spinal cord barrier in ALS patients

brain research 1469 (2012) 114–128

Available online at www.sciencedirect.com

www.elsevier.com/locate/brainres

Research Report

Impaired blood–brain/spinal cord barrier in ALS patients Svitlana Garbuzova-Davisa,b,c,d,n, Diana G. Hernandez-Ontiverosa, Maria C.O. Rodriguesa,g, Edward Hallerf, Aric Frisina-Deyoa, Santhia Mirtyla, Sebastian Sallota, Samuel Saportad, Cesario V. Borlongana,b, Paul R. Sanberga,b,d,e a

Center of Excellence for Aging & Brain Repair, University of South Florida, Morsani College of Medicine, Tampa, FL 33612, USA Department of Neurosurgery and Brain Repair, University of South Florida, Morsani College of Medicine, Tampa, FL 33612, USA c Department of Molecular Pharmacology and Physiology, University of South Florida, Morsani College of Medicine, Tampa, FL 33612, USA d Department of Pathology and Cell Biology, University of South Florida, Morsani College of Medicine, Tampa, FL 33612, USA e Department of Psychiatry, University of South Florida, Morsani College of Medicine, Tampa, FL 33612, USA f Department of Integrative Biology, University of South Florida, Tampa, FL 33620, USA g ~ Preto School of Medicine, University of Sao Paulo, Brazil Department of Internal Medicine, Ribeirao b

ar t ic l e in f o

abs tra ct

Article history:

Vascular pathology, including blood–brain/spinal cord barrier (BBB/BSCB) alterations, has

Accepted 31 May 2012

recently been recognized as a key factor possibly aggravating motor neuron damage,

Available online 27 June 2012

identifying a neurovascular disease signature for ALS. However, BBB/BSCB competence in

Keywords:

sporadic ALS (SALS) is still undetermined. In this study, BBB/BSCB integrity in postmortem

Amyotrophic lateral sclerosis

gray and white matter of medulla and spinal cord tissue from SALS patients and controls

Patients

was investigated. Major findings include (1) endothelial cell damage and pericyte degen-

Blood–brain/spinal cord barrier

eration, (2) severe intra- and extracellular edema, (3) reduced CD31 and CD105 expressions

Impairment

in endothelium, (4) significant accumulation of perivascular collagen IV, and fibrin deposits (5) significantly increased microvascular density in lumbar spinal cord, (6) IgG microvascular leakage, (7) reduced tight junction and adhesion protein expressions. Microvascular barrier abnormalities determined in gray and white matter of the medulla, cervical, and lumbar spinal cord of SALS patients are novel findings. Pervasive barrier damage discovered in ALS may have implications for disease pathogenesis and progression, as well as for uncovering novel therapeutic targets. & 2012 Elsevier B.V. All rights reserved.

Abbreviations: ALS, amyotrophic lateral sclerosis; BBB, barrier; CD31,

PECAM-1; CD105,

EM,

electron microscopy; FALS, familial ALS; GAPDH,

IgG,

immunoglobulin G; JAM-1,

acid; MVD,

blood–brain barrier; BM,

endoglin; CNS, central nervous system; CSF,

basement membrane; BSCB,

cerebrospinal fluid; EC,

glyceraldehyde-3-phosphate dehydrogenase; G93A,

junctional adhesion molecule-1; MMP,

microvascular density; NGS, normal goat serum; RT,

blood–spinal cord

endothelial cell; glycine-93 to alanine;

matrix metalloproteinase; mRNA, messenger ribonucleic

room temperature; SALS, sporadic ALS; SOD1,

superoxide

dismutase 1; TIMP, tissue inhibitor of metalloproteinase; VE-cadherin, vascular endothelial cadherin; ZO-1, zonula occludens 1 n Corresponding author at: Center of Excellence for Aging and Brain Repair, Department of Neurosurgery and Brain Repair, University of South Florida, Morsani College of Medicine, 12901 Bruce B. Downs Blvd. Tampa, FL 33612, USA. Fax: þ1 813 974 3078. E-mail address: [email protected] (S. Garbuzova-Davis). 0006-8993/$ - see front matter & 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.brainres.2012.05.056

brain research 1469 (2012) 114–128

1.

Introduction

Amyotrophic lateral sclerosis (ALS) is a fatal disease characterized by progressive motor neuron degeneration in the brain and spinal cord leading to muscle atrophy, paralysis and death. Sporadic ALS (SALS) dominates, with only 5–10% of cases genetically linked (FALS); 20% of FALS cases show mutations in the Cu/Zn superoxide dismutase (SOD1) gene (Rosen et al., 1993). Clinical presentation and pathology of SALS and FALS are similar. Although numerous hypotheses have been proposed regarding ALS etiopathology (Bruijn et al., 2004; Cleveland and Rothstein, 2001; Rothstein, 2009; Strong et al., 2005), the causes of motor neuron degeneration and pathogenic mechanisms are still uncertain. Existing therapies for ALS are largely palliative. Determining putative mechanism(s) of motor neuron degeneration in ALS is challenging due to the diffuse motor neuron death and the complexity of disease manifestation. Vascular impairment has only recently been recognized as a key factor in ALS, identifying a neurovascular disease signature (GarbuzovaDavis et al., 2011). Evidence of compromised blood–brain barrier (BBB) and/or blood–spinal cord barrier (BSCB) integrity was recently identified in ALS patients and in an animal model of ALS. Structural and functional BBB/BSCB impairments were demonstrated in an animal model of ALS at early stage disease and worsened with disease progression (Garbuzova-Davis et al., 2007a, 2007b; Nicaise et al., 2009a, 2009b; Zhong et al., 2008). Additionally, vascular leakage, decreased capillary length and blood flow, microhemorrhages, reduced expression of basement membrane components and tight junction proteins have been shown in the spinal cords of SOD1 transgenic animals. Importantly, BSCB breakdown was found in SOD1 mutant mice and rats prior to motor neuron degeneration and neurovascular inflammation (Miyazaki et al., 2011; Nicaise et al., 2009a; Zhong et al., 2008), suggesting vascular alteration as an early ALS pathological event. Evidence of BBB/BSCB impairment has also been observed in postmortem tissue from ALS patients. Loss of endothelium integrity as shown by significant reductions of occludin and ZO-1 mRNA was recently observed in spinal cords from ALS patients (Henkel et al., 2009). Decreased perivascular occludin and collagen IV, as well as astrocyte end-feet dissociated from the endothelium were also seen in postmortem ALS spinal cord tissue (Miyazaki et al., 2011). These results strengthen the likelihood that barrier disruption contributes to disease progression and knowledge of this damage may lead to novel therapeutic targets. Thus, impairment of the BBB/BSCB, preceding CNS entry of blood-borne toxins could be a key early factor in ALS pathogenesis, accelerating motor neuron death. The majority of findings on microvascular pathology in ALS, including BBB/BSCB alterations, have been established in mutant SOD1 rodent models, identifying barrier damage during disease development which might similarly occur in FALS patients carrying the SOD1 mutation. However, BBB/BSCB competence, as part of the tightly integrated neurovascular unit, is still largely a mystery in sporadic ALS cases. The aim of this study was to determine integrity of the BBB/BSCB in postmortem tissues from patients with the sporadic form of ALS. A specific focus was analyzing barrier

115

competence in the gray and white matter of the brainstem and spinal cord, CNS structures known to experience motor neuron pathology.

2.

Results

2.1. Ultrastructure of the microvasculature in the brain and spinal cord of ALS patients Ultrastructural integrity of the vessels in the gray and white matter of the brainstem (medulla) and spinal cords (cervical and lumbar) was examined in postmortem tissues from ALS patients and controls using electron microscopy. Microvascular analysis in the gray matter was performed in areas close to motor neuron pools. Microvessels in the white matter were studied in the pyramidal tract (medulla) and lateral funiculi (spinal cord).

2.1.1.

Medulla

Capillaries in control medulla gray and white matter tissues consisted of blood vessels, usually lined with a single layer of endothelial cells (ECs) surrounded by a layer of basement membrane (BM) (Fig. 1A, B). In some capillaries, additional layers of endothelium and BM were observed. Tight junctions were visible between ECs in vessel lumens. Pericytes with well defined nuclei were enveloped by BM. The capillaries were surrounded by astrocytes and myelinated nerve fibers. The neuropil surrounding capillaries was in close, continuous contact with the external BM. Occasional small accumulations of collagen were found between layers of BM, likely due to aging. In tissue from ALS patients, ultrastructure of capillaries in the gray and white matter of the medulla varied. Some capillaries had normal morphology, a single layer of endothelium surrounded by BM and pericyte cytoplasm displaying little evidence of collagen deposits between the EC and BM. Astrocyte cell cytoplasm was in close contact with these capillaries. Other capillaries showed endothelial and pericyte cell damage (Fig. 1C, D). Signs of complete pericyte degeneration and fragments of pericytes in the adjacent extracellular space under the BM were noted. In some capillaries, multiple EC layers, indicating a replication process, were separated by sheets of BM material. Edema was evident in the spaces between the endothelial cells and some endothelial cells contained lipofuscin inclusions (Fig. 1D). Significant protein-filled edema containing free-floating swollen mitochondria surrounded the medulla capillaries (Fig. 1C). In numerous medullar capillaries from ALS patients, large accumulations of disorganized collagen were observed between BM covering the ECs and the limiting BM forming the blood–neural tissue barrier (Fig. 1E, F). Also, both fibrin and Lewy bodies were observed in close proximity to capillaries in some areas of the medulla.

2.1.2.

Cervical spinal cord

The control cervical spinal cord capillaries in gray and white matter consisted of blood vessels with a single layer of ECs surrounded by a BM layer. On occasion, additional layers of endothelium and BM were observed surrounding the capillaries. Parietal cells covered in BM were also observed in gray matter (Fig. 2A). In addition, occasional small pockets of

116

brain research 1469 (2012) 114–128

Fig. 1 – Electron microscope examination of the capillaries in gray and white matter of the medulla from ALS patients. Representative capillaries in gray (A) and white (B) matter of the control medulla were characterized by normal ultrastructural appearance consisting of ECs and tight junctions, surrounded by a single layer of BM, astrocytes, neuropil, and myelinated nerve fibers. Pericytes with well defined nuclei were seen under the capillary BM. Organelles in all cells were well formed. In ALS patients, a capillary in the gray matter (C) of the medulla showed pericyte cell damage; edema was apparent between pericytes. Significant protein-filled edema surrounded the capillaries, which contained free-floating swollen mitochondria. Tight junctions between ECs were well defined. Many myelinated nerves had degenerated. A capillary in the white matter (D) demonstrated complete pericyte degeneration and pericyte fragments were noted in the adjacent extracellular space under the BM. Above and to the left of the capillary, a Lewy body can be seen. Edema was evident between the endothelial cells and outside the capillary. Endothelial cells were swollen and contained lipofuscin inclusions. A large accumulation of disorganized collagen was observed between BM covering the ECs in the gray (E) and white (F) matter of the medulla. EC—endothelial cell, BM—basement membrane, TJ—tight junction, A—axon, P—pericyte, Nu—nucleus, M—mitochondria, Lf—lipofuscin inclusion. Arrowheads in (C) and (D) indicate extracellular and intracellular edema. Magnification in (A), (D) is 5600; in (B), (E) is 4400; in (C), (F), is 3500. collagen were observed between layers of BM in both gray and white matter. The neuropil surrounding capillaries was in close, continual contact with the external BM (Fig. 2B). Ultrastructure of capillaries in gray and white matter of the cervical spinal cord from ALS patients also varied. Some capillaries resembled those observed in control tissue. However, swollen ECs and complete pericyte degeneration were noted under BM in numerous gray matter capillaries (Fig. 2C). Tissue edema with large extracellular spaces at neuropil locations and surrounding the capillaries in gray matter of cervical spinal cord regions was also revealed. In white matter, most

capillaries displayed a thin layer of ECs (Fig. 2D). Basement membranes in these capillaries were separated by collagen deposits as well as by lipofuscin inclusions. In other gray and white matter capillaries from ALS cervical spinal cords, large accumulations of collagen were observed between the basement membrane covering the ECs and the limiting BM forming the blood–neural tissue barrier (Fig. 2E, F).

2.1.3.

Lumbar spinal cord

Similar to medulla and cervical spinal cord tissues, capillaries in control lumbar spinal cord consisted of small blood vessels

brain research 1469 (2012) 114–128

117

Fig. 2 – Electron microscope examination of the capillaries in gray and white matter of the cervical spinal cord from ALS patients. The control cervical spinal cord capillaries in gray matter (A) showed normal ultrastructure consisting of blood vessels with a single layer of ECs surrounded by a layer of BM. Morphologically normal pericytes were enveloped by BM. (B) In a longitudinal section through a capillary in the white matter, an endothelial cell and pericyte were surrounded by a single layer of BM. Several erythrocytes were observed in the lumen of the capillary. Occasional small accumulations of collagen were found between layers of BM. The neuropil surrounding capillaries was in close, continuous contact with the external BM. (C) Capillaries in gray matter of the cervical spinal cord from ALS patients showed swollen ECs and pericyte degeneration. Lipid droplet was determined in ECs. Tissue edema with large extracellular spaces at neuropil locations and surrounding the capillaries in gray matter of cervical spinal cord regions was also revealed. A tight junction between ECs was visible. In white matter (D), a capillary displays a single thin layer of ECs. Basement membranes were separated by collagen deposits. Lipofuscin inclusions were observed within BM collagen expansion. In another capillary in gray matter (E) from the cervical spinal cord of an ALS patient, large accumulations of collagen were accompanied by degenerated pericytes and vacuolated ECs. Also, calcification of collagen fibers was determined. (F) Similarly, a white matter capillary demonstrated extensive collagen accumulation between the BM covering the ECs and the pericyte. EC—endothelial cell, BM—basement membrane, TJ—tight junction, A—axon, P—pericyte, Nu—nucleus, E—erythrocyte, L—lipid droplet, asterisks—lipofuscin inclusions. Arrowheads in (C) indicate extracellular edema. Magnification in (A), (E) is 7100; in (C) is 5600; in (B) is 4400; in (D) is 3500; in (F) is 2800.

in gray and white matter, usually with a single endothelium layer, surrounded by a layer of BM (Fig. 3A, B). Around the capillaries, parietal cells were observed under another BM layer. Astrocyte cell processes were adjacent to the outer surface of the capillary BM (Fig. 3A). Occasional thin layers of collagen, in addition to parietal cell cytoplasm, were observed surrounding capillaries.

In ALS patients, capillaries in the lumbar spinal cord also displayed varied ultrastructure. Some capillaries showed normal endothelium and a thin layer of pericyte cytoplasm comprising the blood–spinal cord barrier. However, intra- and extracellular edema was noted in numerous gray and white matter capillaries (Fig. 3C, D). Degenerated pericytes and EC, indicated by condensed cell cytoplasm and vacuolization, were

118

brain research 1469 (2012) 114–128

Fig. 3 – Electron microscope examination of the capillaries in gray and white matter of the lumbar spinal cord from ALS patients. Similar to medulla and cervical spinal cord tissues, capillaries in control lumbar spinal cord consisted of blood vessels in gray (A) and white matter (B) with a normal-appearing single layer endothelium, surrounded by a layer of BM. Healthy pericytes were observed under another layer of BM. Astrocyte cell processes were adjacent to the outer surface of the capillary BM (A). In ALS patients, a gray matter capillary (C) in the lumbar spinal cord intra- and extracellular edema was noted. An endothelial cell in a capillary has lifted off the basement membrane, creating a fluid-filled pocket in the capillary. Degenerated pericytes and EC vacuolization were observed. Ruptured luminal EC membrane was determined. A capillary in the white matter (D) showed a developing sheath of degenerated cellular debris from a pericyte trapped under the BM. Vacuolated ECs and pericytes were observed. A large lipofuscin deposition was seen in the pericyte cytoplasm. There is a large autophagosome in the EC at the # sign, indicating cell stress. In the lumbar spinal cord capillaries of ALS patients in both gray (E) and white (F) matter, large accumulations of disorganized collagen were observed between BMs. Dark fibrin filaments can be noted within the capillary BM, to the left, in (E), and in extracellular space. Calcified collagen fibers and a degenerated pericyte were noted in (F). EC—endothelial cell, BM—basement membrane, Ast—astrocyte, P—pericyte, Nu—nucleus, asterisks in C—ruptured luminal EC membrane, asterisks in D—vacuoles in pericytes; asterisk in E—lipofuscin inclusion; black # in D—autophagosome in EC. Arrowheads in (C), (D) indicate intra- and extracellular edema. Magnification in (A), (C) is 7100; in (C) is 5600; in (D), (F) is 5600; in (B) is 3500; in (E) is 2800.

also observed. Some capillaries in the lumbar spinal cord of ALS patients showed a developing sheath of cellular debris trapped under the BM that separates the capillary from the spinal cord. Also, ruptured luminal EC membranes were observed in gray

matter capillaries of the lumbar spinal cord (Fig. 3C). In this ruptured capillary, the EC was detached from the BM, trapping plasma proteins. In numerous gray and white matter capillaries from ALS lumbar spinal cord tissue, large accumulations of

brain research 1469 (2012) 114–128

disorganized collagen were observed between BM covering the ECs (Fig. 3E, F). Fig. 3E shows fibrin accumulation within capillary BM and in extracellular space. Fig. 3F also shows pericyte cell debris to the right of the capillary. Importantly, besides large collagen basement membrane accumulations in the majority of microvessels in gray and white matter of the medulla and cervical/lumbar spinal cords from ALS patients, some capillaries contained calcified collagen fibers. Interestingly, severe collagen calcification (Fig. 4A, C, D) was accompanied by the appearance of a Lewy body under the capillary BM in the medulla white matter (Fig. 4B) or outside a spinal cord capillary (Fig. 4D) in one ALS patient. Moreover, Lewy body calcification was noted in the lumbar spinal cord (Fig. 4D). In this particular case, lipofuscin deposits were found within EC and pericytes (Fig. 4B, C) and in the interstitial spaces. The large lipofuscin/lipid deposit in the endothelial cell in Fig. 4B would have resulted in rupture and death of the cell. Also, a large intracellular accumulation of lipofuscin noted in a pericyte may have precipitated cell rupture (Fig. 4C). Scattered areas of fibrin deposition were also observed throughout the medulla and

119

spinal cord in the extracellular tissue in close association with capillaries. In summary, capillary ultrastructural abnormalities were clearly detected in postmortem gray and white matter of the brainstem (medulla) and spinal cord (cervical and lumbar) in ALS patients. Observed vascular endothelium damage led to vascular leakage in the brain and spinal cord.

2.2.

Microvascular collagen expansion

Since we observed a large microvascular accumulation of collagen, measurement of collagen expansion was performed in gray and white matter of the medulla and cervical/lumbar spinal cords of ALS patients and controls via electron micrographs. Results showed a significant increase, 2–2.5 times, of basement membrane collagen accumulation in the majority of medulla (po0.05), cervical (po0.001), and lumbar (po0.001) spinal cord gray and white matter microvessels from ALS patients compared to controls (Fig. 5A). More microvascular collagen expansion was observed in gray matter of the cervical and lumbar spinal cords of ALS patients (2.5470.14 mm and

Fig. 4 – Electron microscope examination of the capillaries in white matter of the medulla, cervical and lumbar spinal cord from ALS patient: collagen calcification and Lewy body. Severe collagen accumulation and calcification (A) were observed in medulla white matter. (B) A Lewy body under the capillary BM and near degenerated pericytes was also determined in the medulla white matter. Swollen ECs and a large lipofuscin/lipid deposit were seen in the same capillary. In white matter of the cervical spinal cord capillary (C), calcification of collagen was also determined along with significant lipofuscin deposits within pericytes. This large intracellular accumulation of lipofuscin noted in a pericyte may have precipitated cell rupture. In the lumbar spinal cord, a white matter capillary (D) also displayed collagen calcifications. A large calcified Lewy body was distinguished outside a lumbar spinal cord capillary. EC—endothelial cell, BM—basement membrane, A—axon, P—pericyte, Nu—nucleus, Lf—lipofuscin inclusion; LB—Lewy body. Arrowheads in (A), (C), (D) indicate collagen calcification. ncalcification in the Lewy body. Magnification in (A), (B) is 8900; in (C) is 2800; in (D) is 3500.

120

brain research 1469 (2012) 114–128

Fig. 5 – Microvascular collagen expansion in the capillaries in the medulla, cervical and lumbar spinal cord from ALS patients. (A) Significant increase of basement membrane collagen accumulation was noted in the majority of medulla (po0.05), cervical (po0.001), and lumbar (po0.001) spinal cord gray and white matter microvessels from ALS patients compared to controls. More microvascular collagen expansion was observed in the cervical and lumbar spinal cords of ALS patients than in controls. po0.05, po0.001. (B) Immunohistochemical analysis for collagen IV confirmed perivascular collagen type IV expansion in gray (arrows in b, d, and f) and white (arrows in h, j, and l) matter of the medulla, cervical and lumbar spinal cord from ALS patients, corroborating the ultrastructural observations. Magnification bar in a–l is 25 lm.

2.3870.13 mm, respectively) compared to controls (0.7470.09 mm and 0.7670.08 mm, respectively). Significantly (po0.001) higher basement membrane collagen accumulation was determined in white matter of the cervical spinal cord vs. white matter capillaries in the medulla or lumbar spinal cord from ALS patients. Immunohistochemical analysis for collagen IV confirmed the presence of this collagen type in the expanded BM in gray and white matter of the medulla and cervical/lumbar spinal cord from ALS patients (Fig. 5B), corroborating the ultrastructural observations.

(po0.0001) two fold increase of MVD in the lumbar spinal cord of ALS patients (601.74731.04 vessels/mm2) compared to controls (308.77731.76 vessels/mm2) (Fig. 6A). In the medulla and cervical spinal cord, there were no differences in MVD between ALS patients and controls. Fig. 6B demonstrates vascular profiles in the gray matter of the medulla and cervical/lumbar spinal cords from ALS patients and controls.

2.3.

Immunohistochemical staining for capillary endothelium in the gray and white matter of the medulla, cervical and lumbar spinal cords was performed with CD31 (PECAM-1) and CD105 (endoglin) markers. The CD31 expressions are shown as dotted lines along the endothelial membrane surface in both gray (Fig. 7A: a, c, e) and white (Fig. 7A: g, i, k)

Microvascular density

Microvascular density (MVD) was measured in the gray matter of the medulla, cervical and lumbar spinal cords from ALS patients and controls using 0.1% Sirius Red staining for basement membrane collagen. Results showed a significant

2.4. Vascular integrity in the gray and white matter of the medulla and spinal cords

brain research 1469 (2012) 114–128

121

leakage in both gray (Fig. 7C: b, d, f) and white (Fig. 7C: h, j, l) matter.

2.5. Characteristics of endothelial cell tight junction and adhesion proteins

Fig. 6 – Microvascular density in the gray matter of the medulla, cervical and lumbar spinal cord from ALS patients. Microvascular density (MVD) was measured in the gray matter of the medulla, cervical and lumbar spinal cords from ALS patients and controls using 0.1% Sirius Red staining for basement membrane collagen. (A) Significant (po0.0001) two fold increase MVD in the lumbar spinal cord was determined in ALS patients compared to controls. In the medulla and cervical spinal cord, there were no differences in MVD between ALS patients and controls. (B) Sirius Red staining demonstrated vascular profiles in the gray matter of the medulla, cervical and lumbar spinal cords from ALS (B, D, F) patients and (A, C, E) controls. GM—gray matter, WM—white matter. Magnification bar in A–F is 200 lm. matter capillaries of control medulla and cervical/lumbar spinal cords. Capillaries in the gray (Fig. 7A: b, d, f) and white (Fig. 7A: h, j, l) matter tissues from ALS patients presented interruptions of CD31 expression in the endothelial lining, indicating endothelial breakage. The ALS tissues also showed asymmetrical increases in capillary wall thickness. Expressions of CD105 are demonstrated as a continuous line along the endothelial surface in capillaries of the gray (Fig. 7B: a, c, e) and white (Fig. 7B: g, i, k) matter in control tissues. Less defined CD105 vessel margins with discontinuity of the endothelial lining were determined in the medulla, cervical and spinal cords, in both gray (Fig. 7B: b, d, f) and white (Fig. 7B: h, j, l) matter, from ALS patients. An asymmetrical increase of CD105 expression in the capillary wall extension was also observed in ALS tissues. Microvascular integrity was also analyzed by IgG staining. In control tissues, IgG was mostly limited to the capillary lumen in the gray (Fig. 7C: a, c, e) and white (Fig. 7C: g, i, k) matter of medulla and cervical/lumbar spinal cords. Vessel walls were sharply outlined. In contrast, numerous capillaries in the medulla, cervical and lumbar spinal cords of the ALS tissues appeared blurry and IgG was dispersed and intruded into the abluminal side of capillaries, indicating microvascular

Tight junction (ZO-1, occludin, claudin-5, and JAM-1) and adhesion (VE-cadherin) proteins were analyzed in separated gray and white matter of the medulla and cervical/lumbar spinal cords from ALS patients and controls using Western blot. The ratio between the band density for each protein vs. GAPDH, a constitutively expressed normalizing protein, was determined. Results showed significant decreases of most tight junction and adhesion protein expressions in both gray and white matter of the medulla and cervical/lumbar cords from ALS patients. The major downregulation of ZO-1 was determined in ALS gray matter of the medulla (po0.01), cervical (po0.05), and lumbar (po0.001) tissues (Fig. 8). In the white matter capillaries, a significant (po0.05) decrease of ZO-1 was seen in the cervical and lumbar spinal cords from ALS patients. Occludin protein expression was significantly (po0.05) reduced only in the medulla and cervical spinal cord in both gray and white matter. Downregulation of claudin-5 was demonstrated in gray and white matter medulla (po0.05) and gray matter cervical spinal cord (po0.05) from ALS patients. No significant differences in protein expression were found in lumbar spinal cord from ALS patients vs. controls for occludin and claudin-5. A significant (po0.01) decrease of JAM-1 was found only in gray matter of the cervical and lumbar spinal cords from ALS patients. However, VE-cadherin expression was significantly (po0.05) downregulated in white matter of ALS medulla and cervical spinal cord. Thus, differences in reductions of tight junction and adhesion protein expressions between the gray and white matter of the medulla, cervical and lumbar spinal cords were determined in ALS patients vs. controls. In the medulla, downregulations of ZO-1, occludin, and claudin-5 were shown in the gray matter and occludin, claudin-5, and VE-cadherin in the white matter. In the cervical spinal cord of ALS patients, diminished expressions of all tight junction, but not adhesion, proteins were demonstrated in gray matter. Lessened expressions of ZO-1, occludin, and VE-cadherin were established in white matter of ALS cervical spinal cord. Interestingly, only diminished ZO-1 expression was shown in both gray and white matter of the lumbar spinal cord from ALS patients whereas reduced JAM-1 expression was determined in gray matter.

3.

Discussion

In the present study, we investigated BBB/BSCB integrity in postmortem tissues from patients with sporadic ALS. A specific focus was determining barrier competence in gray and white matter of the brainstem (medulla) and spinal cord (cervical and lumbar). Major findings include (1) endothelial cell damage and pericyte degeneration compromising BBB/ BSCB integrity, (2) endothelial cells with numerous cytoplasmic vacuoles and membrane rupture, (3) severe edema in the medulla and spinal cord, (4) significant lipofuscin deposits

122

brain research 1469 (2012) 114–128

Fig. 7 – Vascular integrity in the medulla, cervical and lumbar spinal cord from ALS patients. (A) Immunohistochemical staining for CD31 (PECAM-1) expressions were evidenced as a dotted line (arrows) along the endothelial membrane surface in both gray (a, c, e) and white (g, i, k) matter capillaries of control medulla and cervical/lumbar spinal cords. Capillaries in the gray (b, d, f) and white (h, j, l) matter tissues from ALS patients presented interruptions (asterisks) of CD31 expression in the endothelial lining, indicating endothelial breakage. The ALS tissues also showed an asymmetrical increase in the extension of the capillary walls. Magnification bar in a–l is 25 lm. (B) Immunohistochemical staining for CD105 (endoglin) demonstrated as a continuous line (arrows) along the endothelial surface in capillaries of the gray (a, c, e) and white (g, i, k) matter in control tissues. Less defined CD105 vessel margins with discontinuity (asterisks) of the endothelial lining was determined in the medulla, cervical and spinal cords, both gray (b, d, f) and white (h, j, l) matter, from ALS patients. The asymmetrical increase of CD105 expression in the capillary wall extension was also observed in ALS tissues. Magnification bar in a–l is 25 lm. (C) Immunohistochemical staining for IgG in control tissues was mostly limited to within the capillary lumen (arrows) in the gray (a, c, e) and white (g, i, k) matter of medulla and cervical/lumbar spinal cords. The vessel walls were sharply outlined. Numerous capillaries in the medulla, cervical and lumbar spinal cords in the ALS tissues appeared blurry and IgG was dispersed and intruding into the abluminal side (asterisks) of capillaries, indicating microvascular leakage in both gray (b, d, f) and white (h, j, l) matter. Magnification bar in a–l is 25 lm.

within endothelial cells and pericytes, (5) significant accumulations of basement membrane collagen IV and fibrin deposits, (6) significantly increased microvascular density in the lumbar spinal cord, (7) reduced CD31 and CD105 expressions in endothelium, (8) IgG microvascular leakage in the medulla and spinal cord, (8) reductions in most tight junction and adhesion protein expressions in the medulla and spinal cord. Microvascular barrier abnormalities determined in both gray and white matter of the medulla, cervical, and lumbar spinal cord from sporadic ALS patients are novel findings. Newly discovered widespread microvascular damage may impact ALS pathogenesis.

The unique microvascular barriers in the brain (BBB) and spinal cord (BSCB) provide selective transport of molecules and cells from the systemic compartment and prevent passive diffusion of harmful blood solutes (Ballabh et al., 2004; Pardridge, 1999; Vorbrodt and Dobrogowska, 2003). Specific transport systems allow influx of required substances and efflux of cell waste, maintaining CNS homeostasis (Begley and Brightman, 2003; Begley, 2004). BBB/BSCB structural elements – endothelial cells, tight and adherens junctions between endothelial cells, basement membrane, astrocyte end-feet, and pericytes – form a tightly integrated unit regulating the CNS environment. BBB/BSCB dysfunction

brain research 1469 (2012) 114–128

or structural damage would likely lead to an increasingly toxic CNS environment. In this study, analysis of capillaries in the medulla, cervical and lumbar spinal cords from SALS patients demonstrated severe BBB/BSCB impairment in gray and white matter. Although this postmortem examination may not reveal the role of barrier damage in ALS, a compromised BBB/BSCB could accelerate motor neuron degeneration, allowing CNS entry of harmful blood-borne solutes and restricting efflux of cellular waste. Our major finding is endothelial cell damage by vacuolization and even cell membrane rupture which likely impair cell function and may lead to the observed vascular leakage of endogenous IgG. Pericyte deterioration identified by ultrastructural analysis could also significantly contribute to ALS barrier damage. Perivascular collagen expansion in ALS deserves special attention. In our study, significant collagen type IV accumulation and even calcification, extending the basement membrane, were noted in numerous gray and white matter capillaries of the medulla, cervical and lumbar spinal cords. Similar collagen accumulation in cerebral vascular basement membranes and focal necrotic changes in endothelial cells were reported in brain tissues from Alzheimer’s patients (Claudio, 1996). However, Miyazaki et al. (2011) noted decreased perivascular collagen IV in postmortem spinal cord tissues from a small cohort of ALS patients. Ono et al. (1998) also reported reduced perivascular collagen levels in ALS patient spinal cords. The discrepancy between these study results and ours needs clarification. Major basement membrane components: laminin, collagen type IV, entactin, and heparan sulfate proteoglycans (Dermietzel and Krause, 1991; Rutka et al., 1988) support the abluminal surface of the endothelium and restrict passage of macromolecules and cells (Pozzi and Zent, 2009) through the BBB/BSCB. Previous reports of decreased perivascular collagen (Miyazaki et al., 2011; Ono et al., 1998) might reflect matrix metalloproteinase (MMP) activity degrading this basement membrane component. High levels of MMP-2 and MMP-9 have been shown in the CNS tissue, serum and CSF of ALS patients and animal models (Beuche et al., 2000; Fang et al., 2010; Soon et al., 2010). However, an imbalance between MMPs and tissue inhibitors of metalloproteinases (TIMPs) may lead to extracellular matrix breakdown or to collagen deposition. Imbalances between MMPs and TIMPs are present in Parkinson’s, Alzheimer’s and Huntington’s diseases and in cerebral ischemia (Lorenzl et al., 2003a, 2003b). In ALS patients, MMP/TIMP imbalances were noted in sera and cerebrospinal fluid (Niebroj-Dobosz et al., 2010). Since MMPs, enzymes essential for basement membrane integrity (Gasche et al., 2006), are expressed by endothelial cells, neurons, astrocytes and microglia (Candelario-Jalil et al., 2009; Fukuda et al., 2004; Rosenberg, 2002), it is possible that damaged endothelial cells in ALS defectively regulate the MMP pathway and so induce accumulation of basement membrane collagen. It is our belief that the perivascular collagen accumulation we noted occurred over a long period of time and that this accumulation may alter BBB/BSCB transport mechanisms. As a result, motor neurons might suffer from reduced nutrition and increased metabolite levels. Importantly, the perivascular collagen fiber calcifications determined in some capillaries

123

support our contention of long term microvascular damage in ALS. Another finding was pericyte degeneration, observed in most capillaries from ALS medulla and spinal cord segments. Complete pericyte degeneration or fragmentation under the basement membrane was noted. Pericytes, as contractile cells, maintain BBB/BSCB integrity by regulating vessel permeability, blood flow, and vascular tone (Hirschi and D’Amore, 1996; Kutcher and Herman, 2009). The pericyte damage determined via electron microscopy might severely compromise ALS barrier function. It would also explain the 30–45% decreased blood flow through the spinal cord observed in pre-symptomatic G93A SOD1 mice (Zhong et al., 2008) and reduced capillary blood flow in brains of ALS patients correlating with disease severity (Rule et al., 2010), possibly leading to vascular hypoperfusion. Furthermore, hypoxia in ALS patients should be a concern. Blood flow alterations in ALS might reflect altered CNS vascularization. One hypothesis is that ALS is caused by chronic cerebrospinal vascular insufficiency (Arhart, 2010), a theory supported by the chronic hypoxia in the nervous tissue of ALS patients (Iłzecka, 2004). In the lumbar spinal cord of SOD1 mutants, 10–15% reductions in total capillary length were demonstrated prior to inflammatory changes (Zhong et al., 2008). Miyazaki et al. (2011) noted analogous results in spinal cords from a small cohort of ALS patients. Our results, however, demonstrated significantly increased microvascular density in gray matter of the lumbar spinal cord from ALS patients vs. controls. Although capillary density increased slightly in gray matter of the cervical spinal cord, MVD in the lumbar spinal cord segment was higher. Similar to our results, Biron et al. (2011) noted increased brain MVD in Alzheimer’s patients, suggesting a relationship between hypervascularity, neoangiogenesis, and BBB disruption. ALS neovascularization may occur in areas of motor neuron degeneration to compensate for dysfunctional capillaries. This possibility is supported by the 2–3 times higher microvascular collagen expansions measured in numerous gray matter capillaries of ALS lumbar and cervical spinal cord. However, new vessel formation in ALS still needs confirmation. Finally, lost ‘‘tightness’’ between endothelial cells is a hallmark of BBB/BSCB breakdown in ALS. Diminished levels of ZO-1, occludin, and claudin-5 tight junction proteins were detected even before disease onset in ALS mouse models (Zhong et al., 2008). Miyazaki et al. (2011) also observed progressive downregulation of occludin in spinal cord tissues from G93A SOD1 mice preceding motor neuron death. Yet another study, using the G93A SOD1 rat model of ALS, showed reduced mRNA expression of ZO-1 and occludin only at symptomatic stage (Nicaise et al., 2009a). In sporadic and familial ALS patients, lessened mRNA expressions of occludin and ZO-1 were observed in lumbar spinal cord tissue (Henkel et al., 2009). Similarly, analyses of postmortem spinal cord tissue from ALS patients evidenced markedly decreased perivascular occludin (Nicaise et al., 2009b). In agreement with these findings, we showed that primarily ZO-1 expression was reduced in both gray and white matter in all examined ALS tissues. For occludin and claudin-5, diminished expression was found in ALS medulla and cervical

124

brain research 1469 (2012) 114–128

spinal cord. Interestingly, VE-cadherin, a major adhesion junction molecule controlling vascular permeability and leukocyte extravasation (Vestweber, 2008), was reduced only in white matter of the ALS medulla and cervical spinal cord. These results confirm loss of endothelial integrity and indicate BBB/BSCB disruption that might contribute to disease pathogenesis. Inconsistency in diminishing tight and

adhesion junction protein expressions in gray and white matter of ALS medulla, cervical and lumbar spinal cord might be due to involvement of different neurovascular unit components in disease pathogenesis. However, further investigations are necessary. Although a number of similarities in BBB/BSCB damage such as endothelial cell degeneration, capillary leakage,

brain research 1469 (2012) 114–128

perivascular edema, and tight junction protein downregulations are indicated in both mutant SOD1 rodent models of ALS and sporadic ALS patients, some other pathogenic features likely linked to barrier alterations have so far only been identified in patients with sporadic ALS. Pericyte degeneration, perivascular collagen IV expansion, and white matter capillary abnormalities in sporadic ALS patients are significant pathogenic barrier elements which have not been noted in ALS SOD1 animal models. While endothelial cell degeneration might result from oxidative stress induced by the misfolding mutant SOD1 protein in an animal model of ALS, cause(s) of barrier damage in sporadic human ALS are still unknown. In summary, our study results showed severe BBB/BSCB impairment in sporadic ALS patients. Complexity and pervasiveness of barrier alterations are novel aspects which should be strongly considered in ALS pathogenesis. Is barrier damage an initial disease factor or is this impairment just element in ALS pathogenesis? What is the impact of barrier deterioration upon ALS? The answers to these questions are still unclear and less is known about disease initiation in sporadic ALS. Although a recent report showed that numbers of circulating endothelial cells were reduced in the peripheral blood of sporadic ALS patients with moderate or severe disease (Garbuzova-Davis et al., 2010), suggesting altered endothelization during disease, additional markers associated with BBB/BSCB damage during the course of ALS need to be determined. The cause(s) of ALS barrier damage also needs to be identified.

4.

Experimental procedure

4.1.

ALS patients and controls

Frozen postmortem tissues of the brainstem (medulla) and spinal cord (cervical and lumbar segments) from 12 male and 13 female ALS patients and 13 male and 5 female controls obtained from human tissue banks (Human Brain and Spinal Fluid Resource Center, Los Angeles, CA; NICHD Brain and Tissue Bank for Developmental Disorders at the University of Maryland, Baltimore, MD) were used in the study. Ages of ALS subjects and controls ranged from 45 to 83 years (64.871.90 years) and from 39 to 86 years (59.373.78 years), respectively. Average postmortem intervals prior to obtaining tissues were

125

12.8571.31 h (ALS subjects) and 14.6771.24 h (controls). According to medical histories provided by the tissue banks, sporadic ALS was the primary clinical diagnosis for ALS subjects. Two individuals had been diagnosed with probable ALS. Some ALS patients had additional clinical diagnoses: old cerebral infarction (stroke), diabetes type I, moderate hypoxic cerebrum changes, prostate cancer, hypertension, or atherosclerosis. Control individuals perished due to car accident, cancer (no metastasis), or heart failure and had essentially normal brains and spinal cords as confirmed by neuropathological analyses provided by the banks.

4.2.

Tissue preparation

Upon arrival, human tissues were immediately prepared for testing. For electron microscopy (EM) analysis, part of each sample was fixed in 4% paraformaldehyde (PFA) in 0.1 M phosphate buffer (PB), pH 7.2, for 16–24 h at 4 1C. Next day, tissues were cut into 1 mm slices, mapped against a diagram of the whole slice, and regions of interest were removed from slices and fixed overnight in 2.5% glutaraldehyde in 0.1 M PB (Electron Microscopy Sciences, Inc., Hatfield, PA) at 4 1C. The following day, tissues were transferred to a fresh change of buffer and stored for further EM processing. For immunohistochemical analysis, another part of each sample was fixed in 4% PFA in 0.1 M PB, pH 7.2, for 24 h and then cryoprotected in 20% sucrose in 0.1 M PB (pH 7.2) overnight. Coronal medulla and spinal cord sections were cut at 20 mm in a cryostat, mounted on slides and stored at 20 1C. Another part was dissected for the gray and white matter and stored at 80 1C for later Western blot analysis of tight junction proteins.

4.3.

Electron microscopy

BBB/BSCB structural characteristics were identified in the brainstem (medulla) and spinal cords (cervical and lumbar) of ALS patients and compared to age-matched controls by electron microscopy. Structural integrity of microvessels, as described previously (Garbuzova-Davis et al., 2007a), was analyzed in gray matter areas close to motor neuron pools. Microvascular analysis in the white matter was performed in the pyramidal tract (medulla) and lateral funiculi (spinal cord). Briefly, tissue samples were post-fixed in 1% osmium tetroxide (Electron Microscopy Sciences, Inc., Hatfield, PA)

Fig. 8 – Tight junction and adhesion protein expressions in the medulla, cervical and lumbar spinal cord from ALS patients. Tight junction (ZO-1, occludin, claudin-5, and JAM-1) and adhesion (VE-cadherin) proteins were analyzed in separated gray and white matter of the medulla and cervical/lumbar spinal cords from ALS patients and controls using Western blot. The ratio between the band density for each protein and GAPDH was determined. Major downregulation of ZO-1 (A) was determined in ALS gray matter of the medulla (po0.01), cervical (po0.05), and lumbar (po0.001). In the white matter capillaries, significant (po0.05) decrease of ZO-1 was found in the cervical and lumbar spinal cords of ALS patients. (B) For occludin, protein expression was significantly (po0.05) reduced only in the medulla and cervical spinal cord in both gray and white matter. (C) Downregulation of claudin-5 was demonstrated in gray and white matter medulla (po0.05) and gray matter cervical spinal cord (po0.05) from ALS patients. No significant differences were found in lumbar spinal cord from ALS patients vs. controls for occludin and claudin-5. (D) Significant (po0.01) decrease of JAM-1 was found only in gray matter of the cervical and lumbar spinal cords from ALS patients. (E) VE-cadherin expression was significantly (po0.05) downregulated in white matter of ALS medulla and cervical spinal cord. In (a–e), Western immunoblotting in the gray matter of the medulla, cervical and lumbar spinal cord is represented for ZO-1, occludin, claudin-5, JAM-1, and VE-cadherin, respectively.

126

brain research 1469 (2012) 114–128

in 0.1 M PB for 1 h at room temperature (RT). Following osmication, tissues were dehydrated at RT in a graded series of acetone dilutions (30%, 50%, 70%, and 95% acetone in water), 15 min/change. Three 15-min changes in 100% acetone were made and the tissues transferred to a 50:50 mix of acetone and LX112 epoxy resin embedding mix (Ladd Research Industries, Burlington VT). Tissues were infiltrated with this mix for 1 h under vacuum. Tissues were then transferred to a 100% LX112 embedding mix and infiltrated for 1 h on a rotator. Two more 1 h infiltration steps were performed with fresh changes of embedding mix. Tissues were further infiltrated overnight in fresh embedding medium at 4 1C. The following day, tissues were infiltrated in two additional changes of embedding medium at RT, 4 h per change, then embedded in a fresh change of resin in tissue capsules. Blocks were polymerized at 70 1C in an oven overnight and then trimmed and sectioned with a diamond knife on an LKB Huxley ultramicrotome. Thick sections cut at 0.35 mm were placed on glass slides and stained with 1% toluidine blue stain. Thin sections were cut at 80–90 nm, placed on copper grids, and stained with uranyl acetate and lead citrate. Tissue sections were examined and photographed with a FEI Morgagni transmission electron microscope (FEI, Inc., Hillsboro, OR) and Olympus MegaView III digital camera (ResAlta Research Technologies Corp., Golden, CO) at 60 kV.

4.4.

blocking solution of 10% normal goat serum (NGS) and 0.3% Triton X-100 in PBS for 60 min at RT. Mouse monoclonal CD31 (platelet endothelial cell adhesion molecule 1, PECAM-1, 1:50, GeneTex) or CD105 (endoglin, 1:100, Abcam) primary antibodies with 1.5% NGS, and 0.3% Triton X-100 in PBS was applied on the slides overnight at 4 1C. The slides were then rinsed in PBS and incubated with goat anti-mouse secondary antibody conjugated to FITC (1:500-700, Alexa 488, Molecular Probes) for 60 min and after rinsed in PBS coverslipped with Vectashield containing DAPI (Vector Laboratories). A second set of tissue sections was incubated with goat anti-human IgG conjugated to FITC (1:400, Alpha Diagnostics) for 2 h at RT, rinsed three times in PBS and then coverslipped with Vectashield containing DAPI. An epifluorescence microscope (Olympus BX40) was used for tissue examination and imaging. Another set of tissue sections was used for detection of microvascular collagen IV. After pre-incubation with a blocking solution of 10% NGS and 0.3% Triton X-100 in PBS for 60 min at RT, rabbit polyclonal anti-collagen IV (1:200, Abcam) primary antibody was applied on the slides overnight at 4 1C. Next day, tissues were incubated with biotinylated goat antirabbit secondary antibody (1:500, Vector Labs) for 60 min at RT, developed with DAB peroxidase substrate (Vector Labs), and coverslipped with Permount (Fisher Scientific). Tissue images were analyzed through an Olympus DP70 (Olympus) microscope.

Microvascular collagen expansion measurements 4.7.

Electron microscopy images of gray and white matter from the medulla, cervical and lumbar spinal cords from ALS patients and controls were used to measure basement membrane collagen expansion. On the images, 5 capillaries from each analyzed tissue were randomly selected and collagen basement membrane extension was measured in the 4 quadrants of each evaluated capillary. Extensions from ALS patient were compared to similar measures from control group.

4.5.

Microvascular density

Serial medulla and spinal cord tissue sections were rinsed several times in phosphate-buffered saline (PBS) and stained with 0.1% Sirius Red (Direct Red 80, Sigma-Aldrich) for 1 h at room temperature (RT). Tissue slides were then placed in 0.5% acetic acid solution (Sigma-Aldrich) for 10 min and dehydrated in 100% ethanol for 5 min. Slides were coverslipped with Permount (Fisher-Scientific) and examined under a DP70 (Olympus) microscope. Microvascular density (MVD) was measured in gray matter of the medulla and cervical/lumbar spinal cords using Image-Pro software (Media Cybernetics), and vessels with a diameter between 4 and 12 mm were counted. MVD was determined by the ratio of counted vessels per evaluated area, in mm2.

4.6.

Immunohistochemistry

Immunohistochemical staining for capillary endothelium was performed to determine vascular integrity in the gray and white matter of the medulla and spinal cords. Serial tissue sections were rinsed several times in PBS to remove the freezing medium. The tissues were then incubated in a

Western blotting

Dissected gray and white matter tissue samples from the medulla and cervical/lumbar spinal cords from ALS patients and controls were homogenized in 1  cell lysis buffer (Cell Signaling Technology) with 1% protease inhibitor cocktail (Sigma-Aldrich) and centrifuged at 10,000g during 60 min at 4 1C. The supernatant was then aliquoted in 50 ml samples and stored in a 80 1C. Protein assays (BCA Protein Assay, Pierce) were applied to determine the concentration of total protein available in each sample. Electrophoresis gels (4–15% Mini Protean TGX, Bio-Rad) were loaded with 20 mg of total protein per sample per well, and current applied at 90 V, during 90 min. Proteins were transferred to nitrocellulose membranes (Bio-Rad) using a Criterion blotter (Bio-Rad), at 100 V, during 30 min. The membranes were pre-incubated with 5% non-fat dairy milk during 1 h and then incubated overnight at 4 1C with one of the following primary antibodies overnight: rabbit polyclonal occludin (1:2000, Abcam), rabbit monoclonal claudin-5 (1:750, Abcam), rabbit monoclonal vascular-endothelial cadherin (VE-cadherin, 1:750, Abcam), mouse monoclonal junctional adhesion molecule 1 (JAM-1, 1:750, Abcam) or mouse zonula occludens 1 (ZO-1, 1:750, Abcam). The next day, membranes were washed with Tris Buffered Saline (TBS, Pierce) with 0.05% Tween 20 (Fisher-Scientific) and incubated with the corresponding secondary antibody conjugated with horseradish peroxidase (HRP): goat antirabbit (1:5000, Abcam) or rabbit anti-mouse (1:5000, Abcam) for 2 h at RT. The membranes were thoroughly washed with TBS, incubated with HRP substrate (Immobilion Western, Millipore) and placed into a Bio-Rad CCD camera for chemiluminescence band detection. Images were analyzed through Quantity One software (Bio-Rad) and band densities were

brain research 1469 (2012) 114–128

measured. Each membrane was then re-incubated with rabbit anti-GAPDH (1:10,000, Sigma-Aldrich) at 4 1C overnight for normalizing protein in all membranes. On the following day, the membranes were washed and incubated with goat anti-rabbit secondary antibody (1:8000, Abcam) for 2 h at RT. Band detection and density measures were performed as in the previous description. For each protein evaluated (Occludin, Claudin-5, VE-cadherin, JAM-1 and ZO-1), the ratio between the band densities and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was established.

4.8.

Statistical analysis

Data are presented as means7S.E.M. One-way ANOVA with Student–Newman–Keuls Multiple Comparisons test was used with po0.05 considered significant.

Acknowledgments This study was supported by the Muscular Dystrophy Association (Grant #92452). We gratefully acknowledge the Human Brain and Spinal Fluid Resource Center (Los Angeles, CA) and the NICHD Brain and Tissue Bank for Developmental Disorders at the University of Maryland (Baltimore, MD) for providing human tissues.

r e f e r e n c e s

Arhart, R.W., 2010. A possible haemodynamic mechanism for amyotrophic lateral sclerosis. Med. Hypotheses 75, 341–346. Ballabh, P., Braun, A., Nedergaard, M., 2004. The blood–brain barrier: an overview: structure, regulation, and clinical implications. Neurobiol. Dis. 16, 1–13. Begley, D.J., 2004. ABC transporters and the blood–brain barrier. Curr. Pharm. Des. 10, 1295–1312. Begley, D.J., Brightman, M.W., 2003. Structural and functional aspects of the blood–brain barrier. Prog. Drug Res. 61, 39–78. Beuche, W., Yushchenko, M., Ma¨der, M., Maliszewska, M., Felgenhauer, K., Weber, F., 2000. Matrix metalloproteinase-9 is elevated in serum of patients with amyotrophic lateral sclerosis. Neuroreport 11, 3419–3422. Biron, K.E., Dickstein, D.L., Gopaul, R., Jeffries, W.A., 2011. Amyloid triggers extensive cerebral angiogenesis causing blood brain barrier permeability and hypervascularity in Alzheimer’s disease. PLoS One 6, e23789. Bruijn, L.I., Miller, T.M., Cleveland, D.W., 2004. Unraveling the mechanisms involved in motor neuron degeneration in ALS. Annu. Rev. Neurosci. 27, 723–749. Candelario-Jalil, E., Yang, Y., Rosenberg, G.A., 2009. Diverse roles of matrix metalloproteinases and tissue inhibitors of metalloproteinases in neuroinflammation and cerebral ischemia. Neuroscience 158, 983–994. Claudio, L., 1996. Ultrastructural features of the blood–brain barrier in biopsy tissue from Alzheimer’s disease patients. Acta Neuropathol. 91, 6–14. Cleveland, D.W., Rothstein, J.D., 2001. From Charcot to Lou Gehrig: deciphering selective motor neuron death in ALS. Nat. Rev. Neurosci. 2, 806–819. Dermietzel, R., Krause, D., 1991. Molecular anatomy of the blood–brain barrier as defined by immunocytochemistry. Int. Rev. Cytol. 127, 57–109.

127

Fang, L., Teuchert, M., Huber-Abel, F., Schattauer, D., Hendrich, C., Dorst, J., Zettlmeissel, H., Wlaschek, M., Scharffetter-Kochanek, K., Kapfer, T., Tumani, H., Ludolph, A.C., Brettschneider, J., 2010. MMP-2 and MMP-9 are elevated in spinal cord and skin in a mouse model of ALS. J. Neurol. Sci. 294, 51–56. Fukuda, S., Fini, C.A., Mabuchi, T., Koziol, J.A., Eggleston Jr., L.L., del Zoppo, G.J., 2004. Focal cerebral ischemia induces active proteases that degrade microvascular matrix. Stroke 35, 998–1004. Garbuzova-Davis, S., Haller, E., Saporta, S., Kolomey, I., Nicosia, S.V., Sanberg, P.R., 2007a. Ultrastructure of blood–brain barrier and blood–spinal cord barrier in SOD1 mice modeling ALS. Brain Res. 1157, 126–137. Garbuzova-Davis, S., Rodrigues, M.C.O., Hernandez-Ontiveros, D.G., Louis, M.K., Willing, A.E., Borlongan, C.V., Sanberg, P.R., 2011. Amyotrophic lateral sclerosis: a neurovascular disease. Brain Res. 1398, 113–125. Garbuzova-Davis, S., Saporta, S., Haller, E., Kolomey, I., Bennett, S.P., Potter, H., Sanberg, P.R., 2007b. Evidence of compromised blood–spinal cord barrier in early and late symptomatic SOD1 mice modeling ALS. PLoS One 2, e1205. Garbuzova-Davis, S., Woods, R.L., Louis, M.K., Zesiewicz, T.A., Kuzmin-Nichols, N., Sullivan, K.L., Miller, A.M., HernandezOntiveros, D.G., Sanberg, P.R., 2010. Reduction of circulating endothelial cells in peripheral blood of ALS patients. PLoS One 5, e10614. Gasche, Y., Soccal, P.M, Kanemitsu, M., Copin, J.-C., 2006. Matrix metalloproteinases and diseases of the central nervous system with a special emphasis on ischemic brain. Front. Biosci. 11, 1289–1301. Henkel, J.S., Beers, D.R., Wen, S., Bowser, R., Appel, S.H., 2009. Decreased mRNA expression of tight junction proteins in lumbar spinal cords of patients with ALS. Neurology 72, 1614–1616. Hirschi, K.K., D’Amore, P.A., 1996. Pericytes in the microvasculature. Cardiovasc. Res. 32, 687–698. Iłzecka, J., 2004. Cerebrospinal fluid vascular endothelial growth factor in patients with amyotrophic lateral sclerosis. Clin. Neurol. Neurosurg. 106, 289–293. Kutcher, M.E., Herman, I.M., 2009. The pericyte: cellular regulator of microvascular blood flow. Microvasc. Res. 77, 235–246. Lorenzl, S., Albers, D.S., LeWitt, P.A., Chirichigno, J.W., Hilgenberg, S.L., Cudkowicz, M.E., Beal, M.F., 2003a. Tissue inhibitors of matrix metalloproteinases are elevated in cerebrospinal fluid of neurodegenerative diseases. J. Neurol. Sci. 207, 71–76. Lorenzl, S., De Pasquale, G., Segal, A.Z., Beal, M.F., 2003b. Dysregulation of the levels of matrix metalloproteinases and tissue inhibitors of matrix metalloproteinases in the early phase of cerebral ischemia. Stroke 34, e37–38. Miyazaki, K., Ohta, Y., Nagai, M., Morimoto, N., Kurata, T., Takehisa, Y., Ikeda, Y., Matsuura, T., Abe, K., 2011. Disruption of neurovascular unit prior to motor neuron degeneration in amyotrophic lateral sclerosis. J. Neurosci. Res. 89, 718–728. Nicaise, C., Mitrecic, D., Demetter, P., De Decker, R., Authelet, M., Boom, A., Pochet, R., 2009a. Impaired blood–brain and blood–spinal cord barriers in mutant SOD1-linked ALS rat. Brain Res. 1301, 152–162. Nicaise, C., Soyfoo, M.S., Authelet, M., De Decker, R., Bataveljic, D., Delporte, C., Pochet, R., 2009b. Aquaporin-4 overexpression in rat ALS model. Anat. Rec. (Hoboken) 292, 207–213. Niebroj-Dobosz, I., Janik, P., Sokolowska, B., Kwiecinski, H., 2010. Matrix metalloproteinases and their tissue inhibitors in serum and cerebrospinal fluid of patients with amyotrophic lateral sclerosis. Eur. J. Neurol. 17, 226–231. Ono, S., Imai, T., Munakata, S., Takahashi, K., Kanda, F., Hashimoto, K., Yamano, T., Shimizu, N., Nagao, K., Yamauchi, M., 1998. Collagen abnormalities in the spinal cord from patients with amyotrophic lateral sclerosis. J. Neurol. Sci. 160, 140–147.

128

brain research 1469 (2012) 114–128

Pardridge, W.M., 1999. Blood–brain barrier biology and methodology. J. Neurovirol. 5, 556–569. Pozzi, A., Zent, R., 2009. Regulation of endothelial cell functions by basement membrane- and arachidonic acid-derived products. Rev. Syst. Biol. Med. 1, 254–272. Rosen, D.R., Siddique, T., Patterson, D., Figlewicz, D.A., Sapp, P., Hentati, A., Donaldson, D., Goto, J., O’Regan, J.P., Deng, H.X., 1993. Mutations in Cu/Zn superoxide dismutase gene are associated with familial amyotrophic lateral sclerosis. Nature 362, 59–62. Rosenberg, G.A., 2002. Matrix metalloproteinases in neuroinflammation. Glia 39, 279–291. Rothstein, J.D., 2009. Current hypotheses for the underlying biology of amyotrophic lateral sclerosis. Ann. Neurol. 65 (1), S3–9. Rule, R.R., Schuff, N., Miller, R.G., Weiner, M.W., 2010. Gray matter perfusion correlates with disease severity in ALS. Neurology 74, 821–827. Rutka, J.T., Apodaca, G., Stern, R., Rosenblum, M., 1988. The extracellular matrix of the central and peripheral nervous systems: structure and function. J. Neurosurg. 69, 155–170.

Soon, C.P.W., Crouch, P.J., Turner, B.J., McLean, C.A., Laughton, K.M., Atkin, J.D., Masters, C.L., White, A.R., Li, Q.-X., 2010. Serum matrix metalloproteinase-9 activity is dysregulated with disease progression in the mutant SOD1 transgenic mice. Neuromuscul. Disord. 20, 260–266. Strong, M.J., Kesavapany, S., Pant, H.C., 2005. The pathobiology of amyotrophic lateral sclerosis: a proteinopathy? J. Neuropathol. Exp. Neurol. 64, 649–664. Vestweber, D., 2008. VE-cadherin: the major endothelial adhesion molecule controlling cellular junctions and blood vessel formation. Arterioscler. Thromb. Vasc. Biol. 28, 223–232. Vorbrodt, A.W., Dobrogowska, D.H., 2003. Molecular anatomy of intercellular junctions in brain endothelial and epithelial barriers: electron microscopist’s view. Brain Res. Brain Res. Rev. 42, 221–242. Zhong, Z., Deane, R., Ali, Z., Parisi, M., Shapovalov, Y., O’Banion, M.K., Stojanovic, K., Sagare, A., Boillee, S., Cleveland, D.W., Zlokovic, B.V., 2008. ALS-causing SOD1 mutants generate vascular changes prior to motor neuron degeneration. Nat. Neurosci. 11, 420–422.