Blood-spinal cord barrier breakdown and pericyte reductions in amyotrophic lateral sclerosis

Abstract

The blood-brain barrier and blood-spinal cord barrier (BSCB) limit the entry of plasma components and erythrocytes into the central nervous system (CNS). Pericytes play a key role in maintaining blood-CNS barriers. The BSCB is damaged in patients with amyotrophic lateral sclerosis (ALS). Moreover, transgenic ALS rodents and pericyte-deficient mice develop BSCB disruption with erythrocyte extravasation preceding motor neuron dysfunction. Here, we studied whether BSCB disruption with erythrocyte extravasation and pericyte loss are present in human ALS. We show that 11 of 11 cervical cords from ALS patients, but 0 of 5 non-neurodegenerative disorders controls, possess perivascular deposits of erythrocyte-derived hemoglobin and hemosiderin typically 10-50 μm in diameter suggestive of erythrocyte extravasation. Immunostaining for CD235a, a specific marker for erythrocytes, confirmed sporadic erythrocyte extravasation in ALS, but not controls. Quantitative analysis revealed a 3.1-fold increase in perivascular hemoglobin deposits in ALS compared to controls showing hemoglobin confined within the vascular lumen, which correlated with 2.5-fold increase in hemosiderin deposits (r = 0.82, p < 0.01). Spinal cord parenchymal accumulation of plasma-derived immunoglobulin G, fibrin and thrombin was demonstrated in ALS, but not controls. Immunostaining for platelet-derived growth factor receptor-β, a specific marker for CNS pericytes, indicated a 54 % (p < 0.01) reduction in pericyte number in ALS patients compared to controls. Pericyte reduction correlated negatively with the magnitude of BSCB damage as determined by hemoglobin abundance (r = -0.75, p < 0.01). Thus, the BSCB disruption with erythrocyte extravasation and pericyte reductions is present in ALS. Whether similar findings occur in motor cortex and affected brainstem motor nuclei remain to be seen.

Fig. 1

Extravascular deposition of erythrocyte-derived hemoglobin and erythrocyte extravasation in the spinal cord of ALS subjects. a Confocal microscopy analysis of erythrocyte-derived hemoglobin (red) and lectin-positive capillaries (green) in NNDC and sporadic ALS cervical spinal cord anterior horn. b Quantification of extravascular hemoglobin deposits in the cervical spinal cord anterior horn. Mean ± SEM, n = 5 NNDC, eight sporadic (sALS) and three familial (fALS) cases. c Confocal microscopy analysis of CD235a-positive erythrocytes (red) and lectin-positive capillaries (green) in a NNDC control and sporadic ALS cervical spinal cord anterior horn. d Extravasation of erythrocytes in ALS spinal cord anterior horn demonstrated by hematoxylin and eosin staining. In c and d, Arrows denote extravasated erythrocytes in ALS sample

Fig. 2

Perivascular hemosiderin deposition in the spinal cord of ALS subjects. a Bright field microscopy analysis of Prussian blue-positive hemosiderin deposits (blue) and podocalyxin-postive capillaries (brown) in NNDC and sporadic ALS cervical spinal cord anterior horn. b Quantification of perivascular Prussian blue-positive hemosiderin deposits in the cervical spinal cord anterior horn. Mean ± SEM, n = 5 NNDC, eight sporadic (sALS) and three familial (fALS) cases. c Positive correlation between perivascular hemosiderin deposits and extravascular hemoglobin deposition in cervical spinal cord anterior horn. Single data points derived from NNDC, sALS and fALS subjects. r Pearson’s coefficient

Fig. 3

Accumulation of plasma-derived proteins in the spinal cord of ALS subjects. a Confocal microscopy analysis of immunoglobulin G (IgG) (green) and hemoglobin (red) in NNDC and sporadic ALS cervical spinal cord anterior horn. Merged extravascular colocalization of IgG and hemoglobin, white lectin-positive capillary profiles. b Confocal microscopy analysis of plasma-derived fibrin (red), NeuN-positive neurons (green) and lectin-positive capillaries (blue) in NNDC and sporadic ALS cervical spinal cord anterior horn. c Confocal microscopy analysis of plasma-derived thrombin (green), SMI-311-positive neurons (green) and lectin-positive capillaries (blue) in NNDC and sporadic ALS cervical spinal cord anterior horn. Representative images are shown from 5 NNDC and 11 ALS cases

Fig. 4

Pericyte capillary reduction in ALS spinal cord correlates with magnitude of vascular disruption. a Representative confocal microscopy analysis of PDGFRβ-positive pericytes (green) and lectin-positive capillaries (red) in NNDC and sporadic ALS cervical spinal cord anterior horn. b Quantification of PDGFRβ-positive pericyte capillary coverage in cervical spinal cord anterior horn. Mean ± SEM, n = 5 NNDC, eight sporadic (sALS) and three familial (fALS) cases. c Quantification of PDGFRβ-positive pericyte cell number per mm² vascular surface area in cervical spinal cord anterior horn. Mean ± SEM, n = 5 NNDC, eight sALS and three fALS. d Negative correlation between extent of extravascular hemoglobin and pericyte coverage in cervical spinal cord anterior horn. Single data points derived from NNDC, sALS and fALS subjects. r Pearson’s coefficient