Accepted Manuscript The CCAAT/enhancer binding protein delta/miR135a/thrombospondin 1 axis mediates PGE2-induced angiogenesis in Alzheimer’s disease Chiung-Yuan Ko , Yu-Yi Chu , Shuh Narumiya , Jhih-Ying Chi , Tomoyuki Furuyashiki , Tomohiro Aoki , Shao-Ming Wang , Wen-Chang Chang , Ju-Ming Wang PII:

S0197-4580(14)00791-X

DOI:

10.1016/j.neurobiolaging.2014.11.020

Reference:

NBA 9138

To appear in:

Neurobiology of Aging

Received Date: 27 March 2014 Revised Date:

30 September 2014

Accepted Date: 25 November 2014

Please cite this article as: Ko, C.-Y., Chu, Y.-Y., Narumiya, S., Chi, J.-Y., Furuyashiki, T., Aoki, T., Wang, S.-M., Chang, W.-C., Wang, J.-M., The CCAAT/enhancer binding protein delta/miR135a/ thrombospondin 1 axis mediates PGE2-induced angiogenesis in Alzheimer’s disease, Neurobiology of Aging (2015), doi: 10.1016/j.neurobiolaging.2014.11.020. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT The CCAAT/enhancer binding protein delta/miR135a/thrombospondin 1 axis mediates PGE2-induced angiogenesis in Alzheimer’s disease

RI PT

Chiung-Yuan Ko1,2, Yu-Yi Chu4, Shuh Narumiya7, Jhih-Ying Chi6, Tomoyuki Furuyashiki7, Tomohiro Aoki7, Shao-Ming Wang6, Wen-Chang Chang3 and Ju-Ming Wang3,4,5,6

1

M AN U

SC

Ph.D. Program for Neural Regenerative Medicine, College of Medical Science and Technology, 2Center for Neurotrauma and Neuroregeneration, 3Graduate Institute of Medical Sciences, College of Medicine, Taipei Medical University, Taipei 11031, Taiwan, 4Institute of Bioinformatics and Biosignal Transduction, 5Infectious Disease and Signaling Research Center, 6Institute of Basic Medical Sciences, National Cheng Kung University, Tainan 701 and 7Core Research for Evolutional Science and Technology (CREST), Kyoto 606-8501, Japan

AC C

EP

TE D

*To whom all correspondence should be addressed: Ju-Ming Wang Institute of Bioinformatics and Biosignal Transduction College of Bioscience and Biotechnology National Cheng Kung University Tainan 701, Taiwan Phone: 886-6-2757575-31067 Fax: 886-6-2083663 E-mail: [email protected]

ACCEPTED MANUSCRIPT Abstract In Alzheimer’s disease (AD), large populations of endothelial cells undergo angiogenesis due to brain hypoxia and inflammation. Substantial evidence from epidemiological, pathological and clinical reports suggests that vascular factors are

RI PT

critical for the pathogenesis of AD. However, the precise mechanistic correlation between inflammation and angiogenesis in AD has not been well elucidated. Prostaglandin E2 (PGE2), a key factor of the inflammatory response, has been known

SC

to promote angiogenesis. In this study, we demonstrated that PGE2 acts through EP4 receptor and protein kinase A (PKA) to modulate CCAAT/enhancer binding protein

M AN U

delta (CEBPD) abundance in astrocytes. Attenuated vessel formation was observed in the brains of AppTg/Cebpd-/- mice. We showed that miR135a was responsive to the induction of CEBPD and further negatively regulated thrombospondin 1 (THBS1) transcription by directly targeting its 3’-untranslated region (3’UTR) in astrocytes.

TE D

Furthermore, conditioned media from astrocytes expressing miR135a promoted HUVECs tube-like formation, which correlated with the effects of PGE2 on angiogenesis. Our results indicated that CEBPD contributes to the repression of

EP

THBS1 transcription by activating the expression of miR135a in astrocytes following PGE2 treatment. We provided new evidence that astrocytic CEBPD increases

AC C

angiogenesis during AD pathogenesis. This discovery supports the negative influence of CEBPD activation in astrocytes with respect to AD pathogenesis and implies that the CEBPD/miR135a/THBS1 axis could be a therapeutic target of AD.

ACCEPTED MANUSCRIPT 1. Introduction Neuroinflammation has been proposed to be a causative factor in the pathogenesis of AD, Parkinson’s disease (PD), multiple sclerosis, spinal cord injury and stroke (Glass et al., 2010; Czirr and Wyss-Coray, 2012). Prominent activation of

RI PT

inflammatory processes was observed in the brains of AD patients (Akiyama et al., 2000; Ikonomovic et al., 2004; Wyss-Coray, 2006). Cyclooxygenase (COX) is a microsomal enzyme that is present in two isoforms, COX-1 and COX-2, and is

SC

essential for the synthesis of PGE2, a potent inducer of inflammation. Increased levels of PGE2 and COX-2 have been observed in the cerebrospinal fluid of AD patients and

M AN U

these levels also correlated with the degree of AD pathogenesis (Kitamura et al., 1999; Montine et al., 1999; Ho et al., 2001).

Inflammation is increasingly being recognized as a critical mediator of angiogenesis. A variety of studies support the hypothesis that pro-inflammatory

TE D

cytokines and inflammatory cells may modulate angiogenesis by directly or indirectly affecting endothelial cells, resulting in the promotion of the angiogenic process. The brains of AD patients show signs of significant hypervascularity and a high incidence

EP

of microangiopathy (Perlmutter et al., 1990; Desai et al., 2009; Biron et al., 2011). Coincident with this observation in the brains of AD patients, higher levels of

AC C

vasculature have been observed in the AD mouse model (Biron et al., 2011). However, the physiological significance of this angiogenesis and its causes and mechanisms, especially the communications among the neurons, endothelia cells and astrocytes, remain largely unknown. We know that PGE2 plays an important role in the angiogenic process of inflammation (Leahy et al., 2000; Salcedo et al., 2003; Zhang and Daaka, 2011). Recently,

a cycle of angiogenic events was suggested to contribute to Aβ

accumulation and neuronal death (Vagnucci and Li, 2003). Papers also showed that

ACCEPTED MANUSCRIPT the accumulating Aβ peptide in the brain parenchyma of AD patients is responsible for reduced vascular permeability (Bergamaschini et al., 2004), causing vessel disruption and loss of blood flow (Zekry et al., 2002). However, the details of PGE2-mediated signaling in astrocytes and its association with angiogenesis in AD

RI PT

remain unknown. The coactivation of the AMPA/kainate and metabotropic glutamate receptors on astrocytes stimulates these cells to release glutamate through a Ca2+-dependent process mediated by PGE2 (Bezzi et al., 1998). PGE2 can enhance

SC

tumor necrosis factor-α (TNF-α)/interferon-γ-induced astrocytic activation by augmenting the induction of nitric oxide synthase/nitric oxide production through

M AN U

EP2-mediated cross-talk between the cAMP/PKA and IP3/Ca2+ signaling pathways (Hsiao et al., 2007). However, the detailed mechanisms of PGE2 in astrocytes and its consequent effects on AD pathogenesis are still largely unknown. CEBPD belongs to the CCAAT/enhancer-binding protein (C/EBP) family and

TE D

acts as a transcription factor (Ramji and Foka, 2002). CEBPD, in particular, is known to regulate or co-regulate a wide range of inflammatory mediators and participate in signaling in response to interleukin-1β (IL-1β), interleukin-6 (IL-6) and TNFα (Ramji

EP

et al., 1993; Cardinaux et al., 2000). The induction of CEBPD expression has been observed in age-associated disorders, such as neuron degeneration disease (Li et al.,

AC C

2004), atherosclerosis (Yang et al., 2001), type 2 diabetes (Bennett et al., 2010) and rheumatoid arthritis (Chang et al., 2012). However, the physiological functions of CEBPD and its downstream targets in these inflammation-related diseases are poorly understood. Furthermore, IL-1β, IL-6 and TNFα are known to be increased in the pathological forms of neurodegenerative diseases, including AD and PD (Glass et al., 2010). CEBPD accumulates in reactive astrocytes surrounding Aβ peptide deposits in AD (Li et al., 2004; Ko et al., 2012). In response to IL-1β, CEBPD is up-regulated through the MAPK/p38 signaling pathway and is further phosphorylated by GSK3β at

ACCEPTED MANUSCRIPT Ser167 in astrocytes (Ko et al., 2014). The microglia-mediated Aβ clearance by phagocytosis has been mentioned (Simard et al., 2006; Lee and Landreth, 2010) but the phagocytotic activity is reduced in AD patients (Fiala et al., 2005; Fiala et al., 2007). The activation of astrocytic CEBPD was thought to attenuate the

RI PT

macrophage-mediated phagocytosis of damaged neurons and regulate the migration and activation of microglia and macrophages (Ko et al., 2012; Ko et al., 2014).

Thrombospondins (THBSs) are large, oligomeric, extracellular matrix proteins

SC

that mediate cell-cell and cell-matrix interactions by binding to other extracellular matrix proteins and an array of membrane receptors and cytokines. Studies have

M AN U

shown that THBSs participate in cell attachment, proliferation and differentiation, as well as in apoptosis and in the inhibition of angiogenesis (Bornstein, 2001; Adams and Lawler, 2004). The expression of THBS1 in astrocytes is modulated by different secretory factors, including platelet-derived growth factor, transforming growth

TE D

factor-β1, extracellular ATP and oxygen tension (Asch et al., 1986; Scott-Drew and ffrench-Constant, 1997; Song et al., 2002; Tran and Neary, 2006). In addition to promoting the attachment of neurons and the outgrowth of neurites (Neugebauer et al.,

EP

1991; O'Shea et al., 1991; Osterhout et al., 1992), THBS1 also critically supports astrocyte-induced synaptogenesis and neuronal migration in both normal and

AC C

pathological brain tissue, respectively (Christopherson et al., 2005; Blake et al., 2008).

MicroRNAs (miRNAs) are transcribed from the cellular genome and undergo

sequential processing by several RNase complexes to produce small mature miRNAs (≈21-25 nt) that regulate gene expression by targeting one or more mRNAs for translational repression or cleavage (He and Hannon, 2004). The function of miRNAs intersects the RNA-mediated gene-silencing pathways at the post-transcriptional level (Yu and Kumar, 2003). By targeting the mRNAs of protein-coding genes, miRNAs

ACCEPTED MANUSCRIPT play a critical role in a variety of biological processes, including development, cell growth, proliferation and metabolism (Bueno et al., 2008). Recently, it has been suggested that the deregulation of miRNAs participates in inflammation, autoimmunity, neurodegeneration, viral infections, heart diseases and cancer (Kanwar

RI PT

et al., 2010; Sonntag, 2010). To date, more than 400 miRNAs have been identified in human and chimpanzee brains (Berezikov et al., 2006; Landgraf et al., 2007). However, the functions of these miRNAs in the brain, especially in astrocytes, and

SC

how they are linked to the pathogenesis of neurodegeneration are unknown. miR135a is enriched in glial cells and has been shown to induce mitochondria-dependent

M AN U

apoptosis in malignant glioma (Wu et al., 2012), but the function of miR135a in

AC C

EP

TE D

inflamed brains has not been explored.

ACCEPTED MANUSCRIPT 2. Materials and methods 2.1. Materials PGE2 was purchased from Nacalai Tesque, Inc. (Kyoto, Japan). Agonists selective for each EP subtype (ONO-DI-004 for EP1, ONO-AE1-259 for EP2, ONO-AE-248 for

RI PT

EP3 and ONO-AE1-329 for EP4) and an EP4 selective antagonist (ONO-AE3-208) were kindly gifted by the Ono Pharmaceutical Company (Osaka, Japan). The selectivity of these compounds and the pharmacokinetics of ONO-AE3-208 have been

SC

previously reported (Suzawa et al., 2000; Narumiya and FitzGerald, 2001; Kabashima et al., 2002). The cAMP analogs (dibutyryl cAMP, N6-Bnz-cAMP and

M AN U

8-pCPT-2’-O-Me-cAMP), PI3 kinase inhibitor (wortmannin), PKA inhibitor (H-89) and anti-α-Tubulin antibody were purchased from Sigma (St. Louis, MO, USA). An anti-Sp1 antibody was purchased from Millipore (Billerica, MA, USA). Antibodies recognizing CEBPD and THBS1 were purchased from Santa Cruz Biotechnology

TE D

(Santa Cruz, CA, USA). Antibodies recognizing phospho-CREB and CREB were purchased from Cell Signaling Technology (Danvers, MA, USA). An anti-CD31 antibody was purchased from abcam (Cambridge, UK). The TRIzol RNA extraction

EP

reagent, Lipofectamine 2000, Dulbecco’s modified Eagle’s medium (DMEM) and Opti-MEM medium were purchased from Invitrogen (Carlsbad, CA, USA).

AC C

Endothelial cell medium (ECM) was purchased from ScienCell (Carlsbad, CA, USA). The SYBR Premix Ex Taq and the PrimeScript™ RT Reagent Kit were purchased from TaKaRa (Otsu, Shiga, Japan). All oligonucleotides were synthesized by Eurofins MWG Operon Inc. (Tokyo, Japan). Fetal bovine serum (FBS) was purchased from HyClone Laboratories (Logan, UT, USA).

2.2. Animals The AppTg (APPswe/PS1/E9 bigenic) mice were obtained from the Jackson

ACCEPTED MANUSCRIPT Laboratory (stock no. 004462, Bar Harbor, ME, USA). The AppTg mice were crossed with Cebpd−/− mice, a kind gift from Dr. E. Sterneck (Sterneck et al., 1998) on the C57BL/6 genetic background. Female mice heterozygous for AppTg mice was intercrossed with Cebpd−/− homozygous mice; the offspring (AppTg+/−/Cebpd+/−) were

RI PT

then bred to each other to produce the AppTg/Cebpd−/− mice in this study.

2.3. Cell culture

SC

U373MG (human glioblastoma-astrocytoma cell line) cells were cultured in DMEM supplemented with 10% fetal bovine serum, 100 µg/ml streptomycin and 100 units/ml

M AN U

penicillin. HUVECs were obtained from the Bioresource Collection and Research Center (Taiwan) and were maintained in ECM supplemented with 5% fetal bovine serum, 1% endothelial cell growth supplement, 100 µg/ml streptomycin and 100

TE D

units/ml penicillin.

2.4. Primary astroglial preparation

The cerebral cortices were dissected in phosphate-buffered saline, carefully stripped

EP

of their meninges and digested with 0.25% trypsin in Hanks’ Balanced Salt Solution (Invitrogen, Carlsbad, CA, USA) for 25 min at 37°C. Trypsinization was stopped by

AC C

the addition of an equal volume of culture medium containing 0.02% deoxyribonuclease I (Sigma, St. Louis, MO, USA). The solution was pelleted (5 min, 200 g) and then resuspended into a single-cell suspension in culture medium by repeated pipetting, followed by passage through a 105-µm pore mesh. The cells were seeded on plates coated with 20 µg/ml laminin (Invitrogen, Carlsbad, CA, USA). Laminin-coated plates favor astroglial growth and inhibit microglial growth (Milner and Campbell, 2002). When confluent, the cultures were treated with 10 µM cytosine arabinoside for 4 days. The cultures were used one day after the end of the cytosine

ACCEPTED MANUSCRIPT arabinoside treatment. The vast majority of the cells were type I astrocytes and microglia comprised less than 2% of the cells.

2.5. Quantitative real-time polymerase chain reaction (qRT-PCR)

RI PT

Total RNA was extracted using TRIzol according to the manufacturer’s protocol and 1 µg of RNA was subjected to reverse transcription-PCR using the PrimeScript™ RT Reagent Kit. The resultant cDNA was mixed with the SYBR Premix ExTaq kit and

SC

appropriate primers and quantitative PCR was performed in a Thermocycler C1000 (Bio-Rad, Hercules, CA, USA). The following specific primers were used for the

M AN U

qRT-PCR analysis: for CEBPD, 5’-AGCGCAACAACATCGCCGTG-3’ and 5’-GTC GGGTCTGAGGTATGGGTC-3’; for GAPDH, 5’-CCATCACCATCTTCCAGGAG3’ and 5’-CCTGCTTCACCACCTTCTTG-3’; for THBS1, 5’-GACCTGCCACATTC AGGAGT-3’ and 5’-CTTTCTTGCAGGCTTTGGTC-3’; for Thbs1, 5’-CCAAAGCC

TE D

TGCAAGAAAGAC-3’ and 5’-CCTGCTTGTTGCAAACTTGA-3’.

2.6. Microarray analysis

EP

Total RNA was extracted using TRIzol according to the manufacturer’s protocol. Samples were validated with Agilent Human and Primate miRNA one array (Phalanx

AC C

Biotech., Taipei, Taiwan), following the manufacturers’ protocols. All processes were performed by Phalanx Biotech Company (Taipei, Taiwan). Good quality signals were obtained by filtering for scores of p-value < 0.05 in all replicates, M-value of > 6 in all signals and more than 1.5-fold change.

2.7. TaqMan Reverse Transcription (RT)-PCR for miRNA quantification Total RNA was isolated using TRIzol according to the manufacturer’s protocol, reverse-transcribed using the Taq-Man® microRNA reverse transcription kit and

ACCEPTED MANUSCRIPT subjected to real-time PCR using the TaqMan® microRNA assay kit (Applied Biosystems).

2.8. Construction of plasmids

RI PT

The human CEBPD reporter and related reporters containing site-directed mutations were described as previously (Wang et al., 2005). The fragment of 3’-untranslated region (3’UTR) of THBS1 gene was cloned from U373MG cells by PCR with primers:

reverse,

forward,

5’-XbaI-TCTAGATGTAGCTTGTGCAGATGT-3’

SC

THBS1-3’UTR

5’-EcoRV-CTGAGATATCTATTCCAATGGCAATGAG-3’.

The

and

PCR

M AN U

fragment was further subcloned into pGL3-promoter vector. Mutant THBS1 3’UTR reporter was constructed by site-directed mutagenesis following the instructions of the QuikChang Site-directed Mutagenesis Kit (Stratagene, CA, USA) with the primer: THBS1-3’UTR mutant, 5’-AAATTGCAAAGAAAGATATCAGGTCTTCAATACT

TE D

GT-3’.

2.9. Plasmid transfection and reporter gene assay

EP

The cells were replated at an optimal density 24 h before transfection in 6-well plastic plates. The cells were then transfected with plasmids using the FuGENE® HD

AC C

Transfection Reagent (Promega, Madison, WI, USA) according to the manufacturer’s instructions. The total amount of DNA for each experiment was matched to their individual backbone vectors. Opti-MEM media were changed to conditioned media and the cells were incubated for 15 h. After transfection, the luciferase activities of the cell lysates were measured using the luciferase assay system as per the manufacturer’s instructions (Promega, Madison, WI, USA).

2.10. Pre-miR135a and shmiR135a-inducible stable cell construction

ACCEPTED MANUSCRIPT The primers for PCR using pre-mir135a forward (5’-AgeI-AGGCCTCGCTGTTCTC TATGGC-3’) and premir135a reverse (5’-PmeI-TGTCCCCGCCGTGCG-3’) to generate pre-miR135a construction in pAS4w.1.Pneo, a tetracycline inducible system of lentiviral expression vector. The shmiR135a construction in pLAS1w.3xLacO, a expression

vector,

used

the

DNA

fragment

as

follows:

RI PT

lentiviral

5'-TATGGCTTTTTATTCCTATGTGACTCGAGTCACATAGGAATAAAAAGCCAT ATTTTT-3'. Stable U373MG cells containing Pre-miR135a and shmiR135a were

SC

generated by pLAS.AS3w.aOn.Pbsd lentiviral infectants and parental U373MG cells, respectively. Doxycycline (2 µg/ml) was used to induce miR135a expression. IPTG

M AN U

(500 µM) was used to induce shmiR135a expression. The lentiviral expression vectors were obtained from the National RNAi Core Facility located at the Genomic Research Center of Institute of Molecular Biology, Academia Sinica (Taiwan).

TE D

2.11. Western blot analysis

For Western blot analysis, the cells were lysed in modified RIPA buffer containing 50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1 mM EDTA, 1% Nonidet P-40, 0.25%

EP

sodium deoxycholate, 1 mM DTT, 10 mM NaF, 1 mM PMSF, 1 µg/ml aprotinin, 1 µg/ml leupeptin and 1 mM Na3VO4. The lysates were resolved on a sodium dodecyl

AC C

sulfate (SDS)-containing 10% polyacrylamide gel, transferred to a polyvinylidene difluoride (PVDF) nylon membrane and probed with specific antibodies at 4°C overnight. The levels of interested proteins were detected using a horseradish peroxidase-conjugated

antibody

and

were

revealed

using

an

enhanced

chemiluminescence (ECL) Western blot system from Pierce (Rockford, IL, USA).

2.12. Lentiviral knockdown assay Virus was produced from Phoenix cells by cotransfection of the various small hairpin

ACCEPTED MANUSCRIPT RNA (shRNA) expression vectors in combination with pMD2.G and psPAX2. After determining the viral infection efficiency, 10 multiplicity of infection of lentivirus containing shβ-galactosidase (shLacZ) or shCEBPD were used to infect U373MG cells for 96 h. In all lentiviral experiments, medium containing uninfected viruses was

RI PT

removed before conducting further assays. The shRNA sequences in lentiviral expression vectors were: shβ-galactosidase: 5’-CCGGTGTTCGCATTATCCGAACC ATCTCGAGATGGTTCGGATAATGCGAACATTTTTG-3’, shCEBPD (shC7): 5’-

SC

CCGGGCTGTCGGCTGAGAACGAGAACTCGAGTTCTCGTTCTCAGCCGACA GCTTTTT-3’ and shCebpd (shE4) 5’-CCGGCAGCGCCTACATTGACTCCATCTC

M AN U

GAGATGGAGTCAATGTAGGCGCTGTTTTTG-3’. The lentiviral

knockdown

expression vectors were obtained from the National RNAi Core Facility located at the Genomic Research Center of Institute of Molecular Biology, Academia Sinica

TE D

(Taiwan).

2.13. Chromatin immunoprecipitation (ChIP) assay The ChIP assay was performed essentially as described by Wang et al. (Wang et al.,

EP

2006). Briefly, U373MG cells were treated with 1% formaldehyde at room temperature. The cross-linked chromatin was then sonicated to an average size of 500

AC C

bp. The chromatin fragments were then immunoprecipitated with antibodies specific for Sp1, CREB or control rabbit immunoglobulin G (IgG) at 4°C overnight. After the cross-linking was reversed, the immunoprecipitated chromatin was amplified using primers targeting specific regions of the target genomic loci. The following primers were used: 5’-CGAGGAGGTTCCAAGCCCAC-3’ and 5’-GGCTGTCACCTCGCT GGGCC-3’. The amplified DNA products were resolved by agarose gel electrophoresis and confirmed by sequencing.

ACCEPTED MANUSCRIPT 2.14. In vitro capillary tube-like formation assay Forty-eight well plated were pre-coated with Matrigel (BD Biosciences, Mountain View, CA, USA), which was allowed to solidify for 1 h at 37°C. HUVEC endothelial cells were (3 × 104 cells/well) suspended in ECM medium or combined with different

RI PT

conditional media and were added to each well. After 6 h incubation, phase-contrast images of HUVECs seeded on Matrigel were captured by using an inverted light microscope (LEICA CTR 4000, Switzerland) at 10x magnification. Images were

SC

analyzed and quantified by using image analysis services provided by Wimasis

2.15. Immunohistochemistry

M AN U

(Munich, Germany).

Male mouse brain tissue dissections from AppTg or AppTg/Cebpd−/− mice (~15–16 months of age) were postfixed in 4% buffered paraformaldehyde for 24–36 hours,

TE D

cryoprotected, sectioned at 10 µm on a freezing microtome. The frozen sections were treated with protein blocker/antibody diluents (Bio SB, Santa Barbara, CA, USA) for 1 h. In the same buffer solution, the sections were incubated overnight with

EP

anti-CD31 antibody at 4 °C. After incubation, the sections were washed in PBS and incubated in the ABC kit solutions (Vector, Burlingame, CA, USA) for 1.5 h. The

AC C

sections were stained with diaminobenzidine (DAB, Sigma, St. Louis, MO, USA) and mounted on slides and sealed with coverslips. The images were observed with phase contrast on BX51 microscope (Olympus, Tokyo, Japan) with a 40x objective and using DP70 digital camera system combined with DP Controller software (Olympus, Tokyo, Japan). Vessels were counted in four acquired non-overlapping images per slice section of each animal. For the quantification of microvessels, the longitudinal vessel segment between the two nodes (join points) was considered as a vessel unit. All quantifications were performed by an observer blinded to the experimental

ACCEPTED MANUSCRIPT condition of each animal.

2.16. Statistical analysis The data were analyzed by one-tailed unpaired Student’s t test for two groups of

RI PT

experiment using Prism 5 software. For three or more groups of experiment, the data were analyzed by one-way analysis of variance (ANOVA) followed by the

AC C

EP

TE D

M AN U

0.05 was considered significant in all comparisons.

SC

Bonferroni’s multiple comparison test using Prism 5 software. A probability of P

thrombospondin 1 axis mediates PGE2-induced angiogenesis in Alzheimer's disease.

In Alzheimer's disease (AD), large populations of endothelial cells undergo angiogenesis due to brain hypoxia and inflammation. Substantial evidence f...
1MB Sizes 2 Downloads 7 Views