Arteriolar neuropathology in cerebral microvascular disease

Abstract Cerebral microvascular disease (MVD) is an important cause of vascular cognitive impairment. MVD is heterogeneous in aetiology, ranging from universal ageing to the sporadic (hypertension, sporadic cerebral amyloid angiopathy [CAA] and chronic kidney disease) and the genetic (e.g., familial CAA, cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy [CADASIL] and cerebral autosomal recessive arteriopathy with subcortical infarcts and leukoencephalopathy [CARASIL]). The brain parenchymal consequences of MVD predominantly consist of lacunar infarcts (lacunes), microinfarcts, white matter disease of ageing and microhaemorrhages. MVD is characterised by substantial arteriolar neuropathology involving ubiquitous vascular smooth muscle cell (SMC) abnormalities. Cerebral MVD is characterised by a wide variety of arteriolar injuries but only a limited number of parenchymal manifestations. We reason that the cerebral arteriole plays a dominant role in the pathogenesis of each type of MVD. Perturbations in signalling and function (i.e., changes in proliferation, apoptosis, phenotypic switch and migration of SMC) are prominent in the pathogenesis of cerebral MVD, making ‘cerebral angiomyopathy’ an appropriate term to describe the spectrum of pathologic abnormalities. The evidence suggests that the cerebral arteriole acts as both source and mediator of parenchymal injury in MVD.


INTRODUCTION
Cerebral microvascular disease (MVD) refers to a group of conditions that affect small vessels of the brain; they are heterogeneous in their aetiology, and with variable, though often profound, consequences for brain function. MVD is a major public health issue and is an important cause of vascular cognitive impairment [1,2]. Despite its high prevalence, there remain critical unanswered questions regarding its pathogenesis that have resulted in suboptimal treatment efforts and preventive strategies.
The range of MVD entities is well-described. It is encountered universally in ageing and arises sporadically such as in hypertension,  (CARASAL). While this range is substantial, the parenchymal consequences of MVD are limited and consist of lacunar infarcts (lacunes), variably defined as grossly visible cystic lesions less than 1 to 2 cm in greatest diameter [2]; microinfarcts, ischaemic lesions visible only on microscopic examination [2]; white matter disease of ageing [2,3]; and microhaemorrhages [4,5]. Our review addresses the incongruity of MVD, given the wide range of provoking factors versus the relatively limited range of neuropathologic manifestations that affect the brain. It is argued that the cerebral arteriole plays a dominant role in MVD pathogenesis. On one hand, there exists a range of arteriolar pathologic changes that are relatively distinct for each entity. At the same time, these distinctive changes substantially reflect abnormalities of the arteriolar wall characterised by alterations of arteriolar smooth muscle cells (SMC) and lead to occlusion or rupture of cerebral blood vessels. The term cerebral angiomyopathy has been proposed to describe this phenomenon [6].
The parenchymal consequences of these heterogeneous arteriolar changes are limited, suggesting that the abnormalities of cerebral arterioles function as a final common pathway for these structural changes.

CEREBRAL ARTERIOLAR NETWORK AND COMPOSITION
The arteriolar alterations in MVD may lead to common parenchymal consequences, including lacunar infarcts (lacunes), microinfarcts, white matter disease/demyelination and microhaemorrhages. Cerebral arterioles are particularly important given their role as the major site of vascular resistance and primary regulators of cerebral blood flow [7].
The cerebral arteriolar network is divided into three major components: (1) pial arterioles on the surface of the brain; (2) parenchymal/ penetrating arterioles that branch out from pial arterioles to enter the brain parenchyma or arise from deep penetrating arteries in the basal ganglia, thalamus and brainstem; and (3) downstream pre-capillary arterioles [8,9]. The walls of cerebral arterioles, which typically have a diameter in the range of 10-100 μm [10], consist of three layers: tunica intima, tunica media and tunica adventitia [11]. The tunica intima is composed of a single layer of endothelial cells and the basal lamina surrounding endothelium. Endothelial cells in the intima are vital in the regulation of vascular tone by releasing vasoactive factors such as nitric oxide, prostacyclin, thromboxane and endothelin-1 [12], which regulate the contractile state of SMC and ultimately vascular diameter. The tunica media in arterioles consists of one or two complete layers of square-or rectangular-shaped SMC with surrounding elastin and collagen fibres [13]. In larger arterioles with a diameter less than 100 μm, but larger than 40 μm [14], the media may be separated from the innermost intima by the internal elastic lamina. The SMC in the media contract to change the vascular diameter and regulate blood flow between the arteries and capillaries. The tunica adventitia is comprised primarily of collagen fibres and fibroblasts; it provides structural support for the vascular wall and is involved in repair of the vessel wall following injury [15].
Given the complex structure of arterioles and the critical role of each component, the correct identification of cerebral arterioles is of great importance in understanding the pathology of MVD and the development of targeted preventative and treatment strategies. Traditionally, arterioles are distinguished from veins/venules on the basis of smooth muscle content, the shape of the vessel and the presence/ absence of internal elastic lamina. Unlike arteries/arterioles, veins and venules normally have minimal SMC and do not have an elastic lamina [16]. However, small arterioles and venules in the brain parenchyma have similar intramural cell types, that is, SMC and pericytes in the media; both SMC and pericytes contain variable amounts of smooth muscle actin. Pericytes are embedded within the basement membrane of capillaries and play a vital role in blood-brain barrier (BBB) maintenance and cerebral blood flow control. Increase in pericyte numbers in the walls of arterioles of patients with CADASIL is linked to SMC degeneration and vessel wall thickening [17]. Beyond that, the substantial overlap in lumen diameter to wall thickness ratio between arterioles and venules challenges traditional methods that use light microscopy to distinguish arterioles from venules; the sclerotic index is a quantified assessment that may assist with this distinction [18][19][20].

SMOOTH MUSCLE CELLS (SMC) IN VASCULAR REMODELLING
Current understanding of the pathogenic mechanisms underlying MVD is limited due to the challenges in visualising the diseased small vessels via radiographic imaging in vivo. SMC integrity is necessary for cerebrovascular health owing to their principal role in vascular wall contraction and remodelling, involving the phenotypic switch of SMC

Key points
• Cerebral microvascular disease (MVD) has a wide range of aetiologies ranging from universal (ageing) to sporadic and genetic.
• Abnormalities in arterioles are the final common pathway in producing the observable brain parenchymal consequences.
• The term cerebral angiomyopathy describes this spectrum of pathologic abnormalities of the arteriolar wall.
from the quiescent, contractile phenotype to the proliferative, migratory phenotype. Perturbations in SMC signalling and function are suggested to underlie the pathogenesis of cerebral microangiopathies, in particular hypertensive microangiopathy, CAA, CADASIL [21], Fabry Disease [22] and CRV/RVCL [23]. The function of other vascular cell components, for example, endothelial cells, can also be compromised with ageing, hypertension and other risk factors [24]. Important functions of SMC in diseased conditions include apoptosis, phenotypic switch, extracellular matrix degradation, proliferation and contractility.
Inflammatory cell infiltration and genetic changes may modulate SMC functions and involve changes in growth factor signalling and regulation of RNA expression. One of the key factors that has been highlighted recurrently in recent years is transforming growth factor-β (TGF-β), with particular reference to hereditary MVD of the brain [25]. TGF-β isoforms are upregulated and activated in vascular diseases and have an important role in muscle repair and remodelling, as well as regulation and function of myocytes, fibroblasts, immune cells and other vascular cells. Although dysregulation of TGF-β is not primary in all MVD, nor is it the only molecular mechanism underlying MVD arteriolosclerosis, the contribution of TGF-β and vascular basement membrane disruption in the pathogenesis of MVD is clear [25].
The morphologic changes in the vasculature, such as vessel wall disruption and fibrosis, may be highlighted using special stains and immunohistochemistry. These include Verhoeff-Van Gieson staining (for elastin), Masson's trichrome staining (for collagen) [26], α-smooth muscle actin immunohistochemistry (for SMC), Prussian blue (for iron, including haemosiderin) [18], periodic acid-Schiff (PAS) stain (suggesting the presence of granular material within vessel walls and highlighting basement membrane components) [27] and Richardson's staining (for globotriaosylceramide-3) [28]. These histologic and classical tinctorial stains such as PAS and Oil-Red-O enable one to identify some common features of the changes occurring in arterial walls in MVD.

AGEING
Ageing is associated with brain atrophy and lesions such as lacunes, white matter hyperintensities (WMH) and cerebral microbleeds (CMB) as evident on magnetic resonance imaging (MRI) [29]. Ageing is considered an independent risk factor for vascular dysfunction. The bestdescribed arteriolar changes with ageing, primarily in pial arterioles, include a significant decrease in arteriolar number [7] and increase in arteriolar diameter [30]. Increasing age is associated with medial elastin loss with replacement by collagen, gradual intimal thickening evident by a decrease in lumen diameter and an increase in intima-tomedia thickness and adventitial fibrosis. These age-related structural changes contribute to mechanical alterations such as reduced compliance and elasticity and increased arterial stiffness [24,30]. Recent studies in mice suggest ageing is associated with increased wall thickness but reduced wall stress in parenchymal arterioles, which is a potential mechanism to protect these arterioles from vascular injury.
Loss of myogenic tone in large arteries with age may increase the risk of rupture of parenchymal arterioles with blood pressure fluctuations [31]. However, ageing is not associated with changes in distensibility or lumen diameter in parenchymal arterioles [31,32]. Impairment of vascular structure and function with ageing is largely driven by multiple mechanisms involving changes in SMC number, either by affecting their proliferation or by cell death, predominantly by apoptosis. A large number of in vitro [33][34][35][36][37][38][39] and in vivo [34] studies have demonstrated that ageing is associated with increased SMC proliferation, as evidenced by increased expression of stem cell markers [33] and cell cycle activation markers [34] in SMC from humans and rodents. However, conflicting results have shown a loss of proliferation of SMC isolated from aged human donors [40,41] and rodent models [42,43]. The proliferation rate of SMC in each passage in culture is dependent on donor age, with SMC from aged donors more likely to be senescent with impaired proliferative capacity [40]. Decrease in SMC number with advanced age can also be a result of increased apoptosis [34,43].
Vascular SMC have extended life expectancy, and the effect of age on SMC is complex. SMC may initially undergo hyperproliferation early in the ageing process. With advanced age, SMC gradually exhibit cellular senescence, characterised by impaired proliferative capacity, irreversible growth arrest and apoptosis; this contributes to vascular inflammation, loss of arterial function and development of age-related disease [44]. In addition, differences in animal models, experimental conditions and patient comorbidities (among other factors) are likely to underlie the conflicting results in studies of age-related changes in SMC.
Another underlying mechanism relates to phenotypic alterations of SMC. SMC are normally quiescent and contractile to maintain vascular tone. With ageing, they adopt a stiff and pro-migratory phenotype. Aged SMC have an accelerated cell cycle and increased reactive oxygen species (ROS) production compared with their younger counterparts. Aged SMC produce more matrix metalloproteinases (MMP) that promote SMC migration from the media to the intima by detaching cells from the extracellular matrix (ECM); this process contributes to elastin fragmentation and loss with replacement by collagen and resultant arteriolar wall stiffening [24].
Together, these changes in SMC (increase or decrease in proliferation and migration) are key events in ageing that lead to vessel wall thickening and stiffening and to vascular dysfunction that may contribute to age-related MVD. It should be noted that age-related CAA may lead to increased fragility of the vessel wall rather than stiffening, making the vessels prone to bleeding [45]. Animal models allow us to study ageing as an independent risk factor; in humans, age-related vascular remodelling increases the risk of MVD when other risk factors are present, in particular hypertension. Hence, it is very difficult to isolate the effects of ageing from hypertension on the cerebral arterioles.

HYPERTENSION
Hypertension is a major risk factor for the development of ischaemic stroke, intracerebral haemorrhage (ICH) and dementia. It is associated with brain atrophy, microinfarcts and microhaemorrhages ( Figure 1A), which are important factors in the degree and temporal evolution of cognitive impairment. Hypertension has an especially profound effect on parenchymal/penetrating arterioles and pial arterioles [9]. Parenchymal/penetrating arterioles are particularly significant in hypertension because of their limited collateral circulation, and their dysfunction is likely to cause insufficient blood supply and damage to the deep white matter and grey nuclei [9].
The hallmark of the vascular pathology of hypertension is arteriolosclerosis, which causes thickening of the arteriolar wall with SMC degeneration and loss [19,46]. Obstruction of the lumen may cause impaired blood flow leading to oligaemia (hypoperfusive state), lacunar infarcts, microinfarcts and WMH. The adventitial layer undergoes ECM remodelling, leading to collagen deposition and thus arteriolar fibrosis [11]. The severity of arteriolosclerosis is closely related to age and can be exacerbated by hypertension [47]. In hypertension, the affected vessel wall (typically less than 150 μm in diameter) can develop a 'glassy' hyalinized and/or an 'onion-skinning' hyperplastic appearance as a result of degenerated SMC and elastin in the media and proliferated fibroblasts in the adventitia [2] (Figure 1B-E). Fibrinoid necrosis, due to the infiltration of fibrin and fibrin degradation products in the vessel wall, is more common in malignant hypertension and may be seen with severe CAA. Fibrinoid necrosis in subcortical penetrating arterioles causing weakening of the arteriolar wall is likely an underlying cause of ICH in patients with severe hypertension. In patients with severe CAA, it is probably a major contributing factor to cerebral haemorrhage. Hyalinosis and fibrinoid necrosis can be distinguished from each other as hyaline material contains only degenerated SMC and collagen and no fibrin is present [48]. (To avoid confusion, we do not use the term 'lipohyalinosis' [48].) While the exact cellular rearrangements are unclear, gradual hypertension-related changes in the arteriolar wall lie within a spectrum and depend on the severity of other perivascular influences.
Other pathological consequences of hypertension include microatheroma (distal manifestations of atherosclerosis involving larger arterioles) and microaneurysms (segmental dilatation of vessels) [9,49]. The proliferation of mural cells may exert secondary effects that raise vascular pressure and promote SMC degeneration in the upstream feeding arterioles, thereby inducing arteriolar wall rupture and subsequent haemorrhage [50]. CAA is a common MVD associated with Alzheimer's disease [51,52] and is often observed in the brains of the elderly even in the absence of Alzheimer's disease. CAA is characterised by the accumulation of amyloid-β (Aβ) in the walls of small-to-medium-sized arteries and arterioles predominantly located in the leptomeninges and cerebral cortex; these are prone to bleeding due to the replacement of the medial SMC with Aβ and fragility of the affected arteries/arterioles [53]. CAA is an important cause of spontaneous primary lobar haemorrhage in elderly individuals and is also associated with microinfarcts [53,54], microhaemorrhages and superficial siderosis [55]. The spatial distribution of cerebral microhaemorrhages in the brain correlates well with the anatomic distribution of affected small vessels. CMB located in cortical and subcortical regions is considered a surrogate marker of CAA on imaging, and cortical CMB are particularly associated with severe CAA [56]. Figure 2 demonstrates an amyloidladen penetrating arteriole with surrounding haemosiderin-laden macrophages in the parenchyma, consistent with prior haemorrhage. CAA preferentially affects arterioles as they are postulated to be the main routes for perivascular Aβ clearance [57].
Post-mortem examination of human tissue [58] and several transgenic mouse models [59,60] of CAA overexpressing mutant human Aβ precursor protein (APP) have shown sequential arteriolar wall changes. The endothelial layer is relatively preserved, whereas the SMC layer is increasingly disrupted with progressing CAA severity.
Soluble Aβ-induced functional abnormalities such as failure to respond to vasoactive stimuli are early manifestations despite little or no CAA vascular pathology [61]. In the early stages or mild CAA, Aβ is deposited in the external basal lamina in close proximity to the SMC layer and in the adventitia, but with no SMC replacement or disruption [62]. As the disease progresses, Aβ is deposited in the SMC layer, which is disrupted, and the internal basal lamina appears thinner and irregular [61,62]; this is followed by SMC loss and complete replacement by Aβ [59,60]. Advanced CAA is also associated with the formation of microaneurysms and fibrinoid necrosis in Aβ-laden vessels; these are often associated with cerebral lobar haemorrhage [62].
The vascular origin of microhaemorrhages in the context of CAA is complex and remains poorly understood. There are fewer arterioles with Aβ deposition immediately surrounding microhaemorrhages, compared with areas with microinfarcts [63]. Aβ is more frequently observed upstream or downstream from the site of rupture, indicating that microhaemorrhages may occur at a later time point of the disease [63]. Arterioles associated with microhaemorrhages are more likely to be degenerated, allowing extravasation of red blood cells. In regions of microinfarcts, the arterioles appear to be more intact, suggestive of a perfusionmediated phenomenon resulting in infarcts [63]. The spatial relationship between microhaemorrhages and vascular Aβ deposition has been assessed in CAA mice and their wild-type littermates. Rather than originating from CAA-laden vessels, vascular segments without Aβ deposition appeared vulnerable to progressive vessel wall weakening and more prone to bleeding. Leakage sites were most likely to be branch points of penetrating arterioles, but capillaries (BBB leakage) could not be excluded [64].
Familial CAA is rare, generally more severe than sporadic CAA and associated with an earlier age of onset and/or death. Familial CAA is commonly present in the form of autosomal dominant disorders, characterised by mutations in the APP [52], cystatin C3 (CST3) [65] or integral membrane protein 2B [66] genes, leading to the aggregation of Aβ in blood vessel walls. It has been demonstrated that the pathology of familial CAA is age-related and results in both cerebral infarcts and ICH. Beyond leptomeningeal and cortical arteries and arterioles, the affected brain regions in familial CAA are thought to be more extensive and involve the cerebellum and brainstem. In rare cases of patients with severe CAA, profound Aβ deposition is seen in the walls of leptomeningeal vessels but less prominent in parenchymal vessels. The associated neuropathological change features cerebral infarcts rather than lobar haemorrhage as a result of wall thickening and occlusion of large-sized leptomeningeal arteries [53].

CHRONIC KIDNEY DISEASE (CKD)
The kidney and brain share some anatomical and functional similarities; both are characterised by high blood flow rates and dependence on local autoregulation through short, small perforating arterioles. Therefore, the mechanisms underlying the microvascular damage are thought to be similar. Deterioration of these arterioles may markedly accelerate the progression of both renal and cerebral dysfunction. In the kidney, these arterioles, primarily juxtamedullary afferent arterioles, are particularly susceptible to hypertensive injury characterised by hyaline arteriolosclerosis with the replacement of arteriolar SMC by hyaline material.
Hyalinosis of deep penetrating arterioles in the brain is associated with lacunar infarction and deep white matter changes and can cause and elastin degradation in the medial layer [67,70]. Elevated levels of other minerals in CKD (e.g., calcium) also have synergistic effects on inducing arterial calcification via distinct mechanisms affecting SMC [71].
Arterial (intimal and medial) calcification has been associated with CKD in ischaemic stroke patients [72], as well as in young adults undergoing dialysis [73]. Of note, brain microvascular calcification is frequently seen in CKD patients lacking neurologic deficits, for example, in the basal ganglia (arterioles and capillaries) and hippocampal endplate region (capillaries), and (in a rare case) deep cerebellar white matter (arterioles) [69]. However, the prevalence of CKD-related microvascular calcification is likely to be underestimated in routine brain sections. Other studies using thorough sampling and examination of basal ganglia and hippocampal regions have shown that microvascular calcification is commonly seen in the brains of aged individuals [74] as well as patients with familial CAA [75,76]; these findings indicate that microvascular calcification is common in MVD and may be clinically significant. One study proposed that adventitial mesenchymal stem cell-like cells are progenitors of SMC and responsible for driving arterial calcification in CKD [77]. The evidence suggests that calcium and phosphorus metabolism within microvessels plays a key role in vascular abnormalities in CKD-related MVD, which warrants further investigation in both preclinical and clinical studies.

CADASIL
CADASIL is an MVD caused by mutations in the NOTCH3 gene on chromosome 19. CADASIL is the most common genetic cause of stroke. It affects a wide range of age groups, with an early onset in the late 30s [78]. Neuropathological features of CADASIL consist of lacunar infarcts within subcortical white matter, deposition of granular osmiophilic material (GOM) in the media and adventitia of arterioles in both white and grey matter (but especially in the white matter), fibrosis and stenosis of small penetrating arterioles and cerebral microhaemorrhages [79][80][81][82]. GOM can also be found in cutaneous arterioles, a finding which has been used as a diagnostic test for CADASIL; this has largely been superseded by genetic testing [83]. The major components of GOM include the NOTCH3 ectodomain and extracellular matrix proteins. GOM deposition can progress over time, exhibiting alterations in number, size and morphology [84].  [82,93].
Medial SMC loss is suggested to be the primary mechanism underlying the pathogenesis of CARASIL. First, medial SMC loss is widespread in affected cerebral arteries/arterioles of CARASIL patients, regardless of the presence of sclerotic changes [86,93]. Second, medial SMC loss may cause myelin damage and ischaemic changes via impaired autoregulation rather than luminal stenosis and adventitial fibrosis [93,94]. A study combining a mouse model expressing the L364P mutant of the human HTRA1 gene and primary cell culture suggests that CARASIL induces SMC loss by activating apoptosis signalling [95].

FABRY DISEASE
Fabry disease is an X-linked hereditary disorder characterised by the accumulation of globotriaosylceramide-3 (GL-3) in lysosomes, as a result of a mutation in the GLA gene leading to absent or deficient α-galactosidase A enzyme activity [22]. Ischaemic stroke and transient ischaemic attacks are the most common CNS manifestations in affected patients [96,97]. Growing evidence suggests that cerebral MVD is the predominant neuropathology in patients with Fabry disease; WMH are the primary feature, followed by CMB and lacunes [98]. Cerebrovascular changes seen in neuroimaging studies include multifocal lesions in the subcortical, deep and/or periventricular white matter and the subcortical, deep grey matter symmetrically in both cerebral hemispheres.
Ageing is the primary risk factor that affects the lesion load and pattern of distribution; these lesions may precede the onset of neurological symptoms. A higher white matter lesion load is associated with progression of other cerebrovascular abnormalities. Vascular changes include medial thickening and intimal and adventitial fibrosis [99]. Histological  [100]. In vitro studies show that exposure of SMC to GL-3 at concentrations observed in the plasma of Fabry patients induce SMC proliferation [101]. A potent proliferative factor sphingosine-1 phosphate (S1P) in plasma has been identified to be partially responsible for vascular remodelling in this disease. Specifically, higher plasma levels of S1P are seen in Fabry patients compared with healthy controls. Treatment with S1P increase IM thickness in murine aortas and induce SMC proliferation in a dose-dependent manner [102]. SMC changes in Fabry disease are also associated with increased levels of oxidative stress [103], which may further exacerbate endothelial dysfunction.
High circulating levels of MMP (especially MMP-9) in Fabry disease [104] are linked to deleterious effects on the cerebral arterioles, potentially causing internal elastic lamina degradation and migration of SMC from the media to the intimal layer [105]. Ultrastructural examination reveals a thick, multi-laminated basement membrane within larger vessels, or a single lamina densa within arterioles and capillaries [23]. Other histopathological features include vascular wall thickening and hyalinization, luminal narrowing, adventitial fibrosis and in some cases fibrinoid necrosis [23,107,108]. Occasionally, inflammatory cells (CD68 + and CD45 + ) can be seen surrounding vessels with intact SMC and endothelium and less fibrosis, whereas fewer SMC and inflammatory cells are observed in the regions bordering ischaemia [23]. A progressive loss of small blood vessels has been seen, but the exact effect of TREX1 mutation on SMC, and the role of SMC in CRV/RVCL-related vasculopathy, remains largely unknown. One study identified microRNA (miR)-103 as a potent regulator of vascular apoptosis, oxidative stress and angiogenesis via targeting TREX1 in vitro, and its expression was also upregulated in stressed human SMC [109]. This evidence suggests a possible alteration in SMC number in the presence of TREX1, which needs to be confirmed in both preclinical and clinical investigations.
Our histological examinations of COL4 arteriopathy in patients with COL4A1 mutation (c.*32G>A) reveal differential changes in arteriolar walls with prominent SMC loss ( Figure 8). Vascular basement membrane composed of collagen IV, laminin and heparan sulphate proteoglycans forms a three-dimensional protein network to support interactions between SMC with other cellular components [121].
COL4A1/COL4A2 mutations exert effects on SMC behaviour via multiple mechanisms, including inducing SMC apoptosis [117,119], impairing SMC differentiation and maintenance via disruption of vascular basement membrane [118] and promoting the production of a pathogenic, synthetic phenotype of SMC via binding to cellular receptors [120]; these can all cause detrimental effects on vessel wall integrity and function.

CARASAL
In addition to the classic genetic factors (i.e., NOTCH 3, HTRA1, GLA, TREX1 and COLA1/A2), it has been shown that mutations in the CTSA gene encoding cathepsin A can cause a novel MVD, namely CARA-SAL. This can present clinically with ischaemic stroke, cognitive impairment and therapy-resistant hypertension [122,123]. Neuropathological examination shows mild white matter atrophy and small infarcts dispersed widely in the brain, including white matter, deep grey matter, brainstem and cerebellum. This is accompanied by changes in cerebral arterioles including asymmetric fibrous thickening, loss and degeneration of medial SMC, varying degrees of luminal narrowing from stenosis in small arterioles to total occlusion in large arterioles and enlarged adventitia with deposition of collagen fibrils [124].
Homozygous CTSA mutation is thought to cause CARASAL by inter-fering with the function of cathepsin A, which protects against a systemic lysosomal storage disorder by stabilising the lysosomal enzymes β-galactosidase and neuraminidase [124]. Heterozygous CTSA mutation might alter cathepsin A activity in inactivating endothelin-1, therefore impairing blood pressure regulation [124].

CONCLUSIONS
There are many cerebral MVD entities but only a limited range of parenchymal consequences of these disorders. Cerebral MVD has both ischaemic and haemorrhagic components with neuropathology that ranges from lacunar infarcts, microinfarcts and white matter disease to microhaemorrhages. MVD is consistently characterised by substantial arteriolar remodelling (cerebral angiomyopathy) involving alterations in SMC via proliferation, apoptosis, phenotypic switch and/or migration with resultant changes in vessel wall components, diameter and thickness (Table 1). Taken together, these elements strongly support the arteriole acting as both source and mediator of parenchymal injury. Thus, it is the arteriole that is the critical component and final common pathway of cerebral MVD.

ACKNOWLEDGEMENTS
The author(s) disclosed receipt of the following financial support for

CONFLICT OF INTEREST
The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

DATA AVAILABILITY STATEMENT
Not applicable.