Cytokines were measured using the ELISA method and lymphocytes usingflow cytometry. Statistical analysis followed laboratory results. Results:Results showed a statistically significant IL-1�, IL-2, IL-8, IL-10, TNF-�cytokines up-regulation in patients’ blood samples compared to that ofnon-demented age-matched subjects. No statistically significant differ-ences were found relatively as it has been to the IL-6 cytokine levels andthe CD4, CD8, CD4/CD8 lymphocytes’ subpopulation levels betweenpatients and controls. Statistically significant differences were not identi-fied between patients of mild moderate stage versus patients of severestage. Logistic regression revealed three variables that could predict asubject’s probability to develop AD: pre-inflammatory cytokines IL-1�

and TNF-� and anti-inflammatory cytokine IL-10. Conclusion: Data sug-gested that the increase of several cytokines in serum might be a reflectionof brain’s pathological events and manifests the presence of detectableimmuno-biological markers in AD patients. The cytokine profile may havepredictive value in AD.

P3-275 EXPRESSION AND REGULATION OFINFLAMMATORY GENES IN HUMAN BRAINENDOTHELIAL CELLS INDUCED BY BETA-AMYLOID PEPTIDES

Debbie Callaghan1, Aimee Jones1, Vanja Vukic2, Douglas Walker3,Lih-Fen Lue3, John Woulfe2, T. G. Beach3, L. Sue3,Danica Stanimirovic1,2, Wandong Zhang1,2, 1Institute for BiologicalSciences, National Research Council of Canada, Ottawa, ON, Canada;2University of Ottawa, Ottawa, ON, Canada; 3Sun Health ResearchInstitute, Sun City, AZ, USA. Contact e-mail: wandong.zhang@nrc-cnrc.gc.ca

Neuroinflammation is an important mechanism of Alzheimer’s disease(AD)-associated neurodegeneration. Accumulation and deposition of A�

around neurovasculature is found in the majority of AD patients. However,little is known whether A� deposition would result in chronic vascularinflammation and insufficiency, which may contribute to neurodegenera-tion. We showed previously that A�1-40 induces strong expression ofinflammatory genes in primary cultured human brain endothelial cells(HBEC) by semi-quantitative RT-PCR analyses, including IL-1�, MCP-1,TNF-�, IL-8 and ICAM-1. In the present study, we have profiled cytokineexpression at the protein level and shown that A�-treated HBEC secreteimmunoreactive cytokines into media, including MCP-1, MCP-2, GRO,IL6, IL-8, etc. In order to determine whether these cytokines are expressedin AD brain, we have carried out real-time Q-PCR analyses using AD andage-matched nondemented control (ND) brain samples. Increased expres-sion of inflammatory genes, IL-1�, MCP-1, IL-6, and GRO were observedin AD compared to ND brain samples. These results validate our findingsin cultured HBEC. We next investigated transcriptional regulation of in-flammatory genes in A�-treated HBEC. HBEC were treated with A�1-40 orscrambled A�40-1 for 8hr. Nuclear extracts were prepared and TranSignalProtein/DNA Array blots carrying 56 different transcription factors (TF)were used for the analyses (Panomics). Each TF was spotted four times onthe blots and the experiments were repeated. In comparison with controls,five TFs were increased over two fold (AP-1, CREB, GATA, NFATc,GRE) and three TFs were down-regulated over two fold (p53, TR, MEF-1)in A�1-40-treated HBEC. Increased levels of AP-1, CREB and GATA wereconfirmed in AD brain samples by the same array analyses. Sequenceanalyses show that the promoter regions of many inflammatory genes carryAP-1 or/and CREB binding sites. GATA and NFATc are believed tocomplex or associate with AP-1 or/and CREB to regulate cytokine expres-sion in cells. Our studies demonstrate that A� can induce inflammatorygene expression in HBEC and some of these genes are also up-regulated inAD brain and suggest that several TFs (AP-1, CREB, GATA or NFATc)may be involved in transcriptional regulation of the inflammatory genes inA�-treated HBEC and/or in AD brain.

P3-276 MAPKAP KINASE 2 DEFICIENCY IN MICROGLIAINHIBITS PRO-INFLAMMATORY MEDIATORRELEASE AND RESULTANT NEUROTOXICITY:RELEVANCE TO NEUROINFLAMMATION IN ATRANSGENIC MOUSE MODEL OF ALZHEIMER’SDISEASE

Ainsley A. Culbert1, Stephen D. Skaper1, Zoe M. Seymour1,David R. Howlett1, Peter E. Soden1, Florence Guillot1, Laura Facci1,Matthias Gaestel2, Jill C. Richardson1, 1GlaxoSmithKline Research andDevelopment Limited, Harlow, United Kingdom; 2Medical SchoolHannover, Hannover, Germany. Contact e-mail:ainsley.a.culbert@gsk.com

Background: A role for p38 mitogen-activated protein (MAP) kinase inthe pathology of Alzheimer’s disease (AD) has previously been suggested.MAP kinase-activated protein kinase 2 (MAPKAP kinase 2 or MK2) is oneof several kinases directly regulated by p38 MAP kinase. Objective: Wesought to investigate whether MK2 itself plays a role in neuroinflammatoryand neurodegenerative pathology of relevance to AD. Methods: We in-vestigated the role of MK2 in isolated microglial cell cultures, in co-cultures of cortical neurons and microglia, and in the CNS tissue of atransgenic mouse model of AD. Results: MK2 expression was enriched inprimary cultures of cortical microglial cells compared to other CNS celltypes, favouring a function for MK2 in the former. Furthermore, MK2activation and expression were increased in lipopolysaccharide (LPS) �interferon-� (IFN�)-stimulated microglial cells, implicating a role for MK2in eliciting a pro-inflammatory response. Microglia cultured ex vivo fromMK2 deficient (MK2-/-) mice demonstrated significant inhibition in releaseof tumour necrosis factor alpha (TNF-�), KC and macrophage inflamma-tory protein 1 alpha (MIP-1�) on stimulation with LPS�IFN� or amy-loid-� peptide (1-42) compared to MK2�/� microglia. Consistent with aninhibition in pro-inflammatory mediator release, cortical neurons co-cul-tured with LPS�IFN�-stimulated MK2-/- microglia were protected frommicroglial-mediated neuronal cell toxicity. In a transgenic mouse model ofAD in which amyloid precursor protein (APP) and presenilin-1 (PS-1)harbouring familial AD mutations are over-expressed in specific regions ofthe brain, elevated activation and expression of MK2 correlated with�-amyloid deposition, microglial activation and upregulation of TNF-�,MIP-1� and KC gene expression in the same brain regions. Conclusion:Our data propose a role for MK2 in AD brain pathology, for whichneuroinflammation involving cytokines and chemokines and overt neuro-nal loss have been documented.

P3-277 SEVERE NEURODEGENERATION ISINTIMATELY RELATED TO INFLAMMATION INCDK5/P25 INDUCIBLE MICE

David Muyllaert, Dick Terwel, Peter Borghgraef, Fred Van Leuven,Exp. Genetics Group, K.U.Leuven, Leuven, Belgium. Contact e-mail:david.muyllaert@med.kuleuven.be

Background: The molecular relations of amyloid and tau pathology inAlzheimer’s disease (AD) and the precise mechanisms of neurodegenera-tion still remain to be unravelled. Objectives: The third pathologicalhallmark in AD, i.e. persistent inflammation, is proposed to play an essen-tial role in the disease progression by fuelling a vicious cycle that even-tually leads to the severe, diagnostic brain atrophy in AD. Methods: Wegenerated and analysed p25 transgenic mice that express the N-truncatedform of the p35 activator of cdk5 in neurons of hippocampus and forebrainunder tetracycline control (CaMKII-tTA; TET-off system). Results: With-out tetracycline (p25-OFF) lethality is very high in late embryogenesis andpost-natal life (�60 %). Doxycycline in the drinking water of pregnantdams and litters until age 6 weeks alleviated the lethality completely andthis regime was imposed to study repercussions in adult brain. This turnedout to be more severe than expected since after only two weeks withoutDOX (at age 8 weeks) neuronal expression of p25 in hippocampus andcortex was already accompanied by active microgliosis and astrogliosis.

S456 Poster P3:: Tuesday Posters