Microvascular pathology in the spinal cord of severe spinal muscular atrophy patients

doi:10.21203/rs.3.rs-8085055/v1

Download PDF

Research Article

Microvascular pathology in the spinal cord of severe spinal muscular atrophy patients

https://doi.org/10.21203/rs.3.rs-8085055/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Severe spinal muscular atrophy (SMA) is a life-limiting neurodegenerative disease of infancy and early childhood, caused by reduced expression of the ubiquitous survival motor neuron protein (SMN). While current therapies aim to increase SMN levels and preserve motor neurons, significant deficits remain in treated patients and non-neuronal manifestations of SMN deficiency are underexplored. Vascular abnormalities including intrinsic endothelial cell dysfunction, altered vessel morphology, and altered vascular distribution have been reported in preclinical SMA models. Here, we characterised vascular architecture and blood-spinal cord barrier (BSCB) morphology and integrity in post-mortem spinal cord samples from severe SMA patients compared with unaffected controls. Von Willebrand Factor (VWF), a marker of endothelial cell function, was reduced within individual vascular endothelial cells, and associated with ultrastructural endothelial cell oedema, vacuolisation and compromised endothelial integrity. Ultrastructural damage extended to other components of the blood-spinal cord barrier (BSCB) as evidenced by extravascular leakage of fibrinogen into the neural parenchyma and microglial activation consistent with a neuroinflammatory environment. Together, these findings suggest that vascular defects with associated dysfunction of the BSCB are present in the spinal cord of infants with severe SMA. This work adds to a growing body of evidence linking microvascular dysfunction to neurodegeneration in human neurodegenerative diseases. Further studies are warranted to define the contribution of vascular dysfunction to SMA pathogenesis and to assess whether current therapies adequately address this aspect of the disease.

Von Willebrand factor

Endothelial cell

BSCB

Astrocytes

Pericytes

Microglia

Severe spinal muscular atrophy (SMA) is a genetic neuromuscular disorder of infancy and early childhood caused by reduced levels of the survival motor neuron (SMN) protein. SMA primarily manifests pathologically as progressive motor neuron loss, and clinically with skeletal muscle weakness and denervation-induced atrophy [44]. These hallmark features are amenable, at least in part, to treatment with any of three SMN-enhancing therapies, nusinersen (Spinraza; Biogen), risdiplam (Evrysdi; Roche), or onasemnogene abeparvovec (Zolgensma; AveXis, Novartis). Commensurate with the amount of neurodegeneration present before treatment is given, the clinical course and prognosis of SMA is greatly improved [15].

While classic clinical and pathologic studies of SMA have reasonably focused on impaired motor neurons and skeletal muscle, a wider but clinically more subtle dysfunction is suggested by the ubiquitous expression of SMN. Studies in animal models support this, revealing broad, multi-system pathology [31]. Because the extent and distribution of these features differs widely between animal models, and the SMN-expression threshold below which tissue and species’ dysfunction manifests is unknown, the relevance of these extra-neuronal features to humans with SMA is uncertain. With improved survival following central nervous system (CNS)-targeted therapies, there is reason for concern that previously underappreciated, non-motor manifestations of SMA may emerge.

Our focus on the vasculature stems from both clinical and preclinical evidence that suggests vascular endothelium may be another tissue relatively vulnerable to diminished SMN expression. Retinal imaging studies in older children with milder forms of SMA demonstrate reduced capillary branching complexity, a finding mirrored in pathological studies of SMA mouse models [73]. Circulating biomarkers further support the presence of vascular damage, as SMA patients exhibit elevated circulating endothelial cells (CEC) and reduced endothelial progenitors, which correlate with both disease severity and SMN2 copy number [14, 19, 73]. A small number of SMA patients have presented with distal digital necrosis, mirroring tail and ear necrosis in SMA mouse models [6, 58]. Finally, congenital heart defects are more commonly observed in infants with the most severe forms of SMA [57] which are potentially linked to early endothelial dysfunction, as vascular endothelial precursors are essential for heart development [27, 72].

Beyond these clinical indications, pathological findings in both human and animal SMA tissue also reveal a range of vascular abnormalities. Skeletal muscle biopsies from SMA patients across the severity spectrum show reduced vascularisation and disrupted vessel architecture, marked by narrow vessels and simplified capillary beds [63]. Similar vascular defects are reported in systemic organs of SMA models, including the heart, kidneys, and intestines [3, 41, 61, 62]. These vascular abnormalities may arise from intrinsic endothelial defects, as endothelial cells with depleted SMN exhibit impaired angiogenic behaviours, including reduced migration and tube formation, shown in SMN-deficient human umbilical vein endothelial cells (HUVECs) cultures and aortic endothelial cells from severe SMA mouse models [73].

Importantly, vascular pathology may also contribute directly to CNS degeneration. In severe SMA mouse models, loss of the endothelial marker platelet endothelial adhesion molecule 1 (PECAM-1) in spinal cord grey and white matter preceded the onset of motor symptoms, progressed with age and was accompanied by pathological evidence of blood spinal cord barrier (BSCB) changes and tissue hypoxia, supported by increased uptake of in-vivo biomarkers of hypoxia identified pathologically [29, 63]. The spinal vasculature not only sustains neuronal and non-neuronal metabolism but also forms the structural core of the BSCB, regulating immune cell and molecular entry into the CNS [2]. Therefore, endothelial dysfunction within the spinal cord has the potential to alter the neural microenvironment and exacerbate neurodegeneration [21]. Indeed, vascular changes are increasingly recognised across other neurodegenerative diseases such as amyotrophic lateral sclerosis, Alzheimer’s disease, and Parkinson’s disease, though it remains debated whether these represent primary drivers of pathology or secondary consequences of neurodegeneration [12, 24, 49, 64, 71].

These observations suggest that vascular endothelial dysfunction may contribute both to local spinal cord pathology and to wider systemic involvement in SMA. This highlights the need for focused investigation of the human vascular network in SMA. In this study, we characterised the spinal cord vasculature and BSCB in autopsy tissue from a cohort of infants with well-defined, severe SMA and previously quantified motor neuron loss [4]. Our analysis reveals aberrant vascular architecture and BSCB leakage, implicating vascular dysfunction as a potential contributor to disease progression. While centred on the spinal cord, these findings also provide testable hypotheses for extra-neuronal SMA pathology.

Sample collection

Expedited autopsies were conducted under parental- or patient-informed consent in strict observance of the legal and institutional ethical regulations, and material tissue agreements were obtained between collaborating institutes. Paediatric human thoracic spinal cord samples from SMA patients (n=6) and similarly age-matched controls (n=6) were obtained from brain banks: BRAINUK, NIH Neurobiobank, and John Hopkins University Medical School. Tissue samples were dissected at autopsy, fixed in 4% paraformaldehyde at room temperature for 24 hours, washed and stored in PBS. Samples prepared for transmission electron microscopy from SMA patients (n=4), and similarly age-matched controls (n=2) were dissected and fixed in 2% glutaraldehyde at 4^oC for 24 hours. Patient demographics of samples is detailed in Table 1. The following analyses were conducted blinded to disease status (SMA or control).

Immunofluorescence/immunohistochemistry

Histology samples were cryopreserved in 30% sucrose solution with 0.1% sodium azide and embedded in optimum cutting temperature compound. Serial, transverse spinal cord sections were prepared (8µm). Sections were air dried overnight and submerged in 10mM sodium citrate buffer, pH6 (20 mins at 95^oC) for antigen retrieval, cooled in solution and washed (5 mins) in 0.1M PBS. Autofluorescence was quenched using 0.1% Sudan black (20 mins) and sections were washed three times (2x 10 mins PBST (0.1M PBS with 0.1% Tween-20), 1x 0.1M PBS). Sections were incubated (2 hours at 4^oC) in blocking solution (0.4% bovine serum albumin (BSA), 1% Triton X-100 in 0.1M phosphate buffered saline (PBS), 5% donkey serum), and then with diluted primary antibody; anti-Von Willebrand Factor 1:100 (Abcam, ab9378), anti-glial fibrillary acidic protein (GFAP) 1:100 (Abcam, ab53554), anti-haemoglobin 1:200 (R&D Systems, G-134-C), anti-fibrinogen 1:1000 (Millipore, 341552), anti-Iba1 1:500 (Abcam, ab107159), anti-CD68 1:200 (Abcam, ab213363) (overnight at 4^oCc). Sections were washed three times as before (2x 0.1M PBST, 1x 0.1M PBS), then incubated with corresponding diluted secondary antibody; donkey anti-rabbit AF+555 (Invitrogen, A32794), donkey anti-rabbit AF488 (Invitrogen, A21206), donkey anti-goat 488 (Abcam, ab150129), donkey anti-goat Cy3 (Abcam, ab6949) at 1:200 (2 hours at 4^oC) with successive washes as before. For double staining, the protocol was repeated sequentially. Nuclei were counterstained using DAPI (4′,6-diamidino-2-phenylindole) at 1:20,000 (10 mins at room temperature), then washed as before. Sections were mounted using MOWIOL media (10% Mowiol (Sigma-Alrich, 81 381), 20% glycerol, 50% 0.2M Tris buffer pH 8.5, 3% 1,4-diazobicyclooctance in distilled water). Negative control slides which lacked primary antibody were included in each staining run for detection of autofluorescence and non-specific binding.

Immunohistochemistry used the Novolink Polymer Detection kit (Lecia Biosystems, RE7280-K) and DAB visualisation for primary antibodies anti- Ki67, a proliferation marker [11, 25], 1:50 (Abcam, ab16667) and anti-fibrinogen 1:2000 (Millipore, 341552) following the supplied instructions. Sections were coverslipped with DPX mounting media. All stained sections were imaged on an EVOS M5000 inverted microscope.

VWF Quantification

Systematic uniform random sampling was implemented using a random number generator to determine the first section in the series and every 12^th section of spinal cord was immunofluorescently stained with VWF. Sections were imaged at 10x magnification in the grey matter ventral and dorsal horns, and white matter corticospinal tract and dorsal column. Micrographs were analysed using FIJI (Image J) software. For each slide, threshold was set manually to the point of saturation and the de-speckle function was applied to subtract small background noise. Automated thresholding was not applicable due to differences in background fluorescence. A counting frame test probe composed of 12 counting boxes of 17578μm² area was overlain on brightfield micrographs and employed as described in [20], each individual counting box representing a field of view (FOV). FOVs that extended beyond the region of interest were identified and discounted from analysis. Sudan black counterstaining allowed delineation of white and grey matter under brightfield illumination.

VWF+ Area/Field of View: For each included FOV, total VWF stained area was calculated relative to total field of view (area of counting box) and expressed as % area stained. FOV analysed, Control > 150 and SMA >150 per region of interest.

Vessel density: Vessel density/per mm² was calculated for each ROI on each slide by dividing the total number of vessel profiles by the total area of FOV included (17578μm² X number of grid boxes analysed) and converted to vessels/mm². Mean vessel density, in each individual ROI across all sections analysed, was calculated. FOV analysed, Control > 150 and SMA >150 per region of interest. Simple linear regression analyses show change in spinal cord vascular density between ages.

VWF+ Area/vessel profile: Individual VWF⁺ vessel profiles (200 per spinal cord sample) were assessed within the ventral horn. Individual vessel profiles were manually selected and stained area was measured using particle analysis on FIJI (ImageJ), providing the stained area per vessel profile.

Transmission electron microscopy

Epoxy resin-embedded blocks for transmission electron microscopy (TEM) were prepared at Johns Hopkins University (Control n=2, SMA n=4), as described in [35]. Grids were sectioned by the microscopy and histology facility at the University of Aberdeen using a diamond knife to cut approximately 90 nm sections onto formvar/carbon-coated copper grids. The sections were then contrast-stained with Uranyless (Labtech, 11000-200) and Lead citrate. Grids were surveyed systematically: tissue was randomly oriented, and imaging began at the top left corner of a randomly oriented section, proceeding from left to right on each row, and then moving to the row below in a right-to-left direction. The first 50 vessel profiles identified in grey matter were imaged from one section per case. Grey matter was identified by the surrounding neuropil and the absence of myelinated axons around the vessel. Vessel profiles that overlay grids were excluded, thus constraining analysis to vessels less than 30 µm diameter and biasing to smaller vessels in that pool.

Endothelial cell area: The luminal and basal boundaries of the endothelial cells were traced, and the area of the lumen was subtracted from the area enclosed by the outer basal boundary to calculate the endothelial cell area.

Endothelial cell frequency: Determined by calculating the frequency of endothelial cell nuclei observed in vessel images.

Minimum lumen diameter: To determine minimum lumen diameter, electron micrographs were assessed using FIJI (ImageJ) software. Minimum lumen diameter was chosen to reflect the minimum vessel width irrespective of plane of section. Lumenal boundaries of endothelial cells were manually traced, and an ellipse was fitted to the structure to allow maximum and minimum lumen measurements through the centroid.

Morphological EM observations: All observations of morphological differences in endothelial cells, basement membrane, pericytes and astrocytes were identified using published literature as reference [24, 36, 45]. Phenotype frequency determined as a percentage of total vessels assessed.

Extravascular leakage

Fibrinogen/Haemoglobin: Three sections of spinal cord, selected using a random number generator, from control (n=5) and SMA (n=5) tissues were immunohistochemically stained for fibrinogen, VWF and haemoglobin as described above. Extravascular fibrinogen/haemoglobin was determined when staining was observed outside the walls of a vessel in either the grey or white matter of the spinal cord and there was a visible vessel within close proximity/surrounded by positive staining. Extravascular fibrinogen was defined by either a diffuse pattern of fibrinogenpositive staining surrounding a vessel, or a small, clustered deposit of fibrinogenpositive staining in the perivascular space. Extravascular haemoglobin was defined by a region of haemoglobinpositive staining either entirely or partially surrounding a vessel. Number of vessels exhibiting extravascular leakage of fibrinogen and haemoglobin was quantified by counting all visible vessel profiles in a sections, and expressing the number of vessels associated with extravascular fibrinogen of haemoglobin as a percentage of the total.

Microglial Abundance and activation

Iba1/CD68: Three sections of spinal cord, selected using a random number generator, from control (n=4) and SMA (n=4) were immunofluorescently stained for Iba1 (microglial marker) and CD68 (marker of microglial activation) and DAPI. Microglial cell bodies (nuclear DAPI with surrounding Iba1 positive staining) were counted to determine microglial abundance. Microglial activation was determined through co-localised Iba1 positive and CD68 positive staining. Of the total microglial (Iba1 positive) cells observed, the number of activated microglial (Iba1 positive and CD68 positive) cells are presented as a percentage.

Statistical Analysis

Outliers were identified using ROUT method, Q=1%, and removed from the dataset prior to analysis. Descriptive statistics and D’Agostino & Pearson normality tests were conducted to determine distribution of data. If data was of a Gaussian distribution, data is presented as mean +SEM, however if of a non-Gaussian distribution, results are presented as median (interquartile range). Statistical comparison was analysed by Welch’s T-tests. Under all circumstances, ^*P < 0.05, ^**P < 0.01 and ^***P < 0.001.

Patient population

We systematically assessed the spinal cord vasculature and blood–spinal cord barrier integrity in a cohort of infants diagnosed with severe spinal muscular atrophy, and with survival times ranging from 1 to 12 months.

Patient demographics are detailed in Table 1, and information concerning motor neuron loss and spinal cord development and growth for these cases are previously published [4]. All but one SMA case within this cohort, whose genotype remains unknown, possessed two copies of SMN2. This case has the shortest postnatal survival of just 7 days and is likely to reflect either 1 or 2 copies of SMN2, as 3 copies of the SMN2 gene are associated with significantly longer survival [14, 43, 70].

As is inherent in all autopsy-based comparative pathology studies, differences in the cause of death and preterminal health status must be acknowledged. SMA cases died from SMA-related complications that involved extended periods of metabolic distress and prolonged respiratory decline, while controls died from sudden unexplained death syndrome (n=2), pneumonia (n=2), asphyxia (n=1), or unknown causes in three cases—mostly acute events that likely involve less prolonged systemic compromise than SMA cases. Postmortem intervals ranged from 7–32 hours in SMA cases and 14–69 hours in controls, with unknown intervals for two SMA cases and one control.

Table 1 Patient demographics

Summary of case information including age at death (in days (d) or months (m)), cause of death, SMN2 copy number, post-mortem interval (PMI) where known, and use of samples for light microscopy (LM) and/or transmission electron microscopy (TEM).

	Age of death (days/months)	Cause of death	SMN copy number *(SMN1/SMN2)*	PMI (hours)	Application
Control Cases	1d	Unknown	-	25	TEM
	4d	Unknown	-	69	TEM
	9 d	Unknown	-	Unknown	LM
	3.2 m	Sudden unexplained infant death	-	27	LM
	5.2 m	Viral syndrome/focal acute pneumonia	-	31	LM
	7.1 m	Asphyxia	-	24	LM
	9 m	Pneumonia	-	14	LM
	12 m	Sudden infant death syndrome	-	27	LM
SMA Cases	7 d	SMA Type 0	Unknown	Unknown	LM
	1.75 m	SMA Type I	0/2	7	LM TEM
	2.5 m	SMA Type I	0/2	7	LM TEM
	4 m	SMA Type I	0/2	4	LM
	7 m	SMA Type I	0/2	25	LM TEM
	11 m	SMA Type I	0/2	32	LM
	11m	SMA Type I	0/2	Unknown	TEM

VWF expression is decreased in endothelial cells in the SMA spinal cord

Spinal cord vasculature was visualized by immunostaining sections for von Willebrand factor (VWF), a widely used cytosolic marker of endothelial cells and endothelial cell health [59].

As the primary site of neurodegeneration in SMA, we first analysed the ventral horn. In control cases, VWF staining of individual blood vessels was bright and uniform in contrast to SMA cases, where vessels were more heterogeneously stained (Fig. 1a-b). Total VWF positive stained area (expressed as a percentage of total area stained) was significantly lower in SMA ventral horn compared to age-matched controls: control = 1.99% (+0.17), and SMA = 0.79% (+0.10), mean (+SEM), **P <0.01 (Fig. 1c). We extended analyses to also assess afferent and efferent regions and associated tracts involved in motor control. Stained area was also significantly lower in SMA grey matter dorsal horn (*P <0.05) and white matter dorsal column (*P <0.05) but did not reach statistical significance in the white matter corticospinal tract (Pns, Supplementary Fig. 1c, g, k).

To assess whether the decrease in total VWF levels were the result of decreased vessel density in SMA, we quantified the number VWF positive vessel profiles by applying design-based stereology. Vessel density was not lower in SMA cord in any of the 4 regions of interest, rather increasing with increased survival time in both ventral horn (Fig. 1d) and dorsal horn (Supplementary Fig. 1d). Indeed, Ki67 labelling revealed that endothelial cells, were frequently proliferating in the SMA spinal cord, but not in controls (Supplementary Fig. 1m-n). For SMA cases that received artificial ventilation during clinical care, no correlation was observed with increased vascular vessel density and endothelial proliferation. In summary, the decrease in total VWF could not be attributed to lower vessel density in SMA.

Release of VWF from endothelial cells is a marker of dysfunction, therefore as the overall reduction in VWF expression was not associated with vessel loss, we next assessed if it was associated with reduced VWF expression per vessel. VWFstained area was reduced by an average of 60% in individual vessel profiles in SMA spinal cord compared to age-matched controls, control = 102μm² (+12) and SMA = 41μm² (+5), mean (+SEM), **P <0.005 (Fig. 1e). Platelets, which also express VWF, were not observed within vessels, and did not affect our assessments.

Given our observation of reduced VWF in vessel profiles, we next used transmission electron microscopy (TEM) in a subset of age-matched cases (0–1 year) to determine if reduced VWF per vessel profile in SMA spinal cord was due to changes in the size or number of endothelial cells in each vessel profile. However, no differences were observed between SMA and control cases in endothelial cell area (Fig. 1h-j): control = 38.15µm² (+12.56), SMA = 30.61µm² (+3.54), mean (+SEM), nuclear frequency (as a surrogate for endothelial cell number: Fig. 1l): control range = 0.33-0.55, SMA range = 0.23-0.65 (Fig. 2h), or luminal diameter (Fig. 1j-l): control = 5.69µm² (+0.39), SMA = 5.55µm² (+0.43), mean (+SEM).

These findings show that the decrease in total VWF expression is independent of vessel density, cell number or size, rather representing decreased VWF per endothelial cell in the SMA spinal cord.

Endothelial integrity is disrupted in the SMA spinal cord

We extended evaluation of endothelial cell health to the ultrastructural level, examining cross sections of vessels in electron micrographs from four SMA and two control cases (demographics in Table 1). Vessels imaged were likely small arterioles, venules and capillaries, as larger vessels are typically associated with the meninges or are travelling in the plane of the section and rarely observed cross-sectionally [65].

In control spinal cords, vascular endothelial cells displayed classical features: flattened nuclei against the vessel wall, electron-dense cytoplasm, and intact luminal membranes [24] (Fig. 2a). In contrast, SMA vascular endothelial cells exhibited disrupted plasma membranes (Fig. 2b), cytoplasmic oedema, organelle disruption, and vacuolisation (Fig. 2c), all indicative of a loss of cellular integrity and compromised cell health. These abnormalities were specific to endothelium, and surrounding glial cells appeared unaffected.

Endothelial cells form the inner layer of the blood spinal cord barrier (BSCB), establishing a continuous monolayer of cells adjacently connected by tight junctions and anchored to the underlying basement membrane on their basal surface [23]. A healthy endothelial cell from control spinal cord (segmented in yellow) with underlying basement membrane (red dashed line) is shown in Fig. 2d for reference. In SMA, endothelial discontinuities were evident in two of four cases, with gaps in the endothelial lining allowing direct blood cell contact with the basement membrane (Fig. 2e). Basement membrane abnormalities, including detachment from the endothelial layer, 3/4 cases, (Fig. 2f) and focal thickening, 4/4 cases, (Fig. 2g-h) were also observed in SMA but not in controls. Endothelial intracellular oedema and membrane rupture were the most consistent pathological features in SMA cases (Fig. 2i).

Pericyte and Astrocyte End-Feet abnormalities in SMA Spinal Cord

Having identified pathological changes to the endothelial cell layer and basement membrane, we next examined pericytes and astrocyte end-feet, the remaining key cellular components of the BSCB.

In control spinal cord, pericytes displayed classic "bump on the log" morphology, lying adjacent to the endothelium with prominent nuclei, limited cytoplasm and long processes wrapping the vessel wall (Fig. 3a) [7]. In contrast, SMA pericytes showed cytoplasmic swelling, oedema (Fig. 3b), and vacuolisation (Fig. 3c).

Astrocyte end-feet form the outer BSCB layer, and appeared normal in control spinal cord with sparse cytoplasmic filaments [9, 45], (Fig. 3d). This contrasts with the astrocytic end-feet visualized in the SMA cases that displayed abnormal filament accumulation (Fig. 3e) and cytoplasmic vacuolisation (Fig. 3f). Immunofluorescence staining showed a visible increase in glial fibrillary acidic protein (GFAP) expression across all spinal cord regions in SMA cases (Fig. 3g–h).

Evidence of BSCB Compromise in the SMA spinal cord

Given our ultrastructural findings and prior evidence of vascular impairment in both SMA patients and animal models, we next looked for pathological markers of BSCB dysfunction by analysing the distribution of fibrinogen and haemoglobin by immunohistochemistry.

Fibrinogen, a large plasma protein normally restricted to the vascular lumen [51], where it was confined in control spinal cords (Fig. 4a). In contrast, extravascular fibrinogen was detected frequently in SMA spinal cord, presenting as either diffuse staining around the vessel or as focal perivascular deposition (Fig. 4b–c). Extravascular fibrinogen was present in all five SMA cases (median 7. 45%, range 0.95–69.76%) (Fig. 4d), most prominently in the dorsal horn, but not observed in controls (median 0.00%, range 0.00–0.04%).

Haemoglobin, which may be released during or shortly after vascular injury [66], was co-stained with VWF to localise its presence in relation to vessels. In controls, haemoglobin was present intravascularly (Fig. 4e), whereas in SMA spinal cords, extravascular haemoglobin was found surrounding vessels in the neural parenchyma (median 1.17%, range 0.00–18.14%) (Fig. 4f-g). Extravascular haemoglobin was most often seen in the dorsal column.

At the ultrastructural level, extravasated red blood cells were observed in 2 of 4 SMA cases, but not in controls (Fig. 4h), while perivascular oedema was a consistent feature across all SMA spinal cords examined (Fig.4i).

Pronounced Microglial Activation is present throughout the SMA spinal cord

The presence of plasma proteins in the neural parenchyma suggests vulnerability to BSCB disruption in SMA. A downstream consequence is the potential for reactive neuroinflammation associated with glial activation [67]. This was assessed by microglial Iba1 immunofluorescence in all regions of interest. In control spinal cords, microglia exhibited a ramified morphology typical of a resting state (Fig. 5a). In comparison, SMA microglia were clustered, with an ameboid morphology, retracted processes and reduced branching, consistent with an activated and reactive microglial phenotype [47], (Fig 5b). Microglial cell number was broadly similar in SMA and controls (control: 51± 8; SMA: 78 ± 17 Iba1⁺ cells per FOV, Pns <0.05), (Fig. 5c). Microglial activation was increased in SMA cases, where almost all Iba1 positive microglia co-expressed the macrophage activation marker CD68 (97.36% ± 0.29), compared to none in controls (Fig. 5d–f). This reactive microglial phenotype was observed throughout the spinal cord, not only in the ventral horn where motor neuron degeneration occurs, suggesting microglial activation was a widespread characteristic in these cases of SMA.

We have identified abnormalities in SMA spinal cord endothelial cells and BSCB components that are associated with BSCB disruption, leakage and microglial activation. Together these suggest a meaningful, previously underappreciated, role of vascular dysfunction in the pathogenesis of human severe SMA.

Changes in endothelial VWF identifies endothelial call pathology in SMA spinal cord

VWF is a key endothelial protein essential for maintaining vascular health and haemostasis [40]. We observed reduced endothelial VWF in SMA spinal cords and localized this specifically to individual endothelial cells, reflecting lower intracellular levels within endothelial cells. Changes in intracellular VWF can indicate pathology, as it is rapidly released from endothelial cells upon injury or activation [46]. VWF is normally stored in Weibel–Palade bodies and rapidly secreted into the bloodstream in response to stress or damage—a key event in endothelial activation [5, 37, 38, 50, 55, 59]. In SMA, pathological factors such as hypoxia and oxidative stress—both reported in animal models [29, 63]—could trigger this release [8, 26, 53]. In vitro, hypoxic stimulation of HUVECS increases VWF secretion four-fold [5]. Moreover, superoxide anions generated during hypoxia elevate intracellular Ca²⁺, triggering rapid Weibel–Palade body exocytosis [69]. These reactive oxygen species also promote stress fibre formation and disrupt barrier integrity by disassembling tight and adherens junctions through MAP kinase activation [28, 39]. These mechanisms align with the observations in this study of discontinuities in endothelial vessel walls creating gaps in the barrier, detachment of endothelial cells from the basement membrane, and in previous studies, claudin-5 reorganisation in SMA mouse models [63].

Basement membrane, pericytes and astrocyte layers of the BSCB exhibit pathological changes

Endothelial cell injury was accompanied by abnormalities in other BSCB components. The basement membrane (BM) in SMA spinal cords was focally thickened, potentially as the result of endothelial dysfunction and serum protein exposure that alter its normal composition [68]. Oxidative stress can disrupt endothelial metabolism, driving abnormal production or turnover of BM components, contributing to its thickening [22]. Similar BM thickening occurs in Alzheimer’s and Parkinson’s disease, where it is thought to be secondary to vascular injury and oxidative stress, and contributes to impairment of vessel contractility, exacerbating local hypoxia [10, 22].

Pericytes in the SMA spinal cord displayed pronounced cytoplasmic swelling, oedema, and vacuolisation, indicative of cellular stress and dysfunction. Given their roles in vessel stability, angiogenesis regulation, and junctional integrity, pericyte impairment could also exacerbate endothelial detachment, disrupt vessel maturation, and compromise barrier function [16, 17].

Astrocytes in the SMA spinal cord showed cytoplasmic vacuolisation and abnormal glial fibrillary acidic protein (GFAP) accumulation. Consistent with our findings, increased GFAP is reported in a separate study of post-mortem spinal cords from severe SMA patients, and SMA mouse models show reactive astrocytes with morphological and functional changes preceding neuronal death [42, 56]. Astrocytes also contribute to the expression of tight junction proteins, including Occludin, Zona Occludens-1, and Claudin-5, which regulate endothelial cell adherence and BSCB integrity [13]. Dysfunction in astrocytes may therefore further contribute to the endothelial cell detachment observed here and the loss of Claudin-5 reported in SMA mouse models [63]. While the phenotypes observed in pericytes and astrocytes may arise as secondary effects of endothelial injury due to complex intercellular communication, they may also reflect intrinsic defects stemming from SMN protein deficiency.

Pathological changes to the BSCB are associated with barrier dysfunction, leakage and microglial activation

Damage to all components of the BSCB was associated with barrier leakage, demonstrated by the presence of extravascular fibrinogen, a large plasma glycoprotein that enters the neural parenchyma only under pathological conditions [32]. In SMA spinal cords, two patterns of extravascular fibrinogen were observed: diffuse staining surrounding vessels (likely reflecting recent or ongoing leakage) and perivascular deposits (likely remnants of past leakage) [52]. Extravascular fibrinogen drives CNS pathology, triggering microglial activation, promoting neuroinflammation, and inhibiting remyelination and neuronal repair [1, 18, 52, 60].

Extravascular haemoglobin was detected in some SMA cases but did not differ significantly from controls, therefore it may reflect peri-mortem red blood cell lysis or post-mortem artefact. Haemoglobin is known to diffuse following red cell rupture, particularly under conditions of tissue hypoxia or delayed fixation, limiting its utility as a marker of true barrier compromise. The absence of a corresponding increase of extravascular haemoglobin supports our interpretation that fibrinogen leakage reflects true pathological barrier breakdown rather than non-specific post-mortem events.

Increased microglial activation was also present in SMA spinal cords, and extravascular fibrinogen may have contributed to this. When present in the CNS, fibrinogen binds to the integrin receptor Mac-1 (CD11b/CD18) on microglia, triggering intracellular signalling that promotes adhesion, migration, and pro-inflammatory activation [1, 60]. SMA mouse models show reactive microgliosis around motor neurons, characterised by increased M1 and reduced M2 microglial profiles. Initially thought to be secondary to motor neuron degeneration, though more recent work with SMA-derived iPSCs has revealed intrinsic microglial defects in morphology, phagocytosis, and migration [33]. This suggests that microglial activation in SMA might arise from intrinsic pathology and/ or extrinsic triggers such as barrier breakdown, creating a sustained pro-inflammatory environment.

Implications of vascular dysfunction for SMA patients

We have reported extensive pathology in severe SMA patient spinal cords, including all components of the neurovascular unit. Whether these arise secondary to motor neuron loss or are instead a primary consequence of SMN deficiency, this vascular pathology may contribute to a hostile microenvironment for neurons that possibly exacerbate processes of neurodegeneration. While SMN deficiency in motor neurons is a well-established driver of disease, neuron-specific knockdown models do not fully replicate the severity of patient symptoms [48]. Conversely, restoring SMN in peripheral, non-neuronal cells improves survival and motor function, emphasising the contribution of non-neuronal tissues to disease progression [30, 54].

Hypoxia is a critical determinant of neuronal health, and SMN-deficient motor neurons are particularly vulnerable to oxygen deprivation [29]. We propose that endothelial defects in SMA, which could be the result of intrinsic mechanisms through SMN depletion or environmental cues such as hypoxia, drive chronic vascular injury and BSCB dysfunction. This allows toxic blood-derived proteins such as fibrinogen to enter the parenchyma, exacerbating motor neuron degeneration. Even without primary neuronal injury, vascular pathology can create a pro-inflammatory, hypoxic environment that motor neurons are particularly vulnerable to [34].

Limitations of this study

A well-recognised limitation of autopsy-based studies is variability in premorbid health and cause of death between disease and control groups. In our cohort, control infants died from acute causes such as sudden infant death syndrome (SIDS), asphyxia, or other unknown events, likely without prolonged metabolic compromise (Table 1). In contrast, clinical experience suggests that SMA infants often experience extended periods of respiratory failure, malnutrition, and metabolic stress prior to death.

Notably, PMIs were generally longer in control cases, which may also influence tissue preservation and antigenicity. While such confounds cannot be fully controlled in human autopsy studies, complementary findings from SMA animal models—where vascular changes and BSCB abnormalities are also observed [29, 63]—support the disease relevance of our results. Our findings are presented within this context, and we caution against overinterpretation of differences that could reflect systemic pre-terminal factors.

Although three disease-modifying therapies now prolong survival in severe SMA it is unknown if any of these will address vascular pathology. Our findings indicate that vascular defects may contribute to motor neuron vulnerability and disease progression. Therefore, long-term monitoring and therapeutic strategies that consider vascular health may improve outcomes of SMA patients receiving these therapies.

Declarations

Ethics approval and consent to participate

Expedited autopsies were conducted under parental- or patient-informed consent in strict observance of the legal and institutional ethical regulations, and material tissue agreements were obtained between collaborating institutes: University of Aberdeen, Johns Hopkins University School of Medicine, NIH Neurobiobank and BRAIN UK.

Consent for publication

Not Applicable

Competing interests

CJS receives or received grant support from Roche Ltd, Biogen, and Actio Biosciences and has served as a paid advisor, consultant, and/or speaker to Biogen, Roche/Genentech, and Novartis. These arrangements have been reviewed and approved by the Johns Hopkins University in accordance with its conflict-of-interest policies.

Funding

This work was supported by funding from the SMA Foundation and the NIH (R35 NS122306) to CJS and an Anatomical Society PhD Studentship (RG14940-10) to SHP and HA.

Author Contribution

HA and SHP designed the study. CJS and TOC collected SMA autopsy tissues and provided tissues from Johns Hopkins University School of Medicine. HA, HL and BDL performed experiments and analysed data. The first draft of the manuscript was prepared by HA and SHP, and all authors commented and edited the manuscript. All authors read and approved the final manuscript.

Acknowledgement

We would like to acknowledge Patrice Carr at the Department of Neurology, Johns Hopkins University School of Medicine for sample identification, organisation and shipment to the University of Aberdeen. We also wish to thank Professor Peter Dockery, University of Galway, for his advice on the design and application of stereological techniques, and the Microscopy and Histology Core Facility members, Gillian Milne, Lucy Wight and Debbie Wilkinson at the University of Aberdeen.

Data Availability

The datasets generated during the current study available from the corresponding author on reasonable request.

Adams RA, Bauer J, Flick MJ, Sikorski SL, Nuriel T, Lassmann H, Degen JL, Akassoglou K (2007) The fibrin-derived γ377-395 peptide inhibits microglia activation and suppresses relapsing paralysis in central nervous system autoimmune disease. Journal of Experimental Medicine 204:571–582. doi: 10.1084/JEM.20061931,
Alahmari A (2021) Blood-Brain Barrier Overview: Structural and Functional Correlation. Neural Plast 2021:6564585. doi: 10.1155/2021/6564585
Allardyce H, Kuhn D, Hernandez-Gerez E, Hensel N, Huang YT, Faller K, Gillingwater TH, Quondamatteo F, Claus P, Parson SH (2020) Renal pathology in a mouse model of severe Spinal Muscular Atrophy is associated with downregulation of Glial Cell-Line Derived Neurotrophic Factor (GDNF). Hum Mol Genet 29:2365–2378. doi: 10.1093/hmg/ddaa126
Allardyce H, Lawrence BD, Crawford TO, Sumner CJ, Parson SH (2024) A reassessment of spinal cord pathology in severe infantile spinal muscular atrophy. Neuropathol Appl Neurobiol 50:e13013. doi: 10.1111/NAN.13013
Antonova OA, Loktionova SA, Golubeva N V, Romanov YA, Mazurov A V Damage and Activation of Endothelial Cells during in Vitro Hypoxia. Eksperimental’noi Biologii i Meditsiny 144:384–386
Araujo A prufer de QC, Araujo M, Swoboda KJ (2009) Vascular Perfusion Abnormalities in Infants with Spinal Muscular Atrophy. Journal of Pediatrics 155:292–294. doi: 10.1016/j.jpeds.2009.01.071
Attwell D, Mishra A, Hall CN, O’Farrell FM, Dalkara T (2016) What is a pericyte? Journal of Cerebral Blood Flow and Metabolism 36:451–455. doi: 10.1177/0271678X15610340/ASSET/DA388224-4D0B-459B-9ABB-31414409EC5B/ASSETS/IMAGES/LARGE/10.1177_0271678X15610340-FIG2.JPG
Avdonin P V., Tsitrina AA, Mironova GY, Avdonin PP, Zharkikh IL, Nadeev AD, Goncharov N V. (2017) Hydrogen peroxide stimulates exocytosis of von Willebrand factor in human umbilical vein endothelial cells. Biology Bulletin 44:531–537. doi: 10.1134/S106235901705003X/METRICS
Bayir E, Sendemir A (2021) Role of Intermediate Filaments in Blood–Brain Barrier in Health and Disease. Cells 10:1400. doi: 10.3390/CELLS10061400
Bertrand E, Lewandowska E, Stêpieñ T, Szpak GM, Pasennik E, Modzelewska J (2008) Amyloid angiopathy in idiopathic Parkinson’s disease. Immunohistochemical and ultrastructural study
Brown DC, Gatter KC (1990) Monoclonal antibody Ki-67: its use in histopathology. Histopathology 17:489–503. doi: 10.1111/J.1365-2559.1990.TB00788.X
Brown WR, Thore CR (2011) Review: Cerebral microvascular pathology in ageing and neurodegeneration. Neuropathol Appl Neurobiol 37:56–74. doi: 10.1111/J.1365-2990.2010.01139.X
Cabezas R, Ávila M, Gonzalez J, El-Bachá RS, Báez E, García-Segura LM, Coronel JCJ, Capani F, Cardona-Gomez GP, Barreto GE (2014) Astrocytic modulation of blood brain barrier: Perspectives on Parkinson’s disease. Front Cell Neurosci 8:1–11. doi: 10.3389/fncel.2014.00211
Calucho M, Bernal S, Alías L, March F, Venceslá A, Rodríguez-Álvarez FJ, Aller E, Fernández RM, Borrego S, Millán JM, Hernández-Chico C, Cuscó I, Fuentes-Prior P, Tizzano EF (2018) Correlation between SMA type and SMN2 copy number revisited: An analysis of 625 unrelated Spanish patients and a compilation of 2834 reported cases. doi: 10.1016/j.nmd.2018.01.003
Chaytow H, Faller KME, Huang YT, Gillingwater TH (2021) Spinal muscular atrophy: From approved therapies to future therapeutic targets for personalized medicine. Cell Rep Med 2:100346. doi: 10.1016/j.xcrm.2021.100346
Chopra N, Menounos S, Choi JP, Hansbro PM, Diwan AD, Das A (2021) Blood-Spinal Cord Barrier: Its Role in Spinal Disorders and Emerging Therapeutic Strategies. NeuroSci 3:1–27. doi: 10.3390/neurosci3010001
Daneman R, Prat A (2015) The blood–brain barrier. Cold Spring Harb Perspect Biol 7:1–18. doi: 10.1101/cshperspect.a020412
Davalos D, Kyu Ryu J, Merlini M, Baeten KM, Le Moan N, Petersen MA, Deerinck TJ, Smirnoff DS, Bedard C, Hakozaki H, Gonias Murray S, Ling JB, Lassmann H, Degen JL, Ellisman MH, Akassoglou K (2012) Fibrinogen-induced perivascular microglial clustering is required for the development of axonal damage in neuroinflammation. Nature Communications 2012 3:1 3:1–15. doi: 10.1038/ncomms2230
Erdbruegger U, Haubitz M, Woywodt A (2006) Circulating endothelial cells: A novel marker of endothelial damage. Clinica Chimica Acta 373:17–26. doi: 10.1016/J.CCA.2006.05.016
Fabricius K, Barkholt P, Jelsing J, Hansen HH (2017) Application of the physical disector principle for quantification of dopaminergic neuronal loss in a rat 6-hydroxydopamine nigral lesion model of Parkinson’s disease. Front Neuroanat 11:299477. doi: 10.3389/FNANA.2017.00109/BIBTEX
Fang Y, Hsieh Y, Hu C (2023) Endothelial Dysfunction in Neurodegenerative Diseases. Int J Mol Sci 24:1–27
Farkas E, De Jong GI, De Vos RAI, Jansen Steur ENH, Luiten PGM (2000) Pathological features of cerebral cortical capillaries are doubled in Alzheimer’s disease and Parkinson’s disease. Acta Neuropathol 100:395–402. doi: 10.1007/S004010000195/METRICS
Galea I (2021) The blood–brain barrier in systemic infection and inflammation. Cellular & Molecular Immunology 2021 18:11 18:2489–2501. doi: 10.1038/s41423-021-00757-x
Garbuzova-Davis S, Hernandez-Ontiveros DG, Rodrigues MCO, Haller E, Frisina-Deyo A, Mirtyl S, Sallot S, Saporta S, Borlongan C V., Sanberg PR (2012) Impaired blood-brain/spinal cord barrier in ALS patients. Brain Res 1469:114–128. doi: 10.1016/j.brainres.2012.05.056
Gerdes J, Schwab U, Lemke H, Stein H (1983) Production of a mouse monoclonal antibody reactive with a human nuclear antigen associated with cell proliferation. Int J Cancer 31:13–20. doi: 10.1002/IJC.2910310104;CTYPE:STRING:JOURNAL
Goerge T, Niemeyer A, Rogge P, Ossig R, Oberleithner H, Schneider SW (2002) Secretion pores in human endothelial cells during acute hypoxia. J Membr Biol 187:203–211. doi: 10.1007/S00232-001-0164-4
Haack T, Abdelilah-Seyfried S (2016) The force within: endocardial development, mechanotransduction and signalling during cardiac morphogenesis. Development 143:373–386. doi: 10.1242/DEV.131425
Harki O, Tamisier R, Pépin JL, Bailly S, Mahmani A, Gonthier B, Salomon A, Vilgrain I, Faury G, Briançon-Marjollet A (2021) VE-cadherin cleavage in sleep apnoea: new insights into intermittent hypoxia-related endothelial permeability. European Respiratory Journal 58. doi: 10.1183/13993003.04518-2020
Hernandez-Gerez E, Dall’Angelo S, Collinson JM, Fleming IN, Parson SH (2020) Widespread tissue hypoxia dysregulates cell and metabolic pathways in SMA. Ann Clin Transl Neurol 7:1580–1593. doi: 10.1002/acn3.51134
Hua Y, Sahashi K, Rigo F, Hung G, Horev G, Bennett CF, Krainer AR (2011) Peripheral SMN restoration is essential for long-term rescue of a severe spinal muscular atrophy mouse model. Nature 478:123–126. doi: 10.1038/nature10485
Jing C, Yeo J, Darras BT (2020) Overturning the Paradigm of Spinal Muscular Atrophy as Just a Motor Neuron Disease. doi: 10.1016/j.pediatrneurol.2020.01.003
Kadry H, Noorani B, Cucullo L (2020) A blood–brain barrier overview on structure, function, impairment, and biomarkers of integrity. Fluids and Barriers of the CNS 2020 17:1 17:1–24. doi: 10.1186/S12987-020-00230-3
Khayrullina G, Alipio‐Gloria ZA, Deguise M, Gagnon S, Chehade L, Stinson M, Belous N, Bergman EM, Lischka FW, Rotty J, Dalgard CL, Kothary R, Johnson KA, Burnett BG (2022) Survival motor neuron protein deficiency alters microglia reactivity. Glia. doi: 10.1002/GLIA.24177
Knox EG, Aburto MR, Clarke G, Cryan JF, O’Driscoll CM (2022) The blood-brain barrier in aging and neurodegeneration. Molecular Psychiatry 2022 27:6 27:2659–2673. doi: 10.1038/s41380-022-01511-z
Kong L, Valdivia DO, Simon CM, Hassinan CW, Delestrée N, Ramos DM, Park JH, Pilato CM, Xu X, Crowder M, Grzyb CC, King ZA, Petrillo M, Swoboda KJ, Davis C, Lutz CM, Stephan AH, Zhao X, Weetall M, Naryshkin NA, Crawford TO, Mentis GZ, Sumner CJ (2021) Impaired prenatal motor axon development necessitates early therapeutic intervention in severe SMA. Sci Transl Med 13:eabb6871. doi: 10.1126/scitranslmed.abb6871
Krueger M, Mages B, Hobusch C, Michalski D (2019) Endothelial edema precedes blood-brain barrier breakdown in early time points after experimental focal cerebral ischemia. Acta Neuropathol Commun 7:1–17. doi: 10.1186/s40478-019-0671-0
Lenting PJ, Christophe OD, Denis C V. (2015) von Willebrand factor biosynthesis, secretion, and clearance: connecting the far ends. Blood 125:2019–2028. doi: 10.1182/BLOOD-2014-06-528406
Lip GYH, Blann A (1997) von Willebrand factor: A marker of endothelial dysfunction in vascular disorders? Cardiovasc Res 34:255–265. doi: 10.1016/S0008-6363(97)00039-4/2/34-2-255-FIG1.GIF
Makarenko V V., Usatyuk P V., Yuan G, Lee MM, Nanduri J, Natarajan V, Kumar GK, Prabhakar NR (2014) Intermittent hypoxia-induced endothelial barrier dysfunction requires ROS-dependent MAP kinase activation. Am J Physiol Cell Physiol 306:C745. doi: 10.1152/AJPCELL.00313.2013
Manz XD, Bogaard HJ, Aman J (2022) Regulation of VWF (Von Willebrand Factor) in Inflammatory Thrombosis. Arterioscler Thromb Vasc Biol 42:1307–1320. doi: 10.1161/ATVBAHA.122.318179/SUPPL_FILE/ATVB_ATVB-2022-318179D_SUPP2.PDF
Maxwell GK, Szunyogova E, Shorrock HK, Gillingwater TH, Parson SH (2018) Developmental and degenerative cardiac defects in the Taiwanese mouse model of severe spinal muscular atrophy. J Anat 232:965–978. doi: 10.1111/joa.12793
McGivern J V, Patitucci TN, Nord JA, Barabas M-EA, Stucky CL, Ebert AD (2013) Spinal muscular atrophy astrocytes exhibit abnormal calcium regulation and reduced growth factor production. Glia 61:1418–28. doi: 10.1002/glia.22522
Mercuri E, Bertini E, Iannaccone ST (2012) Childhood spinal muscular atrophy: controversies and challenges. Lancet Neurol 11:443–452. doi: 10.1016/S1474-4422(12)70061-3
Monani UR, De Vivo DC (2014) Neurodegeneration in spinal muscular atrophy: from disease phenotype and animal models to therapeutic strategies and beyond. Future Neurol 9:49. doi: 10.2217/FNL.13.58
Nahirney PC, Tremblay ME (2021) Brain Ultrastructure: Putting the Pieces Together. Front Cell Dev Biol 9:1–22. doi: 10.3389/fcell.2021.629503
Nightingale T, Cutler D (2013) The secretion of von Willebrand factor from endothelial cells; an increasingly complicated story. Journal of Thrombosis and Haemostasis 11:192–201. doi: 10.1111/JTH.12225
Paolicelli RC, Sierra A, Stevens B, Tremblay ME, Aguzzi A, Ajami B, Amit I, Audinat E, Bechmann I, Bennett M, Bennett F, Bessis A, Biber K, Bilbo S, Blurton-Jones M, Boddeke E, Brites D, Brône B, Brown GC, Butovsky O, Carson MJ, Castellano B, Colonna M, Cowley SA, Cunningham C, Davalos D, De Jager PL, de Strooper B, Denes A, Eggen BJL, Eyo U, Galea E, Garel S, Ginhoux F, Glass CK, Gokce O, Gomez-Nicola D, González B, Gordon S, Graeber MB, Greenhalgh AD, Gressens P, Greter M, Gutmann DH, Haass C, Heneka MT, Heppner FL, Hong S, Hume DA, Jung S, Kettenmann H, Kipnis J, Koyama R, Lemke G, Lynch M, Majewska A, Malcangio M, Malm T, Mancuso R, Masuda T, Matteoli M, McColl BW, Miron VE, Molofsky AV, Monje M, Mracsko E, Nadjar A, Neher JJ, Neniskyte U, Neumann H, Noda M, Peng B, Peri F, Perry VH, Popovich PG, Pridans C, Priller J, Prinz M, Ragozzino D, Ransohoff RM, Salter MW, Schaefer A, Schafer DP, Schwartz M, Simons M, Smith CJ, Streit WJ, Tay TL, Tsai LH, Verkhratsky A, von Bernhardi R, Wake H, Wittamer V, Wolf SA, Wu LJ, Wyss-Coray T (2022) Microglia states and nomenclature: A field at its crossroads. Neuron 110:3458–3483. doi: 10.1016/J.NEURON.2022.10.020
Park GH, Maeno-Hikichi Y, Awano T, Landmesser LT, Monani UR (2010) Reduced survival of motor neuron (SMN) protein in motor neuronal progenitors functions cell autonomously to cause spinal muscular atrophy in model mice expressing the human centromeric (SMN2) gene. Journal of Neuroscience 30:12005–12019. doi: 10.1523/JNEUROSCI.2208-10.2010
Paul G, Elabi OF (2022) Microvascular Changes in Parkinson’s Disease- Focus on the Neurovascular Unit. Front Aging Neurosci 14:853372. doi: 10.3389/FNAGI.2022.853372/XML/NLM
Paulus P, Jennewein C, Zacharowski K (2011) Biomarkers of endothelial dysfunction: can they help us deciphering systemic inflammation and sepsis? Biomarkers 16. doi: 10.3109/1354750X.2011.587893
Petersen MA, Ryu JK, Akassoglou K (2018) Fibrinogen in neurological diseases: mechanisms, imaging and therapeutics. Nat Rev Neurosci 19:283. doi: 10.1038/NRN.2018.13
Petersen MA, Ryu JK, Akassoglou K (2018) Fibrinogen in neurological diseases: mechanisms, imaging and therapeutics. Nat Rev Neurosci 19. doi: 10.1038/nrn.2018.13
Pinsky DJ, Naka Y, Liao H, Oz MC, Wagner DD, Mayadas TN, Johnson RC, Hynes RO, Heath M, Lawson CA, Stern DM (1996) Hypoxia-induced exocytosis of endothelial cell Weibel-Palade bodies. A mechanism for rapid neutrophil recruitment after cardiac preservation. J Clin Invest 97:493–500. doi: 10.1172/JCI118440
Reilly A, Yaworski R, Beauvais A, Schneider BL, Kothary R (2024) Long term peripheral AAV9-SMN gene therapy promotes survival in a mouse model of spinal muscular atrophy. Hum Mol Genet 33:510–519. doi: 10.1093/HMG/DDAD202
Reinhart K, Bayer O, Brunkhorst F, Meisner M (2002) Markers of endothelial damage in organ dysfunction and sepsis. Crit Care Med 30. doi: 10.1097/00003246-200205001-00021
Rindt H, Feng Z, Mazzasette C, Glascock JJ, Valdivia D, Pyles N, Crawford TO, Swoboda KJ, Patitucci TN, Ebert AD, Sumner CJ, Ko CP, Lorson CL (2015) Astrocytes influence the severity of spinal muscular atrophy. Hum Mol Genet 24:4094. doi: 10.1093/HMG/DDV148
Rudnik-Schöneborn S, Heller R, Berg C, Betzler C, Grimm T, Eggermann T, Eggermann K, Wirth R, Wirth B, Zerres K (2008) Congenital heart disease is a feature of severe infantile spinal muscular atrophy. J Med Genet 45:635–638. doi: 10.1136/jmg.2008.057950
Rudnik-Schöneborn S, Vogelgesang S, Armbrust S, Graul-Neumann L, Fusch C, Zerres K (2010) Digital necroses and vascular thrombosis in severe spinal muscular atrophy. Muscle Nerve 42:144–147. doi: 10.1002/mus.21654
Rusu L, Minshall RD (2018) Endothelial Cell von Willebrand Factor Secretion in Health and Cardiovascular Disease. In: Endothelial Dysfunction - Old Concepts and New Challenges. IntechOpen, pp 147–162
Ryu JK, Mclarnon JG (2009) A leaky blood-brain barrier, fibrinogen infiltration and microglial reactivity in inflamed Alzheimer’s disease brain. J Cell Mol Med 13:2911–2925. doi: 10.1111/j.1582-4934.2008.00434.x
Shababi M, Habibi J, Ma L, Glascock JJ, Sowers JR, Lorson CL (2012) Partial restoration of cardio-vascular defects in a rescued severe model of spinal muscular atrophy. J Mol Cell Cardiol 52:1074–1082. doi: 10.1016/j.yjmcc.2012.01.005
Sintusek P, Catapano F, Angkathunkayul N, Marrosu E, Parson SH, Morgan JE, Muntoni F, Zhou H (2016) Histopathological defects in intestine in severe spinal muscular atrophy mice are improved by systemic antisense oligonucleotide treatment. PLoS One 11:1–15. doi: 10.1371/journal.pone.0155032
Somers E, Lees RD, Hoban K, Sleigh JN, Zhou H, Muntoni F, Talbot K, Gillingwater TH, Parson SH (2016) Vascular Defects and Spinal Cord Hypoxia in Spinal Muscular Atrophy. Ann Neurol 79:217–230. doi: 10.1002/ana.24549
Steinman J, Sun HS, Feng ZP (2021) Microvascular Alterations in Alzheimer’s Disease. Front Cell Neurosci 14:618986. doi: 10.3389/FNCEL.2020.618986/XML/NLM
Sweeney MD, Kisler K, Montagne A, Toga AW, Zlokovic B V. (2018) The role of brain vasculature in neurodegenerative disorders. Nature Neuroscience 2018 21:10 21:1318–1331. doi: 10.1038/s41593-018-0234-x
Sweeney MD, Zhao Z, Montagne A, Nelson AR, Zlokovic B V. (2019) Blood-brain barrier: From physiology to disease and back. Physiol Rev 99:21–78. doi: 10.1152/PHYSREV.00050.2017/ASSET/IMAGES/LARGE/Z9J0041828730010.JPEG
Takata F, Nakagawa S, Matsumoto J, Dohgu S (2021) Blood-Brain Barrier Dysfunction Amplifies the Development of Neuroinflammation: Understanding of Cellular Events in Brain Microvascular Endothelial Cells for Prevention and Treatment of BBB Dysfunction. Front Cell Neurosci 15:661838. doi: 10.3389/FNCEL.2021.661838/FULL
Thomsen MS, Routhe LJ, Moos T (2017) The vascular basement membrane in the healthy and pathological brain. Journal of Cerebral Blood Flow & Metabolism 37:3300. doi: 10.1177/0271678X17722436
Vischer UM, Jornot L, Wollheim CB, Theler JM (1995) Reactive Oxygen Intermediates Induce Regulated Secretion of von Willebrand Factor From Cultured Human Vascular Endothelial Cells. Blood 85:3164–3172. doi: 10.1182/BLOOD.V85.11.3164.BLOODJOURNAL85113164
Wadman RI, Jansen MD, Stam M, Wijngaarde CA, Curial CAD, Medic J, Sodaar P, Schouten J, Vijzelaar R, Lemmink HH, Van Den Berg LH, Groen EJN, Van Der Pol WL (2020) Intragenic and structural variation in the SMN locus and clinical variability in spinal muscular atrophy. Brain Commun 2:1–13. doi: 10.1093/braincomms/fcaa075
Waters S, Swanson MEV, Dieriks B V., Zhang YB, Grimsey NL, Murray HC, Turner C, Waldvogel HJ, Faull RLM, An J, Bowser R, Curtis MA, Dragunow M, Scotter E (2021) Blood-spinal cord barrier leakage is independent of motor neuron pathology in ALS. Acta Neuropathol Commun 9:1–17. doi: 10.1186/s40478-021-01244-0
Wijngaarde CA, Blank AC, Stam M, Wadman RI, Van Den Berg LH, Van Der Pol WL (2017) Cardiac pathology in spinal muscular atrophy: a systematic review. Orphanet J Rare Dis 12:67
Zhou H, Hong Y, Scoto M, Thomson A, Pead E, MacGillivray T, Hernandez-Gerez E, Catapano F, Meng J, Zhang Q, Hunter G, Shorrock HK, Ng TK, Hamida A, Sanson M, Baranello G, Howell K, Gillingwater TH, Brogan P, Thompson DA, Parson SH, Muntoni F (2022) Microvasculopathy in SMA is driven by a reversible autonomous endothelial cell defect. Journal of Clinical Investigation 132:1–15. doi: 10.1172/jci153430

No competing interests reported.

floatimage2.png
Supplementary Fig. 1 VWF expression is decreased in multiple regions of the SMA spinal cord (a-b) Representative micrographs of (a) control and (b) SMA dorsal horn of the spinal cord labelled with VWF, scale, 100µm. (c) Total VWF⁺ staining per FOV in control and SMA dorsal horn (Welch’s T test, *P <0.05), mean +SEM. (d) Mean vascular density (vessels/mm²) in dorsal horn +SEM, with linear regression, 95% CI. (e-f) Representative micrographs of VWF labelled (e) control and (f) SMA corticospinal tract of the spinal cord, scale, 100µm. (g) Total VWF⁺ staining per FOV in control and SMA corticospinal tract (Welch’s T test, where Pns >0.05), mean +SEM. (h) Mean vascular density (vessels/mm²) in corticospinal tract +SEM, with linear regression, 95% CI. Control n=6, SMA n=6. (i-j) Representative micrographs of VWF labelled (i) control and (j) SMA dorsal column of the spinal cord, scale, 100µm. (k) Individual case total VWF⁺ staining per FOV in control and SMA dorsal column (Welch’s T test, *P <0.05), mean +SEM. (l) Mean vascular density (vessels/mm²) in dorsal column +SEM, with linear regression, 95% CI. For all VWF+ Area/FOV analyses, Control n=5, SMA n=5. For all vascular density analyses, Control n=6, SMA n=6. (m) Representative micrograph of ki67 immunolabelled control spinal cord. White arrowheads show non-proliferative endothelial cells and red arrowhead shows a proliferating cell within the spinal cord parenchyma. (n) Representative micrograph of ki67 immunolabelled SMA spinal cord. Black arrowheads show proliferating endothelial cells.

Download PDF

Editorial decision: Revision requested
26 Dec, 2025
Reviews received at journal
11 Dec, 2025
Reviewers agreed at journal
24 Nov, 2025
Reviewers invited by journal
23 Nov, 2025
Editor assigned by journal
14 Nov, 2025
Submission checks completed at journal
13 Nov, 2025
First submitted to journal
11 Nov, 2025

You are reading this latest preprint version

Microvascular pathology in the spinal cord of severe spinal muscular atrophy patients

Status:

Version 1

Abstract

Figures

Introduction

Methods

Results

Discussion

Conclusion

Declarations

Declarations

Funding

Author Contribution

Acknowledgement

Data Availability

References

Additional Declarations

Supplementary Files

Status:

Version 1