Brain Oral Session: Experimental Cerebral Ischemia—Neuroinflammation

947. A cell type specific role for MyD88 following ischemic insult

P.J. Crack, C.E. Downes, C.H.C. Wong and P. Guio

Pharmacology, The University of Melbourne, Melbourne, VIC, Australia

Objectives: Myeloid differentiation factor 88 (MyD88) is an adaptor protein that is integral to many immune linked pathways including certain Toll Like Receptors (TLR). This study is designed to understand the role of MyD88 in coordinating inflammatory pathways in the brain that are activated after ischemia/reperfusion injury. Data generated through this study suggests a complex role for MyD88 during ischemic insult, involving both pro-survival and pro-death signalling.

Methods: Primary fibroblast, neuronal and glial cultures were isolated from wild type (WT) and MyD88 knockout mice (MyD88−/−). A novel membrane-based method of culture was used to study neurones and glia separately, whilst still being able to communicate with one another. WT and MyD88−/− neurons were cultured in the presence of both WT and MyD88−/− glia, these combinations of cells were then exposed to 4 h of oxygen glucose deprivation (OGD). Cellular survival was assessed with a mitochondrial viability assay. Activation by phosphorylation of the TLR system downstream signalling components ERK, JNK, TPL2 and NFκB was measured by Western blot.

Results: Cellular viability following OGD shows that fibroblasts and neurons lacking MyD88 are more susceptible to ischemic injury, suggesting a pro-survival role for MyD88 in these cell types. This correlates with in vivo data showing MyD88−/− mice exposed to mid-cerebral artery occlusion exhibit a two fold increase in infarct volume when compare to WT mice (n = 12). The presence of glia increases the survival of neurons after OGD by 30%, and MyD88−/− glia potentate this increase in survival of neurons by a further 15%, suggesting a pro-death signalling role for MyD88 in glia. MyD88 −/− neurons show delayed phosphorylation of mediators (ERK, JNK, TPL2, and NFκB) known to play a role in the determination of cellular fate as evidenced by Western blots.

Conclusions: Together these data suggest that there is a dual role for MyD88 dependant signalling. These results highlight the role of MyD88 as cell type specific and that responses controlled by MyD88 are more complex than originally thought.

878. Characterization of the immune response in peripheral organs after cerebral ischemia in mice

A. Arac, T.M. Bliss and G.K. Steinberg

Neurosurgery, Stanford University School of Medicine, Stanford, California, USA

Objectives: Inflammation plays an important role in central nervous system ischemia. Clinical and experimental data show acute and prolonged inflammatory response in the brain after stroke. In addition, there are changes in immune cells in peripheral organs following stroke. However, such immune cell changes in blood and secondary lymphoid organs such as lymph nodes and spleen after stroke have not been extensively studied. Here we investigate the response of the immune system in these organs after different stroke severities.

Methods: 12-week-old, male C57BL6J mice were subjected to 30 min (mild) or 90 min (severe) middle cerebral artery occlusion (MCAo) or sham surgery. 96 h later, cells from cervical, axillary and abdominal (extra-peritoneal) lymph nodes, splenocytes and white blood cells from blood were isolated and analyzed by flow cytometry. The brains were processed and infarct sizes were determined from cresyl violet stained sections. Spleen size was also measured.

Results: Spleen size decreased after stroke. Consistent with this, the total number of splenocytes decreased after stroke compared to sham animals (42% and 73%, P<0.05 in 30 mins and 90 mins MCAo groups, respectively, consistent with previous reports. ¹ This decrease in splenocyte number correlated with lesion size, with greater loss of splenocytes measured with increasing lesion size (r = −0.918, P<0.001). Flow cytometric analysis of immune cells in the spleen confirmed a significant decrease after stroke in the total number of all immune cell types tested, except for the following which showed no significant change: NK1.1+ natural killer cells, CD4+ Th cells and CD8+ Tcyt cells in the 30 mins MCAo group and CD4+CD62 L- effector T cells in both stroke groups. The number of all the cell types, except NK1.1+ natural killer cells, also correlated negatively with the infarct size. Although the total number of the different immune cells changed after stroke, the percentage of the different immune cells in the spleen was not different between sham and stroke animals except for: 1) regulatory T cells (Tregs, CD4+CD25+Foxp3+), that increased 42% (P<0.05) after 90 mins MCAo, 2) B cells (B220+), which decreased 17.8% (P<0.05) after 30 mins MCAo and 14% (P<0.05) after 90 mins MCAo. In lymph nodes only Tregs changed and only after 90 mins MCAO, increasing 30% (P<0.05). This increase correlated with infarct size (r = 0.867, P<0.01. In the blood, changes were only seen in granulocytes (Gr1+) and monocytes/macrophages (CD11b+), which significantly increased (P<0.05) in both stroke groups (granulocytes: 213% in 30 mins and 276% in 90 mins MCAo groups; monocytes/macrophages: 185% in 30 mins and 210% in 90 mins MCAo groups).

Conclusions: There is a systemic response of the immune system to stroke. These data are the first to show that some components of this response correlate with infarct size. Modulation of the immune system may provide a new strategy for stroke treatment.

849. Toll-like receptor 2/1 signaling in cerebral ischemia requires the scavenger receptor CD36

T. Abe, M. Shimamura, J. Anrather, P. Zhou, G. Racchumi and C. Iadecola

Neurobiology, Weill Cornell Medical Colledge, New York, New York, USA

Objectives: The scavenger receptor CD36 plays an important role in focal cerebral ischemic injury by triggering NF-κB activation and inducing post-ischemic inflammation. ¹ CD36 can be part of a receptor complex that includes toll-like receptors (TLRs), transmembrane receptors involved in innate immunity. TLR2 form heterodimers with TLR1 or 6, and CD36 may act as a co-receptor for TLR2/6 in macrophages. ² Furthermore, TLR2, like CD36, contributes to cerebral ischemic damage, raising the possibility of a co-receptor interaction between CD36 and TLR2 also in brain injury. Therefore, we investigated whether CD36 is required for TLR2-induced inflammatory signaling and for its deleterious effects on the ischemic brain.

Methods: The TLR2/6 activator FSL-1, the TLR2/1 activator Pam3 and the TLR4 activator LPS were administrated into the cerebral ventricles (ICV) in wild type, TLR2−/− and CD36−/− mice. Inflammatory genes expression in brain was examined by quantitative real-time PCR. The middle cerebral artery was transiently occluded in CD36−/−, or TLR2 −/− mice and, 10 mins after reperfusion, vehicle, FSL-1, Pam3, or LPS was injected ICV. Infarct volume was assessed 3 days later in Nissl stained sections and data corrected for swelling. The microglia-macrophage marker F4/80 and the neutrophil marker MPO were examined immunohistochemically 3 days after ischemia.

Results: ICV injection of the FSL-1 in CD36+/+ mice upregulated inflammatory gene expression (MCP-1 Figure 1A; ICAM-1 19±5 fold induction; ELAM-1 64±17; IL-6 22±4; P<0.05; n = 4/group). FSL-1 induced similar upregulation in CD36−/− mice, as did LPS (P>0.05 from CD36+/+; n = 4/group). In contrast, the inflammatory response induced by Pam3 was markedly attenuated in CD36−/− and TLR2−/− mice. Therefore, TLR2/1 inflammatory signaling in brain requires CD36. ICV injection of FSL-1 or LPS after reperfusion increased stroke volume in CD36−/− mice (Figure 1B). However, Pam3 did not increase stroke volume in either CD36−/− or in TLR2−/− mice (n = 4-8/group; P>0.05). Furthermore, microglial activation and neutrophil infiltration were markedly attenuated in CD36−/− mice compared to WT. Post-ischemic administration of FSL-1, but not Pam3, re-established the cellular inflammatory response in CD36−/− mice.

Figure 1

Conclusions: CD36 is required for TLR2/1-induced neuroinflammation and for the deleterious effects of TLR2/1 ligands in the post-ischemic brain. The findings suggest that in focal cerebral ischemia the endogenous ligands activating TLR2/1 require CD36 for triggering post-ischemic inflammation. Thus, CD36 is a critical sensor of ‘danger signals’ in the post-ischemic brain and, as such, is a potential target for therapeutic interventions.

1048. Genetically-defined mannose-binding lectin deficiency is associated with better stroke outcome in mice and humans

A. Cervera¹, C. Justicia², X. Urra¹, S. Amaro¹, M. Gomez-Choco¹, B. Suarez³, V. Obach¹, I. Perez-De-Puig², J. Jensenius⁴, A.M. Planas², F. Lozano³ and A. Chamorro¹

¹Stroke Unit, Hospital Clinic; ²Department of Brain Ischemia and Neurodegeneration, IIBB-CSIC, IDIBAPS; ³Immunology Department, Hospital Clinic, Barcelona, Spain; ⁴Department of Medical Microbiology and Immunology, University of Aarhus, Aarhus, Denmark

Objectives: Stroke activates humoral and cellular innate immune responses whose contribution to brain injury and/or repair remains incompletely understood. Various innate immune responses, including activation of the complement system,^1,2 are increasingly recognized as players in stroke-induced brain damage. Several lines of evidence support that complement activation in stroke occurred through the non-classical pathways.^3,4 In this translational study, the association of acute stroke outcome with genetically-defined deficiencies of mannose-binding lectin (MBL), a key component of the lectin pathway of complement activation, ⁵ was investigated in humans and mice.

Methods: Functionally relevant MBL and MASP2 polymorphisms were studied in 135 stroke patients admitted into a clinical trial of antibiotic prophylaxis in non-septic stroke (ESPIAS). ⁶ Major clinical endpoints were compared with systemic levels of MBL and MASP-2, and with genetic MBL and MASP2 deficiencies. The time-course of white-cell counts, C-reactive protein (CRP), cytokines, C3 and C4 complement proteins, was studied from baseline to day 90. MBL and MASP-2 circulating protein levels were evaluated with ELISA assays. In experimental studies, brain ischemia was induced in mice by intraluminal occlusion of the middle cerebral artery for 2-hours. Adult male MBL-null mice (KO) and corresponding wild type (WT) mice of the same background (C57/Bl6) were used in this study. Infarct volume and the neurological deficit were evaluated at 48 h. In a subset of animals the expression of myeloperoxidase in brain tissue was evaluated to assess neutrophil recruitment.

Results: Twenty-four (17.8%) patients had MBL2 genotypes associated with MBL-low levels and presented lower CRP, C3 and C4 concentrations than patients with MBL-sufficient genotypes. Genetic deficiencies correlated very well with low levels of circulating proteins. In logistic regression analyses adjusted for age, gender, stroke severity and clinical subtype, MBL-low genotypes and low levels of circulating protein were associated with good functional outcome. In mice, transient cerebral ischemia was induced in WT and KO mice. Ischemia reduced cerebral blood flow to a similar extent in both types of mice. However, infarct volume was significantly smaller (50%) in MBL-deficient mice (n = 10) compared to WT (n = 11), and the neurological deficit was minor in KO mice. Also, MBL-deficiency attenuated neutrophil recruitment to the ischemic tissue.

Conclusions: These results underscore the key role of the MBL pathway in post-stroke injury in both human and mouse, and have not only predictive but also therapeutic value, thus supporting interventions aimed to specifically inhibit this ancient pathway of complement activation.

Acknowledgement: Supported by the Spanish Ministry of Education and Science (CICYT) and by the European Community's Seventh Framework Programme (FP7/2007-201, grant no 201024, European Stroke Network).

899. Vascular adhesion protein-1 (VAP-1) inhibition provides neuroprotection in rats following reversible middle cerebral artery occlusion and 14 days reperfusion

L.Z. Mao, H.L. Xu, F. Vetri and D.A. Pelligrino

Neuroanesthesia Research Laboratory, University of Illinois at Chicago, Chicago, Illinois, USA

Introduction: VAP-1, also called semicarbazide-sensitive amine oxidase (SSAO), is reported to play an important role in adhesion and, especially, endothelial transmigration of multiple leukocyte subsets (i.e., neutrophils, monocytes, lymphocytes). Our laboratory previously reported that treatment with a novel and selective VAP-1/SSAO blocker, LJP-1207, at 6 h reperfusion following transient forebrain ischemia (TFI), prevented neutrophil infiltration into the brain, and provided significant neuroprotection. ¹ A similar level of neuroprotection was seen in TFI rats rendered neutropenic (anti-PMNL antibody treatment). In preliminary experiments, ² we found little or no neuroprotection associated with neutropenia in rats exposed to 1 h right middle cerebral artery occlusion (MCAo) and 72 h reperfusion; yet LJP-1207 treatment was neuroprotective, even when introduced 6 to 12 h post-MCAo. This suggested that the long therapeutic window associated with pharmacologic blockade of VAP-1/SSAO arises from actions toward non-PMNL leukocytes, at least over short (several days) post-ischemic recovery periods. In the present study, we evaluated whether neuroprotection could be extended over longer (2 weeks) post-MCAo recovery.

Methods: We compared control rats and rats treated with the highly-selective, non-hydrazine VAP-1/SSAO blocker, LJP-1586. Neurobehavioral function and histopathology were evaluated in male Sprague-Dawley rats, subjected to 1 h of MCAo (suture model) followed by 14d of reperfusion. Administration of LJP-1586 (10 mg/kg, iv) was initiated at 6 h of reperfusion, followed by repeated doses every 24 h, until sacrifice. Neurobehavioral tests included beam walking speed, which assesses deficits in coordination and integration of motor movement, especially in the hindlimb; and adhesive-removal, which evaluates forelimb somatosensory asymmetries. For histopathology, 30 μm frozen coronal sections were stained with hematoxylin and eosin (H&E), and the infarct area was calculated (with corrections for ischemic versus non-ischemic hemisphere areas).

Results: Treatment with the VAP-1/SSAO inhibitor was associated with smaller infarcts, at 14 days, compared to controls. Representative brain sections (bregma ∼−1.0 mm) are shown in the figure (infarct areas were 19.9 and 8.2 mm² in controls and LJP-treated rats, respectively). Both neurobehavioral function tests indicated a better outcome, at 14 days, in the VAP-1/SSAO-inhibited animals (see Figure).

Figure

Conclusions: In conclusion, VAP-1/SSAO inhibitors are promising neuroprotective compounds with a wide therapeutic window and substantial translational potential.

1056. Brain tryptophan metabolism using small animal pet in a neonatal rabbit model of maternal inflammation induced cerebral palsy

S. Kannan¹, F. Saadani-Makki¹, H. Dai¹, P. Chakraborthy², O. Muzik^1,2, R. Romero³ and D. Chugani¹

¹Pediatrics; ²Radiology, Wayne State University, Detroit, MI; ³Perinatology Research Branch, NICHD, NIH, DHHS, Bethesda, Maryland, USA

Objective: Maternal intrauterine infection has been implicated in the activation of microglia and astrocytes in the neonatal brain, that may lead to neuronal and oligodendrocyte death by production of excitotoxic metabolites, resulting in periventricular leukomalacia (PVL) and cerebral palsy. Indoleamine 2,3-dioxygenase, the rate-limiting enzyme in tryptophan metabolism by the kynurenine pathway, is known to be upregulated in activated microglia and astrocytes. Increased tryptophan metabolism by the kynurenine pathway results in the production of excitotoxic metabolites such as quinolinic acid causing seizures and white matter injury. We hypothesized that there is increased tryptophan metabolism in the periventricular regions in our previously established rabbit model of maternal inflammation induced cerebral palsy, as assessed in vivo by the uptake of [¹¹C]methyl-L- tryptophan (AMT) using microPET.

Material and methods: Pregnant New Zealand White rabbits were injected with saline or 20 μg/Kg of E. Coli lipopolysaccharide (LPS, endotoxin) along the length of the uterus on day 28/31 of gestation. The kits exposed to endotoxin in utero were born with a phenotype of cerebral palsy. The endotoxin (n = 6) and saline exposed kits (n = 5) were compared with term newborn kits born to rabbits that had no surgical intervention (n = 4). Term newborn kits were injected with ¹¹C-AMT intravenously and scanned for 60 min using microPET, followed by MRI scan (4.7T) for anatomical co-registration. 3D regions of interest were drawn for the periventricular region and the whole brain and ratio of tracer uptake at 20 to 40 mins of the scan was determined. After scanning, the kits were euthanized and brains fixed, sectioned and stained for microglial cells (tomato lectin) and astrocytes (glial fibrillary acidic protein).

Results: There was increased uptake of ¹¹C-AMT in the periventricular regions as determined by a significant increase in the ratio of tracer uptake in the periventricular region compared to the whole brain, in the endotoxin kits (mean ratio±SD was 1.46±0.23) when compared to the saline controls (1.11±0.04) (P<0.001) and the no-intervention control kits (1.02±0.10) (P<0.001). Activated microglia and astrocytes as indicated by an increase in density and change in morphology of these cells were noted in periventricular regions and hippocampus of the brain kits exposed to intrauterine inflammation.

Conclusion: Increased tryptophan metabolism is noted in the periventricular regions of newborn kits exposed to maternal inflammation in utero when compared to age matched controls. This increase in tryptophan that is confined to the periventricular regions may be due to the increased presence of activated microglia and astrocytes in these regions. Tryptophan metabolism by the kynurenine pathway may be responsible for production of excitotoxic injury leading to seizures and motor deficits in neonates born to mothers with intrauterine infections.

[¹¹C] AMT uptake in the neonatal rabbit brain.