing the need for further preclinical and clinical testing, and their considerable expense. Conversely, a positive multicentre pRCT will considerably enhance the justification for proceeding to clinical development.

Changing the views of industry, funding agencies and grant reviewers may be necessary to overcome perceptions that pRCTs are incremental and not novel or ground-breaking. For large pRCTs, authorship should be discussed prior to commencing studies and presents a potential issue for many basic science academics where decisions on hiring, promotion and returns on the research excellence framework (REF) are taken on publications and author position.

Biomarkers

Biomarkers can give insights into pathogenesis and suggest potential drug-able targets. They can inform prognosis and be used to ‘enrich’ trial populations with participants who are most likely to have outcomes of interest such as progression to dementia. Biomarkers can be used as surrogate outcome measures for phase II intervention trials, to screen treatment prior to large and expensive definitive phase 3 trials. Finally, biomarkers can be used for diagnostics. In this regard researchers were mindful of the continued debate around the value of biomarkers in diagnosis of AD [127,128]. The necessary properties and performance characteristics of a biomarker will vary with the proposed purpose.

MRI modalities were regarded as offering the greatest utility as SVD biomarkers. A number of different MRI markers of SVD may represent useful markers of disease including recent small subcortical infarcts, lacunes, WMH, brain volume and white matter ultrastructure estimated on diffusion tensor imaging (DTI) or other quantitative methods [129,130]. Increasingly sophisticated approaches to MRI are moving beyond gross anatomy to assess function and connectivity using approaches such as network analysis and these show promise in the mouse pre-clinical hypoperfusion model to predict outcome [131] but need further evaluation in prospective patient cohorts [132]. Computerized tomography (CT) has value, particularly as it is routinely available but lacks sensitivity to early SVD change. For imaging, there is no suitable SVD target for radiolabelling that is analogous to amyloid imaging although PET approaches may be useful to assess parts of the pathophysiological process such as neuroinflammation. For tissue-based markers, use of CSF is under-researched in SVD in direct contrast to its importance in AD. However, the dynamics of the exchange between the peptides in the ISF of the brain and the CSF and their routes in the perivascular compartment still require investigation. To date no blood-based biomarker has shown utility, although with ‘omics’ based technologies new markers may be discovered. Electroencephalogram (EEG) has a role in certain neurodegenerative diseases but has limited application in SVD.

External validation, replication and standardization of putative biomarkers is critical particularly for multicentre work. At present there is no SVD biomarker suitable for clinical use. Future trials and observational cohorts should create biobanks of tissue (blood, CSF) and imaging data that can be interrogated to validate existing biomarkers or discover potential new markers. If biomarkers are to be used as reliable surrogate markers in clinical trials, they must fulfil the following criteria: (1) they must be able to predict clinical outcome, as changes induced by a therapy on a surrogate marker are expected to reflect changes in a clinically meaningful end point; (2) change in a surrogate marker must be detectable prospectively and (3) the sample size required to show therapeutic efficacy should be feasible in the setting of a clinical trial [133].

Sharing, model exchange and potential tissue banks

There is a growing need to address standardization across models, as far as this is possible. This could include standardized protocols for surgery, behavioural testing and clearly defined genetic details in relation to models. Central facilities for provision of genetically modified mice would reduce the genetic drift inherent in many of the frequently used models. However, their utility is dependent on phenotypes. Human tissue banks are well established and provide a useful source of translational material. The development of a tissue bank for preclinical models may be a useful resource which could range from cut sections of tissues through to whole organs with standardization of pathological approaches. Potential issues in relation to animal licensing (e.g. UK Home office) is acknowledged. Tissue banks may facilitate validation of published work in a low cost way, by providing tissues prepared by one published group for laboratory studies to be undertaken by a second group in a remote laboratory. An open access resource where data could be deposited would be particularly useful for negative studies that may not otherwise be published.

Industry perspective on the workshop

Independent of the shape of future research efforts there are a number of critical areas that need to be kept in mind that will help facilitate drug discovery and reduce the overall level of risk. Firstly, the generation of high quality targets/pathways with strong supporting neuropathological, genetic, and/or epidemiological data. This will require the generation and interrogation of well characterized cohorts, using best practices for data generation and sharing that can be found in other pre-competitive settings; these cohorts will be additionally important for longitudinal tracking of disease and the identification of potential biomarkers that can be used to make early Go/NoGo decisions within drug development programmes.

Secondly, there should be a strong focus on generating a robust translational pharmacology package to bridge between preclinical models and clinical development. A number of basic principles underpin this area: understand the temporal expression of target in both the model and the human disease; match the preclinical intervention paradigm to the intended therapeutic intent in the clinic (i.e. prevention versus treatment); demonstrate in preclinical studies drug exposure at the site of action/target engagement, expression of downstream target pharmacology and importantly generate exposure–response relationships (e.g. EC50 values).

Finally, as far as possible standardize procedures/protocols to reduce variability and improve reproducibility of preclinical studies. With this in mind the emergence of the EU-funded Multi-PART initiative provides a network capability for the conduct of randomized preclinical studies in experimental models relevant to SVD. Together these approaches will all serve to ‘de-risk’ clinical development and encourage further industry and academic investment in SVD research in dementia and stroke.

A summary of the workshop proceedings consisting of short lectures and outcomes of discussion groups were provided by a writing group comprise Karen Horsburgh, Joanna Wardlaw, Rhian Touyz, Ken Smith, Anne Joutel, Roxana Carare, Maiken Nedergaard, Zameel Cader, Lisa Saksida, Malcolm Macleod, Mhairi Macrae, Andy Lockhart, Philip Bath, Terence Quinn, Colin Smith, Atticus Hainsworth, Tracy Farr, Lorraine Work, Seth Love, Steve Williams, Michael O'Sullivan, Rajesh Kalaria, Anna Williams, Cheryl Hawkes, Stuart Allan, Patrick Kehoe, Catherine Hall, Jason Berwick, Christopher McCabe, Colin Smith, Tom Van Agtmael, Hugh Markus. The administrative support of Margaret Kinninmont at the BHF Cardiovascular Centre, Glasgow is gratefully appreciated. The participation of representatives from ARUK, (Rosa Sancho), Alzheimer's Society (Claire De May), Stroke Association (Kate Holmes) and Portland Press (Emma Pettengale) is also gratefully acknowledged. P.B. is Stroke Association Professor of Stroke Medicine, and is a NIHR Senior Investigator.

Funding

This work was supported by the ARUK, ARUK Scotland network; Alzheimer's Society; BHF and BHF Cardiovascular centre; MRC-DPUK; the Royal Society Edinburgh; Stroke Association and Fondation Leducq.

Competing interests

Z.C. has received honoraria and consultancy fees from TEVA, Novartis, Orion, Daichii Sankyo, Allergan.

Abbreviations

     
  • AD

    Alzheimer's disease

  •  
  • amyloid

  •  
  • APP

    amyloid precursor protein

  •  
  • AQP4

    aquaporin4

  •  
  • BACE-1

    β-site APP cleaving enzyme 1

  •  
  • BBB

    blood–brain barrier

  •  
  • CADASIL

    Cerebral Autosomal Dominant Arteriopathy with Subcortical Infarcts and Leukoencephalopathy

  •  
  • CARASAL

    Cathepsin A Related Arteriopathy with Strokes and Leukoencephalopathy

  •  
  • CARASIL

    Cerebral Autosomal Recessive Arteriopathy with Subcortical Infarcts and Leukoencephalopathy

  •  
  • CBF

    cerebral blood flow

  •  
  • CSF

    cerebrospinal fluid

  •  
  • CT

    computerized tomography

  •  
  • DTI

    diffusion tensor imaging

  •  
  • EC

    endothelial cell

  •  
  • ECM

    extracellular matrix

  •  
  • EDRF

    endothelium-derived relaxing factor

  •  
  • EEG

    electroencephalogram

  •  
  • ET-1

    endothelin-1

  •  
  • ISF

    interstitial fluid

  •  
  • MMP

    matrix metalloproteinase

  •  
  • MRI

    magnetic resonance imaging

  •  
  • MS

    multiple sclerosis

  •  
  • NGVU

    neuro-glial-vascular unit

  •  
  • NO

    nitric oxide

  •  
  • NOX

    NADPH oxidase

  •  
  • PADMAL

    Pontine Autosomal Dominant Microangiopathy and Leukoencephalopathy

  •  
  • pRCT

    pre-clinical randomized control trial

  •  
  • ROS

    reactive oxygen species

  •  
  • SHRSP

    spontaneously hypertensive stroke prone

  •  
  • SVD

    small vessel disease

  •  
  • TIMP

    tissue inhibitor of metalloproteinase

  •  
  • VaD

    vascular dementia

  •  
  • VCI

    vascular cognitive impairment

  •  
  • WMH

    white matter hyperintensities

  •  
  • WML

    white matter lesions

References

References
1
Bailey
E.L.
,
McCulloch
J.
,
Sudlow
C.
and
Wardlaw
J.M.
(
2009
)
Potential animal models of lacunar stroke: a systematic review
.
Stroke
40
,
e451
e458
[PubMed]
2
Hainsworth
A.H.
,
Allan
S.M.
,
Boltze
J.
,
Cunningham
C.
,
Farris
C.
,
Head
E.
et al. 
(
2017
)
Translational models for vascular cognitive impairment: a review including larger species
.
BMC Med.
15
,
16
3
Wardlaw
J.M.
,
Smith
E.E.
,
Biessels
G.J.
,
Cordonnier
C.
,
Fazekas
F.
,
Frayne
R.
et al. 
(
2013
)
Neuroimaging standards for research into small vessel disease and its contribution to ageing and neurodegeneration: a united approach
.
Lancet Neurol.
12
,
822
838
[PubMed]
4
Esiri
M.M.
,
Wilcock
G.K.
and
Morris
J.H.
(
1997
)
Neuropathological assessment of the lesions of significance in vascular dementia
.
J. Neurol. Neurosurg. Psychiatry
63
,
749
753
[PubMed]
5
Arvanitakis
Z.
,
Capuano
A.W.
,
Leurgans
S.E.
,
Bennett
D.A.
and
Schneider
J.A.
(
2016
)
Relation of cerebral vessel disease to Alzheimer's disease dementia and cognitive function in elderly people: a cross-sectional study
.
Lancet Neurol.
15
,
934
943
[PubMed]
6
Wardlaw
J.M.
,
Allerhand
M.
,
Doubal
F.N.
,
Hernandez
M.V.
,
Morris
Z.
,
Gow
A.J.
et al. 
(
2014
)
Vascular risk factors, large artery atheroma and brain white matter hyperintensities
.
Neurology
82
,
1331
1338
[PubMed]
7
Backhouse
E.V.
,
McHutchison
C.A.
,
Cvoro
V.
,
Shenkin
S.D.
and
Wardlaw
J.M.
(
2017
)
Early life risk factors for cerebrovascular disease: a systematic review and meta-analysis
.
Neurology
88
,
976
984
[PubMed]
8
Wardlaw
J.M.
,
Smith
C.
and
Dichgans
M.
(
2013
)
Mechanisms of sporadic cerebral small vessel disease: insights from neuroimaging
.
Lancet Neurol.
12
,
483
497
[PubMed]
9
Shi
Y.
,
Thrippleton
M.J.
,
Makin
S.D.
,
Marshall
I.
,
Geerlings
M.I.
,
de Craen
A.J.
et al. 
(
2016
)
Cerebral blood flow in small vessel disease: a systematic review and meta-analysis
.
J. Cereb. Blood Flow Metab.
36
,
1653
1667
[PubMed]
10
Nylander
R.
,
Fahlström
M.
,
Rostrup
E.
,
Kullberg
J.
,
Damangir
S.
,
Ahlström
H.
et al. 
(
2017
)
Quantitative and qualitative MRI evaluation of cerebral small vessel disease in an elderly population: a longitudinal study
.
Acta Radiol.
11
Haffner
C.
,
Malik
R.
and
Dichgans
M.
(
2016
)
Genetic factors in cerebral small vessel disease and their impact on stroke and dementia
.
J. Cereb. Blood Flow Metab.
36
,
158
171
[PubMed]
12
Chappell
F.
, ,
Valdes Hernandez
M.
,
Makin
S.
,
Shuler
K.
,
Sakka
E.
,
Dennis
M.S.
et al. 
(
2017
)
Sample size considerations for trials using cerebral white matter hyperintensity progression as an intermediate outcome at 1 year after mild stroke: results from a prospective cohort study
.
Trials
18
,
78
,
[PubMed]
13
METACOHORTS Consortium
(
2016
)
METACOHORTS for the study of vascular disease and its contribution to cognitive decline and neurodegeneration. An initiative of the Joint Programme for Neurodegenerative Disease research
.
Alzheimers Dement.
12
,
1235
1249
[PubMed]
14
Potter
G.M.
,
Doubal
F.N.
,
Jackson
C.A.
et al. 
(
2010
)
Counting cavitating lacunes underestimates the burden of lacunar infarction
.
Stroke
41
,
267
272
[PubMed]
15
Wardlaw
J.M.
,
Chappell
F.M.
,
Valdés Hernández
M.C.
,
Makin
S.D.
,
Staals
J.
,
Shuler
K.
et al. 
(
2017
)
White matter hyperintensity reduction and outcomes after minor stroke
.
Neurology
89
,
1003
1010
[PubMed]
16
van Leijsen
E.M.C.
,
van Uden
I.W.M.
,
Ghafoorian
M.
,
Bergkamp
M.I.
,
Lohner
V.
,
Kooijmans
E.C.M.
et al. 
(
2017
)
Non-linear temporal dynamics of cerebral small vessel disease. The RUN DMC study
.
Neurology
89
,
1569
1577
[PubMed]
17
Jessen
N.A.
,
Munk
A.S.
,
Lundgaard
I.
and
Nedergaard
M.
(
2015
)
The glymphatic system: a beginner's guide
.
Neurochem. Res.
40
,
2583
2599
[PubMed]
18
Lopez
L.
,
Hill
W.D.
,
Harris
S.E.
,
Valdes Hernandez
M.
,
Munoz Maniega
S.
,
Bastin
M.E.
et al. 
(
2015
)
Genes from a translational analysis support a multifactorial nature of white matter hyperintensities
.
Stroke
46
,
341
347
,
[PubMed]
19
Jiwa
N.S.
,
Garrard
P.
and
Hainsworth
A.H.
(
2010
)
Experimental models of vascular dementia and vascular cognitive impairment: a systematic review
.
J. Neurochem.
115
,
814
828
[PubMed]
20
Bailey
E.L.
,
Smith
C.
,
Sudlow
C.L.
and
Wardlaw
J.M.
(
2011
)
Is the spontaneously hypertensive stroke prone rat a pertinent model of sub cortical ischemic stroke? A systematic review
Int. J. Stroke
6
,
434
444
[PubMed]
21
Ogata
J.
,
Fujishima
M.
,
Tamaki
K.
,
Nakatomi
Y.
,
Ishitsuka
T.
and
Omae
T.
(
1981
)
Vascular changes underlying cerebral lesions in stroke-prone spontaneously hypertensive rats. A serial section study
.
Acta Neuropathol.
54
,
183
188
[PubMed]
22
Bailey
E.L.
,
McBride
M.W.
,
Beattie
W.
,
McClure
J.D.
,
Graham
D.
,
Dominiczak
A.F.
et al. 
(
2014
)
Differential gene expression in multiple neurological, inflammatory and connective tissue pathways in a spontaneous model of human small vessel stroke
.
Neuropathol. Appl. Neurobiol.
1940
,
855
872
23
Bailey
E.L.
,
Wardlaw
J.M.
,
Graham
D.
,
Dominiczak
A.F.
,
Sudlow
C.L.
and
Smith
C.
(
2011
)
Cerebral small vessel endothelial structural changes predate hypertension in stroke-prone spontaneously hypertensive rats: a blinded, controlled immunohistochemical study of 5- to 21-week-old rats
.
Neuropathol. Appl. Neurobiol.
37
,
711
726
[PubMed]
24
Brittain
J.F.
,
McCabe
C.
,
Khatun
H.
,
Kaushal
N.
,
Bridges
L.R.
,
Holmes
W.M.
et al. 
(
2013
)
An MRI-histological study of white matter in stroke-free SHRSP
.
J. Cereb. Blood Flow Metab.
33
,
760
763
[PubMed]
25
Holland
P.R.
,
Pannozzo
M.A.
,
Bastin
M.E.
,
McNeilly
A.D.
,
Ferguson
K.J.
,
Caughey
S.
et al. 
(
2015
)
Hypertension fails to disrupt white matter integrity in young or aged Fisher (F44) Cyp1a1Ren2 transgenic rats
.
J. Cereb. Blood Flow Metab.
35
,
188
192
[PubMed]
26
Jalal
F.Y.
,
Yang
Y.
,
Thompson
J.F.
,
Roitbak
T.
and
Rosenberg
G.A.
(
2015
)
Hypoxia-induced neuroinflammatory white-matter injury reduced by minocycline in SHR/SP
.
J. Cereb. Blood Flow Metab.
35
,
1145
1153
[PubMed]
27
Capone
C.
,
Faraco
G.
,
Park
L.
,
Cao
X.
,
Davisson
R.L.
and
Iadecola
C.
(
2011
)
The cerebrovascular dysfunction induced by slow pressor doses of angiotensin II precedes the development of hypertension
.
Am. J. Physiol. Heart Circ. Physiol.
300
,
H397
H407
[PubMed]
28
Meissner
A.
,
Minnerup
J.
,
Soria
G.
and
Planas
A.M.
(
2017
)
Structural and functional brain alterations in a murine model of Angiotensin II-induced hypertension
.
J. Neurochem.
140
,
509
521
[PubMed]
29
Duncombe
J.
,
Kitamura
A.
,
Hase
Y.
,
Ihara
M.
,
Kalaria
R.N.
and
Horsburgh
K.
(
2017
)
Chronic cerebral hypoperfusion: a key mechanism leading to vascular cognitive impairment and dementia (VCID) closing the translational gap between rodent models and human VCID
.
Clin. Sci. (Lond.)
131
,
2451
2468
[PubMed]
30
Shibata
M.
,
Ohtani
R.
,
Ihara
M.
and
Tomimoto
H.
(
2004
)
White matter lesions and glial activation in a novel mouse model of chronic cerebral hypoperfusion
.
Stroke
35
,
2598
2603
[PubMed]
31
Hattori
Y.
,
Enmi
J.
,
Iguchi
S.
,
Saito
S.
,
Yamamoto
Y.
,
Tsuji
M.
et al. 
(
2016
)
Gradual carotid artery stenosis in mice closely replicates hypoperfusive vascular dementia in humans
.
J. Am. Heart Assoc.
5
,
e002757
[PubMed]
32
Coltman
R.
,
Spain
A.
,
Tsenkina
Y.
,
Fowler
J.H.
,
Smith
J.
,
Allerhand
M.
et al. 
(
2011
)
Spatial memory in mice remains intact despite widespread white matter pathology
.
Neurobiol. Aging
32
,
2324.e7
2324.e12
33
Holland
P.R.
,
Searcy
J.L.
,
Salvadores
N.
,
Scullion
G.
,
Chen
G.
,
Lawson
G.
et al. 
(
2015
)
Gliovascular disruption and cognitive deficits in a mouse model with features of small vessel disease
.
J. Cereb. Blood Flow Metab.
35
,
1005
1014
[PubMed]
34
Wardlaw
J.M.
,
Allerhand
M.
,
Eadie
E.
,
Thomas
A.
,
Corley
J.
,
Pattie
A.
et al. 
(
2017
)
Carotid disease at age 73 and cognitive change from age 70 to 76 years: a longitudinal cohort study
.
J. Cereb. Blood Flow Metab.
37
,
3042
3052
[PubMed]
35
Joutel
A.
,
Haddad
I.
,
Ratelade
J.
and
Nelson
M.T.
(
2016
)
Perturbations of the cerebrovascular matrisome: a convergent mechanism in small vessel disease of the brain?
J. Cereb. Blood Flow Metab.
36
,
143
157
[PubMed]
36
Rannikmäe
K.
,
Van Agtmael
T.
,
Woo
D.
,
Markus
H.S.
,
Seshadri
S.
,
Rosand
J.
et al. 
(
2015
)
Common variation in COL4A1/COL4A2 is associated with sporadic cerebral small vessel disease
.
Neurology
84
,
918
926
[PubMed]
37
Iadecola
C.
(
2013
)
The pathobiology of vascular dementia
.
Neuron
80
,
844
866
[PubMed]
38
Salvadores
N.
,
Searcy
J.L.
,
Holland
P.R.
and
Horsburgh
K.
(
2017
)
Chronic cerebral hypoperfusion alters amyloid-β peptide pools leading to cerebral amyloid angiopathy, microinfarcts and hemorrhages in Tg-SwDI mice
.
Clin. Sci. (Lond.)
131
,
2109
2123
[PubMed]
39
Lyros
E.
,
Bakogiannis
C.
,
Liu
Y.
and
Fassbender
K.
(
2014
)
Molecular links between endothelial dysfunction and neurodegeneration in Alzheimer's disease
.
Curr. Alzheimer Res.
11
,
18
26
[PubMed]
40
Bell
R.D.
,
Winkler
E.A.
,
Sagare
A.P.
,
Singh
I.
,
LaRue
B.
,
Deane
R.
et al. 
(
2010
)
Pericytes control key neurovascular functions and neuronal phenotype in the adult brain and during brain aging
.
Neuron
68
,
409
427
[PubMed]
41
Katusic
Z.S.
and
Austin
S.A.
(
2014
)
Endothelial nitric oxide: protector of a healthy mind
.
Eur. Heart J.
35
,
888
894
[PubMed]
42
Faraci
FM.
(
2011
)
Protecting against vascular disease in brain
.
Am. J. Physiol. Heart Circ. Physiol.
300
,
H1566
H1158
[PubMed]
43
Busija
D.W.
,
Rutkai
I.
,
Dutta
S.
and
Katakam
P.V.
(
2016
)
Role of mitochondria in cerebral vascular function: energy production, cellular protection, and regulation of vascular tone
.
Compr. Physiol.
6
,
1529
1548
[PubMed]
44
Yanagida
K.
,
Liu
C.H.
,
Faraco
G.
,
Galvani
S.
,
Smith
H.K.
,
Burg
N.
et al. 
(
2017
)
Size-selective opening of the blood-brain barrier by targeting endothelial sphingosine 1-phosphate receptor 1
.
Proc. Natl. Acad. Sci. USA
114
,
4531
4536
[PubMed]
45
Chrissobolis
S.
,
Banfi
B.
,
Sobey
C.G.
and
Faraci
F.M.
(
2012
)
Role of Nox isoforms in angiotensin II-induced oxidative stress and endothelial dysfunction in brain
.
J. Appl. Physiol.
113
,
184
191
[PubMed]
46
Lourenço
C.F.
,
Ledo
A.
,
Barbosa
R.M.
and
Laranjinha
J.
(
2017
)
Neurovascular-neuroenergetic coupling axis in the brain: master regulation by nitric oxide and consequences in aging and neurodegeneration
.
Free Radic. Biol. Med.
108, 668-682
47
de la Torre
J.C.
(
2012
)
Cerebral hemodynamics and vascular risk factors: setting the stage for Alzheimer's disease
.
J. Alzheimers Dis.
32
,
553
567
[PubMed]
48
Merlini
M.
,
Shi
Y.
,
Keller
S.
,
Savarese
G.
,
Akhmedov
A.
,
Derungs
R.
et al. 
(
2017
)
Reduced nitric oxide bioavailability mediates cerebroarterial dysfunction independent of cerebral amyloid angiopathy in a mouse model of Alzheimer's disease
.
Am. J. Physiol. Heart Circ. Physiol.
312
,
H232
H238
[PubMed]
49
Koizumi
K.
,
Wang
G.
and
Park
L.
(
2016
)
Endothelial dysfunction and amyloid-β-induced neurovascular alterations
.
Cell. Mol. Neurobiol.
36
,
155
165
[PubMed]
50
Basuroy
S.
,
Bhattacharya
S.
,
Leffler
C.W.
and
Parfenova
H.
(
2009
)
Nox4 NADPH oxidase mediates oxidative stress and apoptosis caused by TNF-alpha in cerebral vascular endothelial cells
.
Am. J. Physiol. Cell Physiol.
296
,
C422
C432
[PubMed]
51
Matsumoto
J.
,
Takata
F.
,
Machida
T.
,
Takahashi
H.
,
Soejima
Y.
,
Funakoshi
M.
et al. 
(
2014
)
Tumor necrosis factor-α-stimulated brain pericytes possess a unique cytokine and chemokine release profile and enhance microglial activation
.
Neurosci. Lett.
578
,
133
138
[PubMed]
52
Chrissobolis
S.
,
Drummond
G.R.
,
Faraci
F.M.
and
Sobey
C.G.
(
2014
)
Chronic aldosterone administration causes Nox2-mediated increases in reactive oxygen species production and endothelial dysfunction in the cerebral circulation
.
J. Hypertens.
32
,
1815
1821
[PubMed]
53
Kuroda
J.
,
Ago
T.
,
Nishimura
A.
,
Nakamura
K.
,
Matsuo
R.
,
Wakisaka
Y.
et al. 
(
2014
)
Nox4 is a major source of superoxide production in human brain pericytes
.
J. Vasc. Res.
51
,
429
438
[PubMed]
54
Cahill-Smith
S.
and
Li
J.M.
(
2014
)
Oxidative stress, redox signalling and endothelial dysfunction in ageing-related neurodegenerative diseases: a role of NADPH oxidase 2
.
Br. J. Clin. Pharmacol.
78
,
441
453
[PubMed]
55
Ghosh
N.
,
Ghosh
R.
and
Mandal
S.C.
(
2011
)
Antioxidant protection: a promising therapeutic intervention in neurodegenerative disease
.
Free Radic. Res.
45
,
888
905
[PubMed]
56
Rodionov
R.N.
,
Dayoub
H.
,
Lynch
C.M.
,
Wilson
K.M.
,
Stevens
J.W.
,
Murry
D.J.
et al. 
(
2010
)
Overexpression of dimethylarginine dimethylaminohydrolase protects against cerebral vascular effects of hyperhomocysteinemia
.
Circ. Res.
106
,
551
558
[PubMed]
57
Park
L.
,
Wang
G.
,
Moore
J.
,
Girouard
H.
,
Zhou
P.
,
Anrather
J.
et al. 
(
2014
)
The key role of transient receptor potential melastatin-2 channels in amyloid-beta-induced neurovascular dysfunction
.
Nat. Commun.
5
,
5318
5320
[PubMed]
58
Zhang
L.
,
Papadopoulos
P.
and
Hamel
E.
(
2013
)
Endothelial TRPV4 channels mediate dilation of cerebral arteries: impairment and recovery in cerebrovascular pathologies related to Alzheimer's disease
.
Br. J. Pharmacol.
170
,
661
670
[PubMed]
59
Shefa
U.
,
Yeo
S.G.
,
Kim
M.S.
,
Song
I.O.
,
Jung
J.
,
Jeong
N.Y.
et al. 
(
2017
)
Role of gasotransmitters in oxidative stresses, neuroinflammation, and neuronal repair
.
Biomed. Res. Int.
2017, 1689341
[PubMed]
60
Iadecola
C.
,
Zhang
F.
,
Niwa
K.
,
Eckman
C.
,
Turner
S.K.
,
Fischer
E.
et al. 
(
1999
)
SOD1 rescues cerebral endothelial dysfunction in mice overexpressing amyloid precursor protein
.
Nat. Neurosci.
2
,
157
161
[PubMed]
61
Toth
P.
,
Tarantini
S.
,
Tucsek
Z.
,
Ashpole
N.M.
,
Sosnowska
D.
,
Gautam
T.
et al. 
(
2014
)
Resveratrol treatment rescues neurovascular coupling in aged mice: role of improved cerebromicrovascular endothelial function and downregulation of NADPH oxidase
.
Am. J. Physiol. Heart Circ. Physiol.
306
,
H299
H308
[PubMed]
62
Lucchinetti
C.
,
Bruck
W.
,
Parisi
J.
,
Scheithauer
B.
,
Rodriguez
M.
and
Lassmann
H.
(
2000
)
Heterogeneity of multiple sclerosis lesions: implications for the pathogenesis of demyelination
.
Ann. Neurol.
47
,
707
717
[PubMed]
63
Felts
P.A.
,
Woolston
A.M.
,
Fernando
H.B.
,
Asquith
S.
,
Gregson
N.A.
,
Mizzi
O.J.
et al. 
(
2005
)
Inflammation and primary demyelination induced by the intraspinal injection of lipopolysaccharide
.
Brain
128
,
1649
1666
[PubMed]
64
Marik
C.
,
Felts
P.A.
,
Bauer
J.
,
Lassmann
H.
and
Smith
K.J.
(
2007
)
Lesion genesis in a subset of patients with multiple sclerosis: a role for innate immunity?
Brain
130
,
2800
2815
[PubMed]
65
Sharma
R.
,
Fischer
M.T.
,
Bauer
J.
,
Felts
P.A.
,
Smith
K.J.
,
Misu
T.
et al. 
(
2010
)
Inflammation induced by innate immunity in the central nervous system leads to primary astrocyte dysfunction followed by demyelination
.
Acta Neuropathol.
120
,
223
236
[PubMed]
66
Smith
K.J.
,
Blakemore
W.F.
and
McDonald
W.I.
(
1979
)
Central remyelination restores secure conduction
.
Nature
280
,
395
396
[PubMed]
67
Smith
K.J.
and
Hall
S.M.
(
1994
)
Central demyelination induced in vivo by the calcium ionophore ionomycin
.
Brain
117
,
1351
1356
[PubMed]
68
Desai
R.A.
,
Davies
A.L.
,
Tachrount
M.
,
Kasti
M.
,
Laulund
F.
,
Golay
X.
et al. 
(
2016
)
Cause and prevention of demyelination in a model multiple sclerosis lesion
.
Ann. Neurol.
79
,
591
604
[PubMed]
69
Hynes
R.O.
and
Naba
A.
(
2012
)
Overview of the matrisome – an inventory of extracellular matrix constituents and functions
.
Cold Spring Harb. Perspect. Biol.
4
,
a004903
[PubMed]
70
Gould
D.B.
,
Phalan
F.C.
,
Breedveld
G.J.
,
van Mil
S.E.
,
Smith
R.S.
,
Schimenti
J.C.
et al. 
(
2005
)
Mutations in Col4a1 cause perinatal cerebral hemorrhage and porencephaly
.
Science
308
,
1167
1171
[PubMed]
71
Kuo
D.S.
,
Labelle-Dumais
C.
and
Gould
D.B.
(
2012
)
COL4A1 and COL4A2 mutations and disease: insights into pathogenic mechanisms and potential therapeutic targets
.
Hum. Mol. Genet.
21
,
R97
R110
[PubMed]
72
Verdura
E.
,
Hervé
D.
,
Bergametti
F.
,
Jacquet
C.
,
Morvan
T.
,
Prieto-Morin
C.
et al. 
(
2016
)
Disruption of a miR-29 binding site leading to COL4A1 upregulation causes pontine autosomal dominant microangiopathy with leukoencephalopathy
.
Ann. Neurol.
80
,
741
753
[PubMed]
73
Hara
K.
,
Shiga
A.
,
Fukutake
T.
,
Nozaki
H.
,
Miyashita
A.
,
Yokoseki
A.
et al. 
(
2009
)
Association of HTRA1 mutations and familial ischemic cerebral small-vessel disease
.
N. Engl. J. Med.
360
,
1729
1739
[PubMed]
74
Beaufort
N.
,
Scharrer
E.
,
Kremmer
E.
,
Lux
V.
,
Ehrmann
M.
,
Huber
R.
et al. 
(
2014
)
Cerebral small vessel disease-related protease HtrA1 processes latent TGF-β binding protein 1 and facilitates TGF-β signaling
.
Proc. Natl. Acad. Sci. USA
111
,
16496
16501
[PubMed]
75
Verdura
E.
,
Hervé
D.
,
Scharrer
E.
,
Amador
M.D.M.
,
Guyant-Maréchal
L.
,
Philippi
A.
et al. 
(
2015
)
Heterozygous HTRA1 mutations are associated with autosomal dominant cerebral small vessel disease
.
Brain
138
,
2347
2358
[PubMed]
76
Bugiani
M.
,
Kevelam
S.H.
,
Bakels
H.S.
,
Waisfisz
Q.
,
Ceuterick-de Groote
C.
,
Niessen
H.W.M.
et al. 
(
2016
)
Cathepsin A-related arteriopathy with strokes and leukoencephalopathy (CARASAL)
.
Neurology
87
,
1777
1786
[PubMed]
77
Monet-Leprêtre
M.
,
Haddad
I.
,
Baron-Menguy
C.
,
Fouillot-Panchal
M.
,
Riani
M.
,
Domenga-Denier
V.
et al. 
(
2013
)
Abnormal recruitment of extracellular matrix proteins by excess Notch3 ECD: a new pathomechanism in CADASIL
.
Brain
136
,
1830
1845
[PubMed]
78
Capone
C.
,
Cognat
E.
,
Ghezali
L.
,
Baron-Menguy
C.
,
Aubin
D.
,
Mesnard
L.
et al. 
(
2016
)
Reducing Timp3 or vitronectin ameliorates disease manifestations in CADASIL mice
.
Ann. Neurol.
79
,
387
403
[PubMed]
79
Capone
C.
,
Dabertrand
F.
,
Baron-Menguy
C.
,
Chalaris
A.
,
Ghezali
L.
,
Domenga-Denier
V.
et al. 
(
2016
)
Mechanistic insights into a TIMP3-sensitive pathway constitutively engaged in the regulation of cerebral hemodynamics
.
Elife
5
,
e17536
,
[PubMed]
80
Morris
A.W.
,
Sharp
M.M.
,
Albargothy
N.J.
,
Fernandes
R.
,
Hawkes
C.A.
,
Verma
A.
et al. 
(
2016
)
Vascular basement membranes as pathways for the passage of fluid into and out of the brain
.
Acta Neuropathol.
131
,
725
736
[PubMed]
81
Carare
R.O.
,
Bernardes-Silva
M.
,
Newman
T.A.
,
Page
A.M.
,
Nicoll
J.A.
,
Perry
V.H.
et al. 
(
2008
)
Solutes, but not cells, drain from the brain parenchyma along basement membranes of capillaries and arteries: significance for cerebral amyloid angiopathy and neuroimmunology
.
Neuropathol. Appl. Neurobiol.
34
,
131
144
82
Hawkes
C.A.
,
Hartig
W.
,
Kacza
J.
,
Schliebs
R.
,
Weller
R.O.
,
Nicoll
J.A.
et al. 
(
2011
)
Perivascular drainage of solutes is impaired in the ageing mouse brain and in the presence of cerebral amyloid angiopathy
.
Acta Neuropathol.
121
,
431
443
[PubMed]
83
Hughes
T.M.
,
Kuller
L.H.
,
Barinas-Mitchell
E.J.
,
Mackey
R.H.
,
McDade
E.M.
,
Klunk
W.E.
et al. 
(
2013
)
Pulse wave velocity is associated with beta-amyloid deposition in the brains of very elderly adults
.
Neurology
81
,
1711
1718
[PubMed]
84
Criswell
T.P.
,
Sharp
M.M.
,
Dobson
H.
,
Finucane
C.
,
Weller
R.O.
,
Verma
A.
et al. 
(
2017
)
The structure of the perivascular compartment in the old canine brain: a case study
.
Clin. Sci. (Lond.)
131
,
2737
2744
[PubMed]
85
Kalaria
RN.
(
2016
)
Neuropathological diagnosis of vascular cognitive impairment and vascular dementia with implications for Alzheimer's disease
.
Acta Neuropathol.
131
,
659
685
[PubMed]
86
Iliff
J.J.
,
Wang
M.
,
Liao
Y.
,
Plogg
B.A.
,
Peng
W.
,
Gundersen
G.A.
et al. 
(
2012
)
A paravascular pathway facilitates CSF flow through the brain parenchyma and the clearance of interstitial solutes, including amyloid beta
.
Sci. Transl. Med.
4
,
147ra111
[PubMed]
87
Nedergaard
M.
(
2013
)
Neurosci. Garbage truck of the brain
.
Science
340
,
1529
1530
[PubMed]
88
Louveau
A.
,
Smirnov
I.
,
Keyes
T.J.
,
Eccles
J.D.
,
Rouhani
S.J.
,
Peske
J.D.
et al. 
(
2015
)
Structural and functional features of central nervous system lymphatic vessels
.
Nature
523
,
337
341
[PubMed]
89
Aspelund
A.
,
Antila
S.
,
Proulx
S.T.
,
Karlsen
T.V.
,
Karaman
S.
,
Detmar
M.
et al. 
(
2015
)
A dural lymphatic vascular system that drains brain interstitial fluid and macromolecules
.
J. Exp. Med.
212
,
991
999
[PubMed]
90
Xie
L.
et al. 
(
2013
)
Sleep drives metabolite clearance from the adult brain
.
Science
342
,
373
377
[PubMed]
91
Lee
H.
,
Xie
L.
,
Yu
M.
,
Kang
H.
,
Feng
T.
,
Deane
R.
et al. 
(
2015
)
The effect of body posture on brain glymphatic transport
.
J. Neurosci.
35
,
11034
11044
[PubMed]
92
Kress
B.T.
,
Iliff
J.J.
,
Xia
M.
,
Wang
M.
,
Wei
H.S.
,
Zeppenfeld
D.
et al. 
(
2014
)
Impairment of paravascular clearance pathways in the aging brain
.
Ann. Neurol.
[PubMed]
93
Venkat
P.
,
Chopp
M.
,
Zacharek
A.
,
Cui
C.
,
Zhang
L.
,
Li
Q.
et al. 
(
2017
)
White matter damage and glymphatic dysfunction in a model of vascular dementia in rats with no prior vascular pathologies
.
Neurobiol. Aging
50
,
96
106
[PubMed]
94
Iliff
J.J.
,
Chen
M.J.
,
Plog
B.A.
,
Zeppenfeld
D.M.
,
Soltero
M.
,
Yang
L.
et al. 
(
2014
)
Impairment of glymphatic pathway function promotes tau pathology after traumatic brain injury
.
J. Neurosci.
34
,
16180
16193
[PubMed]
95
Jiang
Q.
,
Zhang
L.
,
Ding
G.
,
Davoodi-Bojd
E.
,
Li
Q.
,
Li
L.
et al. 
(
2016
)
Impairment of the glymphatic system after diabetes
.
J. Cereb. Blood Flow Metab.
,
37
,
1326
1337
,
96
Eide
P.K.
and
Ringstad
G.
(
2015
)
MRI with intrathecal MRI gadolinium contrast medium administration: a possible method to assess glymphatic function in human brain
.
Acta Radiol. Open
4
,
2058460115609635
[PubMed]
,
97
Kiviniemi
V.
,
Wang
X.
,
Korhonen
V.
,
Keinänen
T.
,
Tuovinen
T.
,
Autio
J.
et al. 
(
2015
)
Ultra-fast magnetic resonance encephalography of physiological brain activity – Glymphatic pulsation mechanisms?
J. Cereb. Blood Flow Metab.
[PubMed]
98
van der Meer
A.D.
,
Orlova
V.V.
,
ten Dijke
P.
,
van den Berg
A.
and
Mummery
C.L.
(
2013
)
Three-dimensional co-cultures of human endothelial cells and embryonic stem cell-derived pericytes inside a microfluidic device
.
Lab Chip
13
,
3562
3568
[PubMed]
99
Shi
Y.
,
Inoue
H.
,
Wu
J.C.
and
Yamanaka
S.
(
2017
)
Induced pluripotent stem cell technology: S decade of progress
.
Nat. Rev. Drug Discov.
16
,
115
130
[PubMed]
100
Lippmann
E.S.
,
Azarin
S.M.
,
Kay
J.E.
,
Nessler
R.A.
,
Wilson
H.K.
,
Al-Ahmad
A.
et al. 
(
2012
)
Human blood–brain barrier endothelial cells derived from pluripotent stem cells
.
Nat. Biotechnol.
30
,
783
791
101
Canfield
S.G.
,
Stebbins
M.J.
,
Morales
B.S.
,
Asai
S.W.
,
Vatine
G.D.
,
Svendsen
C.N.
et al. 
(
2017
)
An isogenic blood–brain barrier model comprising brain endothelial cells, astrocytes, and neurons derived from human induced pluripotent stem cells
.
J. Neurochem.
140
,
874
888
[PubMed]
102
Cheung
C.
,
Bernardo
A.S.
,
Trotter
M.W.
,
Pedersen
R.A.
and
Sinha
S.
(
2012
)
Generation of human vascular smooth muscle subtypes provides insight into embryological origin-dependent disease susceptibility
.
Nat. Biotechnol.
30
,
165
173
[PubMed]
103
Moffat
J.G.
,
Vincent
F.
,
Lee
J.A.
,
Eder
J.
and
Prunotto
M.
(
2017
)
Opportunities and challenges in phenotypic drug discovery: an industry perspective
.
Nat. Rev. Drug Discov.
16
,
531
543
[PubMed]
104
Sahakian
B.J.
and
Owen
A.M.
(
1992
)
Computerized assessment in neuropsychiatry using CANTAB: discussion paper
.
J. R. Soc. Med.
85
,
399
402
[PubMed]
105
Levaux
M.-N.
,
Potvin
S.
,
Sepehry
A.A.
,
Sablier
J.
,
Mendrek
A.
,
Stip
E.
(
2007
)
Computerized assessment of cognition in schizophrenia: promises and pitfalls of CANTAB
.
Eur. Psychiatry
22
,
104
115
[PubMed]
106
Barnett
J.H.
,
Robbins
T.W.
,
Leeson
V.C.
,
Sahakian
B.J.
,
Joyce
E.M.
,
Blackwell
A.D.
(
2010
)
Assessing cognitive function in clinical trials of schizophrenia
.
Neurosci. Biobehav. Rev.
34
,
1161
1177
107
Crabbe
J.C.
(
1999
)
Genetics of mouse behavior: interactions with laboratory environment
.
Science
284
,
1670
1672
[PubMed]
108
Romberg
C.
,
Mattson
M.P.
,
Mughal
M.R.
,
Bussey
T.J.
and
Saksida
L.M.
(
2011
)
Impaired attention in the 3xTgAD mouse model of Alzheimer's disease: rescue by donepezil (Aricept)
.
J. Neurosci.
31
,
3500
3507
[PubMed]
109
Nithianantharajah
J.
and
Grant
S.G.N.
(
2013
)
Cognitive components in mice and humans: combining genetics and touchscreens for medical translation
.
Neurobiol. Learn Mem.
[PubMed]
110
Nithianantharajah
J.
,
McKechanie
A.J.
,
Stewart
T.J.
,
Johnstone
M.
,
Blackwood
D.H.
,
St Clair
D.
et al. 
(
2015
)
Bridging the translational divide: identical cognitive touchscreen testing in mice and humans carrying mutations in a disease-relevant homologous gene
.
Sci. Rep.
5
,
14613
,
111
Talpos
J.
and
Steckler
T.
(
2013
)
Touching on translation
.
Cell Tissue Res.
354
,
297
308
[PubMed]
112
Crawley
J.N.
(
2007
)
What's Wrong With My Mouse?
,
John Wiley & Sons
113
Mar
A.C.
,
Horner
A.E.
,
Nilsson
S.R.
,
Alsiö
J.
,
Kent
B.A.
,
Kim
C.H.
et al. 
(
2013
)
The touchscreen operant platform for assessing executive function in rats and mice
.
Nat. Protoc.
8
,
1985
2005
[PubMed]
114
Horner
A.E.
,
Heath
C.J.
,
Hvoslef-Eide
M.
,
Kent
B.A.
,
Kim
C.H.
,
Nilsson
S.R.
et al. 
(
2013
)
The touchscreen operant platform for testing learning and memory in rats and mice
.
Nat. Protoc.
8
,
1961
1984
[PubMed]
115
Oomen
C.A.
,
Hvoslef-Eide
M.
,
Heath
C.J.
,
Mar
A.C.
,
Horner
A.E.
,
Bussey
T.J
et al. 
(
2013
)
The touchscreen operant platform for testing working memory and pattern separation in rats and mice
.
Nat. Protoc.
8
,
2006
2021
[PubMed]
116
Bahor
Z.
,
Liao
J.
,
Macleod
M.R.
,
Bannach-Brown
A.
,
McCann
S.K.
,
Wever
K.E.
et al. 
(
2017
)
Risk of bias reporting in the recent animal focal cerebral ischaemia literature
.
Clin. Sci. (Lond.)
131
,
2525
2532
[PubMed]
117
Sena
E.
,
van der Worp
H.B.
,
Howells
D.
and
Macleod
M.R.
(
2007
)
How can we improve the pre-clinical development of drugs for stroke?
Trends Neurosci.
30
,
433
439
[PubMed]
118
Pedder
H.
,
Vesterinen
H.V.
,
Macleod
M.R.
and
Wardlaw
J.M.
(
2014
)
Systematic review and meta-analysis of interventions tested in animal models of lacunar stroke
.
Stroke
45
,
563
570
[PubMed]
119
Stroke therapy academic industry roundtable (Fisher M, Chair)
(
1999
)
Recommendations for standards regarding preclinical neuroprotective and restorative drug development
.
Stroke
30
,
2752
2758
[PubMed]
120
Fisher
M.
,
Feuerstein
G.
,
Howells
D.W.
,
Hurn
P.D.
,
Kent
T.A.
,
Savitz
S.I.
et al. 
(
2009
)
Update of the stroke therapy academic industry roundtable preclinical recommendations
.
Stroke
40
,
2244
2250
[PubMed]
121
Kilkenny
C.
,
Browne
W.J.
,
Cuthill
I.C.
,
Emerson
M.
and
Altman
D.G.
(
2010
)
Improving bioscience research reporting: the ARRIVE guidelines for reporting animal research
.
PLoS Biol.
8
,
e1000412
[PubMed]
122
RIGOR
(
2012
)
Improving the quality of NINDS-supported preclinical and Clinical Research through rigorous study design and transparent reporting.
123
Percie du Sert
N.
,
Alfieri
A.
,
Allan
S.M.
,
Carswell
H.V.
,
Deuchar
G.A.
,
Farr
T.D.
et al. 
(
2017
)
The IMPROVE Guidelines (ischaemia models: procedural refinements of in vivo experiments)
.
J. Cereb. Blood Flow Metab.
37
,
3488
3517
[PubMed]
124
Bath
P.M.
,
Macleod
M.R.
and
Green
A.R.
(
2009
)
Emulating multicentre clinical stroke trials: a new paradigm for studying novel interventions in experimental models of stroke
.
Int. J. Stroke
4
,
471
479
[PubMed]
125
Llovera
G.
,
Hofmann
K.
,
Roth
S.
,
Salas-Perdoma
A.
,
Ferrer-Ferrer
M.
,
Perego
C.
et al. 
(
2015
)
Results of a preclinical randomized controlled multicentre trial (prct): anti-cd49d treatment for acute brain ischemia
.
Sci. Transl. Med.
7
,
299ra121
[PubMed]
126
Maysami
S.
,
Wong
R.
,
Pradillo
J.
,
Denes
A.
,
Dhungana
H.
,
Malm
T.
et al. 
(
2016
)
A cross-laboratory preclinical study on the effectiveness of interleukin-1 receptor antagonist in stroke
.
J. Cereb. Blood Flow Metab.
36
,
596
605
[PubMed]
127
Noel-Storr
A.H.
,
Flicker
L.
,
Ritchie
C.W.
,
Nguyen
G.H.
,
Gupta
T.
,
Wood
P.
et al. 
(
2013
)
Systematic review of the body of evidence for the use of biomarkers in the diagnosis of dementia
.
Alzheimers Dement.
9
,
e96
e105
[PubMed]
128
Ritchie
C.
,
Smailagic
N.
,
Noel-Storr
A.H.
,
Ukoumunne
O.
,
Ladds
E.C.
and
Martin
S.
(
2017
)
CSF tau and the CSF tau/ABeta ratio for the diagnosis of Alzheimer's disease dementia and other dementias in people with mild cognitive impairment (MCI)
.
Cochrane Database Syst. Rev.
3
,
CD010803
[PubMed]
129
Wardlaw
J.M.
,
Smith
E.E.
,
Biessels
G.J.
,
Cordonnier
C.
,
Fazekas
F.
,
Frayne
R.
et al. 
(
2013
)
Neuroimaging standards for research into small vessel disease and its contribution to ageing and neurodegeneration
.
Lancet Neurol.
12
,
822
838
[PubMed]
130
Lawrence
A.J.
,
Chung
A.W.
,
Morris
R.G.
,
Markus
H.S.
and
Barrick
T.R.
(
2014
)
Structural network efficiency is associated with cognitive impairment in small-vessel disease
.
Neurology
83
,
304
311
[PubMed]
131
Boehm-Sturm
P.
,
Füchtemeier
M.
,
Foddis
M.
,
Mueller
S.
,
Trueman
R.C.
,
Zille
M.
et al. 
(
2017
)
Neuroimaging biomarkers predict brain structural connectivity change in a mouse model of vascular cognitive impairment
.
Stroke
48
,
468
475
[PubMed]
132
Benjamin
P.
,
Zeestraten
E.
,
Lambert
C.
,
Ster
I.C.
,
Williams
O.A.
,
Lawrence
A.J.
et al. 
(
2016
)
Progression of MRI markers in cerebral small vessel disease: sample size considerations for clinical trials
.
J. Cereb. Blood Flow Metab.
36
,
228
240
[PubMed]
133
Quinn
T.J.
and
McCleery
J.
(
2017
)
Diagnosis in vascular dementia, applying 'Cochrane diagnosis rules' to 'dementia diagnostic tools'
.
Clin. Sci. (Lond.)
131
,
729
732
[PubMed]

Author notes

This article is based on the proceedings of a workshop event that was held at the British Heart Foundation Centre of Research Excellence in Glasgow in 25 and 26 January 2017.